SWE-Dev/SWE-Dev-3 · Datasets at Hugging Face

repo stringlengths 7 59	instance_id stringlengths 11 63	base_commit stringlengths 40 40	patch stringlengths 167 798k	problem_statement stringlengths 20 65.2k	hints_text stringlengths 0 142k	created_at timestamp[ns]date 2015-08-30 10:31:05 2024-12-13 16:08:19	FAIL_TO_PASS sequencelengths 0 0	PASS_TO_PASS sequencelengths 0 0
securefederatedai/openfl	securefederatedai__openfl-516	3adc29d3bdc0fcc71f6bd3fc578a1d0fad88a41f	diff --git a/openfl-tutorials/Federated_PyTorch_UNET_Tutorial.ipynb b/openfl-tutorials/Federated_PyTorch_UNET_Tutorial.ipynb index 72f75aef54..7ee6c2e692 100644 --- a/openfl-tutorials/Federated_PyTorch_UNET_Tutorial.ipynb +++ b/openfl-tutorials/Federated_PyTorch_UNET_Tutorial.ipynb @@ -76,8 +76,8 @@ "outputs": [], "source": [ "!wget 'https://datasets.simula.no/downloads/hyper-kvasir/hyper-kvasir-segmented-images.zip' -O kvasir.zip\n", - "ZIP_SHA384 = 'e30d18a772c6520476e55b610a4db457237f151e'\\\n", - " '19182849d54b49ae24699881c1e18e0961f77642be900450ef8b22e7'\n", + "ZIP_SHA384 = ('66cd659d0e8afd8c83408174'\n", + " '1ade2b75dada8d4648b816f2533c8748b1658efa3d49e205415d4116faade2c5810e241e')\n", "validate_file_hash('./kvasir.zip', ZIP_SHA384)\n", "!unzip -n kvasir.zip -d ./data" ] @@ -525,8 +525,19 @@ } ], "metadata": { + "kernelspec": { + "display_name": "Python 3.8.10 64-bit", + "language": "python", + "name": "python3" + }, "language_info": { - "name": "python" + "name": "python", + "version": "3.8.10" + }, + "vscode": { + "interpreter": { + "hash": "916dbcbb3f70747c44a77c7bcd40155683ae19c65e1c03b4aa3499c5328201f1" + } } }, "nbformat": 4, diff --git a/openfl-tutorials/interactive_api/PyTorch_Kvasir_UNet/envoy/kvasir_shard_descriptor.py b/openfl-tutorials/interactive_api/PyTorch_Kvasir_UNet/envoy/kvasir_shard_descriptor.py index 1107d2decf..2a83543250 100644 --- a/openfl-tutorials/interactive_api/PyTorch_Kvasir_UNet/envoy/kvasir_shard_descriptor.py +++ b/openfl-tutorials/interactive_api/PyTorch_Kvasir_UNet/envoy/kvasir_shard_descriptor.py @@ -100,8 +100,8 @@ def download_data(data_folder): "hyper-kvasir/hyper-kvasir-segmented-images.zip'" f' -O {zip_file_path.relative_to(Path.cwd())}' ) - zip_sha384 = ('e30d18a772c6520476e55b610a4db457237f151e' - '19182849d54b49ae24699881c1e18e0961f77642be900450ef8b22e7') + zip_sha384 = ('66cd659d0e8afd8c83408174' + '1ade2b75dada8d4648b816f2533c8748b1658efa3d49e205415d4116faade2c5810e241e') validate_file_hash(zip_file_path, zip_sha384) os.system(f'unzip -n {zip_file_path.relative_to(Path.cwd())}' f' -d {data_folder.relative_to(Path.cwd())}') diff --git a/openfl-tutorials/interactive_api/PyTorch_Kvasir_UNet/envoy/kvasir_shard_descriptor_with_data_splitter.py b/openfl-tutorials/interactive_api/PyTorch_Kvasir_UNet/envoy/kvasir_shard_descriptor_with_data_splitter.py index 687b1ceb87..09bcde87c0 100644 --- a/openfl-tutorials/interactive_api/PyTorch_Kvasir_UNet/envoy/kvasir_shard_descriptor_with_data_splitter.py +++ b/openfl-tutorials/interactive_api/PyTorch_Kvasir_UNet/envoy/kvasir_shard_descriptor_with_data_splitter.py @@ -97,8 +97,8 @@ def download_data(data_folder): " 'https://datasets.simula.no/downloads/hyper-kvasir/" "hyper-kvasir-segmented-images.zip'" f' -O {zip_file_path.relative_to(Path.cwd())}') - zip_sha384 = ('e30d18a772c6520476e55b610a4db457237f151e' - '19182849d54b49ae24699881c1e18e0961f77642be900450ef8b22e7') + zip_sha384 = ('66cd659d0e8afd8c83408174' + '1ade2b75dada8d4648b816f2533c8748b1658efa3d49e205415d4116faade2c5810e241e') validate_file_hash(zip_file_path, zip_sha384) os.system(f'unzip -n {zip_file_path.relative_to(Path.cwd())}' f' -d {data_folder.relative_to(Path.cwd())}') diff --git a/openfl-workspace/torch_unet_kvasir/src/data_loader.py b/openfl-workspace/torch_unet_kvasir/src/data_loader.py index b038a884ef..15db0d63c1 100644 --- a/openfl-workspace/torch_unet_kvasir/src/data_loader.py +++ b/openfl-workspace/torch_unet_kvasir/src/data_loader.py @@ -32,7 +32,7 @@ def read_data(image_path, mask_path): """ img = io.imread(image_path) - assert(img.shape[2] == 3) + assert (img.shape[2] == 3) mask = io.imread(mask_path) return (img, mask[:, :, 0].astype(np.uint8)) @@ -59,7 +59,7 @@ def __init__(self, is_validation, shard_num, collaborator_count, kwargs): self.images_names = self.images_names[shard_num:: collaborator_count] self.is_validation = is_validation - assert(len(self.images_names) > 8) + assert (len(self.images_names) > 8) validation_size = len(self.images_names) // 8 if is_validation: @@ -92,8 +92,8 @@ def __len__(self): def load_kvasir_dataset(): """Load and unzip kvasir dataset.""" - zip_sha384 = ('e30d18a772c6520476e55b610a4db457237f151e' - '19182849d54b49ae24699881c1e18e0961f77642be900450ef8b22e7') + zip_sha384 = ('66cd659d0e8afd8c83408174' + '1ade2b75dada8d4648b816f2533c8748b1658efa3d49e205415d4116faade2c5810e241e') data_url = ('https://datasets.simula.no/downloads/' 'hyper-kvasir/hyper-kvasir-segmented-images.zip') filename = 'kvasir.zip' diff --git a/openfl-workspace/torch_unet_kvasir_gramine_ready/src/data_loader.py b/openfl-workspace/torch_unet_kvasir_gramine_ready/src/data_loader.py index b038a884ef..15db0d63c1 100644 --- a/openfl-workspace/torch_unet_kvasir_gramine_ready/src/data_loader.py +++ b/openfl-workspace/torch_unet_kvasir_gramine_ready/src/data_loader.py @@ -32,7 +32,7 @@ def read_data(image_path, mask_path): """ img = io.imread(image_path) - assert(img.shape[2] == 3) + assert (img.shape[2] == 3) mask = io.imread(mask_path) return (img, mask[:, :, 0].astype(np.uint8)) @@ -59,7 +59,7 @@ def __init__(self, is_validation, shard_num, collaborator_count, kwargs): self.images_names = self.images_names[shard_num:: collaborator_count] self.is_validation = is_validation - assert(len(self.images_names) > 8) + assert (len(self.images_names) > 8) validation_size = len(self.images_names) // 8 if is_validation: @@ -92,8 +92,8 @@ def __len__(self): def load_kvasir_dataset(): """Load and unzip kvasir dataset.""" - zip_sha384 = ('e30d18a772c6520476e55b610a4db457237f151e' - '19182849d54b49ae24699881c1e18e0961f77642be900450ef8b22e7') + zip_sha384 = ('66cd659d0e8afd8c83408174' + '1ade2b75dada8d4648b816f2533c8748b1658efa3d49e205415d4116faade2c5810e241e') data_url = ('https://datasets.simula.no/downloads/' 'hyper-kvasir/hyper-kvasir-segmented-images.zip') filename = 'kvasir.zip'	Example fails: Dataset hash needs correction I faced errors running `torch_unet_kvasir` and `torch_unet_kvasir_gramine_ready` examples. The message is attached [error.log](https://github.com/intel/openfl/files/9597992/error.log)		2022-09-28T15:01:54	[]	[]
securefederatedai/openfl	securefederatedai__openfl-369	b133a439dd44b55f66076aeaaef26f0e60e3119a	diff --git a/openfl/component/aggregator/aggregator.py b/openfl/component/aggregator/aggregator.py index 590cf9f243..c4f092058f 100644 --- a/openfl/component/aggregator/aggregator.py +++ b/openfl/component/aggregator/aggregator.py @@ -11,6 +11,7 @@ from openfl.pipelines import TensorCodec from openfl.protocols import base_pb2 from openfl.protocols import utils +from openfl.utilities import change_tags from openfl.utilities import TaskResultKey from openfl.utilities import TensorKey from openfl.utilities.logs import write_metric @@ -328,7 +329,7 @@ def get_aggregated_tensor(self, collaborator_name, tensor_name, require_lossless: bool round_number: int report: bool - tags: list[str] + tags: tuple[str, ...] Returns: named_tensor : protobuf NamedTensor the tensor requested by the collaborator @@ -341,18 +342,16 @@ def get_aggregated_tensor(self, collaborator_name, tensor_name, else: compress_lossless = False - tags = list(tags) - # TODO the TensorDB doesn't support compressed data yet. # The returned tensor will # be recompressed anyway. if 'compressed' in tags: - tags.remove('compressed') + tags = change_tags(tags, remove_field='compressed') if 'lossy_compressed' in tags: - tags.remove('lossy_compressed') + tags = change_tags(tags, remove_field='lossy_compressed') tensor_key = TensorKey( - tensor_name, self.uuid, round_number, report, tuple(tags) + tensor_name, self.uuid, round_number, report, tags ) tensor_name, origin, round_number, report, tags = tensor_key @@ -583,10 +582,8 @@ def _process_named_tensor(self, named_tensor, collaborator_name): ) dec_name, dec_origin, dec_round_num, dec_report, dec_tags = dec_tk # Need to add the collaborator tag to the resulting tensor - if type(dec_tags) == str: - new_tags = tuple([dec_tags] + [collaborator_name]) - else: - new_tags = tuple(list(dec_tags) + [collaborator_name]) + new_tags = change_tags(dec_tags, add_field=collaborator_name) + # layer.agg.n.trained.delta.col_i decompressed_tensor_key = TensorKey( dec_name, dec_origin, dec_round_num, dec_report, new_tags @@ -599,10 +596,7 @@ def _process_named_tensor(self, named_tensor, collaborator_name): require_lossless=False ) dec_name, dec_origin, dec_round_num, dec_report, dec_tags = dec_tk - if type(dec_tags) == str: - new_tags = tuple([dec_tags] + [collaborator_name]) - else: - new_tags = tuple(list(dec_tags) + [collaborator_name]) + new_tags = change_tags(dec_tags, add_field=collaborator_name) # layer.agg.n.trained.delta.lossy_decompressed.col_i decompressed_tensor_key = TensorKey( dec_name, dec_origin, dec_round_num, dec_report, new_tags @@ -778,11 +772,11 @@ def _compute_validation_related_task_metrics(self, task_name): task_key = TaskResultKey(task_name, collaborators_for_task[0], self.round_number) for tensor_key in self.collaborator_tasks_results[task_key]: tensor_name, origin, round_number, report, tags = tensor_key - assert (tags[-1] == collaborators_for_task[0]), ( + assert (collaborators_for_task[0] in tags), ( f'Tensor {tensor_key} in task {task_name} has not been processed correctly' ) # Strip the collaborator label, and lookup aggregated tensor - new_tags = tuple(tags[:-1]) + new_tags = change_tags(tags, remove_field=collaborators_for_task[0]) agg_tensor_key = TensorKey(tensor_name, origin, round_number, report, new_tags) agg_tensor_name, agg_origin, agg_round_number, agg_report, agg_tags = agg_tensor_key agg_function = WeightedAverage() if 'metric' in tags else task_agg_function diff --git a/openfl/databases/tensor_db.py b/openfl/databases/tensor_db.py index 077e1d2fe7..b412212a19 100644 --- a/openfl/databases/tensor_db.py +++ b/openfl/databases/tensor_db.py @@ -12,6 +12,7 @@ import pandas as pd from openfl.component.aggregation_functions import AggregationFunction +from openfl.utilities import change_tags from openfl.utilities import LocalTensor from openfl.utilities import TensorKey @@ -145,10 +146,7 @@ def get_aggregated_tensor(self, tensor_key: TensorKey, collaborator_weight_dict: return np.array(raw_df.iloc[0]), {} for col in collaborator_names: - if type(tags) == str: - new_tags = tuple([tags] + [col]) - else: - new_tags = tuple(list(tags) + [col]) + new_tags = change_tags(tags, add_field=col) raw_df = self.tensor_db[ (self.tensor_db['tensor_name'] == tensor_name) & (self.tensor_db['origin'] == origin) diff --git a/openfl/federated/task/runner_fe.py b/openfl/federated/task/runner_fe.py index acace8c61b..f76419cb14 100644 --- a/openfl/federated/task/runner_fe.py +++ b/openfl/federated/task/runner_fe.py @@ -4,6 +4,7 @@ import numpy as np +from openfl.utilities import change_tags from openfl.utilities import split_tensor_dict_for_holdouts from openfl.utilities import TensorKey from .runner import TaskRunner @@ -181,7 +182,8 @@ def validate(self, col_name, round_num, input_tensor_dict, kwargs): suffix += '_local' else: suffix += '_agg' - tags = ('metric', suffix) + tags = ('metric',) + tags = change_tags(tags, add_field=suffix) output_tensor_dict = { TensorKey( metric, origin, round_num, True, tags diff --git a/openfl/federated/task/runner_keras.py b/openfl/federated/task/runner_keras.py index 93ac40b1d0..331e01414c 100644 --- a/openfl/federated/task/runner_keras.py +++ b/openfl/federated/task/runner_keras.py @@ -11,6 +11,7 @@ import numpy as np +from openfl.utilities import change_tags from openfl.utilities import Metric from openfl.utilities import split_tensor_dict_for_holdouts from openfl.utilities import TensorKey @@ -230,7 +231,8 @@ def validate(self, col_name, round_num, input_tensor_dict, kwargs): suffix += '_local' else: suffix += '_agg' - tags = ('metric', suffix) + tags = ('metric',) + tags = change_tags(tags, add_field=suffix) output_tensor_dict = { TensorKey(metric, origin, round_num, True, tags): np.array(ret_dict[metric]) diff --git a/openfl/federated/task/runner_pt.py b/openfl/federated/task/runner_pt.py index 2905f09798..0746a48f76 100644 --- a/openfl/federated/task/runner_pt.py +++ b/openfl/federated/task/runner_pt.py @@ -12,6 +12,7 @@ import torch.nn as nn import tqdm +from openfl.utilities import change_tags from openfl.utilities import Metric from openfl.utilities import split_tensor_dict_for_holdouts from openfl.utilities import TensorKey @@ -117,7 +118,8 @@ def validate(self, col_name, round_num, input_tensor_dict, suffix += '_local' else: suffix += '_agg' - tags = ('metric', suffix) + tags = ('metric',) + tags = change_tags(tags, add_field=suffix) # TODO figure out a better way to pass in metric for this pytorch # validate function output_tensor_dict = { diff --git a/openfl/federated/task/task_runner.py b/openfl/federated/task/task_runner.py index 05023e4743..05d8ec0e83 100644 --- a/openfl/federated/task/task_runner.py +++ b/openfl/federated/task/task_runner.py @@ -6,6 +6,7 @@ import numpy as np +from openfl.utilities import change_tags from openfl.utilities import split_tensor_dict_for_holdouts from openfl.utilities import TensorKey @@ -73,7 +74,7 @@ def _prepare_tensorkeys_for_agggregation(self, metric_dict, validation_flag, else: suffix = 'validate' + validation_flag tags = (suffix,) - tags = ('metric', tags) + tags = change_tags(tags, add_field='metric') metric_dict = { TensorKey(metric, origin, round_num, True, tags): np.array(value) for metric, value in metric_dict.items() diff --git a/openfl/pipelines/tensor_codec.py b/openfl/pipelines/tensor_codec.py index 5312a8c0f1..85dfa8c552 100644 --- a/openfl/pipelines/tensor_codec.py +++ b/openfl/pipelines/tensor_codec.py @@ -6,6 +6,7 @@ import numpy as np from openfl.pipelines import NoCompressionPipeline +from openfl.utilities import change_tags from openfl.utilities import TensorKey @@ -67,9 +68,9 @@ def compress(self, tensor_key, data, require_lossless=False, kwargs): # returned ('trained.delta'->'trained.delta.lossy_compressed') tensor_name, origin, round_number, report, tags = tensor_key if not self.compression_pipeline.is_lossy() or require_lossless: - new_tags = tuple(list(tags) + ['compressed']) + new_tags = change_tags(tags, add_field='compressed') else: - new_tags = tuple(list(tags) + ['lossy_compressed']) + new_tags = change_tags(tags, add_field='lossy_compressed') compressed_tensor_key = TensorKey( tensor_name, origin, round_number, report, new_tags) return compressed_tensor_key, compressed_nparray, metadata @@ -121,18 +122,16 @@ def decompress(self, tensor_key, data, transformer_metadata, data, transformer_metadata, kwargs) # Define the decompressed tensorkey that should be returned if 'lossy_compressed' in tags: - lc_idx = tags.index('lossy_compressed') - new_tags = list(tags) - new_tags[lc_idx] = 'lossy_decompressed' + new_tags = change_tags( + tags, add_field='lossy_decompressed', remove_field='lossy_compressed') decompressed_tensor_key = TensorKey( - tensor_name, origin, round_number, report, tuple(new_tags)) + tensor_name, origin, round_number, report, new_tags) elif 'compressed' in tags: # 'compressed' == lossless compression; no need for # compression related tag after decompression - new_tags = list(tags) - new_tags.remove('compressed') + new_tags = change_tags(tags, remove_field='compressed') decompressed_tensor_key = TensorKey( - tensor_name, origin, round_number, report, tuple(new_tags)) + tensor_name, origin, round_number, report, new_tags) else: raise NotImplementedError( 'Decompression is only supported on compressed data') @@ -170,10 +169,7 @@ def generate_delta(tensor_key, nparray, base_model_nparray): assert 'model' not in tags, ( 'The tensorkey should be provided ' 'from the layer with new weights, not the base model') - if type(tags) == str: - new_tags = tuple([tensor_key[3]] + ['delta']) - else: - new_tags = tuple(list(tags) + ['delta']) + new_tags = change_tags(tags, add_field='delta') delta_tensor_key = TensorKey( tensor_name, origin, round_number, report, new_tags) return delta_tensor_key, nparray - base_model_nparray @@ -209,9 +205,7 @@ def apply_delta(tensor_key, delta, base_model_nparray, creates_model=False): # from the base model' # Aggregator UUID has the prefix 'aggregator' if 'aggregator' in origin and not creates_model: - tags = list(tags) - tags.remove('delta') - new_tags = tuple(tags) + new_tags = change_tags(tags, remove_field='delta') new_model_tensor_key = TensorKey( tensor_name, origin, round_number, report, new_tags) else: diff --git a/openfl/transport/grpc/aggregator_server.py b/openfl/transport/grpc/aggregator_server.py index a62f18ccc1..cb19a56d77 100644 --- a/openfl/transport/grpc/aggregator_server.py +++ b/openfl/transport/grpc/aggregator_server.py @@ -185,7 +185,7 @@ def GetAggregatedTensor(self, request, context): # NOQA:N802 require_lossless = request.require_lossless round_number = request.round_number report = request.report - tags = request.tags + tags = tuple(request.tags) named_tensor = self.aggregator.get_aggregated_tensor( collaborator_name, tensor_name, round_number, report, tags, require_lossless) diff --git a/openfl/utilities/utils.py b/openfl/utilities/utils.py index 8bd0feb313..bc04aec990 100644 --- a/openfl/utilities/utils.py +++ b/openfl/utilities/utils.py @@ -236,3 +236,25 @@ def merge_configs( settings.set(key, operation(value)) settings.validators.validate() return settings + + +def change_tags(tags, , add_field=None, remove_field=None) -> Tuple[str, ...]: + """Change tensor tags to add or remove fields. + + Args: + tags(tuple): tensor tags. + add_field(str): add a new tensor tag field. + remove_field(str): remove a tensor tag field. + """ + tags = list(set(tags)) + + if add_field is not None and add_field not in tags: + tags.append(add_field) + if remove_field is not None: + if remove_field in tags: + tags.remove(remove_field) + else: + raise Exception(f'{remove_field} not in tags {tuple(tags)}') + + tags = tuple(sorted(tags)) + return tags	Get organised with tensor tags type. Tensor tags, that are used by the aggregator, can be different types inside the aggregator in different methods. It can be list, tuple, str or <class 'google.protobuf.pyext._message.RepeatedScalarContainer'>. One methods cast it to list, others to tuple. We should unify it.		2022-03-07T20:33:28	[]	[]
securefederatedai/openfl	securefederatedai__openfl-149	8dfe170b5d45e02daa34a2bff63d715ce996caf3	diff --git a/openfl-tutorials/interactive_api/Director_Pytorch_Kvasir_UNET/director_folder/director_config.yaml b/openfl-tutorials/interactive_api/Director_Pytorch_Kvasir_UNET/director_folder/director_config.yaml index 9b45b60869..bc52829c27 100644 --- a/openfl-tutorials/interactive_api/Director_Pytorch_Kvasir_UNET/director_folder/director_config.yaml +++ b/openfl-tutorials/interactive_api/Director_Pytorch_Kvasir_UNET/director_folder/director_config.yaml @@ -1,4 +1,5 @@ settings: - listen_ip: localhost + listen_addr: localhost + listen_port: 50051 sample_shape: ['300', '400', '3'] target_shape: ['300', '400'] \ No newline at end of file diff --git a/openfl-tutorials/interactive_api/Director_Pytorch_Kvasir_UNET/envoy_folder/kvasir_shard_descriptor.py b/openfl-tutorials/interactive_api/Director_Pytorch_Kvasir_UNET/envoy_folder/kvasir_shard_descriptor.py index bb6dc92681..ef7e5f5406 100644 --- a/openfl-tutorials/interactive_api/Director_Pytorch_Kvasir_UNET/envoy_folder/kvasir_shard_descriptor.py +++ b/openfl-tutorials/interactive_api/Director_Pytorch_Kvasir_UNET/envoy_folder/kvasir_shard_descriptor.py @@ -119,11 +119,13 @@ def dataset_description(self) -> str: from openfl.component.envoy.envoy import Envoy shard_name = 'one' - director_uri = 'localhost:50051' + director_host = 'localhost' + director_port = 50051 keeper = Envoy( shard_name=shard_name, - director_uri=director_uri, + director_host=director_host, + director_port=director_port, shard_descriptor=kvasir_sd, tls=True, root_certificate='./cert/root_ca.crt', diff --git a/openfl-tutorials/interactive_api/Director_Pytorch_Kvasir_UNET/envoy_folder/start_envoy.sh b/openfl-tutorials/interactive_api/Director_Pytorch_Kvasir_UNET/envoy_folder/start_envoy.sh index c546d40138..222d3988e0 100644 --- a/openfl-tutorials/interactive_api/Director_Pytorch_Kvasir_UNET/envoy_folder/start_envoy.sh +++ b/openfl-tutorials/interactive_api/Director_Pytorch_Kvasir_UNET/envoy_folder/start_envoy.sh @@ -1,4 +1,4 @@ #!/bin/bash set -e -fx envoy start -n env_one --disable-tls --shard-config-path shard_config.yaml -d localhost:50051 \ No newline at end of file +fx envoy start -n env_one --disable-tls --shard-config-path shard_config.yaml -dh localhost -dp 50051 \ No newline at end of file diff --git a/openfl-tutorials/interactive_api/Director_Pytorch_Kvasir_UNET/envoy_folder/start_envoy_with_tls.sh b/openfl-tutorials/interactive_api/Director_Pytorch_Kvasir_UNET/envoy_folder/start_envoy_with_tls.sh index 295f61b101..b4339cc0cb 100644 --- a/openfl-tutorials/interactive_api/Director_Pytorch_Kvasir_UNET/envoy_folder/start_envoy_with_tls.sh +++ b/openfl-tutorials/interactive_api/Director_Pytorch_Kvasir_UNET/envoy_folder/start_envoy_with_tls.sh @@ -3,4 +3,4 @@ set -e ENVOY_NAME=$1 DIRECTOR_FQDN=$2 -fx envoy start -n "$ENVOY_NAME" --shard-config-path shard_config.yaml -d "$DIRECTOR_FQDN":50051 -rc cert/root_ca.crt -pk cert/"$ENVOY_NAME".key -oc cert/"$ENVOY_NAME".crt \ No newline at end of file +fx envoy start -n "$ENVOY_NAME" --shard-config-path shard_config.yaml -dh"$DIRECTOR_FQDN" -dp 50051 -rc cert/root_ca.crt -pk cert/"$ENVOY_NAME".key -oc cert/"$ENVOY_NAME".crt \ No newline at end of file diff --git a/openfl/component/envoy/envoy.py b/openfl/component/envoy/envoy.py index de3b036779..ab02615b4b 100644 --- a/openfl/component/envoy/envoy.py +++ b/openfl/component/envoy/envoy.py @@ -24,7 +24,7 @@ class Envoy: """Envoy class.""" - def __init__(self, shard_name, director_uri, shard_descriptor, + def __init__(self, , shard_name, director_host, director_port, shard_descriptor, root_certificate: str = None, private_key: str = None, certificate: str = None, tls: bool = True) -> None: """Initialize a envoy object.""" @@ -33,11 +33,15 @@ def __init__(self, shard_name, director_uri, shard_descriptor, root_certificate).absolute() if root_certificate is not None else None self.private_key = Path(private_key).absolute() if root_certificate is not None else None self.certificate = Path(certificate).absolute() if root_certificate is not None else None - self.director_client = ShardDirectorClient(director_uri, shard_name=shard_name, - tls=tls, - root_certificate=root_certificate, - private_key=private_key, - certificate=certificate) + self.director_client = ShardDirectorClient( + director_host=director_host, + director_port=director_port, + shard_name=shard_name, + tls=tls, + root_certificate=root_certificate, + private_key=private_key, + certificate=certificate + ) self.shard_descriptor = shard_descriptor self.executor = ThreadPoolExecutor() self.running_experiments = {} diff --git a/openfl/interface/director.py b/openfl/interface/director.py index 2c182aa42c..bd06dbad6d 100644 --- a/openfl/interface/director.py +++ b/openfl/interface/director.py @@ -48,14 +48,14 @@ def start(director_config_path, tls, root_certificate, private_key, certificate) sample_shape = settings.get('sample_shape', '') target_shape = settings.get('target_shape', '') logger.info(f'Sample shape: {sample_shape}, target shape: {target_shape}') - listen_ip = settings.get('listen_ip') + listen_addr = settings.get('listen_addr') listen_port = settings.get('listen_port') root_certificate = root_certificate or settings.get('root_certificate') private_key = private_key or settings.get('private_key') certificate = certificate or settings.get('certificate') kwargs = {} - if listen_ip: - kwargs['listen_ip'] = listen_ip + if listen_addr: + kwargs['listen_addr'] = listen_addr if listen_port: kwargs['listen_port'] = listen_port director_server = DirectorGRPCServer( diff --git a/openfl/interface/envoy.py b/openfl/interface/envoy.py index ffd0a4fdb1..bde3fc4cd1 100644 --- a/openfl/interface/envoy.py +++ b/openfl/interface/envoy.py @@ -31,8 +31,10 @@ def envoy(context): @envoy.command(name='start') @option('-n', '--shard-name', required=True, help='Current shard name') -@option('-d', '--director-uri', required=True, +@option('-dh', '--director-host', required=True, help='The FQDN of the federation director') +@option('-dp', '--director-port', required=True, + help='The federation director port') @option('--tls/--disable-tls', default=True, is_flag=True, help='Use TLS or not (By default TLS is enabled)') @option('-sc', '--shard-config-path', default='shard_config.yaml', @@ -43,7 +45,7 @@ def envoy(context): help='Path to a private key') @option('-oc', '--public-cert-path', 'certificate', default=None, help='Path to a signed certificate') -def start_(shard_name, director_uri, tls, shard_config_path, +def start_(shard_name, director_host, director_port, tls, shard_config_path, root_certificate, private_key, certificate): """Start the Envoy.""" logger.info('🧿 Starting the Envoy.') @@ -51,7 +53,8 @@ def start_(shard_name, director_uri, tls, shard_config_path, shard_descriptor = shard_descriptor_from_config(shard_config_path) envoy = Envoy( shard_name=shard_name, - director_uri=director_uri, + director_host=director_host, + director_port=director_port, shard_descriptor=shard_descriptor, tls=tls, root_certificate=root_certificate, diff --git a/openfl/interface/interactive_api/federation.py b/openfl/interface/interactive_api/federation.py index 420e696496..0026300412 100644 --- a/openfl/interface/interactive_api/federation.py +++ b/openfl/interface/interactive_api/federation.py @@ -44,8 +44,15 @@ def __init__(self, client_id, director_node_fqdn=None, director_port=None, tls=T self.api_private_key = api_private_key # Create Director client - self.dir_client = DirectorClient(client_id, f'{director_node_fqdn}:{director_port}', - tls, cert_chain, api_private_key, api_cert) + self.dir_client = DirectorClient( + client_id=client_id, + director_host=director_node_fqdn, + director_port=director_port, + tls=tls, + root_certificate=cert_chain, + private_key=api_private_key, + certificate=api_cert + ) # Request sample and target shapes from Director. # This is an internal method for finding out dataset properties in a Federation. diff --git a/openfl/transport/grpc/director_client.py b/openfl/transport/grpc/director_client.py index a21485d29b..be0ccfebcb 100644 --- a/openfl/transport/grpc/director_client.py +++ b/openfl/transport/grpc/director_client.py @@ -25,13 +25,14 @@ class ShardDirectorClient: """The internal director client class.""" - def __init__(self, director_uri, shard_name, tls=True, + def __init__(self, , director_host, director_port, shard_name, tls=True, root_certificate=None, private_key=None, certificate=None) -> None: """Initialize a shard director client object.""" self.shard_name = shard_name + director_addr = f'{director_host}:{director_port}' options = [('grpc.max_message_length', 100 * 1024 * 1024)] if not tls: - channel = grpc.insecure_channel(director_uri, options=options) + channel = grpc.insecure_channel(director_addr, options=options) else: if not (root_certificate and private_key and certificate): raise Exception('No certificates provided') @@ -50,7 +51,7 @@ def __init__(self, director_uri, shard_name, tls=True, private_key=private_key_b, certificate_chain=certificate_b ) - channel = grpc.secure_channel(director_uri, credentials, options=options) + channel = grpc.secure_channel(director_addr, credentials, options=options) self.stub = director_pb2_grpc.FederationDirectorStub(channel) def report_shard_info(self, shard_descriptor) -> bool: @@ -154,13 +155,14 @@ def send_health_check(self, collaborator_name, is_experiment_running, valid_dura class DirectorClient: """Director client class for users.""" - def __init__(self, client_id, director_uri, tls=True, + def __init__(self, , client_id, director_host, director_port, tls=True, root_certificate=None, private_key=None, certificate=None) -> None: """Initialize director client object.""" + director_addr = f'{director_host}:{director_port}' channel_opt = [('grpc.max_send_message_length', 512 1024 * 1024), ('grpc.max_receive_message_length', 512 * 1024 * 1024)] if not tls: - channel = grpc.insecure_channel(director_uri, options=channel_opt) + channel = grpc.insecure_channel(director_addr, options=channel_opt) else: if not (root_certificate and private_key and certificate): raise Exception('No certificates provided') @@ -180,7 +182,7 @@ def __init__(self, client_id, director_uri, tls=True, certificate_chain=certificate_b ) - channel = grpc.secure_channel(director_uri, credentials, options=channel_opt) + channel = grpc.secure_channel(director_addr, credentials, options=channel_opt) self.stub = director_pb2_grpc.FederationDirectorStub(channel) self.client_id = client_id diff --git a/openfl/transport/grpc/director_server.py b/openfl/transport/grpc/director_server.py index 8e96d22aad..3fc955d312 100644 --- a/openfl/transport/grpc/director_server.py +++ b/openfl/transport/grpc/director_server.py @@ -25,12 +25,12 @@ class DirectorGRPCServer(director_pb2_grpc.FederationDirectorServicer): def __init__(self, , director_cls, tls: bool = True, root_certificate: str = None, private_key: str = None, certificate: str = None, - listen_ip='[::]', listen_port=50051, kwargs) -> None: + listen_addr='[::]', listen_port=50051, *kwargs) -> None: """Initialize a director object.""" # TODO: add working directory super().__init__() - self.listen_addr = f'{listen_ip}:{listen_port}' + self.listen_uri = f'{listen_addr}:{listen_port}' self.tls = tls self.root_certificate = None self.private_key = None @@ -86,7 +86,7 @@ async def _run(self): director_pb2_grpc.add_FederationDirectorServicer_to_server(self, self.server) if not self.tls: - self.server.add_insecure_port(self.listen_addr) + self.server.add_insecure_port(self.listen_uri) else: with open(self.private_key, 'rb') as f: private_key_b = f.read() @@ -99,8 +99,8 @@ async def _run(self): root_certificates=root_certificate_b, require_client_auth=True ) - self.server.add_secure_port(self.listen_addr, server_credentials) - logger.info(f'Starting server on {self.listen_addr}') + self.server.add_secure_port(self.listen_uri, server_credentials) + logger.info(f'Starting server on {self.listen_uri}') await self.server.start() await self.server.wait_for_termination()	Need to have a consistent command for passing director's uri As of now, the envoy command asks for director's URI whereas the director for IP and port. It would be great if we could pass in the director's URI at both nodes.		2021-08-13T16:02:34	[]	[]
securefederatedai/openfl	securefederatedai__openfl-104	bd73b749a9ea1b92dbcdd07e639752101d769fc0	diff --git a/.gitignore b/.gitignore index 4d19e080ce..7364deced3 100644 --- a/.gitignore +++ b/.gitignore @@ -1,4 +1,5 @@ .egg-info +.pkl __pycache__ /build /dist @@ -6,3 +7,7 @@ __pycache__ .ipynb_checkpoints venv/* .idea + +.jpg +.crt +.key \ No newline at end of file diff --git a/docs/advanced_topics.rst b/docs/advanced_topics.rst index bb8cd1e953..dafa8b5bc0 100644 --- a/docs/advanced_topics.rst +++ b/docs/advanced_topics.rst @@ -14,5 +14,6 @@ Advanced Topics compression_settings overriding_agg_fn bash_autocomplete_activation + log_metric_callback diff --git a/docs/log_metric_callback.rst b/docs/log_metric_callback.rst new file mode 100644 index 0000000000..46255f6387 --- /dev/null +++ b/docs/log_metric_callback.rst @@ -0,0 +1,69 @@ +.. # Copyright (C) 2020-2021 Intel Corporation +.. # SPDX-License-Identifier: Apache-2.0 + +.. _log_metric_callback: +=============================== +Metric logging callback +=============================== + +------------------------------- +Usage +------------------------------- +\|productName\| allows developers to use custom metric logging functions. This function will call on aggregator node. +In order to define such function, you should: + +Python API +========== +Define function with follow signature: + +.. code-block:: python + + def callback_name(node_name, task_name, metric_name, metric, round_number): + """ + Write metric callback + + Args: + node_name (str): Name of node, which generate metric + task_name (str): Name of task + metric_name (str): Name of metric + metric (np.ndarray): Metric value + round_number (int): Round number + """ + your code +CLI +==== + +Define callback function similar way like in python api in ``src`` folder of your workspace. And provide a way to your function in ``aggregator`` part of ``plan/plan.yaml`` file in your workspace, use ``log_metric_callback`` key: + +.. code-block:: yaml + + aggregator : + defaults : plan/defaults/aggregator.yaml + template : openfl.component.Aggregator + settings : + init_state_path : save/torch_cnn_mnist_init.pbuf + best_state_path : save/torch_cnn_mnist_best.pbuf + last_state_path : save/torch_cnn_mnist_last.pbuf + rounds_to_train : 10 + log_metric_callback : + template : src.mnist_utils.callback_name + + + +Example +======================= + +Below is an example of a log metric callback, which writes metric values to tensorboard + +.. code-block:: python + + from torch.utils.tensorboard import SummaryWriter + + writer = SummaryWriter('./logs/cnn_mnist', flush_secs=5) + + + def write_metric(node_name, task_name, metric_name, metric, round_number): + writer.add_scalar("{}/{}/{}".format(node_name, task_name, metric_name), + metric, round_number) + +Full implementation can be found in ``openfl-tutorials/Federated_Pytorch_MNIST_Tutorial.ipynb`` and in ``torch_cnn_mnist`` workspace \ No newline at end of file diff --git a/docs/mermaid/pki_scheme.mmd b/docs/mermaid/pki_scheme.mmd new file mode 100644 index 0000000000..9e5b69d6ae --- /dev/null +++ b/docs/mermaid/pki_scheme.mmd @@ -0,0 +1,20 @@ +sequenceDiagram +Title: Collaborator Certificate Signing Flow + participant A as Aggregator + participant CA as CA + participant C as Collaborator + CA->>CA: 1. Create CA:<br>`step ca init --password-file pass_file` + CA->>CA: 2. Up HTTPS CA server:<br>`step_ca ca_config.json` + CA->>CA: 3. Generate JWK pair:<br>`step crypto jwk create pub.json priv.json --password-file pass_file` + CA->>CA: 4. Get JWT for aggregator:<br>`step ca token localhost --key priv.json --password-file pass_file --ca-url ca_url` + CA->>A: 5. Copy JWT to aggregator. + A->>CA: 6. Certify node:<br>`step ca certificate localhost agg.crt agg.key --token AbC1d2E..` + Note over A,CA: Get agg.crt + CA->>CA: 7. Get JWT for collaborator:<br>`step ca token col_name --key priv.json --password-file pass_file --ca-url ca_url` + CA->>C: 8. Copy JWT to collaborator. + C->>CA: 9. Certify node:<br>`step ca certificate col_name col_name.crt col_name.key --token AbC1d2E..` + Note over C,CA: Get col_name.crt + CA->>A: 10. Copy root_ca.crt to aggregator + Note over A,CA: This could be done at step 5 with token + CA->>C: 11. Copy root_ca.crt to collaborator + Note over C,CA: This could be done at step 8 with token \ No newline at end of file diff --git a/docs/running_the_federation.certificates.rst b/docs/running_the_federation.certificates.rst index 3443454cae..4ea49d207b 100644 --- a/docs/running_the_federation.certificates.rst +++ b/docs/running_the_federation.certificates.rst @@ -11,7 +11,8 @@ Therefore, security keys and certificates will need to be created for the aggregator and collaborators to negotiate the connection securely. For the :ref:`Hello Federation <running_the_federation>` demo we will run the aggregator and collaborators on the same localhost server -so these configuration steps just need to be done once on that machine. +so these configuration steps just need to be done once on that machine. We have two pki +workflows: manual and semi-automatic (with step-ca). .. note:: @@ -20,10 +21,16 @@ so these configuration steps just need to be done once on that machine. .. _install_certs: .. kroki:: mermaid/CSR_signing.mmd - :caption: Certificate generation and signing + :caption: Manual certificate generation and signing :align: center :type: mermaid +.. _install_certs: + +.. kroki:: mermaid/pki_scheme.mmd + :caption: Step-ca certificate generation and signing + :align: center + :type: mermaid .. _install_certs_agg: diff --git a/docs/running_the_federation.interactive_api.rst b/docs/running_the_federation.interactive_api.rst index 824caa3f12..6acb9e5eb2 100644 --- a/docs/running_the_federation.interactive_api.rst +++ b/docs/running_the_federation.interactive_api.rst @@ -23,6 +23,58 @@ Python Environment =================== Create a virtual Python environment. Please, install only packages that are required for conducting the experiment, since Python environment will be replicated on collaborator nodes. +*************************************** +Certification +************************************** +If you have trusted workspace and connection should not be encrypted you can use :code:`disable_tls` option while starting expirement. +Otherwise it is necessary to certify each node participating in the federation. Certificates allow to use mutual tls connection between nodes. +You can certify nodes by your own pki system or use pki provided by OpenFL. It is based on `step-ca <https://github.com/smallstep/certificates>`_ +as a server and `step <https://github.com/smallstep/cli>`_ as a client utilities. They are downloaded from github during workspace setup. Regardless of the certification method, +paths to certificates on each node are provided at start of experiment. Pki workflow from OpenFL will be discussed below. + +OpenFL PKI workflow +=================== +Openfl PKI pipeline asumes creating local CA with https server which listen signing requests. +Certificates from each node can be signed by requesting to CA server with special token. +Token must be copied to each node by some secure way. Each step is considered in detail below. + +1. Create CA, i.e create root key pair, CA server config and other. + .. code-block:: console + + $ fx pki install -p </path/to/ca/dir> --password <123> --ca-url <host:port> + \| :code:`-p` - path to folder, which will contain ca files. + \| :code:`--password` - password that will encrypts some ca files. + \| :code:`--ca-url` - host and port which ca server will listen + This command will also download `step-ca <https://github.com/smallstep/certificates>`_ and `step <https://github.com/smallstep/cli>`_ binaries from github. + +2. Run CA https server. + .. code-block:: console + + $ fx pki run -p </path/to/ca/dir> + \| :code:`-p` - path to folder, which will contain ca files. + +3. Get token for some node. + + .. code-block:: console + + $ fx pki get-token -n <subject> + \| :code:`-n` - subject name, fqdn for director, collaborator name for envoy or api name for api-layer node + + Run this command on ca side, from ca folder. Output is a token which contains JWT (json web token) from CA server and CA + root certificate concatenated together. This JWT have twenty-four hours time-to-live. + +4. Copy token to node side (director or envoy) by some secure channel and run certify command. + .. code-block:: console + + $ fx pki certify -n <subject> -t <token> + \| :code:`-n` - subject name, fqdn for director, collaborator name for envoy or api name for api-layer node + \| :code:`-t` - output token from previous command + This command call step client, to connect to CA server over https. + Https is provided by root certificate which was copy with JWT. + Server authenticates client by JWT and client authenticates server by root certificate. + +Now signed certificate and private key are stored on current node. Signed certificate has one year time-to-live. You should certify all node that will participate in federation: director, all envoys and api-layer node. + ************************************** Defining a Federated Learning Experiment **************************************** diff --git a/openfl-tutorials/Federated_FedProx_PyTorch_MNIST_Tutorial.ipynb b/openfl-tutorials/Federated_FedProx_PyTorch_MNIST_Tutorial.ipynb index 9419056c12..12936d98e0 100644 --- a/openfl-tutorials/Federated_FedProx_PyTorch_MNIST_Tutorial.ipynb +++ b/openfl-tutorials/Federated_FedProx_PyTorch_MNIST_Tutorial.ipynb @@ -40,7 +40,7 @@ "cell_type": "markdown", "metadata": {}, "source": [ - "After importing the required packages, the next step is setting up our openfl workspace. To do this, simply run the `fx.init()` command as follows:" + "After importing the required packages, the next step is setting Up our openfl workspace. To do this, simply run the `fx.init()` command as follows:" ] }, { diff --git a/openfl-tutorials/Federated_PyTorch_UNET_Tutorial.ipynb b/openfl-tutorials/Federated_PyTorch_UNET_Tutorial.ipynb index a7f116f9ae..b6cc5f81ac 100644 --- a/openfl-tutorials/Federated_PyTorch_UNET_Tutorial.ipynb +++ b/openfl-tutorials/Federated_PyTorch_UNET_Tutorial.ipynb @@ -135,8 +135,11 @@ " def __init__(self, data_path, collaborator_count, collaborator_num, is_validation):\n", " self.images_path = './data/segmented-images/images/'\n", " self.masks_path = './data/segmented-images/masks/'\n", - " self.images_names = [img_name for img_name in sorted(listdir(\n", - " self.images_path)) if len(img_name) > 3 and img_name[-3:] == 'jpg']\n", + " self.images_names = [\n", + " img_name\n", + " for img_name in sorted(listdir(self.images_path))\n", + " if len(img_name) > 3 and img_name[-3:] == 'jpg'\n", + " ]\n", "\n", " self.images_names = self.images_names[collaborator_num:: collaborator_count]\n", " self.is_validation = is_validation\n", @@ -267,9 +270,9 @@ " return score.sum()\n", "\n", "\n", - "class double_conv(nn.Module):\n", + "class DoubleConv(nn.Module):\n", " def __init__(self, in_ch, out_ch):\n", - " super(double_conv, self).__init__()\n", + " super(DoubleConv, self).__init__()\n", " self.in_ch = in_ch\n", " self.out_ch = out_ch\n", " self.conv = nn.Sequential(\n", @@ -286,33 +289,36 @@ " return x\n", "\n", "\n", - "class down(nn.Module):\n", + "class Down(nn.Module):\n", " def __init__(self, in_ch, out_ch):\n", - " super(down, self).__init__()\n", + " super(Down, self).__init__()\n", " self.mpconv = nn.Sequential(\n", - " nn.MaxPool2d(2), double_conv(in_ch, out_ch))\n", + " nn.MaxPool2d(2),\n", + " DoubleConv(in_ch, out_ch)\n", + " )\n", "\n", " def forward(self, x):\n", " x = self.mpconv(x)\n", " return x\n", "\n", "\n", - "class up(nn.Module):\n", + "class Up(nn.Module):\n", " def __init__(self, in_ch, out_ch, bilinear=False):\n", - " super(up, self).__init__()\n", + " super(Up, self).__init__()\n", " self.in_ch = in_ch\n", " self.out_ch = out_ch\n", " if bilinear:\n", - " self.up = nn.Upsample(\n", - " scale_factor=2, mode=\"bilinear\", align_corners=True)\n", - " else:\n", - " self.up = nn.ConvTranspose2d(\n", - " in_ch, in_ch // 2, 2, stride=2\n", + " self.Up = nn.Upsample(\n", + " scale_factor=2,\n", + " mode=\"bilinear\",\n", + " align_corners=True\n", " )\n", - " self.conv = double_conv(in_ch, out_ch)\n", + " else:\n", + " self.Up = nn.ConvTranspose2d(in_ch, in_ch // 2, 2, stride=2)\n", + " self.conv = DoubleConv(in_ch, out_ch)\n", "\n", " def forward(self, x1, x2):\n", - " x1 = self.up(x1)\n", + " x1 = self.Up(x1)\n", " diffY = x2.size()[2] - x1.size()[2]\n", " diffX = x2.size()[3] - x1.size()[3]\n", "\n", @@ -327,15 +333,15 @@ "class UNet(nn.Module):\n", " def __init__(self, n_channels=3, n_classes=1):\n", " super().__init__()\n", - " self.inc = double_conv(n_channels, 64)\n", - " self.down1 = down(64, 128)\n", - " self.down2 = down(128, 256)\n", - " self.down3 = down(256, 512)\n", - " self.down4 = down(512, 1024)\n", - " self.up1 = up(1024, 512)\n", - " self.up2 = up(512, 256)\n", - " self.up3 = up(256, 128)\n", - " self.up4 = up(128, 64)\n", + " self.inc = DoubleConv(n_channels, 64)\n", + " self.down1 = Down(64, 128)\n", + " self.down2 = Down(128, 256)\n", + " self.down3 = Down(256, 512)\n", + " self.down4 = Down(512, 1024)\n", + " self.up1 = Up(1024, 512)\n", + " self.up2 = Up(512, 256)\n", + " self.up3 = Up(256, 128)\n", + " self.up4 = Up(128, 64)\n", " self.outc = nn.Conv2d(64, n_classes, 1)\n", "\n", " def forward(self, x):\n", @@ -558,4 +564,4 @@ }, "nbformat": 4, "nbformat_minor": 4 -} +} \ No newline at end of file diff --git a/openfl-tutorials/Federated_Pytorch_MNIST_Tutorial.ipynb b/openfl-tutorials/Federated_Pytorch_MNIST_Tutorial.ipynb index 52beb21ffb..c889b996f6 100644 --- a/openfl-tutorials/Federated_Pytorch_MNIST_Tutorial.ipynb +++ b/openfl-tutorials/Federated_Pytorch_MNIST_Tutorial.ipynb @@ -49,7 +49,7 @@ "outputs": [], "source": [ "#Setup default workspace, logging, etc.\n", - "fx.init('torch_cnn_mnist')" + "fx.init('torch_cnn_mnist', log_level='METRIC', log_file='./spam_metric.log')" ] }, { @@ -124,6 +124,29 @@ " return F.cross_entropy(input=output,target=target)" ] }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "Here we can define metric logging function. It should has the following signature described below. You can use it to write metrics to tensorboard or some another specific logging." + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "from torch.utils.tensorboard import SummaryWriter\n", + "\n", + "writer = SummaryWriter('./logs/cnn_mnist', flush_secs=5)\n", + "\n", + "\n", + "def write_metric(node_name, task_name, metric_name, metric, round_number):\n", + " writer.add_scalar(\"{}/{}/{}\".format(node_name, task_name, metric_name),\n", + " metric, round_number)" + ] + }, { "cell_type": "code", "execution_count": null, @@ -205,8 +228,12 @@ "metadata": {}, "outputs": [], "source": [ - "#Run experiment, return trained FederatedModel\n", - "final_fl_model = fx.run_experiment(collaborators,{'aggregator.settings.rounds_to_train':5})" + "# Run experiment, return trained FederatedModel\n", + "\n", + "final_fl_model = fx.run_experiment(collaborators, override_config={\n", + " 'aggregator.settings.rounds_to_train': 5,\n", + " 'aggregator.settings.log_metric_callback': write_metric,\n", + "})" ] }, { diff --git a/openfl-tutorials/interactive_api/Director_Pytorch_Kvasir_UNET/director_folder/director_config.yaml b/openfl-tutorials/interactive_api/Director_Pytorch_Kvasir_UNET/director_folder/director_config.yaml new file mode 100644 index 0000000000..9b45b60869 --- /dev/null +++ b/openfl-tutorials/interactive_api/Director_Pytorch_Kvasir_UNET/director_folder/director_config.yaml @@ -0,0 +1,4 @@ +settings: + listen_ip: localhost + sample_shape: ['300', '400', '3'] + target_shape: ['300', '400'] \ No newline at end of file diff --git a/openfl-tutorials/interactive_api/Director_Pytorch_Kvasir_UNET/director_folder/start_director.sh b/openfl-tutorials/interactive_api/Director_Pytorch_Kvasir_UNET/director_folder/start_director.sh new file mode 100644 index 0000000000..5806a6cc0a --- /dev/null +++ b/openfl-tutorials/interactive_api/Director_Pytorch_Kvasir_UNET/director_folder/start_director.sh @@ -0,0 +1,4 @@ +#!/bin/bash +set -e + +fx director start --disable-tls -c director_config.yaml \ No newline at end of file diff --git a/openfl-tutorials/interactive_api/Director_Pytorch_Kvasir_UNET/director_folder/start_director_with_tls.sh b/openfl-tutorials/interactive_api/Director_Pytorch_Kvasir_UNET/director_folder/start_director_with_tls.sh new file mode 100644 index 0000000000..5d6d46a792 --- /dev/null +++ b/openfl-tutorials/interactive_api/Director_Pytorch_Kvasir_UNET/director_folder/start_director_with_tls.sh @@ -0,0 +1,4 @@ +#!/bin/bash +set -e +FQDN=$1 +fx director start -c director_config.yaml -rc cert/root_ca.crt -pk cert/"${FQDN}".key -oc cert/"${FQDN}".crt \ No newline at end of file diff --git a/openfl-tutorials/interactive_api/Director_Pytorch_Kvasir_UNET/envoy_folder/kvasir_shard_descriptor.py b/openfl-tutorials/interactive_api/Director_Pytorch_Kvasir_UNET/envoy_folder/kvasir_shard_descriptor.py new file mode 100644 index 0000000000..69290c1956 --- /dev/null +++ b/openfl-tutorials/interactive_api/Director_Pytorch_Kvasir_UNET/envoy_folder/kvasir_shard_descriptor.py @@ -0,0 +1,135 @@ +# Copyright (C) 2020-2021 Intel Corporation +# SPDX-License-Identifier: Apache-2.0 +"""Kvasir shard descriptor.""" + + +import os +from hashlib import sha384 +from pathlib import Path + +import numpy as np +from PIL import Image + +from openfl.interface.interactive_api.shard_descriptor import ShardDescriptor + + +class KvasirShardDescriptor(ShardDescriptor): + """Shard descriptor class.""" + + def __init__(self, data_folder: str = 'kvasir_data', + rank_worldsize: str = '1,1', + enforce_image_hw: str = None) -> None: + """Initialize KvasirShardDescriptor.""" + super().__init__() + + self.data_folder = Path.cwd() / data_folder + self.download_data(self.data_folder) + + # Settings for resizing data + self.enforce_image_hw = None + if enforce_image_hw is not None: + self.enforce_image_hw = tuple(int(size) for size in enforce_image_hw.split(',')) + # Settings for sharding the dataset + self.rank_worldsize = tuple(int(num) for num in rank_worldsize.split(',')) + + self.images_path = self.data_folder / 'segmented-images' / 'images' + self.masks_path = self.data_folder / 'segmented-images' / 'masks' + + self.images_names = [ + img_name + for img_name in sorted(os.listdir(self.images_path)) + if len(img_name) > 3 and img_name[-3:] == 'jpg' + ] + # Sharding + self.images_names = self.images_names[self.rank_worldsize[0] - 1::self.rank_worldsize[1]] + + # Calculating data and target shapes + sample, target = self[0] + self._sample_shape = [str(dim) for dim in sample.shape] + self._target_shape = [str(dim) for dim in target.shape] + + @staticmethod + def download_data(data_folder): + """Download data.""" + zip_file_path = data_folder / 'kvasir.zip' + os.makedirs(data_folder, exist_ok=True) + os.system('wget -nc' + " 'https://datasets.simula.no/hyper-kvasir/hyper-kvasir-segmented-images.zip'" + f' -O {zip_file_path.relative_to(Path.cwd())}') + zip_sha384 = 'e30d18a772c6520476e55b610a4db457237f151e' \ + '19182849d54b49ae24699881c1e18e0961f77642be900450ef8b22e7' + assert sha384(open(zip_file_path, 'rb').read( + os.path.getsize(zip_file_path))).hexdigest() == zip_sha384 + os.system(f'unzip -n {zip_file_path.relative_to(Path.cwd())}' + f' -d {data_folder.relative_to(Path.cwd())}') + + def __getitem__(self, index): + """Return a item by the index.""" + name = self.images_names[index] + # Reading data + img = Image.open(self.images_path / name) + mask = Image.open(self.masks_path / name) + if self.enforce_image_hw is not None: + # If we need to resize data + # PIL accepts (w,h) tuple, not (h,w) + img = img.resize(self.enforce_image_hw[::-1]) + mask = mask.resize(self.enforce_image_hw[::-1]) + img = np.asarray(img) + mask = np.asarray(mask) + assert img.shape[2] == 3 + + return img, mask[:, :, 0].astype(np.uint8) + + def __len__(self): + """Return the len of the dataset.""" + return len(self.images_names) + + @property + def sample_shape(self): + """Return the sample shape info.""" + return self._sample_shape + + @property + def target_shape(self): + """Return the target shape info.""" + return self._target_shape + + @property + def dataset_description(self) -> str: + """Return the dataset description.""" + return f'Kvasir dataset, shard number {self.rank_worldsize[0]}' \ + f' out of {self.rank_worldsize[1]}' + + +if __name__ == '__main__': + from openfl.interface.cli import setup_logging + setup_logging() + + data_folder = 'data' + rank_worldsize = '1,100' + enforce_image_hw = '529,622' + + kvasir_sd = KvasirShardDescriptor( + data_folder, + rank_worldsize=rank_worldsize, + enforce_image_hw=enforce_image_hw) + + print(kvasir_sd.dataset_description) + print(kvasir_sd.sample_shape, kvasir_sd.target_shape) + + from openfl.component.envoy.envoy import Envoy + + shard_name = 'one' + director_uri = 'localhost:50051' + + keeper = Envoy( + shard_name=shard_name, + director_uri=director_uri, + shard_descriptor=kvasir_sd, + disable_tls=False, + root_ca='./cert/root_ca.crt', + key='./cert/one.key', + cert='./cert/one.crt', + ) + + keeper.start() diff --git a/openfl-tutorials/interactive_api/Director_Pytorch_Kvasir_UNET/envoy_folder/sd_requirements.txt b/openfl-tutorials/interactive_api/Director_Pytorch_Kvasir_UNET/envoy_folder/sd_requirements.txt new file mode 100644 index 0000000000..92606011b1 --- /dev/null +++ b/openfl-tutorials/interactive_api/Director_Pytorch_Kvasir_UNET/envoy_folder/sd_requirements.txt @@ -0,0 +1,2 @@ +numpy +pillow \ No newline at end of file diff --git a/openfl-tutorials/interactive_api/Director_Pytorch_Kvasir_UNET/envoy_folder/shard_config.yaml b/openfl-tutorials/interactive_api/Director_Pytorch_Kvasir_UNET/envoy_folder/shard_config.yaml new file mode 100644 index 0000000000..aba8be3627 --- /dev/null +++ b/openfl-tutorials/interactive_api/Director_Pytorch_Kvasir_UNET/envoy_folder/shard_config.yaml @@ -0,0 +1,5 @@ +template: kvasir_shard_descriptor.KvasirShardDescriptor +params: + data_folder: kvasir_data + rank_worldsize: 1,90 + enforce_image_hw: '300,400' \ No newline at end of file diff --git a/openfl-tutorials/interactive_api/Director_Pytorch_Kvasir_UNET/envoy_folder/start_envoy.sh b/openfl-tutorials/interactive_api/Director_Pytorch_Kvasir_UNET/envoy_folder/start_envoy.sh new file mode 100644 index 0000000000..c546d40138 --- /dev/null +++ b/openfl-tutorials/interactive_api/Director_Pytorch_Kvasir_UNET/envoy_folder/start_envoy.sh @@ -0,0 +1,4 @@ +#!/bin/bash +set -e + +fx envoy start -n env_one --disable-tls --shard-config-path shard_config.yaml -d localhost:50051 \ No newline at end of file diff --git a/openfl-tutorials/interactive_api/Director_Pytorch_Kvasir_UNET/envoy_folder/start_envoy_with_tls.sh b/openfl-tutorials/interactive_api/Director_Pytorch_Kvasir_UNET/envoy_folder/start_envoy_with_tls.sh new file mode 100644 index 0000000000..295f61b101 --- /dev/null +++ b/openfl-tutorials/interactive_api/Director_Pytorch_Kvasir_UNET/envoy_folder/start_envoy_with_tls.sh @@ -0,0 +1,6 @@ +#!/bin/bash +set -e +ENVOY_NAME=$1 +DIRECTOR_FQDN=$2 + +fx envoy start -n "$ENVOY_NAME" --shard-config-path shard_config.yaml -d "$DIRECTOR_FQDN":50051 -rc cert/root_ca.crt -pk cert/"$ENVOY_NAME".key -oc cert/"$ENVOY_NAME".crt \ No newline at end of file diff --git a/openfl-tutorials/interactive_api_tutorials_(experimental)/Pytorch_Kvasir_UNET_workspace/new_python_api_UNET.ipynb b/openfl-tutorials/interactive_api/Director_Pytorch_Kvasir_UNET/workspace/Updated_Kvasir_with_Director.ipynb similarity index 57% rename from openfl-tutorials/interactive_api_tutorials_(experimental)/Pytorch_Kvasir_UNET_workspace/new_python_api_UNET.ipynb rename to openfl-tutorials/interactive_api/Director_Pytorch_Kvasir_UNET/workspace/Updated_Kvasir_with_Director.ipynb index fc4ffe3734..f9081b07b8 100644 --- a/openfl-tutorials/interactive_api_tutorials_(experimental)/Pytorch_Kvasir_UNET_workspace/new_python_api_UNET.ipynb +++ b/openfl-tutorials/interactive_api/Director_Pytorch_Kvasir_UNET/workspace/Updated_Kvasir_with_Director.ipynb @@ -5,10 +5,37 @@ "id": "liquid-jacket", "metadata": {}, "source": [ - "# Federated PyTorch UNET Tutorial\n", + "# Federated Kvasir with Director example\n", "## Using low-level Python API" ] }, + { + "cell_type": "markdown", + "id": "af0579f8", + "metadata": {}, + "source": [ + "# Long-Living entities update\n", + "\n", + "* We now may have director running on another machine.\n", + "* We use Federation API to communicate with Director.\n", + "* Federation object should hold a Director's client (for user service)\n", + "* Keeping in mind that several API instances may be connacted to one Director.\n", + "\n", + "\n", + "* We do not think for now how we start a Director.\n", + "* But it knows the data shape and target shape for the DataScience problem in the Federation.\n", + "* Director holds the list of connected envoys, we do not need to specify it anymore.\n", + "* Director and Envoys are responsible for encrypting connections, we do not need to worry about certs.\n", + "\n", + "\n", + "* Yet we MUST have a cert to communicate to the Director.\n", + "* We MUST know the FQDN of a Director.\n", + "* Director communicates data and target shape to the Federation interface object.\n", + "\n", + "\n", + "* Experiment API may use this info to construct a dummy dataset and a `shard descriptor` stub." + ] + }, { "cell_type": "code", "execution_count": null, @@ -19,180 +46,141 @@ "outputs": [], "source": [ "# Install dependencies if not already installed\n", - "!pip install torchvision==0.8.1\n", - "!pip install scikit-image" + "!pip install torchvision==0.8.1" ] }, { "cell_type": "markdown", - "id": "caring-distinction", + "id": "16986f22", "metadata": {}, "source": [ - "### Describe the model and optimizer" + "# Connect to the Federation" ] }, { "cell_type": "code", "execution_count": null, - "id": "visible-victor", + "id": "4485ac79", "metadata": {}, "outputs": [], "source": [ - "import torch\n", - "import torch.nn as nn\n", - "import torch.optim as optim" + "# Create a federation\n", + "from openfl.interface.interactive_api.federation import Federation\n", + "\n", + "# please use the same identificator that was used in signed certificate\n", + "cliend_id = 'frontend'\n", + "\n", + "# 1) Run with API layer - Director mTLS \n", + "# If the user wants to enable mTLS their must provide CA root chain, and signed key pair to the federation interface\n", + "# cert_chain = 'cert/root_ca.crt'\n", + "# API_certificate = 'cert/frontend.crt'\n", + "# API_private_key = 'cert/frontend.key'\n", + "\n", + "# federation = Federation(client_id='frontend', director_node_fqdn='localhost', director_port='50051', disable_tls=False,\n", + "# cert_chain=cert_chain, api_cert=API_certificate, api_private_key=API_private_key)\n", + "\n", + "# --------------------------------------------------------------------------------------------------------------------\n", + "\n", + "# 2) Run with TLS disabled (trusted environment)\n", + "# Federation can also determine local fqdn automatically\n", + "federation = Federation(client_id='frontend', director_node_fqdn='localhost', director_port='50051', disable_tls=True)\n" ] }, { "cell_type": "code", "execution_count": null, - "id": "foreign-gospel", - "metadata": {}, + "id": "e35802d5", + "metadata": { + "scrolled": true + }, "outputs": [], "source": [ - "\"\"\"\n", - "UNet model definition\n", - "\"\"\"\n", - "from layers import soft_dice_coef, soft_dice_loss, DoubleConv, Down, Up\n", - "\n", - "class UNet(nn.Module):\n", - " def __init__(self, n_channels=3, n_classes=1):\n", - " super().__init__()\n", - " self.inc = DoubleConv(n_channels, 64)\n", - " self.down1 = Down(64, 128)\n", - " self.down2 = Down(128, 256)\n", - " self.down3 = Down(256, 512)\n", - " self.down4 = Down(512, 1024)\n", - " self.up1 = Up(1024, 512)\n", - " self.up2 = Up(512, 256)\n", - " self.up3 = Up(256, 128)\n", - " self.up4 = Up(128, 64)\n", - " self.outc = nn.Conv2d(64, n_classes, 1)\n", - "\n", - " def forward(self, x):\n", - " x1 = self.inc(x)\n", - " x2 = self.down1(x1)\n", - " x3 = self.down2(x2)\n", - " x4 = self.down3(x3)\n", - " x5 = self.down4(x4)\n", - " x = self.up1(x5, x4)\n", - " x = self.up2(x, x3)\n", - " x = self.up3(x, x2)\n", - " x = self.up4(x, x1)\n", - " x = self.outc(x)\n", - " x = torch.sigmoid(x)\n", - " return x\n", - " \n", - " \n", - "model_unet = UNet()" + "shard_registry = federation.get_shard_registry()\n", + "shard_registry" ] }, { "cell_type": "code", "execution_count": null, - "id": "greater-activation", + "id": "67ae50de", "metadata": {}, "outputs": [], "source": [ - "optimizer_adam = optim.Adam(model_unet.parameters(), lr=1e-4)" + "federation.target_shape" ] }, { - "cell_type": "markdown", - "id": "insured-metallic", - "metadata": {}, + "cell_type": "code", + "execution_count": null, + "id": "920216d3", + "metadata": { + "scrolled": true + }, + "outputs": [], "source": [ - "### Prepare data" + "# First, request a dummy_shard_desc that holds information about the federated dataset \n", + "dummy_shard_desc = federation.get_dummy_shard_descriptor(size=10)\n", + "sample, target = dummy_shard_desc[0]" ] }, { "cell_type": "markdown", - "id": "essential-remainder", + "id": "obvious-tyler", "metadata": {}, "source": [ - "We ask user to keep all the test data in `data/` folder under the workspace as it will not be sent to collaborators" + "## Creating a FL experiment using Interactive API" ] }, { "cell_type": "code", "execution_count": null, - "id": "blind-circuit", + "id": "rubber-address", "metadata": {}, "outputs": [], "source": [ - "import os\n", - "from hashlib import sha384\n", - "import PIL\n", - "from torch.utils.data import Dataset, DataLoader\n", - "from torchvision import transforms as tsf\n", - "from skimage import io" + "from openfl.interface.interactive_api.experiment import TaskInterface, DataInterface, ModelInterface, FLExperiment" ] }, { - "cell_type": "code", - "execution_count": null, - "id": "recorded-input", - "metadata": { - "scrolled": true - }, - "outputs": [], + "cell_type": "markdown", + "id": "sustainable-public", + "metadata": {}, "source": [ - "os.makedirs('data', exist_ok=True)\n", - "!wget -nc 'https://datasets.simula.no/hyper-kvasir/hyper-kvasir-segmented-images.zip' -O ./data/kvasir.zip\n", - "ZIP_SHA384 = 'e30d18a772c6520476e55b610a4db457237f151e'\\\n", - " '19182849d54b49ae24699881c1e18e0961f77642be900450ef8b22e7'\n", - "assert sha384(open('./data/kvasir.zip', 'rb').read(\n", - " os.path.getsize('./data/kvasir.zip'))).hexdigest() == ZIP_SHA384\n", - "!unzip -n ./data/kvasir.zip -d ./data" + "### Register dataset" + ] + }, + { + "cell_type": "markdown", + "id": "unlike-texas", + "metadata": {}, + "source": [ + "We extract User dataset class implementation.\n", + "Is it convinient?\n", + "What if the dataset is not a class?" ] }, { "cell_type": "code", "execution_count": null, - "id": "prepared-israel", - "metadata": {}, + "id": "64f37dcf", + "metadata": { + "scrolled": true + }, "outputs": [], "source": [ - "DATA_PATH = './data/segmented-images/'\n", + "import os\n", + "import PIL\n", "import numpy as np\n", + "from torch.utils.data import Dataset, DataLoader, SubsetRandomSampler\n", + "from torchvision import transforms as tsf\n", "\n", - "def read_data(image_path, mask_path):\n", - " \"\"\"\n", - " Read image and mask from disk.\n", - " \"\"\"\n", - " img = io.imread(image_path)\n", - " assert(img.shape[2] == 3)\n", - " mask = io.imread(mask_path)\n", - " return (img, mask[:, :, 0].astype(np.uint8))\n", - "\n", - "\n", - "class KvasirDataset(Dataset):\n", - " \"\"\"\n", - " Kvasir dataset contains 1000 images for all collaborators.\n", - " Args:\n", - " data_path: path to dataset on disk\n", - " collaborator_count: total number of collaborators\n", - " collaborator_num: number of current collaborator\n", - " is_validation: validation option\n", - " \"\"\"\n", - "\n", - " def __init__(self, images_path = './data/segmented-images/images/', \\\n", - " masks_path = './data/segmented-images/masks/',\n", - " validation_fraction=1/8, is_validation=False):\n", - "\n", - " self.images_path = images_path\n", - " self.masks_path = masks_path\n", - " self.images_names = [img_name for img_name in sorted(os.listdir(\n", - " self.images_path)) if len(img_name) > 3 and img_name[-3:] == 'jpg']\n", - "\n", - " assert(len(self.images_names) > 2), \"Too few images\"\n", - " \n", - " validation_size = max(1, int(len(self.images_names) * validation_fraction))\n", + "# Now you can implement you data loaders using dummy_shard_desc\n", + "class KvasirSD(DataInterface, Dataset):\n", + "\n", + " def __init__(self, validation_fraction=1/8, kwargs):\n", + " super().__init__(kwargs)\n", " \n", - " if is_validation:\n", - " self.images_names = self.images_names[-validation_size :]\n", - " else:\n", - " self.images_names = self.images_names[: -validation_size]\n", + " self.validation_fraction = validation_fraction\n", " \n", " # Prepare transforms\n", " self.img_trans = tsf.Compose([\n", @@ -205,239 +193,206 @@ " tsf.Resize((332, 332), interpolation=PIL.Image.NEAREST),\n", " tsf.ToTensor()])\n", " \n", + " @property\n", + " def shard_descriptor(self):\n", + " return self._shard_descriptor\n", + " \n", + " @shard_descriptor.setter\n", + " def shard_descriptor(self, shard_descriptor):\n", + " \"\"\"\n", + " Describe per-collaborator procedures or sharding.\n", + "\n", + " This method will be called during a collaborator initialization.\n", + " Local shard_descriptor will be set by Envoy.\n", + " \"\"\"\n", + " self._shard_descriptor = shard_descriptor\n", + " \n", + " validation_size = max(1, int(len(self.shard_descriptor) * self.validation_fraction))\n", + " \n", + " self.train_indeces = np.arange(len(self.shard_descriptor) - validation_size)\n", + " self.val_indeces = np.arange(len(self.shard_descriptor) - validation_size, len(self.shard_descriptor))\n", + " \n", "\n", " def __getitem__(self, index):\n", - " name = self.images_names[index]\n", - " img, mask = read_data(self.images_path + name, self.masks_path + name)\n", + " img, mask = self.shard_descriptor[index]\n", " img = self.img_trans(img).numpy()\n", " mask = self.mask_trans(mask).numpy()\n", " return img, mask\n", "\n", " def __len__(self):\n", - " return len(self.images_names)" - ] - }, - { - "cell_type": "markdown", - "id": "portuguese-groove", - "metadata": {}, - "source": [ - "### Define Federated Learning tasks" - ] - }, - { - "cell_type": "code", - "execution_count": null, - "id": "enclosed-abuse", - "metadata": {}, - "outputs": [], - "source": [ - "def train(unet_model, train_loader, optimizer, device, loss_fn=soft_dice_loss):\n", + " return len(self.shard_descriptor)\n", " \n", - " function_defined_in_notebook()\n", " \n", - " unet_model.train()\n", - " unet_model.to(device)\n", - "\n", - " losses = []\n", - "\n", - " for data, target in train_loader:\n", - " data, target = torch.tensor(data).to(device), torch.tensor(\n", - " target).to(device, dtype=torch.float32)\n", - " optimizer.zero_grad()\n", - " output = unet_model(data)\n", - " loss = loss_fn(output=output, target=target)\n", - " loss.backward()\n", - " optimizer.step()\n", - " losses.append(loss.detach().cpu().numpy())\n", - " \n", - " return {'train_loss': np.mean(losses),}\n", - "\n", + " def get_train_loader(self, kwargs):\n", + " \"\"\"\n", + " Output of this method will be provided to tasks with optimizer in contract\n", + " \"\"\"\n", + " train_sampler = SubsetRandomSampler(self.train_indeces)\n", + " return DataLoader(\n", + " self, num_workers=8, batch_size=self.kwargs['train_bs'], sampler=train_sampler\n", + " )\n", "\n", - "def validate(unet_model, val_loader, device):\n", - " unet_model.eval()\n", - " unet_model.to(device)\n", + " def get_valid_loader(self, kwargs):\n", + " \"\"\"\n", + " Output of this method will be provided to tasks without optimizer in contract\n", + " \"\"\"\n", + " val_sampler = SubsetRandomSampler(self.val_indeces)\n", + " return DataLoader(self, num_workers=8, batch_size=self.kwargs['valid_bs'], sampler=val_sampler)\n", "\n", - " val_score = 0\n", - " total_samples = 0\n", + " def get_train_data_size(self):\n", + " \"\"\"\n", + " Information for aggregation\n", + " \"\"\"\n", + " return len(self.train_indeces)\n", "\n", - " with torch.no_grad():\n", - " for data, target in val_loader:\n", - " samples = target.shape[0]\n", - " total_samples += samples\n", - " data, target = torch.tensor(data).to(device), \\\n", - " torch.tensor(target).to(device, dtype=torch.int64)\n", - " output = unet_model(data)\n", - " val = soft_dice_coef(output, target)\n", - " val_score += val.sum().cpu().numpy()\n", - " \n", - " return {'dice_coef': val_score / total_samples,}" + " def get_valid_data_size(self):\n", + " \"\"\"\n", + " Information for aggregation\n", + " \"\"\"\n", + " return len(self.val_indeces)" ] }, { "cell_type": "code", "execution_count": null, - "id": "secure-protest", + "id": "d8df35f5", "metadata": {}, "outputs": [], - "source": [] + "source": [ + "fed_dataset = KvasirSD(train_bs=4, valid_bs=8)\n", + "fed_dataset.shard_descriptor = dummy_shard_desc\n", + "for i, (sample, target) in enumerate(fed_dataset.get_train_loader()):\n", + " print(sample.shape)" + ] }, { "cell_type": "markdown", - "id": "obvious-tyler", + "id": "caring-distinction", "metadata": {}, "source": [ - "## Describing FL experiment" + "### Describe a model and optimizer" ] }, { "cell_type": "code", "execution_count": null, - "id": "rubber-address", + "id": "visible-victor", "metadata": {}, "outputs": [], "source": [ - "from openfl.interface.interactive_api.experiment import TaskInterface, DataInterface, ModelInterface, FLExperiment" - ] - }, - { - "cell_type": "markdown", - "id": "caroline-passion", - "metadata": {}, - "source": [ - "### Register model" + "import torch\n", + "import torch.nn as nn\n", + "import torch.optim as optim" ] }, { "cell_type": "code", "execution_count": null, - "id": "handled-teens", + "id": "foreign-gospel", "metadata": {}, "outputs": [], "source": [ - "from copy import deepcopy\n", + "\"\"\"\n", + "UNet model definition\n", + "\"\"\"\n", + "from layers import soft_dice_coef, soft_dice_loss, DoubleConv, Down, Up\n", "\n", - "framework_adapter = 'openfl.plugins.frameworks_adapters.pytorch_adapter.FrameworkAdapterPlugin'\n", - "MI = ModelInterface(model=model_unet, optimizer=optimizer_adam, framework_plugin=framework_adapter)\n", + "class UNet(nn.Module):\n", + " def __init__(self, n_channels=3, n_classes=1):\n", + " super().__init__()\n", + " self.inc = DoubleConv(n_channels, 64)\n", + " self.down1 = Down(64, 128)\n", + " self.down2 = Down(128, 256)\n", + " self.down3 = Down(256, 512)\n", + " self.down4 = Down(512, 1024)\n", + " self.up1 = Up(1024, 512)\n", + " self.up2 = Up(512, 256)\n", + " self.up3 = Up(256, 128)\n", + " self.up4 = Up(128, 64)\n", + " self.outc = nn.Conv2d(64, n_classes, 1)\n", "\n", - "# Save the initial model state\n", - "initial_model = deepcopy(model_unet)" + " def forward(self, x):\n", + " x1 = self.inc(x)\n", + " x2 = self.down1(x1)\n", + " x3 = self.down2(x2)\n", + " x4 = self.down3(x3)\n", + " x5 = self.down4(x4)\n", + " x = self.up1(x5, x4)\n", + " x = self.up2(x, x3)\n", + " x = self.up3(x, x2)\n", + " x = self.up4(x, x1)\n", + " x = self.outc(x)\n", + " x = torch.sigmoid(x)\n", + " return x\n", + " \n", + " \n", + "class UNet(nn.Module):\n", + " def __init__(self, n_channels=3, n_classes=1):\n", + " super().__init__()\n", + " self.inc = DoubleConv(n_channels, 64)\n", + " self.down1 = Down(64, 128)\n", + " self.down2 = Down(128, 256)\n", + " self.down3 = Down(256, 512)\n", + " self.up2 = Up(512, 256)\n", + " self.up3 = Up(256, 128)\n", + " self.up4 = Up(128, 64)\n", + " self.outc = nn.Conv2d(64, n_classes, 1)\n", + "\n", + " def forward(self, x):\n", + " x1 = self.inc(x)\n", + " x2 = self.down1(x1)\n", + " x3 = self.down2(x2)\n", + " x4 = self.down3(x3)\n", + " x = self.up2(x4, x3)\n", + " x = self.up3(x, x2)\n", + " x = self.up4(x, x1)\n", + " x = self.outc(x)\n", + " x = torch.sigmoid(x)\n", + " return x\n", + " \n", + "model_unet = UNet()" ] }, { - "cell_type": "markdown", - "id": "sustainable-public", + "cell_type": "code", + "execution_count": null, + "id": "greater-activation", "metadata": {}, + "outputs": [], "source": [ - "### Register dataset" + "optimizer_adam = optim.Adam(model_unet.parameters(), lr=1e-4)" ] }, { "cell_type": "markdown", - "id": "unlike-texas", + "id": "caroline-passion", "metadata": {}, "source": [ - "We extract User dataset class implementation.\n", - "Is it convinient?\n", - "What if the dataset is not a class?" + "#### Register model" ] }, { "cell_type": "code", "execution_count": null, - "id": "anticipated-coverage", - "metadata": {}, - "outputs": [], - "source": [ - "class UserDataset:\n", - " def __init__(self, path_to_local_data):\n", - " print(f'User Dataset initialized with {path_to_local_data}')\n", - " \n", - " \n", - "class OpenflMixin: \n", - " def _delayed_init(self):\n", - " raise NotImplementedError\n", - " \n", - " \n", - "class FedDataset(OpenflMixin):\n", - " def __init__(self, UserDataset):\n", - " self.user_dataset_class = UserDataset\n", - " print('We implement all abstract methods from mixin in this class')\n", - " \n", - " def _delayed_init(self, data_path):\n", - " print('This method is called on the collaborator node')\n", - " dataset_obj = self.user_dataset_class(data_path)\n", - " \n", - " \n", - "fed_dataset = FedDataset(UserDataset)\n", - "fed_dataset._delayed_init('data path on the collaborator node')" - ] - }, - { - "cell_type": "code", - "execution_count": null, - "id": "verbal-newton", + "id": "handled-teens", "metadata": {}, "outputs": [], "source": [ - "class FedDataset(DataInterface):\n", - " def __init__(self, UserDatasetClass, kwargs):\n", - " self.UserDatasetClass = UserDatasetClass\n", - " self.kwargs = kwargs\n", - " \n", - " def _delayed_init(self, data_path='1,1'):\n", - " # With the next command the local dataset will be loaded on the collaborator node\n", - " # For this example we have the same dataset on the same path, and we will shard it\n", - " # So we use `data_path` information for this purpose.\n", - " self.rank, self.world_size = [int(part) for part in data_path.split(',')]\n", - " \n", - " validation_fraction=1/8\n", - " self.train_set = self.UserDatasetClass(validation_fraction=validation_fraction, is_validation=False)\n", - " self.valid_set = self.UserDatasetClass(validation_fraction=validation_fraction, is_validation=True)\n", - " \n", - " # Do the actual sharding\n", - " self._do_sharding( self.rank, self.world_size)\n", - " \n", - " def _do_sharding(self, rank, world_size):\n", - " # This method relies on the dataset's implementation\n", - " # i.e. coupled in a bad way\n", - " self.train_set.images_names = self.train_set.images_names[ rank-1 :: world_size ]\n", - "\n", - " def get_train_loader(self, kwargs):\n", - " \"\"\"\n", - " Output of this method will be provided to tasks with optimizer in contract\n", - " \"\"\"\n", - " return DataLoader(\n", - " self.train_set, num_workers=8, batch_size=self.kwargs['train_bs'], shuffle=True\n", - " )\n", - "\n", - " def get_valid_loader(self, *kwargs):\n", - " \"\"\"\n", - " Output of this method will be provided to tasks without optimizer in contract\n", - " \"\"\"\n", - " return DataLoader(self.valid_set, num_workers=8, batch_size=self.kwargs['valid_bs'])\n", + "from copy import deepcopy\n", "\n", - " def get_train_data_size(self):\n", - " \"\"\"\n", - " Information for aggregation\n", - " \"\"\"\n", - " return len(self.train_set)\n", + "framework_adapter = 'openfl.plugins.frameworks_adapters.pytorch_adapter.FrameworkAdapterPlugin'\n", + "MI = ModelInterface(model=model_unet, optimizer=optimizer_adam, framework_plugin=framework_adapter)\n", "\n", - " def get_valid_data_size(self):\n", - " \"\"\"\n", - " Information for aggregation\n", - " \"\"\"\n", - " return len(self.valid_set)\n", - " \n", - "fed_dataset = FedDataset(KvasirDataset, train_bs=8, valid_bs=8)" + "# Save the initial model state\n", + "initial_model = deepcopy(model_unet)" ] }, { "cell_type": "markdown", - "id": "marine-kansas", + "id": "portuguese-groove", "metadata": {}, "source": [ - "### Register tasks" + "### Define and register FL tasks" ] }, { @@ -463,6 +418,8 @@ "def train(unet_model, train_loader, optimizer, device, loss_fn=soft_dice_loss, some_parameter=None):\n", " if not torch.cuda.is_available():\n", " device = 'cpu'\n", + " else:\n", + " device = 'cuda'\n", " \n", " function_defined_in_notebook(some_parameter)\n", " \n", @@ -488,6 +445,11 @@ "\n", "@TI.register_fl_task(model='unet_model', data_loader='val_loader', device='device') \n", "def validate(unet_model, val_loader, device):\n", + " if not torch.cuda.is_available():\n", + " device = 'cpu'\n", + " else:\n", + " device = 'cuda'\n", + " \n", " unet_model.eval()\n", " unet_model.to(device)\n", " \n", @@ -517,58 +479,6 @@ "## Time to start a federated learning experiment" ] }, - { - "cell_type": "code", - "execution_count": null, - "id": "silver-baltimore", - "metadata": {}, - "outputs": [], - "source": [ - "# Create a federation\n", - "from openfl.interface.interactive_api.federation import Federation\n", - "\n", - "# 1) Run with aggregator-collaborator mTLS \n", - "# If the user wants to enable mTLS their must provide CA root chain, and signed key pair to the federation interface\n", - "cert_chain = 'cert/cert_chain.crt'\n", - "agg_certificate = 'cert/agg_certificate.crt'\n", - "agg_private_key = 'cert/agg_private.key'\n", - "\n", - "federation = Federation(central_node_fqdn='some.fqdn', disable_tls=False,\n", - " cert_chain=cert_chain, agg_certificate=agg_certificate, agg_private_key=agg_private_key)\n", - "col_data_paths = {'one': '1,1',}\n", - "federation.register_collaborators(col_data_paths=col_data_paths)\n", - "\n", - "# --------------------------------------------------------------------------------------------------------------------\n", - "\n", - "# 2) Run with TLS disabled (trusted environment)\n", - "# Federation can also determine local fqdn automatically\n", - "federation = Federation(central_node_fqdn='localhost', disable_tls=True)\n", - "# First number which is a collaborators rank may also be passed as a cuda device identifier\n", - "col_data_paths = {'one': '1,2',\n", - " 'two': '2,2'}\n", - "federation.register_collaborators(col_data_paths=col_data_paths)" - ] - }, - { - "cell_type": "markdown", - "id": "adad7744", - "metadata": {}, - "source": [ - "#### Certification of an aggregator\n", - " fx workspace certify: creates cert folder and CA as well as cert_chain\n", - "* fx aggregator generate-cert-request --fqdn `FQDN`: you can pass a specific aggregator FQDN if you want\n", - "* fx aggregator certify --fqdn `FQDN` --silent: signes aggregators cert\n", - "<br> After that just pass the paths to required certs to the Federation API\n", - "\n", - "#### Certification of a collaborator\n", - "just follow the usual procedure: <br>\n", - "fx collaborator generate-cert-request -d {DATA_PATH} -n {COL} \n", - "\n", - "fx collaborator certify --request-pkg {COL_DIRECTORY}/{FED_WORKSPACE}/col_{COL}_to_agg_cert_request.zip\n", - "\n", - "fx collaborator certify --import {FED_DIRECTORY}/agg_to_col_{COL}_signed_cert.zip" - ] - }, { "cell_type": "code", "execution_count": null, @@ -577,7 +487,8 @@ "outputs": [], "source": [ "# create an experimnet in federation\n", - "fl_experiment = FLExperiment(federation=federation,)" + "experiment_name = 'kvasir_test_experiment'\n", + "fl_experiment = FLExperiment(federation=federation, experiment_name=experiment_name)" ] }, { @@ -585,23 +496,31 @@ "execution_count": null, "id": "lightweight-causing", "metadata": { - "scrolled": false + "scrolled": true }, "outputs": [], "source": [ "# If I use autoreload I got a pickling error\n", "\n", "# The following command zips the workspace and python requirements to be transfered to collaborator nodes\n", - "fl_experiment.prepare_workspace_distribution(model_provider=MI, task_keeper=TI, data_loader=fed_dataset, rounds_to_train=7, \\\n", - " opt_treatment='CONTINUE_GLOBAL')\n", - "# This command starts the aggregator server\n", - "fl_experiment.start_experiment(model_provider=MI)\n", + "fl_experiment.start(model_provider=MI, \n", + " task_keeper=TI,\n", + " data_loader=fed_dataset,\n", + " rounds_to_train=2,\n", + " opt_treatment='CONTINUE_GLOBAL')\n" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "id": "f1543a36", + "metadata": {}, + "outputs": [], + "source": [ + "# If user want to stop IPython session, then reconnect and check how experiment is going \n", + "# fl_experiment.restore_experiment_state(MI)\n", "\n", - "# When the aggregator server blocks the notebook one can start collaborators\n", - "# For the test run just type console command from the workspace directory:\n", - "# `fx collaborator start -d data.yaml -n {col_name}` for all collaborators\n", - "# For the distributed experiment transfer zipped workspace to the collaborator nodes and run\n", - "# `fx workspace import --archive {workspace_name}.zip` cd to the workspace and start collaborators" + "fl_experiment.stream_metrics()" ] }, { @@ -616,7 +535,9 @@ "cell_type": "code", "execution_count": null, "id": "55acff59", - "metadata": {}, + "metadata": { + "scrolled": true + }, "outputs": [], "source": [ "best_model = fl_experiment.get_best_model()" @@ -625,11 +546,23 @@ { "cell_type": "code", "execution_count": null, - "id": "9fb5718d", + "id": "9479fb7f", "metadata": {}, "outputs": [], "source": [ - "fed_dataset._delayed_init()" + "# We remove exremove_experiment_datamove_experiment_datamove_experiment_datariment data from director\n", + "fl_experiment.remove_experiment_data()" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "id": "75c8aeab", + "metadata": {}, + "outputs": [], + "source": [ + "best_model.inc.conv[0].weight\n", + "# model_unet.inc.conv[0].weight" ] }, { @@ -685,6 +618,70 @@ "# Validating trained model\n", "validate(best_model, fed_dataset.get_valid_loader(), 'cpu')" ] + }, + { + "cell_type": "code", + "execution_count": null, + "id": "2dcdbfeb", + "metadata": {}, + "outputs": [], + "source": [ + "a = (np.zeros((2,4)), np.ones(2,), 2np.ones(2,))\n", + "a = [elem for elem in a if elem.shape==2 else elem.newaxis()]\n", + "np.concatenate(a)" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "id": "70863215", + "metadata": {}, + "outputs": [], + "source": [ + "class A:\n", + " def __init__(self, kwargs):\n", + " print(\"Class A\", kwargs)\n", + " \n", + "class B:\n", + " def __init__(self, kwargs):\n", + " super().__init__()\n", + " print(\"Class B\", kwargs)\n", + " \n", + "class C(B, A):\n", + " def __init__(self, kwargs):\n", + " super().__init__()\n", + " print(\"Class C\", kwargs)\n", + " \n", + "# class A:\n", + "# def __init__(self, kwargs):\n", + "# print(\"Class A\", kwargs)" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "id": "4ba44919", + "metadata": {}, + "outputs": [], + "source": [ + "c = C(x=1, z=5)" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "id": "0637f26f", + "metadata": {}, + "outputs": [], + "source": [] + }, + { + "cell_type": "code", + "execution_count": null, + "id": "e00ff26c", + "metadata": {}, + "outputs": [], + "source": [] } ], "metadata": { @@ -708,4 +705,4 @@ }, "nbformat": 4, "nbformat_minor": 5 -} +} \ No newline at end of file diff --git a/openfl-tutorials/interactive_api/Director_Pytorch_Kvasir_UNET/workspace/layers.py b/openfl-tutorials/interactive_api/Director_Pytorch_Kvasir_UNET/workspace/layers.py new file mode 100644 index 0000000000..5165dcc97e --- /dev/null +++ b/openfl-tutorials/interactive_api/Director_Pytorch_Kvasir_UNET/workspace/layers.py @@ -0,0 +1,100 @@ +"""Layers for Unet model.""" + +import torch +import torch.nn as nn +import torch.nn.functional as F + + +def soft_dice_loss(output, target): + """Calculate loss.""" + num = target.size(0) + m1 = output.view(num, -1) + m2 = target.view(num, -1) + intersection = m1 m2 + score = 2.0 * (intersection.sum(1) + 1) / (m1.sum(1) + m2.sum(1) + 1) + score = 1 - score.sum() / num + return score + + +def soft_dice_coef(output, target): + """Calculate soft DICE coefficient.""" + num = target.size(0) + m1 = output.view(num, -1) + m2 = target.view(num, -1) + intersection = m1 * m2 + score = 2.0 * (intersection.sum(1) + 1) / (m1.sum(1) + m2.sum(1) + 1) + return score.sum() + + +class DoubleConv(nn.Module): + """Pytorch double conv class.""" + + def __init__(self, in_ch, out_ch): + """Initialize layer.""" + super(DoubleConv, self).__init__() + self.in_ch = in_ch + self.out_ch = out_ch + self.conv = nn.Sequential( + nn.Conv2d(in_ch, out_ch, 3, padding=1), + nn.BatchNorm2d(out_ch), + nn.ReLU(inplace=True), + nn.Conv2d(out_ch, out_ch, 3, padding=1), + nn.BatchNorm2d(out_ch), + nn.ReLU(inplace=True), + ) + + def forward(self, x): + """Do forward pass.""" + x = self.conv(x) + return x + + +class Down(nn.Module): + """Pytorch nn module subclass.""" + + def __init__(self, in_ch, out_ch): + """Initialize layer.""" + super(Down, self).__init__() + self.mpconv = nn.Sequential( + nn.MaxPool2d(2), + DoubleConv(in_ch, out_ch) + ) + + def forward(self, x): + """Do forward pass.""" + x = self.mpconv(x) + return x + + +class Up(nn.Module): + """Pytorch nn module subclass.""" + + def __init__(self, in_ch, out_ch, bilinear=False): + """Initialize layer.""" + super(Up, self).__init__() + self.in_ch = in_ch + self.out_ch = out_ch + if bilinear: + self.up = nn.Upsample( + scale_factor=2, + mode='bilinear', + align_corners=True + ) + else: + self.up = nn.ConvTranspose2d(in_ch, in_ch // 2, 2, stride=2) + self.conv = DoubleConv(in_ch, out_ch) + + def forward(self, x1, x2): + """Do forward pass.""" + x1 = self.up(x1) + diff_y = x2.size()[2] - x1.size()[2] + diff_x = x2.size()[3] - x1.size()[3] + + x1 = F.pad( + x1, + (diff_x // 2, diff_x - diff_x // 2, diff_y // 2, diff_y - diff_y // 2) + ) + + x = torch.cat([x2, x1], dim=1) + x = self.conv(x) + return x diff --git a/openfl-tutorials/interactive_api/Director_Pytorch_Kvasir_UNET/workspace/new_python_api_UNET.ipynb b/openfl-tutorials/interactive_api/Director_Pytorch_Kvasir_UNET/workspace/new_python_api_UNET.ipynb new file mode 100644 index 0000000000..b72244349f --- /dev/null +++ b/openfl-tutorials/interactive_api/Director_Pytorch_Kvasir_UNET/workspace/new_python_api_UNET.ipynb @@ -0,0 +1,960 @@ +{ + "cells": [ + { + "cell_type": "markdown", + "id": "liquid-jacket", + "metadata": {}, + "source": [ + "# Federated PyTorch UNET Tutorial\n", + "## Using low-level Python API" + ] + }, + { + "cell_type": "code", + "execution_count": 1, + "id": "alike-sharing", + "metadata": { + "scrolled": true + }, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "Requirement already satisfied: torchvision in /home/davidyuk/.virtualenvs/openfl/lib/python3.8/site-packages (0.9.1)\n", + "Requirement already satisfied: torch==1.8.1 in /home/davidyuk/.virtualenvs/openfl/lib/python3.8/site-packages (from torchvision) (1.8.1)\n", + "Requirement already satisfied: numpy in /home/davidyuk/.virtualenvs/openfl/lib/python3.8/site-packages (from torchvision) (1.20.3)\n", + "Requirement already satisfied: pillow>=4.1.1 in /home/davidyuk/.virtualenvs/openfl/lib/python3.8/site-packages (from torchvision) (8.2.0)\n", + "Requirement already satisfied: typing-extensions in /home/davidyuk/.virtualenvs/openfl/lib/python3.8/site-packages (from torch==1.8.1->torchvision) (3.10.0.0)\n", + "\u001B[33mWARNING: You are using pip version 21.1; however, version 21.1.2 is available.\n", + "You should consider upgrading via the '/home/davidyuk/.virtualenvs/openfl/bin/python -m pip install --upgrade pip' command.\u001B[0m\n" + ] + } + ], + "source": [ + "# Install dependencies if not already installed\n", + "!pip install torchvision" + ] + }, + { + "cell_type": "markdown", + "id": "caring-distinction", + "metadata": {}, + "source": [ + "### Describe the model and optimizer" + ] + }, + { + "cell_type": "code", + "execution_count": 2, + "id": "visible-victor", + "metadata": {}, + "outputs": [], + "source": [ + "import torch\n", + "import torch.nn as nn\n", + "import torch.optim as optim" + ] + }, + { + "cell_type": "code", + "execution_count": 3, + "id": "foreign-gospel", + "metadata": {}, + "outputs": [], + "source": [ + "\"\"\"\n", + "UNet model definition\n", + "\"\"\"\n", + "from layers import soft_dice_coef, soft_dice_loss, DoubleConv, Down, Up\n", + "\n", + "class UNet(nn.Module):\n", + " def __init__(self, n_channels=3, n_classes=1):\n", + " super().__init__()\n", + " self.inc = DoubleConv(n_channels, 64)\n", + " self.down1 = Down(64, 128)\n", + " self.down2 = Down(128, 256)\n", + " self.down3 = Down(256, 512)\n", + " self.down4 = Down(512, 1024)\n", + " self.up1 = Up(1024, 512)\n", + " self.up2 = Up(512, 256)\n", + " self.up3 = Up(256, 128)\n", + " self.up4 = Up(128, 64)\n", + " self.outc = nn.Conv2d(64, n_classes, 1)\n", + "\n", + " def forward(self, x):\n", + " x1 = self.inc(x)\n", + " x2 = self.down1(x1)\n", + " x3 = self.down2(x2)\n", + " x4 = self.down3(x3)\n", + " x5 = self.down4(x4)\n", + " x = self.up1(x5, x4)\n", + " x = self.up2(x, x3)\n", + " x = self.up3(x, x2)\n", + " x = self.up4(x, x1)\n", + " x = self.outc(x)\n", + " x = torch.sigmoid(x)\n", + " return x\n", + " \n", + " \n", + "model_unet = UNet()" + ] + }, + { + "cell_type": "code", + "execution_count": 4, + "id": "24e5cf06", + "metadata": { + "scrolled": true + }, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "class UNet(nn.Module):\n", + " def __init__(self, n_channels=3, n_classes=1):\n", + " super().__init__()\n", + " self.inc = DoubleConv(n_channels, 64)\n", + " self.down1 = Down(64, 128)\n", + " self.down2 = Down(128, 256)\n", + " self.down3 = Down(256, 512)\n", + " self.down4 = Down(512, 1024)\n", + " self.up1 = Up(1024, 512)\n", + " self.up2 = Up(512, 256)\n", + " self.up3 = Up(256, 128)\n", + " self.up4 = Up(128, 64)\n", + " self.outc = nn.Conv2d(64, n_classes, 1)\n", + "\n", + " def forward(self, x):\n", + " x1 = self.inc(x)\n", + " x2 = self.down1(x1)\n", + " x3 = self.down2(x2)\n", + " x4 = self.down3(x3)\n", + " x5 = self.down4(x4)\n", + " x = self.up1(x5, x4)\n", + " x = self.up2(x, x3)\n", + " x = self.up3(x, x2)\n", + " x = self.up4(x, x1)\n", + " x = self.outc(x)\n", + " x = torch.sigmoid(x)\n", + " return x\n", + "\n" + ] + } + ], + "source": [ + "import inspect\n", + "from IPython.core.magics.code import extract_symbols\n", + "\n", + "import sys\n", + "\n", + "def new_getfile(object, _old_getfile=inspect.getfile):\n", + " if not inspect.isclass(object):\n", + " return _old_getfile(object)\n", + " \n", + " # Lookup by parent module (as in current inspect)\n", + " if hasattr(object, '__module__'):\n", + " object_ = sys.modules.get(object.__module__)\n", + " if hasattr(object_, '__file__'):\n", + " return object_.__file__\n", + " \n", + " # If parent module is __main__, lookup by methods (NEW)\n", + " for name, member in inspect.getmembers(object):\n", + " if inspect.isfunction(member) and object.__qualname__ + '.' + member.__name__ == member.__qualname__:\n", + " return inspect.getfile(member)\n", + " else:\n", + " raise TypeError('Source for {!r} not found'.format(object))\n", + "inspect.getfile = new_getfile\n", + "\n", + "obj = UNet\n", + "cell_code = \"\".join(inspect.linecache.getlines(new_getfile(obj)))\n", + "class_code = extract_symbols(cell_code, obj.__name__)[0][0]\n", + "print(class_code)" + ] + }, + { + "cell_type": "code", + "execution_count": 5, + "id": "greater-activation", + "metadata": {}, + "outputs": [], + "source": [ + "optimizer_adam = optim.Adam(model_unet.parameters(), lr=1e-4)" + ] + }, + { + "cell_type": "markdown", + "id": "obvious-tyler", + "metadata": {}, + "source": [ + "## Describing FL experiment" + ] + }, + { + "cell_type": "code", + "execution_count": 6, + "id": "rubber-address", + "metadata": {}, + "outputs": [], + "source": [ + "from openfl.interface.interactive_api.experiment import TaskInterface, DataInterface, ModelInterface, FLExperiment" + ] + }, + { + "cell_type": "markdown", + "id": "caroline-passion", + "metadata": {}, + "source": [ + "### Register model" + ] + }, + { + "cell_type": "code", + "execution_count": 7, + "id": "handled-teens", + "metadata": {}, + "outputs": [], + "source": [ + "from copy import deepcopy\n", + "\n", + "framework_adapter = 'openfl.plugins.frameworks_adapters.pytorch_adapter.FrameworkAdapterPlugin'\n", + "MI = ModelInterface(model=model_unet, optimizer=optimizer_adam, framework_plugin=framework_adapter)\n", + "\n", + "# Save the initial model state\n", + "initial_model = deepcopy(model_unet)" + ] + }, + { + "cell_type": "markdown", + "id": "portuguese-groove", + "metadata": {}, + "source": [ + "### Define Federated Learning tasks" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "id": "secure-protest", + "metadata": {}, + "outputs": [], + "source": [] + }, + { + "cell_type": "markdown", + "id": "sustainable-public", + "metadata": {}, + "source": [ + "### Register dataset" + ] + }, + { + "cell_type": "markdown", + "id": "unlike-texas", + "metadata": {}, + "source": [ + "We extract User dataset class implementation.\n", + "Is it convinient?\n", + "What if the dataset is not a class?" + ] + }, + { + "cell_type": "code", + "execution_count": 8, + "id": "verbal-newton", + "metadata": {}, + "outputs": [], + "source": [ + "# class FedDataset(DataInterface):\n", + "# def __init__(self, UserDatasetClass, kwargs):\n", + "# self.UserDatasetClass = UserDatasetClass\n", + "# self.kwargs = kwargs\n", + " \n", + "# def _delayed_init(self, data_path='1,1'):\n", + "# # With the next command the local dataset will be loaded on the collaborator node\n", + "# # For this example we have the same dataset on the same path, and we will shard it\n", + "# # So we use `data_path` information for this purpose.\n", + "# self.rank, self.world_size = [int(part) for part in data_path.split(',')]\n", + " \n", + "# validation_fraction=1/8\n", + "# self.train_set = self.UserDatasetClass(validation_fraction=validation_fraction, is_validation=False)\n", + "# self.valid_set = self.UserDatasetClass(validation_fraction=validation_fraction, is_validation=True)\n", + " \n", + "# # Do the actual sharding\n", + "# self._do_sharding( self.rank, self.world_size)\n", + " \n", + "# def _do_sharding(self, rank, world_size):\n", + "# # This method relies on the dataset's implementation\n", + "# # i.e. coupled in a bad way\n", + "# self.train_set.images_names = self.train_set.images_names[ rank-1 :: world_size ]\n", + "\n", + "# def get_train_loader(self, kwargs):\n", + "# \"\"\"\n", + "# Output of this method will be provided to tasks with optimizer in contract\n", + "# \"\"\"\n", + "# return DataLoader(\n", + "# self.train_set, num_workers=8, batch_size=self.kwargs['train_bs'], shuffle=True\n", + "# )\n", + "\n", + "# def get_valid_loader(self, kwargs):\n", + "# \"\"\"\n", + "# Output of this method will be provided to tasks without optimizer in contract\n", + "# \"\"\"\n", + "# return DataLoader(self.valid_set, num_workers=8, batch_size=self.kwargs['valid_bs'])\n", + "\n", + "# def get_train_data_size(self):\n", + "# \"\"\"\n", + "# Information for aggregation\n", + "# \"\"\"\n", + "# return len(self.train_set)\n", + "\n", + "# def get_valid_data_size(self):\n", + "# \"\"\"\n", + "# Information for aggregation\n", + "# \"\"\"\n", + "# return len(self.valid_set)\n", + " \n", + "# fed_dataset = FedDataset(KvasirDataset, train_bs=8, valid_bs=8)" + ] + }, + { + "cell_type": "markdown", + "id": "marine-kansas", + "metadata": {}, + "source": [ + "### Register tasks" + ] + }, + { + "cell_type": "code", + "execution_count": 9, + "id": "increasing-builder", + "metadata": {}, + "outputs": [], + "source": [ + "TI = TaskInterface()\n", + "import torch\n", + "\n", + "import tqdm\n", + "\n", + "# The Interactive API supports registering functions definied in main module or imported.\n", + "def function_defined_in_notebook(some_parameter):\n", + " print(f'Also I accept a parameter and it is {some_parameter}')\n", + "\n", + "# Task interface currently supports only standalone functions.\n", + "@TI.add_kwargs({'some_parameter': 42})\n", + "@TI.register_fl_task(model='unet_model', data_loader='train_loader', \\\n", + " device='device', optimizer='optimizer') \n", + "def train(unet_model, train_loader, optimizer, device, loss_fn=soft_dice_loss, some_parameter=None):\n", + " if not torch.cuda.is_available():\n", + " device = 'cpu'\n", + " \n", + " function_defined_in_notebook(some_parameter)\n", + " \n", + " train_loader = tqdm.tqdm(train_loader, desc=\"train\")\n", + " \n", + " unet_model.train()\n", + " unet_model.to(device)\n", + "\n", + " losses = []\n", + "\n", + " for data, target in train_loader:\n", + " data, target = torch.tensor(data).to(device), torch.tensor(\n", + " target).to(device, dtype=torch.float32)\n", + " optimizer.zero_grad()\n", + " output = unet_model(data)\n", + " loss = loss_fn(output=output, target=target)\n", + " loss.backward()\n", + " optimizer.step()\n", + " losses.append(loss.detach().cpu().numpy())\n", + " \n", + " return {'train_loss': np.mean(losses),}\n", + "\n", + "\n", + "@TI.register_fl_task(model='unet_model', data_loader='val_loader', device='device') \n", + "def validate(unet_model, val_loader, device):\n", + " unet_model.eval()\n", + " unet_model.to(device)\n", + " \n", + " val_loader = tqdm.tqdm(val_loader, desc=\"validate\")\n", + "\n", + " val_score = 0\n", + " total_samples = 0\n", + "\n", + " with torch.no_grad():\n", + " for data, target in val_loader:\n", + " samples = target.shape[0]\n", + " total_samples += samples\n", + " data, target = torch.tensor(data).to(device), \\\n", + " torch.tensor(target).to(device, dtype=torch.int64)\n", + " output = unet_model(data)\n", + " val = soft_dice_coef(output, target)\n", + " val_score += val.sum().cpu().numpy()\n", + " \n", + " return {'dice_coef': val_score / total_samples,}" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "id": "b50a39c7", + "metadata": {}, + "outputs": [], + "source": [] + }, + { + "cell_type": "markdown", + "id": "derived-bride", + "metadata": {}, + "source": [ + "## Time to start a federated learning experiment" + ] + }, + { + "cell_type": "code", + "execution_count": 10, + "id": "4485ac79", + "metadata": {}, + "outputs": [], + "source": [ + "# Create a federation\n", + "from openfl.interface.interactive_api.federation import Federation\n", + "\n", + "# 1) Run with API layer - Director mTLS \n", + "# If the user wants to enable mTLS their must provide CA root chain, and signed key pair to the federation interface\n", + "cert_chain = 'cert/cert_chain.crt'\n", + "API_certificate = 'cert/API_certificate.crt'\n", + "API_private_key = 'cert/API_private.key'\n", + "\n", + "# federation = Federation(director_node_fqdn='some.fqdn', disable_tls=False,\n", + "# cert_chain=cert_chain, API_cert=API_certificate, API_private_key=API_private_key)\n", + "\n", + "# --------------------------------------------------------------------------------------------------------------------\n", + "\n", + "# 2) Run with TLS disabled (trusted environment)\n", + "# Federation can also determine local fqdn automatically\n", + "federation = Federation(director_node_fqdn='localhost:50051', disable_tls=True)" + ] + }, + { + "cell_type": "code", + "execution_count": 11, + "id": "e35802d5", + "metadata": {}, + "outputs": [ + { + "data": { + "text/plain": [ + "[node_info {\n", + " name: \"1,2\"\n", + "}\n", + "shard_description: \"Kvasir dataset, shard number 1 out of 2\"\n", + "n_samples: 500\n", + "sample_shape: \"529\"\n", + "sample_shape: \"622\"\n", + "sample_shape: \"3\"\n", + "target_shape: \"529\"\n", + "target_shape: \"622\"\n", + "]" + ] + }, + "execution_count": 11, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "shard_registry = federation.get_shard_registry()\n", + "shard_registry" + ] + }, + { + "cell_type": "code", + "execution_count": 12, + "id": "67ae50de", + "metadata": {}, + "outputs": [ + { + "data": { + "text/plain": [ + "['529', '622']" + ] + }, + "execution_count": 12, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "federation.target_shape" + ] + }, + { + "cell_type": "code", + "execution_count": 13, + "id": "920216d3", + "metadata": { + "scrolled": true + }, + "outputs": [], + "source": [ + "# First, request a dummy_shard_desc that holds information about the federated dataset \n", + "dummy_shard_desc = federation.get_dummy_shard_descriptor(size=10)\n", + "sample, target = dummy_shard_desc[0]" + ] + }, + { + "cell_type": "markdown", + "id": "insured-metallic", + "metadata": {}, + "source": [ + "### Prepare data" + ] + }, + { + "cell_type": "markdown", + "id": "essential-remainder", + "metadata": {}, + "source": [ + "We ask user to keep all the test data in `data/` folder under the workspace as it will not be sent to collaborators" + ] + }, + { + "cell_type": "code", + "execution_count": 14, + "id": "blind-circuit", + "metadata": {}, + "outputs": [], + "source": [ + "import os\n", + "import PIL\n", + "import numpy as np\n", + "from torch.utils.data import Dataset, DataLoader, SubsetRandomSampler\n", + "from torchvision import transforms as tsf" + ] + }, + { + "cell_type": "code", + "execution_count": 15, + "id": "64f37dcf", + "metadata": { + "scrolled": true + }, + "outputs": [], + "source": [ + "# Now you can implement you data loaders using dummy_shard_desc\n", + "class KvasirSD(DataInterface, Dataset):\n", + "\n", + " def __init__(self, validation_fraction=1/8, kwargs):\n", + " super().__init__(kwargs)\n", + " \n", + " self.validation_fraction = validation_fraction\n", + " \n", + " # Prepare transforms\n", + " self.img_trans = tsf.Compose([\n", + " tsf.ToPILImage(),\n", + " tsf.Resize((332, 332)),\n", + " tsf.ToTensor(),\n", + " tsf.Normalize(mean=[0.5, 0.5, 0.5], std=[0.5, 0.5, 0.5])])\n", + " self.mask_trans = tsf.Compose([\n", + " tsf.ToPILImage(),\n", + " tsf.Resize((332, 332), interpolation=PIL.Image.NEAREST),\n", + " tsf.ToTensor()])\n", + " \n", + " @property\n", + " def shard_descriptor(self):\n", + " return self._shard_descriptor\n", + " \n", + " @shard_descriptor.setter\n", + " def shard_descriptor(self, shard_descriptor):\n", + " \"\"\"\n", + " Describe per-collaborator procedures or sharding.\n", + "\n", + " This method will be called during a collaborator initialization.\n", + " Local shard_descriptor will be set by Envoy.\n", + " \"\"\"\n", + " self._shard_descriptor = shard_descriptor\n", + " \n", + " validation_size = max(1, int(len(self.shard_descriptor) * self.validation_fraction))\n", + " \n", + " self.train_indeces = np.arange(len(self.shard_descriptor) - validation_size)\n", + " self.val_indeces = np.arange(len(self.shard_descriptor) - validation_size, len(self.shard_descriptor))\n", + " \n", + "\n", + " def __getitem__(self, index):\n", + " img, mask = self.shard_descriptor[index]\n", + " img = self.img_trans(img).numpy()\n", + " mask = self.mask_trans(mask).numpy()\n", + " return img, mask\n", + "\n", + " def __len__(self):\n", + " return len(self.shard_descriptor)\n", + " \n", + " \n", + " def get_train_loader(self, kwargs):\n", + " \"\"\"\n", + " Output of this method will be provided to tasks with optimizer in contract\n", + " \"\"\"\n", + " train_sampler = SubsetRandomSampler(self.train_indeces)\n", + " return DataLoader(\n", + " self, num_workers=8, batch_size=self.kwargs['train_bs'], sampler=train_sampler\n", + " )\n", + "\n", + " def get_valid_loader(self, kwargs):\n", + " \"\"\"\n", + " Output of this method will be provided to tasks without optimizer in contract\n", + " \"\"\"\n", + " val_sampler = SubsetRandomSampler(self.val_indeces)\n", + " return DataLoader(self, num_workers=8, batch_size=self.kwargs['valid_bs'], sampler=val_sampler)\n", + "\n", + " def get_train_data_size(self):\n", + " \"\"\"\n", + " Information for aggregation\n", + " \"\"\"\n", + " return len(self.train_indeces)\n", + "\n", + " def get_valid_data_size(self):\n", + " \"\"\"\n", + " Information for aggregation\n", + " \"\"\"\n", + " return len(self.val_indeces)" + ] + }, + { + "cell_type": "code", + "execution_count": 16, + "id": "d8df35f5", + "metadata": {}, + "outputs": [ + { + "name": "stderr", + "output_type": "stream", + "text": [ + "/home/davidyuk/.virtualenvs/openfl/lib/python3.8/site-packages/torchvision/transforms/transforms.py:257: UserWarning: Argument interpolation should be of type InterpolationMode instead of int. Please, use InterpolationMode enum.\n", + " warnings.warn(\n" + ] + }, + { + "name": "stdout", + "output_type": "stream", + "text": [ + "torch.Size([8, 3, 332, 332])\n", + "torch.Size([1, 3, 332, 332])\n" + ] + } + ], + "source": [ + "fed_dataset = KvasirSD(train_bs=8, valid_bs=8)\n", + "fed_dataset.shard_descriptor = dummy_shard_desc\n", + "for i, (sample, target) in enumerate(fed_dataset.get_train_loader()):\n", + " print(sample.shape)" + ] + }, + { + "cell_type": "markdown", + "id": "af0579f8", + "metadata": {}, + "source": [ + "# Long-Living entities update\n", + "\n", + "* We now may have director running on another machine.\n", + "* We use Federation API to communicate with Director.\n", + "* Federation object should hold a Director's client (for user service)\n", + "* Keeping in mind that several API instances may be connacted to one Director.\n", + "\n", + "\n", + "* We do not think for now how we start a Director.\n", + "* But it knows the data shape and target shape for the DataScience problem in the Federation.\n", + "* Director holds the list of connected envoys, we do not need to specify it anymore.\n", + "* Director and Envoys are responsible for encrypting connections, we do not need to worry about certs.\n", + "\n", + "\n", + "* Yet we MUST have a cert to communicate to the Director.\n", + "* We MUST know the FQDN of a Director.\n", + "* Director communicates data and target shape to the Federation interface object.\n", + "\n", + "\n", + "* Experiment API may use this info to construct a dummy dataset and a `shard descriptor` stub." + ] + }, + { + "cell_type": "markdown", + "id": "adad7744", + "metadata": {}, + "source": [ + "#### Certification of an aggregator\n", + "* fx workspace certify: creates cert folder and CA as well as cert_chain\n", + "* fx aggregator generate-cert-request --fqdn `FQDN`: you can pass a specific aggregator FQDN if you want\n", + "* fx aggregator certify --fqdn `FQDN` --silent: signes aggregators cert\n", + "<br> After that just pass the paths to required certs to the Federation API\n", + "\n", + "#### Certification of a collaborator\n", + "just follow the usual procedure: <br>\n", + "fx collaborator generate-cert-request -d {DATA_PATH} -n {COL} \n", + "\n", + "fx collaborator certify --request-pkg {COL_DIRECTORY}/{FED_WORKSPACE}/col_{COL}_to_agg_cert_request.zip\n", + "\n", + "fx collaborator certify --import {FED_DIRECTORY}/agg_to_col_{COL}_signed_cert.zip" + ] + }, + { + "cell_type": "code", + "execution_count": 17, + "id": "12f5663f", + "metadata": {}, + "outputs": [ + { + "data": { + "text/plain": [ + "2" + ] + }, + "execution_count": 17, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "# Interface generator\n", + "class IFO:\n", + " def __init__(self):\n", + " self.a = 1\n", + " \n", + "class EXP:\n", + " def __init__(self):\n", + " self.ifo = IFO()\n", + " \n", + " def get_ifo(self):\n", + " return self.ifo\n", + " \n", + "exp = EXP()\n", + "ifo = exp.get_ifo()\n", + "ifo.a = 2\n", + "\n", + "# print(exp.ifo.a)\n", + "exp2 = exp.__class__()\n", + "exp.ifo.a" + ] + }, + { + "cell_type": "code", + "execution_count": 18, + "id": "mature-renewal", + "metadata": {}, + "outputs": [], + "source": [ + "# create an experimnet in federation\n", + "fl_experiment = FLExperiment(federation=federation,)" + ] + }, + { + "cell_type": "code", + "execution_count": 19, + "id": "lightweight-causing", + "metadata": { + "scrolled": true + }, + "outputs": [ + { + "name": "stderr", + "output_type": "stream", + "text": [ + "tried to remove tensor: __opt_state_needed not present in the tensor dict\n", + "tried to remove tensor: __opt_state_needed not present in the tensor dict\n", + "INFO:openfl.transport.grpc.director_client:SetNewExperiment\n", + "WARNING:openfl.transport.grpc.server:gRPC is running on insecure channel with TLS disabled.\n", + "INFO:openfl.transport.grpc.server:Starting Aggregator gRPC Server\n", + "INFO:openfl.component.aggregator.aggregator:Sending tasks to collaborator 1,2 for round 0\n" + ] + } + ], + "source": [ + "# If I use autoreload I got a pickling error\n", + "\n", + "# The following command zips the workspace and python requirements to be transfered to collaborator nodes\n", + "fl_experiment.prepare_workspace_distribution(model_provider=MI, \n", + " task_keeper=TI, data_loader=fed_dataset, rounds_to_train=7, opt_treatment='CONTINUE_GLOBAL')\n", + "\n", + "\n", + "# This command starts the aggregator server\n", + "fl_experiment.start_experiment(model_provider=MI)\n" + ] + }, + { + "cell_type": "markdown", + "id": "8c30b301", + "metadata": {}, + "source": [ + "## Now we validate the best model!" + ] + }, + { + "cell_type": "code", + "execution_count": 20, + "id": "55acff59", + "metadata": {}, + "outputs": [ + { + "ename": "KeyError", + "evalue": "'inc.conv.0.weight'", + "output_type": "error", + "traceback": [ + "\u001B[0;31m---------------------------------------------------------------------------\u001B[0m", + "\u001B[0;31mKeyError\u001B[0m Traceback (most recent call last)", + "\u001B[0;32m<ipython-input-20-a539666cced0>\u001B[0m in \u001B[0;36m<module>\u001B[0;34m\u001B[0m\n\u001B[0;32m----> 1\u001B[0;31m \u001B[0mbest_model\u001B[0m \u001B[0;34m=\u001B[0m \u001B[0mfl_experiment\u001B[0m\u001B[0;34m.\u001B[0m\u001B[0mget_best_model\u001B[0m\u001B[0;34m(\u001B[0m\u001B[0;34m)\u001B[0m\u001B[0;34m\u001B[0m\u001B[0;34m\u001B[0m\u001B[0m\n\u001B[0m", + "\u001B[0;32m~/openfl/openfl/interface/interactive_api/experiment.py\u001B[0m in \u001B[0;36mget_best_model\u001B[0;34m(self)\u001B[0m\n\u001B[1;32m 41\u001B[0m \u001B[0;31m# Next line relies on aggregator inner field where model dicts are stored\u001B[0m\u001B[0;34m\u001B[0m\u001B[0;34m\u001B[0m\u001B[0;34m\u001B[0m\u001B[0m\n\u001B[1;32m 42\u001B[0m \u001B[0mbest_tensor_dict\u001B[0m \u001B[0;34m=\u001B[0m \u001B[0mself\u001B[0m\u001B[0;34m.\u001B[0m\u001B[0mserver\u001B[0m\u001B[0;34m.\u001B[0m\u001B[0maggregator\u001B[0m\u001B[0;34m.\u001B[0m\u001B[0mbest_tensor_dict\u001B[0m\u001B[0;34m\u001B[0m\u001B[0;34m\u001B[0m\u001B[0m\n\u001B[0;32m---> 43\u001B[0;31m \u001B[0mself\u001B[0m\u001B[0;34m.\u001B[0m\u001B[0mtask_runner_stub\u001B[0m\u001B[0;34m.\u001B[0m\u001B[0mrebuild_model\u001B[0m\u001B[0;34m(\u001B[0m\u001B[0mbest_tensor_dict\u001B[0m\u001B[0;34m,\u001B[0m \u001B[0mvalidation\u001B[0m\u001B[0;34m=\u001B[0m\u001B[0;32mTrue\u001B[0m\u001B[0;34m,\u001B[0m \u001B[0mdevice\u001B[0m\u001B[0;34m=\u001B[0m\u001B[0;34m'cpu'\u001B[0m\u001B[0;34m)\u001B[0m\u001B[0;34m\u001B[0m\u001B[0;34m\u001B[0m\u001B[0m\n\u001B[0m\u001B[1;32m 44\u001B[0m \u001B[0;32mreturn\u001B[0m \u001B[0mself\u001B[0m\u001B[0;34m.\u001B[0m\u001B[0mtask_runner_stub\u001B[0m\u001B[0;34m.\u001B[0m\u001B[0mmodel\u001B[0m\u001B[0;34m\u001B[0m\u001B[0;34m\u001B[0m\u001B[0m\n\u001B[1;32m 45\u001B[0m \u001B[0;34m\u001B[0m\u001B[0m\n", + "\u001B[0;32m~/openfl/openfl/federated/task/task_runner.py\u001B[0m in \u001B[0;36mrebuild_model\u001B[0;34m(self, input_tensor_dict, validation, device)\u001B[0m\n\u001B[1;32m 220\u001B[0m \u001B[0;32mif\u001B[0m \u001B[0mself\u001B[0m\u001B[0;34m.\u001B[0m\u001B[0mopt_treatment\u001B[0m \u001B[0;34m==\u001B[0m \u001B[0;34m'RESET'\u001B[0m\u001B[0;34m:\u001B[0m\u001B[0;34m\u001B[0m\u001B[0;34m\u001B[0m\u001B[0m\n\u001B[1;32m 221\u001B[0m \u001B[0mself\u001B[0m\u001B[0;34m.\u001B[0m\u001B[0mreset_opt_vars\u001B[0m\u001B[0;34m(\u001B[0m\u001B[0;34m)\u001B[0m\u001B[0;34m\u001B[0m\u001B[0;34m\u001B[0m\u001B[0m\n\u001B[0;32m--> 222\u001B[0;31m \u001B[0mself\u001B[0m\u001B[0;34m.\u001B[0m\u001B[0mset_tensor_dict\u001B[0m\u001B[0;34m(\u001B[0m\u001B[0minput_tensor_dict\u001B[0m\u001B[0;34m,\u001B[0m \u001B[0mwith_opt_vars\u001B[0m\u001B[0;34m=\u001B[0m\u001B[0;32mFalse\u001B[0m\u001B[0;34m,\u001B[0m \u001B[0mdevice\u001B[0m\u001B[0;34m=\u001B[0m\u001B[0mdevice\u001B[0m\u001B[0;34m)\u001B[0m\u001B[0;34m\u001B[0m\u001B[0;34m\u001B[0m\u001B[0m\n\u001B[0m\u001B[1;32m 223\u001B[0m elif (self.training_round_completed\n\u001B[1;32m 224\u001B[0m and self.opt_treatment == 'CONTINUE_GLOBAL' and not validation):\n", + "\u001B[0;32m~/openfl/openfl/federated/task/task_runner.py\u001B[0m in \u001B[0;36mset_tensor_dict\u001B[0;34m(self, tensor_dict, with_opt_vars, device)\u001B[0m\n\u001B[1;32m 373\u001B[0m \u001B[0mkwargs\u001B[0m \u001B[0;34m=\u001B[0m \u001B[0;34m{\u001B[0m\u001B[0;34m'device'\u001B[0m\u001B[0;34m:\u001B[0m \u001B[0mdevice\u001B[0m\u001B[0;34m,\u001B[0m \u001B[0;34m}\u001B[0m\u001B[0;34m\u001B[0m\u001B[0;34m\u001B[0m\u001B[0m\n\u001B[1;32m 374\u001B[0m \u001B[0;34m\u001B[0m\u001B[0m\n\u001B[0;32m--> 375\u001B[0;31m \u001B[0;32mreturn\u001B[0m \u001B[0mself\u001B[0m\u001B[0;34m.\u001B[0m\u001B[0mframework_adapter\u001B[0m\u001B[0;34m.\u001B[0m\u001B[0mset_tensor_dict\u001B[0m\u001B[0;34m(\u001B[0m\u001B[0;34m\u001B[0m\u001B[0margs\u001B[0m\u001B[0;34m,\u001B[0m \u001B[0;34m\u001B[0m\u001B[0mkwargs\u001B[0m\u001B[0;34m)\u001B[0m\u001B[0;34m\u001B[0m\u001B[0;34m\u001B[0m\u001B[0m\n\u001B[0m", + "\u001B[0;32m~/openfl/openfl/plugins/frameworks_adapters/pytorch_adapter.py\u001B[0m in \u001B[0;36mset_tensor_dict\u001B[0;34m(model, tensor_dict, optimizer, device)\u001B[0m\n\u001B[1;32m 43\u001B[0m \u001B[0;31m# everything\u001B[0m\u001B[0;34m\u001B[0m\u001B[0;34m\u001B[0m\u001B[0;34m\u001B[0m\u001B[0m\n\u001B[1;32m 44\u001B[0m \u001B[0;32mfor\u001B[0m \u001B[0mk\u001B[0m \u001B[0;32min\u001B[0m \u001B[0mmodel\u001B[0m\u001B[0;34m.\u001B[0m\u001B[0mstate_dict\u001B[0m\u001B[0;34m(\u001B[0m\u001B[0;34m)\u001B[0m\u001B[0;34m:\u001B[0m\u001B[0;34m\u001B[0m\u001B[0;34m\u001B[0m\u001B[0m\n\u001B[0;32m---> 45\u001B[0;31m \u001B[0mnew_state\u001B[0m\u001B[0;34m[\u001B[0m\u001B[0mk\u001B[0m\u001B[0;34m]\u001B[0m \u001B[0;34m=\u001B[0m \u001B[0mpt\u001B[0m\u001B[0;34m.\u001B[0m\u001B[0mfrom_numpy\u001B[0m\u001B[0;34m(\u001B[0m\u001B[0mtensor_dict\u001B[0m\u001B[0;34m.\u001B[0m\u001B[0mpop\u001B[0m\u001B[0;34m(\u001B[0m\u001B[0mk\u001B[0m\u001B[0;34m)\u001B[0m\u001B[0;34m)\u001B[0m\u001B[0;34m.\u001B[0m\u001B[0mto\u001B[0m\u001B[0;34m(\u001B[0m\u001B[0mdevice\u001B[0m\u001B[0;34m)\u001B[0m\u001B[0;34m\u001B[0m\u001B[0;34m\u001B[0m\u001B[0m\n\u001B[0m\u001B[1;32m 46\u001B[0m \u001B[0;34m\u001B[0m\u001B[0m\n\u001B[1;32m 47\u001B[0m \u001B[0;31m# set model state\u001B[0m\u001B[0;34m\u001B[0m\u001B[0;34m\u001B[0m\u001B[0;34m\u001B[0m\u001B[0m\n", + "\u001B[0;31mKeyError\u001B[0m: 'inc.conv.0.weight'" + ] + } + ], + "source": [ + "best_model = fl_experiment.get_best_model()" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "id": "9fb5718d", + "metadata": {}, + "outputs": [], + "source": [ + "fed_dataset._delayed_init()" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "id": "a2acb7e6", + "metadata": {}, + "outputs": [], + "source": [ + "# Validating initial model\n", + "validate(initial_model, fed_dataset.get_valid_loader(), 'cpu')" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "id": "c12ca93f", + "metadata": {}, + "outputs": [], + "source": [ + "# Validating trained model\n", + "validate(best_model, fed_dataset.get_valid_loader(), 'cpu')" + ] + }, + { + "cell_type": "markdown", + "id": "1e6734f6", + "metadata": {}, + "source": [ + "## We can tune model further!" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "id": "3940e75e", + "metadata": {}, + "outputs": [], + "source": [ + "MI = ModelInterface(model=best_model, optimizer=optimizer_adam, framework_plugin=framework_adapter)\n", + "fl_experiment.start_experiment(model_provider=MI, task_keeper=TI, data_loader=fed_dataset, rounds_to_train=4, \\\n", + " opt_treatment='CONTINUE_GLOBAL')" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "id": "1bd786d2", + "metadata": {}, + "outputs": [], + "source": [ + "best_model = fl_experiment.get_best_model()\n", + "# Validating trained model\n", + "validate(best_model, fed_dataset.get_valid_loader(), 'cpu')" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "id": "2dcdbfeb", + "metadata": {}, + "outputs": [], + "source": [ + "a = (np.zeros((2,4)), np.ones(2,), 2np.ones(2,))\n", + "a = [elem for elem in a if elem.shape==2 else elem.newaxis()]\n", + "np.concatenate(a)" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "id": "70863215", + "metadata": {}, + "outputs": [], + "source": [ + "class A:\n", + " def __init__(self, kwargs):\n", + " print(\"Class A\", kwargs)\n", + " \n", + "class B:\n", + " def __init__(self, kwargs):\n", + " super().__init__()\n", + " print(\"Class B\", kwargs)\n", + " \n", + "class C(B, A):\n", + " def __init__(self, kwargs):\n", + " super().__init__()\n", + " print(\"Class C\", kwargs)\n", + " \n", + "# class A:\n", + "# def __init__(self, kwargs):\n", + "# print(\"Class A\", kwargs)" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "id": "4ba44919", + "metadata": {}, + "outputs": [], + "source": [ + "c = C(x=1, z=5)" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "id": "0637f26f", + "metadata": {}, + "outputs": [], + "source": [] + } + ], + "metadata": { + "kernelspec": { + "display_name": "Python 3", + "language": "python", + "name": "python3" + }, + "language_info": { + "codemirror_mode": { + "name": "ipython", + "version": 3 + }, + "file_extension": ".py", + "mimetype": "text/x-python", + "name": "python", + "nbconvert_exporter": "python", + "pygments_lexer": "ipython3", + "version": "3.8.5" + } + }, + "nbformat": 4, + "nbformat_minor": 5 +} \ No newline at end of file diff --git a/openfl-tutorials/interactive_api_tutorials_(experimental)/Pytorch_Kvasir_UNET_workspace/layers.py b/openfl-tutorials/interactive_api/Pytorch_Kvasir_UNET/layers.py similarity index 90% rename from openfl-tutorials/interactive_api_tutorials_(experimental)/Pytorch_Kvasir_UNET_workspace/layers.py rename to openfl-tutorials/interactive_api/Pytorch_Kvasir_UNET/layers.py index 75e1e6902e..bf876c1cb3 100644 --- a/openfl-tutorials/interactive_api_tutorials_(experimental)/Pytorch_Kvasir_UNET_workspace/layers.py +++ b/openfl-tutorials/interactive_api/Pytorch_Kvasir_UNET/layers.py @@ -56,7 +56,9 @@ def __init__(self, in_ch, out_ch): """Initialize layer.""" super(Down, self).__init__() self.mpconv = nn.Sequential( - nn.MaxPool2d(2), DoubleConv(in_ch, out_ch)) + nn.MaxPool2d(2), + DoubleConv(in_ch, out_ch) + ) def forward(self, x): """Do forward pass.""" @@ -74,11 +76,12 @@ def __init__(self, in_ch, out_ch, bilinear=False): self.out_ch = out_ch if bilinear: self.up = nn.Upsample( - scale_factor=2, mode='bilinear', align_corners=True) - else: - self.up = nn.ConvTranspose2d( - in_ch, in_ch // 2, 2, stride=2 + scale_factor=2, + mode='bilinear', + align_corners=True ) + else: + self.up = nn.ConvTranspose2d(in_ch, in_ch // 2, 2, stride=2) self.conv = DoubleConv(in_ch, out_ch) def forward(self, x1, x2): diff --git a/openfl-tutorials/interactive_api/Pytorch_Kvasir_UNET/new_python_api_UNET.ipynb b/openfl-tutorials/interactive_api/Pytorch_Kvasir_UNET/new_python_api_UNET.ipynb new file mode 100644 index 0000000000..e236cbea56 --- /dev/null +++ b/openfl-tutorials/interactive_api/Pytorch_Kvasir_UNET/new_python_api_UNET.ipynb @@ -0,0 +1,1159 @@ +{ + "cells": [ + { + "cell_type": "markdown", + "id": "liquid-jacket", + "metadata": {}, + "source": [ + "# Federated PyTorch UNET Tutorial\n", + "## Using low-level Python API" + ] + }, + { + "cell_type": "code", + "execution_count": 1, + "id": "alike-sharing", + "metadata": { + "scrolled": true + }, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "Requirement already satisfied: torchvision in /home/davidyuk/.virtualenvs/openfl/lib/python3.8/site-packages (0.9.1)\n", + "Requirement already satisfied: numpy in /home/davidyuk/.virtualenvs/openfl/lib/python3.8/site-packages (from torchvision) (1.20.3)\n", + "Requirement already satisfied: torch==1.8.1 in /home/davidyuk/.virtualenvs/openfl/lib/python3.8/site-packages (from torchvision) (1.8.1)\n", + "Requirement already satisfied: pillow>=4.1.1 in /home/davidyuk/.virtualenvs/openfl/lib/python3.8/site-packages (from torchvision) (8.2.0)\n", + "Requirement already satisfied: typing-extensions in /home/davidyuk/.virtualenvs/openfl/lib/python3.8/site-packages (from torch==1.8.1->torchvision) (3.10.0.0)\n", + "\u001B[33mWARNING: You are using pip version 21.1; however, version 21.1.2 is available.\n", + "You should consider upgrading via the '/home/davidyuk/.virtualenvs/openfl/bin/python -m pip install --upgrade pip' command.\u001B[0m\n", + "Requirement already satisfied: scikit-image in /home/davidyuk/.virtualenvs/openfl/lib/python3.8/site-packages (0.18.1)\n", + "Requirement already satisfied: matplotlib!=3.0.0,>=2.0.0 in /home/davidyuk/.virtualenvs/openfl/lib/python3.8/site-packages (from scikit-image) (3.4.2)\n", + "Requirement already satisfied: numpy>=1.16.5 in /home/davidyuk/.virtualenvs/openfl/lib/python3.8/site-packages (from scikit-image) (1.20.3)\n", + "Requirement already satisfied: scipy>=1.0.1 in /home/davidyuk/.virtualenvs/openfl/lib/python3.8/site-packages (from scikit-image) (1.6.3)\n", + "Requirement already satisfied: pillow!=7.1.0,!=7.1.1,>=4.3.0 in /home/davidyuk/.virtualenvs/openfl/lib/python3.8/site-packages (from scikit-image) (8.2.0)\n", + "Requirement already satisfied: networkx>=2.0 in /home/davidyuk/.virtualenvs/openfl/lib/python3.8/site-packages (from scikit-image) (2.5.1)\n", + "Requirement already satisfied: PyWavelets>=1.1.1 in /home/davidyuk/.virtualenvs/openfl/lib/python3.8/site-packages (from scikit-image) (1.1.1)\n", + "Requirement already satisfied: tifffile>=2019.7.26 in /home/davidyuk/.virtualenvs/openfl/lib/python3.8/site-packages (from scikit-image) (2021.4.8)\n", + "Requirement already satisfied: imageio>=2.3.0 in /home/davidyuk/.virtualenvs/openfl/lib/python3.8/site-packages (from scikit-image) (2.9.0)\n", + "Requirement already satisfied: cycler>=0.10 in /home/davidyuk/.virtualenvs/openfl/lib/python3.8/site-packages (from matplotlib!=3.0.0,>=2.0.0->scikit-image) (0.10.0)\n", + "Requirement already satisfied: pyparsing>=2.2.1 in /home/davidyuk/.virtualenvs/openfl/lib/python3.8/site-packages (from matplotlib!=3.0.0,>=2.0.0->scikit-image) (2.4.7)\n", + "Requirement already satisfied: python-dateutil>=2.7 in /home/davidyuk/.virtualenvs/openfl/lib/python3.8/site-packages (from matplotlib!=3.0.0,>=2.0.0->scikit-image) (2.8.1)\n", + "Requirement already satisfied: kiwisolver>=1.0.1 in /home/davidyuk/.virtualenvs/openfl/lib/python3.8/site-packages (from matplotlib!=3.0.0,>=2.0.0->scikit-image) (1.3.1)\n", + "Requirement already satisfied: six in /home/davidyuk/.virtualenvs/openfl/lib/python3.8/site-packages (from cycler>=0.10->matplotlib!=3.0.0,>=2.0.0->scikit-image) (1.16.0)\n", + "Requirement already satisfied: decorator<5,>=4.3 in /home/davidyuk/.virtualenvs/openfl/lib/python3.8/site-packages (from networkx>=2.0->scikit-image) (4.4.2)\n", + "\u001B[33mWARNING: You are using pip version 21.1; however, version 21.1.2 is available.\n", + "You should consider upgrading via the '/home/davidyuk/.virtualenvs/openfl/bin/python -m pip install --upgrade pip' command.\u001B[0m\n" + ] + } + ], + "source": [ + "# Install dependencies if not already installed\n", + "!pip install torchvision==0.8.1\n", + "!pip install scikit-image" + ] + }, + { + "cell_type": "markdown", + "id": "caring-distinction", + "metadata": {}, + "source": [ + "### Describe the model and optimizer" + ] + }, + { + "cell_type": "code", + "execution_count": 11, + "id": "visible-victor", + "metadata": {}, + "outputs": [], + "source": [ + "import torch\n", + "import torch.nn as nn\n", + "import torch.optim as optim" + ] + }, + { + "cell_type": "code", + "execution_count": 12, + "id": "foreign-gospel", + "metadata": {}, + "outputs": [], + "source": [ + "\"\"\"\n", + "UNet model definition\n", + "\"\"\"\n", + "from layers import soft_dice_coef, soft_dice_loss, DoubleConv, Down, Up\n", + "\n", + "class UNet(nn.Module):\n", + " def __init__(self, n_channels=3, n_classes=1):\n", + " super().__init__()\n", + " self.inc = DoubleConv(n_channels, 64)\n", + " self.down1 = Down(64, 128)\n", + " self.down2 = Down(128, 256)\n", + " self.down3 = Down(256, 512)\n", + " self.down4 = Down(512, 1024)\n", + " self.up1 = Up(1024, 512)\n", + " self.up2 = Up(512, 256)\n", + " self.up3 = Up(256, 128)\n", + " self.up4 = Up(128, 64)\n", + " self.outc = nn.Conv2d(64, n_classes, 1)\n", + "\n", + " def forward(self, x):\n", + " x1 = self.inc(x)\n", + " x2 = self.down1(x1)\n", + " x3 = self.down2(x2)\n", + " x4 = self.down3(x3)\n", + " x5 = self.down4(x4)\n", + " x = self.up1(x5, x4)\n", + " x = self.up2(x, x3)\n", + " x = self.up3(x, x2)\n", + " x = self.up4(x, x1)\n", + " x = self.outc(x)\n", + " x = torch.sigmoid(x)\n", + " return x\n", + " \n", + " # Lookup by parent module (as in current inspect)\n", + " if hasattr(object, '__module__'):\n", + " object_ = sys.modules.get(object.__module__)\n", + " if hasattr(object_, '__file__'):\n", + " return object_.__file__\n", + " \n", + " # If parent module is __main__, lookup by methods (NEW)\n", + " for name, member in inspect.getmembers(object):\n", + " if inspect.isfunction(member) and object.__qualname__ + '.' + member.__name__ == member.__qualname__:\n", + " return inspect.getfile(member)\n", + " else:\n", + " raise TypeError('Source for {!r} not found'.format(object))\n", + "inspect.getfile = new_getfile\n", + "\n", + "obj = UNet\n", + "cell_code = \"\".join(inspect.linecache.getlines(new_getfile(obj)))\n", + "class_code = extract_symbols(cell_code, obj.__name__)[0][0]\n", + "print(class_code)" + ] + }, + { + "cell_type": "code", + "execution_count": 14, + "id": "greater-activation", + "metadata": {}, + "outputs": [], + "source": [ + "optimizer_adam = optim.Adam(model_unet.parameters(), lr=1e-4)" + ] + }, + { + "cell_type": "markdown", + "id": "insured-metallic", + "metadata": {}, + "source": [ + "### Prepare data" + ] + }, + { + "cell_type": "markdown", + "id": "essential-remainder", + "metadata": {}, + "source": [ + "We ask user to keep all the test data in `data/` folder under the workspace as it will not be sent to collaborators" + ] + }, + { + "cell_type": "code", + "execution_count": 15, + "id": "blind-circuit", + "metadata": {}, + "outputs": [], + "source": [ + "import os\n", + "from hashlib import sha384\n", + "import PIL\n", + "from PIL import Image\n", + "from torch.utils.data import Dataset, DataLoader\n", + "from torchvision import transforms as tsf" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "id": "recorded-input", + "metadata": { + "scrolled": true + }, + "outputs": [], + "source": [ + "os.makedirs('data', exist_ok=True)\n", + "!wget -nc 'https://datasets.simula.no/hyper-kvasir/hyper-kvasir-segmented-images.zip' -O ./data/kvasir.zip\n", + "ZIP_SHA384 = 'e30d18a772c6520476e55b610a4db457237f151e'\\\n", + " '19182849d54b49ae24699881c1e18e0961f77642be900450ef8b22e7'\n", + "assert sha384(open('./data/kvasir.zip', 'rb').read(\n", + " os.path.getsize('./data/kvasir.zip'))).hexdigest() == ZIP_SHA384\n", + "!unzip -n ./data/kvasir.zip -d ./data" + ] + }, + { + "cell_type": "code", + "execution_count": 10, + "id": "5fa108f5", + "metadata": {}, + "outputs": [ + { + "data": { + "text/plain": [ + "<generator object <genexpr> at 0x7f616dce3f90>" + ] + }, + "execution_count": 10, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "rank_worldsize='3,4'\n", + "(int(num) for num in rank_worldsize.split(','))" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "id": "fec3c6d4", + "metadata": {}, + "outputs": [], + "source": [] + }, + { + "cell_type": "code", + "execution_count": 40, + "id": "prepared-israel", + "metadata": {}, + "outputs": [], + "source": [ + "DATA_PATH = './data/segmented-images/'\n", + "import numpy as np\n", + "\n", + "def read_data(image_path, mask_path):\n", + " \"\"\"\n", + " Read image and mask from disk.\n", + " \"\"\"\n", + " img = Image.open(image_path)\n", + " img = np.asarray(img)\n", + " assert(img.shape[2] == 3)\n", + " mask = Image.open(mask_path)\n", + " mask = np.asarray(mask)\n", + " return (img, mask[:, :, 0].astype(np.uint8))\n", + "\n", + "\n", + "class KvasirDataset(Dataset):\n", + " \"\"\"\n", + " Kvasir dataset contains 1000 images for all collaborators.\n", + " Args:\n", + " data_path: path to dataset on disk\n", + " collaborator_count: total number of collaborators\n", + " collaborator_num: number of current collaborator\n", + " is_validation: validation option\n", + " \"\"\"\n", + "\n", + " def __init__(self, images_path = './kvasir_data/segmented-images/images/', \\\n", + " masks_path = './kvasir_data/segmented-images/masks/',\n", + " validation_fraction=1/8, is_validation=False):\n", + "\n", + " self.images_path = images_path\n", + " self.masks_path = masks_path\n", + " self.images_names = [\n", + " img_name\n", + " for img_name in sorted(os.listdir(self.images_path))\n", + " if len(img_name) > 3 and img_name[-3:] == 'jpg'\n", + " ]\n", + "\n", + " assert(len(self.images_names) > 2), \"Too few images\"\n", + " \n", + " validation_size = max(1, int(len(self.images_names) * validation_fraction))\n", + " \n", + " if is_validation:\n", + " self.images_names = self.images_names[-validation_size :]\n", + " else:\n", + " self.images_names = self.images_names[: -validation_size]\n", + "\n", + " # Prepare transforms\n", + " self.img_trans = tsf.Compose([\n", + " tsf.ToPILImage(),\n", + " tsf.Resize((332, 332)),\n", + " tsf.ToTensor(),\n", + " tsf.Normalize(mean=[0.5, 0.5, 0.5], std=[0.5, 0.5, 0.5])])\n", + " self.mask_trans = tsf.Compose([\n", + " tsf.ToPILImage(),\n", + " tsf.Resize((332, 332), interpolation=PIL.Image.NEAREST),\n", + " tsf.ToTensor()])\n", + " \n", + "\n", + " def __getitem__(self, index):\n", + " name = self.images_names[index]\n", + " img, mask = read_data(self.images_path + name, self.masks_path + name)\n", + " img = self.img_trans(img).numpy()\n", + " mask = self.mask_trans(mask).numpy()\n", + " return img, mask\n", + "\n", + " def __len__(self):\n", + " return len(self.images_names)" + ] + }, + { + "cell_type": "code", + "execution_count": 41, + "id": "c17c0944", + "metadata": {}, + "outputs": [], + "source": [ + "img = Image.open('./kvasir_data/segmented-images/images/cju0qkwl35piu0993l0dewei2.jpg')\n", + "img.resize()" + ] + }, + { + "cell_type": "code", + "execution_count": 42, + "id": "9c889655", + "metadata": {}, + "outputs": [ + { + "name": "stderr", + "output_type": "stream", + "text": [ + "/home/davidyuk/.virtualenvs/openfl/lib/python3.8/site-packages/torchvision/transforms/transforms.py:257: UserWarning: Argument interpolation should be of type InterpolationMode instead of int. Please, use InterpolationMode enum.\n", + " warnings.warn(\n" + ] + }, + { + "data": { + "text/plain": [ + "array([[[-1., -1., -1., ..., -1., -1., -1.],\n", + " [-1., -1., -1., ..., -1., -1., -1.],\n", + " [-1., -1., -1., ..., -1., -1., -1.],\n", + " ...,\n", + " [-1., -1., -1., ..., -1., -1., -1.],\n", + " [-1., -1., -1., ..., -1., -1., -1.],\n", + " [-1., -1., -1., ..., -1., -1., -1.]],\n", + "\n", + " [[-1., -1., -1., ..., -1., -1., -1.],\n", + " [-1., -1., -1., ..., -1., -1., -1.],\n", + " [-1., -1., -1., ..., -1., -1., -1.],\n", + " ...,\n", + " [-1., -1., -1., ..., -1., -1., -1.],\n", + " [-1., -1., -1., ..., -1., -1., -1.],\n", + " [-1., -1., -1., ..., -1., -1., -1.]],\n", + "\n", + " [[-1., -1., -1., ..., -1., -1., -1.],\n", + " [-1., -1., -1., ..., -1., -1., -1.],\n", + " [-1., -1., -1., ..., -1., -1., -1.],\n", + " ...,\n", + " [-1., -1., -1., ..., -1., -1., -1.],\n", + " [-1., -1., -1., ..., -1., -1., -1.],\n", + " [-1., -1., -1., ..., -1., -1., -1.]]], dtype=float32)" + ] + }, + "execution_count": 42, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "dset = KvasirDataset()\n", + "dset[0][0]" + ] + }, + { + "cell_type": "markdown", + "id": "portuguese-groove", + "metadata": {}, + "source": [ + "### Define Federated Learning tasks" + ] + }, + { + "cell_type": "code", + "execution_count": 7, + "id": "enclosed-abuse", + "metadata": {}, + "outputs": [], + "source": [ + "def train(unet_model, train_loader, optimizer, device, loss_fn=soft_dice_loss):\n", + " \n", + " function_defined_in_notebook()\n", + " \n", + " unet_model.train()\n", + " unet_model.to(device)\n", + "\n", + " losses = []\n", + "\n", + " for data, target in train_loader:\n", + " data, target = torch.tensor(data).to(device), torch.tensor(\n", + " target).to(device, dtype=torch.float32)\n", + " optimizer.zero_grad()\n", + " output = unet_model(data)\n", + " loss = loss_fn(output=output, target=target)\n", + " loss.backward()\n", + " optimizer.step()\n", + " losses.append(loss.detach().cpu().numpy())\n", + " \n", + " return {'train_loss': np.mean(losses),}\n", + "\n", + "\n", + "def validate(unet_model, val_loader, device):\n", + " unet_model.eval()\n", + " unet_model.to(device)\n", + "\n", + " val_score = 0\n", + " total_samples = 0\n", + "\n", + " with torch.no_grad():\n", + " for data, target in val_loader:\n", + " samples = target.shape[0]\n", + " total_samples += samples\n", + " data, target = torch.tensor(data).to(device), \\\n", + " torch.tensor(target).to(device, dtype=torch.int64)\n", + " output = unet_model(data)\n", + " val = soft_dice_coef(output, target)\n", + " val_score += val.sum().cpu().numpy()\n", + " \n", + " return {'dice_coef': val_score / total_samples,}" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "id": "secure-protest", + "metadata": {}, + "outputs": [], + "source": [] + }, + { + "cell_type": "markdown", + "id": "obvious-tyler", + "metadata": {}, + "source": [ + "## Describing FL experiment" + ] + }, + { + "cell_type": "code", + "execution_count": 1, + "id": "rubber-address", + "metadata": {}, + "outputs": [], + "source": [ + "from openfl.interface.interactive_api.experiment import TaskInterface, DataInterface, ModelInterface, FLExperiment" + ] + }, + { + "cell_type": "code", + "execution_count": 2, + "id": "fd0a70ed", + "metadata": {}, + "outputs": [], + "source": [ + "from torch.utils.data import Dataset, DataLoader\n", + "class A(Dataset, DataInterface):\n", + " pass\n", + "a = A()\n", + "# dir(a)" + ] + }, + { + "cell_type": "markdown", + "id": "caroline-passion", + "metadata": {}, + "source": [ + "### Register model" + ] + }, + { + "cell_type": "code", + "execution_count": 9, + "id": "handled-teens", + "metadata": {}, + "outputs": [], + "source": [ + "from copy import deepcopy\n", + "\n", + "framework_adapter = 'openfl.plugins.frameworks_adapters.pytorch_adapter.FrameworkAdapterPlugin'\n", + "MI = ModelInterface(model=model_unet, optimizer=optimizer_adam, framework_plugin=framework_adapter)\n", + "\n", + "# Save the initial model state\n", + "initial_model = deepcopy(model_unet)" + ] + }, + { + "cell_type": "markdown", + "id": "sustainable-public", + "metadata": {}, + "source": [ + "### Register dataset" + ] + }, + { + "cell_type": "markdown", + "id": "unlike-texas", + "metadata": {}, + "source": [ + "We extract User dataset class implementation.\n", + "Is it convinient?\n", + "What if the dataset is not a class?" + ] + }, + { + "cell_type": "code", + "execution_count": 10, + "id": "anticipated-coverage", + "metadata": {}, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "We implement all abstract methods from mixin in this class\n", + "This method is called on the collaborator node\n", + "User Dataset initialized with data path on the collaborator node\n" + ] + } + ], + "source": [ + "class UserDataset:\n", + " def __init__(self, path_to_local_data):\n", + " print(f'User Dataset initialized with {path_to_local_data}')\n", + " \n", + " \n", + "class OpenflMixin: \n", + " def _delayed_init(self):\n", + " raise NotImplementedError\n", + " \n", + " \n", + "class FedDataset(OpenflMixin):\n", + " def __init__(self, UserDataset):\n", + " self.user_dataset_class = UserDataset\n", + " print('We implement all abstract methods from mixin in this class')\n", + " \n", + " def _delayed_init(self, data_path):\n", + " print('This method is called on the collaborator node')\n", + " dataset_obj = self.user_dataset_class(data_path)\n", + " \n", + " \n", + "fed_dataset = FedDataset(UserDataset)\n", + "fed_dataset._delayed_init('data path on the collaborator node')" + ] + }, + { + "cell_type": "code", + "execution_count": 11, + "id": "verbal-newton", + "metadata": {}, + "outputs": [], + "source": [ + "# class FedDataset(DataInterface):\n", + "# def __init__(self, UserDatasetClass, kwargs):\n", + "# self.UserDatasetClass = UserDatasetClass\n", + "# self.kwargs = kwargs\n", + " \n", + "# def _delayed_init(self, data_path='1,1'):\n", + "# # With the next command the local dataset will be loaded on the collaborator node\n", + "# # For this example we have the same dataset on the same path, and we will shard it\n", + "# # So we use `data_path` information for this purpose.\n", + "# self.rank, self.world_size = [int(part) for part in data_path.split(',')]\n", + " \n", + "# validation_fraction=1/8\n", + "# self.train_set = self.UserDatasetClass(validation_fraction=validation_fraction, is_validation=False)\n", + "# self.valid_set = self.UserDatasetClass(validation_fraction=validation_fraction, is_validation=True)\n", + " \n", + "# # Do the actual sharding\n", + "# self._do_sharding( self.rank, self.world_size)\n", + " \n", + "# def _do_sharding(self, rank, world_size):\n", + "# # This method relies on the dataset's implementation\n", + "# # i.e. coupled in a bad way\n", + "# self.train_set.images_names = self.train_set.images_names[ rank-1 :: world_size ]\n", + "\n", + "# def get_train_loader(self, kwargs):\n", + "# \"\"\"\n", + "# Output of this method will be provided to tasks with optimizer in contract\n", + "# \"\"\"\n", + "# return DataLoader(\n", + "# self.train_set, num_workers=8, batch_size=self.kwargs['train_bs'], shuffle=True\n", + "# )\n", + "\n", + "# def get_valid_loader(self, kwargs):\n", + "# \"\"\"\n", + "# Output of this method will be provided to tasks without optimizer in contract\n", + "# \"\"\"\n", + "# return DataLoader(self.valid_set, num_workers=8, batch_size=self.kwargs['valid_bs'])\n", + "\n", + "# def get_train_data_size(self):\n", + "# \"\"\"\n", + "# Information for aggregation\n", + "# \"\"\"\n", + "# return len(self.train_set)\n", + "\n", + "# def get_valid_data_size(self):\n", + "# \"\"\"\n", + "# Information for aggregation\n", + "# \"\"\"\n", + "# return len(self.valid_set)\n", + " \n", + "# fed_dataset = FedDataset(KvasirDataset, train_bs=8, valid_bs=8)" + ] + }, + { + "cell_type": "markdown", + "id": "marine-kansas", + "metadata": {}, + "source": [ + "### Register tasks" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "id": "increasing-builder", + "metadata": {}, + "outputs": [], + "source": [ + "TI = TaskInterface()\n", + "import torch\n", + "\n", + "import tqdm\n", + "\n", + "# The Interactive API supports registering functions definied in main module or imported.\n", + "def function_defined_in_notebook(some_parameter):\n", + " print(f'Also I accept a parameter and it is {some_parameter}')\n", + "\n", + "# Task interface currently supports only standalone functions.\n", + "@TI.add_kwargs({'some_parameter': 42})\n", + "@TI.register_fl_task(model='unet_model', data_loader='train_loader', \\\n", + " device='device', optimizer='optimizer') \n", + "def train(unet_model, train_loader, optimizer, device, loss_fn=soft_dice_loss, some_parameter=None):\n", + " if not torch.cuda.is_available():\n", + " device = 'cpu'\n", + " \n", + " function_defined_in_notebook(some_parameter)\n", + " \n", + " train_loader = tqdm.tqdm(train_loader, desc=\"train\")\n", + " \n", + " unet_model.train()\n", + " unet_model.to(device)\n", + "\n", + " losses = []\n", + "\n", + " for data, target in train_loader:\n", + " data, target = torch.tensor(data).to(device), torch.tensor(\n", + " target).to(device, dtype=torch.float32)\n", + " optimizer.zero_grad()\n", + " output = unet_model(data)\n", + " loss = loss_fn(output=output, target=target)\n", + " loss.backward()\n", + " optimizer.step()\n", + " losses.append(loss.detach().cpu().numpy())\n", + " \n", + " return {'train_loss': np.mean(losses),}\n", + "\n", + "\n", + "@TI.register_fl_task(model='unet_model', data_loader='val_loader', device='device') \n", + "def validate(unet_model, val_loader, device):\n", + " unet_model.eval()\n", + " unet_model.to(device)\n", + " \n", + " val_loader = tqdm.tqdm(val_loader, desc=\"validate\")\n", + "\n", + " val_score = 0\n", + " total_samples = 0\n", + "\n", + " with torch.no_grad():\n", + " for data, target in val_loader:\n", + " samples = target.shape[0]\n", + " total_samples += samples\n", + " data, target = torch.tensor(data).to(device), \\\n", + " torch.tensor(target).to(device, dtype=torch.int64)\n", + " output = unet_model(data)\n", + " val = soft_dice_coef(output, target)\n", + " val_score += val.sum().cpu().numpy()\n", + " \n", + " return {'dice_coef': val_score / total_samples,}" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "id": "b50a39c7", + "metadata": {}, + "outputs": [], + "source": [] + }, + { + "cell_type": "markdown", + "id": "derived-bride", + "metadata": {}, + "source": [ + "## Time to start a federated learning experiment" + ] + }, + { + "cell_type": "code", + "execution_count": 1, + "id": "4485ac79", + "metadata": {}, + "outputs": [], + "source": [ + "# Create a federation\n", + "from openfl.interface.interactive_api.federation import Federation\n", + "\n", + "# 1) Run with API layer - Director mTLS \n", + "# If the user wants to enable mTLS their must provide CA root chain, and signed key pair to the federation interface\n", + "cert_chain = 'cert/cert_chain.crt'\n", + "API_certificate = 'cert/API_certificate.crt'\n", + "API_private_key = 'cert/API_private.key'\n", + "\n", + "federation = Federation(director_node_fqdn='some.fqdn', disable_tls=False,\n", + " cert_chain=cert_chain, API_cert=API_certificate, API_private_key=API_private_key)\n", + "\n", + "# --------------------------------------------------------------------------------------------------------------------\n", + "\n", + "# 2) Run with TLS disabled (trusted environment)\n", + "# Federation can also determine local fqdn automatically\n", + "federation = Federation(director_node_fqdn='localhost', disable_tls=True)" + ] + }, + { + "cell_type": "code", + "execution_count": 2, + "id": "920216d3", + "metadata": { + "scrolled": true + }, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "(300, 300, 3) (6,) 10\n" + ] + } + ], + "source": [ + "# First, request a dummy_shard_desc that holds information about the federated dataset \n", + "dummy_shard_desc = federation.get_dummy_shard_descriptor(size=10)" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "id": "996f6460", + "metadata": {}, + "outputs": [], + "source": [ + "\"1,6\"" + ] + }, + { + "cell_type": "code", + "execution_count": 17, + "id": "64f37dcf", + "metadata": { + "scrolled": true + }, + "outputs": [ + { + "ename": "NameError", + "evalue": "name 'DataInterface' is not defined", + "output_type": "error", + "traceback": [ + "\u001B[0;31m---------------------------------------------------------------------------\u001B[0m", + "\u001B[0;31mNameError\u001B[0m Traceback (most recent call last)", + "\u001B[0;32m<ipython-input-17-858ae44ef56e>\u001B[0m in \u001B[0;36m<module>\u001B[0;34m\u001B[0m\n\u001B[1;32m 1\u001B[0m \u001B[0;31m# Now you can implement you data loaders using dummy_shard_desc\u001B[0m\u001B[0;34m\u001B[0m\u001B[0;34m\u001B[0m\u001B[0;34m\u001B[0m\u001B[0m\n\u001B[0;32m----> 2\u001B[0;31m \u001B[0;32mclass\u001B[0m \u001B[0mKvasirSD\u001B[0m\u001B[0;34m(\u001B[0m\u001B[0mDataInterface\u001B[0m\u001B[0;34m,\u001B[0m \u001B[0mDataset\u001B[0m\u001B[0;34m)\u001B[0m\u001B[0;34m:\u001B[0m\u001B[0;34m\u001B[0m\u001B[0;34m\u001B[0m\u001B[0m\n\u001B[0m\u001B[1;32m 3\u001B[0m \u001B[0;34m\u001B[0m\u001B[0m\n\u001B[1;32m 4\u001B[0m \u001B[0;32mdef\u001B[0m \u001B[0m__init__\u001B[0m\u001B[0;34m(\u001B[0m\u001B[0mself\u001B[0m\u001B[0;34m,\u001B[0m \u001B[0mshard_descriptor\u001B[0m\u001B[0;34m,\u001B[0m \u001B[0mvalidation_fraction\u001B[0m\u001B[0;34m=\u001B[0m\u001B[0;36m1\u001B[0m\u001B[0;34m/\u001B[0m\u001B[0;36m8\u001B[0m\u001B[0;34m,\u001B[0m \u001B[0mis_validation\u001B[0m\u001B[0;34m=\u001B[0m\u001B[0;32mFalse\u001B[0m\u001B[0;34m,\u001B[0m \u001B[0;34m\u001B[0m\u001B[0mkwargs\u001B[0m\u001B[0;34m)\u001B[0m\u001B[0;34m:\u001B[0m\u001B[0;34m\u001B[0m\u001B[0;34m\u001B[0m\u001B[0m\n\u001B[1;32m 5\u001B[0m \u001B[0msuper\u001B[0m\u001B[0;34m(\u001B[0m\u001B[0;34m)\u001B[0m\u001B[0;34m.\u001B[0m\u001B[0m__init__\u001B[0m\u001B[0;34m(\u001B[0m\u001B[0;34m\u001B[0m\u001B[0mkwargs\u001B[0m\u001B[0;34m)\u001B[0m\u001B[0;34m\u001B[0m\u001B[0;34m\u001B[0m\u001B[0m\n", + "\u001B[0;31mNameError\u001B[0m: name 'DataInterface' is not defined" + ] + } + ], + "source": [ + "# Now you can implement you data loaders using dummy_shard_desc\n", + "class KvasirSD(DataInterface, Dataset):\n", + "\n", + " def __init__(self, shard_descriptor, validation_fraction=1/8, is_validation=False, kwargs):\n", + " super().__init__(kwargs)\n", + "\n", + " self.shard_descriptor = shard_descriptor\n", + " \n", + " self.validation_fraction = validation_fraction\n", + " \n", + " validation_size = max(1, int(len(self.shard_descriptor) * validation_fraction))\n", + " self.train_indeces = list(range())\n", + " \n", + "# if is_validation:\n", + "# self.images_names = self.images_names[-validation_size :]\n", + "# else:\n", + "# self.images_names = self.images_names[: -validation_size]\n", + " \n", + " # Prepare transforms\n", + " self.img_trans = tsf.Compose([\n", + " tsf.ToPILImage(),\n", + " tsf.Resize((332, 332)),\n", + " tsf.ToTensor(),\n", + " tsf.Normalize(mean=[0.5, 0.5, 0.5], std=[0.5, 0.5, 0.5])])\n", + " self.mask_trans = tsf.Compose([\n", + " tsf.ToPILImage(),\n", + " tsf.Resize((332, 332), interpolation=PIL.Image.NEAREST),\n", + " tsf.ToTensor()])\n", + " \n", + "\n", + " def __getitem__(self, index):\n", + " img, mask = self.shard_descriptor[index]\n", + " img = self.img_trans(img).numpy()\n", + " mask = self.mask_trans(mask).numpy()\n", + " return img, mask\n", + "\n", + " def __len__(self):\n", + " return len(self.shard_descriptor)\n", + " \n", + " \n", + " def get_train_loader(self, kwargs):\n", + " \"\"\"\n", + " Output of this method will be provided to tasks with optimizer in contract\n", + " \"\"\"\n", + " return DataLoader(\n", + " self.train_set, num_workers=8, batch_size=self.kwargs['train_bs'], shuffle=True\n", + " )\n", + "\n", + " def get_valid_loader(self, kwargs):\n", + " \"\"\"\n", + " Output of this method will be provided to tasks without optimizer in contract\n", + " \"\"\"\n", + " return DataLoader(self.valid_set, num_workers=8, batch_size=self.kwargs['valid_bs'])\n", + "\n", + " def get_train_data_size(self):\n", + " \"\"\"\n", + " Information for aggregation\n", + " \"\"\"\n", + " return len(self.train_set)\n", + "\n", + " def get_valid_data_size(self):\n", + " \"\"\"\n", + " Information for aggregation\n", + " \"\"\"\n", + " return len(self.valid_set)" + ] + }, + { + "cell_type": "code", + "execution_count": 22, + "id": "d8df35f5", + "metadata": {}, + "outputs": [ + { + "ename": "SyntaxError", + "evalue": "invalid syntax (<ipython-input-22-9778031892f4>, line 1)", + "output_type": "error", + "traceback": [ + "\u001B[0;36m File \u001B[0;32m\"<ipython-input-22-9778031892f4>\"\u001B[0;36m, line \u001B[0;32m1\u001B[0m\n\u001B[0;31m [1:10]\u001B[0m\n\u001B[0m ^\u001B[0m\n\u001B[0;31mSyntaxError\u001B[0m\u001B[0;31m:\u001B[0m invalid syntax\n" + ] + } + ], + "source": [] + }, + { + "cell_type": "markdown", + "id": "af0579f8", + "metadata": {}, + "source": [ + "# Long-Living entities update\n", + "\n", + "* We now may have director running on another machine.\n", + "* We use Federation API to communicate with Director.\n", + "* Federation object should hold a Director's client (for user service)\n", + "* Keeping in mind that several API instances may be connacted to one Director.\n", + "\n", + "\n", + "* We do not think for now how we start a Director.\n", + "* But it knows the data shape and target shape for the DataScience problem in the Federation.\n", + "* Director holds the list of connected envoys, we do not need to specify it anymore.\n", + "* Director and Envoys are responsible for encrypting connections, we do not need to worry about certs.\n", + "\n", + "\n", + "* Yet we MUST have a cert to communicate to the Director.\n", + "* We MUST know the FQDN of a Director.\n", + "* Director communicates data and target shape to the Federation interface object.\n", + "\n", + "\n", + "* Experiment API may use this info to construct a dummy dataset and a `shard descriptor` stub." + ] + }, + { + "cell_type": "markdown", + "id": "adad7744", + "metadata": {}, + "source": [ + "#### Certification of an aggregator\n", + "* fx workspace certify: creates cert folder and CA as well as cert_chain\n", + "* fx aggregator generate-cert-request --fqdn `FQDN`: you can pass a specific aggregator FQDN if you want\n", + "* fx aggregator certify --fqdn `FQDN` --silent: signes aggregators cert\n", + "<br> After that just pass the paths to required certs to the Federation API\n", + "\n", + "#### Certification of a collaborator\n", + "just follow the usual procedure: <br>\n", + "fx collaborator generate-cert-request -d {DATA_PATH} -n {COL} \n", + "\n", + "fx collaborator certify --request-pkg {COL_DIRECTORY}/{FED_WORKSPACE}/col_{COL}_to_agg_cert_request.zip\n", + "\n", + "fx collaborator certify --import {FED_DIRECTORY}/agg_to_col_{COL}_signed_cert.zip" + ] + }, + { + "cell_type": "code", + "execution_count": 2, + "id": "12f5663f", + "metadata": {}, + "outputs": [ + { + "data": { + "text/plain": [ + "2" + ] + }, + "execution_count": 2, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "# Interface generator\n", + "class IFO:\n", + " def __init__(self):\n", + " self.a = 1\n", + " \n", + "class EXP:\n", + " def __init__(self):\n", + " self.ifo = IFO()\n", + " \n", + " def get_ifo(self):\n", + " return self.ifo\n", + " \n", + "exp = EXP()\n", + "ifo = exp.get_ifo()\n", + "ifo.a = 2\n", + "\n", + "# print(exp.ifo.a)\n", + "exp2 = exp.__class__()\n", + "exp.ifo.a" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "id": "mature-renewal", + "metadata": {}, + "outputs": [], + "source": [ + "# create an experimnet in federation\n", + "fl_experiment = FLExperiment(federation=federation,)" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "id": "lightweight-causing", + "metadata": { + "scrolled": false + }, + "outputs": [], + "source": [ + "# If I use autoreload I got a pickling error\n", + "\n", + "# The following command zips the workspace and python requirements to be transfered to collaborator nodes\n", + "fl_experiment.prepare_workspace_distribution(model_provider=MI, task_keeper=TI, data_loader=fed_dataset, rounds_to_train=7, \\\n", + " opt_treatment='CONTINUE_GLOBAL')\n", + "# This command starts the aggregator server. You can pass 'metric' log_level\n", + "fl_experiment.start_experiment(model_provider=MI, log_level='INFO', log_file='federation.log')\n", + "\n", + "# When the aggregator server blocks the notebook one can start collaborators\n", + "# For the test run just type console command from the workspace directory:\n", + "# `fx collaborator start -d data.yaml -n {col_name}` for all collaborators\n", + "# For the distributed experiment transfer zipped workspace to the collaborator nodes and run\n", + "# `fx workspace import --archive {workspace_name}.zip` cd to the workspace and start collaborators" + ] + }, + { + "cell_type": "markdown", + "id": "8c30b301", + "metadata": {}, + "source": [ + "## Now we validate the best model!" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "id": "55acff59", + "metadata": {}, + "outputs": [], + "source": [ + "best_model = fl_experiment.get_best_model()" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "id": "9fb5718d", + "metadata": {}, + "outputs": [], + "source": [ + "fed_dataset._delayed_init()" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "id": "a2acb7e6", + "metadata": {}, + "outputs": [], + "source": [ + "# Validating initial model\n", + "validate(initial_model, fed_dataset.get_valid_loader(), 'cpu')" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "id": "c12ca93f", + "metadata": {}, + "outputs": [], + "source": [ + "# Validating trained model\n", + "validate(best_model, fed_dataset.get_valid_loader(), 'cpu')" + ] + }, + { + "cell_type": "markdown", + "id": "1e6734f6", + "metadata": {}, + "source": [ + "## We can tune model further!" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "id": "3940e75e", + "metadata": {}, + "outputs": [], + "source": [ + "MI = ModelInterface(model=best_model, optimizer=optimizer_adam, framework_plugin=framework_adapter)\n", + "fl_experiment.start_experiment(model_provider=MI, task_keeper=TI, data_loader=fed_dataset, rounds_to_train=4, \\\n", + " opt_treatment='CONTINUE_GLOBAL')" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "id": "1bd786d2", + "metadata": {}, + "outputs": [], + "source": [ + "best_model = fl_experiment.get_best_model()\n", + "# Validating trained model\n", + "validate(best_model, fed_dataset.get_valid_loader(), 'cpu')" + ] + }, + { + "cell_type": "code", + "execution_count": 40, + "id": "2dcdbfeb", + "metadata": {}, + "outputs": [ + { + "ename": "SyntaxError", + "evalue": "invalid syntax (<ipython-input-40-b0f7ed81de85>, line 2)", + "output_type": "error", + "traceback": [ + "\u001B[0;36m File \u001B[0;32m\"<ipython-input-40-b0f7ed81de85>\"\u001B[0;36m, line \u001B[0;32m2\u001B[0m\n\u001B[0;31m a = [elem for elem in a if elem.shape==2 else elem.newaxis()]\u001B[0m\n\u001B[0m ^\u001B[0m\n\u001B[0;31mSyntaxError\u001B[0m\u001B[0;31m:\u001B[0m invalid syntax\n" + ] + } + ], + "source": [ + "a = (np.zeros((2,4)), np.ones(2,), 2np.ones(2,))\n", + "a = [elem for elem in a if elem.shape==2 else elem.newaxis()]\n", + "np.concatenate(a)" + ] + }, + { + "cell_type": "code", + "execution_count": 15, + "id": "70863215", + "metadata": {}, + "outputs": [], + "source": [ + "class A:\n", + " def __init__(self, kwargs):\n", + " print(\"Class A\", kwargs)\n", + " \n", + "class B:\n", + " def __init__(self, kwargs):\n", + " super().__init__()\n", + " print(\"Class B\", kwargs)\n", + " \n", + "class C(B, A):\n", + " def __init__(self, kwargs):\n", + " super().__init__()\n", + " print(\"Class C\", kwargs)\n", + " \n", + "# class A:\n", + "# def __init__(self, kwargs):\n", + "# print(\"Class A\", kwargs)" + ] + }, + { + "cell_type": "code", + "execution_count": 16, + "id": "4ba44919", + "metadata": {}, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "Class A {}\n", + "Class B {}\n", + "Class C {'x': 1, 'z': 5}\n" + ] + } + ], + "source": [ + "c = C(x=1, z=5)" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "id": "0637f26f", + "metadata": {}, + "outputs": [], + "source": [] + } + ], + "metadata": { + "kernelspec": { + "display_name": "Python 3", + "language": "python", + "name": "python3" + }, + "language_info": { + "codemirror_mode": { + "name": "ipython", + "version": 3 + }, + "file_extension": ".py", + "mimetype": "text/x-python", + "name": "python", + "nbconvert_exporter": "python", + "pygments_lexer": "ipython3", + "version": "3.6.9" + } + }, + "nbformat": 4, + "nbformat_minor": 5 +} \ No newline at end of file diff --git a/openfl-tutorials/interactive_api_tutorials_(experimental)/Pytorch_TinyImageNet_workspace/new_python_api_tinyimagenet.ipynb b/openfl-tutorials/interactive_api/Pytorch_TinyImageNet/new_python_api_tinyimagenet.ipynb similarity index 100% rename from openfl-tutorials/interactive_api_tutorials_(experimental)/Pytorch_TinyImageNet_workspace/new_python_api_tinyimagenet.ipynb rename to openfl-tutorials/interactive_api/Pytorch_TinyImageNet/new_python_api_tinyimagenet.ipynb diff --git a/openfl-tutorials/interactive_api_tutorials_(experimental)/Tensorflow_MNIST_workspace/new_python_api_Tensorflow_MNIST.ipynb b/openfl-tutorials/interactive_api/Tensorflow_MNIST/new_python_api_Tensorflow_MNIST.ipynb similarity index 100% rename from openfl-tutorials/interactive_api_tutorials_(experimental)/Tensorflow_MNIST_workspace/new_python_api_Tensorflow_MNIST.ipynb rename to openfl-tutorials/interactive_api/Tensorflow_MNIST/new_python_api_Tensorflow_MNIST.ipynb diff --git a/openfl-workspace/default/director.yaml b/openfl-workspace/default/director.yaml new file mode 100644 index 0000000000..32fd4caa30 --- /dev/null +++ b/openfl-workspace/default/director.yaml @@ -0,0 +1,11 @@ +# Copyright (C) 2020-2021 Intel Corporation +# SPDX-License-Identifier: Apache-2.0 + +# Director's config. +# Parameters: +# 1. sample_shape - sample shape interface unified across the Federation +# 1. target_shape - target shape interface unified across the Federation + +settings: + sample_shape: [] + target_shape: [] \ No newline at end of file diff --git a/openfl-workspace/default/shard_config.yaml b/openfl-workspace/default/shard_config.yaml new file mode 100644 index 0000000000..cb08ac22d2 --- /dev/null +++ b/openfl-workspace/default/shard_config.yaml @@ -0,0 +1,17 @@ +# Copyright (C) 2020-2021 Intel Corporation +# SPDX-License-Identifier: Apache-2.0 + +# To start envoy implement LocalShardDescriptor class in shard_descriptor module. +# Alternatively, point to an implemented Shard Descriptor in 'template' field. + +# Put in 'params' group any argument needed to initialize choosen Shard Descriptor. +# Parameters are optional. For clarity, this example config contains three parameters: +# 1. data_path - string with data location on disk +# 2. sample_shape - shape of sample's numpy representaion that will by return from __getitem__ +# 3. target_shape - shape of target's numpy representaion that will by return from __getitem__ + +template: shard_descriptor.LocalShardDescriptor +params: + data_path: data + sample_shape: [] + target_shape: [] \ No newline at end of file diff --git a/openfl-workspace/default/shard_descriptor.py b/openfl-workspace/default/shard_descriptor.py new file mode 100644 index 0000000000..2f41db0f40 --- /dev/null +++ b/openfl-workspace/default/shard_descriptor.py @@ -0,0 +1,21 @@ +# Copyright (C) 2020-2021 Intel Corporation +# SPDX-License-Identifier: Apache-2.0 +""" +Shard Descriptor template. + +It is recommended to perform tensor manipulations using numpy. +""" + +from openfl.interface.interactive_api.shard_descriptor import ShardDescriptor + + +class LocalShardDescriptor(ShardDescriptor): + """Shard descriptor subclass.""" + + def __init__(self, data_path: str, sample_shape: tuple, target_shape: tuple) -> None: + """ + Initialize local Shard Descriptor. + + Parameters are arbitrary, set up a shard_config.yaml as you need. + """ + super().__init__() diff --git a/openfl-workspace/torch_cnn_mnist/plan/plan.yaml b/openfl-workspace/torch_cnn_mnist/plan/plan.yaml index ffef4de9ac..903d94e27a 100644 --- a/openfl-workspace/torch_cnn_mnist/plan/plan.yaml +++ b/openfl-workspace/torch_cnn_mnist/plan/plan.yaml @@ -5,10 +5,13 @@ aggregator : defaults : plan/defaults/aggregator.yaml template : openfl.component.Aggregator settings : - init_state_path : save/torch_cnn_mnist_init.pbuf - best_state_path : save/torch_cnn_mnist_best.pbuf - last_state_path : save/torch_cnn_mnist_last.pbuf - rounds_to_train : 10 + init_state_path : save/torch_cnn_mnist_init.pbuf + best_state_path : save/torch_cnn_mnist_best.pbuf + last_state_path : save/torch_cnn_mnist_last.pbuf + rounds_to_train : 10 + log_metric_callback : + template : src.mnist_utils.write_metric + collaborator : defaults : plan/defaults/collaborator.yaml diff --git a/openfl-workspace/torch_cnn_mnist/requirements.txt b/openfl-workspace/torch_cnn_mnist/requirements.txt index 9ef102a829..29815a9214 100644 --- a/openfl-workspace/torch_cnn_mnist/requirements.txt +++ b/openfl-workspace/torch_cnn_mnist/requirements.txt @@ -1,2 +1,3 @@ torch >= 1.2.0 torchvision >= 0.9.1 +tensorboard diff --git a/openfl-workspace/torch_cnn_mnist/src/mnist_utils.py b/openfl-workspace/torch_cnn_mnist/src/mnist_utils.py index a03e1e6da2..16ee801b4d 100644 --- a/openfl-workspace/torch_cnn_mnist/src/mnist_utils.py +++ b/openfl-workspace/torch_cnn_mnist/src/mnist_utils.py @@ -6,11 +6,27 @@ from logging import getLogger import numpy as np +from torch.utils.tensorboard import SummaryWriter from torchvision import datasets from torchvision import transforms logger = getLogger(__name__) +writer = None + + +def get_writer(): + """Create global writer object.""" + global writer + if not writer: + writer = SummaryWriter('./logs/cnn_mnist', flush_secs=5) + + +def write_metric(node_name, task_name, metric_name, metric, round_number): + """Write metric callback.""" + get_writer() + writer.add_scalar(f'{node_name}/{task_name}/{metric_name}', metric, round_number) + def one_hot(labels, classes): """ diff --git a/openfl-workspace/torch_unet_kvasir/src/data_loader.py b/openfl-workspace/torch_unet_kvasir/src/data_loader.py index 22b555593a..3ae0b8e65e 100644 --- a/openfl-workspace/torch_unet_kvasir/src/data_loader.py +++ b/openfl-workspace/torch_unet_kvasir/src/data_loader.py @@ -51,8 +51,11 @@ def __init__(self, is_validation, shard_num, collaborator_count, kwargs): """ self.images_path = './data/segmented-images/images/' self.masks_path = './data/segmented-images/masks/' - self.images_names = [img_name for img_name in sorted(listdir( - self.images_path)) if len(img_name) > 3 and img_name[-3:] == 'jpg'] + self.images_names = [ + img_name + for img_name in sorted(listdir(self.images_path)) + if len(img_name) > 3 and img_name[-3:] == 'jpg' + ] self.images_names = self.images_names[shard_num:: collaborator_count] self.is_validation = is_validation diff --git a/openfl-workspace/torch_unet_kvasir/src/pt_unet_parts.py b/openfl-workspace/torch_unet_kvasir/src/pt_unet_parts.py index 1a7318a64f..5883c3ec6a 100644 --- a/openfl-workspace/torch_unet_kvasir/src/pt_unet_parts.py +++ b/openfl-workspace/torch_unet_kvasir/src/pt_unet_parts.py @@ -71,7 +71,8 @@ def __init__(self, in_ch, out_ch): super(Down, self).__init__() self.mpconv = nn.Sequential( nn.MaxPool2d(2), - DoubleConv(in_ch, out_ch)) + DoubleConv(in_ch, out_ch) + ) def forward(self, x): """Run forward.""" diff --git a/openfl/component/aggregator/aggregator.py b/openfl/component/aggregator/aggregator.py index 2a02cd03bd..8e5a35b8b4 100644 --- a/openfl/component/aggregator/aggregator.py +++ b/openfl/component/aggregator/aggregator.py @@ -2,6 +2,7 @@ # SPDX-License-Identifier: Apache-2.0 """Aggregator module.""" +import queue from logging import getLogger from openfl.databases import TensorDB @@ -11,6 +12,7 @@ from openfl.protocols import utils from openfl.utilities import TaskResultKey from openfl.utilities import TensorKey +from openfl.utilities.logs import write_metric class Aggregator: @@ -42,6 +44,7 @@ def __init__(self, assigner, rounds_to_train=256, + log_metric_callback=None, single_col_cert_common_name=None, compression_pipeline=None, db_store_rounds=1, @@ -81,6 +84,7 @@ def __init__(self, self.best_tensor_dict: dict = {} self.last_tensor_dict: dict = {} + self.metric_queue = queue.Queue() self.best_model_score = None if kwargs.get('initial_tensor_dict', None) is not None: @@ -103,6 +107,8 @@ def __init__(self, self.collaborator_task_weight = {} # {TaskResultKey: data_size} + self.log_metric = write_metric + def _load_initial_tensors(self): """ Load all of the tensors required to begin federated learning. @@ -482,7 +488,13 @@ def send_local_task_results(self, collaborator_name, round_number, task_name, tensor_key, nparray = self._process_named_tensor( named_tensor, collaborator_name ) - + if 'metric' in tensor_key.tags: + self.log_metric(tensor_key.tags[-1], task_name, + tensor_key.tensor_name, nparray, round_number) + self.logger.metric(f'Round {round_number}, collaborator {tensor_key.tags[-1]} ' + f'{task_name} result {tensor_key.tensor_name}:\t{nparray}') + self.metric_queue.put((tensor_key.tags[-1], task_name, + tensor_key.tensor_name, nparray, round_number)) task_results.append(tensor_key) # By giving task_key it's own weight, we can support different # training/validation weights @@ -708,7 +720,6 @@ def _compute_validation_related_task_metrics(self, task_name): task_name : str The task name to compute """ - self.logger.info(f'{task_name} task metrics...') # By default, print out all of the metrics that the validation # task sent # This handles getting the subset of collaborators that may be @@ -748,19 +759,25 @@ def _compute_validation_related_task_metrics(self, task_name): if agg_results is None: self.logger.warning( f'Aggregated metric {agg_tensor_name} could not be collected ' - f'for round {self.round_number}. Skipping reporting for this round' - ) + f'for round {self.round_number}. Skipping reporting for this round') if agg_function: - self.logger.info(f'{agg_function} {agg_tensor_name}:\t{agg_results:.4f}') + self.logger.metric(f'Round {round_number}, aggregator: {task_name} ' + f'{agg_function} {agg_tensor_name}:\t{agg_results:.4f}') else: - self.logger.info(f'{agg_tensor_name}:\t{agg_results:.4f}') + self.logger.metric(f'Round {round_number}, aggregator: {task_name} ' + f'{agg_tensor_name}:\t{agg_results:.4f}') + self.log_metric('Aggregator', task_name, tensor_key.tensor_name, + agg_results, round_number) + self.metric_queue.put(('Aggregator', task_name, tensor_key.tensor_name, + agg_results, round_number)) # TODO Add all of the logic for saving the model based # on best accuracy, lowest loss, etc. if 'validate_agg' in tags: # Compare the accuracy of the model, and # potentially save it if self.best_model_score is None or self.best_model_score < agg_results: - self.logger.info(f'Saved the best model with score {agg_results:f}') + self.logger.metric(f'Round {round_number}: saved the best ' + f'model with score {agg_results:f}') self.best_model_score = agg_results self._save_model(round_number, self.best_state_path) if 'trained' in tags: @@ -904,5 +921,5 @@ def _log_big_warning(self): # ##@##, .++:.,# `* @# +. @@@ - #`@ + # `@ , ''' diff --git a/openfl/component/ca/__init__.py b/openfl/component/ca/__init__.py new file mode 100644 index 0000000000..1d77e7fdee --- /dev/null +++ b/openfl/component/ca/__init__.py @@ -0,0 +1,4 @@ +# Copyright (C) 2020-2021 Intel Corporation +# SPDX-License-Identifier: Apache-2.0 + +"""CA package.""" diff --git a/openfl/component/ca/ca.py b/openfl/component/ca/ca.py new file mode 100644 index 0000000000..3bf6aa8458 --- /dev/null +++ b/openfl/component/ca/ca.py @@ -0,0 +1,244 @@ +# Copyright (C) 2020-2021 Intel Corporation +# SPDX-License-Identifier: Apache-2.0 + +"""Aggregator module.""" + + +import base64 +import json +import os +import shutil +import signal +import subprocess +import time +import urllib.request +from logging import getLogger +from pathlib import Path +from subprocess import call + +import requests +from click import confirm + + +logger = getLogger(__name__) + +TOKEN_DELIMITER = '.' + + +def download_step_bin(url, grep_name, architecture, prefix='.', confirmation=True): + """ + Donwload step binaries from github. + + Args: + url: address of latest release + grep_name: name to grep over github assets + architecture: architecture type to grep + prefix: folder path to download + confirmation: request user confirmation or not + """ + if confirmation: + confirm('CA binaries from github will be downloaded now', default=True, abort=True) + result = requests.get(url) + if result.status_code != 200: + logger.warning('Can\'t download binaries from github. Please try lately.') + return + + assets = result.json().get('assets', []) + urls = [ + a['browser_download_url'] + for a in assets + if grep_name in a['name'] and architecture in a['name'] + ] + url = urls[-1] + url = url.replace('https', 'http') + name = url.split('/')[-1] + logger.info(f'Downloading {name}') + urllib.request.urlretrieve(url, f'{prefix}/{name}') + shutil.unpack_archive(f'{prefix}/{name}', f'{prefix}/step') + + +def get_token(name, ca_url, ca_path='.'): + """ + Create authentication token. + + Args: + name: common name for following certificate + (aggregator fqdn or collaborator name) + ca_url: full url of CA server + ca_path: path to ca folder + """ + ca_path = Path(ca_path) + step_config_dir = ca_path / 'step_config' + pki_dir = ca_path / 'cert' + step_path, _ = get_ca_bin_paths(ca_path) + if not step_path: + raise Exception('Step-CA is not installed!\nRun `fx pki install` first') + + priv_json = step_config_dir / 'secrets' / 'priv.json' + pass_file = pki_dir / 'pass_file' + root_crt = step_config_dir / 'certs' / 'root_ca.crt' + try: + token = subprocess.check_output( + f'{step_path} ca token {name} ' + f'--key {priv_json} --root {root_crt} ' + f'--password-file {pass_file} 'f'--ca-url {ca_url}', shell=True) + except subprocess.CalledProcessError as exc: + logger.error(f'Error code {exc.returncode}: {exc.output}') + return + + token = token.strip() + token_b64 = base64.b64encode(token) + + with open(step_config_dir / 'certs' / 'root_ca.crt', mode='rb') as file: + root_ca = file.read() + root_ca_b64 = base64.b64encode(root_ca) + + return TOKEN_DELIMITER.join([ + token_b64.decode('utf-8'), + root_ca_b64.decode('utf-8'), + ]) + + +def get_ca_bin_paths(ca_path): + """Get paths of step binaries.""" + ca_path = Path(ca_path) + step = None + step_ca = None + if (ca_path / 'step').exists(): + dirs = os.listdir(ca_path / 'step') + for dir_ in dirs: + if 'step_' in dir_: + step = ca_path / 'step' / dir_ / 'bin' / 'step' + if 'step-ca' in dir_: + step_ca = ca_path / 'step' / dir_ / 'bin' / 'step-ca' + return step, step_ca + + +def certify(name, cert_path: Path, token_with_cert, ca_path: Path): + """Create an envoy workspace.""" + os.makedirs(cert_path, exist_ok=True) + + token, root_ca = token_with_cert.split(TOKEN_DELIMITER) + token = base64.b64decode(token).decode('utf-8') + root_ca = base64.b64decode(root_ca) + + step_path, _ = get_ca_bin_paths(ca_path) + if not step_path: + url = 'http://api.github.com/repos/smallstep/cli/releases/latest' + download_step_bin(url, 'step_linux', 'amd', prefix=ca_path) + step_path, _ = get_ca_bin_paths(ca_path) + if not step_path: + raise Exception('Step-CA is not installed!\nRun `fx pki install` first') + + with open(f'{cert_path}/root_ca.crt', mode='wb') as file: + file.write(root_ca) + call(f'./{step_path} ca certificate {name} {cert_path}/{name}.crt ' + f'{cert_path}/{name}.key -f --token {token}', shell=True) + + +def remove_ca(ca_path): + """Kill step-ca process and rm ca directory.""" + _check_kill_process('step-ca') + shutil.rmtree(ca_path, ignore_errors=True) + + +def install(ca_path, ca_url, password): + """ + Create certificate authority for federation. + + Args: + ca_path: path to ca directory + ca_url: url for ca server like: 'host:port' + password: Simple password for encrypting root private keys + + """ + logger.info('Creating CA') + + ca_path = Path(ca_path) + ca_path.mkdir(parents=True, exist_ok=True) + step_config_dir = ca_path / 'step_config' + os.environ['STEPPATH'] = str(step_config_dir) + step_path, step_ca_path = get_ca_bin_paths(ca_path) + + if not (step_path and step_ca_path and step_path.exists() and step_ca_path.exists()): + confirm('CA binaries from github will be downloaded now', default=True, abort=True) + url = 'http://api.github.com/repos/smallstep/certificates/releases/latest' + download_step_bin(url, 'step-ca_linux', 'amd', prefix=ca_path, confirmation=False) + url = 'http://api.github.com/repos/smallstep/cli/releases/latest' + download_step_bin(url, 'step_linux', 'amd', prefix=ca_path, confirmation=False) + step_config_dir = ca_path / 'step_config' + if (not step_config_dir.exists() + or confirm('CA exists, do you want to recreate it?', default=True)): + _create_ca(ca_path, ca_url, password) + _configure(step_config_dir) + + +def run_ca(step_ca, pass_file, ca_json): + """Run CA server.""" + if _check_kill_process('step-ca', confirmation=True): + logger.info('Up CA server') + call(f'{step_ca} --password-file {pass_file} {ca_json}', shell=True) + + +def _check_kill_process(pstring, confirmation=False): + """Kill process by name.""" + pids = [] + proc = subprocess.Popen(f'ps ax \| grep {pstring} \| grep -v grep', + shell=True, stdout=subprocess.PIPE) + text = proc.communicate()[0].decode('utf-8') + + for line in text.splitlines(): + fields = line.split() + pids.append(fields[0]) + + if len(pids): + if confirmation and not confirm('CA server is already running. Stop him?', default=True): + return False + for pid in pids: + os.kill(int(pid), signal.SIGKILL) + time.sleep(2) + return True + + +def _create_ca(ca_path: Path, ca_url: str, password: str): + """Create a ca workspace.""" + pki_dir = ca_path / 'cert' + step_config_dir = ca_path / 'step_config' + + pki_dir.mkdir(parents=True, exist_ok=True) + step_config_dir.mkdir(parents=True, exist_ok=True) + + with open(f'{pki_dir}/pass_file', 'w') as f: + f.write(password) + step_path, step_ca_path = get_ca_bin_paths(ca_path) + assert(step_path and step_ca_path and step_path.exists() and step_ca_path.exists()) + + logger.info('Create CA Config') + os.environ['STEPPATH'] = str(step_config_dir) + shutil.rmtree(step_config_dir, ignore_errors=True) + name = ca_url.split(':')[0] + call(f'{step_path} ca init --name name --dns {name} ' + f'--address {ca_url} --provisioner prov ' + f'--password-file {pki_dir}/pass_file', shell=True) + + call(f'{step_path} ca provisioner remove prov --all', shell=True) + call(f'{step_path} crypto jwk create {step_config_dir}/certs/pub.json ' + f'{step_config_dir}/secrets/priv.json --password-file={pki_dir}/pass_file', shell=True) + call( + f'{step_path} ca provisioner add provisioner {step_config_dir}/certs/pub.json', + shell=True + ) + + +def _configure(step_config_dir): + conf_file = step_config_dir / 'config' / 'ca.json' + with open(conf_file, 'r+') as f: + data = json.load(f) + data.setdefault('authority', {}).setdefault('claims', {}) + data['authority']['claims']['maxTLSCertDuration'] = f'{365 * 24}h' + data['authority']['claims']['defaultTLSCertDuration'] = f'{365 * 24}h' + data['authority']['claims']['maxUserSSHCertDuration'] = '24h' + data['authority']['claims']['defaultUserSSHCertDuration'] = '24h' + f.seek(0) + json.dump(data, f, indent=4) + f.truncate() diff --git a/openfl/component/collaborator/collaborator.py b/openfl/component/collaborator/collaborator.py index 840ae54331..b39ce9d09f 100644 --- a/openfl/component/collaborator/collaborator.py +++ b/openfl/component/collaborator/collaborator.py @@ -363,9 +363,9 @@ def send_task_results(self, tensor_dict, round_number, task_name): tensor_name, origin, fl_round, report, tags = tensor if report: - self.logger.info( - f'Sending metric for task {task_name},' - f' round number {round_number}:' + self.logger.metric( + f'Round {round_number}, collaborator {self.collaborator_name} ' + f'is sending metric for task {task_name}:' f' {tensor_name}\t{tensor_dict[tensor]}') self.client.send_local_task_results( diff --git a/openfl/component/director/__init__.py b/openfl/component/director/__init__.py new file mode 100644 index 0000000000..580285bb0e --- /dev/null +++ b/openfl/component/director/__init__.py @@ -0,0 +1,11 @@ +# Copyright (C) 2020-2021 Intel Corporation +# SPDX-License-Identifier: Apache-2.0 + +"""Director package.""" + +from .director import Director + + +__all__ = [ + 'Director', +] diff --git a/openfl/component/director/director.py b/openfl/component/director/director.py new file mode 100644 index 0000000000..81506cf04a --- /dev/null +++ b/openfl/component/director/director.py @@ -0,0 +1,235 @@ +# Copyright (C) 2020-2021 Intel Corporation +# SPDX-License-Identifier: Apache-2.0 + +"""Director module.""" + +import asyncio +import logging +import os +import shutil +import time +import typing +from collections import defaultdict +from pathlib import Path + +from openfl.federated import Plan +from openfl.protocols import director_pb2 + +logger = logging.getLogger(__name__) + + +class Director: + """Director class.""" + + def __init__(self, , disable_tls: bool = False, + root_ca: Path = None, key: Path = None, cert: Path = None, + sample_shape: list = None, target_shape: list = None) -> None: + """Initialize a director object.""" + # TODO: add working directory + super().__init__() + self.sample_shape, self.target_shape = sample_shape, target_shape + self._shard_registry = {} + self.col_exp_queues = defaultdict(asyncio.Queue) + + self.experiment_data = {} # {Experiment name : archive bytes} + # What if two experiments come with the same name from different users? + self.experiments_queue = asyncio.Queue() # experiments waiting to be executed + self.experiment_stash = defaultdict(dict) # Running of finished experiments + # {API name : {experiment name : aggregator}} + + self.disable_tls = disable_tls + self.root_ca = root_ca + self.key = key + self.cert = cert + + def acknowledge_shard(self, shard_info: director_pb2.ShardInfo) -> bool: + """Save shard info to shard registry if it's acceptable.""" + is_accepted = False + if (self.sample_shape != shard_info.sample_shape + or self.target_shape != shard_info.target_shape): + logger.info('Request was not accepted') + return is_accepted + logger.info('Request was accepted') + self._shard_registry[shard_info.node_info.name] = { + 'shard_info': shard_info, + 'is_online': True, + 'is_experiment_running': False + } + is_accepted = True + return is_accepted + + async def set_new_experiment(self, , experiment_name: str, sender_name: str, + tensor_dict: dict, + collaborator_names: typing.Iterable[str], data: bytes) -> bool: + """Set new experiment.""" + # TODO: save to file + self.experiment_data[experiment_name] = data + + self.create_workspace(experiment_name, data) + asyncio.create_task(self._run_aggregator( + sender_name, + tensor_dict, + experiment_name, + collaborator_names + )) + + logger.info(f'New experiment {experiment_name} for ' + f'collaborators {collaborator_names}') + for col_name in collaborator_names: + queue = self.col_exp_queues[col_name] + await queue.put(experiment_name) + + return True + + async def get_trained_model(self, experiment_name: str, caller: str): + """Get trained models.""" + if experiment_name not in self.experiment_stash[caller]: + logger.error('Aggregator has not started yet') + return None, None + + aggregator = self.experiment_stash[caller][experiment_name].aggregator + + if aggregator.last_tensor_dict is None: + logger.error('Aggregator have no aggregated model to return') + return None, None + + return aggregator.best_tensor_dict, aggregator.last_tensor_dict + + def get_experiment_data(self, experiment_name: str) -> bytes: + """Get experiment data.""" + return self.experiment_data.get(experiment_name, b'') + + async def wait_experiment(self, collaborator_name: str) -> str: + """Wait an experiment.""" + queue = self.col_exp_queues[collaborator_name] + experiment_name = await queue.get() + + return experiment_name + + def get_dataset_info(self): + """Get dataset info.""" + return self.sample_shape, self.target_shape + + def get_registered_shards(self) -> list: + """Get registered shard infos.""" + return [shard_status['shard_info'] for shard_status in self._shard_registry.values()] + + async def stream_metrics(self, experiment_name: str, caller: str): + """Stream metrics from the aggregator.""" + aggregator = self.experiment_stash[caller][experiment_name].aggregator + + while not aggregator.all_quit_jobs_sent() or not aggregator.metric_queue.empty(): + # If the aggregator has not fineished the experimnet + # or it finished but metric_queue is not empty we send metrics + + # But here we may have a problem if the new experiment starts too quickly + + while not aggregator.metric_queue.empty(): + metric_origin, task_name, metric_name, metric_value, round_ = \ + aggregator.metric_queue.get() + yield director_pb2.StreamMetricsResponse( + metric_origin=metric_origin, + task_name=task_name, + metric_name=metric_name, + metric_value=float(metric_value), + round=round_) + + # Awaiting quit job sent to collaborators + await asyncio.sleep(5) + + def remove_experiment_data(self, name: str, caller: str): + """Remove experiment data RPC.""" + if name in self.experiment_stash.get(caller, {}): + del self.experiment_stash[caller][name] + + def collaborator_health_check(self, , collaborator_name: str, + is_experiment_running: bool, + valid_duration: int) -> bool: + """Accept health check from envoy.""" + is_accepted = False + shard_info = self._shard_registry.get(collaborator_name) + if not shard_info: + logger.error(f'Unknown shard {collaborator_name}') + return is_accepted + is_accepted = True + + shard_info['is_online']: True + shard_info['is_experiment_running'] = is_experiment_running + shard_info['valid_duration'] = valid_duration + shard_info['last_updated'] = time.time() + + return is_accepted + + def get_envoys(self) -> list: + """Get a status information about envoys.""" + logger.info(f'Shard registry: {self._shard_registry}') + envoy_infos = [] + for envoy in self._shard_registry.values(): + envoy_info = director_pb2.EnvoyInfo( + shard_info=envoy['shard_info'], + is_online=time.time() < envoy['last_updated'] + envoy['valid_duration'], + is_experiment_running=envoy['is_experiment_running'] + ) + envoy_info.valid_duration.seconds = envoy['valid_duration'] + envoy_info.last_updated.seconds = int(envoy['last_updated']) + + envoy_infos.append(envoy_info) + + return envoy_infos + + @staticmethod + def create_workspace(experiment_name, npbytes): + """Create the aggregator workspace.""" + if os.path.exists(experiment_name): + shutil.rmtree(experiment_name) + os.makedirs(experiment_name) + + arch_name = f'{experiment_name}/{experiment_name}' + '.zip' + logger.info(f'arch_name: {arch_name}') + with open(arch_name, 'wb') as content_file: + content_file.write(npbytes) + + shutil.unpack_archive(arch_name, experiment_name) + + async def _run_aggregator( + self, + experiment_sender, + initial_tensor_dict, + experiment_name, + collaborator_names, + plan='plan/plan.yaml', + ): # TODO: path params, change naming + """Run aggregator.""" + cwd = os.getcwd() + os.chdir(f'{cwd}/{experiment_name}') + plan = Plan.parse(plan_config_path=Path(plan)) + + plan.authorized_cols = list(collaborator_names) + + logger.info('🧿 Starting the Aggregator Service.') + aggregator_server = plan.interactive_api_get_server( + tensor_dict=initial_tensor_dict, + chain=self.root_ca, + certificate=self.cert, + private_key=self.key, + disable_tls=self.disable_tls, + ) + self.experiment_stash[experiment_sender][experiment_name] = \ + aggregator_server + + grpc_server = aggregator_server.get_server() + grpc_server.start() + logger.info('Starting Aggregator gRPC Server') + + try: + while not aggregator_server.aggregator.all_quit_jobs_sent(): + # Awaiting quit job sent to collaborators + await asyncio.sleep(10) + except KeyboardInterrupt: + pass + finally: + os.chdir(cwd) + shutil.rmtree(experiment_name) + grpc_server.stop(0) + # Temporary solution to free RAM used by TensorDB + aggregator_server.aggregator.tensor_db.clean_up(0) diff --git a/openfl/component/envoy/__init__.py b/openfl/component/envoy/__init__.py new file mode 100644 index 0000000000..17fe14353f --- /dev/null +++ b/openfl/component/envoy/__init__.py @@ -0,0 +1,4 @@ +# Copyright (C) 2020-2021 Intel Corporation +# SPDX-License-Identifier: Apache-2.0 + +"""Envoy package.""" diff --git a/openfl/component/envoy/envoy.py b/openfl/component/envoy/envoy.py new file mode 100644 index 0000000000..4ae1d8e85c --- /dev/null +++ b/openfl/component/envoy/envoy.py @@ -0,0 +1,112 @@ +# Copyright (C) 2020-2021 Intel Corporation +# SPDX-License-Identifier: Apache-2.0 + +"""Envoy module.""" + +import logging +import os +import sys +import time +from concurrent.futures import ThreadPoolExecutor +from pathlib import Path + +from click import echo + +from openfl.federated import Plan +from openfl.transport.grpc.director_client import ShardDirectorClient + +logger = logging.getLogger(__name__) + + +DEFAULT_TIMEOUT_IN_SECONDS = 60 # TODO: make configurable +DEFAULT_RETRY_TIMEOUT_IN_SECONDS = 5 + + +class Envoy: + """Envoy class.""" + + def __init__(self, shard_name, director_uri, shard_descriptor, + root_ca: str = None, key: str = None, cert: str = None, + disable_tls: bool = False) -> None: + """Initialize a envoy object.""" + self.name = shard_name + self.root_ca = Path(root_ca).absolute() if root_ca is not None else None + self.key = Path(key).absolute() if root_ca is not None else None + self.cert = Path(cert).absolute() if root_ca is not None else None + self.director_client = ShardDirectorClient(director_uri, shard_name=shard_name, + disable_tls=disable_tls, + root_ca=root_ca, key=key, cert=cert) + self.shard_descriptor = shard_descriptor + self.executor = ThreadPoolExecutor() + self.running_experiments = {} + self.is_experiment_running = False + self._health_check_future = None + + def run(self): + """Run of the envoy working cycle.""" + while True: + try: + # Workspace import should not be done by gRPC client! + experiment_name = self.director_client.get_experiment_data() + except Exception as exc: + logger.error(f'Failed to get experiment: {exc}') + time.sleep(DEFAULT_RETRY_TIMEOUT_IN_SECONDS) + continue + self.is_experiment_running = True + try: + self._run_collaborator(experiment_name) + except Exception as exc: + logger.error(f'Collaborator failed: {exc}') + finally: + # Workspace cleaning should not be done by gRPC client! + self.director_client.remove_workspace(experiment_name) + self.is_experiment_running = False + + def send_health_check(self): + """Send health check to the director.""" + logger.info('The health check sender is started.') + while True: + self.director_client.send_health_check( + self.name, + self.is_experiment_running, + DEFAULT_TIMEOUT_IN_SECONDS + ) + time.sleep(DEFAULT_TIMEOUT_IN_SECONDS / 2) + + def _run_collaborator(self, experiment_name, plan='plan/plan.yaml',): + """Run the collaborator for the experiment running.""" + cwd = os.getcwd() + os.chdir(f'{cwd}/{experiment_name}') # TODO: probably it should be another way + + # This is needed for python module finder + sys.path.append(os.getcwd()) + + plan = Plan.parse(plan_config_path=Path(plan)) + + # TODO: Need to restructure data loader config file loader + echo(f'Data = {plan.cols_data_paths}') + logger.info('🧿 Starting a Collaborator Service.') + + col = plan.get_collaborator(self.name, self.root_ca, self.key, + self.cert, shard_descriptor=self.shard_descriptor) + try: + col.run() + finally: + os.chdir(cwd) + + def start(self): + """Start the envoy.""" + try: + is_accepted = self.director_client.report_shard_info(self.shard_descriptor) + except Exception as exc: + logger.exception(str(exc)) + logger.exception('Failed to report shard info') + else: + if is_accepted: + # Shard accepted for participation in the federation + logger.info('Shard accepted') + self._health_check_future = self.executor.submit(self.send_health_check) + self.run() + else: + # Shut down + logger.error('Report shard info was not accepted') diff --git a/openfl/federated/plan/plan.py b/openfl/federated/plan/plan.py index e78585ad4e..e2c1b0a454 100644 --- a/openfl/federated/plan/plan.py +++ b/openfl/federated/plan/plan.py @@ -182,6 +182,27 @@ def build(template, settings, override): return instance + @staticmethod + def import_(template): + """ + Import an instance of a openfl Component or Federated DataLoader/TaskRunner. + + Args: + template: Fully qualified object path + + Returns: + A Python object + """ + class_name = splitext(template)[1].strip('.') + module_path = splitext(template)[0] + Plan.logger.info(f'Importing [red]🡆[/] Object [red]{class_name}[/] ' + f'from [red]{module_path}[/] Module.', + extra={'markup': True}) + module = import_module(module_path) + instance = getattr(module, class_name) + + return instance + def __init__(self): """Initialize.""" self.config = {} # dictionary containing patched plan definition @@ -204,7 +225,7 @@ def __init__(self): self.name_ = None @property - def hash(self): # NOQA + def hash(self): # NOQA """Generate hash for this instance.""" self.hash_ = sha384(dump(self.config).encode('utf-8')) Plan.logger.info(f'FL-Plan hash is [blue]{self.hash_.hexdigest()}[/]', @@ -258,7 +279,16 @@ def get_aggregator(self, tensor_dict=None): defaults[SETTINGS]['authorized_cols'] = self.authorized_cols defaults[SETTINGS]['assigner'] = self.get_assigner() defaults[SETTINGS]['compression_pipeline'] = self.get_tensor_pipe() + log_metric_callback = defaults[SETTINGS].get('log_metric_callback') + + if log_metric_callback: + if isinstance(log_metric_callback, dict): + log_metric_callback = Plan.import_(log_metric_callback) + elif not callable(log_metric_callback): + raise TypeError(f'log_metric_callback should be callable object ' + f'or be import from code part, get {log_metric_callback}') + defaults[SETTINGS]['log_metric_callback'] = log_metric_callback if self.aggregator_ is None: self.aggregator_ = Plan.build(defaults, initial_tensor_dict=tensor_dict) @@ -318,12 +348,9 @@ def get_data_loader(self, collaborator_name): return self.loader_ # Python interactive api - def initialize_data_loader(self, data_loader, collaborator_name): + def initialize_data_loader(self, data_loader, shard_descriptor): """Get data loader.""" - data_path = self.cols_data_paths[ - collaborator_name - ] - data_loader._delayed_init(data_path=data_path) + data_loader.shard_descriptor = shard_descriptor return data_loader # legacy api (TaskRunner subclassing) @@ -372,8 +399,8 @@ def get_core_task_runner(self, data_loader=None, return self.runner_ - def get_collaborator(self, collaborator_name, - task_runner=None, client=None): + def get_collaborator(self, collaborator_name, root_ca=None, key=None, cert=None, + task_runner=None, client=None, shard_descriptor=None): """Get collaborator.""" defaults = self.config.get( 'collaborator', @@ -398,7 +425,7 @@ def get_collaborator(self, collaborator_name, if 'openfl.federated.task.task_runner' in self.config['task_runner']['template']: # Interactive API model_provider, task_keeper, data_loader = self.deserialize_interface_objects() - data_loader = self.initialize_data_loader(data_loader, collaborator_name) + data_loader = self.initialize_data_loader(data_loader, shard_descriptor) defaults[SETTINGS]['task_runner'] = self.get_core_task_runner( data_loader=data_loader, model_provider=model_provider, @@ -416,7 +443,10 @@ def get_collaborator(self, collaborator_name, defaults[SETTINGS]['client'] = self.get_client( collaborator_name, self.aggregator_uuid, - self.federation_uuid + self.federation_uuid, + root_ca, + key, + cert ) if self.collaborator_ is None: @@ -424,13 +454,18 @@ def get_collaborator(self, collaborator_name, return self.collaborator_ - def get_client(self, collaborator_name, aggregator_uuid, federation_uuid): + def get_client(self, collaborator_name, aggregator_uuid, federation_uuid, + root_ca=None, key=None, cert=None): """Get gRPC client for the specified collaborator.""" common_name = collaborator_name - - chain = 'cert/cert_chain.crt' - certificate = f'cert/client/col_{common_name}.crt' - private_key = f'cert/client/col_{common_name}.key' + if root_ca and key and cert: + chain = root_ca + certificate = cert + private_key = key + else: + chain = 'cert/cert_chain.crt' + certificate = f'cert/client/col_{common_name}.crt' + private_key = f'cert/client/col_{common_name}.key' client_args = self.config['network'][SETTINGS] @@ -448,13 +483,18 @@ def get_client(self, collaborator_name, aggregator_uuid, federation_uuid): return self.client_ - def get_server(self): + def get_server(self, root_ca=None, key=None, cert=None, kwargs): """Get gRPC server of the aggregator instance.""" common_name = self.config['network'][SETTINGS]['agg_addr'].lower() - chain = 'cert/cert_chain.crt' - certificate = f'cert/server/agg_{common_name}.crt' - private_key = f'cert/server/agg_{common_name}.key' + if root_ca and key and cert: + chain = '../cert/root_ca.crt' + certificate = f'agg_{common_name}.crt' + private_key = f'agg_{common_name}.key' + else: + chain = 'cert/cert_chain.crt' + certificate = f'cert/server/agg_{common_name}.crt' + private_key = f'cert/server/agg_{common_name}.key' server_args = self.config['network'][SETTINGS] @@ -464,6 +504,8 @@ def get_server(self): server_args['certificate'] = certificate server_args['private_key'] = private_key + server_args.update(kwargs) + server_args['aggregator'] = self.get_aggregator() if self.server_ is None: @@ -471,7 +513,8 @@ def get_server(self): return self.server_ - def interactive_api_get_server(self, tensor_dict, chain, certificate, private_key): + def interactive_api_get_server(self, , tensor_dict, chain, certificate, + private_key, disable_tls): """Get gRPC server of the aggregator instance.""" server_args = self.config['network'][SETTINGS] @@ -479,6 +522,7 @@ def interactive_api_get_server(self, tensor_dict, chain, certificate, private_ke server_args['ca'] = chain server_args['certificate'] = certificate server_args['private_key'] = private_key + server_args['disable_tls'] = disable_tls server_args['aggregator'] = self.get_aggregator(tensor_dict) @@ -489,13 +533,16 @@ def interactive_api_get_server(self, tensor_dict, chain, certificate, private_ke def deserialize_interface_objects(self): """Deserialize objects for TaskRunner.""" - interface_objects = [] serializer = Plan.build( self.config['api_layer']['required_plugin_components']['serializer_plugin'], {}) - for filename in ['model_interface_file', - 'tasks_interface_file', 'dataloader_interface_file']: - interface_objects.append( - serializer.restore_object(self.config['api_layer']['settings'][filename]) - ) + filenames = [ + 'model_interface_file', + 'tasks_interface_file', + 'dataloader_interface_file' + ] + interface_objects = [ + serializer.restore_object(self.config['api_layer']['settings'][filename]) + for filename in filenames + ] model_provider, task_keeper, data_loader = interface_objects return model_provider, task_keeper, data_loader diff --git a/openfl/federated/task/task_runner.py b/openfl/federated/task/task_runner.py index a261fed5ac..755426c6e6 100644 --- a/openfl/federated/task/task_runner.py +++ b/openfl/federated/task/task_runner.py @@ -72,8 +72,8 @@ def _prepare_tensorkeys_for_agggregation(self, metric_dict, validation_flag, else: suffix = 'validate' + validation_flag - tags = ('metric', suffix) - + tags = (suffix,) + tags = ('metric', tags) metric_dict = { TensorKey(metric, origin, round_num, True, tags): np.array(value) for metric, value in metric_dict.items() diff --git a/openfl/interface/cli.py b/openfl/interface/cli.py index e094f7689e..434c2fde49 100755 --- a/openfl/interface/cli.py +++ b/openfl/interface/cli.py @@ -12,39 +12,36 @@ from click import pass_context from click import style +from openfl.utilities import add_log_level -def setup_logging(level='info'): + +def setup_logging(level='info', log_file=None): """Initialize logging settings.""" + import logging from logging import basicConfig - from logging import CRITICAL - from logging import DEBUG - from logging import ERROR - from logging import INFO - from logging import NOTSET - from logging import WARNING + from rich.console import Console from rich.logging import RichHandler - import os - - console = Console(width=160) - levels = \ - { - 'notset': NOTSET, - 'debug': DEBUG, - 'info': INFO, - 'warning': WARNING, - 'error': ERROR, - 'critical': CRITICAL - } + metric = 25 + add_log_level('METRIC', metric) - if level.lower() in ['debug', 'error']: - os.environ['GRPC_VERBOSITY'] = level.upper() + if isinstance(level, str): + level = level.upper() - level = levels.get(level.lower(), levels['notset']) + handlers = [] + if log_file: + fh = logging.FileHandler(log_file) + formatter = logging.Formatter( + '%(asctime)s %(levelname)s %(message)s %(filename)s:%(lineno)d' + ) + fh.setFormatter(formatter) + handlers.append(fh) + console = Console(width=160) + handlers.append(RichHandler(console=console)) basicConfig(level=level, format='%(message)s', - datefmt='[%X]', handlers=[RichHandler(console=console)]) + datefmt='[%X]', handlers=handlers) def disable_warnings(): @@ -129,6 +126,7 @@ def format_help(self, ctx, formatter): @pass_context def cli(context, log_level): """Command-line Interface.""" + import os from sys import argv context.ensure_object(dict) @@ -137,10 +135,11 @@ def cli(context, log_level): context.obj['script'] = argv[0] context.obj['arguments'] = argv[1:] - setup_logging(log_level) + log_file = os.getenv('LOG_FILE') + setup_logging(log_level, log_file) [email protected]() [email protected]_callback() @pass_context def end(context, result, kwargs): """Print the result of the operation.""" @@ -190,7 +189,6 @@ def entry(): file = Path(__file__).resolve() root = file.parent.resolve() # interface root, containing command modules work = Path.cwd().resolve() - path.append(str(root)) path.insert(0, str(work)) diff --git a/openfl/interface/director.py b/openfl/interface/director.py new file mode 100644 index 0000000000..fb294ad4f4 --- /dev/null +++ b/openfl/interface/director.py @@ -0,0 +1,86 @@ +# Copyright (C) 2020-2021 Intel Corporation +# SPDX-License-Identifier: Apache-2.0 +"""Director CLI.""" + +import logging +import shutil +import sys +from pathlib import Path + +import click +from click import group +from click import option +from click import pass_context +from click import Path as ClickPath +from yaml import safe_load + +from openfl.component.director import Director +from openfl.interface.cli_helper import WORKSPACE +from openfl.transport import DirectorGRPCServer + +logger = logging.getLogger(__name__) + + +@group() +@pass_context +def director(context): + """Manage Federated Learning Director.""" + context.obj['group'] = 'director' + + [email protected](name='start') +@option('-c', '--director-config-path', default='director.yaml', + help='The director config file path', type=ClickPath(exists=True)) +@option('--disable-tls', default=False, + is_flag=True) +@option('-rc', '--root-cert-path', 'root_ca', default=None, + help='Path to a root CA cert') +@option('-pk', '--private-key-path', 'key', default=None, + help='Path to a private key') +@option('-oc', '--public-cert-path', 'cert', default=None, + help='Path to a signed certificate') +def start(director_config_path, disable_tls, root_ca, key, cert): + """Start the director service.""" + logger.info('🧿 Starting the Director Service.') + with open(director_config_path) as stream: + director_config = safe_load(stream) + settings = director_config.get('settings', {}) + sample_shape = settings.get('sample_shape', '') + target_shape = settings.get('target_shape', '') + logger.info(f'Sample shape: {sample_shape}, target shape: {target_shape}') + listen_ip = settings.get('listen_ip') + listen_port = settings.get('listen_port') + root_ca = root_ca or settings.get('root_ca') + key = key or settings.get('key') + cert = cert or settings.get('cert') + kwargs = {} + if listen_ip: + kwargs['listen_ip'] = listen_ip + if listen_port: + kwargs['listen_port'] = listen_port + director_server = DirectorGRPCServer( + director_cls=Director, + disable_tls=disable_tls, + sample_shape=sample_shape, + target_shape=target_shape, + root_ca=root_ca, + key=key, + cert=cert, + kwargs + ) + director_server.start() + + [email protected](name='create-workspace') +@option('-p', '--director-path', required=True, + help='The director path', type=ClickPath()) +def create(director_path): + """Create a director workspace.""" + director_path = Path(director_path) + if director_path.exists(): + if not click.confirm('Director workspace already exists. Recreate?', default=True): + sys.exit(1) + shutil.rmtree(director_path) + (director_path / 'cert').mkdir(parents=True, exist_ok=True) + (director_path / 'logs').mkdir(parents=True, exist_ok=True) + shutil.copyfile(WORKSPACE / 'default/director.yaml', director_path / 'director.yaml') diff --git a/openfl/interface/envoy.py b/openfl/interface/envoy.py new file mode 100644 index 0000000000..abff25356c --- /dev/null +++ b/openfl/interface/envoy.py @@ -0,0 +1,102 @@ +# Copyright (C) 2020-2021 Intel Corporation +# SPDX-License-Identifier: Apache-2.0 +"""Envoy CLI.""" + +import logging +import shutil +import sys +from importlib import import_module +from pathlib import Path + +import click +from click import group +from click import option +from click import pass_context +from click import Path as ClickPath +from yaml import safe_load + +from openfl.component.envoy.envoy import Envoy +from openfl.interface.cli_helper import WORKSPACE + +logger = logging.getLogger(__name__) + + +@group() +@pass_context +def envoy(context): + """Manage Federated Learning Envoy.""" + context.obj['group'] = 'envoy' + + [email protected](name='start') +@option('-n', '--shard-name', required=True, + help='Current shard name') +@option('-d', '--director-uri', required=True, + help='The FQDN of the federation director') +@option('--disable-tls', default=False, + is_flag=True) +@option('-sc', '--shard-config-path', default='shard_config.yaml', + help='The shard config path', type=ClickPath(exists=True)) +@option('-rc', '--root-cert-path', 'root_ca', default=None, + help='Path to a root CA cert') +@option('-pk', '--private-key-path', 'key', default=None, + help='Path to a private key') +@option('-oc', '--public-cert-path', 'cert', default=None, + help='Path to a signed certificate') +def start_(shard_name, director_uri, disable_tls, shard_config_path, + root_ca, key, cert): + """Start the Envoy.""" + logger.info('🧿 Starting the Envoy.') + + shard_descriptor = shard_descriptor_from_config(shard_config_path) + envoy = Envoy( + shard_name=shard_name, + director_uri=director_uri, + shard_descriptor=shard_descriptor, + disable_tls=disable_tls, + root_ca=root_ca, + key=key, + cert=cert + ) + + envoy.start() + + [email protected](name='create-workspace') +@option('-p', '--envoy-path', required=True, + help='The Envoy path', type=ClickPath()) +def create(envoy_path): + """Create an envoy workspace.""" + envoy_path = Path(envoy_path) + if envoy_path.exists(): + if not click.confirm('Envoy workspace already exists. Recreate?', + default=True): + sys.exit(1) + shutil.rmtree(envoy_path) + (envoy_path / 'cert').mkdir(parents=True, exist_ok=True) + (envoy_path / 'logs').mkdir(parents=True, exist_ok=True) + (envoy_path / 'data').mkdir(parents=True, exist_ok=True) + shutil.copyfile(WORKSPACE / 'default/shard_config.yaml', + envoy_path / 'shard_config.yaml') + shutil.copyfile(WORKSPACE / 'default/shard_descriptor.py', + envoy_path / 'shard_descriptor.py') + shutil.copyfile(WORKSPACE / 'default/requirements.txt', + envoy_path / 'requirements.txt') + + +def shard_descriptor_from_config(shard_config_path: str): + """Build a shard descriptor from config.""" + with open(shard_config_path) as stream: + shard_config = safe_load(stream) + template = shard_config.get('template') + if not template: + raise Exception(f'You should define a shard ' + f'descriptor template in {shard_config_path}') + class_name = template.split('.')[-1] + module_path = '.'.join(template.split('.')[:-1]) + params = shard_config.get('params', {}) + + module = import_module(module_path) + instance = getattr(module, class_name)(params) + + return instance diff --git a/openfl/interface/interactive_api/experiment.py b/openfl/interface/interactive_api/experiment.py index 0899ac7af7..3baabcd2db 100644 --- a/openfl/interface/interactive_api/experiment.py +++ b/openfl/interface/interactive_api/experiment.py @@ -3,14 +3,17 @@ """Python low-level API module.""" import functools -import logging import os +import time from collections import defaultdict from copy import deepcopy from logging import getLogger from pathlib import Path +from tensorboardX import SummaryWriter + from openfl.federated import Plan +from openfl.interface.cli import setup_logging from openfl.interface.cli_helper import WORKSPACE from openfl.utilities import split_tensor_dict_for_holdouts @@ -18,7 +21,13 @@ class FLExperiment: """Central class for FL experiment orchestration.""" - def __init__(self, federation, serializer_plugin=None) -> None: + def __init__( + self, + federation, + experiment_name: str = None, + serializer_plugin: str = 'openfl.plugins.interface_serializer.' + 'cloudpickle_serializer.CloudpickleSerializer' + ) -> None: """ Initialize an experiment inside a federation. @@ -26,22 +35,90 @@ def __init__(self, federation, serializer_plugin=None) -> None: Information about the data on collaborators is contained on the federation level. """ self.federation = federation + self.experiment_name = experiment_name or 'test-' + time.strftime('%Y%m%d-%H%M%S') + self.summary_writer = None + self.serializer_plugin = serializer_plugin - if serializer_plugin is None: - self.serializer_plugin = \ - 'openfl.plugins.interface_serializer.cloudpickle_serializer.CloudpickleSerializer' - else: - self.serializer_plugin = serializer_plugin + self.experiment_accepted = False self.logger = getLogger(__name__) + setup_logging() + + def _assert_experiment_accepted(self): + """Assure experiment is sent to director.""" + if not self.experiment_accepted: + self.logger.error('The experimnet has not been accepted by director') + self.logger.error( + 'Report the experiment first: ' + 'use the Experiment.start() method.') + raise Exception def get_best_model(self): """Retrieve the model with the best score.""" - # Next line relies on aggregator inner field where model dicts are stored - best_tensor_dict = self.server.aggregator.best_tensor_dict - self.task_runner_stub.rebuild_model(best_tensor_dict, validation=True, device='cpu') + self._assert_experiment_accepted() + tensor_dict = self.federation.dir_client.get_best_model( + experiment_name=self.experiment_name) + + return self._rebuild_model(tensor_dict) + + def get_last_model(self): + """Retrieve the aggregated model after the last round.""" + self._assert_experiment_accepted() + tensor_dict = self.federation.dir_client.get_last_model( + experiment_name=self.experiment_name) + + return self._rebuild_model(tensor_dict) + + def _rebuild_model(self, tensor_dict): + """Use tensor dict to update model weights.""" + if len(tensor_dict) == 0: + self.logger.error('No tensors received from director') + self.logger.error( + 'Possible reasons:\n' + '1. Aggregated model is not ready \n' + '2. Experiment data removed from director' + ) + else: + self.task_runner_stub.rebuild_model(tensor_dict, validation=True, device='cpu') + return self.task_runner_stub.model + def stream_metrics(self, tensorboard_logs: bool = True) -> None: + """Stream metrics.""" + self._assert_experiment_accepted() + for metric_message_dict in self.federation.dir_client.stream_metrics(self.experiment_name): + self.logger.metric( + f'Round {metric_message_dict["round"]}, ' + f'collaborator {metric_message_dict["metric_origin"]} ' + f'{metric_message_dict["task_name"]} result ' + f'{metric_message_dict["metric_name"]}:\t{metric_message_dict["metric_value"]}') + + if tensorboard_logs: + self.write_tensorboard_metric(metric_message_dict) + + def write_tensorboard_metric(self, metric: dict) -> None: + """Write metric callback.""" + if not self.summary_writer: + self.summary_writer = SummaryWriter(f'./logs/{self.experiment_name}', flush_secs=5) + + self.summary_writer.add_scalar( + f'{metric["metric_origin"]}/{metric["task_name"]}/{metric["metric_name"]}', + metric['metric_value'], metric['round']) + + def remove_experiment_data(self): + """Remove experiment data.""" + self._assert_experiment_accepted() + log_message = 'Removing experiment data ' + if self.federation.dir_client.remove_experiment_data( + name=self.experiment_name + ): + log_message += 'succeed.' + self.experiment_accepted = False + else: + log_message += 'failed.' + + self.logger.info(log_message) + def prepare_workspace_distribution( self, model_provider, task_keeper, data_loader, rounds_to_train, @@ -64,41 +141,55 @@ def prepare_workspace_distribution( self._export_python_env() # Compress te workspace to restore it on collaborator - self._pack_the_workspace() + self.arch_path = self._pack_the_workspace() # DO CERTIFICATES exchange - def start_experiment(self, model_provider): - """ - Start the aggregator. - - This method requires model_provider to start an experiment with another - model initialization without workspace redistribution. - """ - # Start the aggregator + def start(self, , model_provider, task_keeper, data_loader, + rounds_to_train, delta_updates=False, opt_treatment='RESET'): + """Prepare experiment and run.""" + self.prepare_workspace_distribution( + model_provider, task_keeper, data_loader, rounds_to_train, + delta_updates=delta_updates, opt_treatment=opt_treatment + ) + self.logger.info('Starting experiment!') self.plan.resolve() - initial_tensor_dict = self._get_initial_tensor_dict(model_provider) - self.server = self.plan.interactive_api_get_server( - initial_tensor_dict, - chain=self.federation.cert_chain, - certificate=self.federation.agg_certificate, - private_key=self.federation.agg_private_key) + try: + response = self.federation.dir_client.set_new_experiment( + name=self.experiment_name, + col_names=self.plan.authorized_cols, + arch_path=self.arch_path, + initial_tensor_dict=initial_tensor_dict + ) + finally: + self.remove_workspace_archive() + + if response.accepted: + self.logger.info('Experiment was accepted and launched.') + self.experiment_accepted = True + else: + self.logger.info('Experiment was not accepted or failed.') - logging.basicConfig(level=logging.INFO) - self.server.serve() + def restore_experiment_state(self, model_provider): + """Restore accepted experimnet object.""" + self.task_runner_stub = self.plan.get_core_task_runner(model_provider=model_provider) + self.experiment_accepted = True @staticmethod def _export_python_env(): """Prepare requirements.txt.""" from pip._internal.operations import freeze requirements_generator = freeze.freeze() + + def is_package_has_version(package: str) -> bool: + return '==' in package and package != 'pkg-resources==0.0.0' + with open('./requirements.txt', 'w') as f: - for package in requirements_generator: - if '==' not in package: - # We do not export dependencies without version + for pack in requirements_generator: + if is_package_has_version(pack): continue - f.write(package + '\n') + f.write(pack + '\n') @staticmethod def _pack_the_workspace(): @@ -115,7 +206,7 @@ def _pack_the_workspace(): archive_name = basename(getcwd()) tmp_dir = 'temp_' + archive_name - makedirs(tmp_dir) + makedirs(tmp_dir, exist_ok=True) ignore = ignore_patterns( '__pycache__', 'data', 'cert', tmp_dir, '.crt', '.key', @@ -123,10 +214,17 @@ def _pack_the_workspace(): copytree('./', tmp_dir + '/workspace', ignore=ignore) - make_archive(archive_name, archive_type, tmp_dir + '/workspace') + arch_path = make_archive(archive_name, archive_type, tmp_dir + '/workspace') rmtree(tmp_dir) + return arch_path + + def remove_workspace_archive(self): + """Remove the workspace archive.""" + os.remove(self.arch_path) + del self.arch_path + def _get_initial_tensor_dict(self, model_provider): """Extract initial weights from the model.""" self.task_runner_stub = self.plan.get_core_task_runner(model_provider=model_provider) @@ -152,9 +250,20 @@ def _prepare_plan(self, model_provider, task_keeper, data_loader, # Change plan name to default one plan.name = 'plan.yaml' - plan.authorized_cols = list(self.federation.col_data_paths.keys()) + # Seems like we still need to fill authorized_cols list + # So aggregator know when to start sending tasks + # We also could change the aggregator logic so it will send tasks to aggregator + # as soon as it connects. This change should be a part of a bigger PR + # brining in fault tolerance changes + shard_registry = self.federation.get_shard_registry() + plan.authorized_cols = [ + name for name, info in shard_registry.items() if info['is_online'] + ] # Network part of the plan - plan.config['network']['settings']['agg_addr'] = self.federation.fqdn + # We keep in mind that an aggregator FQND will be the same as the directors FQDN + # We just choose a port randomly from plan hash + plan.config['network']['settings']['agg_addr'] = \ + self.federation.director_node_fqdn.split(':')[0] # We drop the port plan.config['network']['settings']['disable_tls'] = self.federation.disable_tls # Aggregator part of the plan @@ -356,18 +465,24 @@ class DataInterface: at initialization time for dataloader customization """ - def __init__(self, user_dataset_class, kwargs): + def __init__(self, kwargs): """Initialize DataLoader.""" - self.UserDatasetClass = user_dataset_class self.kwargs = kwargs - def _delayed_init(self, data_path): + @property + def shard_descriptor(self): + """Return shard descriptor.""" + return self._shard_descriptor + + @shard_descriptor.setter + def shard_descriptor(self, shard_descriptor): """ Describe per-collaborator procedures or sharding. This method will be called during a collaborator initialization. - data_path variable will be set according to data.yaml. + Local shard_descriptor will be set by Envoy. """ + self._shard_descriptor = shard_descriptor raise NotImplementedError def get_train_loader(self, *kwargs): diff --git a/openfl/interface/interactive_api/federation.py b/openfl/interface/interactive_api/federation.py index 863ab41de0..910c882cf1 100644 --- a/openfl/interface/interactive_api/federation.py +++ b/openfl/interface/interactive_api/federation.py @@ -2,7 +2,9 @@ # SPDX-License-Identifier: Apache-2.0 """Federation API module.""" +from openfl.transport.grpc.director_client import DirectorClient from openfl.utilities.utils import getfqdn_env +from .shard_descriptor import DummyShardDescriptor class Federation: @@ -13,37 +15,46 @@ class Federation: their local data and network setting to enable communication in federation. """ - def __init__(self, central_node_fqdn=None, disable_tls=False, - cert_chain=None, agg_certificate=None, agg_private_key=None) -> None: + def __init__(self, client_id, director_node_fqdn=None, director_port=None, disable_tls=False, + cert_chain=None, api_cert=None, api_private_key=None) -> None: """ Initialize federation. - Federation API class should be initialized with the aggregator node FQDN + Federation API class should be initialized with the Director node FQDN and encryption settings. One may disable mTLS in trusted environments or - provide paths to a certificate chain to CA, aggregator certificate and + provide paths to a certificate chain to CA, API certificate and pricate key to enable mTLS. + + Args: + - client_id: name of created Frontend API instance. + The same name user certify. + - director_node_fqdn: Address and port a director's service is running on. + User passes here an address with a port. """ - if central_node_fqdn is None: - self.fqdn = getfqdn_env() + self.client_id = client_id + if director_node_fqdn is None: + self.director_node_fqdn = getfqdn_env() else: - self.fqdn = central_node_fqdn + self.director_node_fqdn = director_node_fqdn self.disable_tls = disable_tls self.cert_chain = cert_chain - self.agg_certificate = agg_certificate - self.agg_private_key = agg_private_key + self.api_cert = api_cert + self.api_private_key = api_private_key - def register_collaborators(self, col_data_paths: dict) -> None: - """ - Provide data to be stored in data.yaml. + # Create Director client + self.dir_client = DirectorClient(client_id, f'{director_node_fqdn}:{director_port}', + disable_tls, cert_chain, api_private_key, api_cert) - This method should be used to provide information about collaborators - participating federation. - Arguments: - col_data_paths: dict(collaborator name : local data path) - """ - self.col_data_paths = col_data_paths - with open('./data.yaml', 'w') as f: - for col_name, data_path in self.col_data_paths.items(): - f.write(f'{col_name},{data_path}\n') + # Request sample and target shapes from Director. + # This is an internal method for finding out dataset properties in a Federation. + self.sample_shape, self.target_shape = self.dir_client.get_dataset_info() + + def get_dummy_shard_descriptor(self, size): + """Return a dummy shard descriptor.""" + return DummyShardDescriptor(self.sample_shape, self.target_shape, size) + + def get_shard_registry(self): + """Return a shard registry.""" + return self.dir_client.get_envoys() diff --git a/openfl/interface/interactive_api/shard_descriptor.py b/openfl/interface/interactive_api/shard_descriptor.py new file mode 100644 index 0000000000..a7413be10e --- /dev/null +++ b/openfl/interface/interactive_api/shard_descriptor.py @@ -0,0 +1,68 @@ +# Copyright (C) 2020-2021 Intel Corporation +# SPDX-License-Identifier: Apache-2.0 +"""Shard descriptor.""" + + +import numpy as np + + +class ShardDescriptor: + """Shard descriptor class.""" + + def __len__(self): + """Return the len of the dataset.""" + raise NotImplementedError + + def __getitem__(self, index: int): + """Return a item by the index.""" + raise NotImplementedError + + @property + def sample_shape(self): + """Return the sample shape info.""" + raise NotImplementedError + + @property + def target_shape(self): + """Return the target shape info.""" + raise NotImplementedError + + @property + def dataset_description(self) -> str: + """Return the dataset description.""" + return '' + + +class DummyShardDescriptor(ShardDescriptor): + """Dummy shard descriptor class.""" + + def __init__(self, sample_shape, target_shape, size) -> None: + """Initialize DummyShardDescriptor.""" + self._sample_shape = [int(dim) for dim in sample_shape] + self._target_shape = [int(dim) for dim in target_shape] + self.size = size + self.samples = np.random.randint(0, 255, (self.size, self.sample_shape), np.uint8) + self.targets = np.random.randint(0, 255, (self.size, self.target_shape), np.uint8) + + def __len__(self): + """Return the len of the dataset.""" + return self.size + + def __getitem__(self, index: int): + """Return a item by the index.""" + return self.samples[index], self.targets[index] + + @property + def sample_shape(self): + """Return the sample shape info.""" + return self._sample_shape + + @property + def target_shape(self): + """Return the target shape info.""" + return self._target_shape + + @property + def dataset_description(self) -> str: + """Return the dataset description.""" + return 'Dummy shard descriptor' diff --git a/openfl/interface/pki.py b/openfl/interface/pki.py new file mode 100644 index 0000000000..2f35f83b11 --- /dev/null +++ b/openfl/interface/pki.py @@ -0,0 +1,103 @@ +# Copyright (C) 2020-2021 Intel Corporation +# SPDX-License-Identifier: Apache-2.0 +"""PKI CLI.""" + +import logging +import os +from pathlib import Path + +from click import group +from click import option +from click import pass_context +from click import Path as ClickPath + +from openfl.component.ca.ca import certify +from openfl.component.ca.ca import get_ca_bin_paths +from openfl.component.ca.ca import get_token +from openfl.component.ca.ca import install +from openfl.component.ca.ca import remove_ca +from openfl.component.ca.ca import run_ca + + +logger = logging.getLogger(__name__) + +CA_URL = 'localhost:9123' + + +@group() +@pass_context +def pki(context): + """Manage Step-ca PKI.""" + context.obj['group'] = 'pki' + + [email protected](name='run') +@option('-p', '--ca-path', required=True, + help='The ca path', type=ClickPath()) +def run(ca_path): + """Run CA server.""" + ca_path = Path(ca_path) + step_config_dir = ca_path / 'step_config' + pki_dir = ca_path / 'cert' + pass_file = pki_dir / 'pass_file' + ca_json = step_config_dir / 'config' / 'ca.json' + _, step_ca_path = get_ca_bin_paths(ca_path) + if (not os.path.exists(step_config_dir) or not os.path.exists(pki_dir) + or not os.path.exists(pass_file) or not os.path.exists(ca_json) + or not os.path.exists(step_ca_path)): + logger.warning('CA is not installed or corrupted, please install it first') + return + run_ca(step_ca_path, pass_file, ca_json) + + [email protected](name='install') +@option('-p', '--ca-path', required=True, + help='The ca path', type=ClickPath()) +@option('--password', required=True) +@option('--ca-url', required=False, default=CA_URL) +def install_(ca_path, password, ca_url): + """Create a ca workspace.""" + install(ca_path, ca_url, password) + + [email protected](name='uninstall') +@option('-p', '--ca-path', required=True, + help='The CA path', type=ClickPath()) +def uninstall(ca_path): + """Remove step-CA.""" + remove_ca(ca_path) + + [email protected](name='get-token') +@option('-n', '--name', required=True) +@option('--ca-url', required=False, default=CA_URL) +@option('-p', '--ca-path', default='.', + help='The CA path', type=ClickPath()) +def get_token_(name, ca_url, ca_path): + """ + Create authentication token. + + Args: + name: common name for following certificate + (aggregator fqdn or collaborator name) + ca_url: full url of CA server + ca_path: the path to CA binaries + """ + token = get_token(name, ca_url, ca_path) + print('Token:') + print(token) + + [email protected](name='certify') +@option('-n', '--name', required=True) +@option('-t', '--token', 'token_with_cert', required=True) +@option('-c', '--certs-path', required=False, default=Path('.') / 'cert', + help='The path where certificates will be stored', type=ClickPath()) +@option('-p', '--ca-path', default='.', help='The path to CA client', + type=ClickPath(), required=False) +def certify_(name, token_with_cert, certs_path, ca_path): + """Create an envoy workspace.""" + certs_path = Path(certs_path) + ca_path = Path(ca_path) + certs_path.mkdir(parents=True, exist_ok=True) + certify(name, certs_path, token_with_cert, ca_path) diff --git a/openfl/native/native.py b/openfl/native/native.py index e9c810f935..885cb4bafc 100644 --- a/openfl/native/native.py +++ b/openfl/native/native.py @@ -5,6 +5,7 @@ This file defines openfl entrypoints to be used directly through python (not CLI) """ +import logging import os from copy import copy from logging import getLogger @@ -17,6 +18,7 @@ import openfl.interface.workspace as workspace from openfl.federated import Plan from openfl.protocols import utils +from openfl.utilities import add_log_level from openfl.utilities import split_tensor_dict_for_holdouts logger = getLogger(__name__) @@ -39,12 +41,13 @@ def setup_plan(log_level='CRITICAL'): cols_config = 'plan/cols.yaml' data_config = 'plan/data.yaml' + current_level = logging.root.level getLogger().setLevel(log_level) plan = Plan.parse(plan_config_path=Path(plan_config), cols_config_path=Path(cols_config), data_config_path=Path(data_config), resolve=False) - getLogger().setLevel('INFO') + getLogger().setLevel(current_level) return plan @@ -107,7 +110,7 @@ def unflatten(config, separator='.'): return config -def setup_logging(): +def setup_logging(level='INFO', log_file=None): """Initialize logging settings.""" # Setup logging from logging import basicConfig @@ -117,16 +120,29 @@ def setup_logging(): if True if pkgutil.find_loader('tensorflow') else False: import tensorflow as tf tf.compat.v1.logging.set_verbosity(tf.compat.v1.logging.ERROR) + metric = 25 + add_log_level('METRIC', metric) + + if isinstance(level, str): + level = level.upper() + + handlers = [] + if log_file: + fh = logging.FileHandler(log_file) + formatter = logging.Formatter( + '%(asctime)s %(levelname)s %(message)s %(filename)s:%(lineno)d' + ) + fh.setFormatter(formatter) + handlers.append(fh) + console = Console(width=160) - basicConfig( - level='INFO', - format='%(message)s', - datefmt='[%X]', - handlers=[RichHandler(console=console)] - ) + handlers.append(RichHandler(console=console)) + basicConfig(level=level, format='%(message)s', + datefmt='[%X]', handlers=handlers) -def init(workspace_template: str = 'default', agg_fqdn: str = None, col_names=None): +def init(workspace_template: str = 'default', log_level: str = 'INFO', + log_file: str = None, agg_fqdn: str = None, col_names=None): """ Initialize the openfl package. @@ -149,6 +165,10 @@ def init(workspace_template: str = 'default', agg_fqdn: str = None, col_names=No Other options include are any of the template names [ keras_cnn_mnist, tf_2dunet, tf_cnn_histology, mtorch_cnn_histology, torch_cnn_mnist] + log_level : str + Log level for logging. METRIC level is available + log_file : str + Name of the file in which the log will be duplicated agg_fqdn : str The local node's fully qualified domain name (if it can't be resolved automatically) @@ -174,7 +194,7 @@ def init(workspace_template: str = 'default', agg_fqdn: str = None, col_names=No collaborator.certify(col_name, silent=True) data_path += 1 - setup_logging() + setup_logging(level=log_level, log_file=log_file) def create_collaborator(plan, name, model, aggregator): diff --git a/openfl/protocols/director.proto b/openfl/protocols/director.proto new file mode 100644 index 0000000000..9d9156b873 --- /dev/null +++ b/openfl/protocols/director.proto @@ -0,0 +1,148 @@ +// Copyright (C) 2020 Intel Corporation +// Licensed subject to the terms of the separately executed evaluation license agreement between Intel Corporation and you. +syntax = "proto3"; + +import "google/protobuf/timestamp.proto"; +import "google/protobuf/duration.proto"; +import "federation.proto"; + + +message RequestHeader { + string sender = 1; +} + +// Envoy Messages + +message NodeInfo { + string name = 1; + string adress = 2; + bool cuda_available = 3; + uint32 memory_size = 4; +} + +message ShardInfo { + NodeInfo node_info = 1; + string shard_description = 2; + uint64 n_samples = 3; + // We just pass numpy shapes + repeated string sample_shape = 4; + repeated string target_shape = 5; +} + +message ShardAcknowledgement { + bool accepted = 1; +} + +message WaitExperimentRequest { + string collaborator_name = 1; +} + +message WaitExperimentResponse { + string experiment_name = 1; +} + +message GetExperimentDataRequest { + string experiment_name = 1; + string collaborator_name = 2; +} + +message ExperimentData { + uint32 size = 1; // size, in bytes, of the data sent in npbytes + bytes npbytes = 2; //actual data +} + +// API Messages + +message ExperimentInfo { + RequestHeader header = 1; + string name = 2; + repeated string collaborator_names = 3; + ExperimentData experiment_data = 4; + ModelProto model_proto = 5; +} + +message SetNewExperimentResponse{ + bool accepted = 1; + string tensorboard_address = 2; +} + +message GetTrainedModelRequest { + enum ModelType { + BEST_MODEL = 0; + LAST_MODEL = 1; + } + RequestHeader header = 1; + string experiment_name = 2; + ModelType model_type = 3; +} + +message TrainedModelResponse { + ModelProto model_proto = 1; +} + +message GetDatasetInfoRequest { + RequestHeader header = 1; +} + +message StreamMetricsRequest { + RequestHeader header = 1; + string experiment_name = 2; +} + +message StreamMetricsResponse { + string metric_origin = 1; + string task_name = 2; + string metric_name = 3; + float metric_value = 4; + uint32 round = 5; +} + +message RemoveExperimentRequest { + RequestHeader header = 1; + string experiment_name = 2; +} + +message RemoveExperimentResponse { + bool acknowledgement = 1; +} + +message CollaboratorStatus { + string name = 1; + bool is_experiment_running = 2; + google.protobuf.Duration valid_duration = 3; +} + +message CollaboratorHealthCheckResponse { + bool accepted = 1; +} + +message EnvoyInfo { + ShardInfo shard_info = 1; + bool is_online = 2; + bool is_experiment_running = 3; + google.protobuf.Timestamp last_updated = 4; + google.protobuf.Duration valid_duration = 5; +} + +message GetEnvoysRequest {} + +message GetEnvoysResponse { + repeated EnvoyInfo envoy_infos = 1; +} + +service FederationDirector { + // Envoy RPCs + rpc AcknowledgeShard (ShardInfo) returns (ShardAcknowledgement) {} + // Shard owner could also provide some public data for tests + rpc WaitExperiment (stream WaitExperimentRequest) returns (stream WaitExperimentResponse) {} + rpc GetExperimentData (GetExperimentDataRequest) returns (stream ExperimentData) {} + + // API RPCs + rpc SetNewExperiment (stream ExperimentInfo) returns (SetNewExperimentResponse) {} + rpc GetDatasetInfo (GetDatasetInfoRequest) returns (ShardInfo) {} + rpc GetTrainedModel (GetTrainedModelRequest) returns (TrainedModelResponse) {} + rpc StreamMetrics (StreamMetricsRequest) returns (stream StreamMetricsResponse) {} + rpc RemoveExperimentData (RemoveExperimentRequest) returns (RemoveExperimentResponse) {} + rpc CollaboratorHealthCheck (CollaboratorStatus) returns (CollaboratorHealthCheckResponse) {} + rpc GetEnvoys (GetEnvoysRequest) returns (GetEnvoysResponse) {} +} diff --git a/openfl/protocols/director_pb2.py b/openfl/protocols/director_pb2.py new file mode 100644 index 0000000000..adbaaee0f0 --- /dev/null +++ b/openfl/protocols/director_pb2.py @@ -0,0 +1,1261 @@ +# -- coding: utf-8 -- +# Generated by the protocol buffer compiler. DO NOT EDIT! +# source: director.proto +"""Generated protocol buffer code.""" +from google.protobuf import descriptor as _descriptor +from google.protobuf import message as _message +from google.protobuf import reflection as _reflection +from google.protobuf import symbol_database as _symbol_database +# @@protoc_insertion_point(imports) + +_sym_db = _symbol_database.Default() + + +from google.protobuf import timestamp_pb2 as google_dot_protobuf_dot_timestamp__pb2 +from google.protobuf import duration_pb2 as google_dot_protobuf_dot_duration__pb2 +import openfl.protocols.federation_pb2 as federation__pb2 + + +DESCRIPTOR = _descriptor.FileDescriptor( + name='director.proto', + package='', + syntax='proto3', + serialized_options=None, + create_key=_descriptor._internal_create_key, + serialized_pb=b'\n\x0e\x64irector.proto\x1a\x1fgoogle/protobuf/timestamp.proto\x1a\x1egoogle/protobuf/duration.proto\x1a\x10\x66\x65\x64\x65ration.proto\"\x1f\n\rRequestHeader\x12\x0e\n\x06sender\x18\x01 \x01(\t\"U\n\x08NodeInfo\x12\x0c\n\x04name\x18\x01 \x01(\t\x12\x0e\n\x06\x61\x64ress\x18\x02 \x01(\t\x12\x16\n\x0e\x63uda_available\x18\x03 \x01(\x08\x12\x13\n\x0bmemory_size\x18\x04 \x01(\r\"\x83\x01\n\tShardInfo\x12\x1c\n\tnode_info\x18\x01 \x01(\x0b\x32\t.NodeInfo\x12\x19\n\x11shard_description\x18\x02 \x01(\t\x12\x11\n\tn_samples\x18\x03 \x01(\x04\x12\x14\n\x0csample_shape\x18\x04 \x03(\t\x12\x14\n\x0ctarget_shape\x18\x05 \x03(\t\"(\n\x14ShardAcknowledgement\x12\x10\n\x08\x61\x63\x63\x65pted\x18\x01 \x01(\x08\"2\n\x15WaitExperimentRequest\x12\x19\n\x11\x63ollaborator_name\x18\x01 \x01(\t\"1\n\x16WaitExperimentResponse\x12\x17\n\x0f\x65xperiment_name\x18\x01 \x01(\t\"N\n\x18GetExperimentDataRequest\x12\x17\n\x0f\x65xperiment_name\x18\x01 \x01(\t\x12\x19\n\x11\x63ollaborator_name\x18\x02 \x01(\t\"/\n\x0e\x45xperimentData\x12\x0c\n\x04size\x18\x01 \x01(\r\x12\x0f\n\x07npbytes\x18\x02 \x01(\x0c\"\xa6\x01\n\x0e\x45xperimentInfo\x12\x1e\n\x06header\x18\x01 \x01(\x0b\x32\x0e.RequestHeader\x12\x0c\n\x04name\x18\x02 \x01(\t\x12\x1a\n\x12\x63ollaborator_names\x18\x03 \x03(\t\x12(\n\x0f\x65xperiment_data\x18\x04 \x01(\x0b\x32\x0f.ExperimentData\x12 \n\x0bmodel_proto\x18\x05 \x01(\x0b\x32\x0b.ModelProto\"I\n\x18SetNewExperimentResponse\x12\x10\n\x08\x61\x63\x63\x65pted\x18\x01 \x01(\x08\x12\x1b\n\x13tensorboard_address\x18\x02 \x01(\t\"\xb5\x01\n\x16GetTrainedModelRequest\x12\x1e\n\x06header\x18\x01 \x01(\x0b\x32\x0e.RequestHeader\x12\x17\n\x0f\x65xperiment_name\x18\x02 \x01(\t\x12\x35\n\nmodel_type\x18\x03 \x01(\x0e\x32!.GetTrainedModelRequest.ModelType\"+\n\tModelType\x12\x0e\n\nBEST_MODEL\x10\x00\x12\x0e\n\nLAST_MODEL\x10\x01\"8\n\x14TrainedModelResponse\x12 \n\x0bmodel_proto\x18\x01 \x01(\x0b\x32\x0b.ModelProto\"7\n\x15GetDatasetInfoRequest\x12\x1e\n\x06header\x18\x01 \x01(\x0b\x32\x0e.RequestHeader\"O\n\x14StreamMetricsRequest\x12\x1e\n\x06header\x18\x01 \x01(\x0b\x32\x0e.RequestHeader\x12\x17\n\x0f\x65xperiment_name\x18\x02 \x01(\t\"{\n\x15StreamMetricsResponse\x12\x15\n\rmetric_origin\x18\x01 \x01(\t\x12\x11\n\ttask_name\x18\x02 \x01(\t\x12\x13\n\x0bmetric_name\x18\x03 \x01(\t\x12\x14\n\x0cmetric_value\x18\x04 \x01(\x02\x12\r\n\x05round\x18\x05 \x01(\r\"R\n\x17RemoveExperimentRequest\x12\x1e\n\x06header\x18\x01 \x01(\x0b\x32\x0e.RequestHeader\x12\x17\n\x0f\x65xperiment_name\x18\x02 \x01(\t\"3\n\x18RemoveExperimentResponse\x12\x17\n\x0f\x61\x63knowledgement\x18\x01 \x01(\x08\"t\n\x12\x43ollaboratorStatus\x12\x0c\n\x04name\x18\x01 \x01(\t\x12\x1d\n\x15is_experiment_running\x18\x02 \x01(\x08\x12\x31\n\x0evalid_duration\x18\x03 \x01(\x0b\x32\x19.google.protobuf.Duration\"3\n\x1f\x43ollaboratorHealthCheckResponse\x12\x10\n\x08\x61\x63\x63\x65pted\x18\x01 \x01(\x08\"\xc2\x01\n\tEnvoyInfo\x12\x1e\n\nshard_info\x18\x01 \x01(\x0b\x32\n.ShardInfo\x12\x11\n\tis_online\x18\x02 \x01(\x08\x12\x1d\n\x15is_experiment_running\x18\x03 \x01(\x08\x12\x30\n\x0clast_updated\x18\x04 \x01(\x0b\x32\x1a.google.protobuf.Timestamp\x12\x31\n\x0evalid_duration\x18\x05 \x01(\x0b\x32\x19.google.protobuf.Duration\"\x12\n\x10GetEnvoysRequest\"4\n\x11GetEnvoysResponse\x12\x1f\n\x0b\x65nvoy_infos\x18\x01 \x03(\x0b\x32\n.EnvoyInfo2\xb9\x05\n\x12\x46\x65\x64\x65rationDirector\x12\x37\n\x10\x41\x63knowledgeShard\x12\n.ShardInfo\x1a\x15.ShardAcknowledgement\"\x00\x12G\n\x0eWaitExperiment\x12\x16.WaitExperimentRequest\x1a\x17.WaitExperimentResponse\"\x00(\x01\x30\x01\x12\x43\n\x11GetExperimentData\x12\x19.GetExperimentDataRequest\x1a\x0f.ExperimentData\"\x00\x30\x01\x12\x42\n\x10SetNewExperiment\x12\x0f.ExperimentInfo\x1a\x19.SetNewExperimentResponse\"\x00(\x01\x12\x36\n\x0eGetDatasetInfo\x12\x16.GetDatasetInfoRequest\x1a\n.ShardInfo\"\x00\x12\x43\n\x0fGetTrainedModel\x12\x17.GetTrainedModelRequest\x1a\x15.TrainedModelResponse\"\x00\x12\x42\n\rStreamMetrics\x12\x15.StreamMetricsRequest\x1a\x16.StreamMetricsResponse\"\x00\x30\x01\x12M\n\x14RemoveExperimentData\x12\x18.RemoveExperimentRequest\x1a\x19.RemoveExperimentResponse\"\x00\x12R\n\x17\x43ollaboratorHealthCheck\x12\x13.CollaboratorStatus\x1a .CollaboratorHealthCheckResponse\"\x00\x12\x34\n\tGetEnvoys\x12\x11.GetEnvoysRequest\x1a\x12.GetEnvoysResponse\"\x00\x62\x06proto3' + , + dependencies=[google_dot_protobuf_dot_timestamp__pb2.DESCRIPTOR,google_dot_protobuf_dot_duration__pb2.DESCRIPTOR,federation__pb2.DESCRIPTOR,]) + + + +_GETTRAINEDMODELREQUEST_MODELTYPE = _descriptor.EnumDescriptor( + name='ModelType', + full_name='GetTrainedModelRequest.ModelType', + filename=None, + file=DESCRIPTOR, + create_key=_descriptor._internal_create_key, + values=[ + _descriptor.EnumValueDescriptor( + name='BEST_MODEL', index=0, number=0, + serialized_options=None, + type=None, + create_key=_descriptor._internal_create_key), + _descriptor.EnumValueDescriptor( + name='LAST_MODEL', index=1, number=1, + serialized_options=None, + type=None, + create_key=_descriptor._internal_create_key), + ], + containing_type=None, + serialized_options=None, + serialized_start=1012, + serialized_end=1055, +) +_sym_db.RegisterEnumDescriptor(_GETTRAINEDMODELREQUEST_MODELTYPE) + + +_REQUESTHEADER = _descriptor.Descriptor( + name='RequestHeader', + full_name='RequestHeader', + filename=None, + file=DESCRIPTOR, + containing_type=None, + create_key=_descriptor._internal_create_key, + fields=[ + _descriptor.FieldDescriptor( + name='sender', full_name='RequestHeader.sender', index=0, + number=1, type=9, cpp_type=9, label=1, + has_default_value=False, default_value=b"".decode('utf-8'), + message_type=None, enum_type=None, containing_type=None, + is_extension=False, extension_scope=None, + serialized_options=None, file=DESCRIPTOR, create_key=_descriptor._internal_create_key), + ], + extensions=[ + ], + nested_types=[], + enum_types=[ + ], + serialized_options=None, + is_extendable=False, + syntax='proto3', + extension_ranges=[], + oneofs=[ + ], + serialized_start=101, + serialized_end=132, +) + + +_NODEINFO = _descriptor.Descriptor( + name='NodeInfo', + full_name='NodeInfo', + filename=None, + file=DESCRIPTOR, + containing_type=None, + create_key=_descriptor._internal_create_key, + fields=[ + _descriptor.FieldDescriptor( + name='name', full_name='NodeInfo.name', index=0, + number=1, type=9, cpp_type=9, label=1, + has_default_value=False, default_value=b"".decode('utf-8'), + message_type=None, enum_type=None, containing_type=None, + is_extension=False, extension_scope=None, + serialized_options=None, file=DESCRIPTOR, create_key=_descriptor._internal_create_key), + _descriptor.FieldDescriptor( + name='adress', full_name='NodeInfo.adress', index=1, + number=2, type=9, cpp_type=9, label=1, + has_default_value=False, default_value=b"".decode('utf-8'), + message_type=None, enum_type=None, containing_type=None, + is_extension=False, extension_scope=None, + serialized_options=None, file=DESCRIPTOR, create_key=_descriptor._internal_create_key), + _descriptor.FieldDescriptor( + name='cuda_available', full_name='NodeInfo.cuda_available', index=2, + number=3, type=8, cpp_type=7, label=1, + has_default_value=False, default_value=False, + message_type=None, enum_type=None, containing_type=None, + is_extension=False, extension_scope=None, + serialized_options=None, file=DESCRIPTOR, create_key=_descriptor._internal_create_key), + _descriptor.FieldDescriptor( + name='memory_size', full_name='NodeInfo.memory_size', index=3, + number=4, type=13, cpp_type=3, label=1, + has_default_value=False, default_value=0, + message_type=None, enum_type=None, containing_type=None, + is_extension=False, extension_scope=None, + serialized_options=None, file=DESCRIPTOR, create_key=_descriptor._internal_create_key), + ], + extensions=[ + ], + nested_types=[], + enum_types=[ + ], + serialized_options=None, + is_extendable=False, + syntax='proto3', + extension_ranges=[], + oneofs=[ + ], + serialized_start=134, + serialized_end=219, +) + + +_SHARDINFO = _descriptor.Descriptor( + name='ShardInfo', + full_name='ShardInfo', + filename=None, + file=DESCRIPTOR, + containing_type=None, + create_key=_descriptor._internal_create_key, + fields=[ + _descriptor.FieldDescriptor( + name='node_info', full_name='ShardInfo.node_info', index=0, + number=1, type=11, cpp_type=10, label=1, + has_default_value=False, default_value=None, + message_type=None, enum_type=None, containing_type=None, + is_extension=False, extension_scope=None, + serialized_options=None, file=DESCRIPTOR, create_key=_descriptor._internal_create_key), + _descriptor.FieldDescriptor( + name='shard_description', full_name='ShardInfo.shard_description', index=1, + number=2, type=9, cpp_type=9, label=1, + has_default_value=False, default_value=b"".decode('utf-8'), + message_type=None, enum_type=None, containing_type=None, + is_extension=False, extension_scope=None, + serialized_options=None, file=DESCRIPTOR, create_key=_descriptor._internal_create_key), + _descriptor.FieldDescriptor( + name='n_samples', full_name='ShardInfo.n_samples', index=2, + number=3, type=4, cpp_type=4, label=1, + has_default_value=False, default_value=0, + message_type=None, enum_type=None, containing_type=None, + is_extension=False, extension_scope=None, + serialized_options=None, file=DESCRIPTOR, create_key=_descriptor._internal_create_key), + _descriptor.FieldDescriptor( + name='sample_shape', full_name='ShardInfo.sample_shape', index=3, + number=4, type=9, cpp_type=9, label=3, + has_default_value=False, default_value=[], + message_type=None, enum_type=None, containing_type=None, + is_extension=False, extension_scope=None, + serialized_options=None, file=DESCRIPTOR, create_key=_descriptor._internal_create_key), + _descriptor.FieldDescriptor( + name='target_shape', full_name='ShardInfo.target_shape', index=4, + number=5, type=9, cpp_type=9, label=3, + has_default_value=False, default_value=[], + message_type=None, enum_type=None, containing_type=None, + is_extension=False, extension_scope=None, + serialized_options=None, file=DESCRIPTOR, create_key=_descriptor._internal_create_key), + ], + extensions=[ + ], + nested_types=[], + enum_types=[ + ], + serialized_options=None, + is_extendable=False, + syntax='proto3', + extension_ranges=[], + oneofs=[ + ], + serialized_start=222, + serialized_end=353, +) + + +_SHARDACKNOWLEDGEMENT = _descriptor.Descriptor( + name='ShardAcknowledgement', + full_name='ShardAcknowledgement', + filename=None, + file=DESCRIPTOR, + containing_type=None, + create_key=_descriptor._internal_create_key, + fields=[ + _descriptor.FieldDescriptor( + name='accepted', full_name='ShardAcknowledgement.accepted', index=0, + number=1, type=8, cpp_type=7, label=1, + has_default_value=False, default_value=False, + message_type=None, enum_type=None, containing_type=None, + is_extension=False, extension_scope=None, + serialized_options=None, file=DESCRIPTOR, create_key=_descriptor._internal_create_key), + ], + extensions=[ + ], + nested_types=[], + enum_types=[ + ], + serialized_options=None, + is_extendable=False, + syntax='proto3', + extension_ranges=[], + oneofs=[ + ], + serialized_start=355, + serialized_end=395, +) + + +_WAITEXPERIMENTREQUEST = _descriptor.Descriptor( + name='WaitExperimentRequest', + full_name='WaitExperimentRequest', + filename=None, + file=DESCRIPTOR, + containing_type=None, + create_key=_descriptor._internal_create_key, + fields=[ + _descriptor.FieldDescriptor( + name='collaborator_name', full_name='WaitExperimentRequest.collaborator_name', index=0, + number=1, type=9, cpp_type=9, label=1, + has_default_value=False, default_value=b"".decode('utf-8'), + message_type=None, enum_type=None, containing_type=None, + is_extension=False, extension_scope=None, + serialized_options=None, file=DESCRIPTOR, create_key=_descriptor._internal_create_key), + ], + extensions=[ + ], + nested_types=[], + enum_types=[ + ], + serialized_options=None, + is_extendable=False, + syntax='proto3', + extension_ranges=[], + oneofs=[ + ], + serialized_start=397, + serialized_end=447, +) + + +_WAITEXPERIMENTRESPONSE = _descriptor.Descriptor( + name='WaitExperimentResponse', + full_name='WaitExperimentResponse', + filename=None, + file=DESCRIPTOR, + containing_type=None, + create_key=_descriptor._internal_create_key, + fields=[ + _descriptor.FieldDescriptor( + name='experiment_name', full_name='WaitExperimentResponse.experiment_name', index=0, + number=1, type=9, cpp_type=9, label=1, + has_default_value=False, default_value=b"".decode('utf-8'), + message_type=None, enum_type=None, containing_type=None, + is_extension=False, extension_scope=None, + serialized_options=None, file=DESCRIPTOR, create_key=_descriptor._internal_create_key), + ], + extensions=[ + ], + nested_types=[], + enum_types=[ + ], + serialized_options=None, + is_extendable=False, + syntax='proto3', + extension_ranges=[], + oneofs=[ + ], + serialized_start=449, + serialized_end=498, +) + + +_GETEXPERIMENTDATAREQUEST = _descriptor.Descriptor( + name='GetExperimentDataRequest', + full_name='GetExperimentDataRequest', + filename=None, + file=DESCRIPTOR, + containing_type=None, + create_key=_descriptor._internal_create_key, + fields=[ + _descriptor.FieldDescriptor( + name='experiment_name', full_name='GetExperimentDataRequest.experiment_name', index=0, + number=1, type=9, cpp_type=9, label=1, + has_default_value=False, default_value=b"".decode('utf-8'), + message_type=None, enum_type=None, containing_type=None, + is_extension=False, extension_scope=None, + serialized_options=None, file=DESCRIPTOR, create_key=_descriptor._internal_create_key), + _descriptor.FieldDescriptor( + name='collaborator_name', full_name='GetExperimentDataRequest.collaborator_name', index=1, + number=2, type=9, cpp_type=9, label=1, + has_default_value=False, default_value=b"".decode('utf-8'), + message_type=None, enum_type=None, containing_type=None, + is_extension=False, extension_scope=None, + serialized_options=None, file=DESCRIPTOR, create_key=_descriptor._internal_create_key), + ], + extensions=[ + ], + nested_types=[], + enum_types=[ + ], + serialized_options=None, + is_extendable=False, + syntax='proto3', + extension_ranges=[], + oneofs=[ + ], + serialized_start=500, + serialized_end=578, +) + + +_EXPERIMENTDATA = _descriptor.Descriptor( + name='ExperimentData', + full_name='ExperimentData', + filename=None, + file=DESCRIPTOR, + containing_type=None, + create_key=_descriptor._internal_create_key, + fields=[ + _descriptor.FieldDescriptor( + name='size', full_name='ExperimentData.size', index=0, + number=1, type=13, cpp_type=3, label=1, + has_default_value=False, default_value=0, + message_type=None, enum_type=None, containing_type=None, + is_extension=False, extension_scope=None, + serialized_options=None, file=DESCRIPTOR, create_key=_descriptor._internal_create_key), + _descriptor.FieldDescriptor( + name='npbytes', full_name='ExperimentData.npbytes', index=1, + number=2, type=12, cpp_type=9, label=1, + has_default_value=False, default_value=b"", + message_type=None, enum_type=None, containing_type=None, + is_extension=False, extension_scope=None, + serialized_options=None, file=DESCRIPTOR, create_key=_descriptor._internal_create_key), + ], + extensions=[ + ], + nested_types=[], + enum_types=[ + ], + serialized_options=None, + is_extendable=False, + syntax='proto3', + extension_ranges=[], + oneofs=[ + ], + serialized_start=580, + serialized_end=627, +) + + +_EXPERIMENTINFO = _descriptor.Descriptor( + name='ExperimentInfo', + full_name='ExperimentInfo', + filename=None, + file=DESCRIPTOR, + containing_type=None, + create_key=_descriptor._internal_create_key, + fields=[ + _descriptor.FieldDescriptor( + name='header', full_name='ExperimentInfo.header', index=0, + number=1, type=11, cpp_type=10, label=1, + has_default_value=False, default_value=None, + message_type=None, enum_type=None, containing_type=None, + is_extension=False, extension_scope=None, + serialized_options=None, file=DESCRIPTOR, create_key=_descriptor._internal_create_key), + _descriptor.FieldDescriptor( + name='name', full_name='ExperimentInfo.name', index=1, + number=2, type=9, cpp_type=9, label=1, + has_default_value=False, default_value=b"".decode('utf-8'), + message_type=None, enum_type=None, containing_type=None, + is_extension=False, extension_scope=None, + serialized_options=None, file=DESCRIPTOR, create_key=_descriptor._internal_create_key), + _descriptor.FieldDescriptor( + name='collaborator_names', full_name='ExperimentInfo.collaborator_names', index=2, + number=3, type=9, cpp_type=9, label=3, + has_default_value=False, default_value=[], + message_type=None, enum_type=None, containing_type=None, + is_extension=False, extension_scope=None, + serialized_options=None, file=DESCRIPTOR, create_key=_descriptor._internal_create_key), + _descriptor.FieldDescriptor( + name='experiment_data', full_name='ExperimentInfo.experiment_data', index=3, + number=4, type=11, cpp_type=10, label=1, + has_default_value=False, default_value=None, + message_type=None, enum_type=None, containing_type=None, + is_extension=False, extension_scope=None, + serialized_options=None, file=DESCRIPTOR, create_key=_descriptor._internal_create_key), + _descriptor.FieldDescriptor( + name='model_proto', full_name='ExperimentInfo.model_proto', index=4, + number=5, type=11, cpp_type=10, label=1, + has_default_value=False, default_value=None, + message_type=None, enum_type=None, containing_type=None, + is_extension=False, extension_scope=None, + serialized_options=None, file=DESCRIPTOR, create_key=_descriptor._internal_create_key), + ], + extensions=[ + ], + nested_types=[], + enum_types=[ + ], + serialized_options=None, + is_extendable=False, + syntax='proto3', + extension_ranges=[], + oneofs=[ + ], + serialized_start=630, + serialized_end=796, +) + + +_SETNEWEXPERIMENTRESPONSE = _descriptor.Descriptor( + name='SetNewExperimentResponse', + full_name='SetNewExperimentResponse', + filename=None, + file=DESCRIPTOR, + containing_type=None, + create_key=_descriptor._internal_create_key, + fields=[ + _descriptor.FieldDescriptor( + name='accepted', full_name='SetNewExperimentResponse.accepted', index=0, + number=1, type=8, cpp_type=7, label=1, + has_default_value=False, default_value=False, + message_type=None, enum_type=None, containing_type=None, + is_extension=False, extension_scope=None, + serialized_options=None, file=DESCRIPTOR, create_key=_descriptor._internal_create_key), + _descriptor.FieldDescriptor( + name='tensorboard_address', full_name='SetNewExperimentResponse.tensorboard_address', index=1, + number=2, type=9, cpp_type=9, label=1, + has_default_value=False, default_value=b"".decode('utf-8'), + message_type=None, enum_type=None, containing_type=None, + is_extension=False, extension_scope=None, + serialized_options=None, file=DESCRIPTOR, create_key=_descriptor._internal_create_key), + ], + extensions=[ + ], + nested_types=[], + enum_types=[ + ], + serialized_options=None, + is_extendable=False, + syntax='proto3', + extension_ranges=[], + oneofs=[ + ], + serialized_start=798, + serialized_end=871, +) + + +_GETTRAINEDMODELREQUEST = _descriptor.Descriptor( + name='GetTrainedModelRequest', + full_name='GetTrainedModelRequest', + filename=None, + file=DESCRIPTOR, + containing_type=None, + create_key=_descriptor._internal_create_key, + fields=[ + _descriptor.FieldDescriptor( + name='header', full_name='GetTrainedModelRequest.header', index=0, + number=1, type=11, cpp_type=10, label=1, + has_default_value=False, default_value=None, + message_type=None, enum_type=None, containing_type=None, + is_extension=False, extension_scope=None, + serialized_options=None, file=DESCRIPTOR, create_key=_descriptor._internal_create_key), + _descriptor.FieldDescriptor( + name='experiment_name', full_name='GetTrainedModelRequest.experiment_name', index=1, + number=2, type=9, cpp_type=9, label=1, + has_default_value=False, default_value=b"".decode('utf-8'), + message_type=None, enum_type=None, containing_type=None, + is_extension=False, extension_scope=None, + serialized_options=None, file=DESCRIPTOR, create_key=_descriptor._internal_create_key), + _descriptor.FieldDescriptor( + name='model_type', full_name='GetTrainedModelRequest.model_type', index=2, + number=3, type=14, cpp_type=8, label=1, + has_default_value=False, default_value=0, + message_type=None, enum_type=None, containing_type=None, + is_extension=False, extension_scope=None, + serialized_options=None, file=DESCRIPTOR, create_key=_descriptor._internal_create_key), + ], + extensions=[ + ], + nested_types=[], + enum_types=[ + _GETTRAINEDMODELREQUEST_MODELTYPE, + ], + serialized_options=None, + is_extendable=False, + syntax='proto3', + extension_ranges=[], + oneofs=[ + ], + serialized_start=874, + serialized_end=1055, +) + + +_TRAINEDMODELRESPONSE = _descriptor.Descriptor( + name='TrainedModelResponse', + full_name='TrainedModelResponse', + filename=None, + file=DESCRIPTOR, + containing_type=None, + create_key=_descriptor._internal_create_key, + fields=[ + _descriptor.FieldDescriptor( + name='model_proto', full_name='TrainedModelResponse.model_proto', index=0, + number=1, type=11, cpp_type=10, label=1, + has_default_value=False, default_value=None, + message_type=None, enum_type=None, containing_type=None, + is_extension=False, extension_scope=None, + serialized_options=None, file=DESCRIPTOR, create_key=_descriptor._internal_create_key), + ], + extensions=[ + ], + nested_types=[], + enum_types=[ + ], + serialized_options=None, + is_extendable=False, + syntax='proto3', + extension_ranges=[], + oneofs=[ + ], + serialized_start=1057, + serialized_end=1113, +) + + +_GETDATASETINFOREQUEST = _descriptor.Descriptor( + name='GetDatasetInfoRequest', + full_name='GetDatasetInfoRequest', + filename=None, + file=DESCRIPTOR, + containing_type=None, + create_key=_descriptor._internal_create_key, + fields=[ + _descriptor.FieldDescriptor( + name='header', full_name='GetDatasetInfoRequest.header', index=0, + number=1, type=11, cpp_type=10, label=1, + has_default_value=False, default_value=None, + message_type=None, enum_type=None, containing_type=None, + is_extension=False, extension_scope=None, + serialized_options=None, file=DESCRIPTOR, create_key=_descriptor._internal_create_key), + ], + extensions=[ + ], + nested_types=[], + enum_types=[ + ], + serialized_options=None, + is_extendable=False, + syntax='proto3', + extension_ranges=[], + oneofs=[ + ], + serialized_start=1115, + serialized_end=1170, +) + + +_STREAMMETRICSREQUEST = _descriptor.Descriptor( + name='StreamMetricsRequest', + full_name='StreamMetricsRequest', + filename=None, + file=DESCRIPTOR, + containing_type=None, + create_key=_descriptor._internal_create_key, + fields=[ + _descriptor.FieldDescriptor( + name='header', full_name='StreamMetricsRequest.header', index=0, + number=1, type=11, cpp_type=10, label=1, + has_default_value=False, default_value=None, + message_type=None, enum_type=None, containing_type=None, + is_extension=False, extension_scope=None, + serialized_options=None, file=DESCRIPTOR, create_key=_descriptor._internal_create_key), + _descriptor.FieldDescriptor( + name='experiment_name', full_name='StreamMetricsRequest.experiment_name', index=1, + number=2, type=9, cpp_type=9, label=1, + has_default_value=False, default_value=b"".decode('utf-8'), + message_type=None, enum_type=None, containing_type=None, + is_extension=False, extension_scope=None, + serialized_options=None, file=DESCRIPTOR, create_key=_descriptor._internal_create_key), + ], + extensions=[ + ], + nested_types=[], + enum_types=[ + ], + serialized_options=None, + is_extendable=False, + syntax='proto3', + extension_ranges=[], + oneofs=[ + ], + serialized_start=1172, + serialized_end=1251, +) + + +_STREAMMETRICSRESPONSE = _descriptor.Descriptor( + name='StreamMetricsResponse', + full_name='StreamMetricsResponse', + filename=None, + file=DESCRIPTOR, + containing_type=None, + create_key=_descriptor._internal_create_key, + fields=[ + _descriptor.FieldDescriptor( + name='metric_origin', full_name='StreamMetricsResponse.metric_origin', index=0, + number=1, type=9, cpp_type=9, label=1, + has_default_value=False, default_value=b"".decode('utf-8'), + message_type=None, enum_type=None, containing_type=None, + is_extension=False, extension_scope=None, + serialized_options=None, file=DESCRIPTOR, create_key=_descriptor._internal_create_key), + _descriptor.FieldDescriptor( + name='task_name', full_name='StreamMetricsResponse.task_name', index=1, + number=2, type=9, cpp_type=9, label=1, + has_default_value=False, default_value=b"".decode('utf-8'), + message_type=None, enum_type=None, containing_type=None, + is_extension=False, extension_scope=None, + serialized_options=None, file=DESCRIPTOR, create_key=_descriptor._internal_create_key), + _descriptor.FieldDescriptor( + name='metric_name', full_name='StreamMetricsResponse.metric_name', index=2, + number=3, type=9, cpp_type=9, label=1, + has_default_value=False, default_value=b"".decode('utf-8'), + message_type=None, enum_type=None, containing_type=None, + is_extension=False, extension_scope=None, + serialized_options=None, file=DESCRIPTOR, create_key=_descriptor._internal_create_key), + _descriptor.FieldDescriptor( + name='metric_value', full_name='StreamMetricsResponse.metric_value', index=3, + number=4, type=2, cpp_type=6, label=1, + has_default_value=False, default_value=float(0), + message_type=None, enum_type=None, containing_type=None, + is_extension=False, extension_scope=None, + serialized_options=None, file=DESCRIPTOR, create_key=_descriptor._internal_create_key), + _descriptor.FieldDescriptor( + name='round', full_name='StreamMetricsResponse.round', index=4, + number=5, type=13, cpp_type=3, label=1, + has_default_value=False, default_value=0, + message_type=None, enum_type=None, containing_type=None, + is_extension=False, extension_scope=None, + serialized_options=None, file=DESCRIPTOR, create_key=_descriptor._internal_create_key), + ], + extensions=[ + ], + nested_types=[], + enum_types=[ + ], + serialized_options=None, + is_extendable=False, + syntax='proto3', + extension_ranges=[], + oneofs=[ + ], + serialized_start=1253, + serialized_end=1376, +) + + +_REMOVEEXPERIMENTREQUEST = _descriptor.Descriptor( + name='RemoveExperimentRequest', + full_name='RemoveExperimentRequest', + filename=None, + file=DESCRIPTOR, + containing_type=None, + create_key=_descriptor._internal_create_key, + fields=[ + _descriptor.FieldDescriptor( + name='header', full_name='RemoveExperimentRequest.header', index=0, + number=1, type=11, cpp_type=10, label=1, + has_default_value=False, default_value=None, + message_type=None, enum_type=None, containing_type=None, + is_extension=False, extension_scope=None, + serialized_options=None, file=DESCRIPTOR, create_key=_descriptor._internal_create_key), + _descriptor.FieldDescriptor( + name='experiment_name', full_name='RemoveExperimentRequest.experiment_name', index=1, + number=2, type=9, cpp_type=9, label=1, + has_default_value=False, default_value=b"".decode('utf-8'), + message_type=None, enum_type=None, containing_type=None, + is_extension=False, extension_scope=None, + serialized_options=None, file=DESCRIPTOR, create_key=_descriptor._internal_create_key), + ], + extensions=[ + ], + nested_types=[], + enum_types=[ + ], + serialized_options=None, + is_extendable=False, + syntax='proto3', + extension_ranges=[], + oneofs=[ + ], + serialized_start=1378, + serialized_end=1460, +) + + +_REMOVEEXPERIMENTRESPONSE = _descriptor.Descriptor( + name='RemoveExperimentResponse', + full_name='RemoveExperimentResponse', + filename=None, + file=DESCRIPTOR, + containing_type=None, + create_key=_descriptor._internal_create_key, + fields=[ + _descriptor.FieldDescriptor( + name='acknowledgement', full_name='RemoveExperimentResponse.acknowledgement', index=0, + number=1, type=8, cpp_type=7, label=1, + has_default_value=False, default_value=False, + message_type=None, enum_type=None, containing_type=None, + is_extension=False, extension_scope=None, + serialized_options=None, file=DESCRIPTOR, create_key=_descriptor._internal_create_key), + ], + extensions=[ + ], + nested_types=[], + enum_types=[ + ], + serialized_options=None, + is_extendable=False, + syntax='proto3', + extension_ranges=[], + oneofs=[ + ], + serialized_start=1462, + serialized_end=1513, +) + + +_COLLABORATORSTATUS = _descriptor.Descriptor( + name='CollaboratorStatus', + full_name='CollaboratorStatus', + filename=None, + file=DESCRIPTOR, + containing_type=None, + create_key=_descriptor._internal_create_key, + fields=[ + _descriptor.FieldDescriptor( + name='name', full_name='CollaboratorStatus.name', index=0, + number=1, type=9, cpp_type=9, label=1, + has_default_value=False, default_value=b"".decode('utf-8'), + message_type=None, enum_type=None, containing_type=None, + is_extension=False, extension_scope=None, + serialized_options=None, file=DESCRIPTOR, create_key=_descriptor._internal_create_key), + _descriptor.FieldDescriptor( + name='is_experiment_running', full_name='CollaboratorStatus.is_experiment_running', index=1, + number=2, type=8, cpp_type=7, label=1, + has_default_value=False, default_value=False, + message_type=None, enum_type=None, containing_type=None, + is_extension=False, extension_scope=None, + serialized_options=None, file=DESCRIPTOR, create_key=_descriptor._internal_create_key), + _descriptor.FieldDescriptor( + name='valid_duration', full_name='CollaboratorStatus.valid_duration', index=2, + number=3, type=11, cpp_type=10, label=1, + has_default_value=False, default_value=None, + message_type=None, enum_type=None, containing_type=None, + is_extension=False, extension_scope=None, + serialized_options=None, file=DESCRIPTOR, create_key=_descriptor._internal_create_key), + ], + extensions=[ + ], + nested_types=[], + enum_types=[ + ], + serialized_options=None, + is_extendable=False, + syntax='proto3', + extension_ranges=[], + oneofs=[ + ], + serialized_start=1515, + serialized_end=1631, +) + + +_COLLABORATORHEALTHCHECKRESPONSE = _descriptor.Descriptor( + name='CollaboratorHealthCheckResponse', + full_name='CollaboratorHealthCheckResponse', + filename=None, + file=DESCRIPTOR, + containing_type=None, + create_key=_descriptor._internal_create_key, + fields=[ + _descriptor.FieldDescriptor( + name='accepted', full_name='CollaboratorHealthCheckResponse.accepted', index=0, + number=1, type=8, cpp_type=7, label=1, + has_default_value=False, default_value=False, + message_type=None, enum_type=None, containing_type=None, + is_extension=False, extension_scope=None, + serialized_options=None, file=DESCRIPTOR, create_key=_descriptor._internal_create_key), + ], + extensions=[ + ], + nested_types=[], + enum_types=[ + ], + serialized_options=None, + is_extendable=False, + syntax='proto3', + extension_ranges=[], + oneofs=[ + ], + serialized_start=1633, + serialized_end=1684, +) + + +_ENVOYINFO = _descriptor.Descriptor( + name='EnvoyInfo', + full_name='EnvoyInfo', + filename=None, + file=DESCRIPTOR, + containing_type=None, + create_key=_descriptor._internal_create_key, + fields=[ + _descriptor.FieldDescriptor( + name='shard_info', full_name='EnvoyInfo.shard_info', index=0, + number=1, type=11, cpp_type=10, label=1, + has_default_value=False, default_value=None, + message_type=None, enum_type=None, containing_type=None, + is_extension=False, extension_scope=None, + serialized_options=None, file=DESCRIPTOR, create_key=_descriptor._internal_create_key), + _descriptor.FieldDescriptor( + name='is_online', full_name='EnvoyInfo.is_online', index=1, + number=2, type=8, cpp_type=7, label=1, + has_default_value=False, default_value=False, + message_type=None, enum_type=None, containing_type=None, + is_extension=False, extension_scope=None, + serialized_options=None, file=DESCRIPTOR, create_key=_descriptor._internal_create_key), + _descriptor.FieldDescriptor( + name='is_experiment_running', full_name='EnvoyInfo.is_experiment_running', index=2, + number=3, type=8, cpp_type=7, label=1, + has_default_value=False, default_value=False, + message_type=None, enum_type=None, containing_type=None, + is_extension=False, extension_scope=None, + serialized_options=None, file=DESCRIPTOR, create_key=_descriptor._internal_create_key), + _descriptor.FieldDescriptor( + name='last_updated', full_name='EnvoyInfo.last_updated', index=3, + number=4, type=11, cpp_type=10, label=1, + has_default_value=False, default_value=None, + message_type=None, enum_type=None, containing_type=None, + is_extension=False, extension_scope=None, + serialized_options=None, file=DESCRIPTOR, create_key=_descriptor._internal_create_key), + _descriptor.FieldDescriptor( + name='valid_duration', full_name='EnvoyInfo.valid_duration', index=4, + number=5, type=11, cpp_type=10, label=1, + has_default_value=False, default_value=None, + message_type=None, enum_type=None, containing_type=None, + is_extension=False, extension_scope=None, + serialized_options=None, file=DESCRIPTOR, create_key=_descriptor._internal_create_key), + ], + extensions=[ + ], + nested_types=[], + enum_types=[ + ], + serialized_options=None, + is_extendable=False, + syntax='proto3', + extension_ranges=[], + oneofs=[ + ], + serialized_start=1687, + serialized_end=1881, +) + + +_GETENVOYSREQUEST = _descriptor.Descriptor( + name='GetEnvoysRequest', + full_name='GetEnvoysRequest', + filename=None, + file=DESCRIPTOR, + containing_type=None, + create_key=_descriptor._internal_create_key, + fields=[ + ], + extensions=[ + ], + nested_types=[], + enum_types=[ + ], + serialized_options=None, + is_extendable=False, + syntax='proto3', + extension_ranges=[], + oneofs=[ + ], + serialized_start=1883, + serialized_end=1901, +) + + +_GETENVOYSRESPONSE = _descriptor.Descriptor( + name='GetEnvoysResponse', + full_name='GetEnvoysResponse', + filename=None, + file=DESCRIPTOR, + containing_type=None, + create_key=_descriptor._internal_create_key, + fields=[ + _descriptor.FieldDescriptor( + name='envoy_infos', full_name='GetEnvoysResponse.envoy_infos', index=0, + number=1, type=11, cpp_type=10, label=3, + has_default_value=False, default_value=[], + message_type=None, enum_type=None, containing_type=None, + is_extension=False, extension_scope=None, + serialized_options=None, file=DESCRIPTOR, create_key=_descriptor._internal_create_key), + ], + extensions=[ + ], + nested_types=[], + enum_types=[ + ], + serialized_options=None, + is_extendable=False, + syntax='proto3', + extension_ranges=[], + oneofs=[ + ], + serialized_start=1903, + serialized_end=1955, +) + +_SHARDINFO.fields_by_name['node_info'].message_type = _NODEINFO +_EXPERIMENTINFO.fields_by_name['header'].message_type = _REQUESTHEADER +_EXPERIMENTINFO.fields_by_name['experiment_data'].message_type = _EXPERIMENTDATA +_EXPERIMENTINFO.fields_by_name['model_proto'].message_type = federation__pb2._MODELPROTO +_GETTRAINEDMODELREQUEST.fields_by_name['header'].message_type = _REQUESTHEADER +_GETTRAINEDMODELREQUEST.fields_by_name['model_type'].enum_type = _GETTRAINEDMODELREQUEST_MODELTYPE +_GETTRAINEDMODELREQUEST_MODELTYPE.containing_type = _GETTRAINEDMODELREQUEST +_TRAINEDMODELRESPONSE.fields_by_name['model_proto'].message_type = federation__pb2._MODELPROTO +_GETDATASETINFOREQUEST.fields_by_name['header'].message_type = _REQUESTHEADER +_STREAMMETRICSREQUEST.fields_by_name['header'].message_type = _REQUESTHEADER +_REMOVEEXPERIMENTREQUEST.fields_by_name['header'].message_type = _REQUESTHEADER +_COLLABORATORSTATUS.fields_by_name['valid_duration'].message_type = google_dot_protobuf_dot_duration__pb2._DURATION +_ENVOYINFO.fields_by_name['shard_info'].message_type = _SHARDINFO +_ENVOYINFO.fields_by_name['last_updated'].message_type = google_dot_protobuf_dot_timestamp__pb2._TIMESTAMP +_ENVOYINFO.fields_by_name['valid_duration'].message_type = google_dot_protobuf_dot_duration__pb2._DURATION +_GETENVOYSRESPONSE.fields_by_name['envoy_infos'].message_type = _ENVOYINFO +DESCRIPTOR.message_types_by_name['RequestHeader'] = _REQUESTHEADER +DESCRIPTOR.message_types_by_name['NodeInfo'] = _NODEINFO +DESCRIPTOR.message_types_by_name['ShardInfo'] = _SHARDINFO +DESCRIPTOR.message_types_by_name['ShardAcknowledgement'] = _SHARDACKNOWLEDGEMENT +DESCRIPTOR.message_types_by_name['WaitExperimentRequest'] = _WAITEXPERIMENTREQUEST +DESCRIPTOR.message_types_by_name['WaitExperimentResponse'] = _WAITEXPERIMENTRESPONSE +DESCRIPTOR.message_types_by_name['GetExperimentDataRequest'] = _GETEXPERIMENTDATAREQUEST +DESCRIPTOR.message_types_by_name['ExperimentData'] = _EXPERIMENTDATA +DESCRIPTOR.message_types_by_name['ExperimentInfo'] = _EXPERIMENTINFO +DESCRIPTOR.message_types_by_name['SetNewExperimentResponse'] = _SETNEWEXPERIMENTRESPONSE +DESCRIPTOR.message_types_by_name['GetTrainedModelRequest'] = _GETTRAINEDMODELREQUEST +DESCRIPTOR.message_types_by_name['TrainedModelResponse'] = _TRAINEDMODELRESPONSE +DESCRIPTOR.message_types_by_name['GetDatasetInfoRequest'] = _GETDATASETINFOREQUEST +DESCRIPTOR.message_types_by_name['StreamMetricsRequest'] = _STREAMMETRICSREQUEST +DESCRIPTOR.message_types_by_name['StreamMetricsResponse'] = _STREAMMETRICSRESPONSE +DESCRIPTOR.message_types_by_name['RemoveExperimentRequest'] = _REMOVEEXPERIMENTREQUEST +DESCRIPTOR.message_types_by_name['RemoveExperimentResponse'] = _REMOVEEXPERIMENTRESPONSE +DESCRIPTOR.message_types_by_name['CollaboratorStatus'] = _COLLABORATORSTATUS +DESCRIPTOR.message_types_by_name['CollaboratorHealthCheckResponse'] = _COLLABORATORHEALTHCHECKRESPONSE +DESCRIPTOR.message_types_by_name['EnvoyInfo'] = _ENVOYINFO +DESCRIPTOR.message_types_by_name['GetEnvoysRequest'] = _GETENVOYSREQUEST +DESCRIPTOR.message_types_by_name['GetEnvoysResponse'] = _GETENVOYSRESPONSE +_sym_db.RegisterFileDescriptor(DESCRIPTOR) + +RequestHeader = _reflection.GeneratedProtocolMessageType('RequestHeader', (_message.Message,), { + 'DESCRIPTOR' : _REQUESTHEADER, + '__module__' : 'director_pb2' + # @@protoc_insertion_point(class_scope:RequestHeader) + }) +_sym_db.RegisterMessage(RequestHeader) + +NodeInfo = _reflection.GeneratedProtocolMessageType('NodeInfo', (_message.Message,), { + 'DESCRIPTOR' : _NODEINFO, + '__module__' : 'director_pb2' + # @@protoc_insertion_point(class_scope:NodeInfo) + }) +_sym_db.RegisterMessage(NodeInfo) + +ShardInfo = _reflection.GeneratedProtocolMessageType('ShardInfo', (_message.Message,), { + 'DESCRIPTOR' : _SHARDINFO, + '__module__' : 'director_pb2' + # @@protoc_insertion_point(class_scope:ShardInfo) + }) +_sym_db.RegisterMessage(ShardInfo) + +ShardAcknowledgement = _reflection.GeneratedProtocolMessageType('ShardAcknowledgement', (_message.Message,), { + 'DESCRIPTOR' : _SHARDACKNOWLEDGEMENT, + '__module__' : 'director_pb2' + # @@protoc_insertion_point(class_scope:ShardAcknowledgement) + }) +_sym_db.RegisterMessage(ShardAcknowledgement) + +WaitExperimentRequest = _reflection.GeneratedProtocolMessageType('WaitExperimentRequest', (_message.Message,), { + 'DESCRIPTOR' : _WAITEXPERIMENTREQUEST, + '__module__' : 'director_pb2' + # @@protoc_insertion_point(class_scope:WaitExperimentRequest) + }) +_sym_db.RegisterMessage(WaitExperimentRequest) + +WaitExperimentResponse = _reflection.GeneratedProtocolMessageType('WaitExperimentResponse', (_message.Message,), { + 'DESCRIPTOR' : _WAITEXPERIMENTRESPONSE, + '__module__' : 'director_pb2' + # @@protoc_insertion_point(class_scope:WaitExperimentResponse) + }) +_sym_db.RegisterMessage(WaitExperimentResponse) + +GetExperimentDataRequest = _reflection.GeneratedProtocolMessageType('GetExperimentDataRequest', (_message.Message,), { + 'DESCRIPTOR' : _GETEXPERIMENTDATAREQUEST, + '__module__' : 'director_pb2' + # @@protoc_insertion_point(class_scope:GetExperimentDataRequest) + }) +_sym_db.RegisterMessage(GetExperimentDataRequest) + +ExperimentData = _reflection.GeneratedProtocolMessageType('ExperimentData', (_message.Message,), { + 'DESCRIPTOR' : _EXPERIMENTDATA, + '__module__' : 'director_pb2' + # @@protoc_insertion_point(class_scope:ExperimentData) + }) +_sym_db.RegisterMessage(ExperimentData) + +ExperimentInfo = _reflection.GeneratedProtocolMessageType('ExperimentInfo', (_message.Message,), { + 'DESCRIPTOR' : _EXPERIMENTINFO, + '__module__' : 'director_pb2' + # @@protoc_insertion_point(class_scope:ExperimentInfo) + }) +_sym_db.RegisterMessage(ExperimentInfo) + +SetNewExperimentResponse = _reflection.GeneratedProtocolMessageType('SetNewExperimentResponse', (_message.Message,), { + 'DESCRIPTOR' : _SETNEWEXPERIMENTRESPONSE, + '__module__' : 'director_pb2' + # @@protoc_insertion_point(class_scope:SetNewExperimentResponse) + }) +_sym_db.RegisterMessage(SetNewExperimentResponse) + +GetTrainedModelRequest = _reflection.GeneratedProtocolMessageType('GetTrainedModelRequest', (_message.Message,), { + 'DESCRIPTOR' : _GETTRAINEDMODELREQUEST, + '__module__' : 'director_pb2' + # @@protoc_insertion_point(class_scope:GetTrainedModelRequest) + }) +_sym_db.RegisterMessage(GetTrainedModelRequest) + +TrainedModelResponse = _reflection.GeneratedProtocolMessageType('TrainedModelResponse', (_message.Message,), { + 'DESCRIPTOR' : _TRAINEDMODELRESPONSE, + '__module__' : 'director_pb2' + # @@protoc_insertion_point(class_scope:TrainedModelResponse) + }) +_sym_db.RegisterMessage(TrainedModelResponse) + +GetDatasetInfoRequest = _reflection.GeneratedProtocolMessageType('GetDatasetInfoRequest', (_message.Message,), { + 'DESCRIPTOR' : _GETDATASETINFOREQUEST, + '__module__' : 'director_pb2' + # @@protoc_insertion_point(class_scope:GetDatasetInfoRequest) + }) +_sym_db.RegisterMessage(GetDatasetInfoRequest) + +StreamMetricsRequest = _reflection.GeneratedProtocolMessageType('StreamMetricsRequest', (_message.Message,), { + 'DESCRIPTOR' : _STREAMMETRICSREQUEST, + '__module__' : 'director_pb2' + # @@protoc_insertion_point(class_scope:StreamMetricsRequest) + }) +_sym_db.RegisterMessage(StreamMetricsRequest) + +StreamMetricsResponse = _reflection.GeneratedProtocolMessageType('StreamMetricsResponse', (_message.Message,), { + 'DESCRIPTOR' : _STREAMMETRICSRESPONSE, + '__module__' : 'director_pb2' + # @@protoc_insertion_point(class_scope:StreamMetricsResponse) + }) +_sym_db.RegisterMessage(StreamMetricsResponse) + +RemoveExperimentRequest = _reflection.GeneratedProtocolMessageType('RemoveExperimentRequest', (_message.Message,), { + 'DESCRIPTOR' : _REMOVEEXPERIMENTREQUEST, + '__module__' : 'director_pb2' + # @@protoc_insertion_point(class_scope:RemoveExperimentRequest) + }) +_sym_db.RegisterMessage(RemoveExperimentRequest) + +RemoveExperimentResponse = _reflection.GeneratedProtocolMessageType('RemoveExperimentResponse', (_message.Message,), { + 'DESCRIPTOR' : _REMOVEEXPERIMENTRESPONSE, + '__module__' : 'director_pb2' + # @@protoc_insertion_point(class_scope:RemoveExperimentResponse) + }) +_sym_db.RegisterMessage(RemoveExperimentResponse) + +CollaboratorStatus = _reflection.GeneratedProtocolMessageType('CollaboratorStatus', (_message.Message,), { + 'DESCRIPTOR' : _COLLABORATORSTATUS, + '__module__' : 'director_pb2' + # @@protoc_insertion_point(class_scope:CollaboratorStatus) + }) +_sym_db.RegisterMessage(CollaboratorStatus) + +CollaboratorHealthCheckResponse = _reflection.GeneratedProtocolMessageType('CollaboratorHealthCheckResponse', (_message.Message,), { + 'DESCRIPTOR' : _COLLABORATORHEALTHCHECKRESPONSE, + '__module__' : 'director_pb2' + # @@protoc_insertion_point(class_scope:CollaboratorHealthCheckResponse) + }) +_sym_db.RegisterMessage(CollaboratorHealthCheckResponse) + +EnvoyInfo = _reflection.GeneratedProtocolMessageType('EnvoyInfo', (_message.Message,), { + 'DESCRIPTOR' : _ENVOYINFO, + '__module__' : 'director_pb2' + # @@protoc_insertion_point(class_scope:EnvoyInfo) + }) +_sym_db.RegisterMessage(EnvoyInfo) + +GetEnvoysRequest = _reflection.GeneratedProtocolMessageType('GetEnvoysRequest', (_message.Message,), { + 'DESCRIPTOR' : _GETENVOYSREQUEST, + '__module__' : 'director_pb2' + # @@protoc_insertion_point(class_scope:GetEnvoysRequest) + }) +_sym_db.RegisterMessage(GetEnvoysRequest) + +GetEnvoysResponse = _reflection.GeneratedProtocolMessageType('GetEnvoysResponse', (_message.Message,), { + 'DESCRIPTOR' : _GETENVOYSRESPONSE, + '__module__' : 'director_pb2' + # @@protoc_insertion_point(class_scope:GetEnvoysResponse) + }) +_sym_db.RegisterMessage(GetEnvoysResponse) + + + +_FEDERATIONDIRECTOR = _descriptor.ServiceDescriptor( + name='FederationDirector', + full_name='FederationDirector', + file=DESCRIPTOR, + index=0, + serialized_options=None, + create_key=_descriptor._internal_create_key, + serialized_start=1958, + serialized_end=2655, + methods=[ + _descriptor.MethodDescriptor( + name='AcknowledgeShard', + full_name='FederationDirector.AcknowledgeShard', + index=0, + containing_service=None, + input_type=_SHARDINFO, + output_type=_SHARDACKNOWLEDGEMENT, + serialized_options=None, + create_key=_descriptor._internal_create_key, + ), + _descriptor.MethodDescriptor( + name='WaitExperiment', + full_name='FederationDirector.WaitExperiment', + index=1, + containing_service=None, + input_type=_WAITEXPERIMENTREQUEST, + output_type=_WAITEXPERIMENTRESPONSE, + serialized_options=None, + create_key=_descriptor._internal_create_key, + ), + _descriptor.MethodDescriptor( + name='GetExperimentData', + full_name='FederationDirector.GetExperimentData', + index=2, + containing_service=None, + input_type=_GETEXPERIMENTDATAREQUEST, + output_type=_EXPERIMENTDATA, + serialized_options=None, + create_key=_descriptor._internal_create_key, + ), + _descriptor.MethodDescriptor( + name='SetNewExperiment', + full_name='FederationDirector.SetNewExperiment', + index=3, + containing_service=None, + input_type=_EXPERIMENTINFO, + output_type=_SETNEWEXPERIMENTRESPONSE, + serialized_options=None, + create_key=_descriptor._internal_create_key, + ), + _descriptor.MethodDescriptor( + name='GetDatasetInfo', + full_name='FederationDirector.GetDatasetInfo', + index=4, + containing_service=None, + input_type=_GETDATASETINFOREQUEST, + output_type=_SHARDINFO, + serialized_options=None, + create_key=_descriptor._internal_create_key, + ), + _descriptor.MethodDescriptor( + name='GetTrainedModel', + full_name='FederationDirector.GetTrainedModel', + index=5, + containing_service=None, + input_type=_GETTRAINEDMODELREQUEST, + output_type=_TRAINEDMODELRESPONSE, + serialized_options=None, + create_key=_descriptor._internal_create_key, + ), + _descriptor.MethodDescriptor( + name='StreamMetrics', + full_name='FederationDirector.StreamMetrics', + index=6, + containing_service=None, + input_type=_STREAMMETRICSREQUEST, + output_type=_STREAMMETRICSRESPONSE, + serialized_options=None, + create_key=_descriptor._internal_create_key, + ), + _descriptor.MethodDescriptor( + name='RemoveExperimentData', + full_name='FederationDirector.RemoveExperimentData', + index=7, + containing_service=None, + input_type=_REMOVEEXPERIMENTREQUEST, + output_type=_REMOVEEXPERIMENTRESPONSE, + serialized_options=None, + create_key=_descriptor._internal_create_key, + ), + _descriptor.MethodDescriptor( + name='CollaboratorHealthCheck', + full_name='FederationDirector.CollaboratorHealthCheck', + index=8, + containing_service=None, + input_type=_COLLABORATORSTATUS, + output_type=_COLLABORATORHEALTHCHECKRESPONSE, + serialized_options=None, + create_key=_descriptor._internal_create_key, + ), + _descriptor.MethodDescriptor( + name='GetEnvoys', + full_name='FederationDirector.GetEnvoys', + index=9, + containing_service=None, + input_type=_GETENVOYSREQUEST, + output_type=_GETENVOYSRESPONSE, + serialized_options=None, + create_key=_descriptor._internal_create_key, + ), +]) +_sym_db.RegisterServiceDescriptor(_FEDERATIONDIRECTOR) + +DESCRIPTOR.services_by_name['FederationDirector'] = _FEDERATIONDIRECTOR + +# @@protoc_insertion_point(module_scope) diff --git a/openfl/protocols/director_pb2_grpc.py b/openfl/protocols/director_pb2_grpc.py new file mode 100644 index 0000000000..18064a3c81 --- /dev/null +++ b/openfl/protocols/director_pb2_grpc.py @@ -0,0 +1,366 @@ +# Generated by the gRPC Python protocol compiler plugin. DO NOT EDIT! +"""Client and server classes corresponding to protobuf-defined services.""" +import grpc + +from . import director_pb2 as director__pb2 + + +class FederationDirectorStub(object): + """Missing associated documentation comment in .proto file.""" + + def __init__(self, channel): + """Constructor. + + Args: + channel: A grpc.Channel. + """ + self.AcknowledgeShard = channel.unary_unary( + '/FederationDirector/AcknowledgeShard', + request_serializer=director__pb2.ShardInfo.SerializeToString, + response_deserializer=director__pb2.ShardAcknowledgement.FromString, + ) + self.WaitExperiment = channel.stream_stream( + '/FederationDirector/WaitExperiment', + request_serializer=director__pb2.WaitExperimentRequest.SerializeToString, + response_deserializer=director__pb2.WaitExperimentResponse.FromString, + ) + self.GetExperimentData = channel.unary_stream( + '/FederationDirector/GetExperimentData', + request_serializer=director__pb2.GetExperimentDataRequest.SerializeToString, + response_deserializer=director__pb2.ExperimentData.FromString, + ) + self.SetNewExperiment = channel.stream_unary( + '/FederationDirector/SetNewExperiment', + request_serializer=director__pb2.ExperimentInfo.SerializeToString, + response_deserializer=director__pb2.SetNewExperimentResponse.FromString, + ) + self.GetDatasetInfo = channel.unary_unary( + '/FederationDirector/GetDatasetInfo', + request_serializer=director__pb2.GetDatasetInfoRequest.SerializeToString, + response_deserializer=director__pb2.ShardInfo.FromString, + ) + self.GetTrainedModel = channel.unary_unary( + '/FederationDirector/GetTrainedModel', + request_serializer=director__pb2.GetTrainedModelRequest.SerializeToString, + response_deserializer=director__pb2.TrainedModelResponse.FromString, + ) + self.StreamMetrics = channel.unary_stream( + '/FederationDirector/StreamMetrics', + request_serializer=director__pb2.StreamMetricsRequest.SerializeToString, + response_deserializer=director__pb2.StreamMetricsResponse.FromString, + ) + self.RemoveExperimentData = channel.unary_unary( + '/FederationDirector/RemoveExperimentData', + request_serializer=director__pb2.RemoveExperimentRequest.SerializeToString, + response_deserializer=director__pb2.RemoveExperimentResponse.FromString, + ) + self.CollaboratorHealthCheck = channel.unary_unary( + '/FederationDirector/CollaboratorHealthCheck', + request_serializer=director__pb2.CollaboratorStatus.SerializeToString, + response_deserializer=director__pb2.CollaboratorHealthCheckResponse.FromString, + ) + self.GetEnvoys = channel.unary_unary( + '/FederationDirector/GetEnvoys', + request_serializer=director__pb2.GetEnvoysRequest.SerializeToString, + response_deserializer=director__pb2.GetEnvoysResponse.FromString, + ) + + +class FederationDirectorServicer(object): + """Missing associated documentation comment in .proto file.""" + + def AcknowledgeShard(self, request, context): + """Envoy RPCs + """ + context.set_code(grpc.StatusCode.UNIMPLEMENTED) + context.set_details('Method not implemented!') + raise NotImplementedError('Method not implemented!') + + def WaitExperiment(self, request_iterator, context): + """Shard owner could also provide some public data for tests + """ + context.set_code(grpc.StatusCode.UNIMPLEMENTED) + context.set_details('Method not implemented!') + raise NotImplementedError('Method not implemented!') + + def GetExperimentData(self, request, context): + """Missing associated documentation comment in .proto file.""" + context.set_code(grpc.StatusCode.UNIMPLEMENTED) + context.set_details('Method not implemented!') + raise NotImplementedError('Method not implemented!') + + def SetNewExperiment(self, request_iterator, context): + """API RPCs + """ + context.set_code(grpc.StatusCode.UNIMPLEMENTED) + context.set_details('Method not implemented!') + raise NotImplementedError('Method not implemented!') + + def GetDatasetInfo(self, request, context): + """Missing associated documentation comment in .proto file.""" + context.set_code(grpc.StatusCode.UNIMPLEMENTED) + context.set_details('Method not implemented!') + raise NotImplementedError('Method not implemented!') + + def GetTrainedModel(self, request, context): + """Missing associated documentation comment in .proto file.""" + context.set_code(grpc.StatusCode.UNIMPLEMENTED) + context.set_details('Method not implemented!') + raise NotImplementedError('Method not implemented!') + + def StreamMetrics(self, request, context): + """Missing associated documentation comment in .proto file.""" + context.set_code(grpc.StatusCode.UNIMPLEMENTED) + context.set_details('Method not implemented!') + raise NotImplementedError('Method not implemented!') + + def RemoveExperimentData(self, request, context): + """Missing associated documentation comment in .proto file.""" + context.set_code(grpc.StatusCode.UNIMPLEMENTED) + context.set_details('Method not implemented!') + raise NotImplementedError('Method not implemented!') + + def CollaboratorHealthCheck(self, request, context): + """Missing associated documentation comment in .proto file.""" + context.set_code(grpc.StatusCode.UNIMPLEMENTED) + context.set_details('Method not implemented!') + raise NotImplementedError('Method not implemented!') + + def GetEnvoys(self, request, context): + """Missing associated documentation comment in .proto file.""" + context.set_code(grpc.StatusCode.UNIMPLEMENTED) + context.set_details('Method not implemented!') + raise NotImplementedError('Method not implemented!') + + +def add_FederationDirectorServicer_to_server(servicer, server): + rpc_method_handlers = { + 'AcknowledgeShard': grpc.unary_unary_rpc_method_handler( + servicer.AcknowledgeShard, + request_deserializer=director__pb2.ShardInfo.FromString, + response_serializer=director__pb2.ShardAcknowledgement.SerializeToString, + ), + 'WaitExperiment': grpc.stream_stream_rpc_method_handler( + servicer.WaitExperiment, + request_deserializer=director__pb2.WaitExperimentRequest.FromString, + response_serializer=director__pb2.WaitExperimentResponse.SerializeToString, + ), + 'GetExperimentData': grpc.unary_stream_rpc_method_handler( + servicer.GetExperimentData, + request_deserializer=director__pb2.GetExperimentDataRequest.FromString, + response_serializer=director__pb2.ExperimentData.SerializeToString, + ), + 'SetNewExperiment': grpc.stream_unary_rpc_method_handler( + servicer.SetNewExperiment, + request_deserializer=director__pb2.ExperimentInfo.FromString, + response_serializer=director__pb2.SetNewExperimentResponse.SerializeToString, + ), + 'GetDatasetInfo': grpc.unary_unary_rpc_method_handler( + servicer.GetDatasetInfo, + request_deserializer=director__pb2.GetDatasetInfoRequest.FromString, + response_serializer=director__pb2.ShardInfo.SerializeToString, + ), + 'GetTrainedModel': grpc.unary_unary_rpc_method_handler( + servicer.GetTrainedModel, + request_deserializer=director__pb2.GetTrainedModelRequest.FromString, + response_serializer=director__pb2.TrainedModelResponse.SerializeToString, + ), + 'StreamMetrics': grpc.unary_stream_rpc_method_handler( + servicer.StreamMetrics, + request_deserializer=director__pb2.StreamMetricsRequest.FromString, + response_serializer=director__pb2.StreamMetricsResponse.SerializeToString, + ), + 'RemoveExperimentData': grpc.unary_unary_rpc_method_handler( + servicer.RemoveExperimentData, + request_deserializer=director__pb2.RemoveExperimentRequest.FromString, + response_serializer=director__pb2.RemoveExperimentResponse.SerializeToString, + ), + 'CollaboratorHealthCheck': grpc.unary_unary_rpc_method_handler( + servicer.CollaboratorHealthCheck, + request_deserializer=director__pb2.CollaboratorStatus.FromString, + response_serializer=director__pb2.CollaboratorHealthCheckResponse.SerializeToString, + ), + 'GetEnvoys': grpc.unary_unary_rpc_method_handler( + servicer.GetEnvoys, + request_deserializer=director__pb2.GetEnvoysRequest.FromString, + response_serializer=director__pb2.GetEnvoysResponse.SerializeToString, + ), + } + generic_handler = grpc.method_handlers_generic_handler( + 'FederationDirector', rpc_method_handlers) + server.add_generic_rpc_handlers((generic_handler,)) + + + # This class is part of an EXPERIMENTAL API. +class FederationDirector(object): + """Missing associated documentation comment in .proto file.""" + + @staticmethod + def AcknowledgeShard(request, + target, + options=(), + channel_credentials=None, + call_credentials=None, + insecure=False, + compression=None, + wait_for_ready=None, + timeout=None, + metadata=None): + return grpc.experimental.unary_unary(request, target, '/FederationDirector/AcknowledgeShard', + director__pb2.ShardInfo.SerializeToString, + director__pb2.ShardAcknowledgement.FromString, + options, channel_credentials, + insecure, call_credentials, compression, wait_for_ready, timeout, metadata) + + @staticmethod + def WaitExperiment(request_iterator, + target, + options=(), + channel_credentials=None, + call_credentials=None, + insecure=False, + compression=None, + wait_for_ready=None, + timeout=None, + metadata=None): + return grpc.experimental.stream_stream(request_iterator, target, '/FederationDirector/WaitExperiment', + director__pb2.WaitExperimentRequest.SerializeToString, + director__pb2.WaitExperimentResponse.FromString, + options, channel_credentials, + insecure, call_credentials, compression, wait_for_ready, timeout, metadata) + + @staticmethod + def GetExperimentData(request, + target, + options=(), + channel_credentials=None, + call_credentials=None, + insecure=False, + compression=None, + wait_for_ready=None, + timeout=None, + metadata=None): + return grpc.experimental.unary_stream(request, target, '/FederationDirector/GetExperimentData', + director__pb2.GetExperimentDataRequest.SerializeToString, + director__pb2.ExperimentData.FromString, + options, channel_credentials, + insecure, call_credentials, compression, wait_for_ready, timeout, metadata) + + @staticmethod + def SetNewExperiment(request_iterator, + target, + options=(), + channel_credentials=None, + call_credentials=None, + insecure=False, + compression=None, + wait_for_ready=None, + timeout=None, + metadata=None): + return grpc.experimental.stream_unary(request_iterator, target, '/FederationDirector/SetNewExperiment', + director__pb2.ExperimentInfo.SerializeToString, + director__pb2.SetNewExperimentResponse.FromString, + options, channel_credentials, + insecure, call_credentials, compression, wait_for_ready, timeout, metadata) + + @staticmethod + def GetDatasetInfo(request, + target, + options=(), + channel_credentials=None, + call_credentials=None, + insecure=False, + compression=None, + wait_for_ready=None, + timeout=None, + metadata=None): + return grpc.experimental.unary_unary(request, target, '/FederationDirector/GetDatasetInfo', + director__pb2.GetDatasetInfoRequest.SerializeToString, + director__pb2.ShardInfo.FromString, + options, channel_credentials, + insecure, call_credentials, compression, wait_for_ready, timeout, metadata) + + @staticmethod + def GetTrainedModel(request, + target, + options=(), + channel_credentials=None, + call_credentials=None, + insecure=False, + compression=None, + wait_for_ready=None, + timeout=None, + metadata=None): + return grpc.experimental.unary_unary(request, target, '/FederationDirector/GetTrainedModel', + director__pb2.GetTrainedModelRequest.SerializeToString, + director__pb2.TrainedModelResponse.FromString, + options, channel_credentials, + insecure, call_credentials, compression, wait_for_ready, timeout, metadata) + + @staticmethod + def StreamMetrics(request, + target, + options=(), + channel_credentials=None, + call_credentials=None, + insecure=False, + compression=None, + wait_for_ready=None, + timeout=None, + metadata=None): + return grpc.experimental.unary_stream(request, target, '/FederationDirector/StreamMetrics', + director__pb2.StreamMetricsRequest.SerializeToString, + director__pb2.StreamMetricsResponse.FromString, + options, channel_credentials, + insecure, call_credentials, compression, wait_for_ready, timeout, metadata) + + @staticmethod + def RemoveExperimentData(request, + target, + options=(), + channel_credentials=None, + call_credentials=None, + insecure=False, + compression=None, + wait_for_ready=None, + timeout=None, + metadata=None): + return grpc.experimental.unary_unary(request, target, '/FederationDirector/RemoveExperimentData', + director__pb2.RemoveExperimentRequest.SerializeToString, + director__pb2.RemoveExperimentResponse.FromString, + options, channel_credentials, + insecure, call_credentials, compression, wait_for_ready, timeout, metadata) + + @staticmethod + def CollaboratorHealthCheck(request, + target, + options=(), + channel_credentials=None, + call_credentials=None, + insecure=False, + compression=None, + wait_for_ready=None, + timeout=None, + metadata=None): + return grpc.experimental.unary_unary(request, target, '/FederationDirector/CollaboratorHealthCheck', + director__pb2.CollaboratorStatus.SerializeToString, + director__pb2.CollaboratorHealthCheckResponse.FromString, + options, channel_credentials, + insecure, call_credentials, compression, wait_for_ready, timeout, metadata) + + @staticmethod + def GetEnvoys(request, + target, + options=(), + channel_credentials=None, + call_credentials=None, + insecure=False, + compression=None, + wait_for_ready=None, + timeout=None, + metadata=None): + return grpc.experimental.unary_unary(request, target, '/FederationDirector/GetEnvoys', + director__pb2.GetEnvoysRequest.SerializeToString, + director__pb2.GetEnvoysResponse.FromString, + options, channel_credentials, + insecure, call_credentials, compression, wait_for_ready, timeout, metadata) diff --git a/openfl/protocols/federation.proto b/openfl/protocols/federation.proto index 893204990c..83c01dab97 100644 --- a/openfl/protocols/federation.proto +++ b/openfl/protocols/federation.proto @@ -1,5 +1,6 @@ // Copyright (C) 2020 Intel Corporation // Licensed subject to the terms of the separately executed evaluation license agreement between Intel Corporation and you. +syntax = "proto2"; service Aggregator { rpc GetTasks(TasksRequest) returns (TasksResponse) {} diff --git a/openfl/transport/__init__.py b/openfl/transport/__init__.py index 4953716521..e9307d443c 100644 --- a/openfl/transport/__init__.py +++ b/openfl/transport/__init__.py @@ -5,8 +5,10 @@ from .grpc import AggregatorGRPCServer from .grpc import CollaboratorGRPCClient +from .grpc import DirectorGRPCServer __all__ = [ 'AggregatorGRPCServer', - 'CollaboratorGRPCClient' + 'CollaboratorGRPCClient', + 'DirectorGRPCServer', ] diff --git a/openfl/transport/grpc/__init__.py b/openfl/transport/grpc/__init__.py index bfe365a265..2258bb30aa 100644 --- a/openfl/transport/grpc/__init__.py +++ b/openfl/transport/grpc/__init__.py @@ -4,9 +4,11 @@ """openfl.transport.grpc package.""" from .client import CollaboratorGRPCClient +from .director_server import DirectorGRPCServer from .server import AggregatorGRPCServer __all__ = [ 'AggregatorGRPCServer', 'CollaboratorGRPCClient', + 'DirectorGRPCServer', ] diff --git a/openfl/transport/grpc/client.py b/openfl/transport/grpc/client.py index 70f4d8ee9f..a769b9c781 100644 --- a/openfl/transport/grpc/client.py +++ b/openfl/transport/grpc/client.py @@ -56,6 +56,7 @@ def _intercept_call(self, continuation, client_call_details, request_or_iterator if isinstance(response, grpc.RpcError): # If status code is not in retryable status codes + self.sleeping_policy.logger.info(f'Response code: {response.code()}') if ( self.status_for_retry and response.code() not in self.status_for_retry diff --git a/openfl/transport/grpc/director_client.py b/openfl/transport/grpc/director_client.py new file mode 100644 index 0000000000..ac6c587f93 --- /dev/null +++ b/openfl/transport/grpc/director_client.py @@ -0,0 +1,286 @@ +# Copyright (C) 2020-2021 Intel Corporation +# SPDX-License-Identifier: Apache-2.0 + +"""Director clients module.""" + +import logging +import os +import shutil +import time +from datetime import datetime +from subprocess import check_call +from sys import executable + +import grpc + +from openfl.pipelines import NoCompressionPipeline +from openfl.protocols import director_pb2 +from openfl.protocols import director_pb2_grpc +from openfl.protocols.utils import construct_model_proto +from openfl.protocols.utils import deconstruct_model_proto + + +logger = logging.getLogger(__name__) + + +class ShardDirectorClient: + """The internal director client class.""" + + def __init__(self, director_uri, shard_name, disable_tls=False, + root_ca=None, key=None, cert=None) -> None: + """Initialize a shard director client object.""" + self.shard_name = shard_name + options = [('grpc.max_message_length', 100 1024 * 1024)] + if disable_tls: + channel = grpc.insecure_channel(director_uri, options=options) + else: + if not (root_ca and key and cert): + raise Exception('No certificates provided') + try: + with open(root_ca, 'rb') as f: + root_ca_b = f.read() + with open(key, 'rb') as f: + key_b = f.read() + with open(cert, 'rb') as f: + cert_b = f.read() + except FileNotFoundError as exc: + raise Exception(f'Provided certificate file is not exist: {exc.filename}') + + credentials = grpc.ssl_channel_credentials( + root_certificates=root_ca_b, + private_key=key_b, + certificate_chain=cert_b + ) + channel = grpc.secure_channel(director_uri, credentials, options=options) + self.stub = director_pb2_grpc.FederationDirectorStub(channel) + + def report_shard_info(self, shard_descriptor) -> bool: + """Report shard info to the director.""" + logger.info('Send report AcknowledgeShard') + # True considered as successful registration + shard_info = director_pb2.ShardInfo( + shard_description=shard_descriptor.dataset_description, + n_samples=len(shard_descriptor), + sample_shape=shard_descriptor.sample_shape, + target_shape=shard_descriptor.target_shape + ) + + shard_info.node_info.CopyFrom(self._get_node_info()) + + acknowledgement = self.stub.AcknowledgeShard(shard_info) + return acknowledgement.accepted + + def get_experiment_data(self): + """Get an experiment data from the director.""" + logger.info('Send WaitExperiment request') + response_iter = self.stub.WaitExperiment(self._get_experiment_data()) + logger.info('WaitExperiment response has received') + response = next(response_iter) + experiment_name = response.experiment_name + if not experiment_name: + raise Exception('No experiment') + logger.info(f'Request experiment {experiment_name}') + request = director_pb2.GetExperimentDataRequest( + experiment_name=experiment_name, + collaborator_name=self.shard_name + ) + response_iter = self.stub.GetExperimentData(request) + + self.create_workspace(experiment_name, response_iter) + + return experiment_name + + @staticmethod + def remove_workspace(experiment_name): + """Remove the workspace.""" + shutil.rmtree(experiment_name, ignore_errors=True) + + @staticmethod + def create_workspace(experiment_name, response_iter): + """Create a collaborator workspace for the experiment.""" + if os.path.exists(experiment_name): + shutil.rmtree(experiment_name) + os.makedirs(experiment_name) + + arch_name = f'{experiment_name}/{experiment_name}' + '.zip' + logger.info(f'arch_name: {arch_name}') + with open(arch_name, 'wb') as content_file: + for response in response_iter: + logger.info(f'Size: {response.size}') + if response.size == len(response.npbytes): + content_file.write(response.npbytes) + else: + raise Exception('Broken archive') + + shutil.unpack_archive(arch_name, experiment_name) + os.remove(arch_name) + + requirements_filename = f'./{experiment_name}/requirements.txt' + + if os.path.isfile(requirements_filename): + attempts = 3 + for _ in range(attempts): + try: + check_call([ + executable, '-m', 'pip', 'install', '-r', requirements_filename], + shell=False) + except Exception as exc: + logger.error(f'Failed to install requirements: {exc}') + time.sleep(3) + else: + break + else: + logger.error('No ' + requirements_filename + ' file found.') + + def _get_experiment_data(self): + """Generate the experiment data request.""" + yield director_pb2.WaitExperimentRequest(collaborator_name=self.shard_name) + + def _get_node_info(self): + """Generate a node info message.""" + return director_pb2.NodeInfo(name=self.shard_name) + + def send_health_check(self, collaborator_name, is_experiment_running, valid_duration): + """Send envoy health check.""" + status = director_pb2.CollaboratorStatus( + name=collaborator_name, + is_experiment_running=is_experiment_running, + ) + status.valid_duration.seconds = valid_duration + logger.debug(f'Sending health check status: {status}') + + return self.stub.CollaboratorHealthCheck(status) + + +class DirectorClient: + """Director client class for users.""" + + def __init__(self, client_id, director_uri, disable_tls=False, + root_ca=None, key=None, cert=None) -> None: + """Initialize director client object.""" + channel_opt = [('grpc.max_send_message_length', 512 * 1024 * 1024), + ('grpc.max_receive_message_length', 512 * 1024 * 1024)] + if disable_tls: + channel = grpc.insecure_channel(director_uri, options=channel_opt) + else: + if not (root_ca and key and cert): + raise Exception('No certificates provided') + try: + with open(root_ca, 'rb') as f: + root_ca_b = f.read() + with open(key, 'rb') as f: + key_b = f.read() + with open(cert, 'rb') as f: + cert_b = f.read() + except FileNotFoundError as exc: + raise Exception(f'Provided certificate file is not exist: {exc.filename}') + + credentials = grpc.ssl_channel_credentials( + root_certificates=root_ca_b, + private_key=key_b, + certificate_chain=cert_b + ) + + channel = grpc.secure_channel(director_uri, credentials, options=channel_opt) + self.stub = director_pb2_grpc.FederationDirectorStub(channel) + + self.client_id = client_id + self.header = director_pb2.RequestHeader(sender=self.client_id) + + def set_new_experiment(self, name, col_names, arch_path, + initial_tensor_dict=None): + """Send the new experiment to director to launch.""" + logger.info('SetNewExperiment') + model_proto = None + if initial_tensor_dict: + model_proto = construct_model_proto(initial_tensor_dict, 0, NoCompressionPipeline()) + + with open(arch_path, 'rb') as arch: + def st(): + max_buffer_size = (2 * 1024 * 1024) + chunk = arch.read(max_buffer_size) + while chunk != b'': + if not chunk: + raise StopIteration + # TODO: add hash or/and size to check + experiment_info = director_pb2.ExperimentInfo( + header=self.header, + name=name, + collaborator_names=col_names, + model_proto=model_proto + ) + experiment_info.experiment_data.size = len(chunk) + experiment_info.experiment_data.npbytes = chunk + yield experiment_info + chunk = arch.read(max_buffer_size) + + resp = self.stub.SetNewExperiment(st()) + return resp + + def get_dataset_info(self): + """Request the dataset info from the director.""" + resp = self.stub.GetDatasetInfo(director_pb2.GetDatasetInfoRequest(header=self.header)) + return resp.sample_shape, resp.target_shape + + def _get_trained_model(self, experiment_name, model_type): + """Get trained model RPC.""" + get_model_request = director_pb2.GetTrainedModelRequest( + header=self.header, + experiment_name=experiment_name, + model_type=model_type) + model_proto_response = self.stub.GetTrainedModel(get_model_request) + tensor_dict, _ = deconstruct_model_proto( + model_proto_response.model_proto, + NoCompressionPipeline() + ) + return tensor_dict + + def get_best_model(self, experiment_name): + """Get best model method.""" + model_type = director_pb2.GetTrainedModelRequest.BEST_MODEL + return self._get_trained_model(experiment_name, model_type) + + def get_last_model(self, experiment_name): + """Get last model method.""" + model_type = director_pb2.GetTrainedModelRequest.LAST_MODEL + return self._get_trained_model(experiment_name, model_type) + + def stream_metrics(self, experiment_name): + """Stream metrics RPC.""" + request = director_pb2.StreamMetricsRequest( + header=self.header, + experiment_name=experiment_name) + for metric_message in self.stub.StreamMetrics(request): + yield { + 'metric_origin': metric_message.metric_origin, + 'task_name': metric_message.task_name, + 'metric_name': metric_message.metric_name, + 'metric_value': metric_message.metric_value, + 'round': metric_message.round} + + def remove_experiment_data(self, name): + """Remove experiment data RPC.""" + request = director_pb2.RemoveExperimentRequest( + header=self.header, + experiment_name=name) + response = self.stub.RemoveExperimentData(request) + return response.acknowledgement + + def get_envoys(self, raw_result=False): + """Get envoys info.""" + envoys = self.stub.GetEnvoys(director_pb2.GetEnvoysRequest()) + if raw_result: + return envoys + now = datetime.now().strftime('%Y-%m-%d %H:%M:%S') + result = {} + for envoy in envoys.envoy_infos: + result[envoy.shard_info.node_info.name] = { + 'shard_info': envoy.shard_info, + 'is_online': envoy.is_online or False, + 'is_experiment_running': envoy.is_experiment_running or False, + 'last_updated': datetime.fromtimestamp( + envoy.last_updated.seconds).strftime('%Y-%m-%d %H:%M:%S'), + 'current_time': now, + 'valid_duration': envoy.valid_duration + } + return result diff --git a/openfl/transport/grpc/director_server.py b/openfl/transport/grpc/director_server.py new file mode 100644 index 0000000000..a802b35f44 --- /dev/null +++ b/openfl/transport/grpc/director_server.py @@ -0,0 +1,244 @@ +# Copyright (C) 2020-2021 Intel Corporation +# SPDX-License-Identifier: Apache-2.0 + +"""Director server.""" + +import asyncio +import logging +from pathlib import Path + +from grpc import aio +from grpc import ssl_server_credentials + +from openfl.pipelines import NoCompressionPipeline +from openfl.protocols import director_pb2 +from openfl.protocols import director_pb2_grpc +from openfl.protocols.utils import construct_model_proto +from openfl.protocols.utils import deconstruct_model_proto + +logger = logging.getLogger(__name__) + + +class DirectorGRPCServer(director_pb2_grpc.FederationDirectorServicer): + """Director transport class.""" + + def __init__(self, , director_cls, disable_tls: bool = False, + root_ca: str = None, key: str = None, cert: str = None, + listen_ip='[::]', listen_port=50051, kwargs) -> None: + """Initialize a director object.""" + # TODO: add working directory + super().__init__() + + self.listen_addr = f'{listen_ip}:{listen_port}' + self.disable_tls = disable_tls + self.root_ca = None + self.key = None + self.cert = None + self._fill_certs(root_ca, key, cert) + self.server = None + self.director = director_cls( + disable_tls=self.disable_tls, + root_ca=self.root_ca, + key=self.key, + cert=self.cert, + kwargs + ) + + def _fill_certs(self, root_ca, key, cert): + """Fill certificates.""" + if not self.disable_tls: + if not (root_ca and key and cert): + raise Exception('No certificates provided') + self.root_ca = Path(root_ca).absolute() + self.key = Path(key).absolute() + self.cert = Path(cert).absolute() + + def validate_caller(self, request, context): + """ + Validate the caller. + + Args: + request: The gRPC message request + context: The gRPC context + + Returns: + True - if the caller has valid cert + False - if the callers cert name is invalid + """ + # Can we close the request right here? + if not self.disable_tls: + caller_cert_name = context.auth_context()['x509_common_name'][0].decode('utf-8') + caller_common_name = request.header.sender + if caller_cert_name != caller_common_name: + return False + + return True + + def start(self): + """Launch the director GRPC server.""" + asyncio.run(self._run()) + + async def _run(self): + channel_opt = [('grpc.max_send_message_length', 512 1024 * 1024), + ('grpc.max_receive_message_length', 512 * 1024 * 1024)] + self.server = aio.server(options=channel_opt) + director_pb2_grpc.add_FederationDirectorServicer_to_server(self, self.server) + + if self.disable_tls: + self.server.add_insecure_port(self.listen_addr) + else: + with open(self.key, 'rb') as f: + key_b = f.read() + with open(self.cert, 'rb') as f: + cert_b = f.read() + with open(self.root_ca, 'rb') as f: + root_ca_b = f.read() + server_credentials = ssl_server_credentials( + ((key_b, cert_b),), + root_certificates=root_ca_b, + require_client_auth=True + ) + self.server.add_secure_port(self.listen_addr, server_credentials) + logger.info(f'Starting server on {self.listen_addr}') + await self.server.start() + await self.server.wait_for_termination() + + async def AcknowledgeShard(self, shard_info, context): # NOQA:N802 + """Receive acknowledge shard info.""" + logger.info(f'AcknowledgeShard request has got: {shard_info}') + is_accepted = self.director.acknowledge_shard(shard_info) + reply = director_pb2.ShardAcknowledgement(accepted=is_accepted) + + return reply + + async def SetNewExperiment(self, stream, context): # NOQA:N802 + """Request to set new experiment.""" + logger.info(f'SetNewExperiment request has got {stream}') + # TODO: add streaming reader + npbytes = b'' + async for request in stream: + if request.experiment_data.size == len(request.experiment_data.npbytes): + npbytes += request.experiment_data.npbytes + else: + raise Exception('Bad request') + + if not self.validate_caller(request, context): + # Can we send reject before reading the stream? + return director_pb2.SetNewExperimentResponse(accepted=False) + tensor_dict = None + if request.model_proto: + tensor_dict, _ = deconstruct_model_proto(request.model_proto, NoCompressionPipeline()) + + is_accepted = await self.director.set_new_experiment( + experiment_name=request.name, + sender_name=request.header.sender, + tensor_dict=tensor_dict, + collaborator_names=request.collaborator_names, + data=npbytes + ) + + logger.info('Send response') + return director_pb2.SetNewExperimentResponse(accepted=is_accepted) + + async def GetTrainedModel(self, request, context): # NOQA:N802 + """RPC for retrieving trained models.""" + if not self.validate_caller(request, context): + return director_pb2.TrainedModelResponse() + logger.info('Request GetTrainedModel has got!') + + best_tensor_dict, last_tensor_dict = await self.director.get_trained_model( + request.experiment_name, + request.header.sender + ) + + if request.model_type == director_pb2.GetTrainedModelRequest.BEST_MODEL: + tensor_dict = best_tensor_dict + elif request.model_type == director_pb2.GetTrainedModelRequest.LAST_MODEL: + tensor_dict = last_tensor_dict + else: + logger.error('Incorrect model type') + return director_pb2.TrainedModelResponse() + if not tensor_dict: + return director_pb2.TrainedModelResponse() + + model_proto = construct_model_proto(tensor_dict, 0, NoCompressionPipeline()) + + return director_pb2.TrainedModelResponse(model_proto=model_proto) + + async def GetExperimentData(self, request, context): # NOQA:N802 + """Receive experiment data.""" + # TODO: add size filling + # TODO: add experiment name field + # TODO: rename npbytes to data + content = self.director.get_experiment_data(request.experiment_name) + logger.info(f'Content length: {len(content)}') + max_buffer_size = (2 * 1024 * 1024) + + for i in range(0, len(content), max_buffer_size): + chunk = content[i:i + max_buffer_size] + logger.info(f'Send {len(chunk)} bytes') + yield director_pb2.ExperimentData(size=len(chunk), npbytes=chunk) + + async def WaitExperiment(self, request_iterator, context): # NOQA:N802 + """Request for wait an experiment.""" + logger.info('Request WaitExperiment has got!') + async for msg in request_iterator: + logger.info(msg) + experiment_name = await self.director.wait_experiment(msg.collaborator_name) + logger.info(f'Experiment {experiment_name} was prepared') + + yield director_pb2.WaitExperimentResponse(experiment_name=experiment_name) + + async def GetDatasetInfo(self, request, context): # NOQA:N802 + """Request the info about target and sample shapes in the dataset.""" + logger.info('Request GetDatasetInfo has got!') + if not self.validate_caller(request, context): + return director_pb2.ShardInfo() + + sample_shape, target_shape = self.director.get_dataset_info() + resp = director_pb2.ShardInfo( + sample_shape=sample_shape, + target_shape=target_shape + ) + return resp + + async def StreamMetrics(self, request, context): # NOQA:N802 + """Request to stream metrics from the aggregator to frontend.""" + if not self.validate_caller(request, context): + return + # We should probably set a name to the aggregator and verify it here. + # Moreover, we may save the experiment name in plan.yaml and retrieve it + # during the aggregator initialization + logger.info(f'Request StreamMetrics for {request.experiment_name} experiment has got!') + + async for message in \ + self.director.stream_metrics(request.experiment_name, request.header.sender): + yield message + + async def RemoveExperimentData(self, request, context): # NOQA:N802 + """Remove experiment data RPC.""" + response = director_pb2.RemoveExperimentResponse(acknowledgement=False) + if not self.validate_caller(request, context): + return response + + self.director.remove_experiment_data(request.experiment_name, request.header.sender) + + response.acknowledgement = True + return response + + async def CollaboratorHealthCheck(self, request, context): # NOQA:N802 + """Accept health check from envoy.""" + logger.debug(f'Request CollaboratorHealthCheck has got: {request}') + is_accepted = self.director.collaborator_health_check( + collaborator_name=request.name, + is_experiment_running=request.is_experiment_running, + valid_duration=request.valid_duration.seconds, + ) + + return director_pb2.CollaboratorHealthCheckResponse(accepted=is_accepted) + + async def GetEnvoys(self, request, context): # NOQA:N802 + """Get a status information about envoys.""" + envoy_infos = self.director.get_envoys() + + return director_pb2.GetEnvoysResponse(envoy_infos=envoy_infos) diff --git a/openfl/transport/grpc/server.py b/openfl/transport/grpc/server.py index 801a11f25f..ce58ee1270 100644 --- a/openfl/transport/grpc/server.py +++ b/openfl/transport/grpc/server.py @@ -205,8 +205,8 @@ def SendLocalTaskResults(self, request, context): # NOQA:N802 # turn data stream into local model update return Acknowledgement(header=self.get_header(collaborator_name)) - def serve(self): - """Start an aggregator gRPC service.""" + def get_server(self): + """Return gRPC server.""" self.server = server(ThreadPoolExecutor(max_workers=cpu_count()), options=self.channel_options) @@ -216,8 +216,8 @@ def serve(self): self.logger.warn( 'gRPC is running on insecure channel with TLS disabled.') - - self.server.add_insecure_port(self.uri) + port = self.server.add_insecure_port(self.uri) + self.logger.info(f'Insecure port: {port}') else: @@ -239,8 +239,13 @@ def serve(self): self.server.add_secure_port(self.uri, self.server_credentials) - self.logger.info('Starting Aggregator gRPC Server') + return self.server + def serve(self): + """Start an aggregator gRPC service.""" + self.get_server() + + self.logger.info('Starting Aggregator gRPC Server') self.server.start() try: diff --git a/openfl/utilities/ca.py b/openfl/utilities/ca.py new file mode 100644 index 0000000000..5371635291 --- /dev/null +++ b/openfl/utilities/ca.py @@ -0,0 +1,18 @@ +# Copyright (C) 2020-2021 Intel Corporation +# SPDX-License-Identifier: Apache-2.0 +"""Generic check functions.""" +import os + + +def get_credentials(folder_path): + """Get credentials from folder by template.""" + root_ca, key, cert = None, None, None + if os.path.exists(folder_path): + for f in os.listdir(folder_path): + if '.key' in f: + key = folder_path + '/' + f + if '.crt' in f and 'root_ca' not in f: + cert = folder_path + '/' + f + if 'root_ca' in f: + root_ca = folder_path + '/' + f + return root_ca, key, cert diff --git a/openfl/utilities/logs.py b/openfl/utilities/logs.py new file mode 100644 index 0000000000..3507042058 --- /dev/null +++ b/openfl/utilities/logs.py @@ -0,0 +1,38 @@ +# Copyright (C) 2020-2021 Intel Corporation +# SPDX-License-Identifier: Apache-2.0 + +"""Logs utilities.""" + +import logging + +from rich.console import Console +from rich.logging import RichHandler +from tensorboardX import SummaryWriter + +writer = None + + +def get_writer(): + """Create global writer object.""" + global writer + if not writer: + writer = SummaryWriter('./logs/cnn_mnist', flush_secs=5) + + +def write_metric(node_name, task_name, metric_name, metric, round_number): + """Write metric callback.""" + get_writer() + writer.add_scalar(f'{node_name}/{task_name}/{metric_name}', metric, round_number) + + +def setup_loggers(log_level=logging.INFO): + """Configure loggers.""" + root = logging.getLogger() + root.setLevel(log_level) + console = Console(width=160) + handler = RichHandler(console=console) + formatter = logging.Formatter( + '[%(asctime)s][%(name)s][%(levelname)s] - %(message)s' + ) + handler.setFormatter(formatter) + root.addHandler(handler) diff --git a/openfl/utilities/utils.py b/openfl/utilities/utils.py index 05d55cb181..40cab8fb40 100644 --- a/openfl/utilities/utils.py +++ b/openfl/utilities/utils.py @@ -1,7 +1,10 @@ # Copyright (C) 2020-2021 Intel Corporation # SPDX-License-Identifier: Apache-2.0 """Utilities module.""" + +import logging import os +import shutil from socket import getfqdn import numpy as np @@ -23,6 +26,62 @@ def getfqdn_env(name: str = '') -> str: return getfqdn(name) +def add_log_level(level_name, level_num, method_name=None): + """ + Add a new logging level to the logging module. + + Args: + level_name: name of log level. + level_num: log level value. + method_name: log method wich will use new log level (default = level_name.lower()) + + """ + if not method_name: + method_name = level_name.lower() + + def log_for_level(self, message, args, kwargs): + if self.isEnabledFor(level_num): + self._log(level_num, message, args, kwargs) + + def log_to_root(message, args, *kwargs): + logging.log(level_num, message, args, *kwargs) + + logging.addLevelName(level_num, level_name) + setattr(logging, level_name, level_num) + setattr(logging.getLoggerClass(), method_name, log_for_level) + setattr(logging, method_name, log_to_root) + + +class UnpackWorkspace: + """Workspace context manager. + + It unpacks the archive at the begining of the experiment and removes the archive. + Then in moving to the workspace's folder. + Exiting it removes the workspace + """ + + def __init__(self, archive_path) -> None: + """Initialize workspace context manager.""" + self.cwd = os.getcwd() + + self.workspace_name = os.path.basename(archive_path).split('.')[0] + if os.path.exists(self.workspace_name): + shutil.rmtree(self.workspace_name) + os.makedirs(self.workspace_name) + shutil.unpack_archive(archive_path, self.workspace_name) + + os.remove(archive_path) + + def __enter__(self): + """Enter workspace context manager.""" + os.chdir(f'{self.cwd}/{self.workspace_name}') + + def __exit__(self, exc_type, exc_value, traceback): + """Exit workspace context manager.""" + os.chdir(self.cwd) + shutil.rmtree(self.workspace_name) + + def split_tensor_dict_into_floats_and_non_floats(tensor_dict): """ Split the tensor dictionary into float and non-floating point values. diff --git a/setup.cfg b/setup.cfg index 7d2b39a206..ae06205e93 100644 --- a/setup.cfg +++ b/setup.cfg @@ -7,7 +7,7 @@ ignore = W503, N812 inline-quotes = ' multiline-quotes = ' docstring-quotes = """ -exclude = _pb2* +exclude = _pb2,tests/github/interactive_api,tests/github/interactive_api_director max-line-length = 99 avoid-escape = False import-order-style = smarkets diff --git a/setup.py b/setup.py index 349cc12b6b..20531b565c 100644 --- a/setup.py +++ b/setup.py @@ -26,6 +26,9 @@ 'openfl.component.assigner', 'openfl.component.aggregator', 'openfl.component.collaborator', + 'openfl.component.envoy', + 'openfl.component.director', + 'openfl.component.ca', 'openfl.utilities', 'openfl.utilities.optimizers.torch', 'openfl.utilities.optimizers.keras', @@ -53,8 +56,8 @@ 'numpy', 'pandas', 'protobuf', - 'grpcio==1.34.0', - 'grpcio-tools==1.34.0', + 'grpcio~=1.34.0', + 'grpcio-tools~=1.34.0', 'rich==9.1.0', 'tqdm', 'scikit-learn', @@ -64,6 +67,9 @@ 'flatten_json', 'cryptography>=3.4.6', 'cloudpickle', + 'tensorboardX', + 'tensorboard', + 'requests' ], python_requires='>=3.6, <3.9', project_urls={	Scalable PKIs In avoid manual copying for current PKIs/certificate exchange between Aggregator & Collaborators we need an automatic system for that.		2021-06-04T07:41:00	[]	[]
securefederatedai/openfl	securefederatedai__openfl-62	6a1ea6979dad2f181a000874f1cbf2004e18652f	diff --git a/docs/advanced_topics.rst b/docs/advanced_topics.rst index 7506d2d0a2..2c16c5e7c0 100644 --- a/docs/advanced_topics.rst +++ b/docs/advanced_topics.rst @@ -12,5 +12,6 @@ Advanced Topics multiple_plans compression_settings + overriding_agg_fn diff --git a/docs/overriding_agg_fn.rst b/docs/overriding_agg_fn.rst new file mode 100644 index 0000000000..a4c7d0de82 --- /dev/null +++ b/docs/overriding_agg_fn.rst @@ -0,0 +1,76 @@ +.. # Copyright (C) 2020-2021 Intel Corporation +.. # SPDX-License-Identifier: Apache-2.0 + +.. _overriding_agg_fn: + +Overriding aggregation function +####################### + +Usage +===================== +\|productName\| allows developers to use custom aggregation functions per task. +In order to do this, you should create an implementation of :class:`openfl.component.aggregation_functions.AggregationFunctionInterface` +and pass it as ``tasks.{task_name}.aggregation_type`` parameter of ``override_config`` keyword argument of :func:`openfl.native.run_experiment` native function. + +.. warning:: + + Currently custom aggregation functions supported in native API (``fx.run_experiment(...)``) only. + Command Line Interface always uses FedAvg and this behavior cannot be overriden due to disability of serialization and deserialization of complex Python objects in Plan files. + +``AggregationFunctionInterface`` requires a single ``call`` function. +This function receives tensors for a single parameter from multiple collaborators with additional metadata (see definition of :meth:`openfl.component.aggregation_functions.AggregationFunctionInterface.call`) and returns a single tensor that represents the result of aggregation. + +.. note:: + By default, we use weighted averaging with weights equal to data parts (FedAvg). + +Example +======================= + +Below is an example of custom tensor clipping aggregation function that multiplies all local tensors by 0.3 and averages them according to weights equal to data parts to produce the resulting global tensor. + +.. code-block:: python + + from openfl.component.aggregation_functions import AggregationFunctionInterface + import numpy as np + + class ClippedAveraging(AggregationFunctionInterface): + def __init__(self, ratio): + self.ratio = ratio + + def call(self, + local_tensors, + db_iterator, + tensor_name, + fl_round, + __): + """Aggregate tensors. + + Args: + local_tensors(list[openfl.utilities.LocalTensor]): List of local tensors to aggregate. + db_iterator: iterator over history of all tensors. + Columns: ['tensor_name', 'round', 'tags', 'nparray'] + tensor_name: name of the tensor + fl_round: round number + tags: tuple of tags for this tensor + """ + clipped_tensors = [] + previous_tensor_value = None + for record in db_iterator: + if ( + record['round'] == (fl_round - 1) + and record['tensor_name'] == tensor_name + and 'aggregated' in record['tags'] + and 'delta' not in record['tags'] + ): + previous_tensor_value = record['nparray'] + weights = [] + for local_tensor in local_tensors: + prev_tensor = previous_tensor_value if previous_tensor_value is not None else local_tensor.tensor + delta = local_tensor.tensor - prev_tensor + new_tensor = prev_tensor + delta self.ratio + clipped_tensors.append(new_tensor) + weights.append(local_tensor.weight) + + return np.average(clipped_tensors, weights=weights, axis=0) + +Full implementation can be found at ``openfl-tutorials/Federated_Pytorch_MNIST_custom_aggregation_Tutorial.ipynb`` \ No newline at end of file diff --git a/docs/running_the_federation.notebook.rst b/docs/running_the_federation.notebook.rst index 22f2cb4738..eecbe0c376 100644 --- a/docs/running_the_federation.notebook.rst +++ b/docs/running_the_federation.notebook.rst @@ -47,7 +47,7 @@ At this point you may be wondering what goes into a FL.Plan, and how you can cus .. code-block:: python - print(json.dumps(fx.get_plan(), indent=4, sort_keys=True)) + print(fx.get_plan()) { "aggregator.settings.best_state_path": "save/keras_cnn_mnist_best.pbuf", diff --git a/openfl-tutorials/Federated_Keras_MNIST_Tutorial.ipynb b/openfl-tutorials/Federated_Keras_MNIST_Tutorial.ipynb index bec5ee5ad5..cb570c4f4a 100644 --- a/openfl-tutorials/Federated_Keras_MNIST_Tutorial.ipynb +++ b/openfl-tutorials/Federated_Keras_MNIST_Tutorial.ipynb @@ -205,8 +205,7 @@ "outputs": [], "source": [ "#Get the current values of the plan. Each of these can be overridden\n", - "import json\n", - "print(json.dumps(fx.get_plan(), indent=4, sort_keys=True))" + "print(fx.get_plan())" ] }, { diff --git a/openfl-tutorials/Federated_PyTorch_UNET_Tutorial.ipynb b/openfl-tutorials/Federated_PyTorch_UNET_Tutorial.ipynb index 378e624470..2a7c13a923 100644 --- a/openfl-tutorials/Federated_PyTorch_UNET_Tutorial.ipynb +++ b/openfl-tutorials/Federated_PyTorch_UNET_Tutorial.ipynb @@ -9,9 +9,21 @@ }, { "cell_type": "code", - "execution_count": null, + "execution_count": 1, "metadata": {}, - "outputs": [], + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "Requirement already satisfied: torch in /home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages (1.8.1)\n", + "Requirement already satisfied: typing-extensions in /home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages (from torch) (3.7.4.3)\n", + "Requirement already satisfied: numpy in /home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages (from torch) (1.18.5)\n", + "\u001b[33mWARNING: You are using pip version 21.0.1; however, version 21.1 is available.\n", + "You should consider upgrading via the '/home/itrushkin/.virtualenvs/openfl_research/bin/python -m pip install --upgrade pip' command.\u001b[0m\n" + ] + } + ], "source": [ "# Install dependencies if not already installed\n", "!pip install torch" @@ -26,9 +38,111 @@ }, { "cell_type": "code", - "execution_count": null, + "execution_count": 2, "metadata": {}, - "outputs": [], + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "Creating Workspace Directories\n", + "Creating Workspace Templates\n", + "Successfully installed packages from /home/itrushkin/.local/workspace/requirements.txt.\n", + "\n", + "New workspace directory structure:\n", + "workspace\n", + "├── plan\n", + "│ ├── defaults\n", + "│ │ ├── data_loader.yaml\n", + "│ │ ├── network.yaml\n", + "│ │ ├── tasks_keras.yaml\n", + "│ │ ├── collaborator.yaml\n", + "│ │ ├── assigner.yaml\n", + "│ │ ├── task_runner.yaml\n", + "│ │ ├── defaults\n", + "│ │ ├── tasks_tensorflow.yaml\n", + "│ │ ├── aggregator.yaml\n", + "│ │ ├── tasks_torch.yaml\n", + "│ │ └── tasks_fast_estimator.yaml\n", + "│ ├── plan.yaml\n", + "│ ├── data.yaml\n", + "│ └── cols.yaml\n", + "├── logs\n", + "├── agg_to_col_two_signed_cert.zip\n", + "├── data\n", + "│ └── MNIST\n", + "│ ├── raw\n", + "│ └── processed\n", + "├── cert\n", + "│ ├── ca\n", + "│ │ ├── signing-ca.crt\n", + "│ │ ├── root-ca.crt\n", + "│ │ ├── signing-ca\n", + "│ │ ├── signing-ca.csr\n", + "│ │ └── root-ca\n", + "│ ├── server\n", + "│ │ ├── agg_nnlicv611.inn.intel.com.csr\n", + "│ │ ├── agg_nnlicv611.inn.intel.com.key\n", + "│ │ └── agg_nnlicv611.inn.intel.com.crt\n", + "│ ├── client\n", + "│ │ ├── col_two.crt\n", + "│ │ ├── col_one.crt\n", + "│ │ ├── col_one.key\n", + "│ │ └── col_two.key\n", + "│ ├── cert_chain.crt\n", + "│ ├── col_one.csr\n", + "│ └── col_two.csr\n", + "├── code\n", + "│ ├── __init__.py\n", + "│ ├── data_loader.py\n", + "│ ├── mnist_utils.py\n", + "│ ├── fed_unet_runner.py\n", + "│ ├── ptmnist_inmemory.py\n", + "│ ├── pt_cnn.py\n", + "│ └── pt_unet_parts.py\n", + "├── final_pytorch_model\n", + "├── .workspace\n", + "├── requirements.txt\n", + "├── agg_to_col_one_signed_cert.zip\n", + "└── save\n", + " ├── torch_cnn_mnist_init.pbuf\n", + " ├── torch_cnn_mnist_best.pbuf\n", + " └── torch_cnn_mnist_last.pbuf\n", + "\n", + "15 directories, 42 files\n", + "Setting Up Certificate Authority...\n", + "\n", + "1. Create Root CA\n", + "1.1 Create Directories\n", + "1.2 Create Database\n", + "1.3 Create CA Request and Certificate\n", + "2. Create Signing Certificate\n", + "2.1 Create Directories\n", + "2.2 Create Database\n", + "2.3 Create Signing Certificate CSR\n", + "2.4 Sign Signing Certificate CSR\n", + "3 Create Certificate Chain\n", + "\n", + "Done.\n", + "Creating AGGREGATOR certificate key pair with following settings: CN=\u001b[31mnnlicv611.inn.intel.com\u001b[0m, SAN=\u001b[31mDNS:nnlicv611.inn.intel.com\u001b[0m\n", + " Writing AGGREGATOR certificate key pair to: \u001b[32mcert/server\u001b[0m\n", + "The CSR Hash for file \u001b[32mserver/agg_nnlicv611.inn.intel.com.csr\u001b[0m = \u001b[31mcabebd7a61c27b859c152be3573d77664096762d25e616294f56273fe65bd21d9afe883f2e745c4784a1fa10620d79d9\u001b[0m\n", + " Signing AGGREGATOR certificate\n", + "Creating COLLABORATOR certificate key pair with following settings: CN=\u001b[31mone\u001b[0m, SAN=\u001b[31mDNS:one\u001b[0m\n", + " Moving COLLABORATOR certificate to: \u001b[32mcert/col_one\u001b[0m\n", + "The CSR Hash for file \u001b[32mcol_one.csr\u001b[0m = \u001b[31m5e39b9007a7cedcaab3330907cbf1f471ae221483b2810de42483aec960171e8a95a3e67dd7985477592c873bf8876e2\u001b[0m\n", + " Signing COLLABORATOR certificate\n", + "\n", + "Registering \u001b[32mone\u001b[0m in \u001b[32mplan/cols.yaml\u001b[0m\n", + "Creating COLLABORATOR certificate key pair with following settings: CN=\u001b[31mtwo\u001b[0m, SAN=\u001b[31mDNS:two\u001b[0m\n", + " Moving COLLABORATOR certificate to: \u001b[32mcert/col_two\u001b[0m\n", + "The CSR Hash for file \u001b[32mcol_two.csr\u001b[0m = \u001b[31m54f92e51abdfa46fc18339b0c59f44af5bd55ce3c562daf0254208b49545db231c8ea963e0b2383272ea74a5aa2faa3f\u001b[0m\n", + " Signing COLLABORATOR certificate\n", + "\n", + "Registering \u001b[32mtwo\u001b[0m in \u001b[32mplan/cols.yaml\u001b[0m\n" + ] + } + ], "source": [ "import openfl.native as fx\n", "\n", @@ -38,7 +152,7 @@ }, { "cell_type": "code", - "execution_count": null, + "execution_count": 3, "metadata": { "code_folding": [] }, @@ -73,9 +187,2069 @@ }, { "cell_type": "code", - "execution_count": null, + "execution_count": 4, "metadata": {}, - "outputs": [], + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "--2021-04-29 17:42:32-- https://datasets.simula.no/hyper-kvasir/hyper-kvasir-segmented-images.zip\n", + "Resolving proxy-dmz.intel.com (proxy-dmz.intel.com)... 10.102.248.16\n", + "Connecting to proxy-dmz.intel.com (proxy-dmz.intel.com)\|10.102.248.16\|:912... connected.\n", + "Proxy request sent, awaiting response... 200 OK\n", + "Length: 46251583 (44M) [application/zip]\n", + "Saving to: ‘kvasir.zip’\n", + "\n", + "kvasir.zip 100%[===================>] 44,11M 514KB/s in 98s \n", + "\n", + "2021-04-29 17:44:11 (460 KB/s) - ‘kvasir.zip’ saved [46251583/46251583]\n", + "\n", + "Archive: kvasir.zip\n", + " creating: ./data/segmented-images/\n", + " inflating: ./data/segmented-images/bounding-boxes.json \n", + " creating: ./data/segmented-images/masks/\n", + " inflating: ./data/segmented-images/masks/cju88oh0po9gq0801nge4tgr1.jpg \n", + " inflating: ./data/segmented-images/masks/cju34uhepd3dd0799hs8782ad.jpg \n", + " inflating: ./data/segmented-images/masks/cju8bgdmqrksy0801tozdmraa.jpg \n", + " inflating: ./data/segmented-images/masks/cju5ddda9bkkt0850enzwatb1.jpg \n", + " inflating: ./data/segmented-images/masks/cju0s690hkp960855tjuaqvv0.jpg \n", + " inflating: ./data/segmented-images/masks/cju5nyu31gv8e0871zpk74a2n.jpg \n", + " inflating: ./data/segmented-images/masks/cju358pwtdby20878cg7nm0np.jpg \n", + " inflating: ./data/segmented-images/masks/cju5cu8qkb84x08186jwo8yin.jpg \n", + " inflating: ./data/segmented-images/masks/cju77j66ez52p08019xygi0co.jpg \n", + " inflating: ./data/segmented-images/masks/cju42xpi8lw4w0871ve317a1p.jpg \n", + " inflating: ./data/segmented-images/masks/cju884985nlmx0817vzpax3y4.jpg \n", + " inflating: ./data/segmented-images/masks/cju5fu081d8gc0818l3yylujk.jpg \n", + " inflating: ./data/segmented-images/masks/cju8a2itsq4dv0755ntlovpxe.jpg \n", + " inflating: ./data/segmented-images/masks/cju5bj926aiec07559rshy4wa.jpg \n", + " inflating: ./data/segmented-images/masks/cju1hirfi7ekp0855q0vgm9qq.jpg \n", + " inflating: ./data/segmented-images/masks/cju85bf1algsq0871y9gtlq97.jpg \n", + " inflating: ./data/segmented-images/masks/cju7fd6yt2p740987wkr8exo1.jpg \n", + " inflating: ./data/segmented-images/masks/cju1cqc7n4gpy0855jt246k68.jpg \n", + " inflating: ./data/segmented-images/masks/cju41z76wlgbz0801qdetlvby.jpg \n", + " inflating: ./data/segmented-images/masks/cjyzuio1qgh040763k56deohv.jpg \n", + " inflating: ./data/segmented-images/masks/cju43jcqim2cp08172dvjvyui.jpg \n", + " inflating: ./data/segmented-images/masks/cju8b7aqtr4a00987coba14b7.jpg \n", + " inflating: ./data/segmented-images/masks/cju2zkpdl9h7t0799ix60teqg.jpg \n", + " inflating: ./data/segmented-images/masks/cju5g163vd6mt0817uccuga6u.jpg \n", + " inflating: ./data/segmented-images/masks/cju2z2x3nvd3c099350zgty7w.jpg \n", + " inflating: ./data/segmented-images/masks/cju30u1hbakn808019g15nb8b.jpg \n", + " inflating: ./data/segmented-images/masks/cju2pag1f0s4r0878h52uq83s.jpg \n", + " inflating: ./data/segmented-images/masks/cju5uget8krjy0818kvywd0zu.jpg \n", + " inflating: ./data/segmented-images/masks/cju5yimthmlv80850zhoc90c2.jpg \n", + " inflating: ./data/segmented-images/masks/cju2txjfzv60w098839dcimys.jpg \n", + " inflating: ./data/segmented-images/masks/cju0sxqiclckk08551ycbwhno.jpg \n", + " inflating: ./data/segmented-images/masks/cju1brhsj3rls0855a1vgdlen.jpg \n", + " inflating: ./data/segmented-images/masks/cju8bff9nrfi10850fmfzbf8v.jpg \n", + " inflating: ./data/segmented-images/masks/cju8at3s1qqqx0850hcq8nmnq.jpg \n", + " inflating: ./data/segmented-images/masks/cju1cvkfwqrec0993wbp1jlzm.jpg \n", + " inflating: ./data/segmented-images/masks/cju2zy1e49pqk0878t6ncqn12.jpg \n", + " inflating: ./data/segmented-images/masks/cju3v72v5h1qz0818fggilwtq.jpg \n", + " inflating: ./data/segmented-images/masks/cju8b5p40r2c60987ofa0mu03.jpg \n", + " inflating: ./data/segmented-images/masks/cju1hhj6mxfp90835n3wofrap.jpg \n", + " inflating: ./data/segmented-images/masks/cju6z600qwh4z081700qimgl9.jpg \n", + " inflating: ./data/segmented-images/masks/cju3ul8dogf1z09872y2ecowp.jpg \n", + " inflating: ./data/segmented-images/masks/cju1cu1u2474n0878tt7v4tdr.jpg \n", + " inflating: ./data/segmented-images/masks/cju2t3ibkuecr0835o7si16zv.jpg \n", + " inflating: ./data/segmented-images/masks/cju7b4mtw1n9n080186209f3d.jpg \n", + " inflating: ./data/segmented-images/masks/cju8ayeq7r1fb0818z1junacy.jpg \n", + " inflating: ./data/segmented-images/masks/cju5vcmrqla7i0817x4sp4pqw.jpg \n", + " inflating: ./data/segmented-images/masks/cju83ipu3jwpx0801z5pvguf8.jpg \n", + " inflating: ./data/segmented-images/masks/cju7cp6dw244p0818gncdol4m.jpg \n", + " inflating: ./data/segmented-images/masks/cju2m56cryvqd0801gtn2yp8t.jpg \n", + " inflating: ./data/segmented-images/masks/cju3381d8bz3h07991xtl7ra0.jpg \n", + " inflating: ./data/segmented-images/masks/cju5wqonpm0e60801z88ewmy1.jpg \n", + " inflating: ./data/segmented-images/masks/cju2zr3c3vwb00993jn06bbaz.jpg \n", + " inflating: ./data/segmented-images/masks/cju2zxja9w1eh09933609ho9z.jpg \n", + " inflating: ./data/segmented-images/masks/cju5v8pgplg6k0755rvi2t63h.jpg \n", + " inflating: ./data/segmented-images/masks/cju34fojcctcf0799ebolbvkn.jpg \n", + " inflating: ./data/segmented-images/masks/cju8c5zcbsdfz0801o5t6jag1.jpg \n", + " inflating: ./data/segmented-images/masks/cju5y84q3mdv50817eyp82xf3.jpg \n", + " inflating: ./data/segmented-images/masks/cju88t4fvokxf07558ymyh281.jpg \n", + " inflating: ./data/segmented-images/masks/cju1c8ffau5770835g0g343o8.jpg \n", + " inflating: ./data/segmented-images/masks/cju2hfqnmhisa0993gpleeldd.jpg \n", + " inflating: ./data/segmented-images/masks/cju83h9ysjwe808716nt35oah.jpg \n", + " inflating: ./data/segmented-images/masks/cju1efbr0rqxz09931z0lf4vf.jpg \n", + " inflating: ./data/segmented-images/masks/cju2upu4evw7g08358guwozxv.jpg \n", + " inflating: ./data/segmented-images/masks/cju5wmvsdlx1j0871npgj8j4b.jpg \n", + " inflating: ./data/segmented-images/masks/cju8b3ka8r64u0801fh18hk7l.jpg \n", + " inflating: ./data/segmented-images/masks/cju8a3nhbpwnb0850d37fo2na.jpg \n", + " inflating: ./data/segmented-images/masks/cju2ntxtdzlvu0799xl3j9pan.jpg \n", + " inflating: ./data/segmented-images/masks/cju5vzjoslpj708186z2fusmz.jpg \n", + " inflating: ./data/segmented-images/masks/cju1fuoa4wmc50835qfd11sp9.jpg \n", + " inflating: ./data/segmented-images/masks/cju5vwbr4lhqn0987a1pji0ux.jpg \n", + " inflating: ./data/segmented-images/masks/cju2lyynuymli0855g7fxgbhe.jpg \n", + " inflating: ./data/segmented-images/masks/cju41kd7yl4nm0850gil5qqwh.jpg \n", + " inflating: ./data/segmented-images/masks/cju33jon3ygbj0993pu22a4k6.jpg \n", + " inflating: ./data/segmented-images/masks/cju1gi7jlwyld0835cdf6g6qz.jpg \n", + " inflating: ./data/segmented-images/masks/cju7dqcwi2dz00850gcmr2ert.jpg \n", + " inflating: ./data/segmented-images/masks/cju83kxitjv340987z09m0ezy.jpg \n", + " inflating: ./data/segmented-images/masks/cju2mh8t6p07008350e01tx2a.jpg \n", + " inflating: ./data/segmented-images/masks/cju40taxlkrho0987smigg0x0.jpg \n", + " inflating: ./data/segmented-images/masks/cju31rb7vb6110801p9rhacuw.jpg \n", + " inflating: ./data/segmented-images/masks/cju34zivp3fq80988opxbaqyn.jpg \n", + " inflating: ./data/segmented-images/masks/cju87mrypnb1e0818scv1mxxg.jpg \n", + " inflating: ./data/segmented-images/masks/cju1fjsb4sipq09931lvd8e41.jpg \n", + " inflating: ./data/segmented-images/masks/cju2zrojo9kcd0878ld2epejq.jpg \n", + " inflating: ./data/segmented-images/masks/cju1cbokpuiw70988j4lq1fpi.jpg \n", + " inflating: ./data/segmented-images/masks/cju7b5afm1nfw0801xqm8bf8q.jpg \n", + " inflating: ./data/segmented-images/masks/cju5waeduln160817w0agirve.jpg \n", + " inflating: ./data/segmented-images/masks/cju7do8c72dbo0801vxfzxdc4.jpg \n", + " inflating: ./data/segmented-images/masks/cju15l5ubz9yh0855b3ivdpse.jpg \n", + " inflating: ./data/segmented-images/masks/cju1cdxvz48hw0801i0fjwcnk.jpg \n", + " inflating: ./data/segmented-images/masks/cju2yw4s7z7p20988lmf2gdgd.jpg \n", + " inflating: ./data/segmented-images/masks/cju3v664kh0px0818y4y7wolf.jpg \n", + " inflating: ./data/segmented-images/masks/cju5xneamme2p0801qdf7fdwv.jpg \n", + " inflating: ./data/segmented-images/masks/cju2lcyfgkf5809932fn9gucn.jpg \n", + " inflating: ./data/segmented-images/masks/cju85a8h8llwm07559wxg4t5w.jpg \n", + " inflating: ./data/segmented-images/masks/cju426tomlhll0818fc0i7nvh.jpg \n", + " inflating: ./data/segmented-images/masks/cju0vtox5ain6099360pu62rp.jpg \n", + " inflating: ./data/segmented-images/masks/cju1b75x63ddl0799sdp0i2j3.jpg \n", + " inflating: ./data/segmented-images/masks/cju2phaksnahz0993yxogjcpv.jpg \n", + " inflating: ./data/segmented-images/masks/cju33yemn2qb20988wfjxximx.jpg \n", + " inflating: ./data/segmented-images/masks/cju3128yi0rpu0988o4oo5n8n.jpg \n", + " inflating: ./data/segmented-images/masks/cju7afqon1ip40850ue2308b6.jpg \n", + " inflating: ./data/segmented-images/masks/cju5hl8nee8a40755fm8qjj0o.jpg \n", + " inflating: ./data/segmented-images/masks/cju8abobpqbir08189u01huru.jpg \n", + " inflating: ./data/segmented-images/masks/cju7awzmu1ncs0871hziy65zx.jpg \n", + " inflating: ./data/segmented-images/masks/cju7ejm2l2ncl0801wq6y84nw.jpg \n", + " inflating: ./data/segmented-images/masks/cju8bi8q7rlmn0871abc5ch8k.jpg \n", + " inflating: ./data/segmented-images/masks/cju5xq3tdm9fn0987pbedxdg5.jpg \n", + " inflating: ./data/segmented-images/masks/cju6vvxsev9y30987kespucdg.jpg \n", + " inflating: ./data/segmented-images/masks/cju5fi0yxd3ei0801v7u0yudn.jpg \n", + " inflating: ./data/segmented-images/masks/cju7d7aut2a2p0818z4uxc6cd.jpg \n", + " inflating: ./data/segmented-images/masks/cju8auylgqx0z0871u4o4db7o.jpg \n", + " inflating: ./data/segmented-images/masks/cju1dg44i4z3w0801nyz4p6zf.jpg \n", + " inflating: ./data/segmented-images/masks/cju1dia8hvc6v098827mgffnm.jpg \n", + " inflating: ./data/segmented-images/masks/cju2trtjf4qjd0878a2zle9v9.jpg \n", + " inflating: ./data/segmented-images/masks/cju2t2ivz43i10878zeg8r1br.jpg \n", + " inflating: ./data/segmented-images/masks/cju31t8xd17bk0835rnb893jk.jpg \n", + " inflating: ./data/segmented-images/masks/cju8axq24r4an0755yhv9d4ly.jpg \n", + " inflating: ./data/segmented-images/masks/cju5gucasds9d0801019axylx.jpg \n", + " inflating: ./data/segmented-images/masks/cju6us80mv1b50871ebyq2wxa.jpg \n", + " inflating: ./data/segmented-images/masks/cju2s2527pfyr0993l3h1149a.jpg \n", + " inflating: ./data/segmented-images/masks/cju5klveuff6w0871wbibgh3m.jpg \n", + " inflating: ./data/segmented-images/masks/cju88fpm4o0tl0871w1i6a4ds.jpg \n", + " inflating: ./data/segmented-images/masks/cju1fm3id6gl50801r3fok20c.jpg \n", + " inflating: ./data/segmented-images/masks/cju16d65tzw9d0799ouslsw25.jpg \n", + " inflating: ./data/segmented-images/masks/cju6w733bveoz0817e600tw72.jpg \n", + " inflating: ./data/segmented-images/masks/cju2qs32r1vys07999conmbvx.jpg \n", + " inflating: ./data/segmented-images/masks/cju2zdvjn9h7r08553cp4eed5.jpg \n", + " inflating: ./data/segmented-images/masks/cju2hgsptlfam0835o3b59h1o.jpg \n", + " inflating: ./data/segmented-images/masks/cju5yhgznmkzb0801cji2vi8j.jpg \n", + " inflating: ./data/segmented-images/masks/cju7dvl5m2n4t0755hlnnjjet.jpg \n", + " inflating: ./data/segmented-images/masks/cju6ut4l8va6y0755tyw3vfqq.jpg \n", + " inflating: ./data/segmented-images/masks/cju6ur9l9v9jq0755paud9uka.jpg \n", + " inflating: ./data/segmented-images/masks/cju18ibp219ub08783i6o98g7.jpg \n", + " inflating: ./data/segmented-images/masks/cju2hewssldzx0835ep795xu0.jpg \n", + " inflating: ./data/segmented-images/masks/cju3xvoo2iqlc0817eku2r3wl.jpg \n", + " inflating: ./data/segmented-images/masks/cju2r7h21sj9608354gzks3ae.jpg \n", + " inflating: ./data/segmented-images/masks/cju5k7r0yf98c09878csbxb4d.jpg \n", + " inflating: ./data/segmented-images/masks/cju43h43am1dy08176gwfhmnt.jpg \n", + " inflating: ./data/segmented-images/masks/cju2rnkt22xep0801as160g9t.jpg \n", + " inflating: ./data/segmented-images/masks/cju32l161bi1v07990vm376in.jpg \n", + " inflating: ./data/segmented-images/masks/cju5k3j3uf6de0817hszzfr7n.jpg \n", + " inflating: ./data/segmented-images/masks/cju2hugv9vget0799hhk7ksvg.jpg \n", + " inflating: ./data/segmented-images/masks/cju7aez2x1jtj0871ztezs3oi.jpg \n", + " inflating: ./data/segmented-images/masks/cju1euuc65wm00799m4sjdnnn.jpg \n", + " inflating: ./data/segmented-images/masks/cju330ofbc2l30801th5g3hw6.jpg \n", + " inflating: ./data/segmented-images/masks/cju43lcnum9y10755bjs7z87f.jpg \n", + " inflating: ./data/segmented-images/masks/cju14g8o4xui30878gkgbrvqj.jpg \n", + " inflating: ./data/segmented-images/masks/cju2z45kuzf6d0988nz2c819m.jpg \n", + " inflating: ./data/segmented-images/masks/cju30j1rgadut0801vuyrsnt8.jpg \n", + " inflating: ./data/segmented-images/masks/cju8cgi2kspp308011nxdtjp6.jpg \n", + " inflating: ./data/segmented-images/masks/cju1f79yhsb5w0993txub59ol.jpg \n", + " inflating: ./data/segmented-images/masks/cju33x0f22peh0988g0ln7w5v.jpg \n", + " inflating: ./data/segmented-images/masks/cju85c2d4ln1b0755zz1z3onx.jpg \n", + " inflating: ./data/segmented-images/masks/cju2syxa93yw40799x2iuwabz.jpg \n", + " inflating: ./data/segmented-images/masks/cju3xuj20ivgp0818mij8bjrd.jpg \n", + " inflating: ./data/segmented-images/masks/cju33o12x2jm50988944mxq0v.jpg \n", + " inflating: ./data/segmented-images/masks/cju5thdbrjp1108715xdfx356.jpg \n", + " inflating: ./data/segmented-images/masks/cju17r8il13910799dr2wme2e.jpg \n", + " inflating: ./data/segmented-images/masks/cju3y0pjrj1c30755nxekxccj.jpg \n", + " inflating: ./data/segmented-images/masks/cju7f0ec32txj08184asb8w5f.jpg \n", + " inflating: ./data/segmented-images/masks/cju87zv8lni0o0850hbbecbq6.jpg \n", + " inflating: ./data/segmented-images/masks/cju88trl3ogi208716qvti51b.jpg \n", + " inflating: ./data/segmented-images/masks/cju8bj2ssrmlm0871gc2ug2rs.jpg \n", + " inflating: ./data/segmented-images/masks/cju1egh885m1l0855ci1lt37c.jpg \n", + " inflating: ./data/segmented-images/masks/cju41p90plcsx08018cnzpndc.jpg \n", + " inflating: ./data/segmented-images/masks/cju0tl3uz8blh0993wxvn7ly3.jpg \n", + " inflating: ./data/segmented-images/masks/cju2hx006vidl0799igm81vmh.jpg \n", + " inflating: ./data/segmented-images/masks/cju2spdagu1l50835da1f46fr.jpg \n", + " inflating: ./data/segmented-images/masks/cju34o6dbd2lo0855aqlcy1hs.jpg \n", + " inflating: ./data/segmented-images/masks/cju5ufn3skquf0818dhapnhba.jpg \n", + " inflating: ./data/segmented-images/masks/cju30mm25a53s0799qa5wiqe8.jpg \n", + " inflating: ./data/segmented-images/masks/cju1ddr6p4k5z08780uuuzit2.jpg \n", + " inflating: ./data/segmented-images/masks/cju2zp89k9q1g0855k1x0f1xa.jpg \n", + " inflating: ./data/segmented-images/masks/cju88nroho44508500129f1nh.jpg \n", + " inflating: ./data/segmented-images/masks/cju76o55nymqd0871h31sph9w.jpg \n", + " inflating: ./data/segmented-images/masks/cju160wshltz10993i1gmqxbe.jpg \n", + " inflating: ./data/segmented-images/masks/cju2ouil2mssu0993hvxsed6d.jpg \n", + " inflating: ./data/segmented-images/masks/cju1amqw6p8pw0993d9gc5crl.jpg \n", + " inflating: ./data/segmented-images/masks/cju2rqo702wpx0855fn7d5cxh.jpg \n", + " inflating: ./data/segmented-images/masks/cju843yjskhq30818qre4rwm2.jpg \n", + " inflating: ./data/segmented-images/masks/cju2hjrqcvi2j0801bx1i6gxg.jpg \n", + " inflating: ./data/segmented-images/masks/cju6v5ilsv8hk0850rb5sgh6o.jpg \n", + " inflating: ./data/segmented-images/masks/cju2rmd2rsw9g09888hh1efu0.jpg \n", + " inflating: ./data/segmented-images/masks/cju43kj2pm34f0850l28ahpni.jpg \n", + " inflating: ./data/segmented-images/masks/cju2rxm8rpbaf0993o3qr2oph.jpg \n", + " inflating: ./data/segmented-images/masks/cju5u4pywk81x0817vn9pe14z.jpg \n", + " inflating: ./data/segmented-images/masks/cju88vx2uoocy075531lc63n3.jpg \n", + " inflating: ./data/segmented-images/masks/cju30ov1oah920801mi8thuyg.jpg \n", + " inflating: ./data/segmented-images/masks/cju7ae7bq1f820987toc8si1d.jpg \n", + " inflating: ./data/segmented-images/masks/cju3xiic0ilzp0850lusrb42j.jpg \n", + " inflating: ./data/segmented-images/masks/cju2urqpwvxw70835rvndvtsi.jpg \n", + " inflating: ./data/segmented-images/masks/cju35740hzm0g0993zl5ic246.jpg \n", + " inflating: ./data/segmented-images/masks/cju2rn0hasxri0835nfy3buay.jpg \n", + " inflating: ./data/segmented-images/masks/cju357rxxdaz30878y2esjpjt.jpg \n", + " inflating: ./data/segmented-images/masks/cju7fnfv02tt90801djnix9m8.jpg \n", + " inflating: ./data/segmented-images/masks/cju89z6pqpqfx0817mfv8ixjc.jpg \n", + " inflating: ./data/segmented-images/masks/cju7dz5yy2i7z0801ausi7rna.jpg \n", + " inflating: ./data/segmented-images/masks/cju8ceacrsqkr0755hdz145es.jpg \n", + " inflating: ./data/segmented-images/masks/cju17hw9hr9c5098800fu4u8e.jpg \n", + " inflating: ./data/segmented-images/masks/cju8b542nr81x0871uxnkm9ih.jpg \n", + " inflating: ./data/segmented-images/masks/cju5uhrdwkmsu0817ervv91l8.jpg \n", + " inflating: ./data/segmented-images/masks/cju85dx63lic408017f0l0400.jpg \n", + " inflating: ./data/segmented-images/masks/cju1egx9pvz2n0988eoy8jp23.jpg \n", + " inflating: ./data/segmented-images/masks/cju88v2f9oi8w0871hx9auh01.jpg \n", + " inflating: ./data/segmented-images/masks/cju7ekbo32pft0871fv7kzwb9.jpg \n", + " inflating: ./data/segmented-images/masks/cju5uzmaol56l0817flxh4w9p.jpg \n", + " inflating: ./data/segmented-images/masks/cju6vvb8svhed0801jjcquh5e.jpg \n", + " inflating: ./data/segmented-images/masks/cju8brv16rx7f0818uf5n89pv.jpg \n", + " inflating: ./data/segmented-images/masks/cju2rzpsmtb0f0835jabkbao1.jpg \n", + " inflating: ./data/segmented-images/masks/cju16b6ynq8e40988m8vx0xnj.jpg \n", + " inflating: ./data/segmented-images/masks/cju8914beokbf0850isxpocrk.jpg \n", + " inflating: ./data/segmented-images/masks/cju8chxndsre008015uisl4si.jpg \n", + " inflating: ./data/segmented-images/masks/cju5bwhapakm90987c1v4z46a.jpg \n", + " inflating: ./data/segmented-images/masks/cju7da88w2eod0755wejzynvt.jpg \n", + " inflating: ./data/segmented-images/masks/cju8c6hnxsdvr0801wn0vrsa6.jpg \n", + " inflating: ./data/segmented-images/masks/cju2qu37qobl50993aw7ghcfq.jpg \n", + " inflating: ./data/segmented-images/masks/cju2wxv0hxs2f09884w48v8fi.jpg \n", + " inflating: ./data/segmented-images/masks/cju5ymyd8mmdc0801ry3by1xr.jpg \n", + " inflating: ./data/segmented-images/masks/cju5wrrs0m2af0818vmnajbtw.jpg \n", + " inflating: ./data/segmented-images/masks/cju7ff97z2ow40817u2r83my5.jpg \n", + " inflating: ./data/segmented-images/masks/cju2zpw4q9vzr0801p0lysjdl.jpg \n", + " inflating: ./data/segmented-images/masks/cju77b3wyz4160755qis4ljsb.jpg \n", + " inflating: ./data/segmented-images/masks/cju85citjlnfm0755i4rk5tqj.jpg \n", + " inflating: ./data/segmented-images/masks/cju887ftknop008177nnjt46y.jpg \n", + " inflating: ./data/segmented-images/masks/cju77re6fz5bb0817vp9redjg.jpg \n", + " inflating: ./data/segmented-images/masks/cju33qpdvc9g0087825jhf3s9.jpg \n", + " inflating: ./data/segmented-images/masks/cju8b0jr0r2oi0801jiquetd5.jpg \n", + " inflating: ./data/segmented-images/masks/cju32fhnhbds40799broyoptc.jpg \n", + " inflating: ./data/segmented-images/masks/cju6xmqd9w0250817l5kxfnsk.jpg \n", + " inflating: ./data/segmented-images/masks/cju2wve9v7esz0878mxsdcy04.jpg \n", + " inflating: ./data/segmented-images/masks/cju3xtufwiv9c0818djsc4cqd.jpg \n", + " inflating: ./data/segmented-images/masks/cju8dk7eztzup08182yxko5zh.jpg \n", + " inflating: ./data/segmented-images/masks/cju5hi52odyf90817prvcwg45.jpg \n", + " inflating: ./data/segmented-images/masks/cju7ezs7g2mxm098787atbran.jpg \n", + " inflating: ./data/segmented-images/masks/cju1b0y2e396p08558ois175d.jpg \n", + " inflating: ./data/segmented-images/masks/cju8aeei7q8k308173n9y4klv.jpg \n", + " inflating: ./data/segmented-images/masks/cju1h89h6xbnx08352k2790o9.jpg \n", + " inflating: ./data/segmented-images/masks/cju2xwm1s84l50799i60mq0pu.jpg \n", + " inflating: ./data/segmented-images/masks/cju88cddensj00987788yotmg.jpg \n", + " inflating: ./data/segmented-images/masks/cju3v3ac9gyz30755hfqwyp1i.jpg \n", + " inflating: ./data/segmented-images/masks/cjyzkpsbjdsjq07211dfi4sru.jpg \n", + " inflating: ./data/segmented-images/masks/cju30lncba3ny0878jwnous8n.jpg \n", + " inflating: ./data/segmented-images/masks/cju2lz8vqktne0993fuym6drw.jpg \n", + " inflating: ./data/segmented-images/masks/cju1fmsyf6gxb0801cimx2gle.jpg \n", + " inflating: ./data/segmented-images/masks/cju14pxbaoksp0835qzorx6g6.jpg \n", + " inflating: ./data/segmented-images/masks/cju1djtprvd7b0988thwwrg09.jpg \n", + " inflating: ./data/segmented-images/masks/cju7b2l561oas0871decgslaf.jpg \n", + " inflating: ./data/segmented-images/masks/cju1c4fcu40hl07992b8gj0c8.jpg \n", + " inflating: ./data/segmented-images/masks/cju7eueum2oqn0850rodmx8zo.jpg \n", + " inflating: ./data/segmented-images/masks/cju42py9mlqyd0818u3d1d7ga.jpg \n", + " inflating: ./data/segmented-images/masks/cju84hibuktj80871u519o71q.jpg \n", + " inflating: ./data/segmented-images/masks/cju42tauqlo5p08171l3cuo4b.jpg \n", + " inflating: ./data/segmented-images/masks/cju8bzzy2s66m08016z6mouqt.jpg \n", + " inflating: ./data/segmented-images/masks/cju41lojblbs307555jdci937.jpg \n", + " inflating: ./data/segmented-images/masks/cju85ia1slh220987y7c20sm2.jpg \n", + " inflating: ./data/segmented-images/masks/cju83vvmik9wa08710yeh7cuk.jpg \n", + " inflating: ./data/segmented-images/masks/cju1hp9i2xu8e0988u2dazk7m.jpg \n", + " inflating: ./data/segmented-images/masks/cju5b9oyda4yr0850g9viziyv.jpg \n", + " inflating: ./data/segmented-images/masks/cju2p4ddkmzxj0993p94o62av.jpg \n", + " inflating: ./data/segmented-images/masks/cju8adb60qbiu080188mxpf8d.jpg \n", + " inflating: ./data/segmented-images/masks/cju5es375cnzy0801nkq35ffs.jpg \n", + " inflating: ./data/segmented-images/masks/cju5wphwwlu3m0987hh3ltg88.jpg \n", + " inflating: ./data/segmented-images/masks/cju353d1eda8c07992afde611.jpg \n", + " inflating: ./data/segmented-images/masks/cju34ymm8d6700799uop0cw33.jpg \n", + " inflating: ./data/segmented-images/masks/cju785htizjzo08017tvlhtg4.jpg \n", + " inflating: ./data/segmented-images/masks/cju2qtee81yd708787bsjr75d.jpg \n", + " inflating: ./data/segmented-images/masks/cju34eqjpcpm508788b3lhp97.jpg \n", + " inflating: ./data/segmented-images/masks/cju6z2616wqbk07555bvnuyr1.jpg \n", + " inflating: ./data/segmented-images/masks/cju7aqkue1i2k09879uzcpt8r.jpg \n", + " inflating: ./data/segmented-images/masks/cju2r2obh2bjm08553kng0rh7.jpg \n", + " inflating: ./data/segmented-images/masks/cju17z0qongpa0993de4boim4.jpg \n", + " inflating: ./data/segmented-images/masks/cju774fmayxif0818u2g79usw.jpg \n", + " inflating: ./data/segmented-images/masks/cju1euant5l960878iqj5vvto.jpg \n", + " inflating: ./data/segmented-images/masks/cju2t16vuucaq0835xcpsivn2.jpg \n", + " inflating: ./data/segmented-images/masks/cju1g4nsb6ngy0799l4ezm8ab.jpg \n", + " inflating: ./data/segmented-images/masks/cju2wx0gh7fpz0878wwyd9ep8.jpg \n", + " inflating: ./data/segmented-images/masks/cju5cky5xb0ay0801oxet697t.jpg \n", + " inflating: ./data/segmented-images/masks/cju7ap09p1kz10850ldccjebj.jpg \n", + " inflating: ./data/segmented-images/masks/cju42qet0lsq90871e50xbnuv.jpg \n", + " inflating: ./data/segmented-images/masks/cju32pzh9bpw10855q4vaxfhe.jpg \n", + " inflating: ./data/segmented-images/masks/cju2rga4psq9n09881z519xx0.jpg \n", + " inflating: ./data/segmented-images/masks/cju76lsehyia10987u54vn8rb.jpg \n", + " inflating: ./data/segmented-images/masks/cju7ey10f2rvf0871bwbi9x82.jpg \n", + " inflating: ./data/segmented-images/masks/cju41s6nbleqy0755e2mslg0b.jpg \n", + " inflating: ./data/segmented-images/masks/cju40w3hbkwpn08015rbs3wko.jpg \n", + " inflating: ./data/segmented-images/masks/cju2m71z2ywwv080131bcrsd3.jpg \n", + " inflating: ./data/segmented-images/masks/cju7bd1qu1mx409877xjxibox.jpg \n", + " inflating: ./data/segmented-images/masks/cju7aifyo1p3n07552nxjx51f.jpg \n", + " inflating: ./data/segmented-images/masks/cju35ldepdtlm0801yv79y8vu.jpg \n", + " inflating: ./data/segmented-images/masks/cju2ij9uiic2l09933ljiv6gm.jpg \n", + " inflating: ./data/segmented-images/masks/cju7agj961l2r0818z29iq8yn.jpg \n", + " inflating: ./data/segmented-images/masks/cju6vrs1ov8cr098788h8gs6j.jpg \n", + " inflating: ./data/segmented-images/masks/cju7cue9b232j0801qdzk1ykj.jpg \n", + " inflating: ./data/segmented-images/masks/cju5vbo6jldrt0871jf6f1700.jpg \n", + " inflating: ./data/segmented-images/masks/cju3y9difj6th0801kd1rqm3w.jpg \n", + " inflating: ./data/segmented-images/masks/cju1871y11d6r0799k6cw4yze.jpg \n", + " inflating: ./data/segmented-images/masks/cju2nsmwjlzyl0993jl80chvz.jpg \n", + " inflating: ./data/segmented-images/masks/cju3521y5d5mq0878t3ezsu4p.jpg \n", + " inflating: ./data/segmented-images/masks/cju83rcnzkbsj0755x5anfrcg.jpg \n", + " inflating: ./data/segmented-images/masks/cju8bop5jrsid08716i24fqda.jpg \n", + " inflating: ./data/segmented-images/masks/cju8ando2qqdo0818ck7i1be1.jpg \n", + " inflating: ./data/segmented-images/masks/cju85nr8elly209872w9n5m0s.jpg \n", + " inflating: ./data/segmented-images/masks/cju35oyvd3y850988km12hdz1.jpg \n", + " inflating: ./data/segmented-images/masks/cju8c9akjsdjj0850s67uzlxq.jpg \n", + " inflating: ./data/segmented-images/masks/cju35eg0tdmjt085525sb4bua.jpg \n", + " inflating: ./data/segmented-images/masks/cju5bdwa3aatx0818b79i18zf.jpg \n", + " inflating: ./data/segmented-images/masks/cju5von04litr08718j8po40a.jpg \n", + " inflating: ./data/segmented-images/masks/cju32upim1z7u0988l883nqp6.jpg \n", + " inflating: ./data/segmented-images/masks/cju2zm0axztpe0988r8s9twjr.jpg \n", + " inflating: ./data/segmented-images/masks/cju7epwj82koz098713apjnzo.jpg \n", + " inflating: ./data/segmented-images/masks/cju84ffdzkrjn08183jh1fxmb.jpg \n", + " inflating: ./data/segmented-images/masks/cju2u73dj53oz0878486k8k4b.jpg \n", + " inflating: ./data/segmented-images/masks/cju2nyc5f02m40801ojqbtiea.jpg \n", + " inflating: ./data/segmented-images/masks/cju76l27oyrw907551ri2a7fl.jpg \n", + " inflating: ./data/segmented-images/masks/cju772304yw5t0818vbw8kkjf.jpg \n", + " inflating: ./data/segmented-images/masks/cju6xa0qmvzun0818xjukgncj.jpg \n", + " inflating: ./data/segmented-images/masks/cju7787c5yy3l080159mwqsnj.jpg \n", + " inflating: ./data/segmented-images/masks/cju2y26c588bo07993ksd8eoz.jpg \n", + " inflating: ./data/segmented-images/masks/cju5kre09fhka0850h7b1898j.jpg \n", + " inflating: ./data/segmented-images/masks/cju8cj10qsrau0871o2dr6ai1.jpg \n", + " inflating: ./data/segmented-images/masks/cju2pkwt3r8b90988v2ywq1px.jpg \n", + " inflating: ./data/segmented-images/masks/cju42u5bjlvi10801dc13sskp.jpg \n", + " inflating: ./data/segmented-images/masks/cju7dizi82h2i0755doucgnt3.jpg \n", + " inflating: ./data/segmented-images/masks/cju3x9lttikfb0818a0g104zn.jpg \n", + " inflating: ./data/segmented-images/masks/cju31ugmfb3dz0855xtqshki6.jpg \n", + " inflating: ./data/segmented-images/masks/cju424hy5lckr085073fva1ok.jpg \n", + " inflating: ./data/segmented-images/masks/cju7azuu31mia0801pf9ib5ed.jpg \n", + " inflating: ./data/segmented-images/masks/cju40u30gkuzc0871rq7t666d.jpg \n", + " inflating: ./data/segmented-images/masks/cju8apjewqrk00801k5d71gky.jpg \n", + " inflating: ./data/segmented-images/masks/cju1haab178i70799tk9z8y8x.jpg \n", + " inflating: ./data/segmented-images/masks/cju5wuhm1lwm40987vugqn3vv.jpg \n", + " inflating: ./data/segmented-images/masks/cju7etr3y2p4t0801cdzjj8ab.jpg \n", + " inflating: ./data/segmented-images/masks/cju1dhfok4mhe0878jlgrag0h.jpg \n", + " inflating: ./data/segmented-images/masks/cju2otvvv0l7z0855x7we8cb0.jpg \n", + " inflating: ./data/segmented-images/masks/cju5wj0faly5008187n6530af.jpg \n", + " inflating: ./data/segmented-images/masks/cju7ea4om2l910801bohqjccy.jpg \n", + " inflating: ./data/segmented-images/masks/cju1d50a94qf50855wsowacrc.jpg \n", + " inflating: ./data/segmented-images/masks/cju2oo0wh0bqy0878biujeyhe.jpg \n", + " inflating: ./data/segmented-images/masks/cju18kevfrojc0835bn90f1in.jpg \n", + " inflating: ./data/segmented-images/masks/cju5c7oijaqmq09878qwgqv8n.jpg \n", + " inflating: ./data/segmented-images/masks/cju7bduyq1rjf08719giru9ho.jpg \n", + " inflating: ./data/segmented-images/masks/cju5fw37edaae0801vkwvocn7.jpg \n", + " inflating: ./data/segmented-images/masks/cju83wwn1k55e0850kw6i2d81.jpg \n", + " inflating: ./data/segmented-images/masks/cju0ue769mxii08019zqgdbxn.jpg \n", + " inflating: ./data/segmented-images/masks/cju5hwonqedw10801vsd3w6kk.jpg \n", + " inflating: ./data/segmented-images/masks/cju2uy8ox62jo0801g88hh42z.jpg \n", + " inflating: ./data/segmented-images/masks/cju3tx1qyg0c907552fglumhc.jpg \n", + " inflating: ./data/segmented-images/masks/cju3v11mrgwwb0755u242ygye.jpg \n", + " inflating: ./data/segmented-images/masks/cju2suk42469908015ngmq6f2.jpg \n", + " inflating: ./data/segmented-images/masks/cju7dsrtb2f8i085064kwugfk.jpg \n", + " inflating: ./data/segmented-images/masks/cju8dm2cau2km0818jsv9eeq2.jpg \n", + " inflating: ./data/segmented-images/masks/cju784jpdzeae0987q5ypq883.jpg \n", + " inflating: ./data/segmented-images/masks/cju885ikhnmkn09878s2lqtuh.jpg \n", + " inflating: ./data/segmented-images/masks/cju45ty6zn9oz0850qy4qnck1.jpg \n", + " inflating: ./data/segmented-images/masks/ck2bxlujamu330725szlc2jdu.jpg \n", + " inflating: ./data/segmented-images/masks/cju5yclrymlgj0818k426ud6z.jpg \n", + " inflating: ./data/segmented-images/masks/cju5clr68b48r0755cmuvponm.jpg \n", + " inflating: ./data/segmented-images/masks/cju32srle1xfq083575i3fl75.jpg \n", + " inflating: ./data/segmented-images/masks/cju3ya7goj6at0818v2l5ay7f.jpg \n", + " inflating: ./data/segmented-images/masks/cju1d31sp4d4k0878r3fr02ul.jpg \n", + " inflating: ./data/segmented-images/masks/cju422cm8lfxn0818ojicxejb.jpg \n", + " inflating: ./data/segmented-images/masks/cju783tmkzkqu081803g7q5vk.jpg \n", + " inflating: ./data/segmented-images/masks/cju1h5w4wxajx0835mc954kxy.jpg \n", + " inflating: ./data/segmented-images/masks/cju30k2z40ds308353kdew70n.jpg \n", + " inflating: ./data/segmented-images/masks/cju7et17a2vjk0755e743npl1.jpg \n", + " inflating: ./data/segmented-images/masks/cju85omszllp30850b6rm9mi3.jpg \n", + " inflating: ./data/segmented-images/masks/cju32qr9tbvsj08013pkpjenq.jpg \n", + " inflating: ./data/segmented-images/masks/cju7dp3dw2k4n0755zhe003ad.jpg \n", + " inflating: ./data/segmented-images/masks/cju2ti1du4idn0878giuozonw.jpg \n", + " inflating: ./data/segmented-images/masks/ck2da7fwcjfis07218r1rvm95.jpg \n", + " inflating: ./data/segmented-images/masks/cju2rpa30t07b0835im0erql0.jpg \n", + " inflating: ./data/segmented-images/masks/cju5o1vu9gz8a0818eyy92bns.jpg \n", + " inflating: ./data/segmented-images/masks/cju2xbk0080y80801eghyddi2.jpg \n", + " inflating: ./data/segmented-images/masks/cju3xl264ingx0850rcf0rshj.jpg \n", + " inflating: ./data/segmented-images/masks/cju2z9vlp9j0w0801oag91sy9.jpg \n", + " inflating: ./data/segmented-images/masks/cju77bvg0yv4r0987yh60xmjo.jpg \n", + " inflating: ./data/segmented-images/masks/cju3xwpgviwlx0871rwm15q7v.jpg \n", + " inflating: ./data/segmented-images/masks/cju43o6n7m9nk087191ijwqq9.jpg \n", + " inflating: ./data/segmented-images/masks/cju6v1m1xv07w09870ah3njy1.jpg \n", + " inflating: ./data/segmented-images/masks/cju15ptjtppz40988odsm9azx.jpg \n", + " inflating: ./data/segmented-images/masks/cju6yxyt0wh080871sqpepu47.jpg \n", + " inflating: ./data/segmented-images/masks/cju45lbgznahl08180xz1h7u6.jpg \n", + " inflating: ./data/segmented-images/masks/cju5xopi0md7q0871r1sjc1av.jpg \n", + " inflating: ./data/segmented-images/masks/cju5wtdu4m0im0871mix0yvc0.jpg \n", + " inflating: ./data/segmented-images/masks/cju5f26ebcuai0818xlwh6116.jpg \n", + " inflating: ./data/segmented-images/masks/cju6yywx1whbb0871ksgfgf9f.jpg \n", + " inflating: ./data/segmented-images/masks/cju6vucxvvlda0755j7msqnya.jpg \n", + " inflating: ./data/segmented-images/masks/cju30gxjq0djk0988jytm49rs.jpg \n", + " inflating: ./data/segmented-images/masks/cju303j5r062k098835zxfds5.jpg \n", + " inflating: ./data/segmented-images/masks/cju35c4wzdhow0799h6eq4sgs.jpg \n", + " inflating: ./data/segmented-images/masks/cju5bbtwsa8cl0987wgfsqpao.jpg \n", + " inflating: ./data/segmented-images/masks/cju35atpxdjot0855q46aqrd0.jpg \n", + " inflating: ./data/segmented-images/masks/cju5buy2bal250818ipl6fqwv.jpg \n", + " inflating: ./data/segmented-images/masks/cju3uwz6ogsp10801h2r3bj5l.jpg \n", + " inflating: ./data/segmented-images/masks/cju5wrapcm2290818jsh26ppb.jpg \n", + " inflating: ./data/segmented-images/masks/cju45qbf3n9sa0987oonbkly9.jpg \n", + " inflating: ./data/segmented-images/masks/cju43c92lm5cj0755lorsorfg.jpg \n", + " inflating: ./data/segmented-images/masks/cju2zi4l09f5807991s8do2b2.jpg \n", + " inflating: ./data/segmented-images/masks/cju77vvcwzcm50850lzoykuva.jpg \n", + " inflating: ./data/segmented-images/masks/cju77u1sjz77b0817ft44r3fk.jpg \n", + " inflating: ./data/segmented-images/masks/cju88gx09o2vk0818610zody3.jpg \n", + " inflating: ./data/segmented-images/masks/cju2i03ptvkiu0799xbbd4det.jpg \n", + " inflating: ./data/segmented-images/masks/cju7alcgr1lsr0871riqk84z7.jpg \n", + " inflating: ./data/segmented-images/masks/cju7bb3ss1uoo0755pmhyco7t.jpg \n", + " inflating: ./data/segmented-images/masks/cju5wi6bqlxy90755bu227nvb.jpg \n", + " inflating: ./data/segmented-images/masks/cju7aklv31h4309871m29l4e7.jpg \n", + " inflating: ./data/segmented-images/masks/cju5bf6hxa6m50817rbwettgu.jpg \n", + " inflating: ./data/segmented-images/masks/cju1bhnfitmge0835ynls0l6b.jpg \n", + " inflating: ./data/segmented-images/masks/cju8djdqztu6408506pzhlo18.jpg \n", + " inflating: ./data/segmented-images/masks/cju18849rrsgr0988p90hkygb.jpg \n", + " inflating: ./data/segmented-images/masks/cju8clorgsuwn08714toqb7v6.jpg \n", + " inflating: ./data/segmented-images/masks/cju3ttznuftyf09875t11850w.jpg \n", + " inflating: ./data/segmented-images/masks/cju35mdz73x890835eynq1h9v.jpg \n", + " inflating: ./data/segmented-images/masks/cju5bmhdcafs909878qfzrqzi.jpg \n", + " inflating: ./data/segmented-images/masks/cju8bssulrrcy0987h1vq5060.jpg \n", + " inflating: ./data/segmented-images/masks/cju5huurrecm70801y680y13m.jpg \n", + " inflating: ./data/segmented-images/masks/cju87ox0kncom0801b98hqnd2.jpg \n", + " inflating: ./data/segmented-images/masks/cju30525w04r10835ygp257sb.jpg \n", + " inflating: ./data/segmented-images/masks/cju8chdlqsu620755azjty1tj.jpg \n", + " inflating: ./data/segmented-images/masks/cju30bmab08bi0835mvlr6e0r.jpg \n", + " inflating: ./data/segmented-images/masks/cju5uxjnol2r509871qv2yeia.jpg \n", + " inflating: ./data/segmented-images/masks/cju2saez63gxl08559ucjq3kt.jpg \n", + " inflating: ./data/segmented-images/masks/cju5w7xn0lrkq0801f9k0htgx.jpg \n", + " inflating: ./data/segmented-images/masks/cju7frtqu2xa20818wq8r9fzf.jpg \n", + " inflating: ./data/segmented-images/masks/cju83nwu1jxte0987h1krpfmv.jpg \n", + " inflating: ./data/segmented-images/masks/cju7fbndk2sl608015ravktum.jpg \n", + " inflating: ./data/segmented-images/masks/cju3x7xsaijq80818f0psavav.jpg \n", + " inflating: ./data/segmented-images/masks/cju6x97w4vwua0850x0997r0a.jpg \n", + " inflating: ./data/segmented-images/masks/cju8bpctzrqkr0850zeldv9kt.jpg \n", + " inflating: ./data/segmented-images/masks/cju2yv4imv6cz099314jveiib.jpg \n", + " inflating: ./data/segmented-images/masks/cju42m60jlpcm08186kqppzqv.jpg \n", + " inflating: ./data/segmented-images/masks/cju5jz5fff8c50871hbe6108f.jpg \n", + " inflating: ./data/segmented-images/masks/cju2y8s56ymqr083541ggdsml.jpg \n", + " inflating: ./data/segmented-images/masks/cju773hsyyosz0817pk1e7sjq.jpg \n", + " inflating: ./data/segmented-images/masks/cju35k2fr3vc50988c85qkrwg.jpg \n", + " inflating: ./data/segmented-images/masks/cju40sdwukv3k0755y99ug1k8.jpg \n", + " inflating: ./data/segmented-images/masks/cju2mfjndoz700988b9lc3zeq.jpg \n", + " inflating: ./data/segmented-images/masks/cju7deifq2fzn0755lc8idyh8.jpg \n", + " inflating: ./data/segmented-images/masks/cju7d7tly27h408016fyp5nr7.jpg \n", + " inflating: ./data/segmented-images/masks/cju87kbcen2av0987usezo8kn.jpg \n", + " inflating: ./data/segmented-images/masks/cju1aqqv02qwz0878a5cyhr67.jpg \n", + " inflating: ./data/segmented-images/masks/cju2pjb9v0ywn0878j5g5n69j.jpg \n", + " inflating: ./data/segmented-images/masks/cju1gkndf6yi10801o1qnje19.jpg \n", + " inflating: ./data/segmented-images/masks/cju7apr0c1qqm0755s7msqot4.jpg \n", + " inflating: ./data/segmented-images/masks/cju2qz06823a40878ojcz9ccx.jpg \n", + " inflating: ./data/segmented-images/masks/cju7bc95p1mdm0817yqj5jc6j.jpg \n", + " inflating: ./data/segmented-images/masks/cju1819curo000988pd5xcqme.jpg \n", + " inflating: ./data/segmented-images/masks/cju1fyb1d69et0878muzdak9u.jpg \n", + " inflating: ./data/segmented-images/masks/cju42romflni20817etb9a0fl.jpg \n", + " inflating: ./data/segmented-images/masks/cju13hp5rnbjx0835bf0jowgx.jpg \n", + " inflating: ./data/segmented-images/masks/cju849c23kgnk0817cgv2hw1e.jpg \n", + " inflating: ./data/segmented-images/masks/cju5x15djm7ae0755h8czf6nt.jpg \n", + " inflating: ./data/segmented-images/masks/cju0rx1idathl0835detmsp84.jpg \n", + " inflating: ./data/segmented-images/masks/cju8bqxxurs6i0850mu7mtef9.jpg \n", + " inflating: ./data/segmented-images/masks/cju2tqfgw4oat0799rn0g5b2z.jpg \n", + " inflating: ./data/segmented-images/masks/cju1dnz61vfp40988e78bkjga.jpg \n", + " inflating: ./data/segmented-images/masks/cju33za6l2qy70988jhrlp2ev.jpg \n", + " inflating: ./data/segmented-images/masks/cju42nm68lpyo0818xvvqmupq.jpg \n", + " inflating: ./data/segmented-images/masks/cju8bm24yrrdp081829mbo8ic.jpg \n", + " inflating: ./data/segmented-images/masks/cju6vqarjv7yo0987q4b1btk1.jpg \n", + " inflating: ./data/segmented-images/masks/cju7amjna1ly40871ugiokehb.jpg \n", + " inflating: ./data/segmented-images/masks/cju2pmhtr17a00855cvpelzb0.jpg \n", + " inflating: ./data/segmented-images/masks/cju2zblxw9848087853csbrx1.jpg \n", + " inflating: ./data/segmented-images/masks/cju7fcgbe2z3p07550vaflqdb.jpg \n", + " inflating: ./data/segmented-images/masks/cju7fob3x301u0755x985pmmq.jpg \n", + " inflating: ./data/segmented-images/masks/cju84ih17kp5l09876bkooocl.jpg \n", + " inflating: ./data/segmented-images/masks/cju1hs0za7jha0855vj0mdrjt.jpg \n", + " inflating: ./data/segmented-images/masks/cju5x7iskmad90818frchyfwd.jpg \n", + " inflating: ./data/segmented-images/masks/cju5ccpvqash50850kb4bs22k.jpg \n", + " inflating: ./data/segmented-images/masks/cju5fb86jd1jp0755b1ukbhq5.jpg \n", + " inflating: ./data/segmented-images/masks/cju846ec0kj7z08012o10klrb.jpg \n", + " inflating: ./data/segmented-images/masks/cju33w4sdcivk0855x879zht7.jpg \n", + " inflating: ./data/segmented-images/masks/cju87nkyrnb970801q84m47yt.jpg \n", + " inflating: ./data/segmented-images/masks/cju8cbsyssiqj0871gr4jedjp.jpg \n", + " inflating: ./data/segmented-images/masks/cju77k828z46w0871r0avuoo9.jpg \n", + " inflating: ./data/segmented-images/masks/cju16jgnyzp970878melv7r25.jpg \n", + " inflating: ./data/segmented-images/masks/cju2yyhsp933j0855hp32e012.jpg \n", + " inflating: ./data/segmented-images/masks/cju87z6o6nh73085045bzsx6o.jpg \n", + " inflating: ./data/segmented-images/masks/cju5cjh3xattc0817j2vbulzi.jpg \n", + " inflating: ./data/segmented-images/masks/cju43mkj9m8wb0871qiadahub.jpg \n", + " inflating: ./data/segmented-images/masks/cju8bysfgrzkl081786jwac09.jpg \n", + " inflating: ./data/segmented-images/masks/cju1ftaji6isw0855108yqcse.jpg \n", + " inflating: ./data/segmented-images/masks/cju8cdeazsm8h0801jxifmzur.jpg \n", + " inflating: ./data/segmented-images/masks/cju33belnbyhm0878yxl42233.jpg \n", + " inflating: ./data/segmented-images/masks/cju888fr7nveu0818r9uwtiit.jpg \n", + " inflating: ./data/segmented-images/masks/cju890guyoiti098753yg6cdu.jpg \n", + " inflating: ./data/segmented-images/masks/cju34sh43d8zm08019xbwhc0o.jpg \n", + " inflating: ./data/segmented-images/masks/cju3x2s11ibzi0817kk284k0j.jpg \n", + " inflating: ./data/segmented-images/masks/cju2wzu8wxtgu09880ku9x1pg.jpg \n", + " inflating: ./data/segmented-images/masks/cju77q10sz9ug0801449wu1nu.jpg \n", + " inflating: ./data/segmented-images/masks/cju3280wv1ir009882jze27tc.jpg \n", + " inflating: ./data/segmented-images/masks/cju45jpvfn6c809873pv1i34s.jpg \n", + " inflating: ./data/segmented-images/masks/cju30df2j09dd08351ayx2t6w.jpg \n", + " inflating: ./data/segmented-images/masks/cju17bz250pgd0799u1hqkj5u.jpg \n", + " inflating: ./data/segmented-images/masks/cju2t9tdwuk700835kv0ljmtl.jpg \n", + " inflating: ./data/segmented-images/masks/ck2bxskgxxzfv08386xkqtqdy.jpg \n", + " inflating: ./data/segmented-images/masks/cju7crgxa28550755wbsgqkel.jpg \n", + " inflating: ./data/segmented-images/masks/cju2xa3i4y0160988i679zsqd.jpg \n", + " inflating: ./data/segmented-images/masks/cju45n0oxn5vu08500yfrt9jn.jpg \n", + " inflating: ./data/segmented-images/masks/cju3yb47cj1xq0817zfotbni4.jpg \n", + " inflating: ./data/segmented-images/masks/cju88z8bson4h0871nnd7fdxo.jpg \n", + " inflating: ./data/segmented-images/masks/cju5chrxxawka0871qcj171yz.jpg \n", + " inflating: ./data/segmented-images/masks/cju1dq3x1vgx109889c7wyirg.jpg \n", + " inflating: ./data/segmented-images/masks/cju40r6jrksyk0871wg98zgho.jpg \n", + " inflating: ./data/segmented-images/masks/cju7bmi1v1pnj0987pa52jjok.jpg \n", + " inflating: ./data/segmented-images/masks/cju2zwg05a0oy0801yr73ig7g.jpg \n", + " inflating: ./data/segmented-images/masks/cju3umoh1geet0817cmpef5am.jpg \n", + " inflating: ./data/segmented-images/masks/cju1alwgo30z60855fm3y23sm.jpg \n", + " inflating: ./data/segmented-images/masks/cju2tpfa5uyx408359datxqqj.jpg \n", + " inflating: ./data/segmented-images/masks/cju83syhdk6gs0801rf1rekdl.jpg \n", + " inflating: ./data/segmented-images/masks/cju2htabevq9108015qjei0x7.jpg \n", + " inflating: ./data/segmented-images/masks/cju8bafgqrf4x0818twisk3ea.jpg \n", + " inflating: ./data/segmented-images/masks/cju7b9vcs1luz0987ta60j1dy.jpg \n", + " inflating: ./data/segmented-images/masks/cju2hlm19vjjf0801o69qnber.jpg \n", + " inflating: ./data/segmented-images/masks/cju8ca4geseia0850i2ru11hw.jpg \n", + " inflating: ./data/segmented-images/masks/cju5y7buemcw80987p0r30g9f.jpg \n", + " inflating: ./data/segmented-images/masks/cjyzjzssvd8pq0838f4nolj5l.jpg \n", + " inflating: ./data/segmented-images/masks/cju2oq5570avm079959o20op1.jpg \n", + " inflating: ./data/segmented-images/masks/cju3xhpvvimda0987ygrpzni2.jpg \n", + " inflating: ./data/segmented-images/masks/cju324q101fhe08350wae9cif.jpg \n", + " inflating: ./data/segmented-images/masks/cju8cwy02t9eq08185qn12c02.jpg \n", + " inflating: ./data/segmented-images/masks/cju1ewnoh5z030855vpex9uzt.jpg \n", + " inflating: ./data/segmented-images/masks/cju32csyfblyh080170aa3x5p.jpg \n", + " inflating: ./data/segmented-images/masks/cju5vutu7ll8w0871dfp92n9p.jpg \n", + " inflating: ./data/segmented-images/masks/cju18gzrq18zw0878wbf4ftw6.jpg \n", + " inflating: ./data/segmented-images/masks/cju7cufm7298k0755j09uf3of.jpg \n", + " inflating: ./data/segmented-images/masks/cju2hqt33lmra0988fr5ijv8j.jpg \n", + " inflating: ./data/segmented-images/masks/cju3u1c8tfyqx08503iedc3mx.jpg \n", + " inflating: ./data/segmented-images/masks/cju1f8w0t65en0799m9oacq0q.jpg \n", + " inflating: ./data/segmented-images/masks/cju34m7h536wq0988xz7gx79v.jpg \n", + " inflating: ./data/segmented-images/masks/cju6v6g6kvdw007552x6mb0po.jpg \n", + " inflating: ./data/segmented-images/masks/cju418rckl3ur08012psrx1r1.jpg \n", + " inflating: ./data/segmented-images/masks/cju5ktjwofed70817eg58ef7u.jpg \n", + " inflating: ./data/segmented-images/masks/cju1c6yfz42md08550zgoz3pw.jpg \n", + " inflating: ./data/segmented-images/masks/cju2qfie4rvz508357kad9z5o.jpg \n", + " inflating: ./data/segmented-images/masks/cju5yjq1pmlgc0801z0t24bly.jpg \n", + " inflating: ./data/segmented-images/masks/cju2xjz2ju8pe0993ysv9wg17.jpg \n", + " inflating: ./data/segmented-images/masks/cju5ukkg6kv7u08011x2b6zl5.jpg \n", + " inflating: ./data/segmented-images/masks/cju5xkwzxmf0z0818gk4xabdm.jpg \n", + " inflating: ./data/segmented-images/masks/cju8azmhcr66e0755t61atz72.jpg \n", + " inflating: ./data/segmented-images/masks/cju83u9ftk3ni0987qnhlcinv.jpg \n", + " inflating: ./data/segmented-images/masks/cju3y79ofj3va0871uqfb1mzo.jpg \n", + " inflating: ./data/segmented-images/masks/cju7d4jk723eu0817bqz2n39m.jpg \n", + " inflating: ./data/segmented-images/masks/cju2i6acqvo6l0799u20fift8.jpg \n", + " inflating: ./data/segmented-images/masks/cju7cq6su27qv075574dir0r3.jpg \n", + " inflating: ./data/segmented-images/masks/cju2nfnvxzdkd0878399axlco.jpg \n", + " inflating: ./data/segmented-images/masks/cju306x7w05nb0835cunv799x.jpg \n", + " inflating: ./data/segmented-images/masks/cju5x28nzm7t907558ocq4bt7.jpg \n", + " inflating: ./data/segmented-images/masks/cju7ahtkb1jr90801jck4kbds.jpg \n", + " inflating: ./data/segmented-images/masks/cju83k8fyjsxr0817d6nxs6r4.jpg \n", + " inflating: ./data/segmented-images/masks/cju17x0j4nfc10993y31pvlgs.jpg \n", + " inflating: ./data/segmented-images/masks/cju3x5u2tiihx0818914gzxy1.jpg \n", + " inflating: ./data/segmented-images/masks/cju5f0dezct4q08183ydw11dx.jpg \n", + " inflating: ./data/segmented-images/masks/cju7b3f5h1sm40755i572jden.jpg \n", + " inflating: ./data/segmented-images/masks/cju7ddtz729960801uazp1knc.jpg \n", + " inflating: ./data/segmented-images/masks/cju87xn2snfmv0987sc3d9xnq.jpg \n", + " inflating: ./data/segmented-images/masks/cju13cgqmnhwn0988yrainhcp.jpg \n", + " inflating: ./data/segmented-images/masks/cju2hdr06v2bq0799mbm3bks1.jpg \n", + " inflating: ./data/segmented-images/masks/cju15mhjczc8z0801kit5c6di.jpg \n", + " inflating: ./data/segmented-images/masks/cju88k75inzyb0850ccv5x3vk.jpg \n", + " inflating: ./data/segmented-images/masks/cju8c5223s8j80850b4kealt4.jpg \n", + " inflating: ./data/segmented-images/masks/cju5ft6mcd5q40987rhjgbrr6.jpg \n", + " inflating: ./data/segmented-images/masks/ck2bxknhjvs1x0794iogrq49k.jpg \n", + " inflating: ./data/segmented-images/masks/cju5tgbzhjllu08174ca41eus.jpg \n", + " inflating: ./data/segmented-images/masks/cju5vxuc5loxw0818u8xgf45p.jpg \n", + " inflating: ./data/segmented-images/masks/cju1ats0y372e08011yazcsxm.jpg \n", + " inflating: ./data/segmented-images/masks/cju2ycp1u8g2r0799jslnp7cz.jpg \n", + " inflating: ./data/segmented-images/masks/cju5cetivauok0987ok3e5bre.jpg \n", + " inflating: ./data/segmented-images/masks/cju886ryxnsl50801r93jai7q.jpg \n", + " inflating: ./data/segmented-images/masks/cju2qozsk20cq0855ugrg3cri.jpg \n", + " inflating: ./data/segmented-images/masks/cju2tvrvm53ws0801a0jfjdxg.jpg \n", + " inflating: ./data/segmented-images/masks/cju7f4sc62xqj075597xpmuoy.jpg \n", + " inflating: ./data/segmented-images/masks/cju6v4szov55u0871qmqz3v8n.jpg \n", + " inflating: ./data/segmented-images/masks/cju857ad0l88m0817qx4cwxnf.jpg \n", + " inflating: ./data/segmented-images/masks/cju5ygh1zmmdi0755uod5e17i.jpg \n", + " inflating: ./data/segmented-images/masks/cju8d4jgatgpj0871q2ophhkm.jpg \n", + " inflating: ./data/segmented-images/masks/cju3u4lxmg59o0755rz42b9en.jpg \n", + " inflating: ./data/segmented-images/masks/cju33231uy4gi0993qc7b1jch.jpg \n", + " inflating: ./data/segmented-images/masks/cju858eswlepn0871pzvdrhj1.jpg \n", + " inflating: ./data/segmented-images/masks/cju8c5mxls96t0850wvkvsity.jpg \n", + " inflating: ./data/segmented-images/masks/cju2r11x7sdgx0988o8ule0wl.jpg \n", + " inflating: ./data/segmented-images/masks/cju6wjm81vgsc0987enk9n3pr.jpg \n", + " inflating: ./data/segmented-images/masks/cju2np2k9zi3v079992ypxqkn.jpg \n", + " inflating: ./data/segmented-images/masks/cju3ua8u0g9rg0801uayhdxhu.jpg \n", + " inflating: ./data/segmented-images/masks/cju2p0eveqtdc0835gpi3p93i.jpg \n", + " inflating: ./data/segmented-images/masks/cju1f15k3w4ct0835cmde6ypo.jpg \n", + " inflating: ./data/segmented-images/masks/cju1f320ewfyu0988ndz6blh5.jpg \n", + " inflating: ./data/segmented-images/masks/ck2bxpfgxu2mk0748gsh7xelu.jpg \n", + " inflating: ./data/segmented-images/masks/cju45t5ddnbio0987qtqzx762.jpg \n", + " inflating: ./data/segmented-images/masks/cju2hos57llxm08359g92p6jj.jpg \n", + " inflating: ./data/segmented-images/masks/cju334jzo261t0835yqudnfs1.jpg \n", + " inflating: ./data/segmented-images/masks/cju32jcdabepz0878d0cznmfe.jpg \n", + " inflating: ./data/segmented-images/masks/cju31y80qbawn0801twwm2l5s.jpg \n", + " inflating: ./data/segmented-images/masks/cju8a56vxpy780850r45yu4wk.jpg \n", + " inflating: ./data/segmented-images/masks/cju8amfdtqi4x09871tygrgqe.jpg \n", + " inflating: ./data/segmented-images/masks/cju1f5x1164xv08555654c24r.jpg \n", + " inflating: ./data/segmented-images/masks/cju84gpefknwm098714oq8q61.jpg \n", + " inflating: ./data/segmented-images/masks/cju1csmlc4ht10799b8ymmghg.jpg \n", + " inflating: ./data/segmented-images/masks/cju85rkbnlo1c08503uxcpax1.jpg \n", + " inflating: ./data/segmented-images/masks/cju787jnjzjuj0871p94nck9g.jpg \n", + " inflating: ./data/segmented-images/masks/cju5wcc90lu020850mjrxppv6.jpg \n", + " inflating: ./data/segmented-images/masks/cju2zo0fwzv580988qlijd2xa.jpg \n", + " inflating: ./data/segmented-images/masks/cju5hyi9yegob0755ho3do8en.jpg \n", + " inflating: ./data/segmented-images/masks/cju3xjqtpikx50817tppy6g84.jpg \n", + " inflating: ./data/segmented-images/masks/cjyzu9th0qt4r0a46pyl4zik0.jpg \n", + " inflating: ./data/segmented-images/masks/cju1erep75us208553i4ofwwe.jpg \n", + " inflating: ./data/segmented-images/masks/cju2rkjfwoxys0993x768l1j2.jpg \n", + " inflating: ./data/segmented-images/masks/cju6ywm40wdbo0987pbftsvtg.jpg \n", + " inflating: ./data/segmented-images/masks/cju45v0pungu40871acnwtmu5.jpg \n", + " inflating: ./data/segmented-images/masks/cju8bh8surexp0987o5pzklk1.jpg \n", + " inflating: ./data/segmented-images/masks/cju1gv7106qd008784gk603mg.jpg \n", + " inflating: ./data/segmented-images/masks/cju5ekty5ckzf07550c9u3ckk.jpg \n", + " inflating: ./data/segmented-images/masks/cju0roawvklrq0799vmjorwfv.jpg \n", + " inflating: ./data/segmented-images/masks/cju7f9umg2olj0987fj5y285w.jpg \n", + " inflating: ./data/segmented-images/masks/cju3518w2d838079939fqztbc.jpg \n", + " inflating: ./data/segmented-images/masks/cju16fpvhzypl0799p9phnlx6.jpg \n", + " inflating: ./data/segmented-images/masks/cju5nxkujgscq0817l9gss626.jpg \n", + " inflating: ./data/segmented-images/masks/cju2yb31a8e8u0878wdashg7o.jpg \n", + " inflating: ./data/segmented-images/masks/cju5k503sfa5f0871lx0rpu5y.jpg \n", + " inflating: ./data/segmented-images/masks/cju5y4hgqmk0i08180rjhbwvp.jpg \n", + " inflating: ./data/segmented-images/masks/cju31w6goazci0799n014ly1q.jpg \n", + " inflating: ./data/segmented-images/masks/cju2tjrog4jy30878pawyazqc.jpg \n", + " inflating: ./data/segmented-images/masks/cju5udcufki0s09874ll1dbr5.jpg \n", + " inflating: ./data/segmented-images/masks/cju2osuru0ki00855txo0n3uu.jpg \n", + " inflating: ./data/segmented-images/masks/cju6x0yqbvxqt0755dhxislgb.jpg \n", + " inflating: ./data/segmented-images/masks/cju2trbpkv0c00988hxla5dzz.jpg \n", + " inflating: ./data/segmented-images/masks/cju8567gdlcbq0801dwwyo2jt.jpg \n", + " inflating: ./data/segmented-images/masks/cju7dtb1e2j0t0818deq51ib3.jpg \n", + " inflating: ./data/segmented-images/masks/cju5f8hxdcxxn08188obby0ea.jpg \n", + " inflating: ./data/segmented-images/masks/cju8b2rmgr52s0801p54eyflx.jpg \n", + " inflating: ./data/segmented-images/masks/cju2qh5le1ock0878oahaql7d.jpg \n", + " inflating: ./data/segmented-images/masks/cju8czvnztbf40871b4m7t78w.jpg \n", + " inflating: ./data/segmented-images/masks/cju7dymur2od30755eg8yv2ht.jpg \n", + " inflating: ./data/segmented-images/masks/cju2hw5gjlr5h0988so2qqres.jpg \n", + " inflating: ./data/segmented-images/masks/cju1c3218411b08014g9f6gig.jpg \n", + " inflating: ./data/segmented-images/masks/cju2uzabhs6er0993x3aaf87p.jpg \n", + " inflating: ./data/segmented-images/masks/cju8arof2qpf20850ifr1bnqj.jpg \n", + " inflating: ./data/segmented-images/masks/cju87r56lnkyp0755hz30leew.jpg \n", + " inflating: ./data/segmented-images/masks/cju5xjn5mm78b09871spyqhhr.jpg \n", + " inflating: ./data/segmented-images/masks/cju7f6cqy2ur20818t1saazbm.jpg \n", + " inflating: ./data/segmented-images/masks/cju2y40d8ulqo0993q0adtgtb.jpg \n", + " inflating: ./data/segmented-images/masks/cju7fpfzq2wyf0818xxd1oziv.jpg \n", + " inflating: ./data/segmented-images/masks/cju302fqq9spc0878rrygyzzz.jpg \n", + " inflating: ./data/segmented-images/masks/cju8828oxnool0801qno9luhr.jpg \n", + " inflating: ./data/segmented-images/masks/cju893jmdompz0817xn3g1w4h.jpg \n", + " inflating: ./data/segmented-images/masks/cju6z9a9kwsl007552s49rx6i.jpg \n", + " inflating: ./data/segmented-images/masks/cju15wdt3zla10801odjiw7sy.jpg \n", + " inflating: ./data/segmented-images/masks/cju14hjh2ob2o0835ouz3r5aa.jpg \n", + " inflating: ./data/segmented-images/masks/cju13fwthn9mq0835gacxgy01.jpg \n" + ] + }, + { + "name": "stdout", + "output_type": "stream", + "text": [ + " inflating: ./data/segmented-images/masks/cju85qefyln6v0850szeb9byi.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju6vgdmivcvb08018fra5lnv.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju8b8yair65w09878pyqtr96.jpg \r\n", + " inflating: ./data/segmented-images/masks/ck2395w2mb4vu07480otsu6tw.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju2nguelpmlj0835rojdn097.jpg \r\n", + " inflating: ./data/segmented-images/masks/cjyzkmjy8evns070165gf9dmq.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju0s2a9ekvms080138tjjpxr.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju2rlqdnoz9k0993cpjae3x0.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju7es23b2vcp0755gpbm9s7v.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju5tenjojp1j0755ms4949h2.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju7fq7mm2pw508176uk5ugtx.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju83mki1jv5w0817kubxm31r.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju5c5xc7algd0817pb1ej5yo.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju7d3oc82cho0755dajlwldz.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju83yddek68q0850d2x7zfkm.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju5wkonqlrl409877y8zvnub.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju2i8br1vqtd08784u6vmcjk.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju336l68y7if0993wf092166.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju7d1tvt25bu08019dvw3uff.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju7bgnvb1sf808717qa799ir.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju6x4t13vyw60755gtcf9ndu.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju7er4kc2opa0801anuxc0eb.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju77idwfz36d0871tzfzz51i.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju1cfhyg48bb0799cl5pr2jh.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju88aq6vo1ij0755c2ey7z7n.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju7dn24o296i09871qfxb8s2.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju7bfx651qr80801cs7epotb.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju45pm27n80u08174kyow1gj.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju7b1ygu1msd0801hywhy0mc.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju6z7e4bwgdd0987ogkzq9kt.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju8dpa89u6l80818dj6lldh9.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju5hjxaae3i40850h5z2laf5.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju1g20bdwq6u0835e16xugcd.jpg \r\n", + " inflating: ./data/segmented-images/masks/cjyzlw7f9faqr070129au64sq.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju41r6v2lcww0871ps8k8pf5.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju7db7lp2f400755tntd1ohf.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju43b8daly4408170e5ev06g.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju42dwedlmk60871jbgu4ehi.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju41nz76lcxu0755cya2qefx.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju8aj01yqeqm0850lhdz3xdw.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju6z1bzbwfq50817b2alatvr.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju3tvffffx5f0818t5ov22al.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju5ew4h9cqaf0818rrczkmqh.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju6uy20suzbl0987rzuhz7z9.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju43in5fm22c08175rxziqrk.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju5o4pk9h0720755lgp9jq8m.jpg \r\n", + " inflating: ./data/segmented-images/masks/ck2bxqz3evvg20794iiyv5v2m.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju8buos5rz9b08715lfr0f4f.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju2ma647l0nj0993ot4deq2q.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju35a77vdj4n08556jj2lgmc.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju8bk8oirjhw0817hgkua2w8.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju6wt9jvvn500871hjn3t3g0.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju1b3zgj3d8e0801kpolea6c.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju8d2q30tfhs0801n7lx77xl.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju35fxqyzt5p0993vusm54qz.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju7atnm31if40817pqclnjer.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju3v0fl3gwce0755qkjhzmd4.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju300m3s04fg0988uzupuf7z.jpg \r\n", + " inflating: ./data/segmented-images/masks/cjz14qsk2wci60794un9ozwmw.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju2u2b9o4zvp08788qb9nqxj.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju1fj6axwfp30835ukhuzhw9.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju5jx7jzf7c90871c2i9aiov.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju34ouumcznz07996gg1xq7v.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju2t62nq45jl0799odpufwx6.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju7arvfe1ldu0850erdmphgj.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju5hqz50e7o90850e0prlpa0.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju2lejzcy4pc0878c9rlonot.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju85hguellg50818kwu3s8d1.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju5bhv81abur0850ean02atv.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju428k5fldt108177s6g6f45.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju6wll7wvo3y08502pagos8m.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju35i2e63uxr0835h7zgkg9k.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju15jr8jz8sb0855ukmkswkz.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju1cnnziug1l0835yh4ropyg.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju34repocy5208780gswillm.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju2lberzkdzm09938cl40pog.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju2xlcqxy9c60988vjacdznb.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju2sxf3iqbpv09937iksn8ep.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju6vta3kvazg0817qbeppjtm.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju7evxt12m730987rxivne3x.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju2y0z6g87p10878fpk5d3rq.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju7fazv92ywx0755xov2erga.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju45rj7ln8980850a7821fov.jpg \r\n", + " inflating: ./data/segmented-images/masks/cjyzl833ndne80838pzuq6ila.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju2qn2fzs1vy0988l243cvzy.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju30qbm1ad3x0855znuhpz9u.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju2yg5ht8i4p087800js8hp4.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju84aoa3ktwn0755pfl4gfwd.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju7dlk532dsh0871zvr6qz0r.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju8402x1kcy70801t6kz6bdi.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju5enq1tcn1i0755hnkon787.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju33eqwbcch208012jikwdky.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju7ajnbo1gvm098749rdouk0.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju5i39mreass0817au8p22zy.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju1fr4etsmrr09933u4t4aql.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju5fs6j6d8350801vglraq4u.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju6wi3akvn8r0801px8eligc.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju2omjpeqj5a0988pjdlb8l1.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju42wamblrqn098798r2yyok.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju8c3xs7sauj0801ieyzezr5.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju2raxlosl630988jdbfy9b0.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju87li0zn3yb0817kbwgjiz8.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju2z1nxlzaj40835wj81s1iy.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju42g865lorv07552ytz6xxa.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju7f900s2o0k08175gl1giid.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju1ejj7dvqfa0835ra184v5m.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju76erapykj30871x5eaxh4q.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju2i3hzclw3o0988rrgh911i.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju892fesoq2g0801n0e0jyia.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju88evxanv9r08176zkeovec.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju43gfosm63n08714rpih8pe.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju88msmoo3470817m441j4sg.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju7b10ce1mnm08011c5bwyr4.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju1bm8063nmh07996rsjjemq.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju8dic9mtppa0987swn23wbc.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju2xf8e5y2wm08359vcgk09b.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju2yi9tz8vky0801yqip0xyl.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju7fmvpk2q170987v6i3ola8.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju2tzypl4wss0799ow05oxb9.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju7dmlgf2ebw0871ieqas5fh.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju85mpuglq8k0818d2it6hzb.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju7ctvqn25dy08186g442m1r.jpg \r\n", + " inflating: ./data/segmented-images/masks/cjyzurzvohqnr0794es1itzek.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju8b1v3br45u087189kku66u.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju1cj3f0qi5n0993ut8f49rj.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju7adqyj1jcx08712r1ro5gx.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju34c1xfyz920993itxkkfad.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju2top2ruxxy0988p1svx36g.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju5vi4nxlc530817uoqm2m7a.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju7d2q1k27nf08715zshsckt.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju1gghyjwxt80835vx0wgxw0.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju6uzxk0v83p0801rcwnexdu.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju2qxxko229x08786gvxxhur.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju2xyd9vyi7m098831qcucse.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju88itqbny720987hxizbj5y.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju7ev2b12owa08500bpfpwyw.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju8b4ja9r2s808509d45ma86.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju5boicjagt20871b1fotkh4.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju85fc11ljr40818edpb0inh.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju88l66no10s0850rsda7ej1.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju2nbdpmlmcj0993s1cht0dz.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju2nqapmzvk20801f9us40dx.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju6wuojavt740818b5qcv3iw.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju77g99iyxc00817zqi2ppor.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju3wstckialg0871xs0vevsj.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju85plp7lmkw0850rx42jdpf.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju7bf1lp1shi081835vs84lc.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju1ffnjn6ctm08015perkg37.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju7d8m3b2e210755l8fj1yph.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju7dxffn2eam0817qxosfwch.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju5h57xedz5h0755mjpc8694.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju5vgawslbe30987ndeepc1b.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju3v56bwgy8v0871w14pz8fx.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju6x35ervu2808015c7eoqe4.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju2xs6na81t20878pt6nkfip.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju85l4yjlops0801fvmnwptf.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju16ach3m1da0993r1dq3sn2.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju8bbznkrf5g0871jncffynk.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju33mirdc8mj0799k33wzoes.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju2u4pymvc720988wsxrmi84.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju6v3bb2v7xo085090blubyw.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju88q6h6obpd0871ckmiabbo.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju2zdhsczmn50988z64qwg2q.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju30xqmh0ni00835ix3batv1.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju34ds2531520988qjpqt6e3.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju5eyfe9cpk90987laa7tsl3.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju2s16zp317h0799gr67jqc2.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju8bljw9rqk20801kr54akrl.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju32gzs6xo8x0993r8tedbpb.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju8aw9n1qyg10801jkjlmors.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju5x00l6m5j608503k78ptee.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju5bycdkalkb09875f7bfrvx.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju7dubap2g0w0801fgl42mg9.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju6xifswvwbo0987nibtdr50.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju1expq45zst0855rjqwwj4m.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju2yo1j1v0qz09934o0e683p.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju3yht87j83m08507yk1u1fg.jpg \r\n", + " inflating: ./data/segmented-images/masks/cjyzu3reghjya0794w7pwoi50.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju5ht88gedbu0755xrcuddcx.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju7d9seq29zd0871nzl2uu5m.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju183od81ff608017ekzif89.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju0qkwl35piu0993l0dewei2.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju3xga12iixg0817dijbvjxw.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju8alhigqn2h0801zksudldd.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju7ez7r22qbc08015xfoz2wb.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju2qdj95ru8g09886gfi9rsz.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju40jl7skiuo0817p0smlgg8.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju7el6xv2k520817qxx9wdr5.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju7dhpsc2dnn0818025m6857.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju7dda8w2br20818zhsuz8s7.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju77196iyshb0850ycbto50a.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju40m0rjkpw80871z6n6yg1u.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju7csvlb22fr0850lvm45n3x.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju7dwe282dc309876rco45ts.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju7ecl9i2i060987xawjp4l0.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju88rl5eo94l0850kf5wtrm1.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju7cl8zm1xcu0817ado0jpas.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju8b6rp0r5st0850184f79xt.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju2okvco06xc0799kxe5n1qh.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju2iatlki5u309930zmgkv6h.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju7g7ba42z310987bqzbi2bq.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju0t4oil7vzk099370nun5h9.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju87tyddnnad0755bj0wxahe.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju326h4v1gxw08352px40p7r.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju8c1a0ws7o208181c6lbsom.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju5u8gz4kj5b07552e2wpkwp.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju17otoe119u0799nqcbl8n1.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju2uokeg5jm20799xwgsyz89.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju7dglf226g50987ohbthl19.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju1fb9236a110801yvg0fwju.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju2sggy13na70855tbeoqgha.jpg \r\n", + " inflating: ./data/segmented-images/masks/ck2bxw18mmz1k0725litqq2mc.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju2sevf53lkx08558h5bpaig.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju2qvuj1s9ok0835tp2k4ozh.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju30ftgja7170855xl9bkdm0.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju5eq8c8ck690850vix98hv3.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju2ulk385h170799rlklxob0.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju43eigtm6ev0801mv0m96t1.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju175facms5f0993a5tjikvt.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju1d96gsv62d09881b3wecw2.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju3xzvnzj0hd0755xprz39nj.jpg \r\n", + " inflating: ./data/segmented-images/masks/cjyztzaqtrv430848l8xgcerw.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju89y9h0puti0818i5yw29e6.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju2ysg748ru80878sp6j0gm0.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju2rz4k434s70855wwx3ddtx.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju5eftctcdbj08712gdp989f.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju8aqq8uqmoq0987hphto9gg.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju8bn7m2rmm70817hgxpb1uq.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju5ca9hcatkc0801jzwe7tfx.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju17g6ykn1cs0993dww6qdi8.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju15czxqp3lv0835jvhgzurz.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju323ypb1fbb0988gx5rzudb.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju2r6mt2om21099352pny5gw.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju88y1mwoln50871emyfny1g.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju17v6ih0u7808783zcbg1jy.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju1c0qb4tzi308355wtsnp0y.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju5i5oh2efg60987ez6cpf72.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju0u2g7pmnux0801vkk47ivj.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju2z6ez69g4u0801qwt088lw.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju7ehljc2or70871261br8ai.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju8432cmkgq90871cxe4iptl.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju2zgbj9zmrw0835nnlzxj4c.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju7fen322ou10817ziqkob4k.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju34xspwzenf0993cyzajv9n.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju3y54kwj3nr0801biidlb4e.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju32zhbnc1oy0801iyv1ix6p.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju0u82z3cuma0835wlxrnrjv.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju83qd0yjyht0817ktkfl268.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju412uwlkva50850d1ps1ww7.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju8doa16u5gh0818w1ywda3q.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju7dbppn28nx085097654msi.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju5u6wf0kh1t0755bg1ssixv.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju3uhb79gcgr0871orbrbi3x.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju1drnhbrb9409935wi7vkhg.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju6vifjlv55z0987un6y4zdo.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju87vqa0ndwg0850onjdz7ol.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju8418jhkf7d0818ga2v0xq0.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju2srvy5440s0801y1ba9akr.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju87q6yoneim0871dl4phvkd.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju7d5m0p23kn09871rk7pu3v.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju1cyjb5qtie0993njqne9m3.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju84kplnl1y30755ropua1b0.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju8c82iosagu0817l74s4m5g.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju77t0razbvm080106o56289.jpg \r\n", + " inflating: ./data/segmented-images/masks/cjyzufihqquiw0a46jatrbwln.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju0qx73cjw570799j4n5cjze.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju6xlygpw7bs0818n691jsq4.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju171py4qiha0835u8sl59ds.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju8cattbsivm0818p446wgel.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju3u815rg4ek0850vvhtcvcm.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju30nyxe0gfb0835p256yoju.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju34aozyyy830993bn16u32n.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju7d6ux323ze0987xos3srkx.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju7efffp2ivf0817etg3jehl.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju7dff9529h208503w60lbil.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju1hmff8tkp809931jps6fbr.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju3xeexgii1j0817zs68tb4g.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju34i3qvcyog0855qiejxx5w.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju77afzlz3kp07550x5nafzs.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju3y21quj0ir0818kgjagr15.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju2s9g11pnra0993gn4eh793.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju2ro5jqsy680988pi6qsujw.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju8c2rqzs5t80850d0zky5dy.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju5fydrud94708507vo6oy21.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju2wtwj87kys0855kx6mddzw.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju439oazm2fu0871ma0vvrft.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju5knbbqfipk080128cggukq.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju8a84g0q76m0818hwiggkod.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju84dsvaklpx098750hp83x4.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju2sszfq3uye0878sucelzk2.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju2y5zas8m7f0801d34g5owq.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju7emdni2py40871ivhxjtut.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju7flevb2wii08188otgs9p2.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju6wn57mvooj0850rp78hhy7.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju7ebe962hr409872ovibahw.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju30ia8da2bq0799klnehml2.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju847pxykriq0755268ktrk2.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju5yeqiwmkgl0801fzv2douc.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju320gyvbch60801v2amdi2g.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju45ofd9ne1j0801ri8dup7t.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju2nnqrqzp580855z8mhzgd6.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju2igw4gvxds0878808qj398.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju7druhp2gp308715i6km7be.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju310f6val1v0855xo8tc3gu.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju2ricdv2iys0878sv1adh0u.jpg \r\n", + " inflating: ./data/segmented-images/masks/cjyzk8qieoboa0848ogj51wwm.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju30ajhw09sx0988qyahx9s8.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju3tp94kfstl08181awh6z49.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju16whaj0e7n0855q7b6cjkm.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju3tsh4lfsok0987w6x3a0v1.jpg \r\n", + " inflating: ./data/segmented-images/masks/ck2bxiswtxuw80838qkisqjwz.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju8bw697rwg308177tg8huas.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju32a52lb9rc0799xi40qs00.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju2qqn5ys4uo0988ewrt2ip2.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju414lf2l1lt0801rl3hjllj.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju7avvi51iox0817ym55y6tt.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju430pm2lz0y0755jkhcc3d1.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju2zjcvj9qma0801dk71hhi0.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju7eotqi2qea0871y8yc7tqh.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju410dnfl0960755y8lu8d79.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju32phw2bv130801yj7bkouq.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju3ykamdj9u208503pygyuc8.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju3u39fog1bo0871lxjrabks.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju8ashhnquqr0801rwduzt7d.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju2r91dg2k090801bh0xzbxk.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju2xd75m82720801q4s4ik3n.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju2p91qir00k08350ddfif0w.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju40wto5kxwi0755it190f2k.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju3x4blzieu30850x10uuvbm.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju1dfeupuzlw0835gnxip369.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju1hyolc7aqu0878rrkfn1lr.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju2yljr0yzhw0988ecf271ly.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju8a1jtvpt9m081712iwkca7.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju7eea9b2m0z0801ynqv1fqu.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju2nd7l7z98o0799gfjvyfmw.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju2ueb6j5ado0878vf5md13o.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju30ywtc0oar0835bp2en7ec.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju2uwz9f5yf1085506cfamfx.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju5woy82m07m08505dmjg7g1.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju3uz4o6gr9z0850lhxyxvsj.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju0sr5ghl0nd08789uzf1raf.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju7f5ghb2r5s0801chwkxxh9.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju85je7vlht70817c9jcjwi4.jpg \r\n", + " inflating: ./data/segmented-images/masks/cjyzul1qggwwj07216mhiv5sy.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju84jdl9kv0i0871eog9b3i9.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju0qoxqj9q6s0835b43399p4.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju40poe4kt9s0755f9cnm3h5.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju8dn0c3u2v50801k8rvq02f.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju2x7vw87mu30878hye2ca0m.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju8dqkrqu83i0818ev74qpxq.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju2oi8sq0i2y0801mektzvw8.jpg \r\n", + " creating: ./data/segmented-images/images/\r\n", + " inflating: ./data/segmented-images/images/cju2htabevq9108015qjei0x7.jpg \r\n", + " inflating: ./data/segmented-images/images/cju2r2obh2bjm08553kng0rh7.jpg \r\n", + " inflating: ./data/segmented-images/images/cju2zdvjn9h7r08553cp4eed5.jpg \r\n", + " inflating: ./data/segmented-images/images/cju6v4szov55u0871qmqz3v8n.jpg \r\n", + " inflating: ./data/segmented-images/images/cju16d65tzw9d0799ouslsw25.jpg \r\n", + " inflating: ./data/segmented-images/images/cju2txjfzv60w098839dcimys.jpg \r\n", + " inflating: ./data/segmented-images/images/cju8bff9nrfi10850fmfzbf8v.jpg \r\n", + " inflating: ./data/segmented-images/images/cju3xvoo2iqlc0817eku2r3wl.jpg \r\n", + " inflating: ./data/segmented-images/images/cju5buy2bal250818ipl6fqwv.jpg \r\n", + " inflating: ./data/segmented-images/images/cju7f0ec32txj08184asb8w5f.jpg \r\n", + " inflating: ./data/segmented-images/images/cju31y80qbawn0801twwm2l5s.jpg \r\n", + " inflating: ./data/segmented-images/images/cju3xiic0ilzp0850lusrb42j.jpg \r\n", + " inflating: ./data/segmented-images/images/cju6v6g6kvdw007552x6mb0po.jpg \r\n", + " inflating: ./data/segmented-images/images/cju5hi52odyf90817prvcwg45.jpg \r\n", + " inflating: ./data/segmented-images/images/cju87li0zn3yb0817kbwgjiz8.jpg \r\n", + " inflating: ./data/segmented-images/images/cju357rxxdaz30878y2esjpjt.jpg \r\n", + " inflating: ./data/segmented-images/images/cju35740hzm0g0993zl5ic246.jpg \r\n", + " inflating: ./data/segmented-images/images/cju7da88w2eod0755wejzynvt.jpg \r\n", + " inflating: ./data/segmented-images/images/cju6z2616wqbk07555bvnuyr1.jpg \r\n", + " inflating: ./data/segmented-images/images/cju35oyvd3y850988km12hdz1.jpg \r\n", + " inflating: ./data/segmented-images/images/cju892fesoq2g0801n0e0jyia.jpg \r\n", + " inflating: ./data/segmented-images/images/cju5x7iskmad90818frchyfwd.jpg \r\n", + " inflating: ./data/segmented-images/images/cju6w733bveoz0817e600tw72.jpg \r\n", + " inflating: ./data/segmented-images/images/cju41z76wlgbz0801qdetlvby.jpg \r\n", + " inflating: ./data/segmented-images/images/cju42py9mlqyd0818u3d1d7ga.jpg \r\n", + " inflating: ./data/segmented-images/images/cju8ceacrsqkr0755hdz145es.jpg \r\n", + " inflating: ./data/segmented-images/images/cju42xpi8lw4w0871ve317a1p.jpg \r\n", + " inflating: ./data/segmented-images/images/cju5wj0faly5008187n6530af.jpg \r\n", + " inflating: ./data/segmented-images/images/cju2saez63gxl08559ucjq3kt.jpg \r\n", + " inflating: ./data/segmented-images/images/cju88fpm4o0tl0871w1i6a4ds.jpg \r\n", + " inflating: ./data/segmented-images/images/cju2rnkt22xep0801as160g9t.jpg \r\n", + " inflating: ./data/segmented-images/images/cju87r56lnkyp0755hz30leew.jpg \r\n", + " inflating: ./data/segmented-images/images/cju1egh885m1l0855ci1lt37c.jpg \r\n", + " inflating: ./data/segmented-images/images/cju34eqjpcpm508788b3lhp97.jpg \r\n", + " inflating: ./data/segmented-images/images/cju2t9tdwuk700835kv0ljmtl.jpg \r\n", + " inflating: ./data/segmented-images/images/cju2qxxko229x08786gvxxhur.jpg \r\n", + " inflating: ./data/segmented-images/images/cju8apjewqrk00801k5d71gky.jpg \r\n", + " inflating: ./data/segmented-images/images/cju6ur9l9v9jq0755paud9uka.jpg \r\n", + " inflating: ./data/segmented-images/images/cju843yjskhq30818qre4rwm2.jpg \r\n", + " inflating: ./data/segmented-images/images/cju8bj2ssrmlm0871gc2ug2rs.jpg \r\n", + " inflating: ./data/segmented-images/images/cju2yyhsp933j0855hp32e012.jpg \r\n", + " inflating: ./data/segmented-images/images/cju2top2ruxxy0988p1svx36g.jpg \r\n", + " inflating: ./data/segmented-images/images/cju88trl3ogi208716qvti51b.jpg \r\n", + " inflating: ./data/segmented-images/images/cju5fw37edaae0801vkwvocn7.jpg \r\n", + " inflating: ./data/segmented-images/images/cju17x0j4nfc10993y31pvlgs.jpg \r\n", + " inflating: ./data/segmented-images/images/cju7bd1qu1mx409877xjxibox.jpg \r\n", + " inflating: ./data/segmented-images/images/cju7agj961l2r0818z29iq8yn.jpg \r\n", + " inflating: ./data/segmented-images/images/cju7cue9b232j0801qdzk1ykj.jpg \r\n", + " inflating: ./data/segmented-images/images/cju1cvkfwqrec0993wbp1jlzm.jpg \r\n", + " inflating: ./data/segmented-images/images/cju7b5afm1nfw0801xqm8bf8q.jpg \r\n", + " inflating: ./data/segmented-images/images/cju2qu37qobl50993aw7ghcfq.jpg \r\n", + " inflating: ./data/segmented-images/images/cju5wphwwlu3m0987hh3ltg88.jpg \r\n", + " inflating: ./data/segmented-images/images/cju2ti1du4idn0878giuozonw.jpg \r\n", + " inflating: ./data/segmented-images/images/cju83kxitjv340987z09m0ezy.jpg \r\n", + " inflating: ./data/segmented-images/images/cju2zp89k9q1g0855k1x0f1xa.jpg \r\n", + " inflating: ./data/segmented-images/images/cju6v5ilsv8hk0850rb5sgh6o.jpg \r\n", + " inflating: ./data/segmented-images/images/cju5xkwzxmf0z0818gk4xabdm.jpg \r\n", + " inflating: ./data/segmented-images/images/cju83u9ftk3ni0987qnhlcinv.jpg \r\n", + " inflating: ./data/segmented-images/images/cju8chxndsre008015uisl4si.jpg \r\n", + " inflating: ./data/segmented-images/images/cju1b75x63ddl0799sdp0i2j3.jpg \r\n", + " inflating: ./data/segmented-images/images/cju8arof2qpf20850ifr1bnqj.jpg \r\n", + " inflating: ./data/segmented-images/images/cju5fb86jd1jp0755b1ukbhq5.jpg \r\n", + " inflating: ./data/segmented-images/images/cju6z600qwh4z081700qimgl9.jpg \r\n", + " inflating: ./data/segmented-images/images/cju2yw4s7z7p20988lmf2gdgd.jpg \r\n", + " inflating: ./data/segmented-images/images/cju45qbf3n9sa0987oonbkly9.jpg \r\n", + " inflating: ./data/segmented-images/images/cju8c9akjsdjj0850s67uzlxq.jpg \r\n", + " inflating: ./data/segmented-images/images/cju45jpvfn6c809873pv1i34s.jpg \r\n", + " inflating: ./data/segmented-images/images/cju2lyynuymli0855g7fxgbhe.jpg \r\n", + " inflating: ./data/segmented-images/images/cju3v3ac9gyz30755hfqwyp1i.jpg \r\n", + " inflating: ./data/segmented-images/images/cju77k828z46w0871r0avuoo9.jpg \r\n", + " inflating: ./data/segmented-images/images/cju2z2x3nvd3c099350zgty7w.jpg \r\n", + " inflating: ./data/segmented-images/images/cju41kd7yl4nm0850gil5qqwh.jpg \r\n", + " inflating: ./data/segmented-images/images/cju7b2l561oas0871decgslaf.jpg \r\n", + " inflating: ./data/segmented-images/images/cju7cl8zm1xcu0817ado0jpas.jpg \r\n", + " inflating: ./data/segmented-images/images/cju45ty6zn9oz0850qy4qnck1.jpg \r\n", + " inflating: ./data/segmented-images/images/cju7ae7bq1f820987toc8si1d.jpg \r\n", + " inflating: ./data/segmented-images/images/cju76l27oyrw907551ri2a7fl.jpg \r\n", + " inflating: ./data/segmented-images/images/cju0vtox5ain6099360pu62rp.jpg \r\n", + " inflating: ./data/segmented-images/images/cju1hp9i2xu8e0988u2dazk7m.jpg \r\n", + " inflating: ./data/segmented-images/images/cju88k75inzyb0850ccv5x3vk.jpg \r\n", + " inflating: ./data/segmented-images/images/cju30bmab08bi0835mvlr6e0r.jpg \r\n", + " inflating: ./data/segmented-images/images/cju85bf1algsq0871y9gtlq97.jpg \r\n", + " inflating: ./data/segmented-images/images/cju2uokeg5jm20799xwgsyz89.jpg \r\n", + " inflating: ./data/segmented-images/images/cju1871y11d6r0799k6cw4yze.jpg \r\n", + " inflating: ./data/segmented-images/images/cju2omjpeqj5a0988pjdlb8l1.jpg \r\n", + " inflating: ./data/segmented-images/images/cju7crgxa28550755wbsgqkel.jpg \r\n", + " inflating: ./data/segmented-images/images/cju6xlygpw7bs0818n691jsq4.jpg \r\n", + " inflating: ./data/segmented-images/images/cju7787c5yy3l080159mwqsnj.jpg \r\n", + " inflating: ./data/segmented-images/images/cju1bhnfitmge0835ynls0l6b.jpg \r\n", + " inflating: ./data/segmented-images/images/cju5c7oijaqmq09878qwgqv8n.jpg \r\n", + " inflating: ./data/segmented-images/images/cju7ezs7g2mxm098787atbran.jpg \r\n", + " inflating: ./data/segmented-images/images/cju2qtee81yd708787bsjr75d.jpg \r\n", + " inflating: ./data/segmented-images/images/cju42g865lorv07552ytz6xxa.jpg \r\n" + ] + }, + { + "name": "stdout", + "output_type": "stream", + "text": [ + " inflating: ./data/segmented-images/images/cju6us80mv1b50871ebyq2wxa.jpg \r\n", + " inflating: ./data/segmented-images/images/cju87vqa0ndwg0850onjdz7ol.jpg \r\n", + " inflating: ./data/segmented-images/images/cju2sxf3iqbpv09937iksn8ep.jpg \r\n", + " inflating: ./data/segmented-images/images/cju5wtdu4m0im0871mix0yvc0.jpg \r\n", + " inflating: ./data/segmented-images/images/cju8ando2qqdo0818ck7i1be1.jpg \r\n", + " inflating: ./data/segmented-images/images/cju85plp7lmkw0850rx42jdpf.jpg \r\n", + " inflating: ./data/segmented-images/images/cju2u4pymvc720988wsxrmi84.jpg \r\n", + " inflating: ./data/segmented-images/images/cju35k2fr3vc50988c85qkrwg.jpg \r\n", + " inflating: ./data/segmented-images/images/cju2qz06823a40878ojcz9ccx.jpg \r\n", + " inflating: ./data/segmented-images/images/cju324q101fhe08350wae9cif.jpg \r\n", + " inflating: ./data/segmented-images/images/cju42u5bjlvi10801dc13sskp.jpg \r\n", + " inflating: ./data/segmented-images/images/cju7bb3ss1uoo0755pmhyco7t.jpg \r\n", + " inflating: ./data/segmented-images/images/cju2oo0wh0bqy0878biujeyhe.jpg \r\n", + " inflating: ./data/segmented-images/images/cju87z6o6nh73085045bzsx6o.jpg \r\n", + " inflating: ./data/segmented-images/images/cju1haab178i70799tk9z8y8x.jpg \r\n", + " inflating: ./data/segmented-images/images/cju7dda8w2br20818zhsuz8s7.jpg \r\n", + " inflating: ./data/segmented-images/images/cju2trbpkv0c00988hxla5dzz.jpg \r\n", + " inflating: ./data/segmented-images/images/cju1c6yfz42md08550zgoz3pw.jpg \r\n", + " inflating: ./data/segmented-images/images/cju7dsrtb2f8i085064kwugfk.jpg \r\n", + " inflating: ./data/segmented-images/images/cju2t16vuucaq0835xcpsivn2.jpg \r\n", + " inflating: ./data/segmented-images/images/cju3521y5d5mq0878t3ezsu4p.jpg \r\n", + " inflating: ./data/segmented-images/images/cju2hdr06v2bq0799mbm3bks1.jpg \r\n", + " inflating: ./data/segmented-images/images/cju41s6nbleqy0755e2mslg0b.jpg \r\n", + " inflating: ./data/segmented-images/images/cju13hp5rnbjx0835bf0jowgx.jpg \r\n", + " inflating: ./data/segmented-images/images/cju1fm3id6gl50801r3fok20c.jpg \r\n", + " inflating: ./data/segmented-images/images/cju2pmhtr17a00855cvpelzb0.jpg \r\n", + " inflating: ./data/segmented-images/images/cju2u2b9o4zvp08788qb9nqxj.jpg \r\n", + " inflating: ./data/segmented-images/images/cju8a56vxpy780850r45yu4wk.jpg \r\n", + " inflating: ./data/segmented-images/images/cju6vvb8svhed0801jjcquh5e.jpg \r\n", + " inflating: ./data/segmented-images/images/cju2sggy13na70855tbeoqgha.jpg \r\n", + " inflating: ./data/segmented-images/images/ck2bxknhjvs1x0794iogrq49k.jpg \r\n", + " inflating: ./data/segmented-images/images/cju785htizjzo08017tvlhtg4.jpg \r\n", + " inflating: ./data/segmented-images/images/cju5bdwa3aatx0818b79i18zf.jpg \r\n", + " inflating: ./data/segmented-images/images/cju5yjq1pmlgc0801z0t24bly.jpg \r\n", + " inflating: ./data/segmented-images/images/cju16b6ynq8e40988m8vx0xnj.jpg \r\n", + " inflating: ./data/segmented-images/images/cju41nz76lcxu0755cya2qefx.jpg \r\n", + " inflating: ./data/segmented-images/images/cju5xopi0md7q0871r1sjc1av.jpg \r\n", + " inflating: ./data/segmented-images/images/cju8bafgqrf4x0818twisk3ea.jpg \r\n", + " inflating: ./data/segmented-images/images/cju43b8daly4408170e5ev06g.jpg \r\n", + " inflating: ./data/segmented-images/images/cju0rx1idathl0835detmsp84.jpg \r\n", + " inflating: ./data/segmented-images/images/cju2qvuj1s9ok0835tp2k4ozh.jpg \r\n", + " inflating: ./data/segmented-images/images/cju33w4sdcivk0855x879zht7.jpg \r\n", + " inflating: ./data/segmented-images/images/cju2ycp1u8g2r0799jslnp7cz.jpg \r\n", + " inflating: ./data/segmented-images/images/cju2tjrog4jy30878pawyazqc.jpg \r\n", + " inflating: ./data/segmented-images/images/cju5jz5fff8c50871hbe6108f.jpg \r\n", + " inflating: ./data/segmented-images/images/cju6xa0qmvzun0818xjukgncj.jpg \r\n", + " inflating: ./data/segmented-images/images/cju8bysfgrzkl081786jwac09.jpg \r\n", + " inflating: ./data/segmented-images/images/cju5bhv81abur0850ean02atv.jpg \r\n", + " inflating: ./data/segmented-images/images/cju42nm68lpyo0818xvvqmupq.jpg \r\n", + " inflating: ./data/segmented-images/images/cju6wi3akvn8r0801px8eligc.jpg \r\n", + " inflating: ./data/segmented-images/images/cju1fuoa4wmc50835qfd11sp9.jpg \r\n", + " inflating: ./data/segmented-images/images/cju2tzypl4wss0799ow05oxb9.jpg \r\n", + " inflating: ./data/segmented-images/images/cju8djdqztu6408506pzhlo18.jpg \r\n", + " inflating: ./data/segmented-images/images/cju1fr4etsmrr09933u4t4aql.jpg \r\n", + " inflating: ./data/segmented-images/images/cju2pkwt3r8b90988v2ywq1px.jpg \r\n", + " inflating: ./data/segmented-images/images/cju336l68y7if0993wf092166.jpg \r\n", + " inflating: ./data/segmented-images/images/cju5k3j3uf6de0817hszzfr7n.jpg \r\n", + " inflating: ./data/segmented-images/images/cju5ft6mcd5q40987rhjgbrr6.jpg \r\n", + " inflating: ./data/segmented-images/images/cju7db7lp2f400755tntd1ohf.jpg \r\n", + " inflating: ./data/segmented-images/images/cju893jmdompz0817xn3g1w4h.jpg \r\n", + " inflating: ./data/segmented-images/images/cju1cqc7n4gpy0855jt246k68.jpg \r\n", + " inflating: ./data/segmented-images/images/cju5enq1tcn1i0755hnkon787.jpg \r\n", + " inflating: ./data/segmented-images/images/cju8cgi2kspp308011nxdtjp6.jpg \r\n", + " inflating: ./data/segmented-images/images/cju5hl8nee8a40755fm8qjj0o.jpg \r\n", + " inflating: ./data/segmented-images/images/cju8c5mxls96t0850wvkvsity.jpg \r\n", + " inflating: ./data/segmented-images/images/cju3u1c8tfyqx08503iedc3mx.jpg \r\n", + " inflating: ./data/segmented-images/images/cju89z6pqpqfx0817mfv8ixjc.jpg \r\n", + " inflating: ./data/segmented-images/images/cju6xmqd9w0250817l5kxfnsk.jpg \r\n", + " inflating: ./data/segmented-images/images/cju2m71z2ywwv080131bcrsd3.jpg \r\n", + " inflating: ./data/segmented-images/images/cju34sh43d8zm08019xbwhc0o.jpg \r\n", + " inflating: ./data/segmented-images/images/cju2nyc5f02m40801ojqbtiea.jpg \r\n", + " inflating: ./data/segmented-images/images/cju7fd6yt2p740987wkr8exo1.jpg \r\n", + " inflating: ./data/segmented-images/images/cju846ec0kj7z08012o10klrb.jpg \r\n", + " inflating: ./data/segmented-images/images/cju7ey10f2rvf0871bwbi9x82.jpg \r\n", + " inflating: ./data/segmented-images/images/cju2u73dj53oz0878486k8k4b.jpg \r\n", + " inflating: ./data/segmented-images/images/cju7apr0c1qqm0755s7msqot4.jpg \r\n", + " inflating: ./data/segmented-images/images/cju83h9ysjwe808716nt35oah.jpg \r\n", + " inflating: ./data/segmented-images/images/cju3xzvnzj0hd0755xprz39nj.jpg \r\n", + " inflating: ./data/segmented-images/images/cju7fbndk2sl608015ravktum.jpg \r\n", + " inflating: ./data/segmented-images/images/cju85mpuglq8k0818d2it6hzb.jpg \r\n", + " inflating: ./data/segmented-images/images/cju2zblxw9848087853csbrx1.jpg \r\n", + " inflating: ./data/segmented-images/images/cju6v1m1xv07w09870ah3njy1.jpg \r\n", + " inflating: ./data/segmented-images/images/cju8d4jgatgpj0871q2ophhkm.jpg \r\n", + " inflating: ./data/segmented-images/images/cju2zo0fwzv580988qlijd2xa.jpg \r\n", + " inflating: ./data/segmented-images/images/cju2hewssldzx0835ep795xu0.jpg \r\n", + " inflating: ./data/segmented-images/images/cju1b0y2e396p08558ois175d.jpg \r\n", + " inflating: ./data/segmented-images/images/cju1ftaji6isw0855108yqcse.jpg \r\n", + " inflating: ./data/segmented-images/images/cju7aifyo1p3n07552nxjx51f.jpg \r\n", + " inflating: ./data/segmented-images/images/cju418rckl3ur08012psrx1r1.jpg \r\n", + " inflating: ./data/segmented-images/images/cju6vqarjv7yo0987q4b1btk1.jpg \r\n", + " inflating: ./data/segmented-images/images/cju8adb60qbiu080188mxpf8d.jpg \r\n", + " inflating: ./data/segmented-images/images/cju8aj01yqeqm0850lhdz3xdw.jpg \r\n", + " inflating: ./data/segmented-images/images/cju5uxjnol2r509871qv2yeia.jpg \r\n", + " inflating: ./data/segmented-images/images/cju8aeei7q8k308173n9y4klv.jpg \r\n", + " inflating: ./data/segmented-images/images/cju2hlm19vjjf0801o69qnber.jpg \r\n", + " inflating: ./data/segmented-images/images/cju33o12x2jm50988944mxq0v.jpg \r\n", + " inflating: ./data/segmented-images/images/cju5vwbr4lhqn0987a1pji0ux.jpg \r\n", + " inflating: ./data/segmented-images/images/cju88vx2uoocy075531lc63n3.jpg \r\n", + " inflating: ./data/segmented-images/images/cju7et17a2vjk0755e743npl1.jpg \r\n", + " inflating: ./data/segmented-images/images/cju43mkj9m8wb0871qiadahub.jpg \r\n", + " inflating: ./data/segmented-images/images/cju5o1vu9gz8a0818eyy92bns.jpg \r\n", + " inflating: ./data/segmented-images/images/cju77re6fz5bb0817vp9redjg.jpg \r\n", + " inflating: ./data/segmented-images/images/cju160wshltz10993i1gmqxbe.jpg \r\n", + " inflating: ./data/segmented-images/images/cju34ymm8d6700799uop0cw33.jpg \r\n", + " inflating: ./data/segmented-images/images/cju2rkjfwoxys0993x768l1j2.jpg \r\n", + " inflating: ./data/segmented-images/images/cju783tmkzkqu081803g7q5vk.jpg \r\n", + " inflating: ./data/segmented-images/images/cju7amjna1ly40871ugiokehb.jpg \r\n", + " inflating: ./data/segmented-images/images/cju0sxqiclckk08551ycbwhno.jpg \r\n", + " inflating: ./data/segmented-images/images/cju7f4sc62xqj075597xpmuoy.jpg \r\n", + " inflating: ./data/segmented-images/images/cju45rj7ln8980850a7821fov.jpg \r\n", + " inflating: ./data/segmented-images/images/cju1fjsb4sipq09931lvd8e41.jpg \r\n", + " inflating: ./data/segmented-images/images/cju7fcgbe2z3p07550vaflqdb.jpg \r\n", + " inflating: ./data/segmented-images/images/cju2qdj95ru8g09886gfi9rsz.jpg \r\n", + " inflating: ./data/segmented-images/images/cju8buos5rz9b08715lfr0f4f.jpg \r\n", + " inflating: ./data/segmented-images/images/cju2zr3c3vwb00993jn06bbaz.jpg \r\n", + " inflating: ./data/segmented-images/images/cju1euuc65wm00799m4sjdnnn.jpg \r\n", + " inflating: ./data/segmented-images/images/cju8cj10qsrau0871o2dr6ai1.jpg \r\n", + " inflating: ./data/segmented-images/images/cju5klveuff6w0871wbibgh3m.jpg \r\n", + " inflating: ./data/segmented-images/images/cju5x00l6m5j608503k78ptee.jpg \r\n", + " inflating: ./data/segmented-images/images/cju42qet0lsq90871e50xbnuv.jpg \r\n", + " inflating: ./data/segmented-images/images/cju330ofbc2l30801th5g3hw6.jpg \r\n", + " inflating: ./data/segmented-images/images/cju320gyvbch60801v2amdi2g.jpg \r\n", + " inflating: ./data/segmented-images/images/cju5bf6hxa6m50817rbwettgu.jpg \r\n", + " inflating: ./data/segmented-images/images/cju5thdbrjp1108715xdfx356.jpg \r\n", + " inflating: ./data/segmented-images/images/cju1hhj6mxfp90835n3wofrap.jpg \r\n", + " inflating: ./data/segmented-images/images/ck2bxskgxxzfv08386xkqtqdy.jpg \r\n", + " inflating: ./data/segmented-images/images/cju5hwonqedw10801vsd3w6kk.jpg \r\n", + " inflating: ./data/segmented-images/images/cju888fr7nveu0818r9uwtiit.jpg \r\n", + " inflating: ./data/segmented-images/images/cju1c0qb4tzi308355wtsnp0y.jpg \r\n", + " inflating: ./data/segmented-images/images/cju6wll7wvo3y08502pagos8m.jpg \r\n", + " inflating: ./data/segmented-images/images/cju8b2rmgr52s0801p54eyflx.jpg \r\n", + " inflating: ./data/segmented-images/images/cju6xifswvwbo0987nibtdr50.jpg \r\n", + " inflating: ./data/segmented-images/images/cju45v0pungu40871acnwtmu5.jpg \r\n", + " inflating: ./data/segmented-images/images/cju7d2q1k27nf08715zshsckt.jpg \r\n", + " inflating: ./data/segmented-images/images/cju8azmhcr66e0755t61atz72.jpg \r\n", + " inflating: ./data/segmented-images/images/cju8dic9mtppa0987swn23wbc.jpg \r\n", + " inflating: ./data/segmented-images/images/cju424hy5lckr085073fva1ok.jpg \r\n", + " inflating: ./data/segmented-images/images/cju8bi8q7rlmn0871abc5ch8k.jpg \r\n", + " inflating: ./data/segmented-images/images/cju1d50a94qf50855wsowacrc.jpg \r\n", + " inflating: ./data/segmented-images/images/cju1amqw6p8pw0993d9gc5crl.jpg \r\n", + " inflating: ./data/segmented-images/images/cju5huurrecm70801y680y13m.jpg \r\n", + " inflating: ./data/segmented-images/images/cju1brhsj3rls0855a1vgdlen.jpg \r\n", + " inflating: ./data/segmented-images/images/cju85dx63lic408017f0l0400.jpg \r\n", + " inflating: ./data/segmented-images/images/cju88t4fvokxf07558ymyh281.jpg \r\n", + " inflating: ./data/segmented-images/images/cju31w6goazci0799n014ly1q.jpg \r\n", + " inflating: ./data/segmented-images/images/cju8bbznkrf5g0871jncffynk.jpg \r\n", + " inflating: ./data/segmented-images/images/cju85citjlnfm0755i4rk5tqj.jpg \r\n", + " inflating: ./data/segmented-images/images/cju3x5u2tiihx0818914gzxy1.jpg \r\n", + " inflating: ./data/segmented-images/images/cju1djtprvd7b0988thwwrg09.jpg \r\n", + " inflating: ./data/segmented-images/images/cju7fmvpk2q170987v6i3ola8.jpg \r\n", + " inflating: ./data/segmented-images/images/cju7fq7mm2pw508176uk5ugtx.jpg \r\n", + " inflating: ./data/segmented-images/images/cju774fmayxif0818u2g79usw.jpg \r\n", + " inflating: ./data/segmented-images/images/cju2zrojo9kcd0878ld2epejq.jpg \r\n", + " inflating: ./data/segmented-images/images/cju6z1bzbwfq50817b2alatvr.jpg \r\n", + " inflating: ./data/segmented-images/images/cju6ut4l8va6y0755tyw3vfqq.jpg \r\n", + " inflating: ./data/segmented-images/images/cju5yimthmlv80850zhoc90c2.jpg \r\n", + " inflating: ./data/segmented-images/images/cju5g163vd6mt0817uccuga6u.jpg \r\n", + " inflating: ./data/segmented-images/images/cju84ffdzkrjn08183jh1fxmb.jpg \r\n", + " inflating: ./data/segmented-images/images/cju5y7buemcw80987p0r30g9f.jpg \r\n", + " inflating: ./data/segmented-images/images/cju5bmhdcafs909878qfzrqzi.jpg \r\n", + " inflating: ./data/segmented-images/images/cju8a2itsq4dv0755ntlovpxe.jpg \r\n", + " inflating: ./data/segmented-images/images/cju1gghyjwxt80835vx0wgxw0.jpg \r\n", + " inflating: ./data/segmented-images/images/cju30ov1oah920801mi8thuyg.jpg \r\n", + " inflating: ./data/segmented-images/images/cju5v8pgplg6k0755rvi2t63h.jpg \r\n", + " inflating: ./data/segmented-images/images/cju85nr8elly209872w9n5m0s.jpg \r\n", + " inflating: ./data/segmented-images/images/cju2spdagu1l50835da1f46fr.jpg \r\n", + " inflating: ./data/segmented-images/images/cju7dff9529h208503w60lbil.jpg \r\n", + " inflating: ./data/segmented-images/images/cju7bf1lp1shi081835vs84lc.jpg \r\n", + " inflating: ./data/segmented-images/images/cju32upim1z7u0988l883nqp6.jpg \r\n", + " inflating: ./data/segmented-images/images/cju7evxt12m730987rxivne3x.jpg \r\n", + " inflating: ./data/segmented-images/images/cju2mfjndoz700988b9lc3zeq.jpg \r\n", + " inflating: ./data/segmented-images/images/cju88v2f9oi8w0871hx9auh01.jpg \r\n", + " inflating: ./data/segmented-images/images/cju35a77vdj4n08556jj2lgmc.jpg \r\n", + " inflating: ./data/segmented-images/images/cju1cdxvz48hw0801i0fjwcnk.jpg \r\n", + " inflating: ./data/segmented-images/images/cju8bpctzrqkr0850zeldv9kt.jpg \r\n", + " inflating: ./data/segmented-images/images/cju2nbdpmlmcj0993s1cht0dz.jpg \r\n", + " inflating: ./data/segmented-images/images/cju7ahtkb1jr90801jck4kbds.jpg \r\n", + " inflating: ./data/segmented-images/images/cju8b542nr81x0871uxnkm9ih.jpg \r\n", + " inflating: ./data/segmented-images/images/cju7fpfzq2wyf0818xxd1oziv.jpg \r\n", + " inflating: ./data/segmented-images/images/cju5ufn3skquf0818dhapnhba.jpg \r\n", + " inflating: ./data/segmented-images/images/cju45t5ddnbio0987qtqzx762.jpg \r\n", + " inflating: ./data/segmented-images/images/cju42dwedlmk60871jbgu4ehi.jpg \r\n", + " inflating: ./data/segmented-images/images/cju5f0dezct4q08183ydw11dx.jpg \r\n", + " inflating: ./data/segmented-images/images/cju3xl264ingx0850rcf0rshj.jpg \r\n", + " inflating: ./data/segmented-images/images/cju3tvffffx5f0818t5ov22al.jpg \r\n", + " inflating: ./data/segmented-images/images/cju2ij9uiic2l09933ljiv6gm.jpg \r\n", + " inflating: ./data/segmented-images/images/cju0roawvklrq0799vmjorwfv.jpg \r\n", + " inflating: ./data/segmented-images/images/cju33qpdvc9g0087825jhf3s9.jpg \r\n", + " inflating: ./data/segmented-images/images/cju849c23kgnk0817cgv2hw1e.jpg \r\n", + " inflating: ./data/segmented-images/images/cju2rxm8rpbaf0993o3qr2oph.jpg \r\n", + " inflating: ./data/segmented-images/images/cju40u30gkuzc0871rq7t666d.jpg \r\n", + " inflating: ./data/segmented-images/images/cju3v11mrgwwb0755u242ygye.jpg \r\n", + " inflating: ./data/segmented-images/images/cju7d5m0p23kn09871rk7pu3v.jpg \r\n", + " inflating: ./data/segmented-images/images/cju8bssulrrcy0987h1vq5060.jpg \r\n", + " inflating: ./data/segmented-images/images/cju5y84q3mdv50817eyp82xf3.jpg \r\n", + " inflating: ./data/segmented-images/images/cju43in5fm22c08175rxziqrk.jpg \r\n", + " inflating: ./data/segmented-images/images/cju5wrrs0m2af0818vmnajbtw.jpg \r\n", + " inflating: ./data/segmented-images/images/cju5wi6bqlxy90755bu227nvb.jpg \r\n", + " inflating: ./data/segmented-images/images/cju1cu1u2474n0878tt7v4tdr.jpg \r\n", + " inflating: ./data/segmented-images/images/cju8bn7m2rmm70817hgxpb1uq.jpg \r\n", + " inflating: ./data/segmented-images/images/cju1dhfok4mhe0878jlgrag0h.jpg \r\n", + " inflating: ./data/segmented-images/images/cju45n0oxn5vu08500yfrt9jn.jpg \r\n", + " inflating: ./data/segmented-images/images/cju3y0pjrj1c30755nxekxccj.jpg \r\n", + " inflating: ./data/segmented-images/images/cju858eswlepn0871pzvdrhj1.jpg \r\n", + " inflating: ./data/segmented-images/images/cju88evxanv9r08176zkeovec.jpg \r\n", + " inflating: ./data/segmented-images/images/cju5f8hxdcxxn08188obby0ea.jpg \r\n", + " inflating: ./data/segmented-images/images/cju5bbtwsa8cl0987wgfsqpao.jpg \r\n", + " inflating: ./data/segmented-images/images/cju35i2e63uxr0835h7zgkg9k.jpg \r\n", + " inflating: ./data/segmented-images/images/cju2r7h21sj9608354gzks3ae.jpg \r\n", + " inflating: ./data/segmented-images/images/cju2z9vlp9j0w0801oag91sy9.jpg \r\n", + " inflating: ./data/segmented-images/images/cju2okvco06xc0799kxe5n1qh.jpg \r\n", + " inflating: ./data/segmented-images/images/cju7alcgr1lsr0871riqk84z7.jpg \r\n", + " inflating: ./data/segmented-images/images/cju7azuu31mia0801pf9ib5ed.jpg \r\n", + " inflating: ./data/segmented-images/images/cju2nfnvxzdkd0878399axlco.jpg \r\n", + " inflating: ./data/segmented-images/images/cju34fojcctcf0799ebolbvkn.jpg \r\n", + " inflating: ./data/segmented-images/images/cju8bh8surexp0987o5pzklk1.jpg \r\n", + " inflating: ./data/segmented-images/images/cju8418jhkf7d0818ga2v0xq0.jpg \r\n", + " inflating: ./data/segmented-images/images/cju43o6n7m9nk087191ijwqq9.jpg \r\n", + " inflating: ./data/segmented-images/images/cju2trtjf4qjd0878a2zle9v9.jpg \r\n", + " inflating: ./data/segmented-images/images/cju5vbo6jldrt0871jf6f1700.jpg \r\n", + " inflating: ./data/segmented-images/images/cju3xga12iixg0817dijbvjxw.jpg \r\n", + " inflating: ./data/segmented-images/images/cju7dymur2od30755eg8yv2ht.jpg \r\n", + " inflating: ./data/segmented-images/images/cju5ccpvqash50850kb4bs22k.jpg \r\n", + " inflating: ./data/segmented-images/images/cju8c5zcbsdfz0801o5t6jag1.jpg \r\n", + " inflating: ./data/segmented-images/images/cju2y5zas8m7f0801d34g5owq.jpg \r\n", + " inflating: ./data/segmented-images/images/cju2nnqrqzp580855z8mhzgd6.jpg \r\n", + " inflating: ./data/segmented-images/images/cju3u815rg4ek0850vvhtcvcm.jpg \r\n", + " inflating: ./data/segmented-images/images/cju3xhpvvimda0987ygrpzni2.jpg \r\n", + " inflating: ./data/segmented-images/images/cju3y21quj0ir0818kgjagr15.jpg \r\n", + " inflating: ./data/segmented-images/images/cju30u1hbakn808019g15nb8b.jpg \r\n", + " inflating: ./data/segmented-images/images/cju8c6hnxsdvr0801wn0vrsa6.jpg \r\n", + " inflating: ./data/segmented-images/images/cju5fu081d8gc0818l3yylujk.jpg \r\n", + " inflating: ./data/segmented-images/images/cju77g99iyxc00817zqi2ppor.jpg \r\n", + " inflating: ./data/segmented-images/images/cju30lncba3ny0878jwnous8n.jpg \r\n", + " inflating: ./data/segmented-images/images/cju43h43am1dy08176gwfhmnt.jpg \r\n", + " inflating: ./data/segmented-images/images/cju8cattbsivm0818p446wgel.jpg \r\n", + " inflating: ./data/segmented-images/images/cju1csmlc4ht10799b8ymmghg.jpg \r\n", + " inflating: ./data/segmented-images/images/cju35ldepdtlm0801yv79y8vu.jpg \r\n", + " inflating: ./data/segmented-images/images/cju1euant5l960878iqj5vvto.jpg \r\n", + " inflating: ./data/segmented-images/images/cju87q6yoneim0871dl4phvkd.jpg \r\n", + " inflating: ./data/segmented-images/images/cju5vzjoslpj708186z2fusmz.jpg \r\n", + " inflating: ./data/segmented-images/images/cju5wmvsdlx1j0871npgj8j4b.jpg \r\n", + " inflating: ./data/segmented-images/images/cju83mki1jv5w0817kubxm31r.jpg \r\n", + " inflating: ./data/segmented-images/images/cju40taxlkrho0987smigg0x0.jpg \r\n", + " inflating: ./data/segmented-images/images/cju32pzh9bpw10855q4vaxfhe.jpg \r\n", + " inflating: ./data/segmented-images/images/cju3x9lttikfb0818a0g104zn.jpg \r\n", + " inflating: ./data/segmented-images/images/cju773hsyyosz0817pk1e7sjq.jpg \r\n", + " inflating: ./data/segmented-images/images/cju2zwg05a0oy0801yr73ig7g.jpg \r\n", + " inflating: ./data/segmented-images/images/cju2tvrvm53ws0801a0jfjdxg.jpg \r\n", + " inflating: ./data/segmented-images/images/cju3uwz6ogsp10801h2r3bj5l.jpg \r\n", + " inflating: ./data/segmented-images/images/cju2y0z6g87p10878fpk5d3rq.jpg \r\n", + " inflating: ./data/segmented-images/images/cju32jcdabepz0878d0cznmfe.jpg \r\n", + " inflating: ./data/segmented-images/images/cju15mhjczc8z0801kit5c6di.jpg \r\n", + " inflating: ./data/segmented-images/images/cju8432cmkgq90871cxe4iptl.jpg \r\n", + " inflating: ./data/segmented-images/images/cju5w7xn0lrkq0801f9k0htgx.jpg \r\n", + " inflating: ./data/segmented-images/images/cju7d7aut2a2p0818z4uxc6cd.jpg \r\n", + " inflating: ./data/segmented-images/images/cju2i8br1vqtd08784u6vmcjk.jpg \r\n", + " inflating: ./data/segmented-images/images/cju2qh5le1ock0878oahaql7d.jpg \r\n", + " inflating: ./data/segmented-images/images/ck2bxpfgxu2mk0748gsh7xelu.jpg \r\n", + " inflating: ./data/segmented-images/images/cju87kbcen2av0987usezo8kn.jpg \r\n", + " inflating: ./data/segmented-images/images/cju7bmi1v1pnj0987pa52jjok.jpg \r\n", + " inflating: ./data/segmented-images/images/cju5o4pk9h0720755lgp9jq8m.jpg \r\n", + " inflating: ./data/segmented-images/images/cju7f6cqy2ur20818t1saazbm.jpg \r\n", + " inflating: ./data/segmented-images/images/cju1cyjb5qtie0993njqne9m3.jpg \r\n", + " inflating: ./data/segmented-images/images/ck2bxiswtxuw80838qkisqjwz.jpg \r\n", + " inflating: ./data/segmented-images/images/cju7frtqu2xa20818wq8r9fzf.jpg \r\n", + " inflating: ./data/segmented-images/images/cju31ugmfb3dz0855xtqshki6.jpg \r\n", + " inflating: ./data/segmented-images/images/cju2xjz2ju8pe0993ysv9wg17.jpg \r\n", + " inflating: ./data/segmented-images/images/cju784jpdzeae0987q5ypq883.jpg \r\n" + ] + }, + { + "name": "stdout", + "output_type": "stream", + "text": [ + " inflating: ./data/segmented-images/images/cju8b8yair65w09878pyqtr96.jpg \r\n", + " inflating: ./data/segmented-images/images/cju7cp6dw244p0818gncdol4m.jpg \r\n", + " inflating: ./data/segmented-images/images/cju88aq6vo1ij0755c2ey7z7n.jpg \r\n", + " inflating: ./data/segmented-images/images/cju3v664kh0px0818y4y7wolf.jpg \r\n", + " inflating: ./data/segmented-images/images/cju5x15djm7ae0755h8czf6nt.jpg \r\n", + " inflating: ./data/segmented-images/images/cju88l66no10s0850rsda7ej1.jpg \r\n", + " inflating: ./data/segmented-images/images/cju6ywm40wdbo0987pbftsvtg.jpg \r\n", + " inflating: ./data/segmented-images/images/cju17r8il13910799dr2wme2e.jpg \r\n", + " inflating: ./data/segmented-images/images/cju7cufm7298k0755j09uf3of.jpg \r\n", + " inflating: ./data/segmented-images/images/cju2xf8e5y2wm08359vcgk09b.jpg \r\n", + " inflating: ./data/segmented-images/images/cju7g7ba42z310987bqzbi2bq.jpg \r\n", + " inflating: ./data/segmented-images/images/cju3ul8dogf1z09872y2ecowp.jpg \r\n", + " inflating: ./data/segmented-images/images/cju33eqwbcch208012jikwdky.jpg \r\n", + " inflating: ./data/segmented-images/images/cju2hfqnmhisa0993gpleeldd.jpg \r\n", + " inflating: ./data/segmented-images/images/cju34c1xfyz920993itxkkfad.jpg \r\n", + " inflating: ./data/segmented-images/images/cju43gfosm63n08714rpih8pe.jpg \r\n", + " inflating: ./data/segmented-images/images/cju7dhpsc2dnn0818025m6857.jpg \r\n", + " inflating: ./data/segmented-images/images/cju34ds2531520988qjpqt6e3.jpg \r\n", + " inflating: ./data/segmented-images/images/cju88gx09o2vk0818610zody3.jpg \r\n", + " inflating: ./data/segmented-images/images/cju6uy20suzbl0987rzuhz7z9.jpg \r\n", + " inflating: ./data/segmented-images/images/cju1g20bdwq6u0835e16xugcd.jpg \r\n", + " inflating: ./data/segmented-images/images/cju6x35ervu2808015c7eoqe4.jpg \r\n", + " inflating: ./data/segmented-images/images/cju5eftctcdbj08712gdp989f.jpg \r\n", + " inflating: ./data/segmented-images/images/cju30qbm1ad3x0855znuhpz9u.jpg \r\n", + " inflating: ./data/segmented-images/images/cju7awzmu1ncs0871hziy65zx.jpg \r\n", + " inflating: ./data/segmented-images/images/cju7csvlb22fr0850lvm45n3x.jpg \r\n", + " inflating: ./data/segmented-images/images/cju87xn2snfmv0987sc3d9xnq.jpg \r\n", + " inflating: ./data/segmented-images/images/cju7bgnvb1sf808717qa799ir.jpg \r\n", + " inflating: ./data/segmented-images/images/cju7d7tly27h408016fyp5nr7.jpg \r\n", + " inflating: ./data/segmented-images/images/cju3yb47cj1xq0817zfotbni4.jpg \r\n", + " inflating: ./data/segmented-images/images/cju412uwlkva50850d1ps1ww7.jpg \r\n", + " inflating: ./data/segmented-images/images/cju85rkbnlo1c08503uxcpax1.jpg \r\n", + " inflating: ./data/segmented-images/images/cju88cddensj00987788yotmg.jpg \r\n", + " inflating: ./data/segmented-images/images/cju8b5p40r2c60987ofa0mu03.jpg \r\n", + " inflating: ./data/segmented-images/images/cju13cgqmnhwn0988yrainhcp.jpg \r\n", + " inflating: ./data/segmented-images/images/cju8d2q30tfhs0801n7lx77xl.jpg \r\n", + " inflating: ./data/segmented-images/images/cju2ysg748ru80878sp6j0gm0.jpg \r\n", + " inflating: ./data/segmented-images/images/cju3x7xsaijq80818f0psavav.jpg \r\n", + " inflating: ./data/segmented-images/images/cju85qefyln6v0850szeb9byi.jpg \r\n", + " inflating: ./data/segmented-images/images/cju2zdhsczmn50988z64qwg2q.jpg \r\n", + " inflating: ./data/segmented-images/images/cju7b1ygu1msd0801hywhy0mc.jpg \r\n", + " inflating: ./data/segmented-images/images/cju41lojblbs307555jdci937.jpg \r\n", + " inflating: ./data/segmented-images/images/cju34ouumcznz07996gg1xq7v.jpg \r\n", + " inflating: ./data/segmented-images/images/cju34zivp3fq80988opxbaqyn.jpg \r\n", + " inflating: ./data/segmented-images/images/cju2lberzkdzm09938cl40pog.jpg \r\n", + " inflating: ./data/segmented-images/images/cju6vvxsev9y30987kespucdg.jpg \r\n", + " inflating: ./data/segmented-images/images/cju2raxlosl630988jdbfy9b0.jpg \r\n", + " inflating: ./data/segmented-images/images/cju7atnm31if40817pqclnjer.jpg \r\n", + " inflating: ./data/segmented-images/images/cju5xq3tdm9fn0987pbedxdg5.jpg \r\n", + " inflating: ./data/segmented-images/images/cju183od81ff608017ekzif89.jpg \r\n", + " inflating: ./data/segmented-images/images/cju7aqkue1i2k09879uzcpt8r.jpg \r\n", + " inflating: ./data/segmented-images/images/cju2rn0hasxri0835nfy3buay.jpg \r\n", + " inflating: ./data/segmented-images/images/cju2t3ibkuecr0835o7si16zv.jpg \r\n", + " inflating: ./data/segmented-images/images/cju5x28nzm7t907558ocq4bt7.jpg \r\n", + " inflating: ./data/segmented-images/images/cju787jnjzjuj0871p94nck9g.jpg \r\n", + " inflating: ./data/segmented-images/images/cju2pag1f0s4r0878h52uq83s.jpg \r\n", + " inflating: ./data/segmented-images/images/cju5boicjagt20871b1fotkh4.jpg \r\n", + " inflating: ./data/segmented-images/images/cju3v72v5h1qz0818fggilwtq.jpg \r\n", + " inflating: ./data/segmented-images/images/cju1fj6axwfp30835ukhuzhw9.jpg \r\n", + " inflating: ./data/segmented-images/images/cju2phaksnahz0993yxogjcpv.jpg \r\n", + " inflating: ./data/segmented-images/images/cju87tyddnnad0755bj0wxahe.jpg \r\n", + " inflating: ./data/segmented-images/images/cju77q10sz9ug0801449wu1nu.jpg \r\n", + " inflating: ./data/segmented-images/images/cju35c4wzdhow0799h6eq4sgs.jpg \r\n", + " inflating: ./data/segmented-images/images/cju33yemn2qb20988wfjxximx.jpg \r\n", + " inflating: ./data/segmented-images/images/cju43eigtm6ev0801mv0m96t1.jpg \r\n", + " inflating: ./data/segmented-images/images/cju17z0qongpa0993de4boim4.jpg \r\n", + " inflating: ./data/segmented-images/images/cju7d6ux323ze0987xos3srkx.jpg \r\n", + " inflating: ./data/segmented-images/images/cju32fhnhbds40799broyoptc.jpg \r\n", + " inflating: ./data/segmented-images/images/cju1f320ewfyu0988ndz6blh5.jpg \r\n", + " inflating: ./data/segmented-images/images/cju16jgnyzp970878melv7r25.jpg \r\n", + " inflating: ./data/segmented-images/images/cju77b3wyz4160755qis4ljsb.jpg \r\n", + " inflating: ./data/segmented-images/images/cju5bycdkalkb09875f7bfrvx.jpg \r\n", + " inflating: ./data/segmented-images/images/cju2zpw4q9vzr0801p0lysjdl.jpg \r\n", + " inflating: ./data/segmented-images/images/cju32qr9tbvsj08013pkpjenq.jpg \r\n", + " inflating: ./data/segmented-images/images/cju3uhb79gcgr0871orbrbi3x.jpg \r\n", + " inflating: ./data/segmented-images/images/cju6wt9jvvn500871hjn3t3g0.jpg \r\n", + " inflating: ./data/segmented-images/images/cju2z6ez69g4u0801qwt088lw.jpg \r\n", + " inflating: ./data/segmented-images/images/cju303j5r062k098835zxfds5.jpg \r\n", + " inflating: ./data/segmented-images/images/cju30gxjq0djk0988jytm49rs.jpg \r\n", + " inflating: ./data/segmented-images/images/cju7cq6su27qv075574dir0r3.jpg \r\n", + " inflating: ./data/segmented-images/images/cju3tp94kfstl08181awh6z49.jpg \r\n", + " inflating: ./data/segmented-images/images/cju85l4yjlops0801fvmnwptf.jpg \r\n", + " inflating: ./data/segmented-images/images/cju8cbsyssiqj0871gr4jedjp.jpg \r\n", + " inflating: ./data/segmented-images/images/cju6yywx1whbb0871ksgfgf9f.jpg \r\n", + " inflating: ./data/segmented-images/images/cju1dg44i4z3w0801nyz4p6zf.jpg \r\n", + " inflating: ./data/segmented-images/images/cju31rb7vb6110801p9rhacuw.jpg \r\n", + " inflating: ./data/segmented-images/images/cju2zi4l09f5807991s8do2b2.jpg \r\n", + " inflating: ./data/segmented-images/images/cju1cj3f0qi5n0993ut8f49rj.jpg \r\n", + " inflating: ./data/segmented-images/images/cju8c1a0ws7o208181c6lbsom.jpg \r\n", + " inflating: ./data/segmented-images/images/cju30525w04r10835ygp257sb.jpg \r\n", + " inflating: ./data/segmented-images/images/cju7dlk532dsh0871zvr6qz0r.jpg \r\n", + " inflating: ./data/segmented-images/images/cju5vgawslbe30987ndeepc1b.jpg \r\n", + " inflating: ./data/segmented-images/images/cju6x97w4vwua0850x0997r0a.jpg \r\n", + " inflating: ./data/segmented-images/images/cju35mdz73x890835eynq1h9v.jpg \r\n", + " inflating: ./data/segmented-images/images/cju8bqxxurs6i0850mu7mtef9.jpg \r\n", + " inflating: ./data/segmented-images/images/cju7ea4om2l910801bohqjccy.jpg \r\n", + " inflating: ./data/segmented-images/images/cju8at3s1qqqx0850hcq8nmnq.jpg \r\n", + " inflating: ./data/segmented-images/images/cju6v3bb2v7xo085090blubyw.jpg \r\n", + " inflating: ./data/segmented-images/images/cju885ikhnmkn09878s2lqtuh.jpg \r\n", + " inflating: ./data/segmented-images/images/cju7b10ce1mnm08011c5bwyr4.jpg \r\n", + " inflating: ./data/segmented-images/images/cju88rl5eo94l0850kf5wtrm1.jpg \r\n", + " inflating: ./data/segmented-images/images/cju32srle1xfq083575i3fl75.jpg \r\n", + " inflating: ./data/segmented-images/images/cju34o6dbd2lo0855aqlcy1hs.jpg \r\n", + " inflating: ./data/segmented-images/images/cju1gi7jlwyld0835cdf6g6qz.jpg \r\n", + " inflating: ./data/segmented-images/images/cju3xwpgviwlx0871rwm15q7v.jpg \r\n", + " inflating: ./data/segmented-images/images/cju5knbbqfipk080128cggukq.jpg \r\n", + " inflating: ./data/segmented-images/images/cju8czvnztbf40871b4m7t78w.jpg \r\n", + " inflating: ./data/segmented-images/images/cju5ht88gedbu0755xrcuddcx.jpg \r\n", + " inflating: ./data/segmented-images/images/cju1ffnjn6ctm08015perkg37.jpg \r\n", + " inflating: ./data/segmented-images/images/cju2hgsptlfam0835o3b59h1o.jpg \r\n", + " inflating: ./data/segmented-images/images/cju85c2d4ln1b0755zz1z3onx.jpg \r\n", + " inflating: ./data/segmented-images/images/cju1h89h6xbnx08352k2790o9.jpg \r\n", + " inflating: ./data/segmented-images/images/cju3128yi0rpu0988o4oo5n8n.jpg \r\n", + " inflating: ./data/segmented-images/images/cju426tomlhll0818fc0i7nvh.jpg \r\n", + " inflating: ./data/segmented-images/images/cju3xjqtpikx50817tppy6g84.jpg \r\n", + " inflating: ./data/segmented-images/images/cju8567gdlcbq0801dwwyo2jt.jpg \r\n", + " inflating: ./data/segmented-images/images/cju5von04litr08718j8po40a.jpg \r\n", + " inflating: ./data/segmented-images/images/cju2ma647l0nj0993ot4deq2q.jpg \r\n", + " inflating: ./data/segmented-images/images/cju7arvfe1ldu0850erdmphgj.jpg \r\n", + " inflating: ./data/segmented-images/images/cju2yb31a8e8u0878wdashg7o.jpg \r\n", + " inflating: ./data/segmented-images/images/cju2nsmwjlzyl0993jl80chvz.jpg \r\n", + " inflating: ./data/segmented-images/images/cju88msmoo3470817m441j4sg.jpg \r\n", + " inflating: ./data/segmented-images/images/cju1dq3x1vgx109889c7wyirg.jpg \r\n", + " inflating: ./data/segmented-images/images/cju83qd0yjyht0817ktkfl268.jpg \r\n", + " inflating: ./data/segmented-images/images/cju45ofd9ne1j0801ri8dup7t.jpg \r\n", + " inflating: ./data/segmented-images/images/cju35fxqyzt5p0993vusm54qz.jpg \r\n", + " inflating: ./data/segmented-images/images/cju8b6rp0r5st0850184f79xt.jpg \r\n", + " inflating: ./data/segmented-images/images/cju7d3oc82cho0755dajlwldz.jpg \r\n", + " inflating: ./data/segmented-images/images/cju5cky5xb0ay0801oxet697t.jpg \r\n", + " inflating: ./data/segmented-images/images/cju3y9difj6th0801kd1rqm3w.jpg \r\n", + " inflating: ./data/segmented-images/images/cju6wn57mvooj0850rp78hhy7.jpg \r\n", + " inflating: ./data/segmented-images/images/cju8aw9n1qyg10801jkjlmors.jpg \r\n", + " inflating: ./data/segmented-images/images/cju2hw5gjlr5h0988so2qqres.jpg \r\n", + " inflating: ./data/segmented-images/images/cju428k5fldt108177s6g6f45.jpg \r\n", + " inflating: ./data/segmented-images/images/cju45pm27n80u08174kyow1gj.jpg \r\n", + " inflating: ./data/segmented-images/images/cju5uget8krjy0818kvywd0zu.jpg \r\n", + " inflating: ./data/segmented-images/images/cju2iatlki5u309930zmgkv6h.jpg \r\n", + " inflating: ./data/segmented-images/images/cju2oq5570avm079959o20op1.jpg \r\n", + " inflating: ./data/segmented-images/images/cju171py4qiha0835u8sl59ds.jpg \r\n", + " inflating: ./data/segmented-images/images/cju30nyxe0gfb0835p256yoju.jpg \r\n", + " inflating: ./data/segmented-images/images/cju83rcnzkbsj0755x5anfrcg.jpg \r\n", + " inflating: ./data/segmented-images/images/cju3v0fl3gwce0755qkjhzmd4.jpg \r\n", + " inflating: ./data/segmented-images/images/cju2y8s56ymqr083541ggdsml.jpg \r\n", + " inflating: ./data/segmented-images/images/cju18gzrq18zw0878wbf4ftw6.jpg \r\n", + " inflating: ./data/segmented-images/images/cju8a84g0q76m0818hwiggkod.jpg \r\n", + " inflating: ./data/segmented-images/images/cju8dk7eztzup08182yxko5zh.jpg \r\n", + " inflating: ./data/segmented-images/images/cju2zjcvj9qma0801dk71hhi0.jpg \r\n", + " inflating: ./data/segmented-images/images/cju8b7aqtr4a00987coba14b7.jpg \r\n", + " inflating: ./data/segmented-images/images/cju2hjrqcvi2j0801bx1i6gxg.jpg \r\n", + " inflating: ./data/segmented-images/images/cju17hw9hr9c5098800fu4u8e.jpg \r\n", + " inflating: ./data/segmented-images/images/cju3x2s11ibzi0817kk284k0j.jpg \r\n", + " inflating: ./data/segmented-images/images/cju5vutu7ll8w0871dfp92n9p.jpg \r\n", + " inflating: ./data/segmented-images/images/cju2zxja9w1eh09933609ho9z.jpg \r\n", + " inflating: ./data/segmented-images/images/cju5xneamme2p0801qdf7fdwv.jpg \r\n", + " inflating: ./data/segmented-images/images/cju358pwtdby20878cg7nm0np.jpg \r\n", + " inflating: ./data/segmented-images/images/cju1hs0za7jha0855vj0mdrjt.jpg \r\n", + " inflating: ./data/segmented-images/images/cju2syxa93yw40799x2iuwabz.jpg \r\n", + " inflating: ./data/segmented-images/images/cju5ktjwofed70817eg58ef7u.jpg \r\n", + " inflating: ./data/segmented-images/images/cju3ttznuftyf09875t11850w.jpg \r\n", + " inflating: ./data/segmented-images/images/cju7ctvqn25dy08186g442m1r.jpg \r\n", + " inflating: ./data/segmented-images/images/cju5eyfe9cpk90987laa7tsl3.jpg \r\n", + " inflating: ./data/segmented-images/images/cju310f6val1v0855xo8tc3gu.jpg \r\n", + " inflating: ./data/segmented-images/images/cju306x7w05nb0835cunv799x.jpg \r\n", + " inflating: ./data/segmented-images/images/cju8bop5jrsid08716i24fqda.jpg \r\n", + " inflating: ./data/segmented-images/images/cju5wrapcm2290818jsh26ppb.jpg \r\n", + " inflating: ./data/segmented-images/images/cju7dqcwi2dz00850gcmr2ert.jpg \r\n", + " inflating: ./data/segmented-images/images/cju1g4nsb6ngy0799l4ezm8ab.jpg \r\n", + " inflating: ./data/segmented-images/images/cju7aklv31h4309871m29l4e7.jpg \r\n", + " inflating: ./data/segmented-images/images/cju2i3hzclw3o0988rrgh911i.jpg \r\n", + " inflating: ./data/segmented-images/images/cju7er4kc2opa0801anuxc0eb.jpg \r\n", + " inflating: ./data/segmented-images/images/cju2xlcqxy9c60988vjacdznb.jpg \r\n", + " inflating: ./data/segmented-images/images/cju8bljw9rqk20801kr54akrl.jpg \r\n", + " inflating: ./data/segmented-images/images/cju45lbgznahl08180xz1h7u6.jpg \r\n", + " inflating: ./data/segmented-images/images/cju7b4mtw1n9n080186209f3d.jpg \r\n", + " inflating: ./data/segmented-images/images/cju7f9umg2olj0987fj5y285w.jpg \r\n", + " inflating: ./data/segmented-images/images/cju887ftknop008177nnjt46y.jpg \r\n", + " inflating: ./data/segmented-images/images/cju7b9vcs1luz0987ta60j1dy.jpg \r\n", + " inflating: ./data/segmented-images/images/cju8auylgqx0z0871u4o4db7o.jpg \r\n", + " inflating: ./data/segmented-images/images/cju3ya7goj6at0818v2l5ay7f.jpg \r\n", + " inflating: ./data/segmented-images/images/cju7eueum2oqn0850rodmx8zo.jpg \r\n", + " inflating: ./data/segmented-images/images/cju0u82z3cuma0835wlxrnrjv.jpg \r\n", + " inflating: ./data/segmented-images/images/cju2i03ptvkiu0799xbbd4det.jpg \r\n", + " inflating: ./data/segmented-images/images/cju87mrypnb1e0818scv1mxxg.jpg \r\n", + " inflating: ./data/segmented-images/images/cju7dizi82h2i0755doucgnt3.jpg \r\n", + " inflating: ./data/segmented-images/images/cju5waeduln160817w0agirve.jpg \r\n", + " inflating: ./data/segmented-images/images/cju88nroho44508500129f1nh.jpg \r\n", + " inflating: ./data/segmented-images/images/cju3yht87j83m08507yk1u1fg.jpg \r\n", + " inflating: ./data/segmented-images/images/cju1cfhyg48bb0799cl5pr2jh.jpg \r\n", + " inflating: ./data/segmented-images/images/cju8clorgsuwn08714toqb7v6.jpg \r\n", + " inflating: ./data/segmented-images/images/cju16fpvhzypl0799p9phnlx6.jpg \r\n", + " inflating: ./data/segmented-images/images/cju5wuhm1lwm40987vugqn3vv.jpg \r\n", + " inflating: ./data/segmented-images/images/cju7dz5yy2i7z0801ausi7rna.jpg \r\n", + " inflating: ./data/segmented-images/images/cju2xwm1s84l50799i60mq0pu.jpg \r\n", + " inflating: ./data/segmented-images/images/cju88itqbny720987hxizbj5y.jpg \r\n", + " inflating: ./data/segmented-images/images/cju7ev2b12owa08500bpfpwyw.jpg \r\n", + " inflating: ./data/segmented-images/images/cju84ih17kp5l09876bkooocl.jpg \r\n", + " inflating: ./data/segmented-images/images/cju7afqon1ip40850ue2308b6.jpg \r\n", + " inflating: ./data/segmented-images/images/cju14g8o4xui30878gkgbrvqj.jpg \r\n", + " inflating: ./data/segmented-images/images/cju5cu8qkb84x08186jwo8yin.jpg \r\n", + " inflating: ./data/segmented-images/images/cju5tgbzhjllu08174ca41eus.jpg \r\n", + " inflating: ./data/segmented-images/images/cju1f5x1164xv08555654c24r.jpg \r\n", + " inflating: ./data/segmented-images/images/cju0s690hkp960855tjuaqvv0.jpg \r\n", + " inflating: ./data/segmented-images/images/cju2rzpsmtb0f0835jabkbao1.jpg \r\n", + " inflating: ./data/segmented-images/images/cju7dvl5m2n4t0755hlnnjjet.jpg \r\n", + " inflating: ./data/segmented-images/images/cju7fen322ou10817ziqkob4k.jpg \r\n", + " inflating: ./data/segmented-images/images/cju1cnnziug1l0835yh4ropyg.jpg \r\n", + " inflating: ./data/segmented-images/images/cju88oh0po9gq0801nge4tgr1.jpg \r\n", + " inflating: ./data/segmented-images/images/cju8b3ka8r64u0801fh18hk7l.jpg \r\n", + " inflating: ./data/segmented-images/images/cju18849rrsgr0988p90hkygb.jpg \r\n", + " inflating: ./data/segmented-images/images/cju323ypb1fbb0988gx5rzudb.jpg \r\n", + " inflating: ./data/segmented-images/images/cju6wuojavt740818b5qcv3iw.jpg \r\n", + " inflating: ./data/segmented-images/images/cju0tl3uz8blh0993wxvn7ly3.jpg \r\n", + " inflating: ./data/segmented-images/images/cju2zkpdl9h7t0799ix60teqg.jpg \r\n", + " inflating: ./data/segmented-images/images/cju2urqpwvxw70835rvndvtsi.jpg \r\n", + " inflating: ./data/segmented-images/images/cju5ddda9bkkt0850enzwatb1.jpg \r\n", + " inflating: ./data/segmented-images/images/cju2hx006vidl0799igm81vmh.jpg \r\n", + " inflating: ./data/segmented-images/images/cju5nxkujgscq0817l9gss626.jpg \r\n", + " inflating: ./data/segmented-images/images/cju1dnz61vfp40988e78bkjga.jpg \r\n", + " inflating: ./data/segmented-images/images/cju83syhdk6gs0801rf1rekdl.jpg \r\n", + " inflating: ./data/segmented-images/images/cju42tauqlo5p08171l3cuo4b.jpg \r\n", + " inflating: ./data/segmented-images/images/cju7d1tvt25bu08019dvw3uff.jpg \r\n", + " inflating: ./data/segmented-images/images/cju16whaj0e7n0855q7b6cjkm.jpg \r\n", + " inflating: ./data/segmented-images/images/cju2lcyfgkf5809932fn9gucn.jpg \r\n", + " inflating: ./data/segmented-images/images/cju8c2rqzs5t80850d0zky5dy.jpg \r\n", + " inflating: ./data/segmented-images/images/cju3umoh1geet0817cmpef5am.jpg \r\n", + " inflating: ./data/segmented-images/images/cju34xspwzenf0993cyzajv9n.jpg \r\n", + " inflating: ./data/segmented-images/images/cju2sszfq3uye0878sucelzk2.jpg \r\n", + " inflating: ./data/segmented-images/images/cju84dsvaklpx098750hp83x4.jpg \r\n", + " inflating: ./data/segmented-images/images/cju6wjm81vgsc0987enk9n3pr.jpg \r\n", + " inflating: ./data/segmented-images/images/cju8bk8oirjhw0817hgkua2w8.jpg \r\n", + " inflating: ./data/segmented-images/images/cju8cwy02t9eq08185qn12c02.jpg \r\n", + " inflating: ./data/segmented-images/images/cju2s9g11pnra0993gn4eh793.jpg \r\n", + " inflating: ./data/segmented-images/images/cju2lz8vqktne0993fuym6drw.jpg \r\n", + " inflating: ./data/segmented-images/images/cju8bm24yrrdp081829mbo8ic.jpg \r\n" + ] + }, + { + "name": "stdout", + "output_type": "stream", + "text": [ + " inflating: ./data/segmented-images/images/cju8abobpqbir08189u01huru.jpg \r\n", + " inflating: ./data/segmented-images/images/cju8axq24r4an0755yhv9d4ly.jpg \r\n", + " inflating: ./data/segmented-images/images/cju32csyfblyh080170aa3x5p.jpg \r\n", + " inflating: ./data/segmented-images/images/cju76erapykj30871x5eaxh4q.jpg \r\n", + " inflating: ./data/segmented-images/images/cju77vvcwzcm50850lzoykuva.jpg \r\n", + " inflating: ./data/segmented-images/images/cju30df2j09dd08351ayx2t6w.jpg \r\n", + " inflating: ./data/segmented-images/images/cju43c92lm5cj0755lorsorfg.jpg \r\n", + " inflating: ./data/segmented-images/images/cju5vxuc5loxw0818u8xgf45p.jpg \r\n", + " inflating: ./data/segmented-images/images/cju0t4oil7vzk099370nun5h9.jpg \r\n", + " inflating: ./data/segmented-images/images/cju886ryxnsl50801r93jai7q.jpg \r\n", + " inflating: ./data/segmented-images/images/cju5eq8c8ck690850vix98hv3.jpg \r\n", + " inflating: ./data/segmented-images/images/cju5cjh3xattc0817j2vbulzi.jpg \r\n", + " inflating: ./data/segmented-images/images/cju1fb9236a110801yvg0fwju.jpg \r\n", + " inflating: ./data/segmented-images/images/cju3tsh4lfsok0987w6x3a0v1.jpg \r\n", + " inflating: ./data/segmented-images/images/cju2ricdv2iys0878sv1adh0u.jpg \r\n", + " inflating: ./data/segmented-images/images/cju5uzmaol56l0817flxh4w9p.jpg \r\n", + " inflating: ./data/segmented-images/images/cju42m60jlpcm08186kqppzqv.jpg \r\n", + " inflating: ./data/segmented-images/images/cju7el6xv2k520817qxx9wdr5.jpg \r\n", + " inflating: ./data/segmented-images/images/cju2np2k9zi3v079992ypxqkn.jpg \r\n", + " inflating: ./data/segmented-images/images/cju2zy1e49pqk0878t6ncqn12.jpg \r\n", + " inflating: ./data/segmented-images/images/cju2rz4k434s70855wwx3ddtx.jpg \r\n", + " inflating: ./data/segmented-images/images/cju15l5ubz9yh0855b3ivdpse.jpg \r\n", + " inflating: ./data/segmented-images/images/cju5es375cnzy0801nkq35ffs.jpg \r\n", + " inflating: ./data/segmented-images/images/cju85hguellg50818kwu3s8d1.jpg \r\n", + " inflating: ./data/segmented-images/images/cju5ew4h9cqaf0818rrczkmqh.jpg \r\n", + " inflating: ./data/segmented-images/images/cju7dmlgf2ebw0871ieqas5fh.jpg \r\n", + " inflating: ./data/segmented-images/images/cju77j66ez52p08019xygi0co.jpg \r\n", + " inflating: ./data/segmented-images/images/cju13fwthn9mq0835gacxgy01.jpg \r\n", + " inflating: ./data/segmented-images/images/cju8amfdtqi4x09871tygrgqe.jpg \r\n", + " inflating: ./data/segmented-images/images/cju1hirfi7ekp0855q0vgm9qq.jpg \r\n", + " inflating: ./data/segmented-images/images/cju2yv4imv6cz099314jveiib.jpg \r\n", + " inflating: ./data/segmented-images/images/cju35eg0tdmjt085525sb4bua.jpg \r\n", + " inflating: ./data/segmented-images/images/cju7adqyj1jcx08712r1ro5gx.jpg \r\n", + " inflating: ./data/segmented-images/images/cju8chdlqsu620755azjty1tj.jpg \r\n", + " inflating: ./data/segmented-images/images/cju2qn2fzs1vy0988l243cvzy.jpg \r\n", + " inflating: ./data/segmented-images/images/cju7ebe962hr409872ovibahw.jpg \r\n", + " inflating: ./data/segmented-images/images/cju5k503sfa5f0871lx0rpu5y.jpg \r\n", + " inflating: ./data/segmented-images/images/cju5f26ebcuai0818xlwh6116.jpg \r\n", + " inflating: ./data/segmented-images/images/cju2tqfgw4oat0799rn0g5b2z.jpg \r\n", + " inflating: ./data/segmented-images/images/cju2nqapmzvk20801f9us40dx.jpg \r\n", + " inflating: ./data/segmented-images/images/cju7etr3y2p4t0801cdzjj8ab.jpg \r\n", + " inflating: ./data/segmented-images/images/cju77bvg0yv4r0987yh60xmjo.jpg \r\n", + " inflating: ./data/segmented-images/images/cju83vvmik9wa08710yeh7cuk.jpg \r\n", + " inflating: ./data/segmented-images/images/cju5y4hgqmk0i08180rjhbwvp.jpg \r\n", + " inflating: ./data/segmented-images/images/cju17bz250pgd0799u1hqkj5u.jpg \r\n", + " inflating: ./data/segmented-images/images/cju5i39mreass0817au8p22zy.jpg \r\n", + " inflating: ./data/segmented-images/images/cju41r6v2lcww0871ps8k8pf5.jpg \r\n", + " inflating: ./data/segmented-images/images/cju5vi4nxlc530817uoqm2m7a.jpg \r\n", + " inflating: ./data/segmented-images/images/cju5u4pywk81x0817vn9pe14z.jpg \r\n", + " inflating: ./data/segmented-images/images/cju3y79ofj3va0871uqfb1mzo.jpg \r\n", + " inflating: ./data/segmented-images/images/cju87nkyrnb970801q84m47yt.jpg \r\n", + " inflating: ./data/segmented-images/images/cju1drnhbrb9409935wi7vkhg.jpg \r\n", + " inflating: ./data/segmented-images/images/cju5b9oyda4yr0850g9viziyv.jpg \r\n", + " inflating: ./data/segmented-images/images/cju7fob3x301u0755x985pmmq.jpg \r\n", + " inflating: ./data/segmented-images/images/cju2i6acqvo6l0799u20fift8.jpg \r\n", + " inflating: ./data/segmented-images/images/cju7epwj82koz098713apjnzo.jpg \r\n", + " inflating: ./data/segmented-images/images/cju2m56cryvqd0801gtn2yp8t.jpg \r\n", + " inflating: ./data/segmented-images/images/cju32zhbnc1oy0801iyv1ix6p.jpg \r\n", + " inflating: ./data/segmented-images/images/cju3381d8bz3h07991xtl7ra0.jpg \r\n", + " inflating: ./data/segmented-images/images/cju15jr8jz8sb0855ukmkswkz.jpg \r\n", + " inflating: ./data/segmented-images/images/cju3ua8u0g9rg0801uayhdxhu.jpg \r\n", + " inflating: ./data/segmented-images/images/cju17otoe119u0799nqcbl8n1.jpg \r\n", + " inflating: ./data/segmented-images/images/cju84aoa3ktwn0755pfl4gfwd.jpg \r\n", + " inflating: ./data/segmented-images/images/cju5wkonqlrl409877y8zvnub.jpg \r\n", + " inflating: ./data/segmented-images/images/cju8bw697rwg308177tg8huas.jpg \r\n", + " inflating: ./data/segmented-images/images/cju77u1sjz77b0817ft44r3fk.jpg \r\n", + " inflating: ./data/segmented-images/images/cju7efffp2ivf0817etg3jehl.jpg \r\n", + " inflating: ./data/segmented-images/images/cju890guyoiti098753yg6cdu.jpg \r\n", + " inflating: ./data/segmented-images/images/cju18ibp219ub08783i6o98g7.jpg \r\n", + " inflating: ./data/segmented-images/images/cju32l161bi1v07990vm376in.jpg \r\n", + " inflating: ./data/segmented-images/images/cju30ftgja7170855xl9bkdm0.jpg \r\n", + " inflating: ./data/segmented-images/images/cju1egx9pvz2n0988eoy8jp23.jpg \r\n", + " inflating: ./data/segmented-images/images/cju5u6wf0kh1t0755bg1ssixv.jpg \r\n", + " inflating: ./data/segmented-images/images/cju2rlqdnoz9k0993cpjae3x0.jpg \r\n", + " inflating: ./data/segmented-images/images/cju1cbokpuiw70988j4lq1fpi.jpg \r\n", + " inflating: ./data/segmented-images/images/cju18kevfrojc0835bn90f1in.jpg \r\n", + " inflating: ./data/segmented-images/images/cju5yhgznmkzb0801cji2vi8j.jpg \r\n", + " inflating: ./data/segmented-images/images/cju3y54kwj3nr0801biidlb4e.jpg \r\n", + " inflating: ./data/segmented-images/images/cju5chrxxawka0871qcj171yz.jpg \r\n", + " inflating: ./data/segmented-images/images/cju76o55nymqd0871h31sph9w.jpg \r\n", + " inflating: ./data/segmented-images/images/cju3518w2d838079939fqztbc.jpg \r\n", + " inflating: ./data/segmented-images/images/cju7deifq2fzn0755lc8idyh8.jpg \r\n", + " inflating: ./data/segmented-images/images/cju15wdt3zla10801odjiw7sy.jpg \r\n", + " inflating: ./data/segmented-images/images/cju5hjxaae3i40850h5z2laf5.jpg \r\n", + " inflating: ./data/segmented-images/images/cju2wx0gh7fpz0878wwyd9ep8.jpg \r\n", + " inflating: ./data/segmented-images/images/cju7es23b2vcp0755gpbm9s7v.jpg \r\n", + " inflating: ./data/segmented-images/images/cju8dm2cau2km0818jsv9eeq2.jpg \r\n", + " inflating: ./data/segmented-images/images/cju2y26c588bo07993ksd8eoz.jpg \r\n", + " inflating: ./data/segmented-images/images/cju2suk42469908015ngmq6f2.jpg \r\n", + " inflating: ./data/segmented-images/images/cju1ddr6p4k5z08780uuuzit2.jpg \r\n", + " inflating: ./data/segmented-images/images/cju5vcmrqla7i0817x4sp4pqw.jpg \r\n", + " inflating: ./data/segmented-images/images/cju1gv7106qd008784gk603mg.jpg \r\n", + " inflating: ./data/segmented-images/images/cju6vucxvvlda0755j7msqnya.jpg \r\n", + " inflating: ./data/segmented-images/images/cju7ekbo32pft0871fv7kzwb9.jpg \r\n", + " inflating: ./data/segmented-images/images/cju30ia8da2bq0799klnehml2.jpg \r\n", + " inflating: ./data/segmented-images/images/cju2ntxtdzlvu0799xl3j9pan.jpg \r\n", + " inflating: ./data/segmented-images/images/cju2otvvv0l7z0855x7we8cb0.jpg \r\n", + " inflating: ./data/segmented-images/images/cju8bgdmqrksy0801tozdmraa.jpg \r\n", + " inflating: ./data/segmented-images/images/cju30ajhw09sx0988qyahx9s8.jpg \r\n", + " inflating: ./data/segmented-images/images/cju8c3xs7sauj0801ieyzezr5.jpg \r\n", + " inflating: ./data/segmented-images/images/cju84hibuktj80871u519o71q.jpg \r\n", + " inflating: ./data/segmented-images/images/cju33231uy4gi0993qc7b1jch.jpg \r\n", + " inflating: ./data/segmented-images/images/ck2bxw18mmz1k0725litqq2mc.jpg \r\n", + " inflating: ./data/segmented-images/images/cju88q6h6obpd0871ckmiabbo.jpg \r\n", + " inflating: ./data/segmented-images/images/cju2wtwj87kys0855kx6mddzw.jpg \r\n", + " inflating: ./data/segmented-images/images/cju7ajnbo1gvm098749rdouk0.jpg \r\n", + " inflating: ./data/segmented-images/images/cju8a3nhbpwnb0850d37fo2na.jpg \r\n", + " inflating: ./data/segmented-images/images/cju6vrs1ov8cr098788h8gs6j.jpg \r\n", + " inflating: ./data/segmented-images/images/cju7ez7r22qbc08015xfoz2wb.jpg \r\n", + " inflating: ./data/segmented-images/images/cju3u4lxmg59o0755rz42b9en.jpg \r\n", + " inflating: ./data/segmented-images/images/cju1fmsyf6gxb0801cimx2gle.jpg \r\n", + " inflating: ./data/segmented-images/images/cju7ap09p1kz10850ldccjebj.jpg \r\n", + " inflating: ./data/segmented-images/images/cju1ejj7dvqfa0835ra184v5m.jpg \r\n", + " inflating: ./data/segmented-images/images/cju2wve9v7esz0878mxsdcy04.jpg \r\n", + " inflating: ./data/segmented-images/images/cju2ulk385h170799rlklxob0.jpg \r\n", + " inflating: ./data/segmented-images/images/cju7avvi51iox0817ym55y6tt.jpg \r\n", + " inflating: ./data/segmented-images/images/cju7d4jk723eu0817bqz2n39m.jpg \r\n", + " inflating: ./data/segmented-images/images/cju8brv16rx7f0818uf5n89pv.jpg \r\n", + " inflating: ./data/segmented-images/images/cju2hqt33lmra0988fr5ijv8j.jpg \r\n", + " inflating: ./data/segmented-images/images/cju2upu4evw7g08358guwozxv.jpg \r\n", + " inflating: ./data/segmented-images/images/cju7b3f5h1sm40755i572jden.jpg \r\n", + " inflating: ./data/segmented-images/images/cju6x0yqbvxqt0755dhxislgb.jpg \r\n", + " inflating: ./data/segmented-images/images/cju0ue769mxii08019zqgdbxn.jpg \r\n", + " inflating: ./data/segmented-images/images/cju5clr68b48r0755cmuvponm.jpg \r\n", + " inflating: ./data/segmented-images/images/cju85omszllp30850b6rm9mi3.jpg \r\n", + " inflating: ./data/segmented-images/images/cju2sevf53lkx08558h5bpaig.jpg \r\n", + " inflating: ./data/segmented-images/images/cju2yg5ht8i4p087800js8hp4.jpg \r\n", + " inflating: ./data/segmented-images/images/cju5gucasds9d0801019axylx.jpg \r\n", + " inflating: ./data/segmented-images/images/cju87ox0kncom0801b98hqnd2.jpg \r\n", + " inflating: ./data/segmented-images/images/cju7ecl9i2i060987xawjp4l0.jpg \r\n", + " inflating: ./data/segmented-images/images/cju5nyu31gv8e0871zpk74a2n.jpg \r\n", + " inflating: ./data/segmented-images/images/cju2qs32r1vys07999conmbvx.jpg \r\n", + " inflating: ./data/segmented-images/images/cju7dp3dw2k4n0755zhe003ad.jpg \r\n", + " inflating: ./data/segmented-images/images/cju2zm0axztpe0988r8s9twjr.jpg \r\n", + " inflating: ./data/segmented-images/images/cju1c4fcu40hl07992b8gj0c8.jpg \r\n", + " inflating: ./data/segmented-images/images/cju33za6l2qy70988jhrlp2ev.jpg \r\n", + " inflating: ./data/segmented-images/images/cju2ouil2mssu0993hvxsed6d.jpg \r\n", + " inflating: ./data/segmented-images/images/cju7bc95p1mdm0817yqj5jc6j.jpg \r\n", + " inflating: ./data/segmented-images/images/cju1f15k3w4ct0835cmde6ypo.jpg \r\n", + " inflating: ./data/segmented-images/images/cju40m0rjkpw80871z6n6yg1u.jpg \r\n", + " inflating: ./data/segmented-images/images/cju6vgdmivcvb08018fra5lnv.jpg \r\n", + " inflating: ./data/segmented-images/images/cju1hmff8tkp809931jps6fbr.jpg \r\n", + " inflating: ./data/segmented-images/images/cju8c5223s8j80850b4kealt4.jpg \r\n", + " inflating: ./data/segmented-images/images/cju439oazm2fu0871ma0vvrft.jpg \r\n", + " inflating: ./data/segmented-images/images/cju1819curo000988pd5xcqme.jpg \r\n", + " inflating: ./data/segmented-images/images/cju5fydrud94708507vo6oy21.jpg \r\n", + " inflating: ./data/segmented-images/images/cju2r11x7sdgx0988o8ule0wl.jpg \r\n", + " inflating: ./data/segmented-images/images/cju5wqonpm0e60801z88ewmy1.jpg \r\n", + " inflating: ./data/segmented-images/images/cju76lsehyia10987u54vn8rb.jpg \r\n", + " inflating: ./data/segmented-images/images/cju85ia1slh220987y7c20sm2.jpg \r\n", + " inflating: ./data/segmented-images/images/cju847pxykriq0755268ktrk2.jpg \r\n", + " inflating: ./data/segmented-images/images/cju33jon3ygbj0993pu22a4k6.jpg \r\n", + " inflating: ./data/segmented-images/images/cju3wstckialg0871xs0vevsj.jpg \r\n", + " inflating: ./data/segmented-images/images/cju5fi0yxd3ei0801v7u0yudn.jpg \r\n", + " inflating: ./data/segmented-images/images/cju2s2527pfyr0993l3h1149a.jpg \r\n", + " inflating: ./data/segmented-images/images/cju2tpfa5uyx408359datxqqj.jpg \r\n", + " inflating: ./data/segmented-images/images/cju2xbk0080y80801eghyddi2.jpg \r\n", + " inflating: ./data/segmented-images/images/cju15czxqp3lv0835jvhgzurz.jpg \r\n", + " inflating: ./data/segmented-images/images/cju30j1rgadut0801vuyrsnt8.jpg \r\n", + " inflating: ./data/segmented-images/images/cju77idwfz36d0871tzfzz51i.jpg \r\n", + " inflating: ./data/segmented-images/images/cju14hjh2ob2o0835ouz3r5aa.jpg \r\n", + " inflating: ./data/segmented-images/images/cju5hyi9yegob0755ho3do8en.jpg \r\n", + " inflating: ./data/segmented-images/images/cju3280wv1ir009882jze27tc.jpg \r\n", + " inflating: ./data/segmented-images/images/cju83k8fyjsxr0817d6nxs6r4.jpg \r\n", + " inflating: ./data/segmented-images/images/cju5hqz50e7o90850e0prlpa0.jpg \r\n", + " inflating: ./data/segmented-images/images/cju6uzxk0v83p0801rcwnexdu.jpg \r\n", + " inflating: ./data/segmented-images/images/cju0s2a9ekvms080138tjjpxr.jpg \r\n", + " inflating: ./data/segmented-images/images/cju42wamblrqn098798r2yyok.jpg \r\n", + " inflating: ./data/segmented-images/images/cju34m7h536wq0988xz7gx79v.jpg \r\n", + " inflating: ./data/segmented-images/images/cju17g6ykn1cs0993dww6qdi8.jpg \r\n", + " inflating: ./data/segmented-images/images/cju2s16zp317h0799gr67jqc2.jpg \r\n", + " inflating: ./data/segmented-images/images/cju2wxv0hxs2f09884w48v8fi.jpg \r\n", + " inflating: ./data/segmented-images/images/cju7dubap2g0w0801fgl42mg9.jpg \r\n", + " inflating: ./data/segmented-images/images/cju7do8c72dbo0801vxfzxdc4.jpg \r\n", + " inflating: ./data/segmented-images/images/cju2t2ivz43i10878zeg8r1br.jpg \r\n", + " inflating: ./data/segmented-images/images/cju5u8gz4kj5b07552e2wpkwp.jpg \r\n", + " inflating: ./data/segmented-images/images/cju302fqq9spc0878rrygyzzz.jpg \r\n", + " inflating: ./data/segmented-images/images/cju30k2z40ds308353kdew70n.jpg \r\n", + " inflating: ./data/segmented-images/images/cju1expq45zst0855rjqwwj4m.jpg \r\n", + " inflating: ./data/segmented-images/images/cju43lcnum9y10755bjs7z87f.jpg \r\n", + " inflating: ./data/segmented-images/images/cju43jcqim2cp08172dvjvyui.jpg \r\n", + " inflating: ./data/segmented-images/images/cju3xtufwiv9c0818djsc4cqd.jpg \r\n", + " inflating: ./data/segmented-images/images/cju2qqn5ys4uo0988ewrt2ip2.jpg \r\n", + " inflating: ./data/segmented-images/images/cju422cm8lfxn0818ojicxejb.jpg \r\n", + " inflating: ./data/segmented-images/images/cju2ro5jqsy680988pi6qsujw.jpg \r\n", + " inflating: ./data/segmented-images/images/cju84gpefknwm098714oq8q61.jpg \r\n", + " inflating: ./data/segmented-images/images/cju1b3zgj3d8e0801kpolea6c.jpg \r\n", + " inflating: ./data/segmented-images/images/cju1erep75us208553i4ofwwe.jpg \r\n", + " inflating: ./data/segmented-images/images/cju77196iyshb0850ycbto50a.jpg \r\n", + " inflating: ./data/segmented-images/images/cju1bm8063nmh07996rsjjemq.jpg \r\n", + " inflating: ./data/segmented-images/images/cju772304yw5t0818vbw8kkjf.jpg \r\n", + " inflating: ./data/segmented-images/images/cju2rqo702wpx0855fn7d5cxh.jpg \r\n", + " inflating: ./data/segmented-images/images/cju3v56bwgy8v0871w14pz8fx.jpg \r\n", + " inflating: ./data/segmented-images/images/cju2xs6na81t20878pt6nkfip.jpg \r\n", + " inflating: ./data/segmented-images/images/cju2osuru0ki00855txo0n3uu.jpg \r\n", + " inflating: ./data/segmented-images/images/cju41p90plcsx08018cnzpndc.jpg \r\n", + " inflating: ./data/segmented-images/images/cju30xqmh0ni00835ix3batv1.jpg \r\n", + " inflating: ./data/segmented-images/images/cju2nguelpmlj0835rojdn097.jpg \r\n", + " inflating: ./data/segmented-images/images/cju2srvy5440s0801y1ba9akr.jpg \r\n", + " inflating: ./data/segmented-images/images/cju7dbppn28nx085097654msi.jpg \r\n", + " inflating: ./data/segmented-images/images/cju5c5xc7algd0817pb1ej5yo.jpg \r\n", + " inflating: ./data/segmented-images/images/cju1efbr0rqxz09931z0lf4vf.jpg \r\n", + " inflating: ./data/segmented-images/images/cju8b4ja9r2s808509d45ma86.jpg \r\n", + " inflating: ./data/segmented-images/images/cju33x0f22peh0988g0ln7w5v.jpg \r\n", + " inflating: ./data/segmented-images/images/cju2z1nxlzaj40835wj81s1iy.jpg \r\n", + " inflating: ./data/segmented-images/images/cju7fnfv02tt90801djnix9m8.jpg \r\n", + " inflating: ./data/segmented-images/images/cju2hugv9vget0799hhk7ksvg.jpg \r\n", + " inflating: ./data/segmented-images/images/cju414lf2l1lt0801rl3hjllj.jpg \r\n", + " inflating: ./data/segmented-images/images/cju8b1v3br45u087189kku66u.jpg \r\n", + " inflating: ./data/segmented-images/images/cju7ff97z2ow40817u2r83my5.jpg \r\n", + " inflating: ./data/segmented-images/images/cju8cdeazsm8h0801jxifmzur.jpg \r\n", + " inflating: ./data/segmented-images/images/cju2yo1j1v0qz09934o0e683p.jpg \r\n" + ] + }, + { + "name": "stdout", + "output_type": "stream", + "text": [ + " inflating: ./data/segmented-images/images/cju14pxbaoksp0835qzorx6g6.jpg \r\n", + " inflating: ./data/segmented-images/images/cju5bj926aiec07559rshy4wa.jpg \r\n", + " inflating: ./data/segmented-images/images/cju2yi9tz8vky0801yqip0xyl.jpg \r\n", + " inflating: ./data/segmented-images/images/cju7fazv92ywx0755xov2erga.jpg \r\n", + " inflating: ./data/segmented-images/images/cju8aqq8uqmoq0987hphto9gg.jpg \r\n", + " inflating: ./data/segmented-images/images/cju40jl7skiuo0817p0smlgg8.jpg \r\n", + " inflating: ./data/segmented-images/images/cju2igw4gvxds0878808qj398.jpg \r\n", + " inflating: ./data/segmented-images/images/cju5ekty5ckzf07550c9u3ckk.jpg \r\n", + " inflating: ./data/segmented-images/images/cju5wcc90lu020850mjrxppv6.jpg \r\n", + " inflating: ./data/segmented-images/images/cju7druhp2gp308715i6km7be.jpg \r\n", + " inflating: ./data/segmented-images/images/cju8914beokbf0850isxpocrk.jpg \r\n", + " inflating: ./data/segmented-images/images/cju7ehljc2or70871261br8ai.jpg \r\n", + " inflating: ./data/segmented-images/images/cju83yddek68q0850d2x7zfkm.jpg \r\n", + " inflating: ./data/segmented-images/images/cju2uy8ox62jo0801g88hh42z.jpg \r\n", + " inflating: ./data/segmented-images/images/cju5udcufki0s09874ll1dbr5.jpg \r\n", + " inflating: ./data/segmented-images/images/cju83wwn1k55e0850kw6i2d81.jpg \r\n", + " inflating: ./data/segmented-images/images/cju2uzabhs6er0993x3aaf87p.jpg \r\n", + " inflating: ./data/segmented-images/images/cju300m3s04fg0988uzupuf7z.jpg \r\n", + " inflating: ./data/segmented-images/images/cju1f8w0t65en0799m9oacq0q.jpg \r\n", + " inflating: ./data/segmented-images/images/cju1dia8hvc6v098827mgffnm.jpg \r\n", + " inflating: ./data/segmented-images/images/cju8c82iosagu0817l74s4m5g.jpg \r\n", + " inflating: ./data/segmented-images/images/cju2y40d8ulqo0993q0adtgtb.jpg \r\n", + " inflating: ./data/segmented-images/images/cju2pjb9v0ywn0878j5g5n69j.jpg \r\n", + " inflating: ./data/segmented-images/images/cju7dwe282dc309876rco45ts.jpg \r\n", + " inflating: ./data/segmented-images/images/cju2rga4psq9n09881z519xx0.jpg \r\n", + " inflating: ./data/segmented-images/images/cju2wzu8wxtgu09880ku9x1pg.jpg \r\n", + " inflating: ./data/segmented-images/images/cju3tx1qyg0c907552fglumhc.jpg \r\n", + " inflating: ./data/segmented-images/images/cju7d8m3b2e210755l8fj1yph.jpg \r\n", + " inflating: ./data/segmented-images/images/cju34uhepd3dd0799hs8782ad.jpg \r\n", + " inflating: ./data/segmented-images/images/cju7dxffn2eam0817qxosfwch.jpg \r\n", + " inflating: ./data/segmented-images/images/cju8b0jr0r2oi0801jiquetd5.jpg \r\n", + " inflating: ./data/segmented-images/images/cju34aozyyy830993bn16u32n.jpg \r\n", + " inflating: ./data/segmented-images/images/cju5bwhapakm90987c1v4z46a.jpg \r\n", + " inflating: ./data/segmented-images/images/cju40sdwukv3k0755y99ug1k8.jpg \r\n", + " inflating: ./data/segmented-images/images/cju7emdni2py40871ivhxjtut.jpg \r\n", + " inflating: ./data/segmented-images/images/cju1aqqv02qwz0878a5cyhr67.jpg \r\n", + " inflating: ./data/segmented-images/images/cju85a8h8llwm07559wxg4t5w.jpg \r\n", + " inflating: ./data/segmented-images/images/cju334jzo261t0835yqudnfs1.jpg \r\n", + " inflating: ./data/segmented-images/images/cju5cetivauok0987ok3e5bre.jpg \r\n", + " inflating: ./data/segmented-images/images/cju8dpa89u6l80818dj6lldh9.jpg \r\n", + " inflating: ./data/segmented-images/images/cju88y1mwoln50871emyfny1g.jpg \r\n", + " inflating: ./data/segmented-images/images/cju7bduyq1rjf08719giru9ho.jpg \r\n", + " inflating: ./data/segmented-images/images/cju2t62nq45jl0799odpufwx6.jpg \r\n", + " inflating: ./data/segmented-images/images/cju410dnfl0960755y8lu8d79.jpg \r\n", + " inflating: ./data/segmented-images/images/cju8alhigqn2h0801zksudldd.jpg \r\n", + " inflating: ./data/segmented-images/images/cju884985nlmx0817vzpax3y4.jpg \r\n", + " inflating: ./data/segmented-images/images/cju32gzs6xo8x0993r8tedbpb.jpg \r\n", + " inflating: ./data/segmented-images/images/cju6z7e4bwgdd0987ogkzq9kt.jpg \r\n", + " inflating: ./data/segmented-images/images/cju7d9seq29zd0871nzl2uu5m.jpg \r\n", + " inflating: ./data/segmented-images/images/cju2hos57llxm08359g92p6jj.jpg \r\n", + " inflating: ./data/segmented-images/images/cju5uhrdwkmsu0817ervv91l8.jpg \r\n", + " inflating: ./data/segmented-images/images/cju85fc11ljr40818edpb0inh.jpg \r\n", + " inflating: ./data/segmented-images/images/cju5xjn5mm78b09871spyqhhr.jpg \r\n", + " inflating: ./data/segmented-images/images/cju2rpa30t07b0835im0erql0.jpg \r\n", + " inflating: ./data/segmented-images/images/cju7dtb1e2j0t0818deq51ib3.jpg \r\n", + " inflating: ./data/segmented-images/images/cju8bzzy2s66m08016z6mouqt.jpg \r\n", + " inflating: ./data/segmented-images/images/cju7bfx651qr80801cs7epotb.jpg \r\n", + " inflating: ./data/segmented-images/images/cju0u2g7pmnux0801vkk47ivj.jpg \r\n", + " inflating: ./data/segmented-images/images/cju5h57xedz5h0755mjpc8694.jpg \r\n", + " inflating: ./data/segmented-images/images/cju1d31sp4d4k0878r3fr02ul.jpg \r\n", + " inflating: ./data/segmented-images/images/cju87zv8lni0o0850hbbecbq6.jpg \r\n", + " inflating: ./data/segmented-images/images/cju5ukkg6kv7u08011x2b6zl5.jpg \r\n", + " inflating: ./data/segmented-images/images/cju5k7r0yf98c09878csbxb4d.jpg \r\n", + " inflating: ./data/segmented-images/images/cju32phw2bv130801yj7bkouq.jpg \r\n", + " inflating: ./data/segmented-images/images/cju77t0razbvm080106o56289.jpg \r\n", + " inflating: ./data/segmented-images/images/cju83nwu1jxte0987h1krpfmv.jpg \r\n", + " inflating: ./data/segmented-images/images/cju7aez2x1jtj0871ztezs3oi.jpg \r\n", + " inflating: ./data/segmented-images/images/cju88z8bson4h0871nnd7fdxo.jpg \r\n", + " inflating: ./data/segmented-images/images/cju5i5oh2efg60987ez6cpf72.jpg \r\n", + " inflating: ./data/segmented-images/images/cju8doa16u5gh0818w1ywda3q.jpg \r\n", + " inflating: ./data/segmented-images/images/cju5fs6j6d8350801vglraq4u.jpg \r\n", + " inflating: ./data/segmented-images/images/cju40w3hbkwpn08015rbs3wko.jpg \r\n", + " inflating: ./data/segmented-images/images/cju83ipu3jwpx0801z5pvguf8.jpg \r\n", + " inflating: ./data/segmented-images/images/cju2qfie4rvz508357kad9z5o.jpg \r\n", + " inflating: ./data/segmented-images/images/cju1c3218411b08014g9f6gig.jpg \r\n", + " inflating: ./data/segmented-images/images/cju6vifjlv55z0987un6y4zdo.jpg \r\n", + " inflating: ./data/segmented-images/images/cju430pm2lz0y0755jkhcc3d1.jpg \r\n", + " inflating: ./data/segmented-images/images/cju5ymyd8mmdc0801ry3by1xr.jpg \r\n", + " inflating: ./data/segmented-images/images/cju5ygh1zmmdi0755uod5e17i.jpg \r\n", + " inflating: ./data/segmented-images/images/cju89y9h0puti0818i5yw29e6.jpg \r\n", + " inflating: ./data/segmented-images/images/cju5ca9hcatkc0801jzwe7tfx.jpg \r\n", + " inflating: ./data/segmented-images/images/cju7ejm2l2ncl0801wq6y84nw.jpg \r\n", + " inflating: ./data/segmented-images/images/cju34repocy5208780gswillm.jpg \r\n", + " inflating: ./data/segmented-images/images/cju8ayeq7r1fb0818z1junacy.jpg \r\n", + " inflating: ./data/segmented-images/images/cju1alwgo30z60855fm3y23sm.jpg \r\n", + " inflating: ./data/segmented-images/images/cju7eotqi2qea0871y8yc7tqh.jpg \r\n", + " inflating: ./data/segmented-images/images/cju2r6mt2om21099352pny5gw.jpg \r\n", + " inflating: ./data/segmented-images/images/cju857ad0l88m0817qx4cwxnf.jpg \r\n", + " inflating: ./data/segmented-images/images/cju5tenjojp1j0755ms4949h2.jpg \r\n", + " inflating: ./data/segmented-images/images/cju6z9a9kwsl007552s49rx6i.jpg \r\n", + " inflating: ./data/segmented-images/images/cju34i3qvcyog0855qiejxx5w.jpg \r\n", + " inflating: ./data/segmented-images/images/cju2p4ddkmzxj0993p94o62av.jpg \r\n", + " inflating: ./data/segmented-images/images/cju7f900s2o0k08175gl1giid.jpg \r\n", + " inflating: ./data/segmented-images/images/cju5yeqiwmkgl0801fzv2douc.jpg \r\n", + " inflating: ./data/segmented-images/images/cju175facms5f0993a5tjikvt.jpg \r\n", + " inflating: ./data/segmented-images/images/cju30mm25a53s0799qa5wiqe8.jpg \r\n", + " inflating: ./data/segmented-images/images/cju353d1eda8c07992afde611.jpg \r\n", + " inflating: ./data/segmented-images/images/cju2z45kuzf6d0988nz2c819m.jpg \r\n", + " inflating: ./data/segmented-images/images/cju33belnbyhm0878yxl42233.jpg \r\n", + " inflating: ./data/segmented-images/images/cju7ddtz729960801uazp1knc.jpg \r\n", + " inflating: ./data/segmented-images/images/cju3xeexgii1j0817zs68tb4g.jpg \r\n", + " inflating: ./data/segmented-images/images/cju1h5w4wxajx0835mc954kxy.jpg \r\n", + " inflating: ./data/segmented-images/images/cju1gkndf6yi10801o1qnje19.jpg \r\n", + " inflating: ./data/segmented-images/images/cju7dn24o296i09871qfxb8s2.jpg \r\n", + " inflating: ./data/segmented-images/images/cju6x4t13vyw60755gtcf9ndu.jpg \r\n", + " inflating: ./data/segmented-images/images/cju31t8xd17bk0835rnb893jk.jpg \r\n", + " inflating: ./data/segmented-images/images/cju3xuj20ivgp0818mij8bjrd.jpg \r\n", + " inflating: ./data/segmented-images/images/cju33mirdc8mj0799k33wzoes.jpg \r\n", + " inflating: ./data/segmented-images/images/cju42romflni20817etb9a0fl.jpg \r\n", + " inflating: ./data/segmented-images/images/cju5yclrymlgj0818k426ud6z.jpg \r\n", + " inflating: ./data/segmented-images/images/cju1fyb1d69et0878muzdak9u.jpg \r\n", + " inflating: ./data/segmented-images/images/cju8402x1kcy70801t6kz6bdi.jpg \r\n", + " inflating: ./data/segmented-images/images/cju84kplnl1y30755ropua1b0.jpg \r\n", + " inflating: ./data/segmented-images/images/cju326h4v1gxw08352px40p7r.jpg \r\n", + " inflating: ./data/segmented-images/images/cju40r6jrksyk0871wg98zgho.jpg \r\n", + " inflating: ./data/segmented-images/images/cju77afzlz3kp07550x5nafzs.jpg \r\n", + " inflating: ./data/segmented-images/images/cju0qx73cjw570799j4n5cjze.jpg \r\n", + " inflating: ./data/segmented-images/images/cju1ats0y372e08011yazcsxm.jpg \r\n", + " inflating: ./data/segmented-images/images/cju8ca4geseia0850i2ru11hw.jpg \r\n", + " inflating: ./data/segmented-images/images/cju7dglf226g50987ohbthl19.jpg \r\n", + " inflating: ./data/segmented-images/images/cju43kj2pm34f0850l28ahpni.jpg \r\n", + " inflating: ./data/segmented-images/images/cju6yxyt0wh080871sqpepu47.jpg \r\n", + " inflating: ./data/segmented-images/images/cju2p0eveqtdc0835gpi3p93i.jpg \r\n", + " inflating: ./data/segmented-images/images/cju1c8ffau5770835g0g343o8.jpg \r\n", + " inflating: ./data/segmented-images/images/cju2lejzcy4pc0878c9rlonot.jpg \r\n", + " inflating: ./data/segmented-images/images/cju2zgbj9zmrw0835nnlzxj4c.jpg \r\n", + " inflating: ./data/segmented-images/images/cju1d96gsv62d09881b3wecw2.jpg \r\n", + " inflating: ./data/segmented-images/images/cju5kre09fhka0850h7b1898j.jpg \r\n", + " inflating: ./data/segmented-images/images/cju2mh8t6p07008350e01tx2a.jpg \r\n", + " inflating: ./data/segmented-images/images/cju8828oxnool0801qno9luhr.jpg \r\n", + " inflating: ./data/segmented-images/images/cju5jx7jzf7c90871c2i9aiov.jpg \r\n", + " inflating: ./data/segmented-images/images/cju6vta3kvazg0817qbeppjtm.jpg \r\n", + " inflating: ./data/segmented-images/images/cju2xa3i4y0160988i679zsqd.jpg \r\n", + " inflating: ./data/segmented-images/images/cju1f79yhsb5w0993txub59ol.jpg \r\n", + " inflating: ./data/segmented-images/images/cju16ach3m1da0993r1dq3sn2.jpg \r\n", + " inflating: ./data/segmented-images/images/cju7flevb2wii08188otgs9p2.jpg \r\n", + " inflating: ./data/segmented-images/images/cju2rmd2rsw9g09888hh1efu0.jpg \r\n", + " inflating: ./data/segmented-images/images/cju32a52lb9rc0799xi40qs00.jpg \r\n", + " inflating: ./data/segmented-images/images/cju1ewnoh5z030855vpex9uzt.jpg \r\n", + " inflating: ./data/segmented-images/images/cju0qkwl35piu0993l0dewei2.jpg \r\n", + " inflating: ./data/segmented-images/images/cju15ptjtppz40988odsm9azx.jpg \r\n", + " inflating: ./data/segmented-images/images/cju35atpxdjot0855q46aqrd0.jpg \r\n", + " inflating: ./data/segmented-images/images/cju2xyd9vyi7m098831qcucse.jpg \r\n", + " inflating: ./data/segmented-images/images/cju17v6ih0u7808783zcbg1jy.jpg \r\n", + " inflating: ./data/segmented-images/images/cju2qozsk20cq0855ugrg3cri.jpg \r\n", + " inflating: ./data/segmented-images/images/cju3u39fog1bo0871lxjrabks.jpg \r\n", + " inflating: ./data/segmented-images/images/cju8ashhnquqr0801rwduzt7d.jpg \r\n", + " inflating: ./data/segmented-images/images/cju30ywtc0oar0835bp2en7ec.jpg \r\n", + " inflating: ./data/segmented-images/images/cju7f5ghb2r5s0801chwkxxh9.jpg \r\n", + " inflating: ./data/segmented-images/images/cju3ykamdj9u208503pygyuc8.jpg \r\n", + " inflating: ./data/segmented-images/images/cju2p91qir00k08350ddfif0w.jpg \r\n", + " inflating: ./data/segmented-images/images/cju8a1jtvpt9m081712iwkca7.jpg \r\n", + " inflating: ./data/segmented-images/images/cju1hyolc7aqu0878rrkfn1lr.jpg \r\n", + " inflating: ./data/segmented-images/images/cju3uz4o6gr9z0850lhxyxvsj.jpg \r\n", + " inflating: ./data/segmented-images/images/cju5woy82m07m08505dmjg7g1.jpg \r\n", + " inflating: ./data/segmented-images/images/cju2r91dg2k090801bh0xzbxk.jpg \r\n", + " inflating: ./data/segmented-images/images/cju2ueb6j5ado0878vf5md13o.jpg \r\n", + " inflating: ./data/segmented-images/images/cju2uwz9f5yf1085506cfamfx.jpg \r\n", + " inflating: ./data/segmented-images/images/cju1dfeupuzlw0835gnxip369.jpg \r\n", + " inflating: ./data/segmented-images/images/cju0qoxqj9q6s0835b43399p4.jpg \r\n", + " inflating: ./data/segmented-images/images/cju2x7vw87mu30878hye2ca0m.jpg \r\n", + " inflating: ./data/segmented-images/images/cju40poe4kt9s0755f9cnm3h5.jpg \r\n", + " inflating: ./data/segmented-images/images/cju2nd7l7z98o0799gfjvyfmw.jpg \r\n", + " inflating: ./data/segmented-images/images/cju84jdl9kv0i0871eog9b3i9.jpg \r\n" + ] + }, + { + "name": "stdout", + "output_type": "stream", + "text": [ + " inflating: ./data/segmented-images/images/cju8dn0c3u2v50801k8rvq02f.jpg \r\n", + " inflating: ./data/segmented-images/images/cju3x4blzieu30850x10uuvbm.jpg \r\n", + " inflating: ./data/segmented-images/images/cjyzkmjy8evns070165gf9dmq.jpg \r\n", + " inflating: ./data/segmented-images/images/cju40wto5kxwi0755it190f2k.jpg \r\n", + " inflating: ./data/segmented-images/images/cjyztzaqtrv430848l8xgcerw.jpg \r\n", + " inflating: ./data/segmented-images/images/cjyzurzvohqnr0794es1itzek.jpg \r\n", + " inflating: ./data/segmented-images/images/cju2oi8sq0i2y0801mektzvw8.jpg \r\n", + " inflating: ./data/segmented-images/images/cju7eea9b2m0z0801ynqv1fqu.jpg \r\n", + " inflating: ./data/segmented-images/images/cju8dqkrqu83i0818ev74qpxq.jpg \r\n", + " inflating: ./data/segmented-images/images/cjyzkpsbjdsjq07211dfi4sru.jpg \r\n", + " inflating: ./data/segmented-images/images/cjyzk8qieoboa0848ogj51wwm.jpg \r\n", + " inflating: ./data/segmented-images/images/cjyzu3reghjya0794w7pwoi50.jpg \r\n", + " inflating: ./data/segmented-images/images/cjyzufihqquiw0a46jatrbwln.jpg \r\n", + " inflating: ./data/segmented-images/images/cju85je7vlht70817c9jcjwi4.jpg \r\n", + " inflating: ./data/segmented-images/images/cjz14qsk2wci60794un9ozwmw.jpg \r\n", + " inflating: ./data/segmented-images/images/ck2bxlujamu330725szlc2jdu.jpg \r\n", + " inflating: ./data/segmented-images/images/cju0sr5ghl0nd08789uzf1raf.jpg \r\n", + " inflating: ./data/segmented-images/images/cju2yljr0yzhw0988ecf271ly.jpg \r\n", + " inflating: ./data/segmented-images/images/cju2xd75m82720801q4s4ik3n.jpg \r\n", + " inflating: ./data/segmented-images/images/ck2da7fwcjfis07218r1rvm95.jpg \r\n", + " inflating: ./data/segmented-images/images/ck2395w2mb4vu07480otsu6tw.jpg \r\n", + " inflating: ./data/segmented-images/images/cjyzuio1qgh040763k56deohv.jpg \r\n", + " inflating: ./data/segmented-images/images/cjyzlw7f9faqr070129au64sq.jpg \r\n", + " inflating: ./data/segmented-images/images/cjyzjzssvd8pq0838f4nolj5l.jpg \r\n", + " inflating: ./data/segmented-images/images/cjyzul1qggwwj07216mhiv5sy.jpg \r\n", + " inflating: ./data/segmented-images/images/ck2bxqz3evvg20794iiyv5v2m.jpg \r\n", + " inflating: ./data/segmented-images/images/cjyzl833ndne80838pzuq6ila.jpg \r\n", + " inflating: ./data/segmented-images/images/cjyzu9th0qt4r0a46pyl4zik0.jpg \r\n", + " inflating: ./data/segmented-images/license.txt \r\n" + ] + } + ], "source": [ "!wget 'https://datasets.simula.no/hyper-kvasir/hyper-kvasir-segmented-images.zip' -O kvasir.zip\n", "ZIP_SHA384 = 'e30d18a772c6520476e55b610a4db457237f151e'\\\n", @@ -94,7 +2268,7 @@ }, { "cell_type": "code", - "execution_count": null, + "execution_count": 5, "metadata": {}, "outputs": [], "source": [ @@ -103,7 +2277,7 @@ }, { "cell_type": "code", - "execution_count": null, + "execution_count": 6, "metadata": { "code_folding": [], "scrolled": false @@ -175,7 +2349,7 @@ }, { "cell_type": "code", - "execution_count": null, + "execution_count": 7, "metadata": { "code_folding": [] }, @@ -240,7 +2414,7 @@ }, { "cell_type": "code", - "execution_count": null, + "execution_count": 8, "metadata": { "code_folding": [] }, @@ -404,7 +2578,7 @@ }, { "cell_type": "code", - "execution_count": null, + "execution_count": 9, "metadata": {}, "outputs": [], "source": [ @@ -420,9 +2594,23 @@ }, { "cell_type": "code", - "execution_count": null, + "execution_count": 10, "metadata": {}, - "outputs": [], + "outputs": [ + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\">[17:44:12] </span><span style=\"color: #800000\">WARNING</span> tried to remove tensor: __opt_state_needed not present in the tensor dict <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/utilities/utils.py\"><span style=\"color: #7f7f7f\">utils.py</span></a><span style=\"color: #7f7f7f\">:85</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f752a79ed90>" + ] + }, + "metadata": {}, + "output_type": "display_data" + } + ], "source": [ "# Create a federated model using the pytorch class, optimizer function, and loss function\n", "fl_model = FederatedModel(build_model=UNet, optimizer=optimizer,\n", @@ -431,9 +2619,36 @@ }, { "cell_type": "code", - "execution_count": null, + "execution_count": 11, "metadata": {}, - "outputs": [], + "outputs": [ + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #800000\">WARNING</span> tried to remove tensor: __opt_state_needed not present in the tensor dict <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/utilities/utils.py\"><span style=\"color: #7f7f7f\">utils.py</span></a><span style=\"color: #7f7f7f\">:85</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f7529a69820>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\">[17:44:13] </span><span style=\"color: #800000\">WARNING</span> tried to remove tensor: __opt_state_needed not present in the tensor dict <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/utilities/utils.py\"><span style=\"color: #7f7f7f\">utils.py</span></a><span style=\"color: #7f7f7f\">:85</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f75299b8b20>" + ] + }, + "metadata": {}, + "output_type": "display_data" + } + ], "source": [ "collaborator_models = fl_model.setup(num_collaborators=2)\n", "collaborators = {'one': collaborator_models[0], 'two': collaborator_models[1]}" @@ -448,14 +2663,83 @@ }, { "cell_type": "code", - "execution_count": null, + "execution_count": 12, "metadata": { "code_folding": [] }, - "outputs": [], + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "{\n", + " \"aggregator.settings.best_state_path\": \"save/torch_unet_kvasir_best.pbuf\",\n", + " \"aggregator.settings.db_store_rounds\": 1,\n", + " \"aggregator.settings.init_state_path\": \"save/torch_unet_kvasir_init.pbuf\",\n", + " \"aggregator.settings.last_state_path\": \"save/torch_unet_kvasir_last.pbuf\",\n", + " \"aggregator.settings.rounds_to_train\": 40,\n", + " \"aggregator.template\": \"openfl.component.Aggregator\",\n", + " \"assigner.settings.task_groups\": [\n", + " {\n", + " \"name\": \"train_and_validate\",\n", + " \"percentage\": 1.0,\n", + " \"tasks\": [\n", + " \"aggregated_model_validation\",\n", + " \"train\",\n", + " \"locally_tuned_model_validation\"\n", + " ]\n", + " }\n", + " ],\n", + " \"assigner.template\": \"openfl.component.RandomGroupedAssigner\",\n", + " \"collaborator.settings.db_store_rounds\": 1,\n", + " \"collaborator.settings.delta_updates\": false,\n", + " \"collaborator.settings.epochs_per_round\": 1.0,\n", + " \"collaborator.settings.opt_treatment\": \"RESET\",\n", + " \"collaborator.settings.polling_interval\": 4,\n", + " \"collaborator.template\": \"openfl.component.Collaborator\",\n", + " \"data_loader.settings.batch_size\": 4,\n", + " \"data_loader.settings.collaborator_count\": 2,\n", + " \"data_loader.settings.data_group_name\": \"kvasir\",\n", + " \"data_loader.template\": \"code.data_loader.PyTorchKvasirDataLoader\",\n", + " \"network.settings.agg_addr\": \"nnlicv611.inn.intel.com\",\n", + " \"network.settings.agg_port\": 53966,\n", + " \"network.settings.cert_folder\": \"cert\",\n", + " \"network.settings.client_reconnect_interval\": 5,\n", + " \"network.settings.disable_client_auth\": false,\n", + " \"network.settings.disable_tls\": false,\n", + " \"network.settings.hash_salt\": \"auto\",\n", + " \"network.template\": \"openfl.federation.Network\",\n", + " \"task_runner.settings.n_channels\": 3,\n", + " \"task_runner.settings.n_classes\": 1,\n", + " \"task_runner.template\": \"code.fed_unet_runner.PyTorchFederatedUnet\",\n", + " \"tasks.aggregated_model_validation.function\": \"validate\",\n", + " \"tasks.aggregated_model_validation.kwargs\": {\n", + " \"apply\": \"global\",\n", + " \"metrics\": [\n", + " \"dice_coef\"\n", + " ]\n", + " },\n", + " \"tasks.locally_tuned_model_validation.function\": \"validate\",\n", + " \"tasks.locally_tuned_model_validation.kwargs\": {\n", + " \"apply\": \"local\",\n", + " \"metrics\": [\n", + " \"dice_coef\"\n", + " ]\n", + " },\n", + " \"tasks.settings\": {},\n", + " \"tasks.train.function\": \"train_batches\",\n", + " \"tasks.train.kwargs\": {\n", + " \"metrics\": [\n", + " \"loss\"\n", + " ]\n", + " }\n", + "}\n" + ] + } + ], "source": [ "# Get the current values of the FL plan. Each of these can be overridden\n", - "print(json.dumps(fx.get_plan(), indent=4, sort_keys=True))" + "print(fx.get_plan())" ] }, { @@ -467,11 +2751,4254 @@ }, { "cell_type": "code", - "execution_count": null, + "execution_count": 13, "metadata": { "scrolled": true }, - "outputs": [], + "outputs": [ + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> Updating aggregator.settings.rounds_to_train to <span style=\"color: #000080; font-weight: bold\">30</span><span style=\"color: #808000\">...</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/native/native.py\"><span style=\"color: #7f7f7f\">native.py</span></a><span style=\"color: #7f7f7f\">:82</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f75299ec340>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> Building <span style=\"color: #800000\">🡆</span> Object <span style=\"color: #800000\">NoCompressionPipeline</span> from <span style=\"color: #800000\">openfl.pipelines</span> Module. <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/federated/plan/plan.py\"><span style=\"color: #7f7f7f\">plan.py</span></a><span style=\"color: #7f7f7f\">:162</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f75299ec7f0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> Settings <span style=\"color: #800000\">🡆</span> <span style=\"font-weight: bold\">{}</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/federated/plan/plan.py\"><span style=\"color: #7f7f7f\">plan.py</span></a><span style=\"color: #7f7f7f\">:165</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f75299eca60>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> Override <span style=\"color: #800000\">🡆</span> <span style=\"font-weight: bold\">{}</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/federated/plan/plan.py\"><span style=\"color: #7f7f7f\">plan.py</span></a><span style=\"color: #7f7f7f\">:167</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f75299ec5b0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> Creating Initial Weights File 🠆 save/torch_unet_kvasir_init.pbuf <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/native/native.py\"><span style=\"color: #7f7f7f\">native.py</span></a><span style=\"color: #7f7f7f\">:240</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f7490d61c70>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> Starting Experiment<span style=\"color: #808000\">...</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/native/native.py\"><span style=\"color: #7f7f7f\">native.py</span></a><span style=\"color: #7f7f7f\">:244</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f7490d61ca0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> Building <span style=\"color: #800000\">🡆</span> Object <span style=\"color: #800000\">RandomGroupedAssigner</span> from <span style=\"color: #800000\">openfl.component</span> Module. <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/federated/plan/plan.py\"><span style=\"color: #7f7f7f\">plan.py</span></a><span style=\"color: #7f7f7f\">:162</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f7490d61d90>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> Settings <span style=\"color: #800000\">🡆</span> <span style=\"font-weight: bold\">{</span><span style=\"color: #008000\">'task_groups'</span>: <span style=\"font-weight: bold\">[{</span><span style=\"color: #008000\">'name'</span>: <span style=\"color: #008000\">'train_and_validate'</span>, <span style=\"color: #008000\">'percentage'</span>: <span style=\"color: #000080; font-weight: bold\">1.0</span>, <span style=\"color: #008000\">'tasks'</span>: <span style=\"font-weight: bold\">[</span><span style=\"color: #008000\">'aggregated_model_validation'</span>, <span style=\"color: #008000\">'train'</span>, <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/federated/plan/plan.py\"><span style=\"color: #7f7f7f\">plan.py</span></a><span style=\"color: #7f7f7f\">:165</span>\n", + " <span style=\"color: #008000\">'locally_tuned_model_validation'</span><span style=\"font-weight: bold\">]}]</span>, <span style=\"color: #008000\">'authorized_cols'</span>: <span style=\"font-weight: bold\">[</span><span style=\"color: #008000\">'one'</span>, <span style=\"color: #008000\">'two'</span><span style=\"font-weight: bold\">]</span>, <span style=\"color: #008000\">'rounds_to_train'</span>: <span style=\"color: #000080; font-weight: bold\">40</span>, <span style=\"color: #008000\">'tasks'</span>: <span style=\"font-weight: bold\">{</span><span style=\"color: #008000\">'defaults'</span>: \n", + " <span style=\"color: #008000\">'plan/defaults/tasks_torch.yaml'</span>, <span style=\"color: #008000\">'settings'</span>: <span style=\"font-weight: bold\">{}</span>, <span style=\"color: #008000\">'aggregated_model_validation'</span>: <span style=\"font-weight: bold\">{</span><span style=\"color: #008000\">'function'</span>: <span style=\"color: #008000\">'validate'</span>, <span style=\"color: #008000\">'kwargs'</span>: <span style=\"font-weight: bold\">{</span><span style=\"color: #008000\">'apply'</span>: \n", + " <span style=\"color: #008000\">'global'</span>, <span style=\"color: #008000\">'metrics'</span>: <span style=\"font-weight: bold\">[</span><span style=\"color: #008000\">'dice_coef'</span><span style=\"font-weight: bold\">]}}</span>, <span style=\"color: #008000\">'train'</span>: <span style=\"font-weight: bold\">{</span><span style=\"color: #008000\">'function'</span>: <span style=\"color: #008000\">'train_batches'</span>, <span style=\"color: #008000\">'kwargs'</span>: <span style=\"font-weight: bold\">{</span><span style=\"color: #008000\">'metrics'</span>: <span style=\"font-weight: bold\">[</span><span style=\"color: #008000\">'loss'</span><span style=\"font-weight: bold\">]}}</span>, \n", + " <span style=\"color: #008000\">'locally_tuned_model_validation'</span>: <span style=\"font-weight: bold\">{</span><span style=\"color: #008000\">'function'</span>: <span style=\"color: #008000\">'validate'</span>, <span style=\"color: #008000\">'kwargs'</span>: <span style=\"font-weight: bold\">{</span><span style=\"color: #008000\">'apply'</span>: <span style=\"color: #008000\">'local'</span>, <span style=\"color: #008000\">'metrics'</span>: <span style=\"font-weight: bold\">[</span><span style=\"color: #008000\">'dice_coef'</span><span style=\"font-weight: bold\">]}}}}</span> \n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f7490d61dc0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> Override <span style=\"color: #800000\">🡆</span> <span style=\"font-weight: bold\">{</span><span style=\"color: #008000\">'defaults'</span>: <span style=\"color: #008000\">'plan/defaults/assigner.yaml'</span><span style=\"font-weight: bold\">}</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/federated/plan/plan.py\"><span style=\"color: #7f7f7f\">plan.py</span></a><span style=\"color: #7f7f7f\">:167</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f7490d61af0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> Building <span style=\"color: #800000\">🡆</span> Object <span style=\"color: #800000\">Aggregator</span> from <span style=\"color: #800000\">openfl.component</span> Module. <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/federated/plan/plan.py\"><span style=\"color: #7f7f7f\">plan.py</span></a><span style=\"color: #7f7f7f\">:162</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f7490d2ef70>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> Settings <span style=\"color: #800000\">🡆</span> <span style=\"font-weight: bold\">{</span><span style=\"color: #008000\">'last_state_path'</span>: <span style=\"color: #008000\">'save/torch_unet_kvasir_last.pbuf'</span>, <span style=\"color: #008000\">'db_store_rounds'</span>: <span style=\"color: #000080; font-weight: bold\">1</span>, <span style=\"color: #008000\">'init_state_path'</span>: <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/federated/plan/plan.py\"><span style=\"color: #7f7f7f\">plan.py</span></a><span style=\"color: #7f7f7f\">:165</span>\n", + " <span style=\"color: #008000\">'save/torch_unet_kvasir_init.pbuf'</span>, <span style=\"color: #008000\">'rounds_to_train'</span>: <span style=\"color: #000080; font-weight: bold\">30</span>, <span style=\"color: #008000\">'best_state_path'</span>: <span style=\"color: #008000\">'save/torch_unet_kvasir_best.pbuf'</span>, \n", + " <span style=\"color: #008000\">'aggregator_uuid'</span>: <span style=\"color: #008000\">'aggregator_plan.yaml_851debec'</span>, <span style=\"color: #008000\">'federation_uuid'</span>: <span style=\"color: #008000\">'plan.yaml_851debec'</span>, <span style=\"color: #008000\">'authorized_cols'</span>: <span style=\"font-weight: bold\">[</span><span style=\"color: #008000\">'one'</span>, <span style=\"color: #008000\">'two'</span><span style=\"font-weight: bold\">]</span>, \n", + " <span style=\"color: #008000\">'assigner'</span>: <span style=\"font-weight: bold\"><</span><span style=\"color: #ff00ff; font-weight: bold\">openfl.component.assigner.random_grouped_assigner.RandomGroupedAssigner</span><span style=\"color: #000000\"> object at </span><span style=\"color: #000080; font-weight: bold\">0x7f75299ec520</span><span style=\"font-weight: bold\">>}</span> \n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f7490d2efa0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> Override <span style=\"color: #800000\">🡆</span> <span style=\"font-weight: bold\">{</span><span style=\"color: #008000\">'defaults'</span>: <span style=\"color: #008000\">'plan/defaults/aggregator.yaml'</span><span style=\"font-weight: bold\">}</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/federated/plan/plan.py\"><span style=\"color: #7f7f7f\">plan.py</span></a><span style=\"color: #7f7f7f\">:167</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f7490d2ee80>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> Building <span style=\"color: #800000\">🡆</span> Object <span style=\"color: #800000\">Collaborator</span> from <span style=\"color: #800000\">openfl.component</span> Module. <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/federated/plan/plan.py\"><span style=\"color: #7f7f7f\">plan.py</span></a><span style=\"color: #7f7f7f\">:162</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f7490c9eac0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> Settings <span style=\"color: #800000\">🡆</span> <span style=\"font-weight: bold\">{</span><span style=\"color: #008000\">'db_store_rounds'</span>: <span style=\"color: #000080; font-weight: bold\">1</span>, <span style=\"color: #008000\">'epochs_per_round'</span>: <span style=\"color: #000080; font-weight: bold\">1.0</span>, <span style=\"color: #008000\">'delta_updates'</span>: <span style=\"color: #ff0000; font-style: italic\">False</span>, <span style=\"color: #008000\">'polling_interval'</span>: <span style=\"color: #000080; font-weight: bold\">4</span>, <span style=\"color: #008000\">'opt_treatment'</span>: <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/federated/plan/plan.py\"><span style=\"color: #7f7f7f\">plan.py</span></a><span style=\"color: #7f7f7f\">:165</span>\n", + " <span style=\"color: #008000\">'RESET'</span>, <span style=\"color: #008000\">'collaborator_name'</span>: <span style=\"color: #008000\">'one'</span>, <span style=\"color: #008000\">'aggregator_uuid'</span>: <span style=\"color: #008000\">'aggregator_plan.yaml_851debec'</span>, <span style=\"color: #008000\">'federation_uuid'</span>: \n", + " <span style=\"color: #008000\">'plan.yaml_851debec'</span>, <span style=\"color: #008000\">'task_runner'</span>: <span style=\"font-weight: bold\"><</span><span style=\"color: #ff00ff; font-weight: bold\">openfl.federated.task.fl_model.FederatedModel</span><span style=\"color: #000000\"> object at </span><span style=\"color: #000080; font-weight: bold\">0x7f7529a742b0</span><span style=\"font-weight: bold\">></span>, <span style=\"color: #008000\">'tensor_pipe'</span>: \n", + " <span style=\"font-weight: bold\"><</span><span style=\"color: #ff00ff; font-weight: bold\">openfl.pipelines.no_compression_pipeline.NoCompressionPipeline</span><span style=\"color: #000000\"> object at </span><span style=\"color: #000080; font-weight: bold\">0x7f7529a746a0</span><span style=\"font-weight: bold\">></span>, <span style=\"color: #008000\">'task_config'</span>: <span style=\"font-weight: bold\">{</span><span style=\"color: #008000\">'defaults'</span>: \n", + " <span style=\"color: #008000\">'plan/defaults/tasks_torch.yaml'</span>, <span style=\"color: #008000\">'settings'</span>: <span style=\"font-weight: bold\">{}</span>, <span style=\"color: #008000\">'aggregated_model_validation'</span>: <span style=\"font-weight: bold\">{</span><span style=\"color: #008000\">'function'</span>: <span style=\"color: #008000\">'validate'</span>, <span style=\"color: #008000\">'kwargs'</span>: <span style=\"font-weight: bold\">{</span><span style=\"color: #008000\">'apply'</span>: \n", + " <span style=\"color: #008000\">'global'</span>, <span style=\"color: #008000\">'metrics'</span>: <span style=\"font-weight: bold\">[</span><span style=\"color: #008000\">'dice_coef'</span><span style=\"font-weight: bold\">]}}</span>, <span style=\"color: #008000\">'train'</span>: <span style=\"font-weight: bold\">{</span><span style=\"color: #008000\">'function'</span>: <span style=\"color: #008000\">'train_batches'</span>, <span style=\"color: #008000\">'kwargs'</span>: <span style=\"font-weight: bold\">{</span><span style=\"color: #008000\">'metrics'</span>: <span style=\"font-weight: bold\">[</span><span style=\"color: #008000\">'loss'</span><span style=\"font-weight: bold\">]}}</span>, \n", + " <span style=\"color: #008000\">'locally_tuned_model_validation'</span>: <span style=\"font-weight: bold\">{</span><span style=\"color: #008000\">'function'</span>: <span style=\"color: #008000\">'validate'</span>, <span style=\"color: #008000\">'kwargs'</span>: <span style=\"font-weight: bold\">{</span><span style=\"color: #008000\">'apply'</span>: <span style=\"color: #008000\">'local'</span>, <span style=\"color: #008000\">'metrics'</span>: <span style=\"font-weight: bold\">[</span><span style=\"color: #008000\">'dice_coef'</span><span style=\"font-weight: bold\">]}}}</span>, <span style=\"color: #008000\">'client'</span>: \n", + " <span style=\"font-weight: bold\"><</span><span style=\"color: #ff00ff; font-weight: bold\">openfl.component.aggregator.aggregator.Aggregator</span><span style=\"color: #000000\"> object at </span><span style=\"color: #000080; font-weight: bold\">0x7f7490d61c40</span><span style=\"font-weight: bold\">>}</span> \n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f75299ecf10>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> Override <span style=\"color: #800000\">🡆</span> <span style=\"font-weight: bold\">{</span><span style=\"color: #008000\">'defaults'</span>: <span style=\"color: #008000\">'plan/defaults/collaborator.yaml'</span><span style=\"font-weight: bold\">}</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/federated/plan/plan.py\"><span style=\"color: #7f7f7f\">plan.py</span></a><span style=\"color: #7f7f7f\">:167</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f7490c9eac0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\">[17:44:14] </span><span style=\"color: #000080\">INFO</span> Building <span style=\"color: #800000\">🡆</span> Object <span style=\"color: #800000\">Collaborator</span> from <span style=\"color: #800000\">openfl.component</span> Module. <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/federated/plan/plan.py\"><span style=\"color: #7f7f7f\">plan.py</span></a><span style=\"color: #7f7f7f\">:162</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f7490c9e130>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> Settings <span style=\"color: #800000\">🡆</span> <span style=\"font-weight: bold\">{</span><span style=\"color: #008000\">'db_store_rounds'</span>: <span style=\"color: #000080; font-weight: bold\">1</span>, <span style=\"color: #008000\">'epochs_per_round'</span>: <span style=\"color: #000080; font-weight: bold\">1.0</span>, <span style=\"color: #008000\">'delta_updates'</span>: <span style=\"color: #ff0000; font-style: italic\">False</span>, <span style=\"color: #008000\">'polling_interval'</span>: <span style=\"color: #000080; font-weight: bold\">4</span>, <span style=\"color: #008000\">'opt_treatment'</span>: <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/federated/plan/plan.py\"><span style=\"color: #7f7f7f\">plan.py</span></a><span style=\"color: #7f7f7f\">:165</span>\n", + " <span style=\"color: #008000\">'RESET'</span>, <span style=\"color: #008000\">'collaborator_name'</span>: <span style=\"color: #008000\">'two'</span>, <span style=\"color: #008000\">'aggregator_uuid'</span>: <span style=\"color: #008000\">'aggregator_plan.yaml_851debec'</span>, <span style=\"color: #008000\">'federation_uuid'</span>: \n", + " <span style=\"color: #008000\">'plan.yaml_851debec'</span>, <span style=\"color: #008000\">'task_runner'</span>: <span style=\"font-weight: bold\"><</span><span style=\"color: #ff00ff; font-weight: bold\">openfl.federated.task.fl_model.FederatedModel</span><span style=\"color: #000000\"> object at </span><span style=\"color: #000080; font-weight: bold\">0x7f7529a74490</span><span style=\"font-weight: bold\">></span>, <span style=\"color: #008000\">'tensor_pipe'</span>: \n", + " <span style=\"font-weight: bold\"><</span><span style=\"color: #ff00ff; font-weight: bold\">openfl.pipelines.no_compression_pipeline.NoCompressionPipeline</span><span style=\"color: #000000\"> object at </span><span style=\"color: #000080; font-weight: bold\">0x7f7529a746a0</span><span style=\"font-weight: bold\">></span>, <span style=\"color: #008000\">'task_config'</span>: <span style=\"font-weight: bold\">{</span><span style=\"color: #008000\">'defaults'</span>: \n", + " <span style=\"color: #008000\">'plan/defaults/tasks_torch.yaml'</span>, <span style=\"color: #008000\">'settings'</span>: <span style=\"font-weight: bold\">{}</span>, <span style=\"color: #008000\">'aggregated_model_validation'</span>: <span style=\"font-weight: bold\">{</span><span style=\"color: #008000\">'function'</span>: <span style=\"color: #008000\">'validate'</span>, <span style=\"color: #008000\">'kwargs'</span>: <span style=\"font-weight: bold\">{</span><span style=\"color: #008000\">'apply'</span>: \n", + " <span style=\"color: #008000\">'global'</span>, <span style=\"color: #008000\">'metrics'</span>: <span style=\"font-weight: bold\">[</span><span style=\"color: #008000\">'dice_coef'</span><span style=\"font-weight: bold\">]}}</span>, <span style=\"color: #008000\">'train'</span>: <span style=\"font-weight: bold\">{</span><span style=\"color: #008000\">'function'</span>: <span style=\"color: #008000\">'train_batches'</span>, <span style=\"color: #008000\">'kwargs'</span>: <span style=\"font-weight: bold\">{</span><span style=\"color: #008000\">'metrics'</span>: <span style=\"font-weight: bold\">[</span><span style=\"color: #008000\">'loss'</span><span style=\"font-weight: bold\">]}}</span>, \n", + " <span style=\"color: #008000\">'locally_tuned_model_validation'</span>: <span style=\"font-weight: bold\">{</span><span style=\"color: #008000\">'function'</span>: <span style=\"color: #008000\">'validate'</span>, <span style=\"color: #008000\">'kwargs'</span>: <span style=\"font-weight: bold\">{</span><span style=\"color: #008000\">'apply'</span>: <span style=\"color: #008000\">'local'</span>, <span style=\"color: #008000\">'metrics'</span>: <span style=\"font-weight: bold\">[</span><span style=\"color: #008000\">'dice_coef'</span><span style=\"font-weight: bold\">]}}}</span>, <span style=\"color: #008000\">'client'</span>: \n", + " <span style=\"font-weight: bold\"><</span><span style=\"color: #ff00ff; font-weight: bold\">openfl.component.aggregator.aggregator.Aggregator</span><span style=\"color: #000000\"> object at </span><span style=\"color: #000080; font-weight: bold\">0x7f7490d61c40</span><span style=\"font-weight: bold\">></span>, <span style=\"color: #008000\">'defaults'</span>: <span style=\"color: #008000\">'plan/defaults/collaborator.yaml'</span><span style=\"font-weight: bold\">}</span> \n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f7490c9edf0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> Override <span style=\"color: #800000\">🡆</span> <span style=\"font-weight: bold\">{</span><span style=\"color: #008000\">'defaults'</span>: <span style=\"color: #008000\">'plan/defaults/collaborator.yaml'</span><span style=\"font-weight: bold\">}</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/federated/plan/plan.py\"><span style=\"color: #7f7f7f\">plan.py</span></a><span style=\"color: #7f7f7f\">:167</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f7490c9efd0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> Waiting for tasks<span style=\"color: #808000\">...</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\"><span style=\"color: #7f7f7f\">collaborator.py</span></a><span style=\"color: #7f7f7f\">:156</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f7490c9eee0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> Sending tasks to collaborator one for round <span style=\"color: #000080; font-weight: bold\">0</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\"><span style=\"color: #7f7f7f\">aggregator.py</span></a><span style=\"color: #7f7f7f\">:276</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f7490c9ebb0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> Received the following tasks: <span style=\"font-weight: bold\">[</span><span style=\"color: #008000\">'aggregated_model_validation'</span>, <span style=\"color: #008000\">'train'</span>, <span style=\"color: #008000\">'locally_tuned_model_validation'</span><span style=\"font-weight: bold\">]</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\"><span style=\"color: #7f7f7f\">collaborator.py</span></a><span style=\"color: #7f7f7f\">:144</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f7490c9eca0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "name": "stderr", + "output_type": "stream", + "text": [ + "/home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/federated/task/runner_pt.py:287: UserWarning: The given NumPy array is not writeable, and PyTorch does not support non-writeable tensors. This means you can write to the underlying (supposedly non-writeable) NumPy array using the tensor. You may want to copy the array to protect its data or make it writeable before converting it to a tensor. This type of warning will be suppressed for the rest of this program. (Triggered internally at /pytorch/torch/csrc/utils/tensor_numpy.cpp:143.)\n", + " new_state[k] = pt.from_numpy(tensor_dict.pop(k)).to(device)\n", + "<ipython-input-8-5e98da9ec23c>:124: UserWarning: To copy construct from a tensor, it is recommended to use sourceTensor.clone().detach() or sourceTensor.clone().detach().requires_grad_(True), rather than torch.tensor(sourceTensor).\n", + " torch.tensor(data).to(self.device),\n", + "<ipython-input-8-5e98da9ec23c>:125: UserWarning: To copy construct from a tensor, it is recommended to use sourceTensor.clone().detach() or sourceTensor.clone().detach().requires_grad_(True), rather than torch.tensor(sourceTensor).\n", + " torch.tensor(target).to(self.device),\n" + ] + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\">[17:44:17] </span><span style=\"color: #000080\">INFO</span> Sending metric for task aggregated_model_validation, round number <span style=\"color: #000080; font-weight: bold\">0</span>: dice_coef <span style=\"color: #000080; font-weight: bold\">0.23560654732488817</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\"><span style=\"color: #7f7f7f\">collaborator.py</span></a><span style=\"color: #7f7f7f\">:365</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f74200291c0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> Collaborator one is sending task results for aggregated_model_validation, round <span style=\"color: #000080; font-weight: bold\">0</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\"><span style=\"color: #7f7f7f\">aggregator.py</span></a><span style=\"color: #7f7f7f\">:428</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f74200296a0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "name": "stderr", + "output_type": "stream", + "text": [ + "/home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/federated/task/runner_pt.py:154: UserWarning: To copy construct from a tensor, it is recommended to use sourceTensor.clone().detach() or sourceTensor.clone().detach().requires_grad_(True), rather than torch.tensor(sourceTensor).\n", + " data, target = pt.tensor(data).to(self.device), pt.tensor(\n" + ] + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\">[17:44:37] </span><span style=\"color: #000080\">INFO</span> Sending metric for task train, round number <span style=\"color: #000080; font-weight: bold\">0</span>: soft_dice_loss <span style=\"color: #000080; font-weight: bold\">0.6395286917686462</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\"><span style=\"color: #7f7f7f\">collaborator.py</span></a><span style=\"color: #7f7f7f\">:365</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f7420032f40>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> Collaborator one is sending task results for train, round <span style=\"color: #000080; font-weight: bold\">0</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\"><span style=\"color: #7f7f7f\">aggregator.py</span></a><span style=\"color: #7f7f7f\">:428</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f7420032d30>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\">[17:44:39] </span><span style=\"color: #000080\">INFO</span> Sending metric for task locally_tuned_model_validation, round number <span style=\"color: #000080; font-weight: bold\">0</span>: dice_coef <span style=\"color: #000080; font-weight: bold\">0.43012889739005794</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\"><span style=\"color: #7f7f7f\">collaborator.py</span></a><span style=\"color: #7f7f7f\">:365</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f7490cba880>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> Collaborator one is sending task results for locally_tuned_model_validation, round <span style=\"color: #000080; font-weight: bold\">0</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\"><span style=\"color: #7f7f7f\">aggregator.py</span></a><span style=\"color: #7f7f7f\">:428</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740bf1a7c0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> All tasks completed on one for round <span style=\"color: #000080; font-weight: bold\">0</span><span style=\"color: #808000\">...</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\"><span style=\"color: #7f7f7f\">collaborator.py</span></a><span style=\"color: #7f7f7f\">:148</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f7490cba760>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> Waiting for tasks<span style=\"color: #808000\">...</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\"><span style=\"color: #7f7f7f\">collaborator.py</span></a><span style=\"color: #7f7f7f\">:156</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740bf1a940>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> Sending tasks to collaborator two for round <span style=\"color: #000080; font-weight: bold\">0</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\"><span style=\"color: #7f7f7f\">aggregator.py</span></a><span style=\"color: #7f7f7f\">:276</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f7420029a90>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> Received the following tasks: <span style=\"font-weight: bold\">[</span><span style=\"color: #008000\">'aggregated_model_validation'</span>, <span style=\"color: #008000\">'train'</span>, <span style=\"color: #008000\">'locally_tuned_model_validation'</span><span style=\"font-weight: bold\">]</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\"><span style=\"color: #7f7f7f\">collaborator.py</span></a><span style=\"color: #7f7f7f\">:144</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f7490cc0a90>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\">[17:44:42] </span><span style=\"color: #000080\">INFO</span> Sending metric for task aggregated_model_validation, round number <span style=\"color: #000080; font-weight: bold\">0</span>: dice_coef <span style=\"color: #000080; font-weight: bold\">0.18667999943417887</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\"><span style=\"color: #7f7f7f\">collaborator.py</span></a><span style=\"color: #7f7f7f\">:365</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f73fb224130>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> Collaborator two is sending task results for aggregated_model_validation, round <span style=\"color: #000080; font-weight: bold\">0</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\"><span style=\"color: #7f7f7f\">aggregator.py</span></a><span style=\"color: #7f7f7f\">:428</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740be4c490>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\">[17:45:02] </span><span style=\"color: #000080\">INFO</span> Sending metric for task train, round number <span style=\"color: #000080; font-weight: bold\">0</span>: soft_dice_loss <span style=\"color: #000080; font-weight: bold\">0.6324753165245056</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\"><span style=\"color: #7f7f7f\">collaborator.py</span></a><span style=\"color: #7f7f7f\">:365</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f742002ceb0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> Collaborator two is sending task results for train, round <span style=\"color: #000080; font-weight: bold\">0</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\"><span style=\"color: #7f7f7f\">aggregator.py</span></a><span style=\"color: #7f7f7f\">:428</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f742002c970>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\">[17:45:04] </span><span style=\"color: #000080\">INFO</span> Sending metric for task locally_tuned_model_validation, round number <span style=\"color: #000080; font-weight: bold\">0</span>: dice_coef <span style=\"color: #000080; font-weight: bold\">0.3584949864495185</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\"><span style=\"color: #7f7f7f\">collaborator.py</span></a><span style=\"color: #7f7f7f\">:365</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f742002ca60>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> Collaborator two is sending task results for locally_tuned_model_validation, round <span style=\"color: #000080; font-weight: bold\">0</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\"><span style=\"color: #7f7f7f\">aggregator.py</span></a><span style=\"color: #7f7f7f\">:428</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740be83430>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> train task metrics<span style=\"color: #808000\">...</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\"><span style=\"color: #7f7f7f\">aggregator.py</span></a><span style=\"color: #7f7f7f\">:694</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740be83430>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> soft_dice_loss: <span style=\"color: #000080; font-weight: bold\">0.6360</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\"><span style=\"color: #7f7f7f\">aggregator.py</span></a><span style=\"color: #7f7f7f\">:742</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740be83430>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\">[17:45:09] </span><span style=\"color: #000080\">INFO</span> locally_tuned_model_validation task metrics<span style=\"color: #808000\">...</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\"><span style=\"color: #7f7f7f\">aggregator.py</span></a><span style=\"color: #7f7f7f\">:694</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740bdacac0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> dice_coef: <span style=\"color: #000080; font-weight: bold\">0.3943</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\"><span style=\"color: #7f7f7f\">aggregator.py</span></a><span style=\"color: #7f7f7f\">:742</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740bdacbb0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> aggregated_model_validation task metrics<span style=\"color: #808000\">...</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\"><span style=\"color: #7f7f7f\">aggregator.py</span></a><span style=\"color: #7f7f7f\">:694</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740bdacc10>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> dice_coef: <span style=\"color: #000080; font-weight: bold\">0.2111</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\"><span style=\"color: #7f7f7f\">aggregator.py</span></a><span style=\"color: #7f7f7f\">:742</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740bdac850>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> Saved the best model with score <span style=\"color: #000080; font-weight: bold\">0.211143</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\"><span style=\"color: #7f7f7f\">aggregator.py</span></a><span style=\"color: #7f7f7f\">:749</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740bdac6d0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> Saving round <span style=\"color: #000080; font-weight: bold\">1</span> model<span style=\"color: #808000\">...</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\"><span style=\"color: #7f7f7f\">aggregator.py</span></a><span style=\"color: #7f7f7f\">:783</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740bdacac0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> Starting round <span style=\"color: #000080; font-weight: bold\">1</span><span style=\"color: #808000\">...</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\"><span style=\"color: #7f7f7f\">aggregator.py</span></a><span style=\"color: #7f7f7f\">:791</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740bdac850>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> All tasks completed on two for round <span style=\"color: #000080; font-weight: bold\">0</span><span style=\"color: #808000\">...</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\"><span style=\"color: #7f7f7f\">collaborator.py</span></a><span style=\"color: #7f7f7f\">:148</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740be4cd00>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\">[17:45:10] </span><span style=\"color: #000080\">INFO</span> Waiting for tasks<span style=\"color: #808000\">...</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\"><span style=\"color: #7f7f7f\">collaborator.py</span></a><span style=\"color: #7f7f7f\">:156</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740be5bfa0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> Sending tasks to collaborator one for round <span style=\"color: #000080; font-weight: bold\">1</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\"><span style=\"color: #7f7f7f\">aggregator.py</span></a><span style=\"color: #7f7f7f\">:276</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740be87a30>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> Received the following tasks: <span style=\"font-weight: bold\">[</span><span style=\"color: #008000\">'aggregated_model_validation'</span>, <span style=\"color: #008000\">'train'</span>, <span style=\"color: #008000\">'locally_tuned_model_validation'</span><span style=\"font-weight: bold\">]</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\"><span style=\"color: #7f7f7f\">collaborator.py</span></a><span style=\"color: #7f7f7f\">:144</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740be87b80>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\">[17:45:12] </span><span style=\"color: #000080\">INFO</span> Sending metric for task aggregated_model_validation, round number <span style=\"color: #000080; font-weight: bold\">1</span>: dice_coef <span style=\"color: #000080; font-weight: bold\">0.4241240389885441</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\"><span style=\"color: #7f7f7f\">collaborator.py</span></a><span style=\"color: #7f7f7f\">:365</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f742000ea30>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> Collaborator one is sending task results for aggregated_model_validation, round <span style=\"color: #000080; font-weight: bold\">1</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\"><span style=\"color: #7f7f7f\">aggregator.py</span></a><span style=\"color: #7f7f7f\">:428</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f7420029a60>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\">[17:45:32] </span><span style=\"color: #000080\">INFO</span> Sending metric for task train, round number <span style=\"color: #000080; font-weight: bold\">1</span>: soft_dice_loss <span style=\"color: #000080; font-weight: bold\">0.6150628328323364</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\"><span style=\"color: #7f7f7f\">collaborator.py</span></a><span style=\"color: #7f7f7f\">:365</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f75299eca30>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> Collaborator one is sending task results for train, round <span style=\"color: #000080; font-weight: bold\">1</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\"><span style=\"color: #7f7f7f\">aggregator.py</span></a><span style=\"color: #7f7f7f\">:428</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f7490cc0be0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\">[17:45:34] </span><span style=\"color: #000080\">INFO</span> Sending metric for task locally_tuned_model_validation, round number <span style=\"color: #000080; font-weight: bold\">1</span>: dice_coef <span style=\"color: #000080; font-weight: bold\">0.44148768724933746</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\"><span style=\"color: #7f7f7f\">collaborator.py</span></a><span style=\"color: #7f7f7f\">:365</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f74200299d0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> Collaborator one is sending task results for locally_tuned_model_validation, round <span style=\"color: #000080; font-weight: bold\">1</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\"><span style=\"color: #7f7f7f\">aggregator.py</span></a><span style=\"color: #7f7f7f\">:428</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f74200296d0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> All tasks completed on one for round <span style=\"color: #000080; font-weight: bold\">1</span><span style=\"color: #808000\">...</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\"><span style=\"color: #7f7f7f\">collaborator.py</span></a><span style=\"color: #7f7f7f\">:148</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f7420029640>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> Waiting for tasks<span style=\"color: #808000\">...</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\"><span style=\"color: #7f7f7f\">collaborator.py</span></a><span style=\"color: #7f7f7f\">:156</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f7420029f40>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> Sending tasks to collaborator two for round <span style=\"color: #000080; font-weight: bold\">1</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\"><span style=\"color: #7f7f7f\">aggregator.py</span></a><span style=\"color: #7f7f7f\">:276</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f7420029100>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> Received the following tasks: <span style=\"font-weight: bold\">[</span><span style=\"color: #008000\">'aggregated_model_validation'</span>, <span style=\"color: #008000\">'train'</span>, <span style=\"color: #008000\">'locally_tuned_model_validation'</span><span style=\"font-weight: bold\">]</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\"><span style=\"color: #7f7f7f\">collaborator.py</span></a><span style=\"color: #7f7f7f\">:144</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f7420029190>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\">[17:45:37] </span><span style=\"color: #000080\">INFO</span> Sending metric for task aggregated_model_validation, round number <span style=\"color: #000080; font-weight: bold\">1</span>: dice_coef <span style=\"color: #000080; font-weight: bold\">0.34520076744018063</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\"><span style=\"color: #7f7f7f\">collaborator.py</span></a><span style=\"color: #7f7f7f\">:365</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740be839a0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> Collaborator two is sending task results for aggregated_model_validation, round <span style=\"color: #000080; font-weight: bold\">1</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\"><span style=\"color: #7f7f7f\">aggregator.py</span></a><span style=\"color: #7f7f7f\">:428</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740be4c880>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\">[17:45:57] </span><span style=\"color: #000080\">INFO</span> Sending metric for task train, round number <span style=\"color: #000080; font-weight: bold\">1</span>: soft_dice_loss <span style=\"color: #000080; font-weight: bold\">0.6058140993118286</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\"><span style=\"color: #7f7f7f\">collaborator.py</span></a><span style=\"color: #7f7f7f\">:365</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740be80b20>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> Collaborator two is sending task results for train, round <span style=\"color: #000080; font-weight: bold\">1</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\"><span style=\"color: #7f7f7f\">aggregator.py</span></a><span style=\"color: #7f7f7f\">:428</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740be80910>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\">[17:45:59] </span><span style=\"color: #000080\">INFO</span> Sending metric for task locally_tuned_model_validation, round number <span style=\"color: #000080; font-weight: bold\">1</span>: dice_coef <span style=\"color: #000080; font-weight: bold\">0.3571049436446159</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\"><span style=\"color: #7f7f7f\">collaborator.py</span></a><span style=\"color: #7f7f7f\">:365</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f7490cc05e0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> Collaborator two is sending task results for locally_tuned_model_validation, round <span style=\"color: #000080; font-weight: bold\">1</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\"><span style=\"color: #7f7f7f\">aggregator.py</span></a><span style=\"color: #7f7f7f\">:428</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f7490cc0550>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> train task metrics<span style=\"color: #808000\">...</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\"><span style=\"color: #7f7f7f\">aggregator.py</span></a><span style=\"color: #7f7f7f\">:694</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740be8e220>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> soft_dice_loss: <span style=\"color: #000080; font-weight: bold\">0.6104</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\"><span style=\"color: #7f7f7f\">aggregator.py</span></a><span style=\"color: #7f7f7f\">:742</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740be8efd0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\">[17:46:05] </span><span style=\"color: #000080\">INFO</span> locally_tuned_model_validation task metrics<span style=\"color: #808000\">...</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\"><span style=\"color: #7f7f7f\">aggregator.py</span></a><span style=\"color: #7f7f7f\">:694</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f7490cbae80>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> dice_coef: <span style=\"color: #000080; font-weight: bold\">0.3993</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\"><span style=\"color: #7f7f7f\">aggregator.py</span></a><span style=\"color: #7f7f7f\">:742</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740be8ea30>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> aggregated_model_validation task metrics<span style=\"color: #808000\">...</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\"><span style=\"color: #7f7f7f\">aggregator.py</span></a><span style=\"color: #7f7f7f\">:694</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740bf159a0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> dice_coef: <span style=\"color: #000080; font-weight: bold\">0.3847</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\"><span style=\"color: #7f7f7f\">aggregator.py</span></a><span style=\"color: #7f7f7f\">:742</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740bf0e6d0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> Saved the best model with score <span style=\"color: #000080; font-weight: bold\">0.384662</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\"><span style=\"color: #7f7f7f\">aggregator.py</span></a><span style=\"color: #7f7f7f\">:749</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740bf15940>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\">[17:46:06] </span><span style=\"color: #000080\">INFO</span> Saving round <span style=\"color: #000080; font-weight: bold\">2</span> model<span style=\"color: #808000\">...</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\"><span style=\"color: #7f7f7f\">aggregator.py</span></a><span style=\"color: #7f7f7f\">:783</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740bf0e6d0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> Starting round <span style=\"color: #000080; font-weight: bold\">2</span><span style=\"color: #808000\">...</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\"><span style=\"color: #7f7f7f\">aggregator.py</span></a><span style=\"color: #7f7f7f\">:791</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740bf2d160>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> All tasks completed on two for round <span style=\"color: #000080; font-weight: bold\">1</span><span style=\"color: #808000\">...</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\"><span style=\"color: #7f7f7f\">collaborator.py</span></a><span style=\"color: #7f7f7f\">:148</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f74200327f0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> Waiting for tasks<span style=\"color: #808000\">...</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\"><span style=\"color: #7f7f7f\">collaborator.py</span></a><span style=\"color: #7f7f7f\">:156</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f7420032130>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> Sending tasks to collaborator one for round <span style=\"color: #000080; font-weight: bold\">2</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\"><span style=\"color: #7f7f7f\">aggregator.py</span></a><span style=\"color: #7f7f7f\">:276</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f7490d3feb0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> Received the following tasks: <span style=\"font-weight: bold\">[</span><span style=\"color: #008000\">'aggregated_model_validation'</span>, <span style=\"color: #008000\">'train'</span>, <span style=\"color: #008000\">'locally_tuned_model_validation'</span><span style=\"font-weight: bold\">]</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\"><span style=\"color: #7f7f7f\">collaborator.py</span></a><span style=\"color: #7f7f7f\">:144</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f7420032130>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\">[17:46:09] </span><span style=\"color: #000080\">INFO</span> Sending metric for task aggregated_model_validation, round number <span style=\"color: #000080; font-weight: bold\">2</span>: dice_coef <span style=\"color: #000080; font-weight: bold\">0.43549662059353245</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\"><span style=\"color: #7f7f7f\">collaborator.py</span></a><span style=\"color: #7f7f7f\">:365</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f7490d3fac0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> Collaborator one is sending task results for aggregated_model_validation, round <span style=\"color: #000080; font-weight: bold\">2</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\"><span style=\"color: #7f7f7f\">aggregator.py</span></a><span style=\"color: #7f7f7f\">:428</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740be5be50>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\">[17:46:29] </span><span style=\"color: #000080\">INFO</span> Sending metric for task train, round number <span style=\"color: #000080; font-weight: bold\">2</span>: soft_dice_loss <span style=\"color: #000080; font-weight: bold\">0.5963471531867981</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\"><span style=\"color: #7f7f7f\">collaborator.py</span></a><span style=\"color: #7f7f7f\">:365</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740bf1aac0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> Collaborator one is sending task results for train, round <span style=\"color: #000080; font-weight: bold\">2</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\"><span style=\"color: #7f7f7f\">aggregator.py</span></a><span style=\"color: #7f7f7f\">:428</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740bf1a8e0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\">[17:46:31] </span><span style=\"color: #000080\">INFO</span> Sending metric for task locally_tuned_model_validation, round number <span style=\"color: #000080; font-weight: bold\">2</span>: dice_coef <span style=\"color: #000080; font-weight: bold\">0.4504963544107253</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\"><span style=\"color: #7f7f7f\">collaborator.py</span></a><span style=\"color: #7f7f7f\">:365</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740bffce50>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\">[17:46:32] </span><span style=\"color: #000080\">INFO</span> Collaborator one is sending task results for locally_tuned_model_validation, round <span style=\"color: #000080; font-weight: bold\">2</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\"><span style=\"color: #7f7f7f\">aggregator.py</span></a><span style=\"color: #7f7f7f\">:428</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740bffc5e0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> All tasks completed on one for round <span style=\"color: #000080; font-weight: bold\">2</span><span style=\"color: #808000\">...</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\"><span style=\"color: #7f7f7f\">collaborator.py</span></a><span style=\"color: #7f7f7f\">:148</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740bffc670>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> Waiting for tasks<span style=\"color: #808000\">...</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\"><span style=\"color: #7f7f7f\">collaborator.py</span></a><span style=\"color: #7f7f7f\">:156</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740bffc6d0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> Sending tasks to collaborator two for round <span style=\"color: #000080; font-weight: bold\">2</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\"><span style=\"color: #7f7f7f\">aggregator.py</span></a><span style=\"color: #7f7f7f\">:276</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740bffccd0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> Received the following tasks: <span style=\"font-weight: bold\">[</span><span style=\"color: #008000\">'aggregated_model_validation'</span>, <span style=\"color: #008000\">'train'</span>, <span style=\"color: #008000\">'locally_tuned_model_validation'</span><span style=\"font-weight: bold\">]</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\"><span style=\"color: #7f7f7f\">collaborator.py</span></a><span style=\"color: #7f7f7f\">:144</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740bffc940>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\">[17:46:34] </span><span style=\"color: #000080\">INFO</span> Sending metric for task aggregated_model_validation, round number <span style=\"color: #000080; font-weight: bold\">2</span>: dice_coef <span style=\"color: #000080; font-weight: bold\">0.3556097505554076</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\"><span style=\"color: #7f7f7f\">collaborator.py</span></a><span style=\"color: #7f7f7f\">:365</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740bdacc40>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> Collaborator two is sending task results for aggregated_model_validation, round <span style=\"color: #000080; font-weight: bold\">2</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\"><span style=\"color: #7f7f7f\">aggregator.py</span></a><span style=\"color: #7f7f7f\">:428</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f7490cbaa90>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\">[17:46:55] </span><span style=\"color: #000080\">INFO</span> Sending metric for task train, round number <span style=\"color: #000080; font-weight: bold\">2</span>: soft_dice_loss <span style=\"color: #000080; font-weight: bold\">0.5920241475105286</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\"><span style=\"color: #7f7f7f\">collaborator.py</span></a><span style=\"color: #7f7f7f\">:365</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740bea9d90>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> Collaborator two is sending task results for train, round <span style=\"color: #000080; font-weight: bold\">2</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\"><span style=\"color: #7f7f7f\">aggregator.py</span></a><span style=\"color: #7f7f7f\">:428</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740bea90d0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\">[17:46:57] </span><span style=\"color: #000080\">INFO</span> Sending metric for task locally_tuned_model_validation, round number <span style=\"color: #000080; font-weight: bold\">2</span>: dice_coef <span style=\"color: #000080; font-weight: bold\">0.37100199874370327</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\"><span style=\"color: #7f7f7f\">collaborator.py</span></a><span style=\"color: #7f7f7f\">:365</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f7490d3feb0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> Collaborator two is sending task results for locally_tuned_model_validation, round <span style=\"color: #000080; font-weight: bold\">2</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\"><span style=\"color: #7f7f7f\">aggregator.py</span></a><span style=\"color: #7f7f7f\">:428</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740bf0e910>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> train task metrics<span style=\"color: #808000\">...</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\"><span style=\"color: #7f7f7f\">aggregator.py</span></a><span style=\"color: #7f7f7f\">:694</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740bf0e220>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> soft_dice_loss: <span style=\"color: #000080; font-weight: bold\">0.5942</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\"><span style=\"color: #7f7f7f\">aggregator.py</span></a><span style=\"color: #7f7f7f\">:742</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740bf0e400>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\">[17:47:03] </span><span style=\"color: #000080\">INFO</span> locally_tuned_model_validation task metrics<span style=\"color: #808000\">...</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\"><span style=\"color: #7f7f7f\">aggregator.py</span></a><span style=\"color: #7f7f7f\">:694</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f742002c5e0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> dice_coef: <span style=\"color: #000080; font-weight: bold\">0.4107</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\"><span style=\"color: #7f7f7f\">aggregator.py</span></a><span style=\"color: #7f7f7f\">:742</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f7490d3f7c0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> aggregated_model_validation task metrics<span style=\"color: #808000\">...</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\"><span style=\"color: #7f7f7f\">aggregator.py</span></a><span style=\"color: #7f7f7f\">:694</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740bd1b9a0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> dice_coef: <span style=\"color: #000080; font-weight: bold\">0.3956</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\"><span style=\"color: #7f7f7f\">aggregator.py</span></a><span style=\"color: #7f7f7f\">:742</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740bf15ca0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> Saved the best model with score <span style=\"color: #000080; font-weight: bold\">0.395553</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\"><span style=\"color: #7f7f7f\">aggregator.py</span></a><span style=\"color: #7f7f7f\">:749</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740bd1b820>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\">[17:47:04] </span><span style=\"color: #000080\">INFO</span> Saving round <span style=\"color: #000080; font-weight: bold\">3</span> model<span style=\"color: #808000\">...</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\"><span style=\"color: #7f7f7f\">aggregator.py</span></a><span style=\"color: #7f7f7f\">:783</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f742002cc70>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> Starting round <span style=\"color: #000080; font-weight: bold\">3</span><span style=\"color: #808000\">...</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\"><span style=\"color: #7f7f7f\">aggregator.py</span></a><span style=\"color: #7f7f7f\">:791</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740bfedca0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> All tasks completed on two for round <span style=\"color: #000080; font-weight: bold\">2</span><span style=\"color: #808000\">...</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\"><span style=\"color: #7f7f7f\">collaborator.py</span></a><span style=\"color: #7f7f7f\">:148</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740bfd9e50>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> Waiting for tasks<span style=\"color: #808000\">...</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\"><span style=\"color: #7f7f7f\">collaborator.py</span></a><span style=\"color: #7f7f7f\">:156</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740bf94eb0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> Sending tasks to collaborator one for round <span style=\"color: #000080; font-weight: bold\">3</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\"><span style=\"color: #7f7f7f\">aggregator.py</span></a><span style=\"color: #7f7f7f\">:276</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740bf94310>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> Received the following tasks: <span style=\"font-weight: bold\">[</span><span style=\"color: #008000\">'aggregated_model_validation'</span>, <span style=\"color: #008000\">'train'</span>, <span style=\"color: #008000\">'locally_tuned_model_validation'</span><span style=\"font-weight: bold\">]</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\"><span style=\"color: #7f7f7f\">collaborator.py</span></a><span style=\"color: #7f7f7f\">:144</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740bf94520>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\">[17:47:06] </span><span style=\"color: #000080\">INFO</span> Sending metric for task aggregated_model_validation, round number <span style=\"color: #000080; font-weight: bold\">3</span>: dice_coef <span style=\"color: #000080; font-weight: bold\">0.43500437275055914</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\"><span style=\"color: #7f7f7f\">collaborator.py</span></a><span style=\"color: #7f7f7f\">:365</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f7490cbaa60>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> Collaborator one is sending task results for aggregated_model_validation, round <span style=\"color: #000080; font-weight: bold\">3</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\"><span style=\"color: #7f7f7f\">aggregator.py</span></a><span style=\"color: #7f7f7f\">:428</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740bf2f8e0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\">[17:47:27] </span><span style=\"color: #000080\">INFO</span> Sending metric for task train, round number <span style=\"color: #000080; font-weight: bold\">3</span>: soft_dice_loss <span style=\"color: #000080; font-weight: bold\">0.5834751725196838</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\"><span style=\"color: #7f7f7f\">collaborator.py</span></a><span style=\"color: #7f7f7f\">:365</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f748dc86a30>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> Collaborator one is sending task results for train, round <span style=\"color: #000080; font-weight: bold\">3</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\"><span style=\"color: #7f7f7f\">aggregator.py</span></a><span style=\"color: #7f7f7f\">:428</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f742003f6d0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\">[17:47:30] </span><span style=\"color: #000080\">INFO</span> Sending metric for task locally_tuned_model_validation, round number <span style=\"color: #000080; font-weight: bold\">3</span>: dice_coef <span style=\"color: #000080; font-weight: bold\">0.45610655507733744</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\"><span style=\"color: #7f7f7f\">collaborator.py</span></a><span style=\"color: #7f7f7f\">:365</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740bf146d0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> Collaborator one is sending task results for locally_tuned_model_validation, round <span style=\"color: #000080; font-weight: bold\">3</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\"><span style=\"color: #7f7f7f\">aggregator.py</span></a><span style=\"color: #7f7f7f\">:428</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740bf14580>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> All tasks completed on one for round <span style=\"color: #000080; font-weight: bold\">3</span><span style=\"color: #808000\">...</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\"><span style=\"color: #7f7f7f\">collaborator.py</span></a><span style=\"color: #7f7f7f\">:148</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740bf14b80>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> Waiting for tasks<span style=\"color: #808000\">...</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\"><span style=\"color: #7f7f7f\">collaborator.py</span></a><span style=\"color: #7f7f7f\">:156</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740bf14940>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> Sending tasks to collaborator two for round <span style=\"color: #000080; font-weight: bold\">3</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\"><span style=\"color: #7f7f7f\">aggregator.py</span></a><span style=\"color: #7f7f7f\">:276</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740bf14e20>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> Received the following tasks: <span style=\"font-weight: bold\">[</span><span style=\"color: #008000\">'aggregated_model_validation'</span>, <span style=\"color: #008000\">'train'</span>, <span style=\"color: #008000\">'locally_tuned_model_validation'</span><span style=\"font-weight: bold\">]</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\"><span style=\"color: #7f7f7f\">collaborator.py</span></a><span style=\"color: #7f7f7f\">:144</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740bf14b20>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\">[17:47:32] </span><span style=\"color: #000080\">INFO</span> Sending metric for task aggregated_model_validation, round number <span style=\"color: #000080; font-weight: bold\">3</span>: dice_coef <span style=\"color: #000080; font-weight: bold\">0.3653699919100731</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\"><span style=\"color: #7f7f7f\">collaborator.py</span></a><span style=\"color: #7f7f7f\">:365</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740bf94670>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> Collaborator two is sending task results for aggregated_model_validation, round <span style=\"color: #000080; font-weight: bold\">3</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\"><span style=\"color: #7f7f7f\">aggregator.py</span></a><span style=\"color: #7f7f7f\">:428</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740be8edf0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\">[17:47:53] </span><span style=\"color: #000080\">INFO</span> Sending metric for task train, round number <span style=\"color: #000080; font-weight: bold\">3</span>: soft_dice_loss <span style=\"color: #000080; font-weight: bold\">0.5735182762145996</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\"><span style=\"color: #7f7f7f\">collaborator.py</span></a><span style=\"color: #7f7f7f\">:365</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740bf76e80>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> Collaborator two is sending task results for train, round <span style=\"color: #000080; font-weight: bold\">3</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\"><span style=\"color: #7f7f7f\">aggregator.py</span></a><span style=\"color: #7f7f7f\">:428</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740bf76820>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\">[17:47:56] </span><span style=\"color: #000080\">INFO</span> Sending metric for task locally_tuned_model_validation, round number <span style=\"color: #000080; font-weight: bold\">3</span>: dice_coef <span style=\"color: #000080; font-weight: bold\">0.3892689275164758</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\"><span style=\"color: #7f7f7f\">collaborator.py</span></a><span style=\"color: #7f7f7f\">:365</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740be80340>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> Collaborator two is sending task results for locally_tuned_model_validation, round <span style=\"color: #000080; font-weight: bold\">3</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\"><span style=\"color: #7f7f7f\">aggregator.py</span></a><span style=\"color: #7f7f7f\">:428</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740be80190>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> train task metrics<span style=\"color: #808000\">...</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\"><span style=\"color: #7f7f7f\">aggregator.py</span></a><span style=\"color: #7f7f7f\">:694</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740be800d0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> soft_dice_loss: <span style=\"color: #000080; font-weight: bold\">0.5785</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\"><span style=\"color: #7f7f7f\">aggregator.py</span></a><span style=\"color: #7f7f7f\">:742</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740be801c0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\">[17:48:02] </span><span style=\"color: #000080\">INFO</span> locally_tuned_model_validation task metrics<span style=\"color: #808000\">...</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\"><span style=\"color: #7f7f7f\">aggregator.py</span></a><span style=\"color: #7f7f7f\">:694</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740bfff7f0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> dice_coef: <span style=\"color: #000080; font-weight: bold\">0.4227</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\"><span style=\"color: #7f7f7f\">aggregator.py</span></a><span style=\"color: #7f7f7f\">:742</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740bfff730>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> aggregated_model_validation task metrics<span style=\"color: #808000\">...</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\"><span style=\"color: #7f7f7f\">aggregator.py</span></a><span style=\"color: #7f7f7f\">:694</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740bf2d340>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> dice_coef: <span style=\"color: #000080; font-weight: bold\">0.4002</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\"><span style=\"color: #7f7f7f\">aggregator.py</span></a><span style=\"color: #7f7f7f\">:742</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740bf2d6d0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> Saved the best model with score <span style=\"color: #000080; font-weight: bold\">0.400187</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\"><span style=\"color: #7f7f7f\">aggregator.py</span></a><span style=\"color: #7f7f7f\">:749</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740bfff0a0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> Saving round <span style=\"color: #000080; font-weight: bold\">4</span> model<span style=\"color: #808000\">...</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\"><span style=\"color: #7f7f7f\">aggregator.py</span></a><span style=\"color: #7f7f7f\">:783</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f7420029640>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\">[17:48:03] </span><span style=\"color: #000080\">INFO</span> Starting round <span style=\"color: #000080; font-weight: bold\">4</span><span style=\"color: #808000\">...</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\"><span style=\"color: #7f7f7f\">aggregator.py</span></a><span style=\"color: #7f7f7f\">:791</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740be80580>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> All tasks completed on two for round <span style=\"color: #000080; font-weight: bold\">3</span><span style=\"color: #808000\">...</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\"><span style=\"color: #7f7f7f\">collaborator.py</span></a><span style=\"color: #7f7f7f\">:148</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740bf060d0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> Waiting for tasks<span style=\"color: #808000\">...</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\"><span style=\"color: #7f7f7f\">collaborator.py</span></a><span style=\"color: #7f7f7f\">:156</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f7490c9e9a0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> Sending tasks to collaborator one for round <span style=\"color: #000080; font-weight: bold\">4</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\"><span style=\"color: #7f7f7f\">aggregator.py</span></a><span style=\"color: #7f7f7f\">:276</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740bf060d0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> Received the following tasks: <span style=\"font-weight: bold\">[</span><span style=\"color: #008000\">'aggregated_model_validation'</span>, <span style=\"color: #008000\">'train'</span>, <span style=\"color: #008000\">'locally_tuned_model_validation'</span><span style=\"font-weight: bold\">]</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\"><span style=\"color: #7f7f7f\">collaborator.py</span></a><span style=\"color: #7f7f7f\">:144</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f7490c9e9a0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\">[17:48:05] </span><span style=\"color: #000080\">INFO</span> Sending metric for task aggregated_model_validation, round number <span style=\"color: #000080; font-weight: bold\">4</span>: dice_coef <span style=\"color: #000080; font-weight: bold\">0.4355728107113992</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\"><span style=\"color: #7f7f7f\">collaborator.py</span></a><span style=\"color: #7f7f7f\">:365</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f742003fa00>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> Collaborator one is sending task results for aggregated_model_validation, round <span style=\"color: #000080; font-weight: bold\">4</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\"><span style=\"color: #7f7f7f\">aggregator.py</span></a><span style=\"color: #7f7f7f\">:428</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f742003f730>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\">[17:48:26] </span><span style=\"color: #000080\">INFO</span> Sending metric for task train, round number <span style=\"color: #000080; font-weight: bold\">4</span>: soft_dice_loss <span style=\"color: #000080; font-weight: bold\">0.5637962222099304</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\"><span style=\"color: #7f7f7f\">collaborator.py</span></a><span style=\"color: #7f7f7f\">:365</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740bf76a60>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> Collaborator one is sending task results for train, round <span style=\"color: #000080; font-weight: bold\">4</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\"><span style=\"color: #7f7f7f\">aggregator.py</span></a><span style=\"color: #7f7f7f\">:428</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740bdb93a0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\">[17:48:29] </span><span style=\"color: #000080\">INFO</span> Sending metric for task locally_tuned_model_validation, round number <span style=\"color: #000080; font-weight: bold\">4</span>: dice_coef <span style=\"color: #000080; font-weight: bold\">0.46870911505914503</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\"><span style=\"color: #7f7f7f\">collaborator.py</span></a><span style=\"color: #7f7f7f\">:365</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740bf06730>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> Collaborator one is sending task results for locally_tuned_model_validation, round <span style=\"color: #000080; font-weight: bold\">4</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\"><span style=\"color: #7f7f7f\">aggregator.py</span></a><span style=\"color: #7f7f7f\">:428</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f7420015730>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> All tasks completed on one for round <span style=\"color: #000080; font-weight: bold\">4</span><span style=\"color: #808000\">...</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\"><span style=\"color: #7f7f7f\">collaborator.py</span></a><span style=\"color: #7f7f7f\">:148</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740bf06730>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> Waiting for tasks<span style=\"color: #808000\">...</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\"><span style=\"color: #7f7f7f\">collaborator.py</span></a><span style=\"color: #7f7f7f\">:156</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740bf15f70>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> Sending tasks to collaborator two for round <span style=\"color: #000080; font-weight: bold\">4</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\"><span style=\"color: #7f7f7f\">aggregator.py</span></a><span style=\"color: #7f7f7f\">:276</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740bfffc70>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> Received the following tasks: <span style=\"font-weight: bold\">[</span><span style=\"color: #008000\">'aggregated_model_validation'</span>, <span style=\"color: #008000\">'train'</span>, <span style=\"color: #008000\">'locally_tuned_model_validation'</span><span style=\"font-weight: bold\">]</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\"><span style=\"color: #7f7f7f\">collaborator.py</span></a><span style=\"color: #7f7f7f\">:144</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740bf154f0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\">[17:48:31] </span><span style=\"color: #000080\">INFO</span> Sending metric for task aggregated_model_validation, round number <span style=\"color: #000080; font-weight: bold\">4</span>: dice_coef <span style=\"color: #000080; font-weight: bold\">0.37851148987970046</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\"><span style=\"color: #7f7f7f\">collaborator.py</span></a><span style=\"color: #7f7f7f\">:365</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740be2fb50>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> Collaborator two is sending task results for aggregated_model_validation, round <span style=\"color: #000080; font-weight: bold\">4</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\"><span style=\"color: #7f7f7f\">aggregator.py</span></a><span style=\"color: #7f7f7f\">:428</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740be2fbb0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\">[17:48:52] </span><span style=\"color: #000080\">INFO</span> Sending metric for task train, round number <span style=\"color: #000080; font-weight: bold\">4</span>: soft_dice_loss <span style=\"color: #000080; font-weight: bold\">0.5667974948883057</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\"><span style=\"color: #7f7f7f\">collaborator.py</span></a><span style=\"color: #7f7f7f\">:365</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f742002cd60>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> Collaborator two is sending task results for train, round <span style=\"color: #000080; font-weight: bold\">4</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\"><span style=\"color: #7f7f7f\">aggregator.py</span></a><span style=\"color: #7f7f7f\">:428</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740bdb93a0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\">[17:48:55] </span><span style=\"color: #000080\">INFO</span> Sending metric for task locally_tuned_model_validation, round number <span style=\"color: #000080; font-weight: bold\">4</span>: dice_coef <span style=\"color: #000080; font-weight: bold\">0.3846700369350372</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\"><span style=\"color: #7f7f7f\">collaborator.py</span></a><span style=\"color: #7f7f7f\">:365</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740be2f4f0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> Collaborator two is sending task results for locally_tuned_model_validation, round <span style=\"color: #000080; font-weight: bold\">4</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\"><span style=\"color: #7f7f7f\">aggregator.py</span></a><span style=\"color: #7f7f7f\">:428</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740bf624c0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> train task metrics<span style=\"color: #808000\">...</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\"><span style=\"color: #7f7f7f\">aggregator.py</span></a><span style=\"color: #7f7f7f\">:694</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740bfe3190>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> soft_dice_loss: <span style=\"color: #000080; font-weight: bold\">0.5653</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\"><span style=\"color: #7f7f7f\">aggregator.py</span></a><span style=\"color: #7f7f7f\">:742</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f742003f400>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\">[17:49:01] </span><span style=\"color: #000080\">INFO</span> locally_tuned_model_validation task metrics<span style=\"color: #808000\">...</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\"><span style=\"color: #7f7f7f\">aggregator.py</span></a><span style=\"color: #7f7f7f\">:694</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740bd5c610>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> dice_coef: <span style=\"color: #000080; font-weight: bold\">0.4267</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\"><span style=\"color: #7f7f7f\">aggregator.py</span></a><span style=\"color: #7f7f7f\">:742</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740bf14970>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> aggregated_model_validation task metrics<span style=\"color: #808000\">...</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\"><span style=\"color: #7f7f7f\">aggregator.py</span></a><span style=\"color: #7f7f7f\">:694</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740bf144f0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> dice_coef: <span style=\"color: #000080; font-weight: bold\">0.4070</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\"><span style=\"color: #7f7f7f\">aggregator.py</span></a><span style=\"color: #7f7f7f\">:742</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740bd2e190>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> Saved the best model with score <span style=\"color: #000080; font-weight: bold\">0.407042</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\"><span style=\"color: #7f7f7f\">aggregator.py</span></a><span style=\"color: #7f7f7f\">:749</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740bf2ad00>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> Saving round <span style=\"color: #000080; font-weight: bold\">5</span> model<span style=\"color: #808000\">...</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\"><span style=\"color: #7f7f7f\">aggregator.py</span></a><span style=\"color: #7f7f7f\">:783</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740bd94970>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\">[17:49:02] </span><span style=\"color: #000080\">INFO</span> Starting round <span style=\"color: #000080; font-weight: bold\">5</span><span style=\"color: #808000\">...</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\"><span style=\"color: #7f7f7f\">aggregator.py</span></a><span style=\"color: #7f7f7f\">:791</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740bf15640>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> All tasks completed on two for round <span style=\"color: #000080; font-weight: bold\">4</span><span style=\"color: #808000\">...</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\"><span style=\"color: #7f7f7f\">collaborator.py</span></a><span style=\"color: #7f7f7f\">:148</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740bd9e640>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> Waiting for tasks<span style=\"color: #808000\">...</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\"><span style=\"color: #7f7f7f\">collaborator.py</span></a><span style=\"color: #7f7f7f\">:156</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f7490d3fc70>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> Sending tasks to collaborator one for round <span style=\"color: #000080; font-weight: bold\">5</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\"><span style=\"color: #7f7f7f\">aggregator.py</span></a><span style=\"color: #7f7f7f\">:276</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740bd9e640>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> Received the following tasks: <span style=\"font-weight: bold\">[</span><span style=\"color: #008000\">'aggregated_model_validation'</span>, <span style=\"color: #008000\">'train'</span>, <span style=\"color: #008000\">'locally_tuned_model_validation'</span><span style=\"font-weight: bold\">]</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\"><span style=\"color: #7f7f7f\">collaborator.py</span></a><span style=\"color: #7f7f7f\">:144</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f742002c6d0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\">[17:49:04] </span><span style=\"color: #000080\">INFO</span> Sending metric for task aggregated_model_validation, round number <span style=\"color: #000080; font-weight: bold\">5</span>: dice_coef <span style=\"color: #000080; font-weight: bold\">0.4695041198884287</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\"><span style=\"color: #7f7f7f\">collaborator.py</span></a><span style=\"color: #7f7f7f\">:365</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740bf1ab20>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> Collaborator one is sending task results for aggregated_model_validation, round <span style=\"color: #000080; font-weight: bold\">5</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\"><span style=\"color: #7f7f7f\">aggregator.py</span></a><span style=\"color: #7f7f7f\">:428</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f748dc86040>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\">[17:49:25] </span><span style=\"color: #000080\">INFO</span> Sending metric for task train, round number <span style=\"color: #000080; font-weight: bold\">5</span>: soft_dice_loss <span style=\"color: #000080; font-weight: bold\">0.5579285025596619</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\"><span style=\"color: #7f7f7f\">collaborator.py</span></a><span style=\"color: #7f7f7f\">:365</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740be8ebb0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> Collaborator one is sending task results for train, round <span style=\"color: #000080; font-weight: bold\">5</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\"><span style=\"color: #7f7f7f\">aggregator.py</span></a><span style=\"color: #7f7f7f\">:428</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740be8ebb0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\">[17:49:28] </span><span style=\"color: #000080\">INFO</span> Sending metric for task locally_tuned_model_validation, round number <span style=\"color: #000080; font-weight: bold\">5</span>: dice_coef <span style=\"color: #000080; font-weight: bold\">0.477940563232668</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\"><span style=\"color: #7f7f7f\">collaborator.py</span></a><span style=\"color: #7f7f7f\">:365</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740be8e310>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> Collaborator one is sending task results for locally_tuned_model_validation, round <span style=\"color: #000080; font-weight: bold\">5</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\"><span style=\"color: #7f7f7f\">aggregator.py</span></a><span style=\"color: #7f7f7f\">:428</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740bf15910>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> All tasks completed on one for round <span style=\"color: #000080; font-weight: bold\">5</span><span style=\"color: #808000\">...</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\"><span style=\"color: #7f7f7f\">collaborator.py</span></a><span style=\"color: #7f7f7f\">:148</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740be8e280>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> Waiting for tasks<span style=\"color: #808000\">...</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\"><span style=\"color: #7f7f7f\">collaborator.py</span></a><span style=\"color: #7f7f7f\">:156</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740be2f8b0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> Sending tasks to collaborator two for round <span style=\"color: #000080; font-weight: bold\">5</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\"><span style=\"color: #7f7f7f\">aggregator.py</span></a><span style=\"color: #7f7f7f\">:276</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740bf154f0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> Received the following tasks: <span style=\"font-weight: bold\">[</span><span style=\"color: #008000\">'aggregated_model_validation'</span>, <span style=\"color: #008000\">'train'</span>, <span style=\"color: #008000\">'locally_tuned_model_validation'</span><span style=\"font-weight: bold\">]</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\"><span style=\"color: #7f7f7f\">collaborator.py</span></a><span style=\"color: #7f7f7f\">:144</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740be8e310>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\">[17:49:30] </span><span style=\"color: #000080\">INFO</span> Sending metric for task aggregated_model_validation, round number <span style=\"color: #000080; font-weight: bold\">5</span>: dice_coef <span style=\"color: #000080; font-weight: bold\">0.3998434495541357</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\"><span style=\"color: #7f7f7f\">collaborator.py</span></a><span style=\"color: #7f7f7f\">:365</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740be8e280>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> Collaborator two is sending task results for aggregated_model_validation, round <span style=\"color: #000080; font-weight: bold\">5</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\"><span style=\"color: #7f7f7f\">aggregator.py</span></a><span style=\"color: #7f7f7f\">:428</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f7420029be0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\">[17:49:52] </span><span style=\"color: #000080\">INFO</span> Sending metric for task train, round number <span style=\"color: #000080; font-weight: bold\">5</span>: soft_dice_loss <span style=\"color: #000080; font-weight: bold\">0.5507904291152954</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\"><span style=\"color: #7f7f7f\">collaborator.py</span></a><span style=\"color: #7f7f7f\">:365</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f742003f9a0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> Collaborator two is sending task results for train, round <span style=\"color: #000080; font-weight: bold\">5</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\"><span style=\"color: #7f7f7f\">aggregator.py</span></a><span style=\"color: #7f7f7f\">:428</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740be2fd00>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\">[17:49:54] </span><span style=\"color: #000080\">INFO</span> Sending metric for task locally_tuned_model_validation, round number <span style=\"color: #000080; font-weight: bold\">5</span>: dice_coef <span style=\"color: #000080; font-weight: bold\">0.4104092457602101</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\"><span style=\"color: #7f7f7f\">collaborator.py</span></a><span style=\"color: #7f7f7f\">:365</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740bd5ef40>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> Collaborator two is sending task results for locally_tuned_model_validation, round <span style=\"color: #000080; font-weight: bold\">5</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\"><span style=\"color: #7f7f7f\">aggregator.py</span></a><span style=\"color: #7f7f7f\">:428</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740be1ce20>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> train task metrics<span style=\"color: #808000\">...</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\"><span style=\"color: #7f7f7f\">aggregator.py</span></a><span style=\"color: #7f7f7f\">:694</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740bfe8ca0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> soft_dice_loss: <span style=\"color: #000080; font-weight: bold\">0.5544</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\"><span style=\"color: #7f7f7f\">aggregator.py</span></a><span style=\"color: #7f7f7f\">:742</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f7420015b20>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\">[17:50:00] </span><span style=\"color: #000080\">INFO</span> locally_tuned_model_validation task metrics<span style=\"color: #808000\">...</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\"><span style=\"color: #7f7f7f\">aggregator.py</span></a><span style=\"color: #7f7f7f\">:694</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740be36b20>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> dice_coef: <span style=\"color: #000080; font-weight: bold\">0.4442</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\"><span style=\"color: #7f7f7f\">aggregator.py</span></a><span style=\"color: #7f7f7f\">:742</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740bcf59a0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> aggregated_model_validation task metrics<span style=\"color: #808000\">...</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\"><span style=\"color: #7f7f7f\">aggregator.py</span></a><span style=\"color: #7f7f7f\">:694</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740bcf59a0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> dice_coef: <span style=\"color: #000080; font-weight: bold\">0.4347</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\"><span style=\"color: #7f7f7f\">aggregator.py</span></a><span style=\"color: #7f7f7f\">:742</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740bcf5dc0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> Saved the best model with score <span style=\"color: #000080; font-weight: bold\">0.434674</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\"><span style=\"color: #7f7f7f\">aggregator.py</span></a><span style=\"color: #7f7f7f\">:749</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740be369a0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\">[17:50:01] </span><span style=\"color: #000080\">INFO</span> Saving round <span style=\"color: #000080; font-weight: bold\">6</span> model<span style=\"color: #808000\">...</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\"><span style=\"color: #7f7f7f\">aggregator.py</span></a><span style=\"color: #7f7f7f\">:783</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740bcf59a0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> Starting round <span style=\"color: #000080; font-weight: bold\">6</span><span style=\"color: #808000\">...</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\"><span style=\"color: #7f7f7f\">aggregator.py</span></a><span style=\"color: #7f7f7f\">:791</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740bd94be0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> All tasks completed on two for round <span style=\"color: #000080; font-weight: bold\">5</span><span style=\"color: #808000\">...</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\"><span style=\"color: #7f7f7f\">collaborator.py</span></a><span style=\"color: #7f7f7f\">:148</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740bfff3a0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> Waiting for tasks<span style=\"color: #808000\">...</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\"><span style=\"color: #7f7f7f\">collaborator.py</span></a><span style=\"color: #7f7f7f\">:156</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f748dc865b0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> Sending tasks to collaborator one for round <span style=\"color: #000080; font-weight: bold\">6</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\"><span style=\"color: #7f7f7f\">aggregator.py</span></a><span style=\"color: #7f7f7f\">:276</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740bfff3a0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> Received the following tasks: <span style=\"font-weight: bold\">[</span><span style=\"color: #008000\">'aggregated_model_validation'</span>, <span style=\"color: #008000\">'train'</span>, <span style=\"color: #008000\">'locally_tuned_model_validation'</span><span style=\"font-weight: bold\">]</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\"><span style=\"color: #7f7f7f\">collaborator.py</span></a><span style=\"color: #7f7f7f\">:144</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740bfff3a0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\">[17:50:04] </span><span style=\"color: #000080\">INFO</span> Sending metric for task aggregated_model_validation, round number <span style=\"color: #000080; font-weight: bold\">6</span>: dice_coef <span style=\"color: #000080; font-weight: bold\">0.4847551199697679</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\"><span style=\"color: #7f7f7f\">collaborator.py</span></a><span style=\"color: #7f7f7f\">:365</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f742002cf40>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> Collaborator one is sending task results for aggregated_model_validation, round <span style=\"color: #000080; font-weight: bold\">6</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\"><span style=\"color: #7f7f7f\">aggregator.py</span></a><span style=\"color: #7f7f7f\">:428</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f742002c160>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\">[17:50:25] </span><span style=\"color: #000080\">INFO</span> Sending metric for task train, round number <span style=\"color: #000080; font-weight: bold\">6</span>: soft_dice_loss <span style=\"color: #000080; font-weight: bold\">0.549356997013092</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\"><span style=\"color: #7f7f7f\">collaborator.py</span></a><span style=\"color: #7f7f7f\">:365</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740bcf5d60>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> Collaborator one is sending task results for train, round <span style=\"color: #000080; font-weight: bold\">6</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\"><span style=\"color: #7f7f7f\">aggregator.py</span></a><span style=\"color: #7f7f7f\">:428</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740bcf5220>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\">[17:50:28] </span><span style=\"color: #000080\">INFO</span> Sending metric for task locally_tuned_model_validation, round number <span style=\"color: #000080; font-weight: bold\">6</span>: dice_coef <span style=\"color: #000080; font-weight: bold\">0.49405526922595117</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\"><span style=\"color: #7f7f7f\">collaborator.py</span></a><span style=\"color: #7f7f7f\">:365</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740bfe3250>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> Collaborator one is sending task results for locally_tuned_model_validation, round <span style=\"color: #000080; font-weight: bold\">6</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\"><span style=\"color: #7f7f7f\">aggregator.py</span></a><span style=\"color: #7f7f7f\">:428</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740bfe3670>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> All tasks completed on one for round <span style=\"color: #000080; font-weight: bold\">6</span><span style=\"color: #808000\">...</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\"><span style=\"color: #7f7f7f\">collaborator.py</span></a><span style=\"color: #7f7f7f\">:148</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740bfe3b50>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> Waiting for tasks<span style=\"color: #808000\">...</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\"><span style=\"color: #7f7f7f\">collaborator.py</span></a><span style=\"color: #7f7f7f\">:156</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740bfe3820>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> Sending tasks to collaborator two for round <span style=\"color: #000080; font-weight: bold\">6</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\"><span style=\"color: #7f7f7f\">aggregator.py</span></a><span style=\"color: #7f7f7f\">:276</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740bfe3310>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> Received the following tasks: <span style=\"font-weight: bold\">[</span><span style=\"color: #008000\">'aggregated_model_validation'</span>, <span style=\"color: #008000\">'train'</span>, <span style=\"color: #008000\">'locally_tuned_model_validation'</span><span style=\"font-weight: bold\">]</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\"><span style=\"color: #7f7f7f\">collaborator.py</span></a><span style=\"color: #7f7f7f\">:144</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740bfe3580>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\">[17:50:30] </span><span style=\"color: #000080\">INFO</span> Sending metric for task aggregated_model_validation, round number <span style=\"color: #000080; font-weight: bold\">6</span>: dice_coef <span style=\"color: #000080; font-weight: bold\">0.4081264974609498</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\"><span style=\"color: #7f7f7f\">collaborator.py</span></a><span style=\"color: #7f7f7f\">:365</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740bea51c0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> Collaborator two is sending task results for aggregated_model_validation, round <span style=\"color: #000080; font-weight: bold\">6</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\"><span style=\"color: #7f7f7f\">aggregator.py</span></a><span style=\"color: #7f7f7f\">:428</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740bea5730>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\">[17:50:52] </span><span style=\"color: #000080\">INFO</span> Sending metric for task train, round number <span style=\"color: #000080; font-weight: bold\">6</span>: soft_dice_loss <span style=\"color: #000080; font-weight: bold\">0.5499082803726196</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\"><span style=\"color: #7f7f7f\">collaborator.py</span></a><span style=\"color: #7f7f7f\">:365</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f73fb224880>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> Collaborator two is sending task results for train, round <span style=\"color: #000080; font-weight: bold\">6</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\"><span style=\"color: #7f7f7f\">aggregator.py</span></a><span style=\"color: #7f7f7f\">:428</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f73fb224790>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\">[17:50:55] </span><span style=\"color: #000080\">INFO</span> Sending metric for task locally_tuned_model_validation, round number <span style=\"color: #000080; font-weight: bold\">6</span>: dice_coef <span style=\"color: #000080; font-weight: bold\">0.4086666549405744</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\"><span style=\"color: #7f7f7f\">collaborator.py</span></a><span style=\"color: #7f7f7f\">:365</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740bf06ca0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> Collaborator two is sending task results for locally_tuned_model_validation, round <span style=\"color: #000080; font-weight: bold\">6</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\"><span style=\"color: #7f7f7f\">aggregator.py</span></a><span style=\"color: #7f7f7f\">:428</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740bf2f070>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> train task metrics<span style=\"color: #808000\">...</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\"><span style=\"color: #7f7f7f\">aggregator.py</span></a><span style=\"color: #7f7f7f\">:694</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f73fb224d90>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> soft_dice_loss: <span style=\"color: #000080; font-weight: bold\">0.5496</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\"><span style=\"color: #7f7f7f\">aggregator.py</span></a><span style=\"color: #7f7f7f\">:742</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740bf621c0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\">[17:51:00] </span><span style=\"color: #000080\">INFO</span> locally_tuned_model_validation task metrics<span style=\"color: #808000\">...</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\"><span style=\"color: #7f7f7f\">aggregator.py</span></a><span style=\"color: #7f7f7f\">:694</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740be872b0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> dice_coef: <span style=\"color: #000080; font-weight: bold\">0.4514</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\"><span style=\"color: #7f7f7f\">aggregator.py</span></a><span style=\"color: #7f7f7f\">:742</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740be361c0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> aggregated_model_validation task metrics<span style=\"color: #808000\">...</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\"><span style=\"color: #7f7f7f\">aggregator.py</span></a><span style=\"color: #7f7f7f\">:694</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740bea5220>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> dice_coef: <span style=\"color: #000080; font-weight: bold\">0.4464</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\"><span style=\"color: #7f7f7f\">aggregator.py</span></a><span style=\"color: #7f7f7f\">:742</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740ac917c0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> Saved the best model with score <span style=\"color: #000080; font-weight: bold\">0.446441</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\"><span style=\"color: #7f7f7f\">aggregator.py</span></a><span style=\"color: #7f7f7f\">:749</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740bea5220>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\">[17:51:01] </span><span style=\"color: #000080\">INFO</span> Saving round <span style=\"color: #000080; font-weight: bold\">7</span> model<span style=\"color: #808000\">...</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\"><span style=\"color: #7f7f7f\">aggregator.py</span></a><span style=\"color: #7f7f7f\">:783</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740be872b0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> Starting round <span style=\"color: #000080; font-weight: bold\">7</span><span style=\"color: #808000\">...</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\"><span style=\"color: #7f7f7f\">aggregator.py</span></a><span style=\"color: #7f7f7f\">:791</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740bf62970>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> All tasks completed on two for round <span style=\"color: #000080; font-weight: bold\">6</span><span style=\"color: #808000\">...</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\"><span style=\"color: #7f7f7f\">collaborator.py</span></a><span style=\"color: #7f7f7f\">:148</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740bea9f70>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\">[17:51:02] </span><span style=\"color: #000080\">INFO</span> Waiting for tasks<span style=\"color: #808000\">...</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\"><span style=\"color: #7f7f7f\">collaborator.py</span></a><span style=\"color: #7f7f7f\">:156</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740bfe3520>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> Sending tasks to collaborator one for round <span style=\"color: #000080; font-weight: bold\">7</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\"><span style=\"color: #7f7f7f\">aggregator.py</span></a><span style=\"color: #7f7f7f\">:276</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740bffcc10>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> Received the following tasks: <span style=\"font-weight: bold\">[</span><span style=\"color: #008000\">'aggregated_model_validation'</span>, <span style=\"color: #008000\">'train'</span>, <span style=\"color: #008000\">'locally_tuned_model_validation'</span><span style=\"font-weight: bold\">]</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\"><span style=\"color: #7f7f7f\">collaborator.py</span></a><span style=\"color: #7f7f7f\">:144</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740bfe3d90>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\">[17:51:04] </span><span style=\"color: #000080\">INFO</span> Sending metric for task aggregated_model_validation, round number <span style=\"color: #000080; font-weight: bold\">7</span>: dice_coef <span style=\"color: #000080; font-weight: bold\">0.48679458902728173</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\"><span style=\"color: #7f7f7f\">collaborator.py</span></a><span style=\"color: #7f7f7f\">:365</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f7420032a00>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> Collaborator one is sending task results for aggregated_model_validation, round <span style=\"color: #000080; font-weight: bold\">7</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\"><span style=\"color: #7f7f7f\">aggregator.py</span></a><span style=\"color: #7f7f7f\">:428</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740be36b20>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\">[17:51:25] </span><span style=\"color: #000080\">INFO</span> Sending metric for task train, round number <span style=\"color: #000080; font-weight: bold\">7</span>: soft_dice_loss <span style=\"color: #000080; font-weight: bold\">0.5552520751953125</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\"><span style=\"color: #7f7f7f\">collaborator.py</span></a><span style=\"color: #7f7f7f\">:365</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f7490c9e5b0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> Collaborator one is sending task results for train, round <span style=\"color: #000080; font-weight: bold\">7</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\"><span style=\"color: #7f7f7f\">aggregator.py</span></a><span style=\"color: #7f7f7f\">:428</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f748dc86d30>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\">[17:51:28] </span><span style=\"color: #000080\">INFO</span> Sending metric for task locally_tuned_model_validation, round number <span style=\"color: #000080; font-weight: bold\">7</span>: dice_coef <span style=\"color: #000080; font-weight: bold\">0.4828214414658085</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\"><span style=\"color: #7f7f7f\">collaborator.py</span></a><span style=\"color: #7f7f7f\">:365</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740be83160>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> Collaborator one is sending task results for locally_tuned_model_validation, round <span style=\"color: #000080; font-weight: bold\">7</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\"><span style=\"color: #7f7f7f\">aggregator.py</span></a><span style=\"color: #7f7f7f\">:428</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f742000e160>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> All tasks completed on one for round <span style=\"color: #000080; font-weight: bold\">7</span><span style=\"color: #808000\">...</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\"><span style=\"color: #7f7f7f\">collaborator.py</span></a><span style=\"color: #7f7f7f\">:148</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f742000ee80>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> Waiting for tasks<span style=\"color: #808000\">...</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\"><span style=\"color: #7f7f7f\">collaborator.py</span></a><span style=\"color: #7f7f7f\">:156</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f742000eeb0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> Sending tasks to collaborator two for round <span style=\"color: #000080; font-weight: bold\">7</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\"><span style=\"color: #7f7f7f\">aggregator.py</span></a><span style=\"color: #7f7f7f\">:276</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f748dc854f0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> Received the following tasks: <span style=\"font-weight: bold\">[</span><span style=\"color: #008000\">'aggregated_model_validation'</span>, <span style=\"color: #008000\">'train'</span>, <span style=\"color: #008000\">'locally_tuned_model_validation'</span><span style=\"font-weight: bold\">]</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\"><span style=\"color: #7f7f7f\">collaborator.py</span></a><span style=\"color: #7f7f7f\">:144</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f742000e280>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\">[17:51:31] </span><span style=\"color: #000080\">INFO</span> Sending metric for task aggregated_model_validation, round number <span style=\"color: #000080; font-weight: bold\">7</span>: dice_coef <span style=\"color: #000080; font-weight: bold\">0.41096355165204695</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\"><span style=\"color: #7f7f7f\">collaborator.py</span></a><span style=\"color: #7f7f7f\">:365</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740be8e730>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> Collaborator two is sending task results for aggregated_model_validation, round <span style=\"color: #000080; font-weight: bold\">7</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\"><span style=\"color: #7f7f7f\">aggregator.py</span></a><span style=\"color: #7f7f7f\">:428</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740be8e730>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\">[17:51:52] </span><span style=\"color: #000080\">INFO</span> Sending metric for task train, round number <span style=\"color: #000080; font-weight: bold\">7</span>: soft_dice_loss <span style=\"color: #000080; font-weight: bold\">0.5383422374725342</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\"><span style=\"color: #7f7f7f\">collaborator.py</span></a><span style=\"color: #7f7f7f\">:365</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740bfffbb0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> Collaborator two is sending task results for train, round <span style=\"color: #000080; font-weight: bold\">7</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\"><span style=\"color: #7f7f7f\">aggregator.py</span></a><span style=\"color: #7f7f7f\">:428</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740bd941c0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\">[17:51:55] </span><span style=\"color: #000080\">INFO</span> Sending metric for task locally_tuned_model_validation, round number <span style=\"color: #000080; font-weight: bold\">7</span>: dice_coef <span style=\"color: #000080; font-weight: bold\">0.4175524327062791</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\"><span style=\"color: #7f7f7f\">collaborator.py</span></a><span style=\"color: #7f7f7f\">:365</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740be87dc0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> Collaborator two is sending task results for locally_tuned_model_validation, round <span style=\"color: #000080; font-weight: bold\">7</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\"><span style=\"color: #7f7f7f\">aggregator.py</span></a><span style=\"color: #7f7f7f\">:428</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740be87dc0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> train task metrics<span style=\"color: #808000\">...</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\"><span style=\"color: #7f7f7f\">aggregator.py</span></a><span style=\"color: #7f7f7f\">:694</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740bea5ca0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> soft_dice_loss: <span style=\"color: #000080; font-weight: bold\">0.5468</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\"><span style=\"color: #7f7f7f\">aggregator.py</span></a><span style=\"color: #7f7f7f\">:742</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740bf06d30>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\">[17:52:01] </span><span style=\"color: #000080\">INFO</span> locally_tuned_model_validation task metrics<span style=\"color: #808000\">...</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\"><span style=\"color: #7f7f7f\">aggregator.py</span></a><span style=\"color: #7f7f7f\">:694</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f7490cc0be0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> dice_coef: <span style=\"color: #000080; font-weight: bold\">0.4502</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\"><span style=\"color: #7f7f7f\">aggregator.py</span></a><span style=\"color: #7f7f7f\">:742</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f7490cc0be0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> aggregated_model_validation task metrics<span style=\"color: #808000\">...</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\"><span style=\"color: #7f7f7f\">aggregator.py</span></a><span style=\"color: #7f7f7f\">:694</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740bcdcca0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> dice_coef: <span style=\"color: #000080; font-weight: bold\">0.4489</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\"><span style=\"color: #7f7f7f\">aggregator.py</span></a><span style=\"color: #7f7f7f\">:742</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740bcdcdc0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> Saved the best model with score <span style=\"color: #000080; font-weight: bold\">0.448879</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\"><span style=\"color: #7f7f7f\">aggregator.py</span></a><span style=\"color: #7f7f7f\">:749</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f748dc85d30>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\">[17:52:02] </span><span style=\"color: #000080\">INFO</span> Saving round <span style=\"color: #000080; font-weight: bold\">8</span> model<span style=\"color: #808000\">...</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\"><span style=\"color: #7f7f7f\">aggregator.py</span></a><span style=\"color: #7f7f7f\">:783</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740bfd5580>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> Starting round <span style=\"color: #000080; font-weight: bold\">8</span><span style=\"color: #808000\">...</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\"><span style=\"color: #7f7f7f\">aggregator.py</span></a><span style=\"color: #7f7f7f\">:791</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740be8e730>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> All tasks completed on two for round <span style=\"color: #000080; font-weight: bold\">7</span><span style=\"color: #808000\">...</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\"><span style=\"color: #7f7f7f\">collaborator.py</span></a><span style=\"color: #7f7f7f\">:148</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f73fb224580>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> Waiting for tasks<span style=\"color: #808000\">...</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\"><span style=\"color: #7f7f7f\">collaborator.py</span></a><span style=\"color: #7f7f7f\">:156</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f7490d3f3d0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> Sending tasks to collaborator one for round <span style=\"color: #000080; font-weight: bold\">8</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\"><span style=\"color: #7f7f7f\">aggregator.py</span></a><span style=\"color: #7f7f7f\">:276</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740bd5c340>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> Received the following tasks: <span style=\"font-weight: bold\">[</span><span style=\"color: #008000\">'aggregated_model_validation'</span>, <span style=\"color: #008000\">'train'</span>, <span style=\"color: #008000\">'locally_tuned_model_validation'</span><span style=\"font-weight: bold\">]</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\"><span style=\"color: #7f7f7f\">collaborator.py</span></a><span style=\"color: #7f7f7f\">:144</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f73fb224580>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\">[17:52:05] </span><span style=\"color: #000080\">INFO</span> Sending metric for task aggregated_model_validation, round number <span style=\"color: #000080; font-weight: bold\">8</span>: dice_coef <span style=\"color: #000080; font-weight: bold\">0.49110506234630463</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\"><span style=\"color: #7f7f7f\">collaborator.py</span></a><span style=\"color: #7f7f7f\">:365</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740bd5c340>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> Collaborator one is sending task results for aggregated_model_validation, round <span style=\"color: #000080; font-weight: bold\">8</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\"><span style=\"color: #7f7f7f\">aggregator.py</span></a><span style=\"color: #7f7f7f\">:428</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740be36490>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\">[17:52:26] </span><span style=\"color: #000080\">INFO</span> Sending metric for task train, round number <span style=\"color: #000080; font-weight: bold\">8</span>: soft_dice_loss <span style=\"color: #000080; font-weight: bold\">0.5618945360183716</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\"><span style=\"color: #7f7f7f\">collaborator.py</span></a><span style=\"color: #7f7f7f\">:365</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740b4ba550>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> Collaborator one is sending task results for train, round <span style=\"color: #000080; font-weight: bold\">8</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\"><span style=\"color: #7f7f7f\">aggregator.py</span></a><span style=\"color: #7f7f7f\">:428</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740b4ba220>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\">[17:52:29] </span><span style=\"color: #000080\">INFO</span> Sending metric for task locally_tuned_model_validation, round number <span style=\"color: #000080; font-weight: bold\">8</span>: dice_coef <span style=\"color: #000080; font-weight: bold\">0.46602303943326395</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\"><span style=\"color: #7f7f7f\">collaborator.py</span></a><span style=\"color: #7f7f7f\">:365</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740bd9e130>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> Collaborator one is sending task results for locally_tuned_model_validation, round <span style=\"color: #000080; font-weight: bold\">8</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\"><span style=\"color: #7f7f7f\">aggregator.py</span></a><span style=\"color: #7f7f7f\">:428</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740bd9e070>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> All tasks completed on one for round <span style=\"color: #000080; font-weight: bold\">8</span><span style=\"color: #808000\">...</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\"><span style=\"color: #7f7f7f\">collaborator.py</span></a><span style=\"color: #7f7f7f\">:148</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740bd9ef70>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> Waiting for tasks<span style=\"color: #808000\">...</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\"><span style=\"color: #7f7f7f\">collaborator.py</span></a><span style=\"color: #7f7f7f\">:156</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740bd9eee0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> Sending tasks to collaborator two for round <span style=\"color: #000080; font-weight: bold\">8</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\"><span style=\"color: #7f7f7f\">aggregator.py</span></a><span style=\"color: #7f7f7f\">:276</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740bd9e6a0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> Received the following tasks: <span style=\"font-weight: bold\">[</span><span style=\"color: #008000\">'aggregated_model_validation'</span>, <span style=\"color: #008000\">'train'</span>, <span style=\"color: #008000\">'locally_tuned_model_validation'</span><span style=\"font-weight: bold\">]</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\"><span style=\"color: #7f7f7f\">collaborator.py</span></a><span style=\"color: #7f7f7f\">:144</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740bd9ef70>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\">[17:52:32] </span><span style=\"color: #000080\">INFO</span> Sending metric for task aggregated_model_validation, round number <span style=\"color: #000080; font-weight: bold\">8</span>: dice_coef <span style=\"color: #000080; font-weight: bold\">0.41813477200846516</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\"><span style=\"color: #7f7f7f\">collaborator.py</span></a><span style=\"color: #7f7f7f\">:365</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740bf2fee0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> Collaborator two is sending task results for aggregated_model_validation, round <span style=\"color: #000080; font-weight: bold\">8</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\"><span style=\"color: #7f7f7f\">aggregator.py</span></a><span style=\"color: #7f7f7f\">:428</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740bf2fd30>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\">[17:52:54] </span><span style=\"color: #000080\">INFO</span> Sending metric for task train, round number <span style=\"color: #000080; font-weight: bold\">8</span>: soft_dice_loss <span style=\"color: #000080; font-weight: bold\">0.534250020980835</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\"><span style=\"color: #7f7f7f\">collaborator.py</span></a><span style=\"color: #7f7f7f\">:365</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f7490c9ecd0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> Collaborator two is sending task results for train, round <span style=\"color: #000080; font-weight: bold\">8</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\"><span style=\"color: #7f7f7f\">aggregator.py</span></a><span style=\"color: #7f7f7f\">:428</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740bf06070>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\">[17:52:56] </span><span style=\"color: #000080\">INFO</span> Sending metric for task locally_tuned_model_validation, round number <span style=\"color: #000080; font-weight: bold\">8</span>: dice_coef <span style=\"color: #000080; font-weight: bold\">0.41231230381996403</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\"><span style=\"color: #7f7f7f\">collaborator.py</span></a><span style=\"color: #7f7f7f\">:365</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740bf62520>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> Collaborator two is sending task results for locally_tuned_model_validation, round <span style=\"color: #000080; font-weight: bold\">8</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\"><span style=\"color: #7f7f7f\">aggregator.py</span></a><span style=\"color: #7f7f7f\">:428</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740bf62a00>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> train task metrics<span style=\"color: #808000\">...</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\"><span style=\"color: #7f7f7f\">aggregator.py</span></a><span style=\"color: #7f7f7f\">:694</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740bf62070>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> soft_dice_loss: <span style=\"color: #000080; font-weight: bold\">0.5481</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\"><span style=\"color: #7f7f7f\">aggregator.py</span></a><span style=\"color: #7f7f7f\">:742</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740bf62520>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\">[17:53:02] </span><span style=\"color: #000080\">INFO</span> locally_tuned_model_validation task metrics<span style=\"color: #808000\">...</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\"><span style=\"color: #7f7f7f\">aggregator.py</span></a><span style=\"color: #7f7f7f\">:694</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740bef3ee0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> dice_coef: <span style=\"color: #000080; font-weight: bold\">0.4392</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\"><span style=\"color: #7f7f7f\">aggregator.py</span></a><span style=\"color: #7f7f7f\">:742</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740bee4c10>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> aggregated_model_validation task metrics<span style=\"color: #808000\">...</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\"><span style=\"color: #7f7f7f\">aggregator.py</span></a><span style=\"color: #7f7f7f\">:694</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740bef3550>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> dice_coef: <span style=\"color: #000080; font-weight: bold\">0.4546</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\"><span style=\"color: #7f7f7f\">aggregator.py</span></a><span style=\"color: #7f7f7f\">:742</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740be87610>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> Saved the best model with score <span style=\"color: #000080; font-weight: bold\">0.454620</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\"><span style=\"color: #7f7f7f\">aggregator.py</span></a><span style=\"color: #7f7f7f\">:749</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740bef3b20>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\">[17:53:03] </span><span style=\"color: #000080\">INFO</span> Saving round <span style=\"color: #000080; font-weight: bold\">9</span> model<span style=\"color: #808000\">...</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\"><span style=\"color: #7f7f7f\">aggregator.py</span></a><span style=\"color: #7f7f7f\">:783</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740bee4c10>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> Starting round <span style=\"color: #000080; font-weight: bold\">9</span><span style=\"color: #808000\">...</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\"><span style=\"color: #7f7f7f\">aggregator.py</span></a><span style=\"color: #7f7f7f\">:791</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740ad616a0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> All tasks completed on two for round <span style=\"color: #000080; font-weight: bold\">8</span><span style=\"color: #808000\">...</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\"><span style=\"color: #7f7f7f\">collaborator.py</span></a><span style=\"color: #7f7f7f\">:148</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f742003f400>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> Waiting for tasks<span style=\"color: #808000\">...</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\"><span style=\"color: #7f7f7f\">collaborator.py</span></a><span style=\"color: #7f7f7f\">:156</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740bdb9fd0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> Sending tasks to collaborator one for round <span style=\"color: #000080; font-weight: bold\">9</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\"><span style=\"color: #7f7f7f\">aggregator.py</span></a><span style=\"color: #7f7f7f\">:276</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f748dc85e20>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> Received the following tasks: <span style=\"font-weight: bold\">[</span><span style=\"color: #008000\">'aggregated_model_validation'</span>, <span style=\"color: #008000\">'train'</span>, <span style=\"color: #008000\">'locally_tuned_model_validation'</span><span style=\"font-weight: bold\">]</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\"><span style=\"color: #7f7f7f\">collaborator.py</span></a><span style=\"color: #7f7f7f\">:144</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740bf14280>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\">[17:53:06] </span><span style=\"color: #000080\">INFO</span> Sending metric for task aggregated_model_validation, round number <span style=\"color: #000080; font-weight: bold\">9</span>: dice_coef <span style=\"color: #000080; font-weight: bold\">0.47473019169222924</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\"><span style=\"color: #7f7f7f\">collaborator.py</span></a><span style=\"color: #7f7f7f\">:365</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f73fb224e20>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> Collaborator one is sending task results for aggregated_model_validation, round <span style=\"color: #000080; font-weight: bold\">9</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\"><span style=\"color: #7f7f7f\">aggregator.py</span></a><span style=\"color: #7f7f7f\">:428</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740be5b160>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\">[17:53:28] </span><span style=\"color: #000080\">INFO</span> Sending metric for task train, round number <span style=\"color: #000080; font-weight: bold\">9</span>: soft_dice_loss <span style=\"color: #000080; font-weight: bold\">0.5441964268684387</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\"><span style=\"color: #7f7f7f\">collaborator.py</span></a><span style=\"color: #7f7f7f\">:365</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740bf15280>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> Collaborator one is sending task results for train, round <span style=\"color: #000080; font-weight: bold\">9</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\"><span style=\"color: #7f7f7f\">aggregator.py</span></a><span style=\"color: #7f7f7f\">:428</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740bd5c940>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\">[17:53:30] </span><span style=\"color: #000080\">INFO</span> Sending metric for task locally_tuned_model_validation, round number <span style=\"color: #000080; font-weight: bold\">9</span>: dice_coef <span style=\"color: #000080; font-weight: bold\">0.49820235467726187</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\"><span style=\"color: #7f7f7f\">collaborator.py</span></a><span style=\"color: #7f7f7f\">:365</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740bf76cd0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> Collaborator one is sending task results for locally_tuned_model_validation, round <span style=\"color: #000080; font-weight: bold\">9</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\"><span style=\"color: #7f7f7f\">aggregator.py</span></a><span style=\"color: #7f7f7f\">:428</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740bdac8b0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> All tasks completed on one for round <span style=\"color: #000080; font-weight: bold\">9</span><span style=\"color: #808000\">...</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\"><span style=\"color: #7f7f7f\">collaborator.py</span></a><span style=\"color: #7f7f7f\">:148</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740b5375b0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> Waiting for tasks<span style=\"color: #808000\">...</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\"><span style=\"color: #7f7f7f\">collaborator.py</span></a><span style=\"color: #7f7f7f\">:156</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740bcbb130>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> Sending tasks to collaborator two for round <span style=\"color: #000080; font-weight: bold\">9</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\"><span style=\"color: #7f7f7f\">aggregator.py</span></a><span style=\"color: #7f7f7f\">:276</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740bcbb730>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> Received the following tasks: <span style=\"font-weight: bold\">[</span><span style=\"color: #008000\">'aggregated_model_validation'</span>, <span style=\"color: #008000\">'train'</span>, <span style=\"color: #008000\">'locally_tuned_model_validation'</span><span style=\"font-weight: bold\">]</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\"><span style=\"color: #7f7f7f\">collaborator.py</span></a><span style=\"color: #7f7f7f\">:144</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740bcbb730>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\">[17:53:33] </span><span style=\"color: #000080\">INFO</span> Sending metric for task aggregated_model_validation, round number <span style=\"color: #000080; font-weight: bold\">9</span>: dice_coef <span style=\"color: #000080; font-weight: bold\">0.3963722812552606</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\"><span style=\"color: #7f7f7f\">collaborator.py</span></a><span style=\"color: #7f7f7f\">:365</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740bea5670>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> Collaborator two is sending task results for aggregated_model_validation, round <span style=\"color: #000080; font-weight: bold\">9</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\"><span style=\"color: #7f7f7f\">aggregator.py</span></a><span style=\"color: #7f7f7f\">:428</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740b4bddc0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\">[17:53:55] </span><span style=\"color: #000080\">INFO</span> Sending metric for task train, round number <span style=\"color: #000080; font-weight: bold\">9</span>: soft_dice_loss <span style=\"color: #000080; font-weight: bold\">0.5431901216506958</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\"><span style=\"color: #7f7f7f\">collaborator.py</span></a><span style=\"color: #7f7f7f\">:365</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740bf14c10>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> Collaborator two is sending task results for train, round <span style=\"color: #000080; font-weight: bold\">9</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\"><span style=\"color: #7f7f7f\">aggregator.py</span></a><span style=\"color: #7f7f7f\">:428</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f748dc85100>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\">[17:53:58] </span><span style=\"color: #000080\">INFO</span> Sending metric for task locally_tuned_model_validation, round number <span style=\"color: #000080; font-weight: bold\">9</span>: dice_coef <span style=\"color: #000080; font-weight: bold\">0.4178659223741101</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\"><span style=\"color: #7f7f7f\">collaborator.py</span></a><span style=\"color: #7f7f7f\">:365</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740ad61d30>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> Collaborator two is sending task results for locally_tuned_model_validation, round <span style=\"color: #000080; font-weight: bold\">9</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\"><span style=\"color: #7f7f7f\">aggregator.py</span></a><span style=\"color: #7f7f7f\">:428</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740bee4a30>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> train task metrics<span style=\"color: #808000\">...</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\"><span style=\"color: #7f7f7f\">aggregator.py</span></a><span style=\"color: #7f7f7f\">:694</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740bdc6040>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> soft_dice_loss: <span style=\"color: #000080; font-weight: bold\">0.5437</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\"><span style=\"color: #7f7f7f\">aggregator.py</span></a><span style=\"color: #7f7f7f\">:742</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740ad61550>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\">[17:54:04] </span><span style=\"color: #000080\">INFO</span> locally_tuned_model_validation task metrics<span style=\"color: #808000\">...</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\"><span style=\"color: #7f7f7f\">aggregator.py</span></a><span style=\"color: #7f7f7f\">:694</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740acbfe20>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> dice_coef: <span style=\"color: #000080; font-weight: bold\">0.4580</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\"><span style=\"color: #7f7f7f\">aggregator.py</span></a><span style=\"color: #7f7f7f\">:742</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740bef3eb0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> aggregated_model_validation task metrics<span style=\"color: #808000\">...</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\"><span style=\"color: #7f7f7f\">aggregator.py</span></a><span style=\"color: #7f7f7f\">:694</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740b501040>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> dice_coef: <span style=\"color: #000080; font-weight: bold\">0.4356</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\"><span style=\"color: #7f7f7f\">aggregator.py</span></a><span style=\"color: #7f7f7f\">:742</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740bef3a30>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> Saving round <span style=\"color: #000080; font-weight: bold\">10</span> model<span style=\"color: #808000\">...</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\"><span style=\"color: #7f7f7f\">aggregator.py</span></a><span style=\"color: #7f7f7f\">:783</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740bef3c70>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> Starting round <span style=\"color: #000080; font-weight: bold\">10</span><span style=\"color: #808000\">...</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\"><span style=\"color: #7f7f7f\">aggregator.py</span></a><span style=\"color: #7f7f7f\">:791</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740bea5670>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> All tasks completed on two for round <span style=\"color: #000080; font-weight: bold\">9</span><span style=\"color: #808000\">...</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\"><span style=\"color: #7f7f7f\">collaborator.py</span></a><span style=\"color: #7f7f7f\">:148</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740bea26d0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> Waiting for tasks<span style=\"color: #808000\">...</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\"><span style=\"color: #7f7f7f\">collaborator.py</span></a><span style=\"color: #7f7f7f\">:156</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740bea26d0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> Sending tasks to collaborator one for round <span style=\"color: #000080; font-weight: bold\">10</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\"><span style=\"color: #7f7f7f\">aggregator.py</span></a><span style=\"color: #7f7f7f\">:276</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740be4c4c0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> Received the following tasks: <span style=\"font-weight: bold\">[</span><span style=\"color: #008000\">'aggregated_model_validation'</span>, <span style=\"color: #008000\">'train'</span>, <span style=\"color: #008000\">'locally_tuned_model_validation'</span><span style=\"font-weight: bold\">]</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\"><span style=\"color: #7f7f7f\">collaborator.py</span></a><span style=\"color: #7f7f7f\">:144</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740bf14c10>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\">[17:54:07] </span><span style=\"color: #000080\">INFO</span> Sending metric for task aggregated_model_validation, round number <span style=\"color: #000080; font-weight: bold\">10</span>: dice_coef <span style=\"color: #000080; font-weight: bold\">0.49992890127243533</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\"><span style=\"color: #7f7f7f\">collaborator.py</span></a><span style=\"color: #7f7f7f\">:365</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740bcbbbb0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> Collaborator one is sending task results for aggregated_model_validation, round <span style=\"color: #000080; font-weight: bold\">10</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\"><span style=\"color: #7f7f7f\">aggregator.py</span></a><span style=\"color: #7f7f7f\">:428</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740be87730>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\">[17:54:29] </span><span style=\"color: #000080\">INFO</span> Sending metric for task train, round number <span style=\"color: #000080; font-weight: bold\">10</span>: soft_dice_loss <span style=\"color: #000080; font-weight: bold\">0.5380003452301025</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\"><span style=\"color: #7f7f7f\">collaborator.py</span></a><span style=\"color: #7f7f7f\">:365</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740be87d30>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> Collaborator one is sending task results for train, round <span style=\"color: #000080; font-weight: bold\">10</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\"><span style=\"color: #7f7f7f\">aggregator.py</span></a><span style=\"color: #7f7f7f\">:428</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740bea9bb0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\">[17:54:32] </span><span style=\"color: #000080\">INFO</span> Sending metric for task locally_tuned_model_validation, round number <span style=\"color: #000080; font-weight: bold\">10</span>: dice_coef <span style=\"color: #000080; font-weight: bold\">0.46978584989424677</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\"><span style=\"color: #7f7f7f\">collaborator.py</span></a><span style=\"color: #7f7f7f\">:365</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740bf62070>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> Collaborator one is sending task results for locally_tuned_model_validation, round <span style=\"color: #000080; font-weight: bold\">10</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\"><span style=\"color: #7f7f7f\">aggregator.py</span></a><span style=\"color: #7f7f7f\">:428</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740bee4ee0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> All tasks completed on one for round <span style=\"color: #000080; font-weight: bold\">10</span><span style=\"color: #808000\">...</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\"><span style=\"color: #7f7f7f\">collaborator.py</span></a><span style=\"color: #7f7f7f\">:148</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740bee4250>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> Waiting for tasks<span style=\"color: #808000\">...</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\"><span style=\"color: #7f7f7f\">collaborator.py</span></a><span style=\"color: #7f7f7f\">:156</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740bee4580>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> Sending tasks to collaborator two for round <span style=\"color: #000080; font-weight: bold\">10</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\"><span style=\"color: #7f7f7f\">aggregator.py</span></a><span style=\"color: #7f7f7f\">:276</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740bee4a30>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> Received the following tasks: <span style=\"font-weight: bold\">[</span><span style=\"color: #008000\">'aggregated_model_validation'</span>, <span style=\"color: #008000\">'train'</span>, <span style=\"color: #008000\">'locally_tuned_model_validation'</span><span style=\"font-weight: bold\">]</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\"><span style=\"color: #7f7f7f\">collaborator.py</span></a><span style=\"color: #7f7f7f\">:144</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740bf62490>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\">[17:54:35] </span><span style=\"color: #000080\">INFO</span> Sending metric for task aggregated_model_validation, round number <span style=\"color: #000080; font-weight: bold\">10</span>: dice_coef <span style=\"color: #000080; font-weight: bold\">0.42785738312428995</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\"><span style=\"color: #7f7f7f\">collaborator.py</span></a><span style=\"color: #7f7f7f\">:365</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740bf76340>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> Collaborator two is sending task results for aggregated_model_validation, round <span style=\"color: #000080; font-weight: bold\">10</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\"><span style=\"color: #7f7f7f\">aggregator.py</span></a><span style=\"color: #7f7f7f\">:428</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740acbffd0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "ename": "KeyboardInterrupt", + "evalue": "", + "output_type": "error", + "traceback": [ + "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m", + "\u001b[0;31mKeyboardInterrupt\u001b[0m Traceback (most recent call last)", + "\u001b[0;32m<ipython-input-13-92562d71b5a6>\u001b[0m in \u001b[0;36m<module>\u001b[0;34m\u001b[0m\n\u001b[1;32m 1\u001b[0m \u001b[0;31m# Run experiment, return trained FederatedModel\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m----> 2\u001b[0;31m final_fl_model = fx.run_experiment(\n\u001b[0m\u001b[1;32m 3\u001b[0m collaborators, override_config={'aggregator.settings.rounds_to_train': 30})\n", + "\u001b[0;32m~/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/native/native.py\u001b[0m in \u001b[0;36mrun_experiment\u001b[0;34m(collaborator_dict, override_config)\u001b[0m\n\u001b[1;32m 262\u001b[0m \u001b[0mcollaborator\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mcollaborators\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0mcol\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 263\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 264\u001b[0;31m \u001b[0mcollaborator\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mrun_simulation\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m 265\u001b[0m \u001b[0mmodel_states\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0mcol\u001b[0m\u001b[0;34m]\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mcollaborator\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mtask_runner\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mget_tensor_dict\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mwith_opt_vars\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0;32mTrue\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 266\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n", + "\u001b[0;32m~/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\u001b[0m in \u001b[0;36mrun_simulation\u001b[0;34m(self)\u001b[0m\n\u001b[1;32m 145\u001b[0m 'Received the following tasks: {}'.format(tasks))\n\u001b[1;32m 146\u001b[0m \u001b[0;32mfor\u001b[0m \u001b[0mtask\u001b[0m \u001b[0;32min\u001b[0m \u001b[0mtasks\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 147\u001b[0;31m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mdo_task\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mtask\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mround_number\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m 148\u001b[0m self.logger.info(\n\u001b[1;32m 149\u001b[0m 'All tasks completed on {} for round {}...'.format(\n", + "\u001b[0;32m~/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\u001b[0m in \u001b[0;36mdo_task\u001b[0;34m(self, task, round_number)\u001b[0m\n\u001b[1;32m 196\u001b[0m \u001b[0;31m# now we have whatever the model needs to do the task\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 197\u001b[0m \u001b[0mfunc\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mgetattr\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mtask_runner\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mfunc_name\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 198\u001b[0;31m global_output_tensor_dict, local_output_tensor_dict = func(\n\u001b[0m\u001b[1;32m 199\u001b[0m \u001b[0mcol_name\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mcollaborator_name\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 200\u001b[0m \u001b[0mround_num\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mround_number\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n", + "\u001b[0;32m~/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/federated/task/runner_pt.py\u001b[0m in \u001b[0;36mtrain_batches\u001b[0;34m(self, col_name, round_num, input_tensor_dict, num_batches, use_tqdm, *kwargs)\u001b[0m\n\u001b[1;32m 159\u001b[0m \u001b[0mloss\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mbackward\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 160\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0moptimizer\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mstep\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 161\u001b[0;31m \u001b[0mlosses\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mappend\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mloss\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mdetach\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mcpu\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mnumpy\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m 162\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 163\u001b[0m \u001b[0;31m# Output metric tensors (scalar)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n", + "\u001b[0;31mKeyboardInterrupt\u001b[0m: " + ] + } + ], "source": [ "# Run experiment, return trained FederatedModel\n", "final_fl_model = fx.run_experiment(\n", @@ -550,7 +7077,7 @@ "name": "python", "nbconvert_exporter": "python", "pygments_lexer": "ipython3", - "version": "3.6.12" + "version": "3.8.8" } }, "nbformat": 4, diff --git a/openfl-tutorials/Federated_Pytorch_MNIST_Tutorial.ipynb b/openfl-tutorials/Federated_Pytorch_MNIST_Tutorial.ipynb index d42ef87e5f..8189afe719 100644 --- a/openfl-tutorials/Federated_Pytorch_MNIST_Tutorial.ipynb +++ b/openfl-tutorials/Federated_Pytorch_MNIST_Tutorial.ipynb @@ -9,17 +9,128 @@ }, { "cell_type": "code", - "execution_count": null, + "execution_count": 1, "metadata": {}, - "outputs": [], + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "Processing /home/itrushkin/repos/openfl-fork\n", + "Requirement already satisfied: torch in /home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages (1.8.1)\n", + "Requirement already satisfied: torchvision in /home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages (0.9.1)\n", + "Requirement already satisfied: Click>=7.0 in /home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages (from openfl==1.0.1) (7.1.2)\n", + "Requirement already satisfied: PyYAML>=5.4.1 in /home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages (from openfl==1.0.1) (5.4.1)\n", + "Requirement already satisfied: numpy in /home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages (from openfl==1.0.1) (1.18.5)\n", + "Requirement already satisfied: pandas in /home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages (from openfl==1.0.1) (1.2.3)\n", + "Requirement already satisfied: protobuf in /home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages (from openfl==1.0.1) (3.15.6)\n", + "Requirement already satisfied: grpcio==1.30.0 in /home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages (from openfl==1.0.1) (1.30.0)\n", + "Requirement already satisfied: grpcio-tools==1.30.0 in /home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages (from openfl==1.0.1) (1.30.0)\n", + "Requirement already satisfied: rich==9.1.0 in /home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages (from openfl==1.0.1) (9.1.0)\n", + "Requirement already satisfied: tqdm in /home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages (from openfl==1.0.1) (4.59.0)\n", + "Requirement already satisfied: scikit-learn in /home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages (from openfl==1.0.1) (0.24.1)\n", + "Requirement already satisfied: docker in /home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages (from openfl==1.0.1) (4.4.4)\n", + "Requirement already satisfied: jupyter in /home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages (from openfl==1.0.1) (1.0.0)\n", + "Requirement already satisfied: ipykernel in /home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages (from openfl==1.0.1) (5.5.0)\n", + "Requirement already satisfied: flatten_json in /home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages (from openfl==1.0.1) (0.1.13)\n", + "Requirement already satisfied: cryptography>=3.4.6 in /home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages (from openfl==1.0.1) (3.4.6)\n", + "Requirement already satisfied: six>=1.5.2 in /home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages (from grpcio==1.30.0->openfl==1.0.1) (1.15.0)\n", + "Requirement already satisfied: typing-extensions<4.0.0,>=3.7.4 in /home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages (from rich==9.1.0->openfl==1.0.1) (3.7.4.3)\n", + "Requirement already satisfied: pygments<3.0.0,>=2.6.0 in /home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages (from rich==9.1.0->openfl==1.0.1) (2.8.1)\n", + "Requirement already satisfied: colorama<0.5.0,>=0.4.0 in /home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages (from rich==9.1.0->openfl==1.0.1) (0.4.4)\n", + "Requirement already satisfied: commonmark<0.10.0,>=0.9.0 in /home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages (from rich==9.1.0->openfl==1.0.1) (0.9.1)\n", + "Requirement already satisfied: cffi>=1.12 in /home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages (from cryptography>=3.4.6->openfl==1.0.1) (1.14.5)\n", + "Requirement already satisfied: pycparser in /home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages (from cffi>=1.12->cryptography>=3.4.6->openfl==1.0.1) (2.20)\n", + "Requirement already satisfied: pillow>=4.1.1 in /home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages (from torchvision) (8.1.2)\n", + "Requirement already satisfied: websocket-client>=0.32.0 in /home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages (from docker->openfl==1.0.1) (0.58.0)\n", + "Requirement already satisfied: requests!=2.18.0,>=2.14.2 in /home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages (from docker->openfl==1.0.1) (2.25.1)\n", + "Requirement already satisfied: chardet<5,>=3.0.2 in /home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages (from requests!=2.18.0,>=2.14.2->docker->openfl==1.0.1) (4.0.0)\n", + "Requirement already satisfied: certifi>=2017.4.17 in /home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages (from requests!=2.18.0,>=2.14.2->docker->openfl==1.0.1) (2020.12.5)\n", + "Requirement already satisfied: idna<3,>=2.5 in /home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages (from requests!=2.18.0,>=2.14.2->docker->openfl==1.0.1) (2.10)\n", + "Requirement already satisfied: urllib3<1.27,>=1.21.1 in /home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages (from requests!=2.18.0,>=2.14.2->docker->openfl==1.0.1) (1.26.4)\n", + "Requirement already satisfied: tornado>=4.2 in /home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages (from ipykernel->openfl==1.0.1) (6.1)\n", + "Requirement already satisfied: jupyter-client in /home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages (from ipykernel->openfl==1.0.1) (6.1.12)\n", + "Requirement already satisfied: traitlets>=4.1.0 in /home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages (from ipykernel->openfl==1.0.1) (5.0.5)\n", + "Requirement already satisfied: ipython>=5.0.0 in /home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages (from ipykernel->openfl==1.0.1) (7.21.0)\n", + "Requirement already satisfied: decorator in /home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages (from ipython>=5.0.0->ipykernel->openfl==1.0.1) (4.4.2)\n", + "Requirement already satisfied: setuptools>=18.5 in /home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages (from ipython>=5.0.0->ipykernel->openfl==1.0.1) (54.1.2)\n", + "Requirement already satisfied: jedi>=0.16 in /home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages (from ipython>=5.0.0->ipykernel->openfl==1.0.1) (0.18.0)\n", + "Requirement already satisfied: backcall in /home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages (from ipython>=5.0.0->ipykernel->openfl==1.0.1) (0.2.0)\n", + "Requirement already satisfied: pickleshare in /home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages (from ipython>=5.0.0->ipykernel->openfl==1.0.1) (0.7.5)\n", + "Requirement already satisfied: pexpect>4.3 in /home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages (from ipython>=5.0.0->ipykernel->openfl==1.0.1) (4.8.0)\n", + "Requirement already satisfied: prompt-toolkit!=3.0.0,!=3.0.1,<3.1.0,>=2.0.0 in /home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages (from ipython>=5.0.0->ipykernel->openfl==1.0.1) (3.0.18)\n", + "Requirement already satisfied: parso<0.9.0,>=0.8.0 in /home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages (from jedi>=0.16->ipython>=5.0.0->ipykernel->openfl==1.0.1) (0.8.1)\n", + "Requirement already satisfied: ptyprocess>=0.5 in /home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages (from pexpect>4.3->ipython>=5.0.0->ipykernel->openfl==1.0.1) (0.7.0)\n", + "Requirement already satisfied: wcwidth in /home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages (from prompt-toolkit!=3.0.0,!=3.0.1,<3.1.0,>=2.0.0->ipython>=5.0.0->ipykernel->openfl==1.0.1) (0.2.5)\n", + "Requirement already satisfied: ipython-genutils in /home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages (from traitlets>=4.1.0->ipykernel->openfl==1.0.1) (0.2.0)\n", + "Requirement already satisfied: notebook in /home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages (from jupyter->openfl==1.0.1) (6.3.0)\n", + "Requirement already satisfied: jupyter-console in /home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages (from jupyter->openfl==1.0.1) (6.3.0)\n", + "Requirement already satisfied: ipywidgets in /home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages (from jupyter->openfl==1.0.1) (7.6.3)\n", + "Requirement already satisfied: nbconvert in /home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages (from jupyter->openfl==1.0.1) (6.0.7)\n", + "Requirement already satisfied: qtconsole in /home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages (from jupyter->openfl==1.0.1) (5.0.3)\n", + "Requirement already satisfied: jupyterlab-widgets>=1.0.0 in /home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages (from ipywidgets->jupyter->openfl==1.0.1) (1.0.0)\n", + "Requirement already satisfied: nbformat>=4.2.0 in /home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages (from ipywidgets->jupyter->openfl==1.0.1) (5.1.2)\n", + "Requirement already satisfied: widgetsnbextension~=3.5.0 in /home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages (from ipywidgets->jupyter->openfl==1.0.1) (3.5.1)\n", + "Requirement already satisfied: jupyter-core in /home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages (from nbformat>=4.2.0->ipywidgets->jupyter->openfl==1.0.1) (4.7.1)\n", + "Requirement already satisfied: jsonschema!=2.5.0,>=2.4 in /home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages (from nbformat>=4.2.0->ipywidgets->jupyter->openfl==1.0.1) (3.2.0)\n", + "Requirement already satisfied: pyrsistent>=0.14.0 in /home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages (from jsonschema!=2.5.0,>=2.4->nbformat>=4.2.0->ipywidgets->jupyter->openfl==1.0.1) (0.17.3)\n", + "Requirement already satisfied: attrs>=17.4.0 in /home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages (from jsonschema!=2.5.0,>=2.4->nbformat>=4.2.0->ipywidgets->jupyter->openfl==1.0.1) (20.3.0)\n" + ] + }, + { + "name": "stdout", + "output_type": "stream", + "text": [ + "Requirement already satisfied: jinja2 in /home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages (from notebook->jupyter->openfl==1.0.1) (2.11.3)\n", + "Requirement already satisfied: Send2Trash>=1.5.0 in /home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages (from notebook->jupyter->openfl==1.0.1) (1.5.0)\n", + "Requirement already satisfied: terminado>=0.8.3 in /home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages (from notebook->jupyter->openfl==1.0.1) (0.9.3)\n", + "Requirement already satisfied: pyzmq>=17 in /home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages (from notebook->jupyter->openfl==1.0.1) (22.0.3)\n", + "Requirement already satisfied: prometheus-client in /home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages (from notebook->jupyter->openfl==1.0.1) (0.9.0)\n", + "Requirement already satisfied: argon2-cffi in /home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages (from notebook->jupyter->openfl==1.0.1) (20.1.0)\n", + "Requirement already satisfied: python-dateutil>=2.1 in /home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages (from jupyter-client->ipykernel->openfl==1.0.1) (2.8.1)\n", + "Requirement already satisfied: MarkupSafe>=0.23 in /home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages (from jinja2->notebook->jupyter->openfl==1.0.1) (1.1.1)\n", + "Requirement already satisfied: entrypoints>=0.2.2 in /home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages (from nbconvert->jupyter->openfl==1.0.1) (0.3)\n", + "Requirement already satisfied: nbclient<0.6.0,>=0.5.0 in /home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages (from nbconvert->jupyter->openfl==1.0.1) (0.5.3)\n", + "Requirement already satisfied: defusedxml in /home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages (from nbconvert->jupyter->openfl==1.0.1) (0.7.1)\n", + "Requirement already satisfied: bleach in /home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages (from nbconvert->jupyter->openfl==1.0.1) (3.3.0)\n", + "Requirement already satisfied: jupyterlab-pygments in /home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages (from nbconvert->jupyter->openfl==1.0.1) (0.1.2)\n", + "Requirement already satisfied: testpath in /home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages (from nbconvert->jupyter->openfl==1.0.1) (0.4.4)\n", + "Requirement already satisfied: mistune<2,>=0.8.1 in /home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages (from nbconvert->jupyter->openfl==1.0.1) (0.8.4)\n", + "Requirement already satisfied: pandocfilters>=1.4.1 in /home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages (from nbconvert->jupyter->openfl==1.0.1) (1.4.3)\n", + "Requirement already satisfied: nest-asyncio in /home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages (from nbclient<0.6.0,>=0.5.0->nbconvert->jupyter->openfl==1.0.1) (1.5.1)\n", + "Requirement already satisfied: async-generator in /home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages (from nbclient<0.6.0,>=0.5.0->nbconvert->jupyter->openfl==1.0.1) (1.10)\n", + "Requirement already satisfied: packaging in /home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages (from bleach->nbconvert->jupyter->openfl==1.0.1) (20.9)\n", + "Requirement already satisfied: webencodings in /home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages (from bleach->nbconvert->jupyter->openfl==1.0.1) (0.5.1)\n", + "Requirement already satisfied: pyparsing>=2.0.2 in /home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages (from packaging->bleach->nbconvert->jupyter->openfl==1.0.1) (2.4.7)\n", + "Requirement already satisfied: pytz>=2017.3 in /home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages (from pandas->openfl==1.0.1) (2021.1)\n", + "Requirement already satisfied: qtpy in /home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages (from qtconsole->jupyter->openfl==1.0.1) (1.9.0)\n", + "Requirement already satisfied: scipy>=0.19.1 in /home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages (from scikit-learn->openfl==1.0.1) (1.6.1)\n", + "Requirement already satisfied: threadpoolctl>=2.0.0 in /home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages (from scikit-learn->openfl==1.0.1) (2.1.0)\n", + "Requirement already satisfied: joblib>=0.11 in /home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages (from scikit-learn->openfl==1.0.1) (1.0.1)\n", + "Building wheels for collected packages: openfl\n", + " Building wheel for openfl (setup.py) ... \u001b[?25ldone\n", + "\u001b[?25h Created wheel for openfl: filename=openfl-1.0.1-py3-none-any.whl size=4329637 sha256=3c7a0cc242c9c32265142a0c4b0efa4c76cad2db2da12cbcc929c0805919c5bf\n", + " Stored in directory: /home/itrushkin/.cache/pip/wheels/ed/ad/82/f26baafab4be32d052fb76a6f56c9a76ce6903cc3eed7bb394\n", + "Successfully built openfl\n", + "Installing collected packages: openfl\n", + " Attempting uninstall: openfl\n", + " Found existing installation: openfl 1.0.1\n", + " Uninstalling openfl-1.0.1:\n", + " Successfully uninstalled openfl-1.0.1\n", + "Successfully installed openfl-1.0.1\n", + "\u001b[33mWARNING: You are using pip version 21.0.1; however, version 21.1 is available.\n", + "You should consider upgrading via the '/home/itrushkin/.virtualenvs/openfl_research/bin/python -m pip install --upgrade pip' command.\u001b[0m\n" + ] + } + ], "source": [ "#Install dependencies if not already installed\n", - "!pip install torch torchvision" + "!pip install torch torchvision .." ] }, { "cell_type": "code", - "execution_count": null, + "execution_count": 2, "metadata": {}, "outputs": [], "source": [ @@ -44,9 +155,108 @@ }, { "cell_type": "code", - "execution_count": null, + "execution_count": 3, "metadata": {}, - "outputs": [], + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "Creating Workspace Directories\n", + "Creating Workspace Templates\n", + "Successfully installed packages from /home/itrushkin/.local/workspace/requirements.txt.\n", + "\n", + "New workspace directory structure:\n", + "workspace\n", + "├── plan\n", + "│ ├── defaults\n", + "│ │ ├── data_loader.yaml\n", + "│ │ ├── network.yaml\n", + "│ │ ├── tasks_keras.yaml\n", + "│ │ ├── collaborator.yaml\n", + "│ │ ├── assigner.yaml\n", + "│ │ ├── task_runner.yaml\n", + "│ │ ├── defaults\n", + "│ │ ├── tasks_tensorflow.yaml\n", + "│ │ ├── aggregator.yaml\n", + "│ │ ├── tasks_torch.yaml\n", + "│ │ └── tasks_fast_estimator.yaml\n", + "│ ├── plan.yaml\n", + "│ ├── data.yaml\n", + "│ └── cols.yaml\n", + "├── logs\n", + "├── agg_to_col_two_signed_cert.zip\n", + "├── data\n", + "│ └── MNIST\n", + "│ ├── raw\n", + "│ └── processed\n", + "├── cert\n", + "│ ├── ca\n", + "│ │ ├── signing-ca.crt\n", + "│ │ ├── root-ca.crt\n", + "│ │ ├── signing-ca\n", + "│ │ ├── signing-ca.csr\n", + "│ │ └── root-ca\n", + "│ ├── server\n", + "│ │ ├── agg_nnlicv611.inn.intel.com.csr\n", + "│ │ ├── agg_nnlicv611.inn.intel.com.key\n", + "│ │ └── agg_nnlicv611.inn.intel.com.crt\n", + "│ ├── client\n", + "│ │ ├── col_two.crt\n", + "│ │ ├── col_one.crt\n", + "│ │ ├── col_one.key\n", + "│ │ └── col_two.key\n", + "│ ├── cert_chain.crt\n", + "│ ├── col_one.csr\n", + "│ └── col_two.csr\n", + "├── code\n", + "│ ├── __init__.py\n", + "│ ├── mnist_utils.py\n", + "│ ├── ptmnist_inmemory.py\n", + "│ └── pt_cnn.py\n", + "├── final_pytorch_model\n", + "├── .workspace\n", + "├── requirements.txt\n", + "├── agg_to_col_one_signed_cert.zip\n", + "└── save\n", + " ├── torch_cnn_mnist_init.pbuf\n", + " ├── torch_cnn_mnist_best.pbuf\n", + " └── torch_cnn_mnist_last.pbuf\n", + "\n", + "15 directories, 39 files\n", + "Setting Up Certificate Authority...\n", + "\n", + "1. Create Root CA\n", + "1.1 Create Directories\n", + "1.2 Create Database\n", + "1.3 Create CA Request and Certificate\n", + "2. Create Signing Certificate\n", + "2.1 Create Directories\n", + "2.2 Create Database\n", + "2.3 Create Signing Certificate CSR\n", + "2.4 Sign Signing Certificate CSR\n", + "3 Create Certificate Chain\n", + "\n", + "Done.\n", + "Creating AGGREGATOR certificate key pair with following settings: CN=\u001b[31mnnlicv611.inn.intel.com\u001b[0m, SAN=\u001b[31mDNS:nnlicv611.inn.intel.com\u001b[0m\n", + " Writing AGGREGATOR certificate key pair to: \u001b[32mcert/server\u001b[0m\n", + "The CSR Hash for file \u001b[32mserver/agg_nnlicv611.inn.intel.com.csr\u001b[0m = \u001b[31m269da4be6e73f7a492ef3ff73a71eae14eaef3a228427542d0226e29e279222711f8bad6f0e0b62686c19d723c2f46eb\u001b[0m\n", + " Signing AGGREGATOR certificate\n", + "Creating COLLABORATOR certificate key pair with following settings: CN=\u001b[31mone\u001b[0m, SAN=\u001b[31mDNS:one\u001b[0m\n", + " Moving COLLABORATOR certificate to: \u001b[32mcert/col_one\u001b[0m\n", + "The CSR Hash for file \u001b[32mcol_one.csr\u001b[0m = \u001b[31m2c758f295d8b7a321b7af8fcd15c097e327cd53dbafe3393cbe8a6d71f728bf2f1175859e3f90477ace2b022c9b34d6a\u001b[0m\n", + " Signing COLLABORATOR certificate\n", + "\n", + "Registering \u001b[32mone\u001b[0m in \u001b[32mplan/cols.yaml\u001b[0m\n", + "Creating COLLABORATOR certificate key pair with following settings: CN=\u001b[31mtwo\u001b[0m, SAN=\u001b[31mDNS:two\u001b[0m\n", + " Moving COLLABORATOR certificate to: \u001b[32mcert/col_two\u001b[0m\n", + "The CSR Hash for file \u001b[32mcol_two.csr\u001b[0m = \u001b[31m7ff429d1e8baa9f2a6cfb9b26624ea9bbf78c3cd4bca2990061488dabffdd165ded3d51af1e2334469ede41b11af42b8\u001b[0m\n", + " Signing COLLABORATOR certificate\n", + "\n", + "Registering \u001b[32mtwo\u001b[0m in \u001b[32mplan/cols.yaml\u001b[0m\n" + ] + } + ], "source": [ "#Setup default workspace, logging, etc.\n", "fx.init('torch_cnn_mnist')" @@ -61,9 +271,24 @@ }, { "cell_type": "code", - "execution_count": null, + "execution_count": 4, "metadata": {}, - "outputs": [], + "outputs": [ + { + "name": "stderr", + "output_type": "stream", + "text": [ + "/home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/torchvision/datasets/mnist.py:64: UserWarning: train_data has been renamed data\n", + " warnings.warn(\"train_data has been renamed data\")\n", + "/home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/torchvision/datasets/mnist.py:54: UserWarning: train_labels has been renamed targets\n", + " warnings.warn(\"train_labels has been renamed targets\")\n", + "/home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/torchvision/datasets/mnist.py:69: UserWarning: test_data has been renamed data\n", + " warnings.warn(\"test_data has been renamed data\")\n", + "/home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/torchvision/datasets/mnist.py:59: UserWarning: test_labels has been renamed targets\n", + " warnings.warn(\"test_labels has been renamed targets\")\n" + ] + } + ], "source": [ "def one_hot(labels, classes):\n", " return np.eye(classes)[labels]\n", @@ -88,7 +313,7 @@ }, { "cell_type": "code", - "execution_count": null, + "execution_count": 5, "metadata": {}, "outputs": [], "source": [ @@ -126,9 +351,23 @@ }, { "cell_type": "code", - "execution_count": null, + "execution_count": 6, "metadata": {}, - "outputs": [], + "outputs": [ + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\">[17:00:50] </span><span style=\"color: #800000\">WARNING</span> tried to remove tensor: __opt_state_needed not present in the tensor dict <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/utilities/utils.py\"><span style=\"color: #7f7f7f\">utils.py</span></a><span style=\"color: #7f7f7f\">:85</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f1e44b9a3d0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + } + ], "source": [ "\n", "#Create a federated model using the pytorch class, lambda optimizer function, and loss function\n", @@ -144,9 +383,36 @@ }, { "cell_type": "code", - "execution_count": null, + "execution_count": 7, "metadata": {}, - "outputs": [], + "outputs": [ + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #800000\">WARNING</span> tried to remove tensor: __opt_state_needed not present in the tensor dict <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/utilities/utils.py\"><span style=\"color: #7f7f7f\">utils.py</span></a><span style=\"color: #7f7f7f\">:85</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f1e5cf39340>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #800000\">WARNING</span> tried to remove tensor: __opt_state_needed not present in the tensor dict <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/utilities/utils.py\"><span style=\"color: #7f7f7f\">utils.py</span></a><span style=\"color: #7f7f7f\">:85</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f1e44b44e20>" + ] + }, + "metadata": {}, + "output_type": "display_data" + } + ], "source": [ "collaborator_models = fl_model.setup(num_collaborators=2)\n", "collaborators = {'one':collaborator_models[0],'two':collaborator_models[1]}#, 'three':collaborator_models[2]}" @@ -154,9 +420,25 @@ }, { "cell_type": "code", - "execution_count": null, + "execution_count": 8, "metadata": {}, - "outputs": [], + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "Original training data size: 60000\n", + "Original validation data size: 10000\n", + "\n", + "Collaborator one's training data size: 30000\n", + "Collaborator one's validation data size: 5000\n", + "\n", + "Collaborator two's training data size: 30000\n", + "Collaborator two's validation data size: 5000\n", + "\n" + ] + } + ], "source": [ "#Original MNIST dataset\n", "print(f'Original training data size: {len(train_images)}')\n", @@ -184,13 +466,78 @@ }, { "cell_type": "code", - "execution_count": null, + "execution_count": 9, "metadata": {}, - "outputs": [], + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "{\n", + " \"aggregator.settings.best_state_path\": \"save/torch_cnn_mnist_best.pbuf\",\n", + " \"aggregator.settings.db_store_rounds\": 1,\n", + " \"aggregator.settings.init_state_path\": \"save/torch_cnn_mnist_init.pbuf\",\n", + " \"aggregator.settings.last_state_path\": \"save/torch_cnn_mnist_last.pbuf\",\n", + " \"aggregator.settings.rounds_to_train\": 10,\n", + " \"aggregator.template\": \"openfl.component.Aggregator\",\n", + " \"assigner.settings.task_groups\": [\n", + " {\n", + " \"name\": \"train_and_validate\",\n", + " \"percentage\": 1.0,\n", + " \"tasks\": [\n", + " \"aggregated_model_validation\",\n", + " \"train\",\n", + " \"locally_tuned_model_validation\"\n", + " ]\n", + " }\n", + " ],\n", + " \"assigner.template\": \"openfl.component.RandomGroupedAssigner\",\n", + " \"collaborator.settings.db_store_rounds\": 1,\n", + " \"collaborator.settings.delta_updates\": false,\n", + " \"collaborator.settings.opt_treatment\": \"RESET\",\n", + " \"collaborator.template\": \"openfl.component.Collaborator\",\n", + " \"data_loader.settings.batch_size\": 256,\n", + " \"data_loader.settings.collaborator_count\": 2,\n", + " \"data_loader.settings.data_group_name\": \"mnist\",\n", + " \"data_loader.template\": \"code.ptmnist_inmemory.PyTorchMNISTInMemory\",\n", + " \"network.settings.agg_addr\": \"nnlicv611.inn.intel.com\",\n", + " \"network.settings.agg_port\": 56041,\n", + " \"network.settings.cert_folder\": \"cert\",\n", + " \"network.settings.client_reconnect_interval\": 5,\n", + " \"network.settings.disable_client_auth\": false,\n", + " \"network.settings.disable_tls\": false,\n", + " \"network.settings.hash_salt\": \"auto\",\n", + " \"network.template\": \"openfl.federation.Network\",\n", + " \"task_runner.settings\": {},\n", + " \"task_runner.template\": \"code.pt_cnn.PyTorchCNN\",\n", + " \"tasks.aggregated_model_validation.function\": \"validate\",\n", + " \"tasks.aggregated_model_validation.kwargs\": {\n", + " \"apply\": \"global\",\n", + " \"metrics\": [\n", + " \"acc\"\n", + " ]\n", + " },\n", + " \"tasks.locally_tuned_model_validation.function\": \"validate\",\n", + " \"tasks.locally_tuned_model_validation.kwargs\": {\n", + " \"apply\": \"local\",\n", + " \"metrics\": [\n", + " \"acc\"\n", + " ]\n", + " },\n", + " \"tasks.settings\": {},\n", + " \"tasks.train.function\": \"train_batches\",\n", + " \"tasks.train.kwargs\": {\n", + " \"metrics\": [\n", + " \"loss\"\n", + " ]\n", + " }\n", + "}\n" + ] + } + ], "source": [ " #Get the current values of the plan. Each of these can be overridden\n", - "import json\n", - "print(json.dumps(fx.get_plan(), indent=4, sort_keys=True))" + "print(fx.get_plan())" ] }, { @@ -202,9 +549,2168 @@ }, { "cell_type": "code", - "execution_count": null, + "execution_count": 10, "metadata": {}, - "outputs": [], + "outputs": [ + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> Updating aggregator.settings.rounds_to_train to <span style=\"color: #000080; font-weight: bold\">5</span><span style=\"color: #808000\">...</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/native/native.py\"><span style=\"color: #7f7f7f\">native.py</span></a><span style=\"color: #7f7f7f\">:82</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f1e44b07040>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> Building <span style=\"color: #800000\">🡆</span> Object <span style=\"color: #800000\">NoCompressionPipeline</span> from <span style=\"color: #800000\">openfl.pipelines</span> Module. <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/federated/plan/plan.py\"><span style=\"color: #7f7f7f\">plan.py</span></a><span style=\"color: #7f7f7f\">:162</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f1e44b076d0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> Settings <span style=\"color: #800000\">🡆</span> <span style=\"font-weight: bold\">{}</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/federated/plan/plan.py\"><span style=\"color: #7f7f7f\">plan.py</span></a><span style=\"color: #7f7f7f\">:165</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f1e44b077c0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> Override <span style=\"color: #800000\">🡆</span> <span style=\"font-weight: bold\">{}</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/federated/plan/plan.py\"><span style=\"color: #7f7f7f\">plan.py</span></a><span style=\"color: #7f7f7f\">:167</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f1e44b07940>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> Creating Initial Weights File 🠆 save/torch_cnn_mnist_init.pbuf <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/native/native.py\"><span style=\"color: #7f7f7f\">native.py</span></a><span style=\"color: #7f7f7f\">:240</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f1d58ece5e0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> Starting Experiment<span style=\"color: #808000\">...</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/native/native.py\"><span style=\"color: #7f7f7f\">native.py</span></a><span style=\"color: #7f7f7f\">:244</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f1d58ece550>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> Building <span style=\"color: #800000\">🡆</span> Object <span style=\"color: #800000\">RandomGroupedAssigner</span> from <span style=\"color: #800000\">openfl.component</span> Module. <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/federated/plan/plan.py\"><span style=\"color: #7f7f7f\">plan.py</span></a><span style=\"color: #7f7f7f\">:162</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f1d58ece730>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> Settings <span style=\"color: #800000\">🡆</span> <span style=\"font-weight: bold\">{</span><span style=\"color: #008000\">'task_groups'</span>: <span style=\"font-weight: bold\">[{</span><span style=\"color: #008000\">'name'</span>: <span style=\"color: #008000\">'train_and_validate'</span>, <span style=\"color: #008000\">'percentage'</span>: <span style=\"color: #000080; font-weight: bold\">1.0</span>, <span style=\"color: #008000\">'tasks'</span>: <span style=\"font-weight: bold\">[</span><span style=\"color: #008000\">'aggregated_model_validation'</span>, <span style=\"color: #008000\">'train'</span>, <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/federated/plan/plan.py\"><span style=\"color: #7f7f7f\">plan.py</span></a><span style=\"color: #7f7f7f\">:165</span>\n", + " <span style=\"color: #008000\">'locally_tuned_model_validation'</span><span style=\"font-weight: bold\">]}]</span>, <span style=\"color: #008000\">'authorized_cols'</span>: <span style=\"font-weight: bold\">[</span><span style=\"color: #008000\">'one'</span>, <span style=\"color: #008000\">'two'</span><span style=\"font-weight: bold\">]</span>, <span style=\"color: #008000\">'rounds_to_train'</span>: <span style=\"color: #000080; font-weight: bold\">10</span>, <span style=\"color: #008000\">'tasks'</span>: <span style=\"font-weight: bold\">{</span><span style=\"color: #008000\">'settings'</span>: <span style=\"font-weight: bold\">{}</span>, \n", + " <span style=\"color: #008000\">'defaults'</span>: <span style=\"color: #008000\">'plan/defaults/tasks_torch.yaml'</span>, <span style=\"color: #008000\">'locally_tuned_model_validation'</span>: <span style=\"font-weight: bold\">{</span><span style=\"color: #008000\">'function'</span>: <span style=\"color: #008000\">'validate'</span>, <span style=\"color: #008000\">'kwargs'</span>: <span style=\"font-weight: bold\">{</span><span style=\"color: #008000\">'apply'</span>: \n", + " <span style=\"color: #008000\">'local'</span>, <span style=\"color: #008000\">'metrics'</span>: <span style=\"font-weight: bold\">[</span><span style=\"color: #008000\">'acc'</span><span style=\"font-weight: bold\">]}}</span>, <span style=\"color: #008000\">'train'</span>: <span style=\"font-weight: bold\">{</span><span style=\"color: #008000\">'function'</span>: <span style=\"color: #008000\">'train_batches'</span>, <span style=\"color: #008000\">'kwargs'</span>: <span style=\"font-weight: bold\">{</span><span style=\"color: #008000\">'metrics'</span>: <span style=\"font-weight: bold\">[</span><span style=\"color: #008000\">'loss'</span><span style=\"font-weight: bold\">]}}</span>, \n", + " <span style=\"color: #008000\">'aggregated_model_validation'</span>: <span style=\"font-weight: bold\">{</span><span style=\"color: #008000\">'function'</span>: <span style=\"color: #008000\">'validate'</span>, <span style=\"color: #008000\">'kwargs'</span>: <span style=\"font-weight: bold\">{</span><span style=\"color: #008000\">'apply'</span>: <span style=\"color: #008000\">'global'</span>, <span style=\"color: #008000\">'metrics'</span>: <span style=\"font-weight: bold\">[</span><span style=\"color: #008000\">'acc'</span><span style=\"font-weight: bold\">]}}}}</span> \n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f1d58ece7f0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> Override <span style=\"color: #800000\">🡆</span> <span style=\"font-weight: bold\">{</span><span style=\"color: #008000\">'defaults'</span>: <span style=\"color: #008000\">'plan/defaults/assigner.yaml'</span><span style=\"font-weight: bold\">}</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/federated/plan/plan.py\"><span style=\"color: #7f7f7f\">plan.py</span></a><span style=\"color: #7f7f7f\">:167</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f1d58ece400>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> Building <span style=\"color: #800000\">🡆</span> Object <span style=\"color: #800000\">Aggregator</span> from <span style=\"color: #800000\">openfl.component</span> Module. <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/federated/plan/plan.py\"><span style=\"color: #7f7f7f\">plan.py</span></a><span style=\"color: #7f7f7f\">:162</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f1d58e837c0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> Settings <span style=\"color: #800000\">🡆</span> <span style=\"font-weight: bold\">{</span><span style=\"color: #008000\">'init_state_path'</span>: <span style=\"color: #008000\">'save/torch_cnn_mnist_init.pbuf'</span>, <span style=\"color: #008000\">'last_state_path'</span>: <span style=\"color: #008000\">'save/torch_cnn_mnist_last.pbuf'</span>, <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/federated/plan/plan.py\"><span style=\"color: #7f7f7f\">plan.py</span></a><span style=\"color: #7f7f7f\">:165</span>\n", + " <span style=\"color: #008000\">'rounds_to_train'</span>: <span style=\"color: #000080; font-weight: bold\">5</span>, <span style=\"color: #008000\">'best_state_path'</span>: <span style=\"color: #008000\">'save/torch_cnn_mnist_best.pbuf'</span>, <span style=\"color: #008000\">'db_store_rounds'</span>: <span style=\"color: #000080; font-weight: bold\">1</span>, <span style=\"color: #008000\">'aggregator_uuid'</span>: \n", + " <span style=\"color: #008000\">'aggregator_plan.yaml_acf3e6ef'</span>, <span style=\"color: #008000\">'federation_uuid'</span>: <span style=\"color: #008000\">'plan.yaml_acf3e6ef'</span>, <span style=\"color: #008000\">'authorized_cols'</span>: <span style=\"font-weight: bold\">[</span><span style=\"color: #008000\">'one'</span>, <span style=\"color: #008000\">'two'</span><span style=\"font-weight: bold\">]</span>, <span style=\"color: #008000\">'assigner'</span>: \n", + " <span style=\"font-weight: bold\"><</span><span style=\"color: #ff00ff; font-weight: bold\">openfl.component.assigner.random_grouped_assigner.RandomGroupedAssigner</span><span style=\"color: #000000\"> object at </span><span style=\"color: #000080; font-weight: bold\">0x7f1e44b07550</span><span style=\"font-weight: bold\">>}</span> \n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f1d58e83850>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> Override <span style=\"color: #800000\">🡆</span> <span style=\"font-weight: bold\">{</span><span style=\"color: #008000\">'defaults'</span>: <span style=\"color: #008000\">'plan/defaults/aggregator.yaml'</span><span style=\"font-weight: bold\">}</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/federated/plan/plan.py\"><span style=\"color: #7f7f7f\">plan.py</span></a><span style=\"color: #7f7f7f\">:167</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f1d58e836d0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> Building <span style=\"color: #800000\">🡆</span> Object <span style=\"color: #800000\">Collaborator</span> from <span style=\"color: #800000\">openfl.component</span> Module. <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/federated/plan/plan.py\"><span style=\"color: #7f7f7f\">plan.py</span></a><span style=\"color: #7f7f7f\">:162</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f1d58e30af0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> Settings <span style=\"color: #800000\">🡆</span> <span style=\"font-weight: bold\">{</span><span style=\"color: #008000\">'opt_treatment'</span>: <span style=\"color: #008000\">'RESET'</span>, <span style=\"color: #008000\">'delta_updates'</span>: <span style=\"color: #ff0000; font-style: italic\">False</span>, <span style=\"color: #008000\">'db_store_rounds'</span>: <span style=\"color: #000080; font-weight: bold\">1</span>, <span style=\"color: #008000\">'collaborator_name'</span>: <span style=\"color: #008000\">'one'</span>, <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/federated/plan/plan.py\"><span style=\"color: #7f7f7f\">plan.py</span></a><span style=\"color: #7f7f7f\">:165</span>\n", + " <span style=\"color: #008000\">'aggregator_uuid'</span>: <span style=\"color: #008000\">'aggregator_plan.yaml_acf3e6ef'</span>, <span style=\"color: #008000\">'federation_uuid'</span>: <span style=\"color: #008000\">'plan.yaml_acf3e6ef'</span>, <span style=\"color: #008000\">'task_runner'</span>: \n", + " <span style=\"font-weight: bold\"><</span><span style=\"color: #ff00ff; font-weight: bold\">openfl.federated.task.fl_model.FederatedModel</span><span style=\"color: #000000\"> object at </span><span style=\"color: #000080; font-weight: bold\">0x7f1e44b9a5b0</span><span style=\"font-weight: bold\">></span>, <span style=\"color: #008000\">'tensor_pipe'</span>: \n", + " <span style=\"font-weight: bold\"><</span><span style=\"color: #ff00ff; font-weight: bold\">openfl.pipelines.no_compression_pipeline.NoCompressionPipeline</span><span style=\"color: #000000\"> object at </span><span style=\"color: #000080; font-weight: bold\">0x7f1e44b07be0</span><span style=\"font-weight: bold\">></span>, <span style=\"color: #008000\">'task_config'</span>: <span style=\"font-weight: bold\">{</span><span style=\"color: #008000\">'settings'</span>: <span style=\"font-weight: bold\">{}</span>, \n", + " <span style=\"color: #008000\">'defaults'</span>: <span style=\"color: #008000\">'plan/defaults/tasks_torch.yaml'</span>, <span style=\"color: #008000\">'locally_tuned_model_validation'</span>: <span style=\"font-weight: bold\">{</span><span style=\"color: #008000\">'function'</span>: <span style=\"color: #008000\">'validate'</span>, <span style=\"color: #008000\">'kwargs'</span>: <span style=\"font-weight: bold\">{</span><span style=\"color: #008000\">'apply'</span>: \n", + " <span style=\"color: #008000\">'local'</span>, <span style=\"color: #008000\">'metrics'</span>: <span style=\"font-weight: bold\">[</span><span style=\"color: #008000\">'acc'</span><span style=\"font-weight: bold\">]}}</span>, <span style=\"color: #008000\">'train'</span>: <span style=\"font-weight: bold\">{</span><span style=\"color: #008000\">'function'</span>: <span style=\"color: #008000\">'train_batches'</span>, <span style=\"color: #008000\">'kwargs'</span>: <span style=\"font-weight: bold\">{</span><span style=\"color: #008000\">'metrics'</span>: <span style=\"font-weight: bold\">[</span><span style=\"color: #008000\">'loss'</span><span style=\"font-weight: bold\">]}}</span>, \n", + " <span style=\"color: #008000\">'aggregated_model_validation'</span>: <span style=\"font-weight: bold\">{</span><span style=\"color: #008000\">'function'</span>: <span style=\"color: #008000\">'validate'</span>, <span style=\"color: #008000\">'kwargs'</span>: <span style=\"font-weight: bold\">{</span><span style=\"color: #008000\">'apply'</span>: <span style=\"color: #008000\">'global'</span>, <span style=\"color: #008000\">'metrics'</span>: <span style=\"font-weight: bold\">[</span><span style=\"color: #008000\">'acc'</span><span style=\"font-weight: bold\">]}}}</span>, <span style=\"color: #008000\">'client'</span>: \n", + " <span style=\"font-weight: bold\"><</span><span style=\"color: #ff00ff; font-weight: bold\">openfl.component.aggregator.aggregator.Aggregator</span><span style=\"color: #000000\"> object at </span><span style=\"color: #000080; font-weight: bold\">0x7f1d58ece520</span><span style=\"font-weight: bold\">>}</span> \n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f1d58e30a60>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> Override <span style=\"color: #800000\">🡆</span> <span style=\"font-weight: bold\">{</span><span style=\"color: #008000\">'defaults'</span>: <span style=\"color: #008000\">'plan/defaults/collaborator.yaml'</span><span style=\"font-weight: bold\">}</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/federated/plan/plan.py\"><span style=\"color: #7f7f7f\">plan.py</span></a><span style=\"color: #7f7f7f\">:167</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f1d58e302e0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> Building <span style=\"color: #800000\">🡆</span> Object <span style=\"color: #800000\">Collaborator</span> from <span style=\"color: #800000\">openfl.component</span> Module. <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/federated/plan/plan.py\"><span style=\"color: #7f7f7f\">plan.py</span></a><span style=\"color: #7f7f7f\">:162</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f1d58e30f70>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> Settings <span style=\"color: #800000\">🡆</span> <span style=\"font-weight: bold\">{</span><span style=\"color: #008000\">'opt_treatment'</span>: <span style=\"color: #008000\">'RESET'</span>, <span style=\"color: #008000\">'delta_updates'</span>: <span style=\"color: #ff0000; font-style: italic\">False</span>, <span style=\"color: #008000\">'db_store_rounds'</span>: <span style=\"color: #000080; font-weight: bold\">1</span>, <span style=\"color: #008000\">'collaborator_name'</span>: <span style=\"color: #008000\">'two'</span>, <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/federated/plan/plan.py\"><span style=\"color: #7f7f7f\">plan.py</span></a><span style=\"color: #7f7f7f\">:165</span>\n", + " <span style=\"color: #008000\">'aggregator_uuid'</span>: <span style=\"color: #008000\">'aggregator_plan.yaml_acf3e6ef'</span>, <span style=\"color: #008000\">'federation_uuid'</span>: <span style=\"color: #008000\">'plan.yaml_acf3e6ef'</span>, <span style=\"color: #008000\">'task_runner'</span>: \n", + " <span style=\"font-weight: bold\"><</span><span style=\"color: #ff00ff; font-weight: bold\">openfl.federated.task.fl_model.FederatedModel</span><span style=\"color: #000000\"> object at </span><span style=\"color: #000080; font-weight: bold\">0x7f1e44b9ac40</span><span style=\"font-weight: bold\">></span>, <span style=\"color: #008000\">'tensor_pipe'</span>: \n", + " <span style=\"font-weight: bold\"><</span><span style=\"color: #ff00ff; font-weight: bold\">openfl.pipelines.no_compression_pipeline.NoCompressionPipeline</span><span style=\"color: #000000\"> object at </span><span style=\"color: #000080; font-weight: bold\">0x7f1e44b07be0</span><span style=\"font-weight: bold\">></span>, <span style=\"color: #008000\">'task_config'</span>: <span style=\"font-weight: bold\">{</span><span style=\"color: #008000\">'settings'</span>: <span style=\"font-weight: bold\">{}</span>, \n", + " <span style=\"color: #008000\">'defaults'</span>: <span style=\"color: #008000\">'plan/defaults/tasks_torch.yaml'</span>, <span style=\"color: #008000\">'locally_tuned_model_validation'</span>: <span style=\"font-weight: bold\">{</span><span style=\"color: #008000\">'function'</span>: <span style=\"color: #008000\">'validate'</span>, <span style=\"color: #008000\">'kwargs'</span>: <span style=\"font-weight: bold\">{</span><span style=\"color: #008000\">'apply'</span>: \n", + " <span style=\"color: #008000\">'local'</span>, <span style=\"color: #008000\">'metrics'</span>: <span style=\"font-weight: bold\">[</span><span style=\"color: #008000\">'acc'</span><span style=\"font-weight: bold\">]}}</span>, <span style=\"color: #008000\">'train'</span>: <span style=\"font-weight: bold\">{</span><span style=\"color: #008000\">'function'</span>: <span style=\"color: #008000\">'train_batches'</span>, <span style=\"color: #008000\">'kwargs'</span>: <span style=\"font-weight: bold\">{</span><span style=\"color: #008000\">'metrics'</span>: <span style=\"font-weight: bold\">[</span><span style=\"color: #008000\">'loss'</span><span style=\"font-weight: bold\">]}}</span>, \n", + " <span style=\"color: #008000\">'aggregated_model_validation'</span>: <span style=\"font-weight: bold\">{</span><span style=\"color: #008000\">'function'</span>: <span style=\"color: #008000\">'validate'</span>, <span style=\"color: #008000\">'kwargs'</span>: <span style=\"font-weight: bold\">{</span><span style=\"color: #008000\">'apply'</span>: <span style=\"color: #008000\">'global'</span>, <span style=\"color: #008000\">'metrics'</span>: <span style=\"font-weight: bold\">[</span><span style=\"color: #008000\">'acc'</span><span style=\"font-weight: bold\">]}}}</span>, <span style=\"color: #008000\">'client'</span>: \n", + " <span style=\"font-weight: bold\"><</span><span style=\"color: #ff00ff; font-weight: bold\">openfl.component.aggregator.aggregator.Aggregator</span><span style=\"color: #000000\"> object at </span><span style=\"color: #000080; font-weight: bold\">0x7f1d58ece520</span><span style=\"font-weight: bold\">></span>, <span style=\"color: #008000\">'defaults'</span>: <span style=\"color: #008000\">'plan/defaults/collaborator.yaml'</span><span style=\"font-weight: bold\">}</span> \n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f1d58e30fa0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> Override <span style=\"color: #800000\">🡆</span> <span style=\"font-weight: bold\">{</span><span style=\"color: #008000\">'defaults'</span>: <span style=\"color: #008000\">'plan/defaults/collaborator.yaml'</span><span style=\"font-weight: bold\">}</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/federated/plan/plan.py\"><span style=\"color: #7f7f7f\">plan.py</span></a><span style=\"color: #7f7f7f\">:167</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f1d58e309d0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> Waiting for tasks<span style=\"color: #808000\">...</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\"><span style=\"color: #7f7f7f\">collaborator.py</span></a><span style=\"color: #7f7f7f\">:156</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f1d58e83af0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> Sending tasks to collaborator one for round <span style=\"color: #000080; font-weight: bold\">0</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\"><span style=\"color: #7f7f7f\">aggregator.py</span></a><span style=\"color: #7f7f7f\">:276</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f1d58e83a90>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> Received the following tasks: <span style=\"font-weight: bold\">[</span><span style=\"color: #008000\">'aggregated_model_validation'</span>, <span style=\"color: #008000\">'train'</span>, <span style=\"color: #008000\">'locally_tuned_model_validation'</span><span style=\"font-weight: bold\">]</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\"><span style=\"color: #7f7f7f\">collaborator.py</span></a><span style=\"color: #7f7f7f\">:144</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f1d58e30250>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "name": "stderr", + "output_type": "stream", + "text": [ + "/home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/federated/task/runner_pt.py:287: UserWarning: The given NumPy array is not writeable, and PyTorch does not support non-writeable tensors. This means you can write to the underlying (supposedly non-writeable) NumPy array using the tensor. You may want to copy the array to protect its data or make it writeable before converting it to a tensor. This type of warning will be suppressed for the rest of this program. (Triggered internally at /pytorch/torch/csrc/utils/tensor_numpy.cpp:143.)\n", + " new_state[k] = pt.from_numpy(tensor_dict.pop(k)).to(device)\n", + "/home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/federated/task/runner_pt.py:100: UserWarning: To copy construct from a tensor, it is recommended to use sourceTensor.clone().detach() or sourceTensor.clone().detach().requires_grad_(True), rather than torch.tensor(sourceTensor).\n", + " data, target = pt.tensor(data).to(self.device), pt.tensor(\n" + ] + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\">[17:00:52] </span><span style=\"color: #000080\">INFO</span> Sending metric for task aggregated_model_validation, round number <span style=\"color: #000080; font-weight: bold\">0</span>: acc <span style=\"color: #000080; font-weight: bold\">0.1226</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\"><span style=\"color: #7f7f7f\">collaborator.py</span></a><span style=\"color: #7f7f7f\">:365</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f1d58e30d00>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> Collaborator one is sending task results for aggregated_model_validation, round <span style=\"color: #000080; font-weight: bold\">0</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\"><span style=\"color: #7f7f7f\">aggregator.py</span></a><span style=\"color: #7f7f7f\">:428</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f1d58e30eb0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "name": "stderr", + "output_type": "stream", + "text": [ + "/home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/federated/task/runner_pt.py:154: UserWarning: To copy construct from a tensor, it is recommended to use sourceTensor.clone().detach() or sourceTensor.clone().detach().requires_grad_(True), rather than torch.tensor(sourceTensor).\n", + " data, target = pt.tensor(data).to(self.device), pt.tensor(\n" + ] + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\">[17:00:56] </span><span style=\"color: #000080\">INFO</span> Sending metric for task train, round number <span style=\"color: #000080; font-weight: bold\">0</span>: cross_entropy <span style=\"color: #000080; font-weight: bold\">0.11886847764253616</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\"><span style=\"color: #7f7f7f\">collaborator.py</span></a><span style=\"color: #7f7f7f\">:365</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f1d58e178e0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> Collaborator one is sending task results for train, round <span style=\"color: #000080; font-weight: bold\">0</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\"><span style=\"color: #7f7f7f\">aggregator.py</span></a><span style=\"color: #7f7f7f\">:428</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f1d58e17f10>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> Sending metric for task locally_tuned_model_validation, round number <span style=\"color: #000080; font-weight: bold\">0</span>: acc <span style=\"color: #000080; font-weight: bold\">0.954</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\"><span style=\"color: #7f7f7f\">collaborator.py</span></a><span style=\"color: #7f7f7f\">:365</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f1d58e17490>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> Collaborator one is sending task results for locally_tuned_model_validation, round <span style=\"color: #000080; font-weight: bold\">0</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\"><span style=\"color: #7f7f7f\">aggregator.py</span></a><span style=\"color: #7f7f7f\">:428</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f1d58e17760>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> All tasks completed on one for round <span style=\"color: #000080; font-weight: bold\">0</span><span style=\"color: #808000\">...</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\"><span style=\"color: #7f7f7f\">collaborator.py</span></a><span style=\"color: #7f7f7f\">:148</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f1d58e17430>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> Waiting for tasks<span style=\"color: #808000\">...</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\"><span style=\"color: #7f7f7f\">collaborator.py</span></a><span style=\"color: #7f7f7f\">:156</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f1d58e17c10>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> Sending tasks to collaborator two for round <span style=\"color: #000080; font-weight: bold\">0</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\"><span style=\"color: #7f7f7f\">aggregator.py</span></a><span style=\"color: #7f7f7f\">:276</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f1d58e17880>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> Received the following tasks: <span style=\"font-weight: bold\">[</span><span style=\"color: #008000\">'aggregated_model_validation'</span>, <span style=\"color: #008000\">'train'</span>, <span style=\"color: #008000\">'locally_tuned_model_validation'</span><span style=\"font-weight: bold\">]</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\"><span style=\"color: #7f7f7f\">collaborator.py</span></a><span style=\"color: #7f7f7f\">:144</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f1d58e17880>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> Sending metric for task aggregated_model_validation, round number <span style=\"color: #000080; font-weight: bold\">0</span>: acc <span style=\"color: #000080; font-weight: bold\">0.1176</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\"><span style=\"color: #7f7f7f\">collaborator.py</span></a><span style=\"color: #7f7f7f\">:365</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f1d58e17760>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> Collaborator two is sending task results for aggregated_model_validation, round <span style=\"color: #000080; font-weight: bold\">0</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\"><span style=\"color: #7f7f7f\">aggregator.py</span></a><span style=\"color: #7f7f7f\">:428</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f1d58e83bb0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\">[17:01:01] </span><span style=\"color: #000080\">INFO</span> Sending metric for task train, round number <span style=\"color: #000080; font-weight: bold\">0</span>: cross_entropy <span style=\"color: #000080; font-weight: bold\">0.11838725954294205</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\"><span style=\"color: #7f7f7f\">collaborator.py</span></a><span style=\"color: #7f7f7f\">:365</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f1d58e30c70>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> Collaborator two is sending task results for train, round <span style=\"color: #000080; font-weight: bold\">0</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\"><span style=\"color: #7f7f7f\">aggregator.py</span></a><span style=\"color: #7f7f7f\">:428</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f1d58e30dc0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> Sending metric for task locally_tuned_model_validation, round number <span style=\"color: #000080; font-weight: bold\">0</span>: acc <span style=\"color: #000080; font-weight: bold\">0.953</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\"><span style=\"color: #7f7f7f\">collaborator.py</span></a><span style=\"color: #7f7f7f\">:365</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f1d54711340>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> Collaborator two is sending task results for locally_tuned_model_validation, round <span style=\"color: #000080; font-weight: bold\">0</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\"><span style=\"color: #7f7f7f\">aggregator.py</span></a><span style=\"color: #7f7f7f\">:428</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f1d547110d0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> aggregated_model_validation task metrics<span style=\"color: #808000\">...</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\"><span style=\"color: #7f7f7f\">aggregator.py</span></a><span style=\"color: #7f7f7f\">:694</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f1d54711070>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> acc: <span style=\"color: #000080; font-weight: bold\">0.1201</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\"><span style=\"color: #7f7f7f\">aggregator.py</span></a><span style=\"color: #7f7f7f\">:742</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f1d58e83d90>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> Saved the best model with score <span style=\"color: #000080; font-weight: bold\">0.120100</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\"><span style=\"color: #7f7f7f\">aggregator.py</span></a><span style=\"color: #7f7f7f\">:749</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f1d54711220>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> train task metrics<span style=\"color: #808000\">...</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\"><span style=\"color: #7f7f7f\">aggregator.py</span></a><span style=\"color: #7f7f7f\">:694</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f1d547111f0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> cross_entropy: <span style=\"color: #000080; font-weight: bold\">0.1186</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\"><span style=\"color: #7f7f7f\">aggregator.py</span></a><span style=\"color: #7f7f7f\">:742</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f1d58e83550>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> locally_tuned_model_validation task metrics<span style=\"color: #808000\">...</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\"><span style=\"color: #7f7f7f\">aggregator.py</span></a><span style=\"color: #7f7f7f\">:694</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f1d54705b80>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> acc: <span style=\"color: #000080; font-weight: bold\">0.9535</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\"><span style=\"color: #7f7f7f\">aggregator.py</span></a><span style=\"color: #7f7f7f\">:742</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f1d54711220>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> Saving round <span style=\"color: #000080; font-weight: bold\">1</span> model<span style=\"color: #808000\">...</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\"><span style=\"color: #7f7f7f\">aggregator.py</span></a><span style=\"color: #7f7f7f\">:783</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f1d58e83700>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> Starting round <span style=\"color: #000080; font-weight: bold\">1</span><span style=\"color: #808000\">...</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\"><span style=\"color: #7f7f7f\">aggregator.py</span></a><span style=\"color: #7f7f7f\">:791</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f1d5472d7f0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> All tasks completed on two for round <span style=\"color: #000080; font-weight: bold\">0</span><span style=\"color: #808000\">...</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\"><span style=\"color: #7f7f7f\">collaborator.py</span></a><span style=\"color: #7f7f7f\">:148</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f1d54705f40>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> Waiting for tasks<span style=\"color: #808000\">...</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\"><span style=\"color: #7f7f7f\">collaborator.py</span></a><span style=\"color: #7f7f7f\">:156</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f1d5472d4c0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> Sending tasks to collaborator one for round <span style=\"color: #000080; font-weight: bold\">1</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\"><span style=\"color: #7f7f7f\">aggregator.py</span></a><span style=\"color: #7f7f7f\">:276</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f1d5472dc70>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> Received the following tasks: <span style=\"font-weight: bold\">[</span><span style=\"color: #008000\">'aggregated_model_validation'</span>, <span style=\"color: #008000\">'train'</span>, <span style=\"color: #008000\">'locally_tuned_model_validation'</span><span style=\"font-weight: bold\">]</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\"><span style=\"color: #7f7f7f\">collaborator.py</span></a><span style=\"color: #7f7f7f\">:144</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f1d58e30100>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> Sending metric for task aggregated_model_validation, round number <span style=\"color: #000080; font-weight: bold\">1</span>: acc <span style=\"color: #000080; font-weight: bold\">0.952</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\"><span style=\"color: #7f7f7f\">collaborator.py</span></a><span style=\"color: #7f7f7f\">:365</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f1d58e30c40>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> Collaborator one is sending task results for aggregated_model_validation, round <span style=\"color: #000080; font-weight: bold\">1</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\"><span style=\"color: #7f7f7f\">aggregator.py</span></a><span style=\"color: #7f7f7f\">:428</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f1d58e30e50>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\">[17:01:05] </span><span style=\"color: #000080\">INFO</span> Sending metric for task train, round number <span style=\"color: #000080; font-weight: bold\">1</span>: cross_entropy <span style=\"color: #000080; font-weight: bold\">0.08371877670288086</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\"><span style=\"color: #7f7f7f\">collaborator.py</span></a><span style=\"color: #7f7f7f\">:365</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f1d546e2580>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> Collaborator one is sending task results for train, round <span style=\"color: #000080; font-weight: bold\">1</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\"><span style=\"color: #7f7f7f\">aggregator.py</span></a><span style=\"color: #7f7f7f\">:428</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f1d546e2d60>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> Sending metric for task locally_tuned_model_validation, round number <span style=\"color: #000080; font-weight: bold\">1</span>: acc <span style=\"color: #000080; font-weight: bold\">0.9558</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\"><span style=\"color: #7f7f7f\">collaborator.py</span></a><span style=\"color: #7f7f7f\">:365</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f1d547118b0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> Collaborator one is sending task results for locally_tuned_model_validation, round <span style=\"color: #000080; font-weight: bold\">1</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\"><span style=\"color: #7f7f7f\">aggregator.py</span></a><span style=\"color: #7f7f7f\">:428</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f1d546e20a0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> All tasks completed on one for round <span style=\"color: #000080; font-weight: bold\">1</span><span style=\"color: #808000\">...</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\"><span style=\"color: #7f7f7f\">collaborator.py</span></a><span style=\"color: #7f7f7f\">:148</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f1d546e25e0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> Waiting for tasks<span style=\"color: #808000\">...</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\"><span style=\"color: #7f7f7f\">collaborator.py</span></a><span style=\"color: #7f7f7f\">:156</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f1d546e29a0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> Sending tasks to collaborator two for round <span style=\"color: #000080; font-weight: bold\">1</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\"><span style=\"color: #7f7f7f\">aggregator.py</span></a><span style=\"color: #7f7f7f\">:276</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f1d546e25e0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> Received the following tasks: <span style=\"font-weight: bold\">[</span><span style=\"color: #008000\">'aggregated_model_validation'</span>, <span style=\"color: #008000\">'train'</span>, <span style=\"color: #008000\">'locally_tuned_model_validation'</span><span style=\"font-weight: bold\">]</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\"><span style=\"color: #7f7f7f\">collaborator.py</span></a><span style=\"color: #7f7f7f\">:144</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f1d546e23a0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> Sending metric for task aggregated_model_validation, round number <span style=\"color: #000080; font-weight: bold\">1</span>: acc <span style=\"color: #000080; font-weight: bold\">0.957</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\"><span style=\"color: #7f7f7f\">collaborator.py</span></a><span style=\"color: #7f7f7f\">:365</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f1d58e83940>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> Collaborator two is sending task results for aggregated_model_validation, round <span style=\"color: #000080; font-weight: bold\">1</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\"><span style=\"color: #7f7f7f\">aggregator.py</span></a><span style=\"color: #7f7f7f\">:428</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f1d58e83df0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\">[17:01:09] </span><span style=\"color: #000080\">INFO</span> Sending metric for task train, round number <span style=\"color: #000080; font-weight: bold\">1</span>: cross_entropy <span style=\"color: #000080; font-weight: bold\">0.08402730524539948</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\"><span style=\"color: #7f7f7f\">collaborator.py</span></a><span style=\"color: #7f7f7f\">:365</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f1d54705fd0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> Collaborator two is sending task results for train, round <span style=\"color: #000080; font-weight: bold\">1</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\"><span style=\"color: #7f7f7f\">aggregator.py</span></a><span style=\"color: #7f7f7f\">:428</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f1d54705b80>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> Sending metric for task locally_tuned_model_validation, round number <span style=\"color: #000080; font-weight: bold\">1</span>: acc <span style=\"color: #000080; font-weight: bold\">0.9676</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\"><span style=\"color: #7f7f7f\">collaborator.py</span></a><span style=\"color: #7f7f7f\">:365</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f1d58e17e50>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> Collaborator two is sending task results for locally_tuned_model_validation, round <span style=\"color: #000080; font-weight: bold\">1</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\"><span style=\"color: #7f7f7f\">aggregator.py</span></a><span style=\"color: #7f7f7f\">:428</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f1d58e171f0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> aggregated_model_validation task metrics<span style=\"color: #808000\">...</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\"><span style=\"color: #7f7f7f\">aggregator.py</span></a><span style=\"color: #7f7f7f\">:694</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f1d546d84f0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\">[17:01:10] </span><span style=\"color: #000080\">INFO</span> acc: <span style=\"color: #000080; font-weight: bold\">0.9545</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\"><span style=\"color: #7f7f7f\">aggregator.py</span></a><span style=\"color: #7f7f7f\">:742</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f1d546d8730>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> Saved the best model with score <span style=\"color: #000080; font-weight: bold\">0.954500</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\"><span style=\"color: #7f7f7f\">aggregator.py</span></a><span style=\"color: #7f7f7f\">:749</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f1d546d8df0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> train task metrics<span style=\"color: #808000\">...</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\"><span style=\"color: #7f7f7f\">aggregator.py</span></a><span style=\"color: #7f7f7f\">:694</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f1d546e2d90>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> cross_entropy: <span style=\"color: #000080; font-weight: bold\">0.0839</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\"><span style=\"color: #7f7f7f\">aggregator.py</span></a><span style=\"color: #7f7f7f\">:742</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f1d546e2070>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> locally_tuned_model_validation task metrics<span style=\"color: #808000\">...</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\"><span style=\"color: #7f7f7f\">aggregator.py</span></a><span style=\"color: #7f7f7f\">:694</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f1d546e25e0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> acc: <span style=\"color: #000080; font-weight: bold\">0.9617</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\"><span style=\"color: #7f7f7f\">aggregator.py</span></a><span style=\"color: #7f7f7f\">:742</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f1d546e2700>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> Saving round <span style=\"color: #000080; font-weight: bold\">2</span> model<span style=\"color: #808000\">...</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\"><span style=\"color: #7f7f7f\">aggregator.py</span></a><span style=\"color: #7f7f7f\">:783</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f1d54707b50>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> Starting round <span style=\"color: #000080; font-weight: bold\">2</span><span style=\"color: #808000\">...</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\"><span style=\"color: #7f7f7f\">aggregator.py</span></a><span style=\"color: #7f7f7f\">:791</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f1d546dccd0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> All tasks completed on two for round <span style=\"color: #000080; font-weight: bold\">1</span><span style=\"color: #808000\">...</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\"><span style=\"color: #7f7f7f\">collaborator.py</span></a><span style=\"color: #7f7f7f\">:148</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f1d54705f40>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> Waiting for tasks<span style=\"color: #808000\">...</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\"><span style=\"color: #7f7f7f\">collaborator.py</span></a><span style=\"color: #7f7f7f\">:156</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f1d546d8190>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> Sending tasks to collaborator one for round <span style=\"color: #000080; font-weight: bold\">2</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\"><span style=\"color: #7f7f7f\">aggregator.py</span></a><span style=\"color: #7f7f7f\">:276</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f1d58e30d00>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> Received the following tasks: <span style=\"font-weight: bold\">[</span><span style=\"color: #008000\">'aggregated_model_validation'</span>, <span style=\"color: #008000\">'train'</span>, <span style=\"color: #008000\">'locally_tuned_model_validation'</span><span style=\"font-weight: bold\">]</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\"><span style=\"color: #7f7f7f\">collaborator.py</span></a><span style=\"color: #7f7f7f\">:144</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f1d58e30e50>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> Sending metric for task aggregated_model_validation, round number <span style=\"color: #000080; font-weight: bold\">2</span>: acc <span style=\"color: #000080; font-weight: bold\">0.9632</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\"><span style=\"color: #7f7f7f\">collaborator.py</span></a><span style=\"color: #7f7f7f\">:365</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f1d58e30130>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> Collaborator one is sending task results for aggregated_model_validation, round <span style=\"color: #000080; font-weight: bold\">2</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\"><span style=\"color: #7f7f7f\">aggregator.py</span></a><span style=\"color: #7f7f7f\">:428</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f1d546e29d0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\">[17:01:14] </span><span style=\"color: #000080\">INFO</span> Sending metric for task train, round number <span style=\"color: #000080; font-weight: bold\">2</span>: cross_entropy <span style=\"color: #000080; font-weight: bold\">0.07936933636665344</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\"><span style=\"color: #7f7f7f\">collaborator.py</span></a><span style=\"color: #7f7f7f\">:365</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f1d546d8190>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> Collaborator one is sending task results for train, round <span style=\"color: #000080; font-weight: bold\">2</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\"><span style=\"color: #7f7f7f\">aggregator.py</span></a><span style=\"color: #7f7f7f\">:428</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f1d58e30fa0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> Sending metric for task locally_tuned_model_validation, round number <span style=\"color: #000080; font-weight: bold\">2</span>: acc <span style=\"color: #000080; font-weight: bold\">0.9772</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\"><span style=\"color: #7f7f7f\">collaborator.py</span></a><span style=\"color: #7f7f7f\">:365</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f1d5472dc40>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> Collaborator one is sending task results for locally_tuned_model_validation, round <span style=\"color: #000080; font-weight: bold\">2</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\"><span style=\"color: #7f7f7f\">aggregator.py</span></a><span style=\"color: #7f7f7f\">:428</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f1d58e17640>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> All tasks completed on one for round <span style=\"color: #000080; font-weight: bold\">2</span><span style=\"color: #808000\">...</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\"><span style=\"color: #7f7f7f\">collaborator.py</span></a><span style=\"color: #7f7f7f\">:148</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f1d54707e20>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> Waiting for tasks<span style=\"color: #808000\">...</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\"><span style=\"color: #7f7f7f\">collaborator.py</span></a><span style=\"color: #7f7f7f\">:156</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f1d58e177f0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> Sending tasks to collaborator two for round <span style=\"color: #000080; font-weight: bold\">2</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\"><span style=\"color: #7f7f7f\">aggregator.py</span></a><span style=\"color: #7f7f7f\">:276</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f1d58e171f0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> Received the following tasks: <span style=\"font-weight: bold\">[</span><span style=\"color: #008000\">'aggregated_model_validation'</span>, <span style=\"color: #008000\">'train'</span>, <span style=\"color: #008000\">'locally_tuned_model_validation'</span><span style=\"font-weight: bold\">]</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\"><span style=\"color: #7f7f7f\">collaborator.py</span></a><span style=\"color: #7f7f7f\">:144</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f1d58e171f0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> Sending metric for task aggregated_model_validation, round number <span style=\"color: #000080; font-weight: bold\">2</span>: acc <span style=\"color: #000080; font-weight: bold\">0.9678</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\"><span style=\"color: #7f7f7f\">collaborator.py</span></a><span style=\"color: #7f7f7f\">:365</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f1d547118b0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> Collaborator two is sending task results for aggregated_model_validation, round <span style=\"color: #000080; font-weight: bold\">2</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\"><span style=\"color: #7f7f7f\">aggregator.py</span></a><span style=\"color: #7f7f7f\">:428</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f1d54711070>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\">[17:01:19] </span><span style=\"color: #000080\">INFO</span> Sending metric for task train, round number <span style=\"color: #000080; font-weight: bold\">2</span>: cross_entropy <span style=\"color: #000080; font-weight: bold\">0.07971804589033127</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\"><span style=\"color: #7f7f7f\">collaborator.py</span></a><span style=\"color: #7f7f7f\">:365</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f1d546e29d0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> Collaborator two is sending task results for train, round <span style=\"color: #000080; font-weight: bold\">2</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\"><span style=\"color: #7f7f7f\">aggregator.py</span></a><span style=\"color: #7f7f7f\">:428</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f1d54711a90>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> Sending metric for task locally_tuned_model_validation, round number <span style=\"color: #000080; font-weight: bold\">2</span>: acc <span style=\"color: #000080; font-weight: bold\">0.9778</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\"><span style=\"color: #7f7f7f\">collaborator.py</span></a><span style=\"color: #7f7f7f\">:365</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f1d546bf550>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> Collaborator two is sending task results for locally_tuned_model_validation, round <span style=\"color: #000080; font-weight: bold\">2</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\"><span style=\"color: #7f7f7f\">aggregator.py</span></a><span style=\"color: #7f7f7f\">:428</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f1d546bf9d0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> aggregated_model_validation task metrics<span style=\"color: #808000\">...</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\"><span style=\"color: #7f7f7f\">aggregator.py</span></a><span style=\"color: #7f7f7f\">:694</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f1d546f8070>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> acc: <span style=\"color: #000080; font-weight: bold\">0.9655</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\"><span style=\"color: #7f7f7f\">aggregator.py</span></a><span style=\"color: #7f7f7f\">:742</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f1d546bf850>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> Saved the best model with score <span style=\"color: #000080; font-weight: bold\">0.965500</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\"><span style=\"color: #7f7f7f\">aggregator.py</span></a><span style=\"color: #7f7f7f\">:749</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f1d546f8700>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> train task metrics<span style=\"color: #808000\">...</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\"><span style=\"color: #7f7f7f\">aggregator.py</span></a><span style=\"color: #7f7f7f\">:694</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f1d546f8580>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> cross_entropy: <span style=\"color: #000080; font-weight: bold\">0.0795</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\"><span style=\"color: #7f7f7f\">aggregator.py</span></a><span style=\"color: #7f7f7f\">:742</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f1d58e171f0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> locally_tuned_model_validation task metrics<span style=\"color: #808000\">...</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\"><span style=\"color: #7f7f7f\">aggregator.py</span></a><span style=\"color: #7f7f7f\">:694</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f1d54696760>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> acc: <span style=\"color: #000080; font-weight: bold\">0.9775</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\"><span style=\"color: #7f7f7f\">aggregator.py</span></a><span style=\"color: #7f7f7f\">:742</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f1d58e17880>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> Saving round <span style=\"color: #000080; font-weight: bold\">3</span> model<span style=\"color: #808000\">...</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\"><span style=\"color: #7f7f7f\">aggregator.py</span></a><span style=\"color: #7f7f7f\">:783</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f1d58e17ee0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> Starting round <span style=\"color: #000080; font-weight: bold\">3</span><span style=\"color: #808000\">...</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\"><span style=\"color: #7f7f7f\">aggregator.py</span></a><span style=\"color: #7f7f7f\">:791</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f1d546f8fd0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> All tasks completed on two for round <span style=\"color: #000080; font-weight: bold\">2</span><span style=\"color: #808000\">...</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\"><span style=\"color: #7f7f7f\">collaborator.py</span></a><span style=\"color: #7f7f7f\">:148</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f1d546d8d30>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> Waiting for tasks<span style=\"color: #808000\">...</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\"><span style=\"color: #7f7f7f\">collaborator.py</span></a><span style=\"color: #7f7f7f\">:156</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f1d546bfee0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> Sending tasks to collaborator one for round <span style=\"color: #000080; font-weight: bold\">3</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\"><span style=\"color: #7f7f7f\">aggregator.py</span></a><span style=\"color: #7f7f7f\">:276</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f1d58e30e50>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> Received the following tasks: <span style=\"font-weight: bold\">[</span><span style=\"color: #008000\">'aggregated_model_validation'</span>, <span style=\"color: #008000\">'train'</span>, <span style=\"color: #008000\">'locally_tuned_model_validation'</span><span style=\"font-weight: bold\">]</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\"><span style=\"color: #7f7f7f\">collaborator.py</span></a><span style=\"color: #7f7f7f\">:144</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f1d58e30130>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> Sending metric for task aggregated_model_validation, round number <span style=\"color: #000080; font-weight: bold\">3</span>: acc <span style=\"color: #000080; font-weight: bold\">0.9794</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\"><span style=\"color: #7f7f7f\">collaborator.py</span></a><span style=\"color: #7f7f7f\">:365</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f1d58e30fa0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> Collaborator one is sending task results for aggregated_model_validation, round <span style=\"color: #000080; font-weight: bold\">3</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\"><span style=\"color: #7f7f7f\">aggregator.py</span></a><span style=\"color: #7f7f7f\">:428</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f1d546bfc10>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\">[17:01:23] </span><span style=\"color: #000080\">INFO</span> Sending metric for task train, round number <span style=\"color: #000080; font-weight: bold\">3</span>: cross_entropy <span style=\"color: #000080; font-weight: bold\">0.07742393016815186</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\"><span style=\"color: #7f7f7f\">collaborator.py</span></a><span style=\"color: #7f7f7f\">:365</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f1d54707b50>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> Collaborator one is sending task results for train, round <span style=\"color: #000080; font-weight: bold\">3</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\"><span style=\"color: #7f7f7f\">aggregator.py</span></a><span style=\"color: #7f7f7f\">:428</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f1d547118b0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> Sending metric for task locally_tuned_model_validation, round number <span style=\"color: #000080; font-weight: bold\">3</span>: acc <span style=\"color: #000080; font-weight: bold\">0.9756</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\"><span style=\"color: #7f7f7f\">collaborator.py</span></a><span style=\"color: #7f7f7f\">:365</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f1d546c6400>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> Collaborator one is sending task results for locally_tuned_model_validation, round <span style=\"color: #000080; font-weight: bold\">3</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\"><span style=\"color: #7f7f7f\">aggregator.py</span></a><span style=\"color: #7f7f7f\">:428</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f1d546c6310>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> All tasks completed on one for round <span style=\"color: #000080; font-weight: bold\">3</span><span style=\"color: #808000\">...</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\"><span style=\"color: #7f7f7f\">collaborator.py</span></a><span style=\"color: #7f7f7f\">:148</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f1d546e2df0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\">[17:01:24] </span><span style=\"color: #000080\">INFO</span> Waiting for tasks<span style=\"color: #808000\">...</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\"><span style=\"color: #7f7f7f\">collaborator.py</span></a><span style=\"color: #7f7f7f\">:156</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f1d546e2df0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> Sending tasks to collaborator two for round <span style=\"color: #000080; font-weight: bold\">3</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\"><span style=\"color: #7f7f7f\">aggregator.py</span></a><span style=\"color: #7f7f7f\">:276</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f1d546c6040>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> Received the following tasks: <span style=\"font-weight: bold\">[</span><span style=\"color: #008000\">'aggregated_model_validation'</span>, <span style=\"color: #008000\">'train'</span>, <span style=\"color: #008000\">'locally_tuned_model_validation'</span><span style=\"font-weight: bold\">]</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\"><span style=\"color: #7f7f7f\">collaborator.py</span></a><span style=\"color: #7f7f7f\">:144</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f1d54696d00>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> Sending metric for task aggregated_model_validation, round number <span style=\"color: #000080; font-weight: bold\">3</span>: acc <span style=\"color: #000080; font-weight: bold\">0.9806</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\"><span style=\"color: #7f7f7f\">collaborator.py</span></a><span style=\"color: #7f7f7f\">:365</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f1d546c6bb0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> Collaborator two is sending task results for aggregated_model_validation, round <span style=\"color: #000080; font-weight: bold\">3</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\"><span style=\"color: #7f7f7f\">aggregator.py</span></a><span style=\"color: #7f7f7f\">:428</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f1d546c6370>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\">[17:01:28] </span><span style=\"color: #000080\">INFO</span> Sending metric for task train, round number <span style=\"color: #000080; font-weight: bold\">3</span>: cross_entropy <span style=\"color: #000080; font-weight: bold\">0.07761752605438232</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\"><span style=\"color: #7f7f7f\">collaborator.py</span></a><span style=\"color: #7f7f7f\">:365</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f1d546bfcd0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> Collaborator two is sending task results for train, round <span style=\"color: #000080; font-weight: bold\">3</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\"><span style=\"color: #7f7f7f\">aggregator.py</span></a><span style=\"color: #7f7f7f\">:428</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f1d546e2700>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> Sending metric for task locally_tuned_model_validation, round number <span style=\"color: #000080; font-weight: bold\">3</span>: acc <span style=\"color: #000080; font-weight: bold\">0.9834</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\"><span style=\"color: #7f7f7f\">collaborator.py</span></a><span style=\"color: #7f7f7f\">:365</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f1d546c6400>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> Collaborator two is sending task results for locally_tuned_model_validation, round <span style=\"color: #000080; font-weight: bold\">3</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\"><span style=\"color: #7f7f7f\">aggregator.py</span></a><span style=\"color: #7f7f7f\">:428</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f1d546c6310>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> aggregated_model_validation task metrics<span style=\"color: #808000\">...</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\"><span style=\"color: #7f7f7f\">aggregator.py</span></a><span style=\"color: #7f7f7f\">:694</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f1d546a8370>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> acc: <span style=\"color: #000080; font-weight: bold\">0.9800</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\"><span style=\"color: #7f7f7f\">aggregator.py</span></a><span style=\"color: #7f7f7f\">:742</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f1d546c6880>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> Saved the best model with score <span style=\"color: #000080; font-weight: bold\">0.980000</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\"><span style=\"color: #7f7f7f\">aggregator.py</span></a><span style=\"color: #7f7f7f\">:749</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f1d546a8610>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> train task metrics<span style=\"color: #808000\">...</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\"><span style=\"color: #7f7f7f\">aggregator.py</span></a><span style=\"color: #7f7f7f\">:694</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f1d546a8b20>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> cross_entropy: <span style=\"color: #000080; font-weight: bold\">0.0775</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\"><span style=\"color: #7f7f7f\">aggregator.py</span></a><span style=\"color: #7f7f7f\">:742</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f1d546c6520>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> locally_tuned_model_validation task metrics<span style=\"color: #808000\">...</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\"><span style=\"color: #7f7f7f\">aggregator.py</span></a><span style=\"color: #7f7f7f\">:694</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f1d546b8be0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> acc: <span style=\"color: #000080; font-weight: bold\">0.9795</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\"><span style=\"color: #7f7f7f\">aggregator.py</span></a><span style=\"color: #7f7f7f\">:742</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f1d546b8bb0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> Saving round <span style=\"color: #000080; font-weight: bold\">4</span> model<span style=\"color: #808000\">...</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\"><span style=\"color: #7f7f7f\">aggregator.py</span></a><span style=\"color: #7f7f7f\">:783</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f1d546c6e20>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> Starting round <span style=\"color: #000080; font-weight: bold\">4</span><span style=\"color: #808000\">...</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\"><span style=\"color: #7f7f7f\">aggregator.py</span></a><span style=\"color: #7f7f7f\">:791</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f1d546b8400>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> All tasks completed on two for round <span style=\"color: #000080; font-weight: bold\">3</span><span style=\"color: #808000\">...</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\"><span style=\"color: #7f7f7f\">collaborator.py</span></a><span style=\"color: #7f7f7f\">:148</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f1d54707880>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> Waiting for tasks<span style=\"color: #808000\">...</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\"><span style=\"color: #7f7f7f\">collaborator.py</span></a><span style=\"color: #7f7f7f\">:156</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f1d54696d00>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> Sending tasks to collaborator one for round <span style=\"color: #000080; font-weight: bold\">4</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\"><span style=\"color: #7f7f7f\">aggregator.py</span></a><span style=\"color: #7f7f7f\">:276</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f1d58e17880>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> Received the following tasks: <span style=\"font-weight: bold\">[</span><span style=\"color: #008000\">'aggregated_model_validation'</span>, <span style=\"color: #008000\">'train'</span>, <span style=\"color: #008000\">'locally_tuned_model_validation'</span><span style=\"font-weight: bold\">]</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\"><span style=\"color: #7f7f7f\">collaborator.py</span></a><span style=\"color: #7f7f7f\">:144</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f1d546f8dc0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> Sending metric for task aggregated_model_validation, round number <span style=\"color: #000080; font-weight: bold\">4</span>: acc <span style=\"color: #000080; font-weight: bold\">0.9792</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\"><span style=\"color: #7f7f7f\">collaborator.py</span></a><span style=\"color: #7f7f7f\">:365</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f1d5468c4c0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> Collaborator one is sending task results for aggregated_model_validation, round <span style=\"color: #000080; font-weight: bold\">4</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\"><span style=\"color: #7f7f7f\">aggregator.py</span></a><span style=\"color: #7f7f7f\">:428</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f1d5468cdc0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\">[17:01:33] </span><span style=\"color: #000080\">INFO</span> Sending metric for task train, round number <span style=\"color: #000080; font-weight: bold\">4</span>: cross_entropy <span style=\"color: #000080; font-weight: bold\">0.07607017457485199</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\"><span style=\"color: #7f7f7f\">collaborator.py</span></a><span style=\"color: #7f7f7f\">:365</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f1d5469c340>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> Collaborator one is sending task results for train, round <span style=\"color: #000080; font-weight: bold\">4</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\"><span style=\"color: #7f7f7f\">aggregator.py</span></a><span style=\"color: #7f7f7f\">:428</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f1d5469ce50>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> Sending metric for task locally_tuned_model_validation, round number <span style=\"color: #000080; font-weight: bold\">4</span>: acc <span style=\"color: #000080; font-weight: bold\">0.9804</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\"><span style=\"color: #7f7f7f\">collaborator.py</span></a><span style=\"color: #7f7f7f\">:365</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f1d58e17880>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> Collaborator one is sending task results for locally_tuned_model_validation, round <span style=\"color: #000080; font-weight: bold\">4</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\"><span style=\"color: #7f7f7f\">aggregator.py</span></a><span style=\"color: #7f7f7f\">:428</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f1d58e17880>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> All tasks completed on one for round <span style=\"color: #000080; font-weight: bold\">4</span><span style=\"color: #808000\">...</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\"><span style=\"color: #7f7f7f\">collaborator.py</span></a><span style=\"color: #7f7f7f\">:148</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f1d546c6370>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> Waiting for tasks<span style=\"color: #808000\">...</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\"><span style=\"color: #7f7f7f\">collaborator.py</span></a><span style=\"color: #7f7f7f\">:156</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f1d546e2700>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> Sending tasks to collaborator two for round <span style=\"color: #000080; font-weight: bold\">4</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\"><span style=\"color: #7f7f7f\">aggregator.py</span></a><span style=\"color: #7f7f7f\">:276</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f1d546b8df0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> Received the following tasks: <span style=\"font-weight: bold\">[</span><span style=\"color: #008000\">'aggregated_model_validation'</span>, <span style=\"color: #008000\">'train'</span>, <span style=\"color: #008000\">'locally_tuned_model_validation'</span><span style=\"font-weight: bold\">]</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\"><span style=\"color: #7f7f7f\">collaborator.py</span></a><span style=\"color: #7f7f7f\">:144</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f1d58e83790>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> Sending metric for task aggregated_model_validation, round number <span style=\"color: #000080; font-weight: bold\">4</span>: acc <span style=\"color: #000080; font-weight: bold\">0.9812</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\"><span style=\"color: #7f7f7f\">collaborator.py</span></a><span style=\"color: #7f7f7f\">:365</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f1d546e2940>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> Collaborator two is sending task results for aggregated_model_validation, round <span style=\"color: #000080; font-weight: bold\">4</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\"><span style=\"color: #7f7f7f\">aggregator.py</span></a><span style=\"color: #7f7f7f\">:428</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f1d54705f40>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\">[17:01:37] </span><span style=\"color: #000080\">INFO</span> Sending metric for task train, round number <span style=\"color: #000080; font-weight: bold\">4</span>: cross_entropy <span style=\"color: #000080; font-weight: bold\">0.07631511241197586</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\"><span style=\"color: #7f7f7f\">collaborator.py</span></a><span style=\"color: #7f7f7f\">:365</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f1d546a80a0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> Collaborator two is sending task results for train, round <span style=\"color: #000080; font-weight: bold\">4</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\"><span style=\"color: #7f7f7f\">aggregator.py</span></a><span style=\"color: #7f7f7f\">:428</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f1d546a80a0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> Sending metric for task locally_tuned_model_validation, round number <span style=\"color: #000080; font-weight: bold\">4</span>: acc <span style=\"color: #000080; font-weight: bold\">0.9836</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\"><span style=\"color: #7f7f7f\">collaborator.py</span></a><span style=\"color: #7f7f7f\">:365</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f1d546c6370>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> Collaborator two is sending task results for locally_tuned_model_validation, round <span style=\"color: #000080; font-weight: bold\">4</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\"><span style=\"color: #7f7f7f\">aggregator.py</span></a><span style=\"color: #7f7f7f\">:428</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f1d546b1760>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> aggregated_model_validation task metrics<span style=\"color: #808000\">...</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\"><span style=\"color: #7f7f7f\">aggregator.py</span></a><span style=\"color: #7f7f7f\">:694</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f1d546b1eb0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> acc: <span style=\"color: #000080; font-weight: bold\">0.9802</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\"><span style=\"color: #7f7f7f\">aggregator.py</span></a><span style=\"color: #7f7f7f\">:742</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f1d5469c940>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> Saved the best model with score <span style=\"color: #000080; font-weight: bold\">0.980200</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\"><span style=\"color: #7f7f7f\">aggregator.py</span></a><span style=\"color: #7f7f7f\">:749</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f1d5469ca00>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> train task metrics<span style=\"color: #808000\">...</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\"><span style=\"color: #7f7f7f\">aggregator.py</span></a><span style=\"color: #7f7f7f\">:694</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f1d546b1b20>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> cross_entropy: <span style=\"color: #000080; font-weight: bold\">0.0762</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\"><span style=\"color: #7f7f7f\">aggregator.py</span></a><span style=\"color: #7f7f7f\">:742</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f1d54711a90>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\">[17:01:38] </span><span style=\"color: #000080\">INFO</span> locally_tuned_model_validation task metrics<span style=\"color: #808000\">...</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\"><span style=\"color: #7f7f7f\">aggregator.py</span></a><span style=\"color: #7f7f7f\">:694</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f1d546b1790>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> acc: <span style=\"color: #000080; font-weight: bold\">0.9820</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\"><span style=\"color: #7f7f7f\">aggregator.py</span></a><span style=\"color: #7f7f7f\">:742</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f1d5469cc70>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> Saving round <span style=\"color: #000080; font-weight: bold\">5</span> model<span style=\"color: #808000\">...</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\"><span style=\"color: #7f7f7f\">aggregator.py</span></a><span style=\"color: #7f7f7f\">:783</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f1d5469ce50>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> Experiment Completed. Cleaning up<span style=\"color: #808000\">...</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\"><span style=\"color: #7f7f7f\">aggregator.py</span></a><span style=\"color: #7f7f7f\">:789</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f1d5469de50>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf\"> </span><span style=\"color: #000080\">INFO</span> All tasks completed on two for round <span style=\"color: #000080; font-weight: bold\">4</span><span style=\"color: #808000\">...</span> <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\"><span style=\"color: #7f7f7f\">collaborator.py</span></a><span style=\"color: #7f7f7f\">:148</span>\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f1d546d8d30>" + ] + }, + "metadata": {}, + "output_type": "display_data" + } + ], "source": [ "#Run experiment, return trained FederatedModel\n", "final_fl_model = fx.run_experiment(collaborators,{'aggregator.settings.rounds_to_train':5})" @@ -212,7 +2718,7 @@ }, { "cell_type": "code", - "execution_count": null, + "execution_count": 11, "metadata": {}, "outputs": [], "source": [ diff --git a/openfl-tutorials/Federated_Pytorch_MNIST_custom_aggregation_Tutorial.ipynb b/openfl-tutorials/Federated_Pytorch_MNIST_custom_aggregation_Tutorial.ipynb new file mode 100644 index 0000000000..d635a51d30 --- /dev/null +++ b/openfl-tutorials/Federated_Pytorch_MNIST_custom_aggregation_Tutorial.ipynb @@ -0,0 +1,430 @@ +{ + "cells": [ + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "# Federated PyTorch MNIST Tutorial" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "#Install dependencies if not already installed\n", + "!pip install torch torchvision" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "import numpy as np\n", + "import torch\n", + "import torch.nn as nn\n", + "import torch.nn.functional as F\n", + "import torch.optim as optim\n", + "\n", + "import torchvision\n", + "import torchvision.transforms as transforms\n", + "import openfl.native as fx\n", + "from openfl.federated import FederatedModel,FederatedDataSet\n", + "\n", + "torch.manual_seed(0)\n", + "np.random.seed(0)" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "After importing the required packages, the next step is setting up our openfl workspace. To do this, simply run the `fx.init()` command as follows:" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "#Setup default workspace, logging, etc.\n", + "fx.init('torch_cnn_mnist')" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "Now we are ready to define our dataset and model to perform federated learning on. The dataset should be composed of a numpy arrayWe start with a simple fully connected model that is trained on the MNIST dataset. " + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "def one_hot(labels, classes):\n", + " return np.eye(classes)[labels]\n", + "\n", + "transform = transforms.Compose(\n", + " [transforms.ToTensor(),\n", + " transforms.Normalize((0.5, 0.5, 0.5), (0.5, 0.5, 0.5))])\n", + "\n", + "trainset = torchvision.datasets.MNIST(root='./data', train=True,\n", + " download=True, transform=transform)\n", + "\n", + "train_images,train_labels = trainset.train_data, np.array(trainset.train_labels)\n", + "train_images = torch.from_numpy(np.expand_dims(train_images, axis=1)).float()\n", + "train_labels = one_hot(train_labels,10)\n", + "\n", + "validset = torchvision.datasets.MNIST(root='./data', train=False,\n", + " download=True, transform=transform)\n", + "\n", + "valid_images,valid_labels = validset.test_data, np.array(validset.test_labels)\n", + "valid_images = torch.from_numpy(np.expand_dims(valid_images, axis=1)).float()\n", + "valid_labels = one_hot(valid_labels,10)" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "feature_shape = train_images.shape[1]\n", + "classes = 10\n", + "\n", + "fl_data = FederatedDataSet(train_images,train_labels,valid_images,valid_labels,batch_size=32,num_classes=classes)\n", + "\n", + "class Net(nn.Module):\n", + " def __init__(self):\n", + " super(Net, self).__init__()\n", + " self.conv1 = nn.Conv2d(1, 16, 3)\n", + " self.pool = nn.MaxPool2d(2, 2)\n", + " self.conv2 = nn.Conv2d(16, 32, 3)\n", + " self.fc1 = nn.Linear(32 5 * 5, 32)\n", + " self.fc2 = nn.Linear(32, 84)\n", + " self.fc3 = nn.Linear(84, 10)\n", + "\n", + " def forward(self, x):\n", + " x = self.pool(F.relu(self.conv1(x)))\n", + " x = self.pool(F.relu(self.conv2(x)))\n", + " x = x.view(x.size(0),-1)\n", + " x = F.relu(self.fc1(x))\n", + " x = F.relu(self.fc2(x))\n", + " x = self.fc3(x)\n", + " return x\n", + " \n", + "optimizer = lambda x: optim.Adam(x, lr=1e-4)\n", + "\n", + "def cross_entropy(output, target):\n", + " \"\"\"Binary cross-entropy metric\n", + " \"\"\"\n", + " return F.cross_entropy(input=output,target=torch.argmax(target, dim=1))" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "\n", + "#Create a federated model using the pytorch class, lambda optimizer function, and loss function\n", + "fl_model = FederatedModel(build_model=Net,optimizer=optimizer,loss_fn=cross_entropy,data_loader=fl_data)" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "The `FederatedModel` object is a wrapper around your Keras, Tensorflow or PyTorch model that makes it compatible with openfl. It provides built in federated training and validation functions that we will see used below. Using it's `setup` function, collaborator models and datasets can be automatically defined for the experiment. " + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "collaborator_models = fl_model.setup(num_collaborators=10)\n", + "collaborators = {str(i): collaborator_models[i] for i in range(10)}#, 'three':collaborator_models[2]}" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "#Original MNIST dataset\n", + "print(f'Original training data size: {len(train_images)}')\n", + "print(f'Original validation data size: {len(valid_images)}\\n')\n", + "\n", + "#Collaborator one's data\n", + "print(f'Collaborator one\\'s training data size: {len(collaborator_models[0].data_loader.X_train)}')\n", + "print(f'Collaborator one\\'s validation data size: {len(collaborator_models[0].data_loader.X_valid)}\\n')\n", + "\n", + "#Collaborator two's data\n", + "print(f'Collaborator two\\'s training data size: {len(collaborator_models[1].data_loader.X_train)}')\n", + "print(f'Collaborator two\\'s validation data size: {len(collaborator_models[1].data_loader.X_valid)}\\n')\n", + "\n", + "#Collaborator three's data\n", + "#print(f'Collaborator three\\'s training data size: {len(collaborator_models[2].data_loader.X_train)}')\n", + "#print(f'Collaborator three\\'s validation data size: {len(collaborator_models[2].data_loader.X_valid)}')" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "We can see the current plan values by running the `fx.get_plan()` function" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + " #Get the current values of the plan. Each of these can be overridden\n", + "print(fx.get_plan())" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "Now we are ready to run our experiment. If we want to pass in custom plan settings, we can easily do that with the `override_config` parameter" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "from openfl.component.aggregation_functions import AggregationFunctionInterface\n", + "import numpy as np\n", + "\n", + "class ExponentialSmoothingAveraging(AggregationFunctionInterface):\n", + " \"\"\"\n", + " Averaging via exponential smoothing.\n", + " \n", + " In order to use this mechanism properly you should specify `aggregator.settings.db_store_rounds` \n", + " in `override_config` keyword argument of `run_experiment` function. \n", + " It should be equal to the number of rounds you want to include in smoothing window.\n", + " \n", + " Args:\n", + " alpha(float): Smoothing term.\n", + " \"\"\"\n", + " def __init__(self, alpha=0.9):\n", + " self.alpha = alpha\n", + " \n", + " def call(self,\n", + " local_tensors,\n", + " db_iterator,\n", + " tensor_name,\n", + " fl_round,\n", + " tags):\n", + " \"\"\"Aggregate tensors.\n", + "\n", + " Args:\n", + " local_tensors(list[openfl.utilities.LocalTensor]): List of local tensors to aggregate.\n", + " db_iterator: iterator over history of all tensors.\n", + " Columns: ['tensor_name', 'round', 'tags', 'nparray']\n", + " tensor_name: name of the tensor\n", + " fl_round: round number\n", + " tags: tuple of tags for this tensor\n", + " \"\"\"\n", + " tensors, weights = zip([(x.tensor, x.weight) for x in local_tensors])\n", + " tensors, weights = np.array(tensors), np.array(weights)\n", + " average = np.average(tensors, weights=weights, axis=0)\n", + " if 'metric' in tags:\n", + " return average\n", + " previous_tensor_values = []\n", + " for record in db_iterator:\n", + " if (\n", + " record['tensor_name'] == tensor_name\n", + " and 'aggregated' in record['tags']\n", + " and 'delta' not in record['tags']\n", + " ):\n", + " previous_tensor_values.append(record['nparray'])\n", + " for i, x in enumerate(previous_tensor_values):\n", + " previous_tensor_values[i] = x self.alpha * (1 - self.alpha) ** i\n", + " smoothing_term = np.sum(previous_tensor_values, axis=0)\n", + " return self.alpha * average + (1 - self.alpha) * smoothing_term" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "from openfl.component.aggregation_functions import AggregationFunctionInterface\n", + "import numpy as np\n", + "\n", + "class ClippedAveraging(AggregationFunctionInterface):\n", + " def __init__(self, ratio):\n", + " \"\"\"Average clipped tensors.\n", + " \n", + " Args:\n", + " ratio(float): Ratio to multiply with a tensor for clipping\n", + " \"\"\"\n", + " self.ratio = ratio\n", + " \n", + " def call(self,\n", + " local_tensors,\n", + " db_iterator,\n", + " tensor_name,\n", + " fl_round,\n", + " __):\n", + " \"\"\"Aggregate tensors.\n", + "\n", + " Args:\n", + " local_tensors(list[openfl.utilities.LocalTensor]): List of local tensors to aggregate.\n", + " db_iterator: iterator over history of all tensors.\n", + " Columns: ['tensor_name', 'round', 'tags', 'nparray']\n", + " tensor_name: name of the tensor\n", + " fl_round: round number\n", + " tags: tuple of tags for this tensor\n", + " \"\"\"\n", + " clipped_tensors = []\n", + " previous_tensor_value = None\n", + " for record in db_iterator:\n", + " if (\n", + " record['round'] == (fl_round - 1)\n", + " and record['tensor_name'] == tensor_name\n", + " and 'aggregated' in record['tags']\n", + " and 'delta' not in record['tags']\n", + " ):\n", + " previous_tensor_value = record['nparray']\n", + " weights = []\n", + " for local_tensor in local_tensors:\n", + " prev_tensor = previous_tensor_value if previous_tensor_value is not None else local_tensor.tensor\n", + " delta = local_tensor.tensor - prev_tensor\n", + " new_tensor = prev_tensor + delta self.ratio\n", + " clipped_tensors.append(new_tensor)\n", + " weights.append(local_tensor.weight)\n", + "\n", + " return np.average(clipped_tensors, weights=weights, axis=0)" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "from openfl.component.aggregation_functions import AggregationFunctionInterface\n", + "\n", + "class ConditionalThresholdAveraging(AggregationFunctionInterface):\n", + " def __init__(self, threshold_fn, metric_name='acc', tags=['metric', 'validate_local']):\n", + " \"\"\"Average tensors by metric value on previous round.\n", + " If no tensors match threshold condition, a simple weighted averaging will be performed.\n", + " \n", + " Args:\n", + " threshold_fn(callable): function to define a threshold for each round.\n", + " Has single argument `round_number`. \n", + " Returns threshold value above which collaborators are allowed to participate in aggregation.\n", + " metric_name(str): name of the metric to trace. Can be either 'acc' or 'loss'.\n", + " tags(Tuple[str]): tags of the metric tensor.\n", + " \"\"\"\n", + " self.metric_name = metric_name\n", + " self.threshold_fn = threshold_fn\n", + " self.tags = tags\n", + " self.logged_round = -1\n", + " \n", + " def call(self,\n", + " local_tensors,\n", + " db_iterator,\n", + " tensor_name,\n", + " fl_round,\n", + " __):\n", + " \"\"\"Aggregate tensors.\n", + "\n", + " Args:\n", + " local_tensors(list[openfl.utilities.LocalTensor]): List of local tensors to aggregate.\n", + " db_iterator: iterator over history of all tensors.\n", + " Columns: ['tensor_name', 'round', 'tags', 'nparray']\n", + " tensor_name: name of the tensor\n", + " fl_round: round number\n", + " tags: tuple of tags for this tensor\n", + " \"\"\"\n", + " selected_tensors = []\n", + " selected_weights = []\n", + " for record in db_iterator:\n", + " for local_tensor in local_tensors:\n", + " tags = set(self.tags + [local_tensor.col_name])\n", + " if (\n", + " tags <= set(record['tags']) \n", + " and record['round'] == fl_round\n", + " and record['tensor_name'] == self.metric_name\n", + " and record['nparray'] >= self.threshold_fn(fl_round)\n", + " ):\n", + " selected_tensors.append(local_tensor.tensor)\n", + " selected_weights.append(local_tensor.weight)\n", + " if not selected_tensors:\n", + " if self.logged_round < fl_round:\n", + " fx.logger.warning('No collaborators match threshold condition. Performing simple averaging...')\n", + " selected_tensors = [local_tensor.tensor for tensor in local_tensors]\n", + " selected_weights = [local_tensor.weight for tensor in local_tensors]\n", + " if self.logged_round < fl_round:\n", + " self.logged_round += 1\n", + " return np.average(selected_tensors, weights=selected_weights, axis=0)" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "#Run experiment, return trained FederatedModel\n", + "final_fl_model = fx.run_experiment(collaborators,\n", + " {\n", + " 'aggregator.settings.rounds_to_train':5,\n", + " 'aggregator.settings.db_store_rounds':5,\n", + " 'tasks.train.aggregation_type': ConditionalThresholdAveraging(lambda round_num: 0.85 + 0.03 round_num)\n", + " })" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "#Save final model\n", + "final_fl_model.save_native('final_pytorch_model')" + ] + } + ], + "metadata": { + "kernelspec": { + "display_name": "Python 3", + "language": "python", + "name": "python3" + }, + "language_info": { + "codemirror_mode": { + "name": "ipython", + "version": 3 + }, + "file_extension": ".py", + "mimetype": "text/x-python", + "name": "python", + "nbconvert_exporter": "python", + "pygments_lexer": "ipython3", + "version": "3.8.8" + } + }, + "nbformat": 4, + "nbformat_minor": 4 +} diff --git a/openfl/component/aggregation_functions/__init__.py b/openfl/component/aggregation_functions/__init__.py new file mode 100644 index 0000000000..f0d8247f67 --- /dev/null +++ b/openfl/component/aggregation_functions/__init__.py @@ -0,0 +1,11 @@ +# Copyright (C) 2020-2021 Intel Corporation +# SPDX-License-Identifier: Apache-2.0 + +"""Aggregation functions package.""" + +from .median import Median +from .weighted_average import WeightedAverage +from .geometric_median import GeometricMedian +from .interface import AggregationFunctionInterface + +__all__ = ['Median', 'WeightedAverage', 'GeometricMedian', 'AggregationFunctionInterface'] diff --git a/openfl/component/aggregation_functions/geometric_median.py b/openfl/component/aggregation_functions/geometric_median.py new file mode 100644 index 0000000000..5f19c1adfc --- /dev/null +++ b/openfl/component/aggregation_functions/geometric_median.py @@ -0,0 +1,62 @@ +# Copyright (C) 2020-2021 Intel Corporation +# SPDX-License-Identifier: Apache-2.0 + +"""Geometric median module.""" + +from .interface import AggregationFunctionInterface +import numpy as np +from .weighted_average import weighted_average + + +def _geometric_median_objective(median, tensors, weights): + """Compute geometric median objective.""" + return sum([w * _l2dist(median, x) for w, x in zip(weights, tensors)]) + + +def geometric_median(tensors, weights, maxiter=4, eps=1e-5, ftol=1e-6): + """Compute geometric median of tensors with weights using Weiszfeld's Algorithm.""" + weights = np.asarray(weights) / sum(weights) + median = weighted_average(tensors, weights) + num_oracle_calls = 1 + + obj_val = _geometric_median_objective(median, tensors, weights) + + for _ in range(maxiter): + prev_obj_val = obj_val + weights = np.asarray([w / max(eps, _l2dist(median, x)) for w, x in zip(weights, tensors)]) + weights = weights / weights.sum() + median = weighted_average(tensors, weights) + num_oracle_calls += 1 + obj_val = _geometric_median_objective(median, tensors, weights) + if abs(prev_obj_val - obj_val) < ftol * obj_val: + break + return median + + +def _l2dist(p1, p2): + """L2 distance between p1, p2, each of which is a list of nd-arrays.""" + if p1.ndim != p2.ndim: + raise RuntimeError('Tensor shapes should be equal') + if p1.ndim < 2: + return _l2dist([np.expand_dims(x, axis=0) for x in [p1, p2]]) + return np.linalg.norm([np.linalg.norm(x1 - x2) for x1, x2 in zip(p1, p2)]) + + +class GeometricMedian(AggregationFunctionInterface): + """Geometric median aggregation.""" + + def call(self, local_tensors, _) -> np.ndarray: + """Aggregate tensors. + + Args: + agg_tensor_dict: Dict of (collaborator name, tensor) pairs to aggregate. + weights: array of floats representing data partition (sum up to 1) + db_iterator: iterator over history of all tensors. + Columns: ['tensor_name', 'round', 'tags', 'nparray'] + tensor_name: name of the tensor + fl_round: round number + tags: tuple of tags for this tensor + """ + tensors, weights = zip([(x.tensor, x.weight) for x in local_tensors]) + tensors, weights = np.array(tensors), np.array(weights) + return geometric_median(tensors, weights) diff --git a/openfl/component/aggregation_functions/interface.py b/openfl/component/aggregation_functions/interface.py new file mode 100644 index 0000000000..bdc3a62686 --- /dev/null +++ b/openfl/component/aggregation_functions/interface.py @@ -0,0 +1,41 @@ +# Copyright (C) 2020-2021 Intel Corporation +# SPDX-License-Identifier: Apache-2.0 +"""Aggregation function interface module.""" +from typing import Iterator, Tuple, List +from openfl.utilities import LocalTensor +import numpy as np +from abc import ABC, abstractmethod + +import pandas as pd + + +class AggregationFunctionInterface(ABC): + """Interface for specifying aggregation function.""" + + @abstractmethod + def call(self, + local_tensors: List[LocalTensor], + weights: np.ndarray, + db_iterator: Iterator[pd.Series], + tensor_name: str, + fl_round: int, + tags: Tuple[str]) -> np.ndarray: + """Aggregate tensors. + + Args: + local_tensors(list[openfl.utilities.LocalTensor]): List of local tensors to aggregate. + db_iterator: iterator over history of all tensors. + Columns: ['tensor_name', 'round', 'tags', 'nparray'] + tensor_name: name of the tensor + fl_round: round number + tags: tuple of tags for this tensor + """ + raise NotImplementedError + + def __call__(self, local_tensors, + db_iterator, + tensor_name, + fl_round, + tags): + """Use magic function for ease.""" + return self.call(local_tensors, db_iterator, tensor_name, fl_round, tags) diff --git a/openfl/component/aggregation_functions/median.py b/openfl/component/aggregation_functions/median.py new file mode 100644 index 0000000000..81ef01c05c --- /dev/null +++ b/openfl/component/aggregation_functions/median.py @@ -0,0 +1,26 @@ +# Copyright (C) 2020-2021 Intel Corporation +# SPDX-License-Identifier: Apache-2.0 + +"""Median module.""" + +from .interface import AggregationFunctionInterface +import numpy as np + + +class Median(AggregationFunctionInterface): + """Median aggregation.""" + + def call(self, local_tensors, _): + """Aggregate tensors. + + Args: + agg_tensor_dict: Dict of (collaborator name, tensor) pairs to aggregate. + weights: array of floats representing data partition (sum up to 1) + db_iterator: iterator over history of all tensors. + Columns: ['tensor_name', 'round', 'tags', 'nparray'] + tensor_name: name of the tensor + fl_round: round number + tags: tuple of tags for this tensor + """ + tensors = np.array([x.tensor for x in local_tensors]) + return np.median(tensors, axis=0) diff --git a/openfl/component/aggregation_functions/weighted_average.py b/openfl/component/aggregation_functions/weighted_average.py new file mode 100644 index 0000000000..e7ab596b4d --- /dev/null +++ b/openfl/component/aggregation_functions/weighted_average.py @@ -0,0 +1,31 @@ +# Copyright (C) 2020-2021 Intel Corporation +# SPDX-License-Identifier: Apache-2.0 + +"""Federated averaging module.""" + +from .interface import AggregationFunctionInterface +import numpy as np + + +def weighted_average(tensors, weights): + """Compute average.""" + return np.average(tensors, weights=weights, axis=0) + + +class WeightedAverage(AggregationFunctionInterface): + """Weighted average aggregation.""" + + def call(self, local_tensors, _): + """Aggregate tensors. + + Args: + agg_tensor_dict: Dict of (collaborator name, tensor) pairs to aggregate. + weights: array of floats representing data partition (sum up to 1) + db_iterator: iterator over history of all tensors. + Columns: ['tensor_name', 'round', 'tags', 'nparray'] + tensor_name: name of the tensor + fl_round: round number + tags: tuple of tags for this tensor + """ + tensors, weights = zip([(x.tensor, x.weight) for x in local_tensors]) + return weighted_average(tensors, weights) diff --git a/openfl/component/aggregator/aggregator.py b/openfl/component/aggregator/aggregator.py index 9efc5ec159..719cf9257a 100644 --- a/openfl/component/aggregator/aggregator.py +++ b/openfl/component/aggregator/aggregator.py @@ -741,7 +741,7 @@ def _compute_validation_related_task_metrics(self, task_name): # collaborator in our subset, and apply the correct # transformations to the tensorkey to resolve the aggregated # tensor for that round - agg_functions = self.assigner.get_aggregation_type_for_task(task_name) + agg_function = self.assigner.get_aggregation_type_for_task(task_name) task_key = TaskResultKey(task_name, collaborators_for_task[0], self.round_number) for tensor_key in self.collaborator_tasks_results[task_key]: tensor_name, origin, round_number, report, tags = tensor_key @@ -752,8 +752,8 @@ def _compute_validation_related_task_metrics(self, task_name): new_tags = tuple(list(tags[:-1])) agg_tensor_key = TensorKey(tensor_name, origin, round_number, report, new_tags) agg_tensor_name, agg_origin, agg_round_number, agg_report, agg_tags = agg_tensor_key - agg_results, agg_metadata_dict = self.tensor_db.get_aggregated_tensor( - agg_tensor_key, collaborator_weight_dict, agg_functions) + agg_results = self.tensor_db.get_aggregated_tensor( + agg_tensor_key, collaborator_weight_dict, aggregation_function=agg_function) if report: # Print the aggregated metric if agg_results is None: @@ -761,16 +761,12 @@ def _compute_validation_related_task_metrics(self, task_name): 'Aggregated metric {} could not be collected for round {}. ' 'Skipping reporting for this round'.format( agg_tensor_name, self.round_number)) - if agg_functions is not None: + if agg_function: self.logger.info('{0} {1}:\t{2:.4f}'.format( - agg_functions[0], agg_tensor_name, agg_results) + agg_function, agg_tensor_name, agg_results) ) else: self.logger.info('{0}:\t{1:.4f}'.format(agg_tensor_name, agg_results)) - for met in agg_metadata_dict: - self.logger.info('{0} {1}:\t{2:.4f}'.format( - met, agg_tensor_name, agg_metadata_dict[met]) - ) # TODO Add all of the logic for saving the model based # on best accuracy, lowest loss, etc. if 'validate_agg' in tags: diff --git a/openfl/databases/tensor_db.py b/openfl/databases/tensor_db.py index 6e439deb38..ff251a7404 100644 --- a/openfl/databases/tensor_db.py +++ b/openfl/databases/tensor_db.py @@ -8,7 +8,9 @@ from threading import Lock -from openfl.utilities import TensorKey +from openfl.utilities import TensorKey, LocalTensor +from openfl.component.aggregation_functions import (WeightedAverage, Median, GeometricMedian, + AggregationFunctionInterface) class TensorDB: @@ -20,6 +22,12 @@ class TensorDB: collaborator and aggregator has its own TensorDB. """ + aggregation_fns = { + 'weighted_average': WeightedAverage(), + 'median': Median(), + 'geometric_median': GeometricMedian() + } + def __init__(self): """Initialize.""" self.tensor_db = pd.DataFrame([], columns=[ @@ -101,7 +109,7 @@ def get_tensor_from_cache(self, tensor_key): return np.array(df['nparray'].iloc[0]) def get_aggregated_tensor(self, tensor_key, collaborator_weight_dict, - aggregation_functions): + aggregation_function=None): """ Determine whether all of the collaborator tensors are present for a given tensor key. @@ -113,16 +121,25 @@ def get_aggregated_tensor(self, tensor_key, collaborator_weight_dict, compute weighted average of all collaborators collaborator_weight_dict: List of collaborator names in federation and their respective weights - aggregation_functions: Call the underlying numpy aggregation + aggregation_function: Call the underlying numpy aggregation function. Default is just the weighted - average. ONLY THE FIRST FUNCTION will get - recorded as the aggregated tensor - + average. Returns: weighted_nparray if all collaborator values are present None if not all values are present """ + if aggregation_function is None: + aggregation_function = 'weighted_average' + if aggregation_function in self.aggregation_fns: + aggregation_function = self.aggregation_fns[aggregation_function] + elif not isinstance(aggregation_function, AggregationFunctionInterface): + raise NotImplementedError( + 'Aggregation function should either ' + + f'implement {AggregationFunctionInterface.__name__} interface ' + + 'or be one of ' + + f'{list(TensorDB.aggregation_fns.keys())}') + if len(collaborator_weight_dict) != 0: assert (np.abs( 1.0 - sum(collaborator_weight_dict.values()) @@ -130,9 +147,6 @@ def get_aggregated_tensor(self, tensor_key, collaborator_weight_dict, 'Collaborator weights do not sum to 1.0:' \ ' {}'.format(collaborator_weight_dict) - if aggregation_functions is None: - aggregation_functions = ['weighted_average'] - collaborator_names = collaborator_weight_dict.keys() agg_tensor_dict = {} @@ -164,41 +178,30 @@ def get_aggregated_tensor(self, tensor_key, collaborator_weight_dict, col, TensorKey( tensor_name, origin, report, fl_round, new_tags ))) - return None, {} + return None else: agg_tensor_dict[col] = raw_df.iloc[0] # agg_tensor_dict[col] = agg_tensor_dict[col] # * collaborator_weight_dict[col] - - concat_nparray = np.array(list(agg_tensor_dict.values())) - - aggregated_tensorkey_is_set = False - agg_metadata_dict = {} - - for aggregation_function in aggregation_functions: - if aggregation_function == 'weighted_average': - agg_nparray = np.average( - concat_nparray, - weights=np.array(list(collaborator_weight_dict.values())), - axis=0 - ) - else: - agg_func = getattr(np, aggregation_function, None) - if callable(agg_func): - agg_nparray = agg_func(concat_nparray, axis=0) - else: - raise KeyError( - '{} is not a valid numpy function'.format( - aggregation_function)) - - # TODO This enforces that only the first aggregation function - # will be associated with - if not aggregated_tensorkey_is_set: - # Cache aggregated tensor in TensorDB - self.cache_tensor({tensor_key: agg_nparray}) - aggregated_tensorkey_is_set = True - primary_agg = agg_nparray - else: - agg_metadata_dict[aggregation_function] = np.array(agg_nparray) - - return np.array(primary_agg), agg_metadata_dict + local_tensors = [LocalTensor(col_name=col_name, + tensor=agg_tensor_dict[col_name], + weight=collaborator_weight_dict[col_name]) + for col_name in collaborator_names] + + db_iterator = self._iterate() + agg_nparray = aggregation_function(local_tensors, + db_iterator, + tensor_name, + fl_round, + tags) + self.cache_tensor({tensor_key: agg_nparray}) + + return np.array(agg_nparray) + + def _iterate(self, order_by='round', ascending=False): + columns = ['round', 'nparray', 'tensor_name', 'tags'] + + for _, row in self.tensor_db[columns]\ + .sort_values(by=order_by, ascending=ascending)\ + .iterrows(): + yield row diff --git a/openfl/federated/plan/plan.py b/openfl/federated/plan/plan.py index 0a0b27a1da..8062f34c8d 100644 --- a/openfl/federated/plan/plan.py +++ b/openfl/federated/plan/plan.py @@ -49,11 +49,11 @@ def ignore_aliases(self, data): if frozen_yaml_path.exists(): Plan.logger.info(f"{yaml_path.name} is already frozen") return - frozen_yaml_path.write_text(dump(config, Dumper=NoAliasDumper)) + frozen_yaml_path.write_text(dump(config)) frozen_yaml_path.chmod(0o400) Plan.logger.info(f"{yaml_path.name} frozen successfully") else: - yaml_path.write_text(dump(config, Dumper=NoAliasDumper)) + yaml_path.write_text(dump(config)) @staticmethod def Parse(plan_config_path: Path, cols_config_path: Path = None, diff --git a/openfl/native/native.py b/openfl/native/native.py index 7e3e7a55fc..09e60f04f4 100644 --- a/openfl/native/native.py +++ b/openfl/native/native.py @@ -24,7 +24,7 @@ WORKSPACE_PREFIX = os.path.join(os.path.expanduser('~'), '.local', 'workspace') -def setup_plan(save=True): +def setup_plan(log_level='CRITICAL'): """ Dump the plan with all defaults + overrides set. @@ -38,42 +38,20 @@ def setup_plan(save=True): plan_config = 'plan/plan.yaml' cols_config = 'plan/cols.yaml' data_config = 'plan/data.yaml' + + getLogger().setLevel(log_level) plan = Plan.Parse(plan_config_path=Path(plan_config), cols_config_path=Path(cols_config), - data_config_path=Path(data_config)) - if save: - Plan.Dump(Path(plan_config), plan.config) + data_config_path=Path(data_config), + resolve=False) + getLogger().setLevel('INFO') return plan -def get_plan(return_complete=False): - """ - Return the flattened dictionary associated with the plan. - - To read the output in a human readable format, we recommend interpreting it - as follows: - - ``` - print(json.dumps(fx.get_plan(), indent=4, sort_keys=True)) - ``` - - Args: - return_complete : bool (default=False) - By default will not print the default file locations for each of - the templates - - Returns: - plan : dict - flattened dictionary of the current plan - """ - getLogger().setLevel('CRITICAL') - - plan_config = setup_plan().config - - getLogger().setLevel('INFO') - - flattened_config = flatten_preserve_lists(plan_config, '.')[0] +def flatten(config, return_complete=False): + """Flatten nested config.""" + flattened_config = flatten_preserve_lists(config, '.')[0] if not return_complete: keys_to_remove = [ k for k, v in flattened_config.items() @@ -98,16 +76,18 @@ def update_plan(override_config): Returns: None """ - plan_path = 'plan/plan.yaml' - flat_plan_config = get_plan(return_complete=True) + plan = setup_plan() + flat_plan_config = flatten(plan.config, return_complete=True) for k, v in override_config.items(): if k in flat_plan_config: - flat_plan_config[k] = v logger.info(f'Updating {k} to {v}... ') else: - logger.info(f'Key {k} not found in plan. Ignoring... ') - plan_config = unflatten(flat_plan_config, '.') - Plan.Dump(Path(plan_path), plan_config) + # TODO: We probably need to validate the new key somehow + logger.warn(f'Did not find {k} in config. Make sure it should exist. Creating...') + flat_plan_config[k] = v + plan.config = unflatten(flat_plan_config, '.') + plan.resolve() + return plan def unflatten(config, separator='.'): @@ -236,19 +216,7 @@ def run_experiment(collaborator_dict, override_config={}): path.insert(0, str(work)) # Update the plan if necessary - if len(override_config) > 0: - update_plan(override_config) - - # TODO: Fix this implementation. The full plan parsing is reused here, - # but the model and data will be overwritten based on user specifications - plan_config = 'plan/plan.yaml' - cols_config = 'plan/cols.yaml' - data_config = 'plan/data.yaml' - - plan = Plan.Parse(plan_config_path=Path(plan_config), - cols_config_path=Path(cols_config), - data_config_path=Path(data_config)) - + plan = update_plan(override_config) # Overwrite plan values plan.authorized_cols = list(collaborator_dict) tensor_pipe = plan.get_tensor_pipe() @@ -296,3 +264,11 @@ def run_experiment(collaborator_dict, override_config={}): model.rebuild_model( rounds_to_train - 1, aggregator.last_tensor_dict, validation=True) return model + + +def get_plan(indent=4, sort_keys=True): + """Get string representation of current Plan.""" + import json + plan = setup_plan() + flat_plan_config = flatten(plan.config) + return json.dumps(flat_plan_config, indent=indent, sort_keys=sort_keys) diff --git a/openfl/utilities/types.py b/openfl/utilities/types.py index a9699246c7..1d752f4d72 100644 --- a/openfl/utilities/types.py +++ b/openfl/utilities/types.py @@ -9,3 +9,4 @@ TaskResultKey = namedtuple('TaskResultKey', ['task_name', 'owner', 'round_number']) Metric = namedtuple('Metric', ['name', 'value']) +LocalTensor = namedtuple('LocalTensor', ['col_name', 'tensor', 'weight']) diff --git a/setup.py b/setup.py index 8bafe8ec3b..f606b16b48 100644 --- a/setup.py +++ b/setup.py @@ -44,6 +44,7 @@ 'openfl-workspace', 'openfl-docker', 'openfl-tutorials', + 'openfl.component.aggregation_functions' ], include_package_data=True, install_requires=[	Simple aggregation schemes on aggregator: geometric mean + median It would be great to have not only _Avg_ operator for tensor aggregation, but also few other simple operations like _Geometric Mean_, _Median_, etc.	@alexey-gruzdev, we do have some of this in place: I believe we interpret one of the flplan args as the aggregation function. If the name doesn't match our pre-defined strings, we assume it is a numpy function and invoke it. I'd like the end-result to be the same as others: you can declare arbitrary python classes in the flplan that provide the functionality, along with init kwargs, the function to invoke, and kwargs for that function. Essentially, our same generic task declaration method we use for the collaborators. This is one of the reasons Patrick and I have been talking about making the aggregator essentially "just another task runner". This would also have a nice advantage in that we wouldn't have to hard-code the logic that chooses which tensors are inputs for the aggregation function. Instead, it would use the same generic logic that determines the tensor input requirements for any task. @msheller that's great, if we have it in place - could you provide a simple example for the reference? Ideally, for better extendibility it would be great to have a python function to be passed to Python API entities & being used by OpenFL further. ``` def custom_aggregation(a: torch.Tensor, b: torch.Tensor) -> torch.Tensor: return (a + b)/2 fl_model = fx.run_experiment(collaborators, agg_function=custom_aggregation, override_config={'aggregator.settings.rounds_to_train':5}) ``` > you can declare arbitrary python classes in the flplan that provide the functionality, along with init kwargs, the function to invoke, and kwargs for that function Maybe I am wrong, however from Python user's perspective - is it a right place to define such kind of functionalities in FLPlan (.yaml)? @alexey-gruzdev This is a feature I added a few months ago. A limited set of custom aggregation types are supported at the task level. They can be called in the following way: ``` plan.yaml ... tasks: aggregated_model_validation: aggregation_type: median function: validate kwargs: apply: global batch_size: 32 metrics: - accuracy ... ``` and you can call it through the current Python API like this: ``` fl_model = fx.run_experiment(collaborators, agg_function=custom_aggregation, override_config={'aggregator.settings.rounds_to_train':5, 'tasks.aggregated_model_validation.aggregation_type': 'median'}) ``` These aggregation types are currently limited to simple numpy functions (see [here](https://github.com/intel/openfl/blob/develop/openfl/databases/tensor_db.py#L186-L192)). As both of you suggest, we should provide a way to define custom functions that can be applied to model weights and task metrics. There is likely a need to apply a specific aggregation function to certain task output, and possibly another for how to combine model weights. Because of this level of specificity my view is that the plan is the appropriate place to embed this information, but we could extend our existing python APIs so that the user has a cleaner way for expressing the function they want to use without modifying the plan directly.	2021-04-27T12:30:14	[]	[]
securefederatedai/openfl	securefederatedai__openfl-28	be9ea6ebf16f80c7ed84a775e61269c36375e984	diff --git a/openfl/native/native.py b/openfl/native/native.py index d085335351..702116ad89 100644 --- a/openfl/native/native.py +++ b/openfl/native/native.py @@ -263,7 +263,7 @@ def run_experiment(collaborator_dict, override_config={}): rounds_to_train = plan.config['aggregator']['settings']['rounds_to_train'] tensor_dict, holdout_params = split_tensor_dict_for_holdouts( logger, - plan.runner_.get_tensor_dict(False) + model.get_tensor_dict(False) ) model_snap = utils.construct_model_proto(tensor_dict=tensor_dict, @@ -285,7 +285,7 @@ def run_experiment(collaborator_dict, override_config={}): # Create the collaborators collaborators = { collaborator: create_collaborator( - plan, collaborator, model, aggregator + plan, collaborator, collaborator_dict[collaborator], aggregator ) for collaborator in plan.authorized_cols } @@ -293,14 +293,9 @@ def run_experiment(collaborator_dict, override_config={}): for col in plan.authorized_cols: collaborator = collaborators[col] - model.set_data_loader(collaborator_dict[col].data_loader) - - if round_num != 0: - model.rebuild_model(round_num, model_states[col]) collaborator.run_simulation() - - model_states[col] = model.get_tensor_dict(with_opt_vars=True) + model_states[col] = collaborator.task_runner.get_tensor_dict(with_opt_vars=True) # Set the weights for the final model model.rebuild_model(	Same validation metric values across collaborators in Jupyter notebooks Describe the bug When launching the federated training in Jupyter notebooks of `openfl-tutorials` folder, I noticed that different collaborators achieve the same validation metric values. That could be possible if collaborators had the same data, but the data is randomly split. Looks like there is an issue of defining the Data Loader for each collaborator. To Reproduce Steps to reproduce the behavior: 1. Open `openfl-tutorials/Federated_PyTorch_UNET_Tutorial.ipynb` in Jupyter. 2. Execute all cells. Expected behavior Collaborators have different metrics due to random data split. Screenshot ![image](https://user-images.githubusercontent.com/76161256/110151259-53b6a080-7df1-11eb-86bc-7ff5f787a4e9.png)		2021-03-09T11:30:51	[]	[]
TeamLEGWORK/LEGWORK	TeamLEGWORK__LEGWORK-112	3edb13b81d67d15977eead0a3e1849a4415f6fea	diff --git a/changelog.md b/changelog.md index f6cf0df..94da070 100644 --- a/changelog.md +++ b/changelog.md @@ -166,6 +166,10 @@ having two when we could just set `f_dom=2 f_orb` - Ensure `theta` is a co-latitude in `amplitude_modulation` ## 0.4.6 -TW 13/04/22 +TW 13/04/23 - [Issue [#109](https://github.com/TeamLEGWORK/LEGWORK/issues/109)] Upgrade `numpy` version - - Upgrade required to avoid error in `np.nan_to_num` usage - thanks to Jakob Stegmann for raising this \ No newline at end of file + - Upgrade required to avoid error in `np.nan_to_num` usage - thanks to Jakob Stegmann for raising this + +## 0.4.7 +TW 12/10/23 +- [Issue [#111](https://github.com/TeamLEGWORK/LEGWORK/issues/111)] Allow `plot_sources_on_sc` to change the underlying sensitivity curve directly with `sc_vis_settings` diff --git a/legwork/_version.py b/legwork/_version.py index 3dd3d2d..a34b2f6 100644 --- a/legwork/_version.py +++ b/legwork/_version.py @@ -1,1 +1,1 @@ -__version__ = "0.4.6" +__version__ = "0.4.7" diff --git a/legwork/evol.py b/legwork/evol.py index ae64836..10092cf 100644 --- a/legwork/evol.py +++ b/legwork/evol.py @@ -15,7 +15,7 @@ 'determine_stationarity'] -@jit +@jit(nopython=True) def de_dt(e, times, beta, c_0): # pragma: no cover """Compute eccentricity time derivative diff --git a/legwork/source.py b/legwork/source.py index 000f2c7..1f70beb 100644 --- a/legwork/source.py +++ b/legwork/source.py @@ -1192,7 +1192,7 @@ def plot_source_variables(self, xstr, ystr=None, which_sources=None, return vis.plot_1D_dist(x=x[which_sources], weights=weights, kwargs) def plot_sources_on_sc(self, snr_cutoff=0, fig=None, ax=None, show=True, - label="Stationary sources", kwargs): # pragma: no cover + label="Stationary sources", sc_vis_settings={}, kwargs): # pragma: no cover """Plot all sources in the class on the sensitivity curve Parameters @@ -1214,6 +1214,10 @@ def plot_sources_on_sc(self, snr_cutoff=0, fig=None, ax=None, show=True, label : `str` Label to use for the plotted points + sc_vis_settings : `dict` + Dictionary of parameters to pass to :meth:`~legwork.visualisation.plot_sensitivity_curve`, + e.g. {"y_quantity": "h_c"} will plot characteristic strain instead of ASD + kwargs : `various` Keyword arguments to be passed to plotting functions @@ -1250,7 +1254,7 @@ def plot_sources_on_sc(self, snr_cutoff=0, fig=None, ax=None, show=True, weights = self.weights[stat] if self.weights is not None else None fig, ax = vis.plot_sources_on_sc(f_dom=f_dom, snr=self.snr[stat], weights=weights, snr_cutoff=snr_cutoff, show=show, fig=fig, ax=ax, - label=label, self._sc_params, kwargs) + label=label, sc_vis_settings=sc_vis_settings, self._sc_params, kwargs) # show warnings for evolving sources circ_evol = self.get_source_mask(circular=True, stationary=False) diff --git a/legwork/visualisation.py b/legwork/visualisation.py index 0bd367c..4a72ba6 100644 --- a/legwork/visualisation.py +++ b/legwork/visualisation.py @@ -374,7 +374,7 @@ def plot_sensitivity_curve(frequency_range=None, y_quantity="ASD", fig=None, ax= def plot_sources_on_sc(f_dom, snr, weights=None, snr_cutoff=0, t_obs="auto", instrument="LISA", custom_psd=None, L="auto", approximate_R=False, - confusion_noise="auto", fig=None, ax=None, show=True, kwargs): + confusion_noise="auto", fig=None, ax=None, show=True, sc_vis_settings={}, kwargs): """Overlay stationary sources on the LISA sensitivity curve. Each source is plotted at its max snr harmonic frequency such that that its height above the curve is @@ -429,6 +429,10 @@ def plot_sources_on_sc(f_dom, snr, weights=None, snr_cutoff=0, t_obs="auto", show : `boolean` Whether to immediately show the plot or only return the Figure and Axis + sc_vis_settings : `dict` + Dictionary of parameters to pass to :meth:`~legwork.visualisation.plot_sensitivity_curve`, + e.g. {"y_quantity": "h_c"} will plot characteristic strain instead of ASD + kwargs : `various` This function is a wrapper on :func:`legwork.visualisation.plot_2D_dist` and each kwarg is passed directly to this function. For example, you can write `disttype="kde"` for a kde density plot @@ -446,7 +450,7 @@ def plot_sources_on_sc(f_dom, snr, weights=None, snr_cutoff=0, t_obs="auto", if fig is None or ax is None: fig, ax = plot_sensitivity_curve(show=False, t_obs=t_obs, instrument=instrument, L=L, custom_psd=custom_psd, approximate_R=approximate_R, - confusion_noise=confusion_noise) + confusion_noise=confusion_noise, sc_vis_settings) # work out which binaries are above the cutoff detectable = snr > snr_cutoff @@ -455,12 +459,14 @@ def plot_sources_on_sc(f_dom, snr, weights=None, snr_cutoff=0, t_obs="auto", return fig, ax # calculate asd that makes it so height above curve is snr - asd = snr[detectable] * np.sqrt(psd.power_spectral_density(f_dom[detectable])) + asd = (snr[detectable] * np.sqrt(psd.power_spectral_density(f_dom[detectable]))).to(u.Hz*(-1/2)) + h_c = asd np.sqrt(f_dom[detectable]) + use_h_c = ("y_quantity" in sc_vis_settings and sc_vis_settings["y_quantity"] == "h_c") # plot either a scatter or density plot of the detectable binaries ylims = ax.get_ylim() weights = weights[detectable] if weights is not None else None - fig, ax = plot_2D_dist(x=f_dom[detectable], y=asd.to(u.Hz(-1/2)), weights=weights, + fig, ax = plot_2D_dist(x=f_dom[detectable], y=h_c if use_h_c else asd, weights=weights, fig=fig, ax=ax, show=False, kwargs) ax.set_ylim(ylims)	y axis of the frequency versus ASD plot In the plot_sensitive_curve function, it provides an option of y-axis to be "h_c", while in the plot_source_on_sc function, there is no such an option. Sometimes, it would be necessary to have a such plot. Therefore, it would be good if the plot_source_on_sce function. Best wishes, Hailiang Chen		2023-10-12T15:36:57	[]	[]
TeamLEGWORK/LEGWORK	TeamLEGWORK__LEGWORK-72	f95affca969255a21e4d84afa01170882e2d543e	diff --git a/changelog.md b/changelog.md index 7734470..0c4b755 100644 --- a/changelog.md +++ b/changelog.md @@ -76,4 +76,15 @@ Small changes to visualisation code and updates to tutorials/demos with the new ## 0.1.7 TW 05/01/21 -- Link GitHub releases to Zenodo \ No newline at end of file +- Link GitHub releases to Zenodo + +## 0.2.0 +TW 05/01/21 +- A couple of changes to how confusion noise is handled + - End user can now access confusion noise functions directly through `get_confusion_noise` + - Added confusion noise fit from Huang+20 to be used with TianQin + - Added confusion noise fit from Thiele+21 which is based on a WDWD population with a metallicity dependent binary fraction +- TianQin psd function now include the confusion noise +- Change defaults used in `source` and `psd` + - Often defaults for arm length, observation time and confusion noise were previously LISA related, LEGWORK now automatically works out the defaults based on what instrument is chosen +- Bug fix: in `visualisation` avoid mixing floats with Quantities when filling in a sensitivity curve diff --git a/legwork/_version.py b/legwork/_version.py index 283b03a..d3ec452 100644 --- a/legwork/_version.py +++ b/legwork/_version.py @@ -1,1 +1,1 @@ -__version__ = "0.1.7" \ No newline at end of file +__version__ = "0.2.0" diff --git a/legwork/psd.py b/legwork/psd.py index 5d2c7a3..6a36029 100644 --- a/legwork/psd.py +++ b/legwork/psd.py @@ -88,10 +88,11 @@ def lisa_psd(f, t_obs=4 * u.yr, L=2.5e9 * u.m, approximate_R=False, confusion_no approximate_R : `boolean` Whether to approximate the response function (default: no) - confusion_noise : `various` - Galactic confusion noise. Acceptable inputs are 'robson19' (the confusion noise from Robson+19), - `None` (don't include confusion noise) or a custom function that gives the confusion noise at each - frequency for a given mission length where it would be called by running `noise(f, t_obs)` + confusion_noise : `various` + Galactic confusion noise. Acceptable inputs are one of the values listed in + :meth:`legwork.psd.get_confusion_noise` or a custom function that gives the confusion noise at each + frequency for a given mission length where it would be called by running `noise(f, t_obs)` and return + a value with units of inverse Hertz Returns ------- @@ -116,23 +117,6 @@ def Poms(f): def Pacc(f): return (3e-15)*2 (1 + (0.4e-3 / f)*2) (1 + (f / (8e-3))*4) - # galactic confusion noise (Robson+ Eq. 14) - def Sc(f, t_obs): - - # parameters from Robson+ 2019 Table 1 - lengths = np.array([0.5, 1.0, 2.0, 4.0]) u.yr - alpha = [0.133, 0.171, 0.165, 0.138] - beta = [243.0, 292.0, 299.0, -221.0] - kappa = [482.0, 1020.0, 611.0, 521.0] - gamma = [917.0, 1680.0, 1340.0, 1680.0] - fk = [2.58e-3, 2.15e-3, 1.73e-3, 1.13e-3] - - # find index of the closest length to inputted observation time - ind = np.abs(t_obs - lengths).argmin() - - return 9e-45 * f*(-7 / 3.) np.exp(-f*(alpha[ind]) + beta[ind] f * np.sin(kappa[ind] * f)) \ - * (1 + np.tanh(gamma[ind] * (fk[ind] - f))) - # calculate response function (either exactly or with approximation) fstar = (const.c / (2 * np.pi * L)).to(u.Hz).value if approximate_R: @@ -140,15 +124,11 @@ def Sc(f, t_obs): else: R = load_response_function(f, fstar) - # use confusion noise from Robson+19 - if confusion_noise == "robson19": - cn = Sc(f, t_obs) - # use a custom function for the confusion noise - elif confusion_noise is not None: - cn = confusion_noise(f, t_obs) - # don't include any confusion noise + # get confusion noise + if isinstance(confusion_noise, str) or confusion_noise is None: + cn = get_confusion_noise(f=f * u.Hz, t_obs=t_obs, model=confusion_noise).value else: - cn = np.zeros(np.shape(f)) if isinstance(f, (list, np.ndarray)) else 0.0 + cn = confusion_noise(f, t_obs).value L = L.to(u.m).value @@ -160,7 +140,7 @@ def Sc(f, t_obs): return psd / u.Hz -def tianqin_psd(f, L=np.sqrt(3) * 1e5 * u.km, t_obs=None, approximate_R=None, confusion_noise=None): +def tianqin_psd(f, L=np.sqrt(3) * 1e5 * u.km, t_obs=5 * u.yr, approximate_R=None, confusion_noise="huang20"): """Calculates the effective TianQin power spectral density sensitivity curve Using Eq. 13 from Huang+20, calculate the effective TianQin PSD for the sensitivity curve @@ -178,13 +158,16 @@ def tianqin_psd(f, L=np.sqrt(3) * 1e5 * u.km, t_obs=None, approximate_R=None, co Arm length t_obs : `float` - Ignored for this function + Observation time (default 5 years) approximate_R : `boolean` Ignored for this function - confusion_noise : `various` - Ignored for this function + confusion_noise : `various` + Galactic confusion noise. Acceptable inputs are one of the values listed in + :meth:`legwork.psd.get_confusion_noise` or a custom function that gives the confusion noise at each + frequency for a given mission length where it would be called by running `noise(f, t_obs)` and return + a value with units of inverse Hertz Returns ------- @@ -196,11 +179,19 @@ def tianqin_psd(f, L=np.sqrt(3) * 1e5 * u.km, t_obs=None, approximate_R=None, co Sx = 1e-24 * u.m*2 u.Hz*(-1) psd = 10 / (3 L*2) (4 * Sa / (2 * np.pi * f)*4 (1 + (1e-4 * u.Hz / f)) + Sx) \ * (1 + 0.6 * (f / fstar)2) + + # get confusion noise + if isinstance(confusion_noise, str) or confusion_noise is None: + cn = get_confusion_noise(f=f, t_obs=t_obs, model=confusion_noise) + else: + cn = confusion_noise(f, t_obs) + psd += cn + return psd.to(u.Hz(-1)) -def power_spectral_density(f, instrument="LISA", custom_psd=None, t_obs=4 * u.yr, L=None, - approximate_R=False, confusion_noise="robson19"): +def power_spectral_density(f, instrument="LISA", custom_psd=None, t_obs="auto", L="auto", + approximate_R=False, confusion_noise="auto"): """Calculates the effective power spectral density for all instruments. Parameters @@ -216,7 +207,7 @@ def power_spectral_density(f, instrument="LISA", custom_psd=None, t_obs=4 * u.yr even if it ignores some. t_obs : `float` - Observation time (default 4 years) + Observation time (default 4 years for LISA and 5 years for TianQin) L : `float` LISA arm length in metres @@ -224,10 +215,12 @@ def power_spectral_density(f, instrument="LISA", custom_psd=None, t_obs=4 * u.yr approximate_R : `boolean` Whether to approximate the response function (default: no) - confusion_noise : `various` - Galactic confusion noise. Acceptable inputs are 'robson19' (the confusion noise from Robson+19), - `None` (don't include confusion noise) or a custom function that gives the confusion noise at each - frequency for a given mission length where it would be called by running `noise(f, t_obs)` + confusion_noise : `various` + Galactic confusion noise. Acceptable inputs are either one of the values listed in + :meth:`legwork.psd.get_confusion_noise`, "auto" (automatically selects confusion noise based on + `instrument` - 'robson19' if LISA and 'huang20' if TianQin), or a custom function that gives the + confusion noise at each frequency for a given mission length where it would be called by running + `noise(f, t_obs)` and return a value with units of inverse Hertz Returns ------- @@ -235,12 +228,20 @@ def power_spectral_density(f, instrument="LISA", custom_psd=None, t_obs=4 * u.yr Effective power strain spectral density """ if instrument == "LISA": - if L is None: - L = 2.5e9 * u.m + # update any auto values to be instrument specific + L = 2.5e9 * u.m if L == "auto" else L + confusion_noise = "robson19" if confusion_noise == "auto" else confusion_noise + t_obs = 4 * u.yr if t_obs == "auto" else t_obs + + # calculate psd psd = lisa_psd(f=f, L=L, t_obs=t_obs, approximate_R=approximate_R, confusion_noise=confusion_noise) elif instrument == "TianQin": - if L is None: - L = np.sqrt(3) * 1e5 * u.km + # update any auto values to be instrument specific + L = np.sqrt(3) * 1e5 * u.km if L == "auto" else L + confusion_noise = "huang20" if confusion_noise == "auto" else confusion_noise + t_obs = 5 * u.yr if t_obs == "auto" else t_obs + + # calculate psd psd = tianqin_psd(f=f, L=L, t_obs=t_obs, approximate_R=approximate_R, confusion_noise=confusion_noise) elif instrument == "custom": psd = custom_psd(f=f, L=L, t_obs=t_obs, approximate_R=approximate_R, confusion_noise=confusion_noise) @@ -248,3 +249,160 @@ def power_spectral_density(f, instrument="LISA", custom_psd=None, t_obs=4 * u.yr raise ValueError("instrument: `{}` not recognised".format(instrument)) return psd + + +def get_confusion_noise_robson19(f, t_obs=4 * u.yr): + """Calculate the confusion noise using the model from Robson+19 Eq. 14 and Table 1 + + Also note that this fit is designed based on LISA sensitivity and so it is likely not sensible to apply + it to TianQin or other missions. + + Parameters + ---------- + f : `float/array` + Frequencies at which to calculate the confusion noise, must have units of frequency + t_obs : `float`, optional + Mission length, parameters are defined for 0.5, 1, 2 and 4 years, the closest mission length to the + one inputted will be used. By default 4 years. + + Returns + ------- + confusion_noise : `float/array` + The confusion noise at each frequency + """ + # erase the units for speed + f = f.to(u.Hz).value + + # parameters from Robson+ 2019 Table 1 + lengths = np.array([0.5, 1.0, 2.0, 4.0]) * u.yr + alpha = [0.133, 0.171, 0.165, 0.138] + beta = [243.0, 292.0, 299.0, -221.0] + kappa = [482.0, 1020.0, 611.0, 521.0] + gamma = [917.0, 1680.0, 1340.0, 1680.0] + fk = [2.58e-3, 2.15e-3, 1.73e-3, 1.13e-3] + + # find index of the closest length to inputted observation time + ind = np.abs(t_obs - lengths).argmin() + + confusion_noise = 9e-45 * f*(-7 / 3.) np.exp(-f*(alpha[ind]) + beta[ind] + f * np.sin(kappa[ind] * f))\ + * (1 + np.tanh(gamma[ind] * (fk[ind] - f))) * u.Hz*(-1) + return confusion_noise + + +def get_confusion_noise_huang20(f, t_obs=5 u.yr): + """Calculate the confusion noise using the model from Huang+20 Table II. Note that we set the confusion + noise to be exactly 0 outside of the range [1e-4, 1] Hz as the fits are not designed to be used outside + of this range. + + Also note that this fit is designed based on TianQin sensitivity and so it is likely not sensible to apply + it to LISA or other missions. + + Parameters + ---------- + f : `float/array` + Frequencies at which to calculate the confusion noise, must have units of frequency + t_obs : `float`, optional + Mission length, parameters are defined for 0.5, 1, 2, 4 and 5 years, the closest mission length to the + one inputted will be used. By default 5 years. + + Returns + ------- + confusion_noise : `float/array` + The confusion noise at each frequency + """ + + # define the mission lengths and corresponding coefficients + lengths = np.array([0.5, 1.0, 2.0, 4.0, 5.0]) * u.yr + coefficients = np.array([ + [-18.6, -18.6, -18.6, -18.6, -18.6], + [-1.22, -1.13, -1.45, -1.43, -1.51], + [0.009, -0.945, 0.315, -0.687, -0.71], + [-1.87, -1.02, -1.19, 0.24, -1.13], + [0.65, 4.05, -4.48, -0.15, -0.83], + [3.6, -4.5, 10.8, -1.8, 13.2], + [-4.6, -0.5, -9.4, -3.2, -19.1] + ]) + + # find the nearest mission length + ind = np.abs(t_obs - lengths).argmin() + + # start with no confusion noise and add each coefficient using Huang+20 Table 1 + f = f.to(u.Hz).value + confusion_noise = np.zeros_like(f) + x = np.log10(f / 1e-3) + for i in range(len(coefficients)): + xi = x*i + ai = coefficients[i][ind] + confusion_noise += ai xi + confusion_noise = 10*(confusion_noise) u.Hz*(-1/2) + + # remove confusion noise outside of TianQin regime (fit doesn't apply outside of regime) + confusion_noise[np.logical_or(f < 1e-4, f > 1e0)] = 0.0 u.Hz(-1/2) + + # square the result as Huang+20 given the sensitivity not psd + return confusion_noise2 + + +def get_confusion_noise_thiele21(f): + """Calculate the confusion noise using the model from Thiele+20 Eq. 16 and Table 1. This fit uses a + metallicity-dependent binary fraction. + + Note: This fit only applies to LISA and only when the mission length is 4 years. + + Parameters + ---------- + f : `float/array` + Frequencies at which to calculate the confusion noise, must have units of frequency + + Returns + ------- + confusion_noise : `float/array` + The confusion noise at each frequency + """ + x = np.log10(f.to(u.Hz).value) + t_obs = 4 * u.yr + coefficients = [-540.315450, -554.061115, -233.748834, -44.021504, -3.112634] + confusion_noise = 10*np.poly1d(coefficients[::-1])(x) t_obs.to(u.s).value + return confusion_noise * u.Hz*(-1) + + +def get_confusion_noise(f, model, t_obs="auto"): + """Calculate the confusion noise for a particular model + + Parameters + ---------- + f : `float/array` + Frequencies at which to calculate the confusion noise, must have units of frequency + model : str, optional + Which model to use for the confusion noise. Must be one of 'robson19', 'huang20', 'thiele21' or None. + t_obs : `float`, optional + Mission length. Default is 4 years for robson19 and thiele21 and 5 years for huang20. + + Returns + ------- + confusion_noise : `float/array` + The confusion noise at each frequency. + + Raises + ------ + ValueError + When a model other than those defined above is used. + """ + if model == "robson19": + t_obs = 4 u.yr if t_obs == "auto" else t_obs + return get_confusion_noise_robson19(f=f, t_obs=t_obs) + elif model == "huang20": + t_obs = 5 * u.yr if t_obs == "auto" else t_obs + return get_confusion_noise_huang20(f=f, t_obs=t_obs) + elif model == "thiele21": + if t_obs == 4 * u.yr or t_obs == "auto": + return get_confusion_noise_thiele21(f=f) + else: + error = "Invalid mission length: Thiele+21 confusion noise is only fit for `t_obs=4u.yr`" + raise ValueError(error) + elif model is None: + f = f.to(u.Hz).value + return np.zeros_like(f) u.Hz*(-1) if isinstance(f, (list, np.ndarray)) else 0.0 u.Hz*(-1) + else: + raise ValueError("confusion noise model: `{}` not recognised".format(model)) diff --git a/legwork/source.py b/legwork/source.py index c33afd6..0e2ec42 100644 --- a/legwork/source.py +++ b/legwork/source.py @@ -76,7 +76,7 @@ class Source(): sc_params : `dict` Parameters for interpolated sensitivity curve. Include any of ``instrument``, ``custom_psd``, ``t_obs``, ``L``, ``approximate_R`` and ``confusion_noise``. Default values are: "LISA", None, - 4 years, 2.5e9, 19.09e-3, False and 'robson19'. + "auto", "auto", False and "auto". Attributes ---------- @@ -170,10 +170,10 @@ def __init__(self, m_1, m_2, ecc, dist, n_proc=1, f_orb=None, a=None, position=N default_sc_params = { "instrument": "LISA", "custom_psd": None, - "t_obs": 4 u.yr, - "L": 2.5e9 * u.m, + "t_obs": "auto", + "L": "auto", "approximate_R": False, - "confusion_noise": 'robson19' + "confusion_noise": 'auto' } default_sc_params.update(sc_params) self._sc_params = default_sc_params @@ -347,10 +347,10 @@ def update_sc_params(self, sc_params): default_sc_params = { "instrument": "LISA", "custom_psd": None, - "t_obs": 4 * u.yr, - "L": 2.5e9 * u.m, + "t_obs": "auto", + "L": "auto", "approximate_R": False, - "confusion_noise": "robson19" + "confusion_noise": "auto" } if sc_params is not None: default_sc_params.update(sc_params) @@ -358,7 +358,7 @@ def update_sc_params(self, sc_params): self._sc_params = default_sc_params self.set_sc() - def get_source_mask(self, circular=None, stationary=None, t_obs=4 * u.yr): + def get_source_mask(self, circular=None, stationary=None, t_obs=None): """Produce a mask of the sources. Create a mask based on whether binaries are circular or eccentric and stationary or evolving. @@ -373,7 +373,7 @@ def get_source_mask(self, circular=None, stationary=None, t_obs=4 * u.yr): ``None`` means either, ``True`` means only stationary binaries and ``False`` means only evolving t_obs : `float` - Observation time + Observation time, default is value from `self._sc_params` Returns ------- @@ -392,6 +392,9 @@ def get_source_mask(self, circular=None, stationary=None, t_obs=4 * u.yr): if stationary is None: stat_mask = np.repeat(True, self.n_sources) elif stationary is True or stationary is False: + t_obs = self._sc_params["t_obs"] if t_obs is None else t_obs + if t_obs == "auto": + t_obs = 4 * u.yr if self._sc_params["instrument"] == "LISA" else 5 * u.yr stat_mask = evol.determine_stationarity(m_c=self.m_c, f_orb_i=self.f_orb, t_evol=t_obs, ecc_i=self.ecc, stat_tol=self.stat_tol) if stationary is False: @@ -603,14 +606,14 @@ def get_snr(self, t_obs=None, instrument=None, custom_psd=None, n_step=100, verbose=verbose) return snr - def get_snr_stationary(self, t_obs=4 * u.yr, instrument="LISA", custom_psd=None, which_sources=None, + def get_snr_stationary(self, t_obs=None, instrument=None, custom_psd=None, which_sources=None, verbose=False): """Computes the SNR assuming a stationary binary Parameters ---------- t_obs : `array` - Observation duration (default: 4 years) + Observation duration (default: follow sc_params) instrument : `{{ 'LISA', 'TianQin', 'custom' }}` Instrument to observe with. If 'custom' then ``custom_psd`` must be supplied. @@ -633,6 +636,11 @@ def get_snr_stationary(self, t_obs=4 * u.yr, instrument="LISA", custom_psd=None, if which_sources is None: which_sources = np.repeat(True, self.n_sources) + instrument = self._sc_params["instrument"] if instrument is None else instrument + t_obs = self._sc_params["t_obs"] if t_obs is None else t_obs + if t_obs == "auto": + t_obs = 4 * u.yr if instrument == "LISA" else 5 * u.yr + insp_sources = np.logical_and(which_sources, np.logical_not(self.merged)) snr = np.zeros(self.n_sources) e_mask = np.logical_and(self.ecc > self.ecc_tol, insp_sources) @@ -703,14 +711,14 @@ def get_snr_stationary(self, t_obs=4 * u.yr, instrument="LISA", custom_psd=None, return snr[which_sources] - def get_snr_evolving(self, t_obs, instrument="LISA", custom_psd=None, n_step=100, which_sources=None, + def get_snr_evolving(self, t_obs=None, instrument=None, custom_psd=None, n_step=100, which_sources=None, verbose=False): """Computes the SNR assuming an evolving binary Parameters ---------- t_obs : `array` - Observation duration (default: 4 years) + Observation duration (default: follow sc_params) instrument : `{{ 'LISA', 'TianQin', 'custom' }}` Instrument to observe with. If 'custom' then ``custom_psd`` must be supplied. @@ -738,6 +746,11 @@ def get_snr_evolving(self, t_obs, instrument="LISA", custom_psd=None, n_step=100 if which_sources is None: which_sources = np.repeat(True, self.n_sources) + instrument = self._sc_params["instrument"] if instrument is None else instrument + t_obs = self._sc_params["t_obs"] if t_obs is None else t_obs + if t_obs == "auto": + t_obs = 4 * u.yr if instrument == "LISA" else 5 * u.yr + insp_sources = np.logical_and(which_sources, np.logical_not(self.merged)) e_mask = np.logical_and(self.ecc > self.ecc_tol, insp_sources) c_mask = np.logical_and(self.ecc <= self.ecc_tol, insp_sources) @@ -1125,7 +1138,7 @@ def plot_sources_on_sc(self, snr_cutoff=0, fig=None, ax=None, show=True, *kwarg class Stationary(Source): """Subclass for sources that are stationary""" - def get_snr(self, t_obs=4 u.yr, instrument="LISA", custom_psd=None, verbose=False): + def get_snr(self, t_obs=None, instrument=None, custom_psd=None, verbose=False): self.snr = self.get_snr_stationary(t_obs=t_obs, instrument=instrument, custom_psd=custom_psd, verbose=verbose) return self.snr @@ -1134,7 +1147,7 @@ def get_snr(self, t_obs=4 * u.yr, instrument="LISA", custom_psd=None, verbose=Fa class Evolving(Source): """Subclass for sources that are evolving""" - def get_snr(self, t_obs=4 * u.yr, instrument="LISA", custom_psd=None, n_step=100, verbose=False): + def get_snr(self, t_obs=None, instrument=None, custom_psd=None, n_step=100, verbose=False): self.snr = self.get_snr_evolving(t_obs=t_obs, n_step=n_step, instrument=instrument, custom_psd=custom_psd, verbose=verbose) return self.snr diff --git a/legwork/visualisation.py b/legwork/visualisation.py index aa5b7b4..90b510a 100644 --- a/legwork/visualisation.py +++ b/legwork/visualisation.py @@ -311,7 +311,7 @@ def plot_sensitivity_curve(frequency_range=None, y_quantity="ASD", fig=None, ax= # plot the curve and fill if needed ax.loglog(frequency_range, noise_amplitude, color=color, label=label, linewidth=linewidth) if fill: - ax.fill_between(frequency_range, 0, noise_amplitude, alpha=alpha, color=color) + ax.fill_between(frequency_range, 0 * noise_amplitude.unit, noise_amplitude, alpha=alpha, color=color) # adjust labels, sizes and frequency limits to plot is flush to the edges ax.set_xlabel(r'Frequency [$\rm Hz$]') @@ -329,9 +329,9 @@ def plot_sensitivity_curve(frequency_range=None, y_quantity="ASD", fig=None, ax= return fig, ax -def plot_sources_on_sc_circ_stat(f_orb, h_0_2, snr, weights=None, snr_cutoff=0, t_obs=4 * u.yr, - instrument="LISA", custom_psd=None, L=2.5e9 * u.m, approximate_R=False, - confusion_noise=True, fig=None, ax=None, show=True, kwargs): +def plot_sources_on_sc_circ_stat(f_orb, h_0_2, snr, weights=None, snr_cutoff=0, t_obs="auto", + instrument="LISA", custom_psd=None, L="auto", approximate_R=False, + confusion_noise="auto", fig=None, ax=None, show=True, kwargs): """Overlay circular/stationary sources on the LISA sensitivity curve. Each source is plotted at its gravitational wave frequency (n = 2) such that its height above the curve @@ -362,18 +362,20 @@ def plot_sources_on_sc_circ_stat(f_orb, h_0_2, snr, weights=None, snr_cutoff=0, even if it ignores some. t_obs : `float` - Observation time (default 4 years) + Observation time (default "auto") L : `float` - LISA arm length in metres + Arm length approximate_R : `boolean` Whether to approximate the response function (default: no) - confusion_noise : `various` - Galactic confusion noise. Acceptable inputs are 'robson19' (the confusion noise from Robson+19), - `None` (don't include confusion noise) or a custom function that gives the confusion noise at each - frequency for a given mission length where it would be called by running `noise(f, t_obs)` + confusion_noise : `various` + Galactic confusion noise. Acceptable inputs are either one of the values listed in + :meth:`legwork.psd.get_confusion_noise`, "auto" (automatically selects confusion noise based on + `instrument` - 'robson19' if LISA and 'huang20' if TianQin), or a custom function that gives the + confusion noise at each frequency for a given mission length where it would be called by running + `noise(f, t_obs)` and return a value with units of inverse Hertz fig: `matplotlib Figure` A figure on which to plot the distribution. Both `ax` and `fig` must be supplied for either to be used @@ -425,9 +427,9 @@ def plot_sources_on_sc_circ_stat(f_orb, h_0_2, snr, weights=None, snr_cutoff=0, return fig, ax -def plot_sources_on_sc_ecc_stat(f_dom, snr, weights=None, snr_cutoff=0, t_obs=4 * u.yr, - instrument="LISA", custom_psd=None, L=2.5e9 * u.m, approximate_R=False, - confusion_noise=True, fig=None, ax=None, show=True, kwargs): +def plot_sources_on_sc_ecc_stat(f_dom, snr, weights=None, snr_cutoff=0, t_obs="auto", + instrument="LISA", custom_psd=None, L="auto", approximate_R=False, + confusion_noise="auto", fig=None, ax=None, show=True, kwargs): """Overlay eccentric/stationary sources on the LISA sensitivity curve. Each source is plotted at its max snr harmonic frequency such that that its height above the curve is @@ -455,18 +457,20 @@ def plot_sources_on_sc_ecc_stat(f_dom, snr, weights=None, snr_cutoff=0, t_obs=4 even if it ignores some. t_obs : `float` - Observation time (default 4 years) + Observation time (default auto) L : `float` - LISA arm length in metres + Arm length in metres approximate_R : `boolean` Whether to approximate the response function (default: no) - confusion_noise : `various` - Galactic confusion noise. Acceptable inputs are 'robson19' (the confusion noise from Robson+19), - `None` (don't include confusion noise) or a custom function that gives the confusion noise at each - frequency for a given mission length where it would be called by running `noise(f, t_obs)` + confusion_noise : `various` + Galactic confusion noise. Acceptable inputs are either one of the values listed in + :meth:`legwork.psd.get_confusion_noise`, "auto" (automatically selects confusion noise based on + `instrument` - 'robson19' if LISA and 'huang20' if TianQin), or a custom function that gives the + confusion noise at each frequency for a given mission length where it would be called by running + `noise(f, t_obs)` and return a value with units of inverse Hertz fig: `matplotlib Figure` A figure on which to plot the distribution. Both `ax` and `fig` must be supplied for either to be used	Visualisation function mixes quantity with dimensionless constants This doesn't always produce an error, it seems to depend on your version of astropy/python but either way needs to be fixed. I'll sort it out in the next PR, the exact function is `plot_sensitivity_curve` when `fill=True`. A quick fix for this in the meantime is to include this at the top of any files that use this plotting function ``` from astropy.visualization import quantity_support quantity_support() ```		2022-01-05T17:55:39	[]	[]
PyFilesystem/pyfilesystem2	PyFilesystem__pyfilesystem2-518	a6ea045e766c76bae2e2fde19294c8275773ffde	diff --git a/CHANGELOG.md b/CHANGELOG.md index ab0d827c..8a3f73be 100644 --- a/CHANGELOG.md +++ b/CHANGELOG.md @@ -13,6 +13,8 @@ and this project adheres to [Semantic Versioning](http://semver.org/). - Support more lenient usernames and group names in FTP servers ([#507](https://github.com/PyFilesystem/pyfilesystem2/pull/507)). Closes [#506](https://github.com/PyFilesystem/pyfilesystem2/issues/506). +- Removed dependency on pytz ([#518](https://github.com/PyFilesystem/pyfilesystem2/pull/518)). + Closes [#516](https://github.com/PyFilesystem/pyfilesystem2/issues/518). ### Fixed diff --git a/fs/_ftp_parse.py b/fs/_ftp_parse.py index 0061f875..42c0720e 100644 --- a/fs/_ftp_parse.py +++ b/fs/_ftp_parse.py @@ -3,17 +3,20 @@ from __future__ import unicode_literals import unicodedata -import datetime import re import time +from datetime import datetime -from pytz import UTC +try: + from datetime import timezone +except ImportError: + from ._tzcompat import timezone # type: ignore from .enums import ResourceType from .permissions import Permissions -EPOCH_DT = datetime.datetime.fromtimestamp(0, UTC) +EPOCH_DT = datetime.fromtimestamp(0, timezone.utc) RE_LINUX = re.compile( @@ -98,7 +101,7 @@ def _parse_time(t, formats): day = _t.tm_mday hour = _t.tm_hour minutes = _t.tm_min - dt = datetime.datetime(year, month, day, hour, minutes, tzinfo=UTC) + dt = datetime(year, month, day, hour, minutes, tzinfo=timezone.utc) epoch_time = (dt - EPOCH_DT).total_seconds() return epoch_time diff --git a/fs/_tzcompat.py b/fs/_tzcompat.py new file mode 100644 index 00000000..282859b2 --- /dev/null +++ b/fs/_tzcompat.py @@ -0,0 +1,30 @@ +"""Compatibility shim for python2's lack of datetime.timezone. + +This is the example code from the Python 2 documentation: +https://docs.python.org/2.7/library/datetime.html#tzinfo-objects +""" + +from datetime import tzinfo, timedelta + + +ZERO = timedelta(0) + + +class UTC(tzinfo): + """UTC""" + + def utcoffset(self, dt): + return ZERO + + def tzname(self, dt): + return "UTC" + + def dst(self, dt): + return ZERO + + +utc = UTC() + + +class timezone: + utc = utc diff --git a/fs/time.py b/fs/time.py index cdf06061..52a59a77 100644 --- a/fs/time.py +++ b/fs/time.py @@ -7,17 +7,16 @@ import typing from calendar import timegm from datetime import datetime -from pytz import UTC, timezone + +try: + from datetime import timezone +except ImportError: + from ._tzcompat import timezone # type: ignore if typing.TYPE_CHECKING: from typing import Optional -utcfromtimestamp = datetime.utcfromtimestamp -utclocalize = UTC.localize -GMT = timezone("GMT") - - def datetime_to_epoch(d): # type: (datetime) -> int """Convert datetime to epoch.""" @@ -39,4 +38,6 @@ def epoch_to_datetime(t): # noqa: D103 def epoch_to_datetime(t): # type: (Optional[int]) -> Optional[datetime] """Convert epoch time to a UTC datetime.""" - return utclocalize(utcfromtimestamp(t)) if t is not None else None + if t is None: + return None + return datetime.fromtimestamp(t, tz=timezone.utc) diff --git a/setup.cfg b/setup.cfg index d6aa8943..06dac8f0 100644 --- a/setup.cfg +++ b/setup.cfg @@ -43,7 +43,6 @@ setup_requires = setuptools >=38.3.0 install_requires = appdirs~=1.4.3 - pytz setuptools six ~=1.10 enum34 ~=1.1.6 ; python_version < '3.4'	Move away from pytz pytz is known for doing funky things that aren't really compatible with the rest of the python ecosystem and in some cases simply don't make sense. Their timezone objects aren't even compatible with actual Python tzinfo objects. Here's a discussion from matplotlib where they decided to drop it in favour of dateutil, which has also been recommended by the python documentation for years now: https://github.com/matplotlib/matplotlib/issues/10443/ Alternatively (if you're willing to drop support for Python 3.8 and below), Python 3.9 has a new built-in zoneinfo module in the stdlib that provides the common use case of just loading timezone objects by name with accurate timezone info.	This article provides a great writeup of why pytz is dangerous: https://blog.ganssle.io/articles/2018/03/pytz-fastest-footgun.html > This article provides a great writeup of why pytz is dangerous: > > https://blog.ganssle.io/articles/2018/03/pytz-fastest-footgun.html Interesting. `pytz` doesn't [appear to be used](https://github.com/PyFilesystem/pyfilesystem2/search?q=pytz) in too many places in PyFilesystem, so do you wanna create a suitable PR? Sure!	2022-01-28T06:21:03	[]	[]
PyFilesystem/pyfilesystem2	PyFilesystem__pyfilesystem2-475	b88f43204960d69db6efb96e01f52b4f9c9c551e	diff --git a/CHANGELOG.md b/CHANGELOG.md index 62cc085c..f7b98bc1 100644 --- a/CHANGELOG.md +++ b/CHANGELOG.md @@ -40,6 +40,8 @@ and this project adheres to [Semantic Versioning](http://semver.org/). - `WrapCachedDir.isdir` and `WrapCachedDir.isfile` raising a `ResourceNotFound` error on non-existing path ([#470](https://github.com/PyFilesystem/pyfilesystem2/pull/470)). - `FTPFS` not listing certain entries with sticky/SUID/SGID permissions set by Linux server ([#473](https://github.com/PyFilesystem/pyfilesystem2/pull/473)). Closes [#451](https://github.com/PyFilesystem/pyfilesystem2/issues/451). +- `scandir` iterator not being closed explicitly in `OSFS.scandir`, occasionally causing a `ResourceWarning` + to be thrown. Closes [#311](https://github.com/PyFilesystem/pyfilesystem2/issues/311). ## [2.4.12] - 2021-01-14 diff --git a/fs/osfs.py b/fs/osfs.py index 3b35541c..5beb16bf 100644 --- a/fs/osfs.py +++ b/fs/osfs.py @@ -475,6 +475,7 @@ def _scandir(self, path, namespaces=None): # type: (Text, Optional[Collection[Text]]) -> Iterator[Info] self.check() namespaces = namespaces or () + requires_stat = not {"details", "stat", "access"}.isdisjoint(namespaces) _path = self.validatepath(path) if _WINDOWS_PLATFORM: sys_path = os.path.join( @@ -483,39 +484,47 @@ def _scandir(self, path, namespaces=None): else: sys_path = self._to_sys_path(_path) # type: ignore with convert_os_errors("scandir", path, directory=True): - for dir_entry in scandir(sys_path): - info = { - "basic": { - "name": fsdecode(dir_entry.name), - "is_dir": dir_entry.is_dir(), - } - } - if "details" in namespaces: - stat_result = dir_entry.stat() - info["details"] = self._make_details_from_stat(stat_result) - if "stat" in namespaces: - stat_result = dir_entry.stat() - info["stat"] = { - k: getattr(stat_result, k) - for k in dir(stat_result) - if k.startswith("st_") - } - if "lstat" in namespaces: - lstat_result = dir_entry.stat(follow_symlinks=False) - info["lstat"] = { - k: getattr(lstat_result, k) - for k in dir(lstat_result) - if k.startswith("st_") + scandir_iter = scandir(sys_path) + try: + for dir_entry in scandir_iter: + info = { + "basic": { + "name": fsdecode(dir_entry.name), + "is_dir": dir_entry.is_dir(), + } } - if "link" in namespaces: - info["link"] = self._make_link_info( - os.path.join(sys_path, dir_entry.name) - ) - if "access" in namespaces: - stat_result = dir_entry.stat() - info["access"] = self._make_access_from_stat(stat_result) - - yield Info(info) + if requires_stat: + stat_result = dir_entry.stat() + if "details" in namespaces: + info["details"] = self._make_details_from_stat( + stat_result + ) + if "stat" in namespaces: + info["stat"] = { + k: getattr(stat_result, k) + for k in dir(stat_result) + if k.startswith("st_") + } + if "access" in namespaces: + info["access"] = self._make_access_from_stat( + stat_result + ) + if "lstat" in namespaces: + lstat_result = dir_entry.stat(follow_symlinks=False) + info["lstat"] = { + k: getattr(lstat_result, k) + for k in dir(lstat_result) + if k.startswith("st_") + } + if "link" in namespaces: + info["link"] = self._make_link_info( + os.path.join(sys_path, dir_entry.name) + ) + + yield Info(info) + finally: + if sys.version_info >= (3, 6): + scandir_iter.close() else:	ResourceWarning: unclosed scandir iterator Getting this warning under Python 3.7 for scandir(): /home/ajung/.buildout/eggs/fs-2.4.8-py3.7.egg/fs/osfs.py:521: ResourceWarning: unclosed scandir iterator <posix.ScandirIterator object at 0x7f61d00596b0> yield Info(info)	Do you have some code to reproduce this? This will happen if you have an iterator to scandir that isn't full exhausted when it goes out of scope. It's not immediately obvious why it wouldn't close the native scandir iterator at the same time.	2021-03-27T03:21:07	[]	[]
PyFilesystem/pyfilesystem2	PyFilesystem__pyfilesystem2-474	60b0a6347e3bbcee4a2a6261a5e1b3166c45e192	diff --git a/CHANGELOG.md b/CHANGELOG.md index aacded93..62cc085c 100644 --- a/CHANGELOG.md +++ b/CHANGELOG.md @@ -13,6 +13,8 @@ and this project adheres to [Semantic Versioning](http://semver.org/). - Added FTP over TLS (FTPS) support to FTPFS. Closes [#437](https://github.com/PyFilesystem/pyfilesystem2/issues/437), [#449](https://github.com/PyFilesystem/pyfilesystem2/pull/449). +- `PathError` now supports wrapping an exception using the `exc` argument. + Closes [#453](https://github.com/PyFilesystem/pyfilesystem2/issues/453). ### Changed diff --git a/fs/errors.py b/fs/errors.py index 25625e28..2448c7a6 100644 --- a/fs/errors.py +++ b/fs/errors.py @@ -139,13 +139,14 @@ class PathError(FSError): default_message = "path '{path}' is invalid" - def __init__(self, path, msg=None): # noqa: D107 - # type: (Text, Optional[Text]) -> None + def __init__(self, path, msg=None, exc=None): # noqa: D107 + # type: (Text, Optional[Text], Optional[Exception]) -> None self.path = path + self.exc = exc super(PathError, self).__init__(msg=msg) def __reduce__(self): - return type(self), (self.path, self._msg) + return type(self), (self.path, self._msg, self.exc) class NoSysPath(PathError):	convert_os_errors error when converting errno.ENAMETOOLONG to PathError Using convert_os_errors to convert errno.ENAMETOOLONG to PathError results in unexpected keyword arguement 'exc' error. Seems that [this line](https://github.com/PyFilesystem/pyfilesystem2/blob/8784fd61797fac654511f16fd50c38c8a1675f16/fs/error_tools.py#L89) will set exc=exc_value however PathError does not take in exc parameter. The following example code: ``` import errno from fs.error_tools import convert_os_errors from fs import errors as fserrors try: from fs import open_fs f = open_fs('/tmp') f.open('x'*500, mode='w') except OSError as e: with convert_os_errors("foo", "test"): raise e ``` results in: TypeError: __init__() got an unexpected keyword argument 'exc'		2021-03-27T02:24:49	[]	[]
PyFilesystem/pyfilesystem2	PyFilesystem__pyfilesystem2-473	4efe083ebb1a43aafd0e171b3b20d6b546c86756	diff --git a/CHANGELOG.md b/CHANGELOG.md index c99ffcc3..aacded93 100644 --- a/CHANGELOG.md +++ b/CHANGELOG.md @@ -25,7 +25,7 @@ and this project adheres to [Semantic Versioning](http://semver.org/). test suites. - `FSTestCases` now builds the large data required for `upload` and `download` tests only once in order to reduce the total testing time. -- `MemoryFS.move` and `MemoryFS.movedir` will now avoid copying data. +- `MemoryFS.move` and `MemoryFS.movedir` will now avoid copying data. Closes [#452](https://github.com/PyFilesystem/pyfilesystem2/issues/452). ### Fixed @@ -36,6 +36,8 @@ and this project adheres to [Semantic Versioning](http://semver.org/). - Avoid creating a new connection on every call of `FTPFS.upload`. Closes [#455](https://github.com/PyFilesystem/pyfilesystem2/issues/455). - `WrapReadOnly.removetree` not raising a `ResourceReadOnly` when called. Closes [#468](https://github.com/PyFilesystem/pyfilesystem2/issues/468). - `WrapCachedDir.isdir` and `WrapCachedDir.isfile` raising a `ResourceNotFound` error on non-existing path ([#470](https://github.com/PyFilesystem/pyfilesystem2/pull/470)). +- `FTPFS` not listing certain entries with sticky/SUID/SGID permissions set by Linux server ([#473](https://github.com/PyFilesystem/pyfilesystem2/pull/473)). + Closes [#451](https://github.com/PyFilesystem/pyfilesystem2/issues/451). ## [2.4.12] - 2021-01-14 diff --git a/fs/_ftp_parse.py b/fs/_ftp_parse.py index 9e15a265..a9088ab4 100644 --- a/fs/_ftp_parse.py +++ b/fs/_ftp_parse.py @@ -19,7 +19,8 @@ RE_LINUX = re.compile( r""" ^ - ([ldrwx-]{10}) + ([-dlpscbD]) + ([r-][w-][xsS-][r-][w-][xsS-][r-][w-][xtT-][\.\+]?) \s+? (\d+) \s+? @@ -110,14 +111,14 @@ def _decode_linux_time(mtime): def decode_linux(line, match): - perms, links, uid, gid, size, mtime, name = match.groups() - is_link = perms.startswith("l") - is_dir = perms.startswith("d") or is_link + ty, perms, links, uid, gid, size, mtime, name = match.groups() + is_link = ty == "l" + is_dir = ty == "d" or is_link if is_link: name, _, _link_name = name.partition("->") name = name.strip() _link_name = _link_name.strip() - permissions = Permissions.parse(perms[1:]) + permissions = Permissions.parse(perms) mtime_epoch = _decode_linux_time(mtime)	FTPFS does not list certain folders It seems that the pub folder is not listed and igrnoed. ftplib does show the folder. This is with python 3.8. ```python import fs fs.__version__ # 2.4.12 from fs.ftpfs import FTPFS a = FTP(host='ftp.ensemblgenomes.org', user="anonymous", passwd="") a.dir() """" drwxr-sr-x 72 ftp ftp 8192 Feb 16 02:00 pub -rw-r--r-- 1 ftp ftp 25 Jan 06 22:41 robots.txt """" from ftplib import FTP a = FTPFS(host='ftp.ensemblgenomes.org') a.listdir('.') # ['robots.txt'] ```	Hmm, could be an issue with the #439, which changed the FTP list parsing. I'll have a look when I can. Thanks @althonos!	2021-03-26T18:30:40	[]	[]
PyFilesystem/pyfilesystem2	PyFilesystem__pyfilesystem2-472	99d5314cd15a36c408e3ce90be25a4aa09bb8391	diff --git a/CHANGELOG.md b/CHANGELOG.md index 0b80ff81..c99ffcc3 100644 --- a/CHANGELOG.md +++ b/CHANGELOG.md @@ -20,11 +20,13 @@ and this project adheres to [Semantic Versioning](http://semver.org/). Closes [#445](https://github.com/PyFilesystem/pyfilesystem2/pull/445). - Migrate continuous integration from Travis-CI to GitHub Actions and introduce several linters again in the build steps ([#448](https://github.com/PyFilesystem/pyfilesystem2/pull/448)). - Closes [#446](https://github.com/PyFilesystem/pyfilesystem2/pull/446). + Closes [#446](https://github.com/PyFilesystem/pyfilesystem2/issues/446). - Stop requiring `pytest` to run tests, allowing any test runner supporting `unittest`-style test suites. - `FSTestCases` now builds the large data required for `upload` and `download` tests only once in order to reduce the total testing time. +- `MemoryFS.move` and `MemoryFS.movedir` will now avoid copying data. + Closes [#452](https://github.com/PyFilesystem/pyfilesystem2/issues/452). ### Fixed diff --git a/fs/memoryfs.py b/fs/memoryfs.py index 30e9c21f..7965a135 100644 --- a/fs/memoryfs.py +++ b/fs/memoryfs.py @@ -38,6 +38,7 @@ SupportsInt, Union, Text, + Tuple, ) from .base import _OpendirFactory from .info import RawInfo @@ -274,6 +275,10 @@ def remove_entry(self, name): # type: (Text) -> None del self._dir[name] + def clear(self): + # type: () -> None + self._dir.clear() + def __contains__(self, name): # type: (object) -> bool return name in self._dir @@ -294,6 +299,21 @@ def remove_open_file(self, memory_file): # type: (_MemoryFile) -> None self._open_files.remove(memory_file) + def to_info(self, namespaces=None): + # type: (Optional[Collection[Text]]) -> Info + namespaces = namespaces or () + info = {"basic": {"name": self.name, "is_dir": self.is_dir}} + if "details" in namespaces: + info["details"] = { + "_write": ["accessed", "modified"], + "type": int(self.resource_type), + "size": self.size, + "accessed": self.accessed_time, + "modified": self.modified_time, + "created": self.created_time, + } + return Info(info) + @six.python_2_unicode_compatible class MemoryFS(FS): @@ -368,33 +388,24 @@ def close(self): def getinfo(self, path, namespaces=None): # type: (Text, Optional[Collection[Text]]) -> Info - namespaces = namespaces or () _path = self.validatepath(path) dir_entry = self._get_dir_entry(_path) if dir_entry is None: raise errors.ResourceNotFound(path) - info = {"basic": {"name": dir_entry.name, "is_dir": dir_entry.is_dir}} - if "details" in namespaces: - info["details"] = { - "_write": ["accessed", "modified"], - "type": int(dir_entry.resource_type), - "size": dir_entry.size, - "accessed": dir_entry.accessed_time, - "modified": dir_entry.modified_time, - "created": dir_entry.created_time, - } - return Info(info) + return dir_entry.to_info(namespaces=namespaces) def listdir(self, path): # type: (Text) -> List[Text] self.check() _path = self.validatepath(path) with self._lock: + # locate and validate the entry corresponding to the given path dir_entry = self._get_dir_entry(_path) if dir_entry is None: raise errors.ResourceNotFound(path) if not dir_entry.is_dir: raise errors.DirectoryExpected(path) + # return the filenames in the order they were created return dir_entry.list() if typing.TYPE_CHECKING: @@ -433,6 +444,46 @@ def makedir( parent_dir.set_entry(dir_name, new_dir) return self.opendir(path) + def move(self, src_path, dst_path, overwrite=False): + src_dir, src_name = split(self.validatepath(src_path)) + dst_dir, dst_name = split(self.validatepath(dst_path)) + + with self._lock: + src_dir_entry = self._get_dir_entry(src_dir) + if src_dir_entry is None or src_name not in src_dir_entry: + raise errors.ResourceNotFound(src_path) + src_entry = src_dir_entry.get_entry(src_name) + if src_entry.is_dir: + raise errors.FileExpected(src_path) + + dst_dir_entry = self._get_dir_entry(dst_dir) + if dst_dir_entry is None: + raise errors.ResourceNotFound(dst_path) + elif not overwrite and dst_name in dst_dir_entry: + raise errors.DestinationExists(dst_path) + + dst_dir_entry.set_entry(dst_name, src_entry) + src_dir_entry.remove_entry(src_name) + + def movedir(self, src_path, dst_path, create=False): + src_dir, src_name = split(self.validatepath(src_path)) + dst_dir, dst_name = split(self.validatepath(dst_path)) + + with self._lock: + src_dir_entry = self._get_dir_entry(src_dir) + if src_dir_entry is None or src_name not in src_dir_entry: + raise errors.ResourceNotFound(src_path) + src_entry = src_dir_entry.get_entry(src_name) + if not src_entry.is_dir: + raise errors.DirectoryExpected(src_path) + + dst_dir_entry = self._get_dir_entry(dst_dir) + if dst_dir_entry is None or (not create and dst_name not in dst_dir_entry): + raise errors.ResourceNotFound(dst_path) + + dst_dir_entry.set_entry(dst_name, src_entry) + src_dir_entry.remove_entry(src_name) + def openbin(self, path, mode="r", buffering=-1, options): # type: (Text, Text, int, Any) -> BinaryIO _mode = Mode(mode) @@ -499,12 +550,29 @@ def remove(self, path): def removedir(self, path): # type: (Text) -> None + # make sure we are not removing root _path = self.validatepath(path) - if _path == "/": raise errors.RemoveRootError() + # make sure the directory is empty + if not self.isempty(path): + raise errors.DirectoryNotEmpty(path) + # we can now delegate to removetree since we confirmed that + # * path exists (isempty) + # * path is a folder (isempty) + # * path is not root + self.removetree(_path) + + def removetree(self, path): + # type: (Text) -> None + _path = self.validatepath(path) with self._lock: + + if _path == "/": + self.root.clear() + return + dir_path, file_name = split(_path) parent_dir_entry = self._get_dir_entry(dir_path) @@ -515,11 +583,34 @@ def removedir(self, path): if not dir_dir_entry.is_dir: raise errors.DirectoryExpected(path) - if len(dir_dir_entry): - raise errors.DirectoryNotEmpty(path) - parent_dir_entry.remove_entry(file_name) + def scandir( + self, + path, # type: Text + namespaces=None, # type: Optional[Collection[Text]] + page=None, # type: Optional[Tuple[int, int]] + ): + # type: (...) -> Iterator[Info] + self.check() + _path = self.validatepath(path) + with self._lock: + # locate and validate the entry corresponding to the given path + dir_entry = self._get_dir_entry(_path) + if dir_entry is None: + raise errors.ResourceNotFound(path) + if not dir_entry.is_dir: + raise errors.DirectoryExpected(path) + # if paging was requested, slice the filenames + filenames = dir_entry.list() + if page is not None: + start, end = page + filenames = filenames[start:end] + # yield info with the right namespaces + for name in filenames: + entry = typing.cast(_DirEntry, dir_entry.get_entry(name)) + yield entry.to_info(namespaces=namespaces) + def setinfo(self, path, info): # type: (Text, RawInfo) -> None _path = self.validatepath(path)	How to rename files in MemFS? Hi, there doesn't seem to be a way to rename files in MemFS without copying them around. Looks like MemFS tries to apply `os.rename()` if the file can be seen by the OS (which is not the case for MemFS). The only remaining way is to use something like `fs.move` that duplicates the file before removing the old one. On systems with limited memory this obviously doesn't work. Example: a MemFS with a total size of 10 GB and a single file of 8 GB lying on it. How to rename this file?	Hmmm, interesting point... https://docs.pyfilesystem.org/en/latest/guide.html#moving-and-copying says "These move and copy methods are optimized where possible, and depending on the implementation, they may be more performant than reading and writing files.". However it looks like MemFS doesn't have optimised `move` or `movedir` methods and instead just drops through to the base `fs.move`, which as you note does a copy-then-delete. I'm not familiar enough with the MemFS to know if there's a good reason _why_ it doesn't provide optimised `move` / `movedir` methods though (and I don't have time to look into it ATM). But feel free to experiment and put a PR together... :wink: MemFs certainly could do a rename operation rather than a copy. It just hasn’t been implemented.	2021-03-26T14:06:18	[]	[]
PyFilesystem/pyfilesystem2	PyFilesystem__pyfilesystem2-457	64d7a52e1f3037f33937f91e9db812cafdbbafc0	diff --git a/CHANGELOG.md b/CHANGELOG.md index 31cb320c..4dfe9058 100644 --- a/CHANGELOG.md +++ b/CHANGELOG.md @@ -31,6 +31,7 @@ and this project adheres to [Semantic Versioning](http://semver.org/). - Make `FTPFile`, `MemoryFile` and `RawWrapper` accept [`array.array`](https://docs.python.org/3/library/array.html) arguments for the `write` and `writelines` methods, as expected by their base class [`io.RawIOBase`](https://docs.python.org/3/library/io.html#io.RawIOBase). - Various documentation issues, including `MemoryFS` docstring not rendering properly. +- Avoid creating a new connection on every call of `FTPFS.upload`. Closes [#455](https://github.com/PyFilesystem/pyfilesystem2/issues/455). ## [2.4.12] - 2021-01-14 diff --git a/fs/ftpfs.py b/fs/ftpfs.py index 3d8b0944..c5929ee1 100644 --- a/fs/ftpfs.py +++ b/fs/ftpfs.py @@ -814,11 +814,10 @@ def upload(self, path, file, chunk_size=None, options): # type: (Text, BinaryIO, Optional[int], Any) -> None _path = self.validatepath(path) with self._lock: - with self._manage_ftp() as ftp: - with ftp_errors(self, path): - ftp.storbinary( - str("STOR ") + _encode(_path, self.ftp.encoding), file - ) + with ftp_errors(self, path): + self.ftp.storbinary( + str("STOR ") + _encode(_path, self.ftp.encoding), file + ) def writebytes(self, path, contents): # type: (Text, ByteString) -> None	fs.ftpfs.FTPFS.upload opens new connections Hi, I'm looking at this function: https://github.com/PyFilesystem/pyfilesystem2/blob/64d7a52e1f3037f33937f91e9db812cafdbbafc0/fs/ftpfs.py#L813 `_manage_ftp` always opens a new FTP connection, together with user login, "FEAT" command and so on. This is done everytime one wants to copy a file to an FTPFS. Is there a good reason why that code doesn't just use `self._ftp`, the connection that should be open for the lifetime of the FTPFS object? Otherwise, I'd like to create a PR to fix that.	Hmmm, true, I'm actually wondering too. I know it makes sense to open a new connection when opening a new file, to guarantee the file and the `FTPFS` can be used in parallel, but i don't think it makes sense for the shortcut methods (like `upload` or `readbytes`) to create a new connection. @willmcgugan , any opinion? Maybe there was a reason for doing so when you wrote the code that I am unaware of. > I know it makes sense to open a new connection when opening a new file, to guarantee the file and the FTPFS can be used in parallel, but i don't think it makes sense for the shortcut methods (like upload or readbytes) to create a new connection. Yeah, FTP offers no way of multiplexing file downloads that I know of, so multiple instances are the only way to achieve that. Re shortcut methods, I guess we could re-use the main connection used for listing directories, but there would still need to be one per thread to make it thread-safe. I wouldn't be surprised if there were wins that could be made for copying files sequentially. Especially for many small files. But in general I would expect downloading in parallel to eclipse those wins... > but there would still need to be one per thread to make it thread-safe. Wouldn't locking the `FTPFS` when accessing the main connection be sufficient though? Sure. That would limit performance though. Do we do that already? I don’t recall. In the snippet linked by @atollk the lock is acquired before `_manage_ftp` is called, yes. In that case re-using the connection may be beneficial. @atollk you’re welcome to tackle that, assuming @althonos approves. Fine by me! There are probably other places in `fs.ftpfs` where this could be beneficial, but we can figure that out if you want to open a PR.	2021-03-19T08:15:22	[]	[]
NREL/routee-compass	NREL__routee-compass-260	7ca47d112b954cd668d05c7539d908462d5bf986	diff --git a/rust/routee-compass-core/src/algorithm/component/scc.rs b/rust/routee-compass-core/src/algorithm/component/scc.rs index e6de4381..8cc5e10d 100644 --- a/rust/routee-compass-core/src/algorithm/component/scc.rs +++ b/rust/routee-compass-core/src/algorithm/component/scc.rs @@ -1,5 +1,5 @@ -use crate::model::road_network::graph::Graph; -use crate::model::road_network::{graph_error::GraphError, vertex_id::VertexId}; +use crate::model::network::graph::Graph; +use crate::model::network::{network_error::NetworkError, vertex_id::VertexId}; use std::collections::HashSet; /// Conducts a depth-first search (DFS) on a directed graph. @@ -21,23 +21,23 @@ use std::collections::HashSet; /// pub fn depth_first_search( graph: &Graph, - vertex: VertexId, + vertex: &VertexId, visited: &mut HashSet<VertexId>, stack: &mut Vec<VertexId>, -) -> Result<(), GraphError> { - if visited.contains(&vertex) { +) -> Result<(), NetworkError> { + if visited.contains(vertex) { return Ok(()); } - visited.insert(vertex); + visited.insert(vertex); - let edges = graph.out_edges(vertex)?; + let edges = graph.out_edges(vertex); for edge in edges { - let dst = graph.dst_vertex_id(edge)?; - depth_first_search(graph, dst, visited, stack)?; + let dst = graph.dst_vertex_id(&edge)?; + depth_first_search(graph, &dst, visited, stack)?; } - stack.push(vertex); + stack.push(vertex); Ok(()) } @@ -61,23 +61,23 @@ pub fn depth_first_search( /// pub fn reverse_depth_first_search( graph: &Graph, - vertex: VertexId, + vertex: &VertexId, visited: &mut HashSet<VertexId>, stack: &mut Vec<VertexId>, -) -> Result<(), GraphError> { - if visited.contains(&vertex) { +) -> Result<(), NetworkError> { + if visited.contains(vertex) { return Ok(()); } - visited.insert(vertex); + visited.insert(vertex); - let edges = graph.in_edges(vertex)?; + let edges = graph.in_edges(vertex); for edge in edges { - let src = graph.src_vertex_id(edge)?; - reverse_depth_first_search(graph, src, visited, stack)?; + let src = graph.src_vertex_id(&edge)?; + reverse_depth_first_search(graph, &src, visited, stack)?; } - stack.push(vertex); + stack.push(vertex); Ok(()) } @@ -94,14 +94,15 @@ pub fn reverse_depth_first_search( /// /// Returns an error if the `graph` has an issue like a non-existing vertex. /// -pub fn all_strongly_connected_componenets(graph: &Graph) -> Result<Vec<Vec<VertexId>>, GraphError> { +pub fn all_strongly_connected_componenets( + graph: &Graph, +) -> Result<Vec<Vec<VertexId>>, NetworkError> { let mut visited: HashSet<VertexId> = HashSet::new(); let mut container: Vec<VertexId> = Vec::new(); - let mut result: Vec<Vec<VertexId>> = Vec::new(); for vertex_id in graph.vertex_ids() { - depth_first_search(graph, vertex_id, &mut visited, &mut container)?; + depth_first_search(graph, &vertex_id, &mut visited, &mut container)?; } visited.clear(); @@ -112,7 +113,7 @@ pub fn all_strongly_connected_componenets(graph: &Graph) -> Result<Vec<Vec<Verte } let mut component: Vec<VertexId> = Vec::new(); - reverse_depth_first_search(graph, vertex_id, &mut visited, &mut component)?; + reverse_depth_first_search(graph, &vertex_id, &mut visited, &mut component)?; result.push(component); } @@ -131,7 +132,7 @@ pub fn all_strongly_connected_componenets(graph: &Graph) -> Result<Vec<Vec<Verte /// /// Returns an error if the `graph` has an issue like a non-existing vertex. /// -pub fn largest_strongly_connected_component(graph: &Graph) -> Result<Vec<VertexId>, GraphError> { +pub fn largest_strongly_connected_component(graph: &Graph) -> Result<Vec<VertexId>, NetworkError> { let components = all_strongly_connected_componenets(graph)?; let mut largest_component: Vec<VertexId> = Vec::new(); @@ -149,7 +150,7 @@ pub fn largest_strongly_connected_component(graph: &Graph) -> Result<Vec<VertexI mod tests { use super::; use crate::{ - model::property::{edge::Edge, vertex::Vertex}, + model::network::{Edge, Vertex}, util::compact_ordered_hash_map::CompactOrderedHashMap, }; diff --git a/rust/routee-compass-core/src/algorithm/search/a_star/a_star_algorithm.rs b/rust/routee-compass-core/src/algorithm/search/a_star/a_star_algorithm.rs index 0bcceecf..641d130a 100644 --- a/rust/routee-compass-core/src/algorithm/search/a_star/a_star_algorithm.rs +++ b/rust/routee-compass-core/src/algorithm/search/a_star/a_star_algorithm.rs @@ -4,8 +4,8 @@ use crate::algorithm::search::search_error::SearchError; use crate::algorithm::search::search_instance::SearchInstance; use crate::algorithm::search::search_result::SearchResult; use crate::algorithm::search::search_tree_branch::SearchTreeBranch; -use crate::model::road_network::edge_id::EdgeId; -use crate::model::road_network::vertex_id::VertexId; +use crate::model::network::edge_id::EdgeId; +use crate::model::network::vertex_id::VertexId; use crate::model::unit::as_f64::AsF64; use crate::model::unit::cost::ReverseCost; use crate::model::unit::Cost; @@ -61,7 +61,7 @@ pub fn run_a_star( let last_edge_id = get_last_traversed_edge_id(&current_vertex_id, &source, &solution)?; let last_edge = match last_edge_id { - Some(id) => Some(si.directed_graph.get_edge(id)?), + Some(id) => Some(si.directed_graph.get_edge(&id)?), None => None, }; @@ -72,7 +72,7 @@ pub fn run_a_star( solution .get(&current_vertex_id) .ok_or_else(\|\| { - SearchError::InternalSearchError(format!( + SearchError::InternalError(format!( "expected vertex id {} missing from solution", current_vertex_id )) @@ -83,9 +83,9 @@ pub fn run_a_star( }; // visit all neighbors of this source vertex - let incident_edge_iterator = direction.get_incident_edges(&current_vertex_id, si)?; + let incident_edge_iterator = direction.get_incident_edges(&current_vertex_id, si); for edge_id in incident_edge_iterator { - let e = si.directed_graph.get_edge(edge_id)?; + let e = si.directed_graph.get_edge(edge_id)?; let terminal_vertex_id = direction.terminal_vertex_id(e); let key_vertex_id = direction.tree_key_vertex_id(e); @@ -188,8 +188,8 @@ pub fn run_a_star_edge_oriented( si: &SearchInstance, ) -> Result<SearchResult, SearchError> { // 1. guard against edge conditions (src==dst, src.dst_v == dst.src_v) - let e1_src = si.directed_graph.src_vertex_id(source)?; - let e1_dst = si.directed_graph.dst_vertex_id(source)?; + let e1_src = si.directed_graph.src_vertex_id(&source)?; + let e1_dst = si.directed_graph.dst_vertex_id(&source)?; let src_et = EdgeTraversal { edge_id: source, access_cost: Cost::ZERO, @@ -217,8 +217,8 @@ pub fn run_a_star_edge_oriented( Ok(updated) } Some(target_edge) => { - let e2_src = si.directed_graph.src_vertex_id(target_edge)?; - let e2_dst = si.directed_graph.dst_vertex_id(target_edge)?; + let e2_src = si.directed_graph.src_vertex_id(&target_edge)?; + let e2_dst = si.directed_graph.dst_vertex_id(&target_edge)?; if source == target_edge { Ok(SearchResult::default()) @@ -254,12 +254,17 @@ pub fn run_a_star_edge_oriented( } = run_a_star(e1_dst, Some(e2_src), direction, weight_factor, si)?; if tree.is_empty() { - return Err(SearchError::NoPathExists(e1_dst, e2_src)); + return Err(SearchError::NoPathExistsBetweenVertices(e1_dst, e2_src)); } let final_state = &tree .get(&e2_src) - .ok_or_else(\|\| SearchError::VertexMissingFromSearchTree(e2_src))? + .ok_or_else(\|\| { + SearchError::InternalError(format!( + "resulting tree missing vertex {} expected via backtrack", + e2_src + )) + })? .edge_traversal .result_state; let dst_et = EdgeTraversal { @@ -313,7 +318,10 @@ fn advance_search( target: Option<VertexId>, ) -> Result<Option<VertexId>, SearchError> { match (cost.pop(), target) { - (None, Some(target_vertex_id)) => Err(SearchError::NoPathExists(source, target_vertex_id)), + (None, Some(target_vertex_id)) => Err(SearchError::NoPathExistsBetweenVertices( + source, + target_vertex_id, + )), (None, None) => Ok(None), (Some((current_v, _)), Some(target_v)) if current_v == target_v => Ok(None), (Some((current_vertex_id, _)), _) => Ok(Some(current_vertex_id)), @@ -347,7 +355,7 @@ fn get_last_traversed_edge_id( let edge_id = tree .get(this_vertex_id) .ok_or_else(\|\| { - SearchError::InternalSearchError(format!( + SearchError::InternalError(format!( "expected vertex id {} missing from solution", this_vertex_id )) @@ -368,10 +376,10 @@ mod tests { use crate::model::cost::cost_model::CostModel; use crate::model::cost::vehicle::vehicle_cost_rate::VehicleCostRate; use crate::model::frontier::default::no_restriction::NoRestriction; - use crate::model::property::edge::Edge; - use crate::model::property::vertex::Vertex; - use crate::model::road_network::edge_id::EdgeId; - use crate::model::road_network::graph::Graph; + use crate::model::network::edge_id::EdgeId; + use crate::model::network::graph::Graph; + use crate::model::network::Edge; + use crate::model::network::Vertex; use crate::model::state::state_feature::StateFeature; use crate::model::state::state_model::StateModel; use crate::model::termination::termination_model::TerminationModel; diff --git a/rust/routee-compass-core/src/algorithm/search/a_star/bidirectional_a_star_algorithm.rs b/rust/routee-compass-core/src/algorithm/search/a_star/bidirectional_ops.rs similarity index 89% rename from rust/routee-compass-core/src/algorithm/search/a_star/bidirectional_a_star_algorithm.rs rename to rust/routee-compass-core/src/algorithm/search/a_star/bidirectional_ops.rs index 882adf9e..398f7731 100644 --- a/rust/routee-compass-core/src/algorithm/search/a_star/bidirectional_a_star_algorithm.rs +++ b/rust/routee-compass-core/src/algorithm/search/a_star/bidirectional_ops.rs @@ -23,12 +23,7 @@ pub fn reorient_reverse_route( ) -> Result<Vec<EdgeTraversal>, SearchError> { // get the final edge id and state for the forward traversal let (final_fwd_edge_id, mut acc_state) = match fwd_route.last() { - None => ( - None, - si.state_model - .initial_state() - .map_err(SearchError::StateError)?, - ), + None => (None, si.state_model.initial_state()?), Some(last_edge) => (Some(last_edge.edge_id), last_edge.result_state.clone()), }; @@ -46,7 +41,7 @@ pub fn reorient_reverse_route( let mut result: Vec<EdgeTraversal> = Vec::with_capacity(rev_route.len()); for (prev_opt, next_opt) in edge_ids.iter().tuple_windows() { let next = next_opt.ok_or_else(\|\| { - SearchError::InternalSearchError(String::from("next_opt should never be None")) + SearchError::InternalError(String::from("next_opt should never be None")) })?; let et = EdgeTraversal::forward_traversal(next, prev_opt, &acc_state, si)?; acc_state = et.result_state.clone(); @@ -72,7 +67,7 @@ pub fn route_contains_loop( ) -> Result<bool, SearchError> { let src_vertices = route .iter() - .map(\|e\| si.directed_graph.src_vertex_id(e.edge_id)) + .map(\|e\| si.directed_graph.src_vertex_id(&e.edge_id)) .collect::<Result<Vec<_>, _>>()?; Ok(src_vertices.iter().unique().collect_vec().len() < src_vertices.len()) } diff --git a/rust/routee-compass-core/src/algorithm/search/a_star/mod.rs b/rust/routee-compass-core/src/algorithm/search/a_star/mod.rs index 7177a9bb..c60a7533 100644 --- a/rust/routee-compass-core/src/algorithm/search/a_star/mod.rs +++ b/rust/routee-compass-core/src/algorithm/search/a_star/mod.rs @@ -1,2 +1,2 @@ pub mod a_star_algorithm; -pub mod bidirectional_a_star_algorithm; +pub mod bidirectional_ops; diff --git a/rust/routee-compass-core/src/algorithm/search/backtrack.rs b/rust/routee-compass-core/src/algorithm/search/backtrack.rs index 307f0e20..8670a278 100644 --- a/rust/routee-compass-core/src/algorithm/search/backtrack.rs +++ b/rust/routee-compass-core/src/algorithm/search/backtrack.rs @@ -1,7 +1,7 @@ use super::{ edge_traversal::EdgeTraversal, search_error::SearchError, search_tree_branch::SearchTreeBranch, }; -use crate::model::road_network::{edge_id::EdgeId, graph::Graph, vertex_id::VertexId}; +use crate::model::network::{edge_id::EdgeId, graph::Graph, vertex_id::VertexId}; use std::{ collections::{HashMap, HashSet}, sync::Arc, @@ -24,12 +24,16 @@ pub fn vertex_oriented_route( } let traversal = solution .get(&this_vertex) - .ok_or_else(\|\| SearchError::VertexMissingFromSearchTree(this_vertex))?; + .ok_or(SearchError::InternalError(format!( + "resulting tree missing vertex {} expected via backtrack", + this_vertex + )))?; let first_visit = visited.insert(traversal.edge_traversal.edge_id); if !first_visit { - return Err(SearchError::LoopInSearchResult( - traversal.edge_traversal.edge_id, - )); + return Err(SearchError::InternalError(format!( + "loop in search result, edge {} visited more than once", + traversal.edge_traversal.edge_id + ))); } result.push(traversal.edge_traversal.clone()); this_vertex = traversal.terminal_vertex; @@ -45,11 +49,7 @@ pub fn edge_oriented_route( solution: &HashMap<VertexId, SearchTreeBranch>, graph: Arc<Graph>, ) -> Result<Vec<EdgeTraversal>, SearchError> { - let o_v = graph - .src_vertex_id(source_id) - .map_err(SearchError::GraphError)?; - let d_v = graph - .dst_vertex_id(target_id) - .map_err(SearchError::GraphError)?; + let o_v = graph.src_vertex_id(&source_id)?; + let d_v = graph.dst_vertex_id(&target_id)?; vertex_oriented_route(o_v, d_v, solution) } diff --git a/rust/routee-compass-core/src/algorithm/search/direction.rs b/rust/routee-compass-core/src/algorithm/search/direction.rs index 6c49c337..d51bd5ae 100644 --- a/rust/routee-compass-core/src/algorithm/search/direction.rs +++ b/rust/routee-compass-core/src/algorithm/search/direction.rs @@ -1,11 +1,8 @@ use super::{ edge_traversal::EdgeTraversal, search_error::SearchError, search_instance::SearchInstance, }; -use crate::model::{ - property::edge::Edge, - road_network::{edge_id::EdgeId, graph_error::GraphError, vertex_id::VertexId}, - traversal::state::state_variable::StateVar, -}; +use crate::model::network::{Edge, EdgeId, VertexId}; +use crate::model::traversal::state::state_variable::StateVar; use serde::{Deserialize, Serialize}; #[derive(Copy, Clone, Serialize, Deserialize, Default)] @@ -21,10 +18,10 @@ impl Direction { &'a self, vertex_id: &VertexId, si: &'a SearchInstance, - ) -> Result<Box<dyn Iterator<Item = &EdgeId> + 'a>, GraphError> { + ) -> Box<dyn Iterator<Item = &'a EdgeId> + 'a> { match self { - Direction::Forward => si.directed_graph.out_edges_iter(vertex_id), - Direction::Reverse => si.directed_graph.in_edges_iter(vertex_id), + Direction::Forward => si.directed_graph.out_edges_iter(vertex_id), + Direction::Reverse => si.directed_graph.in_edges_iter(vertex_id), } } diff --git a/rust/routee-compass-core/src/algorithm/search/edge_traversal.rs b/rust/routee-compass-core/src/algorithm/search/edge_traversal.rs index 786ce83a..675975ed 100644 --- a/rust/routee-compass-core/src/algorithm/search/edge_traversal.rs +++ b/rust/routee-compass-core/src/algorithm/search/edge_traversal.rs @@ -1,6 +1,6 @@ use super::search_error::SearchError; use super::search_instance::SearchInstance; -use crate::model::road_network::edge_id::EdgeId; +use crate::model::network::edge_id::EdgeId; use crate::model::traversal::state::state_variable::StateVar; use crate::model::traversal::state::traversal_state::TraversalState; use crate::model::unit::Cost; @@ -39,7 +39,6 @@ impl EdgeTraversal { /// /// # Arguments /// - /// `next_edge_id` - the edge to traverse /// * `prev_edge_id_opt` - the previously traversed edge, if exists, for access costs /// * `prev_state` - the state before traversal, positioned closer to the destination @@ -58,22 +57,13 @@ impl EdgeTraversal { let mut access_cost = Cost::ZERO; // find this traversal in the graph - let traversal_trajectory = si - .directed_graph - .edge_triplet_attrs(next_edge_id) - .map_err(SearchError::GraphError)?; + let traversal_trajectory = si.directed_graph.edge_triplet(&next_edge_id)?; // perform access traversal for (v2)-[next]->(v3) // access cost for (v1)-[prev]->(v2)-[next]->(v3) if let Some(prev_edge_id) = prev_edge_id_opt { - let e1 = si - .directed_graph - .get_edge(prev_edge_id) - .map_err(SearchError::GraphError)?; - let v1 = si - .directed_graph - .get_vertex(e1.src_vertex_id) - .map_err(SearchError::GraphError)?; + let e1 = si.directed_graph.get_edge(&prev_edge_id)?; + let v1 = si.directed_graph.get_vertex(&e1.src_vertex_id)?; let (v2, e2, v3) = traversal_trajectory; let access_trajectory = (v1, e1, v2, e2, v3); @@ -83,20 +73,20 @@ impl EdgeTraversal { let ac = si .cost_model - .access_cost(e1, e2, prev_state, &result_state) - .map_err(SearchError::CostError)?; + .access_cost(e1, e2, prev_state, &result_state)?; access_cost = access_cost + ac; } - si.traversal_model - .traverse_edge(traversal_trajectory, &mut result_state, &si.state_model) - .map_err(SearchError::TraversalModelFailure)?; + si.traversal_model.traverse_edge( + traversal_trajectory, + &mut result_state, + &si.state_model, + )?; let (_, edge, _) = traversal_trajectory; let total_cost = si .cost_model - .traversal_cost(edge, prev_state, &result_state) - .map_err(SearchError::CostError)?; + .traversal_cost(edge, prev_state, &result_state)?; let traversal_cost = total_cost - access_cost; let result = EdgeTraversal { @@ -137,22 +127,13 @@ impl EdgeTraversal { let mut access_cost = Cost::ZERO; // find this traversal in the graph - let traversal_trajectory = si - .directed_graph - .edge_triplet_attrs(prev_edge_id) - .map_err(SearchError::GraphError)?; + let traversal_trajectory = si.directed_graph.edge_triplet(&prev_edge_id)?; // perform access traversal for (v1)-[prev]->(v2) // access cost for (v1)-[prev]->(v2)-[next]->(v3) if let Some(next_edge_id) = next_edge_id_opt { - let e2 = si - .directed_graph - .get_edge(next_edge_id) - .map_err(SearchError::GraphError)?; - let v3 = si - .directed_graph - .get_vertex(e2.dst_vertex_id) - .map_err(SearchError::GraphError)?; + let e2 = si.directed_graph.get_edge(&next_edge_id)?; + let v3 = si.directed_graph.get_vertex(&e2.dst_vertex_id)?; let (v1, e1, v2) = traversal_trajectory; let access_trajectory = (v1, e1, v2, e2, v3); @@ -162,20 +143,20 @@ impl EdgeTraversal { let ac = si .cost_model - .access_cost(e1, e2, prev_state, &result_state) - .map_err(SearchError::CostError)?; + .access_cost(e1, e2, prev_state, &result_state)?; access_cost = access_cost + ac; } - si.traversal_model - .traverse_edge(traversal_trajectory, &mut result_state, &si.state_model) - .map_err(SearchError::TraversalModelFailure)?; + si.traversal_model.traverse_edge( + traversal_trajectory, + &mut result_state, + &si.state_model, + )?; let (_, edge, _) = traversal_trajectory; let total_cost = si .cost_model - .traversal_cost(edge, prev_state, &result_state) - .map_err(SearchError::CostError)?; + .traversal_cost(edge, prev_state, &result_state)?; let traversal_cost = total_cost - access_cost; let result = EdgeTraversal { diff --git a/rust/routee-compass-core/src/algorithm/search/ksp/ksp_query.rs b/rust/routee-compass-core/src/algorithm/search/ksp/ksp_query.rs index d9714d95..4b4093ef 100644 --- a/rust/routee-compass-core/src/algorithm/search/ksp/ksp_query.rs +++ b/rust/routee-compass-core/src/algorithm/search/ksp/ksp_query.rs @@ -1,6 +1,4 @@ -use crate::{ - algorithm::search::search_error::SearchError, model::road_network::vertex_id::VertexId, -}; +use crate::{algorithm::search::search_error::SearchError, model::network::vertex_id::VertexId}; #[derive(Debug, Clone)] pub struct KspQuery<'a> { diff --git a/rust/routee-compass-core/src/algorithm/search/ksp/single_via_paths_algorithm.rs b/rust/routee-compass-core/src/algorithm/search/ksp/single_via_paths_algorithm.rs index 5f6bc9b7..5ce9ef7c 100644 --- a/rust/routee-compass-core/src/algorithm/search/ksp/single_via_paths_algorithm.rs +++ b/rust/routee-compass-core/src/algorithm/search/ksp/single_via_paths_algorithm.rs @@ -3,12 +3,12 @@ use itertools::Itertools; use super::{ksp_query::KspQuery, ksp_termination_criteria::KspTerminationCriteria}; use crate::{ algorithm::search::{ - a_star::bidirectional_a_star_algorithm, backtrack, direction::Direction, - edge_traversal::EdgeTraversal, search_algorithm::SearchAlgorithm, - search_algorithm_result::SearchAlgorithmResult, search_error::SearchError, - search_instance::SearchInstance, util::route_similarity_function::RouteSimilarityFunction, + a_star::bidirectional_ops, backtrack, direction::Direction, edge_traversal::EdgeTraversal, + search_algorithm::SearchAlgorithm, search_algorithm_result::SearchAlgorithmResult, + search_error::SearchError, search_instance::SearchInstance, + util::route_similarity_function::RouteSimilarityFunction, }, - model::{road_network::vertex_id::VertexId, unit::cost::ReverseCost}, + model::{network::vertex_id::VertexId, unit::cost::ReverseCost}, util::priority_queue::InternalPriorityQueue, }; use std::collections::HashMap; @@ -45,23 +45,23 @@ pub fn run( si, )?; if fwd_trees.len() != 1 { - Err(SearchError::InternalSearchError(format!( + Err(SearchError::InternalError(format!( "ksp solver fwd trees count should be exactly 1, found {}", fwd_trees.len() )))?; } if rev_trees.len() != 1 { - Err(SearchError::InternalSearchError(format!( + Err(SearchError::InternalError(format!( "ksp solver rev trees count should be exactly 1, found {}", rev_trees.len() )))?; } - let fwd_tree = fwd_trees.first().ok_or_else(\|\| { - SearchError::InternalSearchError(String::from("cannot retrieve fwd tree 0")) - })?; - let rev_tree = rev_trees.first().ok_or_else(\|\| { - SearchError::InternalSearchError(String::from("cannot retrieve rev tree 0")) - })?; + let fwd_tree = fwd_trees + .first() + .ok_or_else(\|\| SearchError::InternalError(String::from("cannot retrieve fwd tree 0")))?; + let rev_tree = rev_trees + .first() + .ok_or_else(\|\| SearchError::InternalError(String::from("cannot retrieve rev tree 0")))?; // find intersection vertices let rev_vertices = rev_trees.iter().flatten().collect::<HashMap<_, _>>(); @@ -114,15 +114,12 @@ pub fn run( intersection_vertex_id, rev_tree, )?; - let rev_route = bidirectional_a_star_algorithm::reorient_reverse_route( - &fwd_route, - &rev_route_backward, - si, - )?; + let rev_route = + bidirectional_ops::reorient_reverse_route(&fwd_route, &rev_route_backward, si)?; let this_route = fwd_route.into_iter().chain(rev_route).collect::<Vec<_>>(); // test loop - if bidirectional_a_star_algorithm::route_contains_loop(&this_route, si)? { + if bidirectional_ops::route_contains_loop(&this_route, si)? { log::debug!("ksp:{} contains loop", ksp_it); accept_route = false; } diff --git a/rust/routee-compass-core/src/algorithm/search/ksp/yens_algorithm.rs b/rust/routee-compass-core/src/algorithm/search/ksp/yens_algorithm.rs index d0525ab0..b480cfc0 100644 --- a/rust/routee-compass-core/src/algorithm/search/ksp/yens_algorithm.rs +++ b/rust/routee-compass-core/src/algorithm/search/ksp/yens_algorithm.rs @@ -6,7 +6,7 @@ use crate::{ search_instance::SearchInstance, util::edge_cut_frontier_model::EdgeCutFrontierModel, util::route_similarity_function::RouteSimilarityFunction, }, - model::{road_network::edge_id::EdgeId, unit::Cost}, + model::{network::edge_id::EdgeId, unit::Cost}, }; use itertools::Itertools; use std::{collections::HashSet, sync::Arc}; @@ -53,7 +53,7 @@ pub fn run( accepted .last() .cloned() - .ok_or(SearchError::InternalSearchError(String::from( + .ok_or(SearchError::InternalError(String::from( "at least one route should be in routes", )))?; @@ -65,12 +65,12 @@ pub fn run( let spur_edge_traversal = root_path .last() - .ok_or(SearchError::InternalSearchError(String::from( + .ok_or(SearchError::InternalError(String::from( "root path is empty", )))?; let spur_vertex_id = si .directed_graph - .get_edge(spur_edge_traversal.edge_id)? + .get_edge(&spur_edge_traversal.edge_id)? .dst_vertex_id; // cut frontier edges based on previous paths with matching root path @@ -149,7 +149,7 @@ pub fn run( fn get_first_route(res: &SearchAlgorithmResult) -> Result<&Vec<EdgeTraversal>, SearchError> { res.routes .first() - .ok_or(SearchError::InternalSearchError(String::from( + .ok_or(SearchError::InternalError(String::from( "no empty results should be stored in routes", ))) } diff --git a/rust/routee-compass-core/src/algorithm/search/mod.rs b/rust/routee-compass-core/src/algorithm/search/mod.rs index aaef9d27..989db776 100644 --- a/rust/routee-compass-core/src/algorithm/search/mod.rs +++ b/rust/routee-compass-core/src/algorithm/search/mod.rs @@ -1,5 +1,5 @@ use self::search_tree_branch::SearchTreeBranch; -use crate::model::road_network::vertex_id::VertexId; +use crate::model::network::vertex_id::VertexId; use std::collections::HashMap; pub mod a_star; diff --git a/rust/routee-compass-core/src/algorithm/search/search_algorithm.rs b/rust/routee-compass-core/src/algorithm/search/search_algorithm.rs index a09c9253..34fe869f 100644 --- a/rust/routee-compass-core/src/algorithm/search/search_algorithm.rs +++ b/rust/routee-compass-core/src/algorithm/search/search_algorithm.rs @@ -9,7 +9,7 @@ use super::search_instance::SearchInstance; use super::search_tree_branch::SearchTreeBranch; use super::util::route_similarity_function::RouteSimilarityFunction; use super::{a_star::a_star_algorithm, direction::Direction}; -use crate::model::road_network::{edge_id::EdgeId, vertex_id::VertexId}; +use crate::model::network::{edge_id::EdgeId, vertex_id::VertexId}; use crate::model::unit::Cost; use serde::{Deserialize, Serialize}; @@ -182,8 +182,8 @@ pub fn run_edge_oriented( si: &SearchInstance, ) -> Result<SearchAlgorithmResult, SearchError> { // 1. guard against edge conditions (src==dst, src.dst_v == dst.src_v) - let e1_src = si.directed_graph.src_vertex_id(source)?; - let e1_dst = si.directed_graph.dst_vertex_id(source)?; + let e1_src = si.directed_graph.src_vertex_id(&source)?; + let e1_dst = si.directed_graph.dst_vertex_id(&source)?; let src_et = EdgeTraversal { edge_id: source, access_cost: Cost::ZERO, @@ -218,8 +218,8 @@ pub fn run_edge_oriented( Ok(updated) } Some(target_edge) => { - let e2_src = si.directed_graph.src_vertex_id(target_edge)?; - let e2_dst = si.directed_graph.dst_vertex_id(target_edge)?; + let e2_src = si.directed_graph.src_vertex_id(&target_edge)?; + let e2_dst = si.directed_graph.dst_vertex_id(&target_edge)?; if source == target_edge { Ok(SearchAlgorithmResult::default()) @@ -258,14 +258,14 @@ pub fn run_edge_oriented( } = alg.run_vertex_oriented(e1_dst, Some(e2_src), query, direction, si)?; if trees.is_empty() { - return Err(SearchError::NoPathExists(e1_dst, e2_src)); + return Err(SearchError::NoPathExistsBetweenVertices(e1_dst, e2_src)); } // it is possible that the search already found these vertices. one major edge // case is when the trip starts with a u-turn. for route in routes.iter_mut() { let final_state = route.last().ok_or_else(\|\| { - SearchError::InternalSearchError(String::from("found empty result route")) + SearchError::InternalError(String::from("found empty result route")) })?; let dst_et = EdgeTraversal { diff --git a/rust/routee-compass-core/src/algorithm/search/search_algorithm_result.rs b/rust/routee-compass-core/src/algorithm/search/search_algorithm_result.rs index cd14a504..93fb6e4c 100644 --- a/rust/routee-compass-core/src/algorithm/search/search_algorithm_result.rs +++ b/rust/routee-compass-core/src/algorithm/search/search_algorithm_result.rs @@ -1,5 +1,5 @@ use super::{edge_traversal::EdgeTraversal, search_tree_branch::SearchTreeBranch}; -use crate::model::road_network::vertex_id::VertexId; +use crate::model::network::vertex_id::VertexId; use std::collections::HashMap; #[derive(Default)] diff --git a/rust/routee-compass-core/src/algorithm/search/search_error.rs b/rust/routee-compass-core/src/algorithm/search/search_error.rs index 0777d61b..fad6be4d 100644 --- a/rust/routee-compass-core/src/algorithm/search/search_error.rs +++ b/rust/routee-compass-core/src/algorithm/search/search_error.rs @@ -1,43 +1,60 @@ use crate::model::{ access::access_model_error::AccessModelError, - cost::cost_error::CostError, + cost::cost_model_error::CostModelError, frontier::frontier_model_error::FrontierModelError, - road_network::{edge_id::EdgeId, graph_error::GraphError, vertex_id::VertexId}, - state::state_error::StateError, + network::{edge_id::EdgeId, network_error::NetworkError, vertex_id::VertexId}, + state::state_model_error::StateModelError, termination::termination_model_error::TerminationModelError, traversal::traversal_model_error::TraversalModelError, }; #[derive(thiserror::Error, Debug)] pub enum SearchError { - #[error("distance heuristic can only be provided when there is a target")] - DistanceHeuristicWithNoTarget, - #[error(transparent)] - StateError(#[from] StateError), - #[error(transparent)] - GraphError(#[from] GraphError), - #[error(transparent)] - TerminationModelError(#[from] TerminationModelError), - #[error(transparent)] - TraversalModelFailure(#[from] TraversalModelError), - #[error(transparent)] - AccessModelFailure(#[from] AccessModelError), - #[error(transparent)] - FrontierModelFailure(#[from] FrontierModelError), - #[error(transparent)] - CostError(#[from] CostError), - #[error("loop in search result revisits edge {0}")] - LoopInSearchResult(EdgeId), + #[error("failure building search algorithm: {0}")] + BuildError(String), + #[error("The search failed due to state model error. The state model is responsible for updates to the state of the search at each increment of traversal. Please review the [state] section of your Compass configuration. Source: {source}")] + StateFailure { + #[from] + source: StateModelError, + }, + #[error("The search failed due to a road network error. Please review the [graph] section of your Compass configuration. Source: {source}")] + NetworkFailure { + #[from] + source: NetworkError, + }, + #[error("The search failed due to termination model error. Please review the [termination] section of your Compass Configuration to make changes. Source: {source}")] + TerminationModelFailure { + #[from] + source: TerminationModelError, + }, + #[error("The search failed due to traversal model error. The traversal model performs edge traversals by updating the various search dimensions, based on the state and cost models. Please review the [traversal] section of your Compass Configuration. Source: {source}")] + TraversalModelFailure { + #[from] + source: TraversalModelError, + }, + #[error("The search failed due to access model error. The access model performs edge-to-edge transitions by updating the various search dimensions, based on the state and cost models. Please review the [access] section of your Compass Configuration. Source: {source}")] + AccessModelFailure { + #[from] + source: AccessModelError, + }, + #[error("The search failed due to frontier model error. The frontier model restricts access to edges. Please review the [frontier] section of your Compass Configuration. Source: {source}")] + FrontierModelFailure { + #[from] + source: FrontierModelError, + }, + #[error("The search failed due to cost model error. The cost model interprets a delta of search state dimensions as having a cost value, which is minimized by the search. Please see the [cost] section of your Compass Configuration and additionally any query-time overrides. Source: {source}")] + CostFailure { + #[from] + source: CostModelError, + }, #[error("query terminated due to {0}")] QueryTerminated(String), #[error("no path exists between vertices {0} and {1}")] - NoPathExists(VertexId, VertexId), - #[error("search tree is missing linked vertex {0}")] - VertexMissingFromSearchTree(VertexId), + NoPathExistsBetweenVertices(VertexId, VertexId), + #[error("no path exists between edges {0} and {1}")] + NoPathExistsBetweenEdges(EdgeId, EdgeId), #[error("error accessing shared read-only dataset: {0}")] ReadOnlyPoisonError(String), - #[error("failure building search algorithm: {0}")] - BuildError(String), #[error("internal error due to search logic: {0}")] - InternalSearchError(String), + InternalError(String), } diff --git a/rust/routee-compass-core/src/algorithm/search/search_instance.rs b/rust/routee-compass-core/src/algorithm/search/search_instance.rs index b8d50e8b..e71e665c 100644 --- a/rust/routee-compass-core/src/algorithm/search/search_instance.rs +++ b/rust/routee-compass-core/src/algorithm/search/search_instance.rs @@ -3,7 +3,7 @@ use crate::model::{ access::access_model::AccessModel, cost::cost_model::CostModel, frontier::frontier_model::FrontierModel, - road_network::{graph::Graph, vertex_id::VertexId}, + network::{graph::Graph, vertex_id::VertexId}, state::state_model::StateModel, termination::termination_model::TerminationModel, traversal::{state::state_variable::StateVar, traversal_model::TraversalModel}, @@ -32,8 +32,8 @@ impl SearchInstance { dst: VertexId, state: &[StateVar], ) -> Result<Cost, SearchError> { - let src = self.directed_graph.get_vertex(src)?; - let dst = self.directed_graph.get_vertex(dst)?; + let src = self.directed_graph.get_vertex(&src)?; + let dst = self.directed_graph.get_vertex(&dst)?; let mut dst_state = state.to_vec(); self.traversal_model diff --git a/rust/routee-compass-core/src/algorithm/search/search_result.rs b/rust/routee-compass-core/src/algorithm/search/search_result.rs index c3538f31..e414fe92 100644 --- a/rust/routee-compass-core/src/algorithm/search/search_result.rs +++ b/rust/routee-compass-core/src/algorithm/search/search_result.rs @@ -1,5 +1,5 @@ use super::search_tree_branch::SearchTreeBranch; -use crate::model::road_network::vertex_id::VertexId; +use crate::model::network::vertex_id::VertexId; use std::collections::HashMap; #[derive(Default)] diff --git a/rust/routee-compass-core/src/algorithm/search/search_tree_branch.rs b/rust/routee-compass-core/src/algorithm/search/search_tree_branch.rs index 1845de44..92409d34 100644 --- a/rust/routee-compass-core/src/algorithm/search/search_tree_branch.rs +++ b/rust/routee-compass-core/src/algorithm/search/search_tree_branch.rs @@ -1,5 +1,5 @@ use crate::{ - algorithm::search::edge_traversal::EdgeTraversal, model::road_network::vertex_id::VertexId, + algorithm::search::edge_traversal::EdgeTraversal, model::network::vertex_id::VertexId, }; use allocative::Allocative; use serde::Serialize; diff --git a/rust/routee-compass-core/src/algorithm/search/util/edge_cut_frontier_model.rs b/rust/routee-compass-core/src/algorithm/search/util/edge_cut_frontier_model.rs index d243ab68..58b591fb 100644 --- a/rust/routee-compass-core/src/algorithm/search/util/edge_cut_frontier_model.rs +++ b/rust/routee-compass-core/src/algorithm/search/util/edge_cut_frontier_model.rs @@ -1,7 +1,9 @@ +use crate::model::{ + frontier::frontier_model::FrontierModel, + network::{Edge, EdgeId}, +}; use std::{collections::HashSet, sync::Arc}; -use crate::model::{frontier::frontier_model::FrontierModel, road_network::edge_id::EdgeId}; - /// A wrapper of the user-generated FrontierModel which prohibits traversals /// on selected edges. algorithms can create this wrapper with a set of "cut edges" /// and the search will not allow traversal of these edges. @@ -25,9 +27,9 @@ impl EdgeCutFrontierModel { impl FrontierModel for EdgeCutFrontierModel { fn valid_frontier( &self, - edge: &crate::model::property::edge::Edge, + edge: &Edge, state: &[crate::model::traversal::state::state_variable::StateVar], - previous_edge: Option<&crate::model::property::edge::Edge>, + previous_edge: Option<&Edge>, state_model: &crate::model::state::state_model::StateModel, ) -> Result<bool, crate::model::frontier::frontier_model_error::FrontierModelError> { if self.cut_edges.contains(&edge.edge_id) { diff --git a/rust/routee-compass-core/src/algorithm/search/util/route_similarity_function.rs b/rust/routee-compass-core/src/algorithm/search/util/route_similarity_function.rs index babad35e..cc219898 100644 --- a/rust/routee-compass-core/src/algorithm/search/util/route_similarity_function.rs +++ b/rust/routee-compass-core/src/algorithm/search/util/route_similarity_function.rs @@ -1,12 +1,11 @@ -use std::collections::{HashMap, HashSet}; - use crate::{ algorithm::search::{ edge_traversal::EdgeTraversal, search_error::SearchError, search_instance::SearchInstance, }, - model::{road_network::edge_id::EdgeId, unit::as_f64::AsF64}, + model::{network::edge_id::EdgeId, unit::as_f64::AsF64}, }; use serde::{Deserialize, Serialize}; +use std::collections::{HashMap, HashSet}; #[derive(Serialize, Deserialize, Default, Clone)] #[serde(rename_all = "snake_case", tag = "type")] @@ -21,6 +20,8 @@ pub enum RouteSimilarityFunction { }, } +type DistanceFunction<'a> = Box<dyn Fn(&'_ EdgeId) -> Result<f64, SearchError> + 'a>; + impl RouteSimilarityFunction { /// tests for similarity between two paths. /// @@ -79,15 +80,15 @@ impl RouteSimilarityFunction { match self { RouteSimilarityFunction::AcceptAll => Ok(0.0), RouteSimilarityFunction::EdgeIdCosineSimilarity { threshold: _ } => { - let unit_dist_fn = Box::new(\|_\| Ok(1.0)); + let unit_dist_fn = Box::new(\|_: &EdgeId\| Ok(1.0)); cos_similarity(a, b, unit_dist_fn) } RouteSimilarityFunction::DistanceWeightedCosineSimilarity { threshold: _ } => { - let dist_fn = Box::new(\|edge_id\| { + let dist_fn = Box::new(\|edge_id: &EdgeId\| { si.directed_graph .get_edge(edge_id) .map(\|edge\| edge.distance.as_f64()) - .map_err(SearchError::GraphError) + .map_err(SearchError::from) }); cos_similarity(a, b, dist_fn) } @@ -106,18 +107,18 @@ impl RouteSimilarityFunction { /// # Returns /// /// the cosine similarity of the routes, a value from -1 to 1 -fn cos_similarity<'a>( +fn cos_similarity( a: &[&EdgeTraversal], b: &[&EdgeTraversal], - dist_fn: Box<dyn Fn(EdgeId) -> Result<f64, SearchError> + 'a>, + dist_fn: DistanceFunction<'_>, ) -> Result<f64, SearchError> { let a_map = a .iter() - .map(\|e\| dist_fn(e.edge_id).map(\|dist\| (e.edge_id, dist))) + .map(\|e\| dist_fn(&e.edge_id).map(\|dist\| (e.edge_id, dist))) .collect::<Result<HashMap<_, _>, _>>()?; let b_map = b .iter() - .map(\|e\| dist_fn(e.edge_id).map(\|dist\| (e.edge_id, dist))) + .map(\|e\| dist_fn(&e.edge_id).map(\|dist\| (e.edge_id, dist))) .collect::<Result<HashMap<_, _>, _>>()?; let numer: f64 = a_map diff --git a/rust/routee-compass-core/src/doc.md b/rust/routee-compass-core/src/doc.md index e39a9208..bbb2ee72 100644 --- a/rust/routee-compass-core/src/doc.md +++ b/rust/routee-compass-core/src/doc.md @@ -20,8 +20,8 @@ This can be broken up into the following modules: RouteE Compass takes a layered approach to modeling the road network. At the core is the [Graph] model. The edges and vertices are stored in `vec`s and carry the minimal data required for most search applications: -- [`crate::model::property::edge::Edge`] records store distance in meters -- [`crate::model::property::vertex::Vertex`] records store coordinate locations, assumed in WGS84 projection +- [`crate::model::road_network::Edge`] records store distance in meters +- [`crate::model::road_network::Vertex`] records store coordinate locations, assumed in WGS84 projection A forward and reverse adjacency list describes connectivity in the graph using the indices of the edges and vertices in their respective `vec`s. diff --git a/rust/routee-compass-core/src/model/access/access_model.rs b/rust/routee-compass-core/src/model/access/access_model.rs index f6f0597e..02254ebf 100644 --- a/rust/routee-compass-core/src/model/access/access_model.rs +++ b/rust/routee-compass-core/src/model/access/access_model.rs @@ -1,6 +1,6 @@ use super::access_model_error::AccessModelError; use crate::model::{ - property::{edge::Edge, vertex::Vertex}, + network::{Edge, Vertex}, state::{state_feature::StateFeature, state_model::StateModel}, traversal::state::state_variable::StateVar, }; diff --git a/rust/routee-compass-core/src/model/access/access_model_error.rs b/rust/routee-compass-core/src/model/access/access_model_error.rs index 1787a77d..b28307ec 100644 --- a/rust/routee-compass-core/src/model/access/access_model_error.rs +++ b/rust/routee-compass-core/src/model/access/access_model_error.rs @@ -1,13 +1,14 @@ -use crate::model::state::state_error::StateError; +use crate::model::state::state_model_error::StateModelError; #[derive(thiserror::Error, Debug)] pub enum AccessModelError { #[error("error while executing access model {name}: {error}")] RuntimeError { name: String, error: String }, - #[error(transparent)] - StateError(#[from] StateError), - #[error(transparent)] - SerdeJsonError(#[from] serde_json::Error), + #[error("access model failed due to state transition error: {source}")] + StateError { + #[from] + source: StateModelError, + }, #[error("{0}")] BuildError(String), } diff --git a/rust/routee-compass-core/src/model/access/default/combined_model.rs b/rust/routee-compass-core/src/model/access/default/combined_model.rs index 118bc62f..ad350a2b 100644 --- a/rust/routee-compass-core/src/model/access/default/combined_model.rs +++ b/rust/routee-compass-core/src/model/access/default/combined_model.rs @@ -3,7 +3,7 @@ use crate::model::{ access_model::AccessModel, access_model_error::AccessModelError, access_model_service::AccessModelService, }, - property::{edge::Edge, vertex::Vertex}, + network::{Edge, Vertex}, state::{state_feature::StateFeature, state_model::StateModel}, traversal::state::state_variable::StateVar, }; diff --git a/rust/routee-compass-core/src/model/access/default/no_access_model.rs b/rust/routee-compass-core/src/model/access/default/no_access_model.rs index 236adc53..812ab2c6 100644 --- a/rust/routee-compass-core/src/model/access/default/no_access_model.rs +++ b/rust/routee-compass-core/src/model/access/default/no_access_model.rs @@ -1,9 +1,11 @@ -use std::sync::Arc; - -use crate::model::access::{ - access_model::AccessModel, access_model_builder::AccessModelBuilder, - access_model_service::AccessModelService, +use crate::model::{ + access::{ + access_model::AccessModel, access_model_builder::AccessModelBuilder, + access_model_service::AccessModelService, + }, + network::{Edge, Vertex}, }; +use std::sync::Arc; #[derive(Clone, Debug)] pub struct NoAccessModel {} @@ -15,15 +17,9 @@ impl AccessModel for NoAccessModel { fn access_edge( &self, - _traversal: ( - &crate::model::property::vertex::Vertex, - &crate::model::property::edge::Edge, - &crate::model::property::vertex::Vertex, - &crate::model::property::edge::Edge, - &crate::model::property::vertex::Vertex, - ), - _state: &mut Vec<crate::model::traversal::state::state_variable::StateVar>, - _state_model: &crate::model::state::state_model::StateModel, + _: (&Vertex, &Edge, &Vertex, &Edge, &Vertex), + _: &mut Vec<crate::model::traversal::state::state_variable::StateVar>, + _: &crate::model::state::state_model::StateModel, ) -> Result<(), crate::model::access::access_model_error::AccessModelError> { Ok(()) } diff --git a/rust/routee-compass-core/src/model/access/default/turn_delays/turn_delay_access_model.rs b/rust/routee-compass-core/src/model/access/default/turn_delays/turn_delay_access_model.rs index a019c2ab..917d19ac 100644 --- a/rust/routee-compass-core/src/model/access/default/turn_delays/turn_delay_access_model.rs +++ b/rust/routee-compass-core/src/model/access/default/turn_delays/turn_delay_access_model.rs @@ -1,7 +1,7 @@ use super::turn_delay_access_model_engine::TurnDelayAccessModelEngine; use crate::model::{ access::{access_model::AccessModel, access_model_error::AccessModelError}, - property::{edge::Edge, vertex::Vertex}, + network::{Edge, Vertex}, state::{state_feature::StateFeature, state_model::StateModel}, traversal::state::state_variable::StateVar, }; diff --git a/rust/routee-compass-core/src/model/access/default/turn_delays/turn_delay_access_model_engine.rs b/rust/routee-compass-core/src/model/access/default/turn_delays/turn_delay_access_model_engine.rs index 7fe963fa..32e8ab15 100644 --- a/rust/routee-compass-core/src/model/access/default/turn_delays/turn_delay_access_model_engine.rs +++ b/rust/routee-compass-core/src/model/access/default/turn_delays/turn_delay_access_model_engine.rs @@ -1,12 +1,9 @@ -use crate::model::access::access_model_error::AccessModelError; -use crate::model::property::edge::Edge; -use crate::model::property::vertex::Vertex; -use crate::model::road_network::edge_id::EdgeId; -use crate::model::unit::{Time, TimeUnit}; - use super::edge_heading::EdgeHeading; use super::turn::Turn; use super::turn_delay_model::TurnDelayModel; +use crate::model::access::access_model_error::AccessModelError; +use crate::model::network::{Edge, EdgeId, Vertex}; +use crate::model::unit::{Time, TimeUnit}; pub struct TurnDelayAccessModelEngine { pub edge_headings: Box<[EdgeHeading]>, diff --git a/rust/routee-compass-core/src/model/cost/cost_aggregation.rs b/rust/routee-compass-core/src/model/cost/cost_aggregation.rs index e0094239..968be6ba 100644 --- a/rust/routee-compass-core/src/model/cost/cost_aggregation.rs +++ b/rust/routee-compass-core/src/model/cost/cost_aggregation.rs @@ -1,7 +1,7 @@ use crate::model::unit::{as_f64::AsF64, Cost}; use serde::{Deserialize, Serialize}; -use super::cost_error::CostError; +use super::cost_model_error::CostModelError; #[derive(Deserialize, Serialize, Clone, Copy, Debug, Default)] #[serde(rename_all = "snake_case")] @@ -29,8 +29,8 @@ impl CostAggregation { pub fn agg_iter<'a>( &self, - costs: impl Iterator<Item = Result<(&'a String, Cost), CostError>>, - ) -> Result<Cost, CostError> { + costs: impl Iterator<Item = Result<(&'a String, Cost), CostModelError>>, + ) -> Result<Cost, CostModelError> { match self { CostAggregation::Sum => { let mut sum = Cost::ZERO; diff --git a/rust/routee-compass-core/src/model/cost/cost_model.rs b/rust/routee-compass-core/src/model/cost/cost_model.rs index 44bfd6c4..99b3578d 100644 --- a/rust/routee-compass-core/src/model/cost/cost_model.rs +++ b/rust/routee-compass-core/src/model/cost/cost_model.rs @@ -2,8 +2,8 @@ use super::cost_aggregation::CostAggregation; use super::cost_ops; use super::network::network_cost_rate::NetworkCostRate; use super::vehicle::vehicle_cost_rate::VehicleCostRate; -use crate::model::cost::cost_error::CostError; -use crate::model::property::edge::Edge; +use crate::model::cost::cost_model_error::CostModelError; +use crate::model::network::Edge; use crate::model::state::state_model::StateModel; use crate::model::traversal::state::state_variable::StateVar; use crate::model::unit::Cost; @@ -52,7 +52,7 @@ impl CostModel { network_rate_mapping: Arc<HashMap<String, NetworkCostRate>>, cost_aggregation: CostAggregation, state_model: Arc<StateModel>, - ) -> Result<CostModel, CostError> { + ) -> Result<CostModel, CostModelError> { let mut indices = vec![]; let mut weights = vec![]; let mut vehicle_rates = vec![]; @@ -75,7 +75,7 @@ impl CostModel { } if weights.iter().sum::<f64>() == 0.0 { - return Err(CostError::InvalidCostVariables); + return Err(CostModelError::InvalidCostVariables); } Ok(CostModel { feature_indices: indices, @@ -102,7 +102,7 @@ impl CostModel { edge: &Edge, prev_state: &[StateVar], next_state: &[StateVar], - ) -> Result<Cost, CostError> { + ) -> Result<Cost, CostModelError> { let vehicle_cost = cost_ops::calculate_vehicle_costs( (prev_state, next_state), &self.feature_indices, @@ -146,7 +146,7 @@ impl CostModel { next_edge: &Edge, prev_state: &[StateVar], next_state: &[StateVar], - ) -> Result<Cost, CostError> { + ) -> Result<Cost, CostModelError> { let vehicle_cost = cost_ops::calculate_vehicle_costs( (prev_state, next_state), &self.feature_indices, @@ -184,7 +184,7 @@ impl CostModel { &self, src_state: &[StateVar], dst_state: &[StateVar], - ) -> Result<Cost, CostError> { + ) -> Result<Cost, CostModelError> { let vehicle_cost = cost_ops::calculate_vehicle_costs( (src_state, dst_state), &self.feature_indices, @@ -207,14 +207,14 @@ impl CostModel { /// A JSON serialized version of the state. This does not need to include /// additional details such as the units (kph, hours, etc), which can be /// summarized in the serialize_state_info method. - pub fn serialize_cost(&self, state: &[StateVar]) -> Result<serde_json::Value, CostError> { + pub fn serialize_cost(&self, state: &[StateVar]) -> Result<serde_json::Value, CostModelError> { let mut state_variable_costs = self .feature_indices .iter() .map(move \|(name, idx)\| { let state_var = state .get(idx) - .ok_or_else(\|\| CostError::StateIndexOutOfBounds(idx, name.clone()))?; + .ok_or_else(\|\| CostModelError::StateIndexOutOfBounds(idx, name.clone()))?; let rate = self.vehicle_rates.get(idx).ok_or_else(\|\| { let alternatives = self @@ -224,7 +224,7 @@ impl CostModel { .map(\|(n, _)\| n.to_string()) .collect::<Vec<_>>() .join(","); - CostError::StateVariableNotFound( + CostModelError::StateVariableNotFound( name.clone(), String::from("vehicle cost rates while serializing cost"), alternatives, @@ -233,7 +233,7 @@ impl CostModel { let cost = rate.map_value(state_var); Ok((name.clone(), cost)) }) - .collect::<Result<HashMap<String, Cost>, CostError>>()?; + .collect::<Result<HashMap<String, Cost>, CostModelError>>()?; let total_cost = state_variable_costs .values() @@ -255,27 +255,27 @@ impl CostModel { /// /// JSON containing information such as the units (kph, hours, etc) or other /// traversal info (charge events, days traveled, etc) - pub fn serialize_cost_info(&self) -> Result<serde_json::Value, CostError> { + pub fn serialize_cost_info(&self) -> Result<serde_json::Value, CostModelError> { let mut result = serde_json::Map::with_capacity(self.feature_indices.len()); for (name, index) in self.feature_indices.iter() { let weight = self .weights .get(index) - .ok_or(CostError::CostVectorOutOfBounds( + .ok_or(CostModelError::CostVectorOutOfBounds( index, String::from(Self::WEIGHTS), ))?; let veh_rate = self.vehicle_rates .get(index) - .ok_or(CostError::CostVectorOutOfBounds( + .ok_or(CostModelError::CostVectorOutOfBounds( index, String::from(Self::VEHICLE_RATES), ))?; let net_rate = self.network_rates .get(index) - .ok_or(CostError::CostVectorOutOfBounds( + .ok_or(CostModelError::CostVectorOutOfBounds( index, String::from(Self::NETWORK_RATES), ))?; diff --git a/rust/routee-compass-core/src/model/cost/cost_error.rs b/rust/routee-compass-core/src/model/cost/cost_model_error.rs similarity index 82% rename from rust/routee-compass-core/src/model/cost/cost_error.rs rename to rust/routee-compass-core/src/model/cost/cost_model_error.rs index 71a9cc7a..6e67aa42 100644 --- a/rust/routee-compass-core/src/model/cost/cost_error.rs +++ b/rust/routee-compass-core/src/model/cost/cost_model_error.rs @@ -1,10 +1,7 @@ #[derive(thiserror::Error, Debug)] -pub enum CostError { - #[error("failure reading CSV: {source}")] - CsvIoError { - #[from] - source: csv::Error, - }, +pub enum CostModelError { + #[error("{0}")] + BuildError(String), #[error("invalid cost model configuration: {0}")] InvalidConfiguration(String), #[error( diff --git a/rust/routee-compass-core/src/model/cost/cost_ops.rs b/rust/routee-compass-core/src/model/cost/cost_ops.rs index aa6bbd39..db54b22b 100644 --- a/rust/routee-compass-core/src/model/cost/cost_ops.rs +++ b/rust/routee-compass-core/src/model/cost/cost_ops.rs @@ -1,8 +1,8 @@ use super::{ - cost_aggregation::CostAggregation, cost_error::CostError, + cost_aggregation::CostAggregation, cost_model_error::CostModelError, network::network_cost_rate::NetworkCostRate, vehicle::vehicle_cost_rate::VehicleCostRate, }; -use crate::model::{property::edge::Edge, traversal::state::state_variable::StateVar, unit::Cost}; +use crate::model::{network::Edge, traversal::state::state_variable::StateVar, unit::Cost}; /// steps through each state variable and assigns vehicle costs related to that variable /// due to an edge access + traversal event. @@ -17,25 +17,25 @@ pub fn calculate_vehicle_costs( weights: &[f64], rates: &[VehicleCostRate], cost_aggregation: &CostAggregation, -) -> Result<Cost, CostError> { +) -> Result<Cost, CostModelError> { let (prev_state, next_state) = state_sequence; let costs = indices.iter().map(\|(name, state_idx)\| { // compute delta let prev_state_var = prev_state .get(state_idx) - .ok_or_else(\|\| CostError::StateIndexOutOfBounds(state_idx, name.clone()))?; + .ok_or_else(\|\| CostModelError::StateIndexOutOfBounds(state_idx, name.clone()))?; let next_state_var = next_state .get(state_idx) - .ok_or_else(\|\| CostError::StateIndexOutOfBounds(state_idx, name.clone()))?; + .ok_or_else(\|\| CostModelError::StateIndexOutOfBounds(state_idx, name.clone()))?; let delta: StateVar = next_state_var - prev_state_var; // collect weight and vehicle cost rate let mapping = rates.get(state_idx).ok_or_else(\|\| { - CostError::CostVectorOutOfBounds(state_idx, String::from("vehicle_rates")) + CostModelError::CostVectorOutOfBounds(state_idx, String::from("vehicle_rates")) + })?; + let weight = weights.get(state_idx).ok_or_else(\|\| { + CostModelError::CostVectorOutOfBounds(state_idx, String::from("weights")) })?; - let weight = weights - .get(state_idx) - .ok_or_else(\|\| CostError::CostVectorOutOfBounds(state_idx, String::from("weights")))?; // compute cost let delta_cost = mapping.map_value(delta); @@ -53,22 +53,22 @@ pub fn calculate_network_traversal_costs( weights: &[f64], rates: &[NetworkCostRate], cost_aggregation: &CostAggregation, -) -> Result<Cost, CostError> { +) -> Result<Cost, CostModelError> { let (prev_state, next_state) = state_sequence; let costs = indices.iter().map(\|(name, state_idx)\| { let prev_state_var = prev_state .get(state_idx) - .ok_or_else(\|\| CostError::StateIndexOutOfBounds(state_idx, name.clone()))?; + .ok_or_else(\|\| CostModelError::StateIndexOutOfBounds(state_idx, name.clone()))?; let next_state_var = next_state .get(state_idx) - .ok_or_else(\|\| CostError::StateIndexOutOfBounds(state_idx, name.clone()))?; + .ok_or_else(\|\| CostModelError::StateIndexOutOfBounds(state_idx, name.clone()))?; // determine weight and access cost for this state feature - let weight = weights - .get(state_idx) - .ok_or_else(\|\| CostError::CostVectorOutOfBounds(state_idx, String::from("weights")))?; + let weight = weights.get(state_idx).ok_or_else(\|\| { + CostModelError::CostVectorOutOfBounds(state_idx, String::from("weights")) + })?; let rate = rates.get(state_idx).ok_or_else(\|\| { - CostError::CostVectorOutOfBounds(state_idx, String::from("network_cost_rate")) + CostModelError::CostVectorOutOfBounds(state_idx, String::from("network_cost_rate")) })?; let access_cost = rate.traversal_cost(prev_state_var, next_state_var, edge)?; let cost = access_cost weight; @@ -85,7 +85,7 @@ pub fn calculate_network_access_costs( weights: &[f64], rates: &[NetworkCostRate], cost_aggregation: &CostAggregation, -) -> Result<Cost, CostError> { +) -> Result<Cost, CostModelError> { let (prev_state, next_state) = state_sequence; let (prev_edge, next_edge) = edge_sequence; let costs = indices.iter().map(\|(name, idx)\| match rates.get(idx) { @@ -93,10 +93,10 @@ pub fn calculate_network_access_costs( Some(m) => { let prev_state_var = prev_state .get(idx) - .ok_or_else(\|\| CostError::StateIndexOutOfBounds(idx, name.clone()))?; + .ok_or_else(\|\| CostModelError::StateIndexOutOfBounds(idx, name.clone()))?; let next_state_var = next_state .get(idx) - .ok_or_else(\|\| CostError::StateIndexOutOfBounds(idx, name.clone()))?; + .ok_or_else(\|\| CostModelError::StateIndexOutOfBounds(idx, name.clone()))?; let access_cost = m.access_cost(prev_state_var, next_state_var, prev_edge, next_edge)?; let coefficient = weights.get(idx).unwrap_or(&1.0); diff --git a/rust/routee-compass-core/src/model/cost/mod.rs b/rust/routee-compass-core/src/model/cost/mod.rs index 6fee0752..ee9ff94f 100644 --- a/rust/routee-compass-core/src/model/cost/mod.rs +++ b/rust/routee-compass-core/src/model/cost/mod.rs @@ -1,6 +1,6 @@ pub mod cost_aggregation; -pub mod cost_error; pub mod cost_model; +pub mod cost_model_error; pub mod cost_ops; pub mod network; pub mod vehicle; diff --git a/rust/routee-compass-core/src/model/cost/network/network_access_cost_row.rs b/rust/routee-compass-core/src/model/cost/network/network_access_cost_row.rs index d073af28..1f1bcb26 100644 --- a/rust/routee-compass-core/src/model/cost/network/network_access_cost_row.rs +++ b/rust/routee-compass-core/src/model/cost/network/network_access_cost_row.rs @@ -1,4 +1,4 @@ -use crate::model::{road_network::edge_id::EdgeId, unit::Cost}; +use crate::model::{network::edge_id::EdgeId, unit::Cost}; use serde::{Deserialize, Serialize}; #[derive(Serialize, Deserialize)] diff --git a/rust/routee-compass-core/src/model/cost/network/network_cost_rate.rs b/rust/routee-compass-core/src/model/cost/network/network_cost_rate.rs index e77cdee0..80303232 100644 --- a/rust/routee-compass-core/src/model/cost/network/network_cost_rate.rs +++ b/rust/routee-compass-core/src/model/cost/network/network_cost_rate.rs @@ -1,7 +1,7 @@ -use crate::model::cost::cost_error::CostError; -use crate::model::property::edge::Edge; +use crate::model::network::Edge; +use crate::model::traversal::state::state_variable::StateVar; use crate::model::unit::Cost; -use crate::model::{road_network::edge_id::EdgeId, traversal::state::state_variable::StateVar}; +use crate::model::{cost::cost_model_error::CostModelError, network::EdgeId}; use serde::{Deserialize, Serialize}; use std::collections::HashMap; @@ -30,7 +30,7 @@ impl NetworkCostRate { _prev_state_var: StateVar, _next_state_var: StateVar, edge: &Edge, - ) -> Result<Cost, CostError> { + ) -> Result<Cost, CostModelError> { match self { NetworkCostRate::Zero => Ok(Cost::ZERO), NetworkCostRate::EdgeEdgeLookup { lookup: _ } => Ok(Cost::ZERO), @@ -42,7 +42,7 @@ impl NetworkCostRate { let mapped = mappings .iter() .map(\|f\| f.traversal_cost(_prev_state_var, _next_state_var, edge)) - .collect::<Result<Vec<Cost>, CostError>>()?; + .collect::<Result<Vec<Cost>, CostModelError>>()?; let cost = mapped.iter().fold(Cost::ZERO, \|a, b\| a + b); Ok(cost) @@ -58,7 +58,7 @@ impl NetworkCostRate { /// `next_state_var` - the state variable after accessing the next edge origin /// * `prev_edge` - the edge traversed to reach the next_edge (or none if at origin) /// * `next_edge` - the edge we are attempting to access (not yet traversed) - + /// /// # Result /// /// the Cost value for that state, a real number that is aggregated with @@ -69,7 +69,7 @@ impl NetworkCostRate { _next_state_var: StateVar, prev_edge: &Edge, next_edge: &Edge, - ) -> Result<Cost, CostError> { + ) -> Result<Cost, CostModelError> { match self { NetworkCostRate::Zero => Ok(Cost::ZERO), NetworkCostRate::EdgeLookup { lookup: _ } => Ok(Cost::ZERO), @@ -83,7 +83,7 @@ impl NetworkCostRate { let mapped = mappings .iter() .map(\|f\| f.access_cost(_prev_state_var, _next_state_var, prev_edge, next_edge)) - .collect::<Result<Vec<Cost>, CostError>>()?; + .collect::<Result<Vec<Cost>, CostModelError>>()?; let cost = mapped.iter().fold(Cost::ZERO, \|a, b\| a + b); Ok(cost) diff --git a/rust/routee-compass-core/src/model/cost/network/network_cost_rate_builder.rs b/rust/routee-compass-core/src/model/cost/network/network_cost_rate_builder.rs index dbc32000..f421d7e6 100644 --- a/rust/routee-compass-core/src/model/cost/network/network_cost_rate_builder.rs +++ b/rust/routee-compass-core/src/model/cost/network/network_cost_rate_builder.rs @@ -1,5 +1,5 @@ use super::network_cost_rate::NetworkCostRate; -use crate::model::cost::cost_error::CostError; +use crate::model::cost::cost_model_error::CostModelError; use crate::{ model::cost::network::{ network_access_cost_row::NetworkAccessUtilityRow, @@ -22,14 +22,19 @@ pub enum NetworkCostRateBuilder { } impl NetworkCostRateBuilder { - pub fn build(&self) -> Result<NetworkCostRate, CostError> { + pub fn build(&self) -> Result<NetworkCostRate, CostModelError> { use NetworkCostRate as NCM; use NetworkCostRateBuilder as Builder; match self { Builder::EdgeLookupBuilder { cost_input_file } => { let lookup = read_utils::from_csv::<NetworkTraversalUtilityRow>(cost_input_file, true, None) - .map_err(\|source\| CostError::CsvIoError { source })? + .map_err(\|source\| { + CostModelError::BuildError(format!( + "failure reading file {}: {}", + cost_input_file, source + )) + })? .iter() .map(\|row\| (row.edge_id, row.cost)) .collect::<HashMap<_, _>>(); @@ -38,7 +43,12 @@ impl NetworkCostRateBuilder { Builder::EdgeEdgeLookupBuilder { cost_input_file } => { let lookup = read_utils::from_csv::<NetworkAccessUtilityRow>(cost_input_file, true, None) - .map_err(\|source\| CostError::CsvIoError { source })? + .map_err(\|source\| { + CostModelError::BuildError(format!( + "failure reading file {}: {}", + cost_input_file, source + )) + })? .iter() .map(\|row\| ((row.source, row.destination), row.cost)) .collect::<HashMap<_, _>>(); @@ -49,7 +59,7 @@ impl NetworkCostRateBuilder { let mappings = builders .iter() .map(\|b\| b.build()) - .collect::<Result<Vec<_>, CostError>>()?; + .collect::<Result<Vec<_>, CostModelError>>()?; Ok(NCM::Combined(mappings)) } } diff --git a/rust/routee-compass-core/src/model/cost/network/network_traversal_cost_row.rs b/rust/routee-compass-core/src/model/cost/network/network_traversal_cost_row.rs index 550e42cd..88cc4aa9 100644 --- a/rust/routee-compass-core/src/model/cost/network/network_traversal_cost_row.rs +++ b/rust/routee-compass-core/src/model/cost/network/network_traversal_cost_row.rs @@ -1,4 +1,4 @@ -use crate::model::{road_network::edge_id::EdgeId, unit::Cost}; +use crate::model::{network::edge_id::EdgeId, unit::Cost}; use serde::{Deserialize, Serialize}; #[derive(Serialize, Deserialize)] diff --git a/rust/routee-compass-core/src/model/frontier/frontier_model.rs b/rust/routee-compass-core/src/model/frontier/frontier_model.rs index 71734df7..135704d7 100644 --- a/rust/routee-compass-core/src/model/frontier/frontier_model.rs +++ b/rust/routee-compass-core/src/model/frontier/frontier_model.rs @@ -1,7 +1,6 @@ use super::frontier_model_error::FrontierModelError; use crate::model::{ - property::edge::Edge, state::state_model::StateModel, - traversal::state::state_variable::StateVar, + network::Edge, state::state_model::StateModel, traversal::state::state_variable::StateVar, }; /// Validates edge and traversal states. Provides an API for removing edges from diff --git a/rust/routee-compass-core/src/model/frontier/frontier_model_error.rs b/rust/routee-compass-core/src/model/frontier/frontier_model_error.rs index d177a33b..bdee0c42 100644 --- a/rust/routee-compass-core/src/model/frontier/frontier_model_error.rs +++ b/rust/routee-compass-core/src/model/frontier/frontier_model_error.rs @@ -2,6 +2,6 @@ pub enum FrontierModelError { #[error("failure building frontier model: {0}")] BuildError(String), - #[error("edge id {0} missing from frontier model file")] - MissingIndex(String), + #[error("{0}")] + FrontierModelError(String), } diff --git a/rust/routee-compass-core/src/model/mod.rs b/rust/routee-compass-core/src/model/mod.rs index cf79cd37..c2922796 100644 --- a/rust/routee-compass-core/src/model/mod.rs +++ b/rust/routee-compass-core/src/model/mod.rs @@ -1,8 +1,7 @@ pub mod access; pub mod cost; pub mod frontier; -pub mod property; -pub mod road_network; +pub mod network; pub mod state; pub mod termination; pub mod traversal; diff --git a/rust/routee-compass-core/src/model/property/edge.rs b/rust/routee-compass-core/src/model/network/edge.rs similarity index 93% rename from rust/routee-compass-core/src/model/property/edge.rs rename to rust/routee-compass-core/src/model/network/edge.rs index 6a7a44b4..2f3bd972 100644 --- a/rust/routee-compass-core/src/model/property/edge.rs +++ b/rust/routee-compass-core/src/model/network/edge.rs @@ -1,4 +1,4 @@ -use crate::model::road_network::{edge_id::EdgeId, vertex_id::VertexId}; +use crate::model::network::{edge_id::EdgeId, vertex_id::VertexId}; use crate::model::unit::Distance; use allocative::Allocative; use serde::{Deserialize, Serialize}; diff --git a/rust/routee-compass-core/src/model/road_network/edge_id.rs b/rust/routee-compass-core/src/model/network/edge_id.rs similarity index 100% rename from rust/routee-compass-core/src/model/road_network/edge_id.rs rename to rust/routee-compass-core/src/model/network/edge_id.rs diff --git a/rust/routee-compass-core/src/model/road_network/edge_loader.rs b/rust/routee-compass-core/src/model/network/edge_loader.rs similarity index 84% rename from rust/routee-compass-core/src/model/road_network/edge_loader.rs rename to rust/routee-compass-core/src/model/network/edge_loader.rs index 3e20243f..95c3c6f5 100644 --- a/rust/routee-compass-core/src/model/road_network/edge_loader.rs +++ b/rust/routee-compass-core/src/model/network/edge_loader.rs @@ -1,10 +1,5 @@ -use crate::{ - model::{ - property::edge::Edge, - road_network::{edge_id::EdgeId, graph_error::GraphError, vertex_id::VertexId}, - }, - util::{compact_ordered_hash_map::CompactOrderedHashMap, fs::read_utils}, -}; +use super::{Edge, EdgeId, NetworkError, VertexId}; +use crate::util::{compact_ordered_hash_map::CompactOrderedHashMap, fs::read_utils}; use kdam::Bar; use kdam::BarExt; use std::{collections::HashSet, path::PathBuf}; @@ -22,7 +17,7 @@ pub struct EdgeLoaderConfig { } impl TryFrom<EdgeLoaderConfig> for EdgeLoader { - type Error = GraphError; + type Error = NetworkError; fn try_from(c: EdgeLoaderConfig) -> Result<Self, Self::Error> { let mut adj: Vec<CompactOrderedHashMap<EdgeId, VertexId>> = @@ -35,7 +30,9 @@ impl TryFrom<EdgeLoaderConfig> for EdgeLoader { .animation("fillup") .desc("edge list") .build() - .map_err(\|e\| GraphError::ProgressBarBuildError(String::from("edge list"), e))?; + .map_err(\|e\| { + NetworkError::InternalError(format!("could not build progress bar: {}", e)) + })?; let mut missing_vertices: HashSet<VertexId> = HashSet::new(); let cb = Box::new(\|edge: &Edge\| { @@ -61,7 +58,7 @@ impl TryFrom<EdgeLoaderConfig> for EdgeLoader { let edges = read_utils::from_csv(&c.edge_list_csv, true, Some(cb))?; - println!(); + eprintln!(); let result = EdgeLoader { edges, adj: adj.into_boxed_slice(), diff --git a/rust/routee-compass-core/src/model/road_network/graph.rs b/rust/routee-compass-core/src/model/network/graph.rs similarity index 74% rename from rust/routee-compass-core/src/model/road_network/graph.rs rename to rust/routee-compass-core/src/model/network/graph.rs index efe53dbc..402b55bc 100644 --- a/rust/routee-compass-core/src/model/road_network/graph.rs +++ b/rust/routee-compass-core/src/model/network/graph.rs @@ -1,14 +1,12 @@ +use super::{Edge, EdgeId, NetworkError, Vertex, VertexId}; use crate::algorithm::search::direction::Direction; -use crate::model::property::edge::Edge; -use crate::model::property::vertex::Vertex; -use crate::model::road_network::graph_error::GraphError; -use crate::model::road_network::{edge_id::EdgeId, vertex_id::VertexId}; use crate::util::compact_ordered_hash_map::CompactOrderedHashMap; use std::path::Path; use super::graph_loader::graph_from_files; use allocative::Allocative; +use itertools::Itertools; /// Road network topology represented as an adjacency list. /// The `EdgeId` and `VertexId` values correspond to edge and @@ -61,7 +59,7 @@ impl Graph { n_edges: Option<usize>, n_vertices: Option<usize>, verbose: Option<bool>, - ) -> Result<Graph, GraphError> { + ) -> Result<Graph, NetworkError> { graph_from_files(edge_list_csv, vertex_list_csv, n_edges, n_vertices, verbose) } /// number of edges in the Graph @@ -99,9 +97,9 @@ impl Graph { /// # Returns /// /// The associated `Edge` or an error if the id is missing - pub fn get_edge(&self, edge_id: EdgeId) -> Result<&Edge, GraphError> { + pub fn get_edge(&self, edge_id: &EdgeId) -> Result<&Edge, NetworkError> { match self.edges.get(edge_id.0) { - None => Err(GraphError::EdgeAttributeNotFound { edge_id }), + None => Err(NetworkError::EdgeNotFound(edge_id)), Some(edge) => Ok(edge), } } @@ -115,9 +113,9 @@ impl Graph { /// # Returns /// /// The associated `Vertex` or an error if the id is missing - pub fn get_vertex(&self, vertex_id: VertexId) -> Result<&Vertex, GraphError> { + pub fn get_vertex(&self, vertex_id: &VertexId) -> Result<&Vertex, NetworkError> { match self.vertices.get(vertex_id.0) { - None => Err(GraphError::VertexAttributeNotFound { vertex_id }), + None => Err(NetworkError::VertexNotFound(vertex_id)), Some(vertex) => Ok(vertex), } } @@ -132,26 +130,17 @@ impl Graph { /// /// A list of `EdgeIds` for outbound edges that leave this `VertexId`, or an error /// if the vertex is missing from the Graph adjacency matrix. - pub fn out_edges(&self, src: VertexId) -> Result<Vec<EdgeId>, GraphError> { - match self.adj.get(src.0) { - None => Err(GraphError::VertexWithoutOutEdges { vertex_id: src }), - Some(out_map) => { - let edge_ids = out_map.keys().cloned().collect(); - Ok(edge_ids) - } - } + pub fn out_edges(&self, src: &VertexId) -> Vec<EdgeId> { + self.out_edges_iter(src).cloned().collect_vec() } pub fn out_edges_iter<'a>( &'a self, - src: VertexId, - ) -> Result<Box<dyn Iterator<Item = &EdgeId> + 'a>, GraphError> { + src: &VertexId, + ) -> Box<dyn Iterator<Item = &'a EdgeId> + 'a> { match self.adj.get(src.0) { - None => Err(GraphError::VertexWithoutOutEdges { vertex_id: src }), - Some(out_map) => { - let edge_ids = out_map.keys(); - Ok(edge_ids) - } + None => Box::new(std::iter::empty()), + Some(out_map) => out_map.keys(), } } @@ -165,26 +154,17 @@ impl Graph { /// /// A list of `EdgeIds` for inbound edges that arrive at this `VertexId`, or an error /// if the vertex is missing from the Graph adjacency matrix. - pub fn in_edges(&self, dst: VertexId) -> Result<Vec<EdgeId>, GraphError> { - match self.rev.get(dst.0) { - None => Err(GraphError::VertexWithoutInEdges { vertex_id: dst }), - Some(in_map) => { - let edge_ids = in_map.keys().cloned().collect(); - Ok(edge_ids) - } - } + pub fn in_edges(&self, dst: &VertexId) -> Vec<EdgeId> { + self.in_edges_iter(dst).cloned().collect_vec() } pub fn in_edges_iter<'a>( &'a self, - dst: VertexId, - ) -> Result<Box<dyn Iterator<Item = &EdgeId> + 'a>, GraphError> { + dst: &VertexId, + ) -> Box<dyn Iterator<Item = &'a EdgeId> + 'a> { match self.rev.get(dst.0) { - None => Err(GraphError::VertexWithoutInEdges { vertex_id: dst }), - Some(in_map) => { - let edge_ids = in_map.keys(); - Ok(edge_ids) - } + None => Box::new(std::iter::empty()), + Some(out_map) => out_map.keys(), } } @@ -197,7 +177,7 @@ impl Graph { /// # Returns /// /// The source `VertexId` of an `Edge` or an error if the edge is missing - pub fn src_vertex_id(&self, edge_id: EdgeId) -> Result<VertexId, GraphError> { + pub fn src_vertex_id(&self, edge_id: &EdgeId) -> Result<VertexId, NetworkError> { self.get_edge(edge_id).map(\|e\| e.src_vertex_id) } @@ -210,7 +190,7 @@ impl Graph { /// # Returns /// /// The destination `VertexId` of an `Edge` or an error if the edge is missing - pub fn dst_vertex_id(&self, edge_id: EdgeId) -> Result<VertexId, GraphError> { + pub fn dst_vertex_id(&self, edge_id: &EdgeId) -> Result<VertexId, NetworkError> { self.get_edge(edge_id).map(\|e\| e.dst_vertex_id) } @@ -225,17 +205,35 @@ impl Graph { /// # Returns /// /// The incident `EdgeId`s or an error if the vertex is not connected. - pub fn incident_edges( - &self, - vertex_id: VertexId, - direction: Direction, - ) -> Result<Vec<EdgeId>, GraphError> { + pub fn incident_edges(&self, vertex_id: &VertexId, direction: &Direction) -> Vec<EdgeId> { match direction { Direction::Forward => self.out_edges(vertex_id), Direction::Reverse => self.in_edges(vertex_id), } } + /// helper function to give incident edges to a vertex based on a + /// traversal direction. + /// + /// # Arguments + /// + /// `vertex_id` - vertex to find edges which connect to it + /// * `direction` - whether to find out edges (Forward) or in edges (Reverse) + /// + /// # Returns + /// + /// The incident `EdgeId`s or an error if the vertex is not connected. + pub fn incident_edges_iter<'a>( + &'a self, + vertex_id: &VertexId, + direction: &Direction, + ) -> Box<dyn Iterator<Item = &'a EdgeId> + 'a> { + match direction { + Direction::Forward => self.out_edges_iter(vertex_id), + Direction::Reverse => self.in_edges_iter(vertex_id), + } + } + /// helper function to give the incident vertex to an edge based on a /// traversal direction. /// @@ -249,9 +247,9 @@ impl Graph { /// The incident `VertexId` of an edge or an error if the edge is missing pub fn incident_vertex( &self, - edge_id: EdgeId, - direction: Direction, - ) -> Result<VertexId, GraphError> { + edge_id: &EdgeId, + direction: &Direction, + ) -> Result<VertexId, NetworkError> { match direction { Direction::Forward => self.dst_vertex_id(edge_id), Direction::Reverse => self.src_vertex_id(edge_id), @@ -268,13 +266,13 @@ impl Graph { /// /// The triplet of attributes surrounding one `Edge` or an error if /// any id is invalid. - pub fn edge_triplet_attrs( + pub fn edge_triplet( &self, - edge_id: EdgeId, - ) -> Result<(&Vertex, &Edge, &Vertex), GraphError> { + edge_id: &EdgeId, + ) -> Result<(&Vertex, &Edge, &Vertex), NetworkError> { let edge = self.get_edge(edge_id)?; - let src = self.get_vertex(edge.src_vertex_id)?; - let dst = self.get_vertex(edge.dst_vertex_id)?; + let src = self.get_vertex(&edge.src_vertex_id)?; + let dst = self.get_vertex(&edge.dst_vertex_id)?; Ok((src, edge, dst)) } @@ -294,16 +292,15 @@ impl Graph { /// /// For each edge connected to this `vertex_id` traversable via the provided `direction`, /// a triplet of the `EdgeId` and it's connecting `VertexId`s. - pub fn incident_triplets( + pub fn incident_triplet_ids( &self, - vertex_id: VertexId, - direction: Direction, - ) -> Result<Vec<(VertexId, EdgeId, VertexId)>, GraphError> { - self.incident_edges(vertex_id, direction)? - .into_iter() + vertex_id: &VertexId, + direction: &Direction, + ) -> Result<Vec<(VertexId, EdgeId, VertexId)>, NetworkError> { + self.incident_edges_iter(vertex_id, direction) .map(\|edge_id\| { let terminal_vid = self.incident_vertex(edge_id, direction)?; - Ok((vertex_id, edge_id, terminal_vid)) + Ok((vertex_id, edge_id, terminal_vid)) }) .collect() } @@ -325,15 +322,15 @@ impl Graph { /// a triplet of the `Edge` and it's connecting `Vertex`s. pub fn incident_triplet_attributes( &self, - vertex_id: VertexId, - direction: Direction, - ) -> Result<Vec<(&Vertex, &Edge, &Vertex)>, GraphError> { - self.incident_triplets(vertex_id, direction)? + vertex_id: &VertexId, + direction: &Direction, + ) -> Result<Vec<(&Vertex, &Edge, &Vertex)>, NetworkError> { + self.incident_triplet_ids(vertex_id, direction)? .iter() .map(\|(src_id, edge_id, dst_id)\| { - let src = self.get_vertex(src_id)?; - let edge = self.get_edge(edge_id)?; - let dst = self.get_vertex(dst_id)?; + let src = self.get_vertex(src_id)?; + let edge = self.get_edge(edge_id)?; + let dst = self.get_vertex(dst_id)?; Ok((src, edge, dst)) }) .collect() diff --git a/rust/routee-compass-core/src/model/road_network/graph_loader.rs b/rust/routee-compass-core/src/model/network/graph_loader.rs similarity index 76% rename from rust/routee-compass-core/src/model/road_network/graph_loader.rs rename to rust/routee-compass-core/src/model/network/graph_loader.rs index bdbd9941..a7f44211 100644 --- a/rust/routee-compass-core/src/model/road_network/graph_loader.rs +++ b/rust/routee-compass-core/src/model/network/graph_loader.rs @@ -1,15 +1,11 @@ -use std::path::Path; - -use log::warn; - -use crate::{model::property::vertex::Vertex, util::fs::fs_utils::line_count}; - use super::{ edge_loader::{EdgeLoader, EdgeLoaderConfig}, - graph::Graph, - graph_error::GraphError, vertex_loader::VertexLoaderConfig, + Graph, NetworkError, Vertex, }; +use crate::util::fs::fs_utils::line_count; +use log::warn; +use std::path::Path; pub fn graph_from_files<P: AsRef<Path>>( edge_list_csv: &P, @@ -17,7 +13,7 @@ pub fn graph_from_files<P: AsRef<Path>>( n_edges: Option<usize>, n_vertices: Option<usize>, verbose: Option<bool>, -) -> Result<Graph, GraphError> { +) -> Result<Graph, NetworkError> { let verbose = verbose.unwrap_or(false); let n_edges = match n_edges { Some(n) => n, @@ -63,7 +59,7 @@ pub fn graph_from_files<P: AsRef<Path>>( Ok(graph) } -fn get_n_edges<P: AsRef<Path>>(edge_list_csv: &P) -> Result<usize, GraphError> { +fn get_n_edges<P: AsRef<Path>>(edge_list_csv: &P) -> Result<usize, NetworkError> { // check if the extension is .gz let is_gzip = edge_list_csv .as_ref() @@ -73,14 +69,16 @@ fn get_n_edges<P: AsRef<Path>>(edge_list_csv: &P) -> Result<usize, GraphError> { .unwrap_or(false); let n = line_count(edge_list_csv, is_gzip)?; if n < 1 { - return Err(GraphError::EmptyFileSource { - filename: edge_list_csv.as_ref().to_path_buf(), - }); + let msg = match edge_list_csv.as_ref().to_str() { + Some(path_str) => format!("edge file {} is empty", path_str), + None => String::from("edge file is empty"), + }; + return Err(NetworkError::DatasetError(msg)); } Ok(n - 1) // drop count of header line } -fn get_n_vertices<P: AsRef<Path>>(vertex_list_csv: &P) -> Result<usize, GraphError> { +fn get_n_vertices<P: AsRef<Path>>(vertex_list_csv: &P) -> Result<usize, NetworkError> { let is_gzip = vertex_list_csv .as_ref() .to_path_buf() @@ -89,9 +87,11 @@ fn get_n_vertices<P: AsRef<Path>>(vertex_list_csv: &P) -> Result<usize, GraphErr .unwrap_or(false); let n = line_count(vertex_list_csv, is_gzip)?; if n < 1 { - return Err(GraphError::EmptyFileSource { - filename: vertex_list_csv.as_ref().to_path_buf(), - }); + let msg = match vertex_list_csv.as_ref().to_str() { + Some(path_str) => format!("vertex file {} is empty", path_str), + None => String::from("vertex file is empty"), + }; + return Err(NetworkError::DatasetError(msg)); } Ok(n - 1) // drop count of header line } diff --git a/rust/routee-compass-core/src/model/network/mod.rs b/rust/routee-compass-core/src/model/network/mod.rs new file mode 100644 index 00000000..82e63b5d --- /dev/null +++ b/rust/routee-compass-core/src/model/network/mod.rs @@ -0,0 +1,16 @@ +pub mod edge; +pub mod edge_id; +pub mod edge_loader; +pub mod graph; +pub mod graph_loader; +pub mod network_error; +pub mod vertex; +pub mod vertex_id; +pub mod vertex_loader; + +pub use edge::Edge; +pub use edge_id::EdgeId; +pub use graph::Graph; +pub use network_error::NetworkError; +pub use vertex::Vertex; +pub use vertex_id::VertexId; diff --git a/rust/routee-compass-core/src/model/network/network_error.rs b/rust/routee-compass-core/src/model/network/network_error.rs new file mode 100644 index 00000000..96c4ccae --- /dev/null +++ b/rust/routee-compass-core/src/model/network/network_error.rs @@ -0,0 +1,25 @@ +use crate::model::network::{edge_id::EdgeId, vertex_id::VertexId}; + +#[derive(thiserror::Error, Debug)] +pub enum NetworkError { + #[error("edge attribute not found for edge {0}")] + EdgeNotFound(EdgeId), + #[error("vertex attribute not found for vertex {0}")] + VertexNotFound(VertexId), + #[error("Error with graph attribute {0}: {1}")] + AttributeError(String, String), + #[error("error with provided dataset: {0}")] + DatasetError(String), + #[error("failure reading graph data from file: {source}")] + IOError { + #[from] + source: std::io::Error, + }, + #[error("failure reading graph data from CSV: {source}")] + CsvError { + #[from] + source: csv::Error, + }, + #[error("{0}")] + InternalError(String), +} diff --git a/rust/routee-compass-core/src/model/property/vertex.rs b/rust/routee-compass-core/src/model/network/vertex.rs similarity index 94% rename from rust/routee-compass-core/src/model/property/vertex.rs rename to rust/routee-compass-core/src/model/network/vertex.rs index f0d8138a..8c711453 100644 --- a/rust/routee-compass-core/src/model/property/vertex.rs +++ b/rust/routee-compass-core/src/model/network/vertex.rs @@ -1,4 +1,6 @@ -use crate::{model::road_network::vertex_id::VertexId, util::geo::coord::InternalCoord}; +use std::fmt::Display; + +use crate::{model::network::vertex_id::VertexId, util::geo::coord::InternalCoord}; use allocative::Allocative; use geo::coord; use serde::de; @@ -32,6 +34,12 @@ impl Vertex { } } +impl Display for Vertex { + fn fmt(&self, f: &mut std::fmt::Formatter<'_>) -> std::fmt::Result { + write!(f, "Vertex {} ({},{})", self.vertex_id, self.x(), self.y()) + } +} + const VERTEX_ID: &str = "vertex_id"; const X_COORDINATE: &str = "x"; const Y_COORDINATE: &str = "y"; @@ -132,7 +140,7 @@ mod tests { use geo::Coord; use crate::{ - model::{property::vertex::Vertex, road_network::vertex_id::VertexId}, + model::network::{Vertex, VertexId}, util::geo::coord::InternalCoord, }; @@ -141,7 +149,7 @@ mod tests { let filepath = PathBuf::from(env!("CARGO_MANIFEST_DIR")) .join("src") .join("model") - .join("property") + .join("network") .join("test") .join("vertices.csv"); diff --git a/rust/routee-compass-core/src/model/road_network/vertex_id.rs b/rust/routee-compass-core/src/model/network/vertex_id.rs similarity index 100% rename from rust/routee-compass-core/src/model/road_network/vertex_id.rs rename to rust/routee-compass-core/src/model/network/vertex_id.rs diff --git a/rust/routee-compass-core/src/model/road_network/vertex_loader.rs b/rust/routee-compass-core/src/model/network/vertex_loader.rs similarity index 75% rename from rust/routee-compass-core/src/model/road_network/vertex_loader.rs rename to rust/routee-compass-core/src/model/network/vertex_loader.rs index 9b2aa55d..c97dc0a5 100644 --- a/rust/routee-compass-core/src/model/road_network/vertex_loader.rs +++ b/rust/routee-compass-core/src/model/network/vertex_loader.rs @@ -1,7 +1,6 @@ use std::path::PathBuf; -use crate::model::property::vertex::Vertex; -use crate::model::road_network::graph_error::GraphError; +use super::{NetworkError, Vertex}; use crate::util::fs::read_utils; use kdam::{Bar, BarExt}; @@ -11,7 +10,7 @@ pub struct VertexLoaderConfig { } impl TryFrom<VertexLoaderConfig> for Box<[Vertex]> { - type Error = GraphError; + type Error = NetworkError; fn try_from(conf: VertexLoaderConfig) -> Result<Self, Self::Error> { let mut processed: usize = 0; @@ -20,7 +19,9 @@ impl TryFrom<VertexLoaderConfig> for Box<[Vertex]> { .animation("fillup") .desc("vertex list") .build() - .map_err(\|e\| GraphError::ProgressBarBuildError(String::from("vertex list"), e))?; + .map_err(\|e\| { + NetworkError::InternalError(format!("could not build progress bar: {}", e)) + })?; let cb = Box::new(\|_v: &Vertex\| { let _ = pb.update(1); @@ -28,7 +29,7 @@ impl TryFrom<VertexLoaderConfig> for Box<[Vertex]> { }); let result: Box<[Vertex]> = read_utils::from_csv(&conf.vertex_list_csv, true, Some(cb))?; - println!(); + eprintln!(); Ok(result) } } diff --git a/rust/routee-compass-core/src/model/property/mod.rs b/rust/routee-compass-core/src/model/property/mod.rs deleted file mode 100644 index dbda5dc7..00000000 --- a/rust/routee-compass-core/src/model/property/mod.rs +++ /dev/null @@ -1,2 +0,0 @@ -pub mod edge; -pub mod vertex; diff --git a/rust/routee-compass-core/src/model/road_network/graph_error.rs b/rust/routee-compass-core/src/model/road_network/graph_error.rs deleted file mode 100644 index bc89eb51..00000000 --- a/rust/routee-compass-core/src/model/road_network/graph_error.rs +++ /dev/null @@ -1,39 +0,0 @@ -use std::path::PathBuf; - -use crate::model::road_network::{edge_id::EdgeId, vertex_id::VertexId}; - -#[derive(thiserror::Error, Debug)] -pub enum GraphError { - #[error("edge {edge_id} not found")] - EdgeIdNotFound { edge_id: EdgeId }, - #[error("edge attribute not found for edge {edge_id}")] - EdgeAttributeNotFound { edge_id: EdgeId }, - #[error("vertex {vertex_id} not found")] - VertexIdNotFound { vertex_id: VertexId }, - #[error("vertex attribute not found for vertex {vertex_id}")] - VertexAttributeNotFound { vertex_id: VertexId }, - #[error("vertex without out edges in graph")] - VertexWithoutOutEdges { vertex_id: VertexId }, - #[error("vertex without in edges in graph")] - VertexWithoutInEdges { vertex_id: VertexId }, - #[error("error in test setup")] - TestError, - #[error("Error with graph attribute {0}: {1}")] - AttributeError(String, String), - #[error("{filename} file source was empty")] - EmptyFileSource { filename: PathBuf }, - #[error("failure reading TomTom graph: {source}")] - IOError { - #[from] - source: std::io::Error, - }, - #[error("csv error: {source}")] - CsvError { - #[from] - source: csv::Error, - }, - #[error("internal error: adjacency list missing vertex {0}")] - AdjacencyVertexMissing(VertexId), - #[error("error creating progress bar for {0}: {1}")] - ProgressBarBuildError(String, String), -} diff --git a/rust/routee-compass-core/src/model/road_network/mod.rs b/rust/routee-compass-core/src/model/road_network/mod.rs deleted file mode 100644 index 2e41655a..00000000 --- a/rust/routee-compass-core/src/model/road_network/mod.rs +++ /dev/null @@ -1,7 +0,0 @@ -pub mod edge_id; -pub mod edge_loader; -pub mod graph; -pub mod graph_error; -pub mod graph_loader; -pub mod vertex_id; -pub mod vertex_loader; diff --git a/rust/routee-compass-core/src/model/state/custom_feature_format.rs b/rust/routee-compass-core/src/model/state/custom_feature_format.rs index 2243a3e5..e347d0ac 100644 --- a/rust/routee-compass-core/src/model/state/custom_feature_format.rs +++ b/rust/routee-compass-core/src/model/state/custom_feature_format.rs @@ -1,6 +1,6 @@ use std::fmt::Display; -use super::{state_error::StateError, unit_codec_name::UnitCodecType}; +use super::{state_model_error::StateModelError, unit_codec_name::UnitCodecType}; use crate::model::traversal::state::state_variable::StateVar; use ordered_float::OrderedFloat; use serde::{Deserialize, Serialize}; @@ -63,7 +63,7 @@ impl CustomFeatureFormat { } } - pub fn initial(&self) -> Result<StateVar, StateError> { + pub fn initial(&self) -> Result<StateVar, StateModelError> { match self { CustomFeatureFormat::FloatingPoint { initial } => self.encode_f64(initial), CustomFeatureFormat::SignedInteger { initial } => self.encode_i64(initial), @@ -72,37 +72,37 @@ impl CustomFeatureFormat { } } - pub fn encode_f64(&self, value: &f64) -> Result<StateVar, StateError> { + pub fn encode_f64(&self, value: &f64) -> Result<StateVar, StateModelError> { match self { CustomFeatureFormat::FloatingPoint { initial: _ } => Ok(StateVar(value)), - _ => Err(StateError::EncodeError( + _ => Err(StateModelError::EncodeError( UnitCodecType::FloatingPoint.to_string(), self.name(), )), } } - pub fn encode_i64(&self, value: &i64) -> Result<StateVar, StateError> { + pub fn encode_i64(&self, value: &i64) -> Result<StateVar, StateModelError> { match self { CustomFeatureFormat::SignedInteger { initial: _ } => Ok(StateVar(value as f64)), - _ => Err(StateError::EncodeError( + _ => Err(StateModelError::EncodeError( UnitCodecType::SignedInteger.to_string(), self.name(), )), } } - pub fn encode_u64(&self, value: &u64) -> Result<StateVar, StateError> { + pub fn encode_u64(&self, value: &u64) -> Result<StateVar, StateModelError> { match self { CustomFeatureFormat::UnsignedInteger { initial: _ } => Ok(StateVar(value as f64)), - _ => Err(StateError::EncodeError( + _ => Err(StateModelError::EncodeError( UnitCodecType::UnsignedInteger.to_string(), self.name(), )), } } - pub fn encode_bool(&self, value: &bool) -> Result<StateVar, StateError> { + pub fn encode_bool(&self, value: &bool) -> Result<StateVar, StateModelError> { match self { CustomFeatureFormat::Boolean { initial: _ } => { if value { @@ -111,38 +111,38 @@ impl CustomFeatureFormat { Ok(StateVar(0.0)) } } - _ => Err(StateError::EncodeError( + _ => Err(StateModelError::EncodeError( UnitCodecType::Boolean.to_string(), self.name(), )), } } - pub fn decode_f64(&self, value: &StateVar) -> Result<f64, StateError> { + pub fn decode_f64(&self, value: &StateVar) -> Result<f64, StateModelError> { match self { CustomFeatureFormat::FloatingPoint { initial: _ } => Ok(value.0), - _ => Err(StateError::DecodeError( + _ => Err(StateModelError::DecodeError( value, UnitCodecType::FloatingPoint.to_string(), self.name(), )), } } - pub fn decode_i64(&self, value: &StateVar) -> Result<i64, StateError> { + pub fn decode_i64(&self, value: &StateVar) -> Result<i64, StateModelError> { match self { CustomFeatureFormat::SignedInteger { initial: _ } => Ok(value.0 as i64), - _ => Err(StateError::DecodeError( + _ => Err(StateModelError::DecodeError( value, UnitCodecType::SignedInteger.to_string(), self.name(), )), } } - pub fn decode_u64(&self, value: &StateVar) -> Result<u64, StateError> { + pub fn decode_u64(&self, value: &StateVar) -> Result<u64, StateModelError> { match self { CustomFeatureFormat::UnsignedInteger { initial: _ } => { if value < &StateVar::ZERO { - Err(StateError::ValueError( + Err(StateModelError::ValueError( value, UnitCodecType::UnsignedInteger.to_string(), )) @@ -150,14 +150,14 @@ impl CustomFeatureFormat { Ok(value.0 as u64) } } - _ => Err(StateError::DecodeError( + _ => Err(StateModelError::DecodeError( value, UnitCodecType::UnsignedInteger.to_string(), self.name(), )), } } - pub fn decode_bool(&self, value: &StateVar) -> Result<bool, StateError> { + pub fn decode_bool(&self, value: &StateVar) -> Result<bool, StateModelError> { match self { CustomFeatureFormat::Boolean { initial: _ } => { let is_false = value.0 == 0.0; @@ -167,7 +167,7 @@ impl CustomFeatureFormat { Ok(true) } } - _ => Err(StateError::DecodeError( + _ => Err(StateModelError::DecodeError( value, UnitCodecType::Boolean.to_string(), self.name(), diff --git a/rust/routee-compass-core/src/model/state/mod.rs b/rust/routee-compass-core/src/model/state/mod.rs index 8514bab2..e4ef7d2e 100644 --- a/rust/routee-compass-core/src/model/state/mod.rs +++ b/rust/routee-compass-core/src/model/state/mod.rs @@ -1,7 +1,7 @@ pub mod custom_feature_format; pub mod indexed_state_feature; -pub mod state_error; pub mod state_feature; pub mod state_model; +pub mod state_model_error; pub mod unit_codec_name; pub mod update_operation; diff --git a/rust/routee-compass-core/src/model/state/state_feature.rs b/rust/routee-compass-core/src/model/state/state_feature.rs index dc3f5edc..dbda3cc1 100644 --- a/rust/routee-compass-core/src/model/state/state_feature.rs +++ b/rust/routee-compass-core/src/model/state/state_feature.rs @@ -1,6 +1,6 @@ use std::fmt::Display; -use super::{custom_feature_format::CustomFeatureFormat, state_error::StateError}; +use super::{custom_feature_format::CustomFeatureFormat, state_model_error::StateModelError}; use crate::model::{traversal::state::state_variable::StateVar, unit}; use serde::{Deserialize, Serialize}; @@ -195,7 +195,7 @@ impl StateFeature { } } - pub fn get_initial(&self) -> Result<StateVar, StateError> { + pub fn get_initial(&self) -> Result<StateVar, StateModelError> { match self { StateFeature::Distance { distance_unit: _, @@ -217,53 +217,53 @@ impl StateFeature { } } - pub fn get_distance_unit(&self) -> Result<unit::DistanceUnit, StateError> { + pub fn get_distance_unit(&self) -> Result<unit::DistanceUnit, StateModelError> { match self { StateFeature::Distance { distance_unit: unit, initial: _, } => Ok(unit), - _ => Err(StateError::UnexpectedFeatureUnit( + _ => Err(StateModelError::UnexpectedFeatureUnit( String::from("distance"), self.get_feature_type(), )), } } - pub fn get_time_unit(&self) -> Result<unit::TimeUnit, StateError> { + pub fn get_time_unit(&self) -> Result<unit::TimeUnit, StateModelError> { match self { StateFeature::Time { time_unit: unit, initial: _, } => Ok(unit), - _ => Err(StateError::UnexpectedFeatureUnit( + _ => Err(StateModelError::UnexpectedFeatureUnit( String::from("time"), self.get_feature_type(), )), } } - pub fn get_energy_unit(&self) -> Result<unit::EnergyUnit, StateError> { + pub fn get_energy_unit(&self) -> Result<unit::EnergyUnit, StateModelError> { match self { StateFeature::Energy { energy_unit, initial: _, } => Ok(energy_unit), - _ => Err(StateError::UnexpectedFeatureUnit( + _ => Err(StateModelError::UnexpectedFeatureUnit( String::from("energy"), self.get_feature_type(), )), } } - pub fn get_custom_feature_format(&self) -> Result<&CustomFeatureFormat, StateError> { + pub fn get_custom_feature_format(&self) -> Result<&CustomFeatureFormat, StateModelError> { match self { StateFeature::Custom { r#type: _, unit: _, format, } => Ok(format), - _ => Err(StateError::UnexpectedFeatureUnit( + _ => Err(StateModelError::UnexpectedFeatureUnit( self.get_feature_unit_name(), self.get_feature_type(), )), diff --git a/rust/routee-compass-core/src/model/state/state_model.rs b/rust/routee-compass-core/src/model/state/state_model.rs index cb8a9f0e..93ee9d66 100644 --- a/rust/routee-compass-core/src/model/state/state_model.rs +++ b/rust/routee-compass-core/src/model/state/state_model.rs @@ -1,6 +1,6 @@ use super::{ - custom_feature_format::CustomFeatureFormat, state_error::StateError, - state_feature::StateFeature, update_operation::UpdateOperation, + custom_feature_format::CustomFeatureFormat, state_feature::StateFeature, + state_model_error::StateModelError, update_operation::UpdateOperation, }; use crate::util::compact_ordered_hash_map::CompactOrderedHashMap; use crate::{ @@ -44,7 +44,10 @@ impl StateModel { /// /// # Arguments /// `query` - JSON search query contents containing state model information - pub fn extend(&self, entries: Vec<(String, StateFeature)>) -> Result<StateModel, StateError> { + pub fn extend( + &self, + entries: Vec<(String, StateFeature)>, + ) -> Result<StateModel, StateModelError> { let mut map = self .0 .iter() @@ -64,7 +67,7 @@ impl StateModel { .iter() .map(\|(k, old, new)\| format!("{} old: {} \| new: {}", k, old, new)) .join(", "); - Err(StateError::BuildError(format!( + Err(StateModelError::BuildError(format!( "new state features overwriting existing: {}", msg ))) @@ -105,7 +108,7 @@ impl StateModel { /// # Returns /// /// an initialized, "zero"-valued traversal state, or an error - pub fn initial_state(&self) -> Result<Vec<StateVar>, StateError> { + pub fn initial_state(&self) -> Result<Vec<StateVar>, StateModelError> { self.0 .iter() .map(\|(_, feature)\| { @@ -130,7 +133,7 @@ impl StateModel { state: &[StateVar], name: &String, unit: &DistanceUnit, - ) -> Result<Distance, StateError> { + ) -> Result<Distance, StateModelError> { let value = self.get_state_variable(state, name)?; let feature = self.get_feature(name)?; let result = feature.get_distance_unit()?.convert(&value.into(), unit); @@ -151,7 +154,7 @@ impl StateModel { state: &[StateVar], name: &String, unit: &TimeUnit, - ) -> Result<Time, StateError> { + ) -> Result<Time, StateModelError> { let value = self.get_state_variable(state, name)?; let feature = self.get_feature(name)?; let result = feature.get_time_unit()?.convert(&value.into(), unit); @@ -172,7 +175,7 @@ impl StateModel { state: &[StateVar], name: &String, unit: &EnergyUnit, - ) -> Result<Energy, StateError> { + ) -> Result<Energy, StateModelError> { let value = self.get_state_variable(state, name)?; let feature = self.get_feature(name)?; let result = feature.get_energy_unit()?.convert(&value.into(), unit); @@ -187,7 +190,11 @@ impl StateModel { /// # Returns /// /// the expected value or an error - pub fn get_custom_f64(&self, state: &[StateVar], name: &String) -> Result<f64, StateError> { + pub fn get_custom_f64( + &self, + state: &[StateVar], + name: &String, + ) -> Result<f64, StateModelError> { let (value, format) = self.get_custom_state_variable(state, name)?; let result = format.decode_f64(&value)?; Ok(result) @@ -201,7 +208,11 @@ impl StateModel { /// # Returns /// /// the expected value or an error - pub fn get_custom_i64(&self, state: &[StateVar], name: &String) -> Result<i64, StateError> { + pub fn get_custom_i64( + &self, + state: &[StateVar], + name: &String, + ) -> Result<i64, StateModelError> { let (value, format) = self.get_custom_state_variable(state, name)?; let result = format.decode_i64(&value)?; Ok(result) @@ -215,7 +226,11 @@ impl StateModel { /// # Returns /// /// the expected value or an error - pub fn get_custom_u64(&self, state: &[StateVar], name: &String) -> Result<u64, StateError> { + pub fn get_custom_u64( + &self, + state: &[StateVar], + name: &String, + ) -> Result<u64, StateModelError> { let (value, format) = self.get_custom_state_variable(state, name)?; let result = format.decode_u64(&value)?; Ok(result) @@ -229,7 +244,11 @@ impl StateModel { /// # Returns /// /// the expected value or an error - pub fn get_custom_bool(&self, state: &[StateVar], name: &String) -> Result<bool, StateError> { + pub fn get_custom_bool( + &self, + state: &[StateVar], + name: &String, + ) -> Result<bool, StateModelError> { let (value, format) = self.get_custom_state_variable(state, name)?; let result = format.decode_bool(&value)?; Ok(result) @@ -250,7 +269,7 @@ impl StateModel { &self, state: &[StateVar], name: &String, - ) -> Result<(StateVar, &CustomFeatureFormat), StateError> { + ) -> Result<(StateVar, &CustomFeatureFormat), StateModelError> { let value = self.get_state_variable(state, name)?; let feature = self.get_feature(name)?; let format = feature.get_custom_feature_format()?; @@ -273,7 +292,7 @@ impl StateModel { prev: &[StateVar], next: &[StateVar], name: &String, - ) -> Result<StateVar, StateError> { + ) -> Result<StateVar, StateModelError> { let prev_val = self.get_state_variable(prev, name)?; let next_val = self.get_state_variable(next, name)?; Ok(next_val - prev_val) @@ -286,7 +305,7 @@ impl StateModel { name: &String, distance: &Distance, from_unit: &DistanceUnit, - ) -> Result<(), StateError> { + ) -> Result<(), StateModelError> { let prev_distance = self.get_distance(state, name, from_unit)?; let next_distance = prev_distance + distance; self.set_distance(state, name, &next_distance, from_unit) @@ -299,7 +318,7 @@ impl StateModel { name: &String, time: &Time, from_unit: &TimeUnit, - ) -> Result<(), StateError> { + ) -> Result<(), StateModelError> { let prev_time = self.get_time(state, name, from_unit)?; let next_time = prev_time + time; self.set_time(state, name, &next_time, from_unit) @@ -312,7 +331,7 @@ impl StateModel { name: &String, energy: &Energy, from_unit: &EnergyUnit, - ) -> Result<(), StateError> { + ) -> Result<(), StateModelError> { let prev_energy = self.get_energy(state, name, from_unit)?; let next_energy = prev_energy + energy; self.set_energy(state, name, &next_energy, from_unit) @@ -324,7 +343,7 @@ impl StateModel { name: &String, distance: &Distance, from_unit: &DistanceUnit, - ) -> Result<(), StateError> { + ) -> Result<(), StateModelError> { let feature = self.get_feature(name)?; let to_unit = feature.get_distance_unit()?; let value = from_unit.convert(distance, &to_unit); @@ -337,7 +356,7 @@ impl StateModel { name: &String, time: &Time, from_unit: &TimeUnit, - ) -> Result<(), StateError> { + ) -> Result<(), StateModelError> { let feature = self.get_feature(name)?; let to_unit = feature.get_time_unit()?; let value = from_unit.convert(time, &to_unit); @@ -350,7 +369,7 @@ impl StateModel { name: &String, energy: &Energy, from_unit: &EnergyUnit, - ) -> Result<(), StateError> { + ) -> Result<(), StateModelError> { let feature = self.get_feature(name)?; let to_unit = feature.get_energy_unit()?; let value = from_unit.convert(energy, &to_unit); @@ -362,7 +381,7 @@ impl StateModel { state: &mut [StateVar], name: &String, value: &f64, - ) -> Result<(), StateError> { + ) -> Result<(), StateModelError> { let feature = self.get_feature(name)?; let format = feature.get_custom_feature_format()?; let encoded_value = format.encode_f64(value)?; @@ -374,7 +393,7 @@ impl StateModel { state: &mut [StateVar], name: &String, value: &i64, - ) -> Result<(), StateError> { + ) -> Result<(), StateModelError> { let feature = self.get_feature(name)?; let format = feature.get_custom_feature_format()?; let encoded_value = format.encode_i64(value)?; @@ -386,7 +405,7 @@ impl StateModel { state: &mut [StateVar], name: &String, value: &u64, - ) -> Result<(), StateError> { + ) -> Result<(), StateModelError> { let feature = self.get_feature(name)?; let format = feature.get_custom_feature_format()?; let encoded_value = format.encode_u64(value)?; @@ -398,7 +417,7 @@ impl StateModel { state: &mut [StateVar], name: &String, value: &bool, - ) -> Result<(), StateError> { + ) -> Result<(), StateModelError> { let feature = self.get_feature(name)?; let format = feature.get_custom_feature_format()?; let encoded_value = format.encode_bool(value)?; @@ -443,9 +462,9 @@ impl StateModel { self.0.iter().map(\|(k, _)\| k.clone()).join(",") } - fn get_feature(&self, feature_name: &String) -> Result<&StateFeature, StateError> { + fn get_feature(&self, feature_name: &String) -> Result<&StateFeature, StateModelError> { self.0.get(feature_name).ok_or_else(\|\| { - StateError::UnknownStateVariableName(feature_name.clone(), self.get_names()) + StateModelError::UnknownStateVariableName(feature_name.clone(), self.get_names()) }) } @@ -454,13 +473,12 @@ impl StateModel { &self, state: &[StateVar], name: &String, - ) -> Result<StateVar, StateError> { - let idx = self - .0 - .get_index(name) - .ok_or_else(\|\| StateError::UnknownStateVariableName(name.clone(), self.get_names()))?; + ) -> Result<StateVar, StateModelError> { + let idx = self.0.get_index(name).ok_or_else(\|\| { + StateModelError::UnknownStateVariableName(name.clone(), self.get_names()) + })?; let value = state.get(idx).ok_or_else(\|\| { - StateError::RuntimeError(format!( + StateModelError::RuntimeError(format!( "state index {} for {} is out of range for state vector with {} entries", idx, name, @@ -476,14 +494,16 @@ impl StateModel { name: &String, value: &StateVar, op: UpdateOperation, - ) -> Result<(), StateError> { - let index = self - .0 - .get_index(name) - .ok_or_else(\|\| StateError::UnknownStateVariableName(name.clone(), self.get_names()))?; + ) -> Result<(), StateModelError> { + let index = self.0.get_index(name).ok_or_else(\|\| { + StateModelError::UnknownStateVariableName(name.clone(), self.get_names()) + })?; let prev = state .get(index) - .ok_or(StateError::InvalidStateVariableIndex(index, state.len()))?; + .ok_or(StateModelError::InvalidStateVariableIndex( + index, + state.len(), + ))?; let updated = op.perform_operation(prev, value); state[index] = updated; Ok(()) @@ -491,7 +511,7 @@ impl StateModel { } impl<'a> TryFrom<&'a serde_json::Value> for StateModel { - type Error = StateError; + type Error = StateModelError; /// builds a new state model from a JSON array of deserialized StateFeatures. /// the size of the JSON object matches the size of the feature vector. downstream @@ -526,11 +546,11 @@ impl<'a> TryFrom<&'a serde_json::Value> for StateModel { /// } /// } /// ``` - fn try_from(json: &'a serde_json::Value) -> Result<StateModel, StateError> { + fn try_from(json: &'a serde_json::Value) -> Result<StateModel, StateModelError> { let tuples = json .as_object() .ok_or_else(\|\| { - StateError::BuildError(String::from( + StateModelError::BuildError(String::from( "expected state model configuration to be a JSON object {}", )) })? @@ -538,7 +558,7 @@ impl<'a> TryFrom<&'a serde_json::Value> for StateModel { .map(\|(feature_name, feature_json)\| { let feature = serde_json::from_value::<StateFeature>(feature_json.clone()) .map_err(\|e\| { - StateError::BuildError(format!( + StateModelError::BuildError(format!( "unable to parse state feature row with name '{}' contents '{}' due to: {}", feature_name.clone(), feature_json.clone(), diff --git a/rust/routee-compass-core/src/model/state/state_error.rs b/rust/routee-compass-core/src/model/state/state_model_error.rs similarity index 97% rename from rust/routee-compass-core/src/model/state/state_error.rs rename to rust/routee-compass-core/src/model/state/state_model_error.rs index 8f39c607..22b5b606 100644 --- a/rust/routee-compass-core/src/model/state/state_error.rs +++ b/rust/routee-compass-core/src/model/state/state_model_error.rs @@ -1,7 +1,7 @@ use crate::model::traversal::state::state_variable::StateVar; #[derive(thiserror::Error, Debug)] -pub enum StateError { +pub enum StateModelError { #[error("attempting to encode {0} as state variable when value is a {1}")] EncodeError(String, String), #[error("attempting to decode {0} as a {1} when codec expects a {2}")] diff --git a/rust/routee-compass-core/src/model/state/update_operation.rs b/rust/routee-compass-core/src/model/state/update_operation.rs index 07f7ddae..bcea3751 100644 --- a/rust/routee-compass-core/src/model/state/update_operation.rs +++ b/rust/routee-compass-core/src/model/state/update_operation.rs @@ -1,8 +1,8 @@ use crate::model::traversal::state::state_variable::StateVar; - -/// describes an arbitrary state update operation. // pub(crate) type GenericStateUpdateOp = Box<dyn Fn(&StateVar, &StateVar) -> StateVar>; +/// describes an arbitrary state update operation. +/// /// represents the type of arithmetic operation used to update a state variable. /// the specific index of a state variable is hidden via the StateModel, which /// makes life harder, but protects against all sorts of indexing errors. diff --git a/rust/routee-compass-core/src/model/traversal/default/distance_traversal_model.rs b/rust/routee-compass-core/src/model/traversal/default/distance_traversal_model.rs index e347d1f6..d75e64f6 100644 --- a/rust/routee-compass-core/src/model/traversal/default/distance_traversal_model.rs +++ b/rust/routee-compass-core/src/model/traversal/default/distance_traversal_model.rs @@ -1,12 +1,12 @@ +use crate::model::network::{Edge, Vertex}; use crate::model::state::state_feature::StateFeature; use crate::model::state::state_model::StateModel; use crate::model::traversal::traversal_model::TraversalModel; +use crate::model::traversal::{ + state::state_variable::StateVar, traversal_model_error::TraversalModelError, +}; use crate::model::unit::DistanceUnit; use crate::model::unit::BASE_DISTANCE_UNIT; -use crate::model::{ - property::{edge::Edge, vertex::Vertex}, - traversal::{state::state_variable::StateVar, traversal_model_error::TraversalModelError}, -}; use crate::util::geo::haversine; /// A simple traversal model that uses the edge distance as the cost of traversal. @@ -49,7 +49,12 @@ impl TraversalModel for DistanceTraversalModel { let (src, dst) = od; let distance = haversine::coord_distance(&src.coordinate, &dst.coordinate, self.distance_unit) - .map_err(TraversalModelError::NumericError)?; + .map_err(\|e\| { + TraversalModelError::TraversalModelFailure(format!( + "could not compute haversine distance between {} and {}: {}", + src, dst, e + )) + })?; state_model.add_distance( state, &Self::DISTANCE.into(), diff --git a/rust/routee-compass-core/src/model/traversal/default/speed_traversal_engine.rs b/rust/routee-compass-core/src/model/traversal/default/speed_traversal_engine.rs index 93ef4363..8539c0db 100644 --- a/rust/routee-compass-core/src/model/traversal/default/speed_traversal_engine.rs +++ b/rust/routee-compass-core/src/model/traversal/default/speed_traversal_engine.rs @@ -25,10 +25,11 @@ impl SpeedTraversalEngine { let speed_table: Box<[Speed]> = read_utils::read_raw_file(speed_table_path, read_decoders::default, None).map_err( \|e\| { - TraversalModelError::FileReadError( - speed_table_path.as_ref().to_path_buf(), - e.to_string(), - ) + TraversalModelError::BuildError(format!( + "cannot read {} due to {}", + speed_table_path.as_ref().to_str().unwrap_or_default(), + e, + )) }, )?; let max_speed = get_max_speed(&speed_table)?; diff --git a/rust/routee-compass-core/src/model/traversal/default/speed_traversal_model.rs b/rust/routee-compass-core/src/model/traversal/default/speed_traversal_model.rs index 28d7ee20..706d3a9b 100644 --- a/rust/routee-compass-core/src/model/traversal/default/speed_traversal_model.rs +++ b/rust/routee-compass-core/src/model/traversal/default/speed_traversal_model.rs @@ -1,11 +1,11 @@ use super::speed_traversal_engine::SpeedTraversalEngine; -use crate::model::road_network::edge_id::EdgeId; +use crate::model::network::edge_id::EdgeId; +use crate::model::network::{Edge, Vertex}; use crate::model::state::state_feature::StateFeature; use crate::model::state::state_model::StateModel; use crate::model::traversal::traversal_model::TraversalModel; use crate::model::unit::{Distance, Time, BASE_DISTANCE_UNIT}; use crate::model::{ - property::{edge::Edge, vertex::Vertex}, traversal::{state::state_variable::StateVar, traversal_model_error::TraversalModelError}, unit::Speed, }; @@ -66,7 +66,12 @@ impl TraversalModel for SpeedTraversalModel { let (src, dst) = od; let distance = haversine::coord_distance(&src.coordinate, &dst.coordinate, self.engine.distance_unit) - .map_err(TraversalModelError::NumericError)?; + .map_err(\|e\| { + TraversalModelError::TraversalModelFailure(format!( + "could not compute haversine distance between {} and {}: {}", + src, dst, e + )) + })?; if distance == Distance::ZERO { return Ok(()); @@ -118,11 +123,10 @@ impl TraversalModel for SpeedTraversalModel { /// look up a speed from the speed table pub fn get_speed(speed_table: &[Speed], edge_id: EdgeId) -> Result<Speed, TraversalModelError> { let speed: &Speed = speed_table.get(edge_id.as_usize()).ok_or_else(\|\| { - TraversalModelError::MissingIdInTabularCostFunction( - format!("{}", edge_id), - String::from("EdgeId"), - String::from("speed table"), - ) + TraversalModelError::TraversalModelFailure(format!( + "could not find expected index {} in speed table", + edge_id + )) })?; Ok(speed) } @@ -130,11 +134,8 @@ pub fn get_speed(speed_table: &[Speed], edge_id: EdgeId) -> Result<Speed, Traver #[cfg(test)] mod tests { use super::; + use crate::model::network::{Edge, EdgeId, Vertex, VertexId}; use crate::model::unit::{Distance, DistanceUnit, SpeedUnit, TimeUnit}; - use crate::model::{ - property::{edge::Edge, vertex::Vertex}, - road_network::{edge_id::EdgeId, vertex_id::VertexId}, - }; use crate::util::geo::coord::InternalCoord; use geo::coord; use std::path::PathBuf; diff --git a/rust/routee-compass-core/src/model/traversal/traversal_model.rs b/rust/routee-compass-core/src/model/traversal/traversal_model.rs index 53547170..9fd889e4 100644 --- a/rust/routee-compass-core/src/model/traversal/traversal_model.rs +++ b/rust/routee-compass-core/src/model/traversal/traversal_model.rs @@ -1,6 +1,6 @@ use super::state::state_variable::StateVar; use super::traversal_model_error::TraversalModelError; -use crate::model::property::{edge::Edge, vertex::Vertex}; +use crate::model::network::{Edge, Vertex}; use crate::model::state::state_feature::StateFeature; use crate::model::state::state_model::StateModel; diff --git a/rust/routee-compass-core/src/model/traversal/traversal_model_error.rs b/rust/routee-compass-core/src/model/traversal/traversal_model_error.rs index ecba24af..8b5d393f 100644 --- a/rust/routee-compass-core/src/model/traversal/traversal_model_error.rs +++ b/rust/routee-compass-core/src/model/traversal/traversal_model_error.rs @@ -1,32 +1,34 @@ -use super::state::traversal_state::TraversalState; -use crate::model::road_network::graph_error::GraphError; -use crate::model::state::state_error::StateError; +use crate::model::network::network_error::NetworkError; +use crate::model::state::state_model_error::StateModelError; use crate::model::unit::UnitError; use crate::util::cache_policy::cache_error::CacheError; -use std::path::PathBuf; #[derive(thiserror::Error, Debug)] pub enum TraversalModelError { - #[error("failure building traversal model from file {0}: {1}")] - FileReadError(PathBuf, String), #[error("failure building traversal model: {0}")] BuildError(String), - #[error("numeric error during calculation: {0}")] - NumericError(String), - #[error("index {0} for {1} not found on search state {2:?}")] - StateVectorIndexOutOfBounds(usize, String, TraversalState), - #[error("id {0} for id type {1} not found in tabular edge cost function {2}")] - MissingIdInTabularCostFunction(String, String, String), + #[error("{0}")] + TraversalModelFailure(String), #[error("internal error: {0}")] InternalError(String), - #[error(transparent)] - TraversalUnitsError(#[from] UnitError), - #[error(transparent)] - CacheError(#[from] CacheError), - #[error(transparent)] - GraphError(#[from] GraphError), - #[error(transparent)] - StateError(#[from] StateError), - #[error("prediction model failed with error {0}")] - PredictionModel(String), + #[error("failure executing traversal model due to numeric units: {source}")] + UnitsFailure { + #[from] + source: UnitError, + }, + #[error("failure executing traversal model due to cache: {source}")] + CacheFailure { + #[from] + source: CacheError, + }, + #[error("failure executing traversal model due to network: {source}")] + NetworkFailure { + #[from] + source: NetworkError, + }, + #[error("failure executing traversal model due to search state: {source}")] + StateError { + #[from] + source: StateModelError, + }, } diff --git a/rust/routee-compass-core/src/util/fs/read_utils.rs b/rust/routee-compass-core/src/util/fs/read_utils.rs index 9a039eeb..39f0b386 100644 --- a/rust/routee-compass-core/src/util/fs/read_utils.rs +++ b/rust/routee-compass-core/src/util/fs/read_utils.rs @@ -34,13 +34,12 @@ where .trim(csv::Trim::Fields) .from_reader(r) .into_deserialize::<T>() - .map(move \|r\| { - if let Ok(t) = &r { + .inspect(move \|r\| { + if let Ok(t) = r { if let Some(cb) = &mut row_callback { cb(t); } } - r }); Ok(Box::new(reader)) diff --git a/rust/routee-compass-core/src/util/io_utils.rs b/rust/routee-compass-core/src/util/io_utils.rs index 347a2fa7..f42d4c51 100644 --- a/rust/routee-compass-core/src/util/io_utils.rs +++ b/rust/routee-compass-core/src/util/io_utils.rs @@ -26,7 +26,7 @@ pub fn with_progress_bar<R, E>( let result = closure(cb); - println!(); // create a newline once the progress bar is complete + eprintln!(); // create a newline once the progress bar is complete result } diff --git a/rust/routee-compass-core/src/util/multiset.rs b/rust/routee-compass-core/src/util/multiset.rs index fb2ecc6c..ec6a3d74 100644 --- a/rust/routee-compass-core/src/util/multiset.rs +++ b/rust/routee-compass-core/src/util/multiset.rs @@ -19,7 +19,7 @@ where } } -impl<'a, T> Iterator for MultiSet<'a, T> +impl<T> Iterator for MultiSet<'_, T> where T: Clone + Copy, { diff --git a/rust/routee-compass-macros/src/lib.rs b/rust/routee-compass-macros/src/lib.rs index defbc6f4..bd3b0b2b 100644 --- a/rust/routee-compass-macros/src/lib.rs +++ b/rust/routee-compass-macros/src/lib.rs @@ -50,21 +50,11 @@ pub fn pybindings(_attr: TokenStream, item: TokenStream) -> TokenStream { )) }) } - fn graph_get_out_edge_ids(&self, vertex_id: usize) -> PyResult<Vec<usize>> { - CompassAppBindings::graph_get_out_edge_ids(self, vertex_id).map_err(\|e\| { - PyException::new_err(format!( - "error retrieving out edge ids for vertex_id {}: {}", - vertex_id, e - )) - }) + fn graph_get_out_edge_ids(&self, vertex_id: usize) -> Vec<usize> { + CompassAppBindings::graph_get_out_edge_ids(self, vertex_id) } - fn graph_get_in_edge_ids(&self, vertex_id: usize) -> PyResult<Vec<usize>> { - CompassAppBindings::graph_get_in_edge_ids(self, vertex_id).map_err(\|e\| { - PyException::new_err(format!( - "error retrieving in edge ids for vertex_id {}: {}", - vertex_id, e - )) - }) + fn graph_get_in_edge_ids(&self, vertex_id: usize) -> Vec<usize> { + CompassAppBindings::graph_get_in_edge_ids(self, vertex_id) } #[staticmethod] pub fn _from_config_toml_string( diff --git a/rust/routee-compass-powertrain/src/routee/energy_model_ops.rs b/rust/routee-compass-powertrain/src/routee/energy_model_ops.rs index eb3238ce..236b3dc6 100644 --- a/rust/routee-compass-powertrain/src/routee/energy_model_ops.rs +++ b/rust/routee-compass-powertrain/src/routee/energy_model_ops.rs @@ -1,5 +1,5 @@ use routee_compass_core::model::{ - access::default::turn_delays::edge_heading::EdgeHeading, road_network::edge_id::EdgeId, + access::default::turn_delays::edge_heading::EdgeHeading, network::edge_id::EdgeId, traversal::traversal_model_error::TraversalModelError, unit::Grade, }; @@ -14,11 +14,10 @@ pub fn get_grade( None => Ok(Grade::ZERO), Some(gt) => { let grade: &Grade = gt.get(edge_id.as_usize()).ok_or_else(\|\| { - TraversalModelError::MissingIdInTabularCostFunction( - format!("{}", edge_id), - String::from("EdgeId"), - String::from("grade table"), - ) + TraversalModelError::TraversalModelFailure(format!( + "missing index {} from grade table", + edge_id + )) })?; Ok(grade) } @@ -31,11 +30,10 @@ pub fn get_headings( edge_id: EdgeId, ) -> Result<EdgeHeading, TraversalModelError> { let heading: &EdgeHeading = headings_table.get(edge_id.as_usize()).ok_or_else(\|\| { - TraversalModelError::MissingIdInTabularCostFunction( - format!("{}", edge_id), - String::from("EdgeId"), - String::from("headings table"), - ) + TraversalModelError::TraversalModelFailure(format!( + "missing index {} from headings table", + edge_id + )) })?; Ok(heading) } diff --git a/rust/routee-compass-powertrain/src/routee/energy_model_service.rs b/rust/routee-compass-powertrain/src/routee/energy_model_service.rs index 5153c113..faeac742 100644 --- a/rust/routee-compass-powertrain/src/routee/energy_model_service.rs +++ b/rust/routee-compass-powertrain/src/routee/energy_model_service.rs @@ -38,7 +38,11 @@ impl EnergyModelService { let grade_table: Arc<Option<Box<[Grade]>>> = match grade_table_path_option { Some(gtp) => Arc::new(Some( read_utils::read_raw_file(gtp, read_decoders::default, None).map_err(\|e\| { - TraversalModelError::FileReadError(gtp.as_ref().to_path_buf(), e.to_string()) + TraversalModelError::BuildError(format!( + "failure reading grade table {} due to {}", + gtp.as_ref().to_str().unwrap_or_default(), + e + )) })?, )), None => Arc::new(None), diff --git a/rust/routee-compass-powertrain/src/routee/energy_traversal_model.rs b/rust/routee-compass-powertrain/src/routee/energy_traversal_model.rs index 61743a58..7c2d400e 100644 --- a/rust/routee-compass-powertrain/src/routee/energy_traversal_model.rs +++ b/rust/routee-compass-powertrain/src/routee/energy_traversal_model.rs @@ -1,8 +1,7 @@ use super::energy_model_ops::get_grade; use super::energy_model_service::EnergyModelService; use super::vehicle::vehicle_type::VehicleType; -use routee_compass_core::model::property::edge::Edge; -use routee_compass_core::model::property::vertex::Vertex; +use routee_compass_core::model::network::{Edge, Vertex}; use routee_compass_core::model::state::state_feature::StateFeature; use routee_compass_core::model::state::state_model::StateModel; use routee_compass_core::model::traversal::state::state_variable::StateVar; @@ -92,7 +91,12 @@ impl TraversalModel for EnergyTraversalModel { &dst.coordinate, self.energy_model_service.distance_unit, ) - .map_err(TraversalModelError::NumericError)?; + .map_err(\|e\| { + TraversalModelError::TraversalModelFailure(format!( + "could not compute haversine distance between {} and {}: {}", + src, dst, e + )) + })?; if distance == Distance::ZERO { return Ok(()); @@ -162,8 +166,7 @@ mod tests { use geo::coord; use routee_compass_core::{ model::{ - property::{edge::Edge, vertex::Vertex}, - road_network::{edge_id::EdgeId, vertex_id::VertexId}, + network::{Edge, EdgeId, Vertex, VertexId}, traversal::default::{ speed_traversal_engine::SpeedTraversalEngine, speed_traversal_service::SpeedLookupService, diff --git a/rust/routee-compass-powertrain/src/routee/prediction/interpolation/interpolation_speed_grade_model.rs b/rust/routee-compass-powertrain/src/routee/prediction/interpolation/interpolation_speed_grade_model.rs index 279f17aa..52385cfb 100644 --- a/rust/routee-compass-powertrain/src/routee/prediction/interpolation/interpolation_speed_grade_model.rs +++ b/rust/routee-compass-powertrain/src/routee/prediction/interpolation/interpolation_speed_grade_model.rs @@ -36,28 +36,28 @@ impl PredictionModel for InterpolationSpeedGradeModel { let (min_speed, max_speed, min_grade, max_grade) = match &self.interpolator { interp::Interpolator::Interp2D(interp) => ( interp.x.first().ok_or_else(\|\| { - TraversalModelError::PredictionModel( - "Could not get first x-value; are x-values empty?".to_string(), + TraversalModelError::TraversalModelFailure( + "Could not get first x-value from powertrain interpolation result; are x-values empty?".to_string(), ) })?, interp.x.last().ok_or_else(\|\| { - TraversalModelError::PredictionModel( - "Could not get last x-value; are x-values empty?".to_string(), + TraversalModelError::TraversalModelFailure( + "Could not get last x-value from powertrain interpolation result; are x-values empty?".to_string(), ) })?, interp.y.first().ok_or_else(\|\| { - TraversalModelError::PredictionModel( - "Could not get first y-value; are y-values empty?".to_string(), + TraversalModelError::TraversalModelFailure( + "Could not get first y-value from powertrain interpolation result; are y-values empty?".to_string(), ) })?, interp.y.last().ok_or_else(\|\| { - TraversalModelError::PredictionModel( - "Could not get last y-value; are y-values empty?".to_string(), + TraversalModelError::TraversalModelFailure( + "Could not get last y-value from powertrain interpolation result; are y-values empty?".to_string(), ) })?, ), _ => { - return Err(TraversalModelError::PredictionModel( + return Err(TraversalModelError::TraversalModelFailure( "Only 2-D interpolators are currently supported".to_string(), )) } @@ -70,7 +70,10 @@ impl PredictionModel for InterpolationSpeedGradeModel { .interpolator .interpolate(&[speed_value, grade_value], &interp::Strategy::Linear) .map_err(\|e\| { - TraversalModelError::PredictionModel(format!("Failed to interpolate: {}", e)) + TraversalModelError::TraversalModelFailure(format!( + "Failed to interpolate speed/grade model output during prediction: {}", + e + )) })?; let energy_rate = EnergyRate::new(y); @@ -119,13 +122,11 @@ impl InterpolationSpeedGradeModel { for speed_value in speed_values.clone().into_iter() { let mut row: Vec<f64> = Vec::new(); for grade_value in grade_values.clone().into_iter() { - let (energy, _energy_unit) = model - .predict( - (Speed::new(speed_value), speed_unit), - (Grade::new(grade_value), grade_unit), - (distance, distance_unit), - ) - .map_err(\|e\| TraversalModelError::PredictionModel(e.to_string()))?; + let (energy, _energy_unit) = model.predict( + (Speed::new(speed_value), speed_unit), + (Grade::new(grade_value), grade_unit), + (distance, distance_unit), + )?; row.push(energy.as_f64()); } values.push(row); @@ -133,8 +134,8 @@ impl InterpolationSpeedGradeModel { let interpolator = interp::Interpolator::Interp2D( interp::Interp2D::new(speed_values, grade_values, values).map_err(\|e\| { - TraversalModelError::PredictionModel(format!( - "Failed to validate interpolation model: {}", + TraversalModelError::TraversalModelFailure(format!( + "Failed to build interpolation speed/grade model: {}", e )) })?, diff --git a/rust/routee-compass-powertrain/src/routee/prediction/prediction_model_ops.rs b/rust/routee-compass-powertrain/src/routee/prediction/prediction_model_ops.rs index 0aad686c..fe6467a7 100644 --- a/rust/routee-compass-powertrain/src/routee/prediction/prediction_model_ops.rs +++ b/rust/routee-compass-powertrain/src/routee/prediction/prediction_model_ops.rs @@ -113,7 +113,7 @@ pub fn find_min_energy_rate( (speed, SpeedUnit::MilesPerHour), (grade, GradeUnit::Percent), ) - .map_err(\|e\| TraversalModelError::PredictionModel(e.to_string()))?; + .map_err(\|e\| TraversalModelError::BuildError(format!("failure while executing grid search for minimum energy rate in prediction model: {}", e)))?; if energy_rate < minimum_energy_rate { minimum_energy_rate = energy_rate; } diff --git a/rust/routee-compass-powertrain/src/routee/prediction/smartcore/smartcore_speed_grade_model.rs b/rust/routee-compass-powertrain/src/routee/prediction/smartcore/smartcore_speed_grade_model.rs index 5d527f30..13296c69 100644 --- a/rust/routee-compass-powertrain/src/routee/prediction/smartcore/smartcore_speed_grade_model.rs +++ b/rust/routee-compass-powertrain/src/routee/prediction/smartcore/smartcore_speed_grade_model.rs @@ -27,10 +27,12 @@ impl PredictionModel for SmartcoreSpeedGradeModel { let speed_value = speed_unit.convert(&speed, &self.speed_unit).as_f64(); let grade_value = grade_unit.convert(&grade, &self.grade_unit).as_f64(); let x = DenseMatrix::from_2d_vec(&vec![vec![speed_value, grade_value]]); - let y = self - .rf - .predict(&x) - .map_err(\|e\| TraversalModelError::PredictionModel(e.to_string()))?; + let y = self.rf.predict(&x).map_err(\|e\| { + TraversalModelError::TraversalModelFailure(format!( + "failure running underlying Smartcore random forest energy prediction: {}", + e + )) + })?; let energy_rate = EnergyRate::new(y[0]); Ok((energy_rate, self.energy_rate_unit)) @@ -46,17 +48,19 @@ impl SmartcoreSpeedGradeModel { ) -> Result<Self, TraversalModelError> { // Load random forest binary file let rf_binary = std::fs::read(routee_model_path).map_err(\|e\| { - TraversalModelError::FileReadError( - routee_model_path.as_ref().to_path_buf(), - e.to_string(), - ) + TraversalModelError::BuildError(format!( + "failure reading smartcore binary text file {} due to {}", + routee_model_path.as_ref().to_str().unwrap_or_default(), + e + )) })?; let rf: RandomForestRegressor<f64, f64, DenseMatrix<f64>, Vec<f64>> = bincode::deserialize(&rf_binary).map_err(\|e\| { - TraversalModelError::FileReadError( - routee_model_path.as_ref().to_path_buf(), - e.to_string(), - ) + TraversalModelError::BuildError(format!( + "failure deserializing smartcore model {} due to {}", + routee_model_path.as_ref().to_str().unwrap_or_default(), + e + )) })?; Ok(SmartcoreSpeedGradeModel { rf, diff --git a/rust/routee-compass-powertrain/src/routee/vehicle/vehicle_ops.rs b/rust/routee-compass-powertrain/src/routee/vehicle/vehicle_ops.rs index 5c005683..fef1899a 100644 --- a/rust/routee-compass-powertrain/src/routee/vehicle/vehicle_ops.rs +++ b/rust/routee-compass-powertrain/src/routee/vehicle/vehicle_ops.rs @@ -1,5 +1,5 @@ use routee_compass_core::model::{ - state::{state_error::StateError, state_model::StateModel}, + state::{state_model::StateModel, state_model_error::StateModelError}, traversal::state::state_variable::StateVar, unit::{as_f64::AsF64, Energy}, }; @@ -22,7 +22,7 @@ pub fn update_soc_percent( delta: &Energy, max: &Energy, state_model: &StateModel, -) -> Result<(), StateError> { +) -> Result<(), StateModelError> { let start_soc = state_model.get_custom_f64(state, &feature_name.into())?; let start_battery = max.as_f64() (start_soc / 100.0); let current_soc = soc_from_battery_and_delta(&Energy::new(start_battery), delta, max); diff --git a/rust/routee-compass/src/app/bindings.rs b/rust/routee-compass/src/app/bindings.rs index 4a87f90b..88d6a711 100644 --- a/rust/routee-compass/src/app/bindings.rs +++ b/rust/routee-compass/src/app/bindings.rs @@ -1,9 +1,10 @@ use std::str::FromStr; +use itertools::Itertools; use routee_compass_core::{ algorithm::search::direction::Direction, model::{ - road_network::{edge_id::EdgeId, vertex_id::VertexId}, + network::{edge_id::EdgeId, vertex_id::VertexId}, unit::{as_f64::AsF64, DistanceUnit}, }, }; @@ -101,7 +102,7 @@ pub trait CompassAppBindings { let edge_id_internal = EdgeId(edge_id); self.app() .search_app - .get_edge_origin(edge_id_internal) + .get_edge_origin(&edge_id_internal) .map(\|o\| o.0) } @@ -116,7 +117,7 @@ pub trait CompassAppBindings { let edge_id_internal = EdgeId(edge_id); self.app() .search_app - .get_edge_destination(edge_id_internal) + .get_edge_destination(&edge_id_internal) .map(\|o\| o.0) } @@ -150,7 +151,7 @@ pub trait CompassAppBindings { let edge_id_internal = EdgeId(edge_id); self.app() .search_app - .get_edge_distance(edge_id_internal, du_internal) + .get_edge_distance(&edge_id_internal, du_internal) .map(\|o\| o.as_f64()) } @@ -161,12 +162,14 @@ pub trait CompassAppBindings { /// /// # Returns /// * the ids of the edges incident to the vertex in the forward direction - fn graph_get_out_edge_ids(&self, vertex_id: usize) -> Result<Vec<usize>, CompassAppError> { + fn graph_get_out_edge_ids(&self, vertex_id: usize) -> Vec<usize> { let vertex_id_internal = VertexId(vertex_id); self.app() .search_app - .get_incident_edge_ids(vertex_id_internal, Direction::Forward) - .map(\|es\| es.iter().map(\|e\| e.0).collect()) + .get_incident_edge_ids(&vertex_id_internal, &Direction::Forward) + .into_iter() + .map(\|e\| e.0) + .collect_vec() } /// Get the ids of the edges incident to a vertex in the reverse direction @@ -176,12 +179,15 @@ pub trait CompassAppBindings { /// /// # Returns /// * the ids of the edges incident to the vertex in the reverse direction - fn graph_get_in_edge_ids(&self, vertex_id: usize) -> Result<Vec<usize>, CompassAppError> { + fn graph_get_in_edge_ids(&self, vertex_id: usize) -> Vec<usize> { let vertex_id_internal = VertexId(vertex_id); self.app() .search_app - .get_incident_edge_ids(vertex_id_internal, Direction::Reverse) - .map(\|es\| es.iter().map(\|e\| e.0).collect()) + .get_incident_edge_ids(&vertex_id_internal, &Direction::Reverse) + // .map(\|es\| es.iter().map(\|e\| e.0).collect()) + .into_iter() + .map(\|e\| e.0) + .collect_vec() } /// Runs a set of queries and returns the results diff --git a/rust/routee-compass/src/app/cli/cli_args.rs b/rust/routee-compass/src/app/cli/cli_args.rs index b1fa5214..5f920e1c 100644 --- a/rust/routee-compass/src/app/cli/cli_args.rs +++ b/rust/routee-compass/src/app/cli/cli_args.rs @@ -52,7 +52,7 @@ impl CliArgs { if c > 0 { Ok(Some(c as usize)) } else { - Err(CompassAppError::InvalidInput(format!( + Err(CompassAppError::CompassFailure(format!( "chunksize must be positive, found {}", c ))) diff --git a/rust/routee-compass/src/app/cli/run.rs b/rust/routee-compass/src/app/cli/run.rs index ff438d12..ba319fdb 100644 --- a/rust/routee-compass/src/app/cli/run.rs +++ b/rust/routee-compass/src/app/cli/run.rs @@ -43,7 +43,7 @@ pub fn command_line_runner( // read user file containing JSON query/queries let query_file = File::open(args.query_file.clone()).map_err(\|_e\| { - CompassAppError::NoInputFile(format!("Could not find query file {}", args.query_file)) + CompassAppError::BuildFailure(format!("Could not find query file {}", args.query_file)) })?; // execute queries on app @@ -70,8 +70,7 @@ fn run_json( run_config: Option<&Value>, ) -> Result<(), CompassAppError> { let reader = BufReader::new(query_file); - let user_json: serde_json::Value = - serde_json::from_reader(reader).map_err(CompassAppError::CodecError)?; + let user_json: serde_json::Value = serde_json::from_reader(reader)?; let user_queries = user_json.get_queries()?; let results = compass_app.run(user_queries, run_config)?; for result in results.iter() { @@ -106,7 +105,10 @@ fn run_newline_json( CompassConfigurationError::SerdeDeserializationError(e), )), }, - Err(e) => Either::Right(CompassAppError::IOError(e)), + Err(e) => Either::Right(CompassAppError::CompassFailure(format!( + "failed to parse query row due to: {}", + e + ))), }); // run Compass on this chunk of queries diff --git a/rust/routee-compass/src/app/compass/compass_app.rs b/rust/routee-compass/src/app/compass/compass_app.rs index d922706b..801c0e7d 100644 --- a/rust/routee-compass/src/app/compass/compass_app.rs +++ b/rust/routee-compass/src/app/compass/compass_app.rs @@ -351,7 +351,9 @@ impl CompassApp { .animation("fillup") .desc("input plugins") .build() - .map_err(CompassAppError::UXError)?; + .map_err(\|e\| { + CompassAppError::InternalError(format!("could not build progress bar: {}", e)) + })?; let input_pb_shared = Arc::new(Mutex::new(input_pb)); // input plugins need to be flattened, and queries that fail input processing need to be @@ -378,7 +380,7 @@ impl CompassApp { }) .unzip(); - println!(); + eprintln!(); // unpack input plugin results let (processed_inputs_nested, error_inputs_nested) = input_plugin_result; @@ -416,7 +418,9 @@ impl CompassApp { .animation("fillup") .desc("search") .build() - .map_err(CompassAppError::UXError)?; + .map_err(\|e\| { + CompassAppError::InternalError(format!("could not build progress bar: {}", e)) + })?; let search_pb_shared = Arc::new(Mutex::new(search_pb)); // run parallel searches as organized by the (optional) load balancing policy @@ -439,6 +443,7 @@ impl CompassApp { search_pb_shared, )?, }; + eprintln!(); let run_result = run_query_result.chain(error_inputs).collect(); Ok(run_result) @@ -572,7 +577,7 @@ pub fn apply_input_plugins( let mut plugin_state = serde_json::Value::Array(vec![query.clone()]); for plugin in plugins { let p = plugin.clone(); - let op: in_ops::ArrayOp = Rc::new(\|q\| p.process(q)); + let op: in_ops::InputArrayOp = Rc::new(\|q\| p.process(q)); in_ops::json_array_op(&mut plugin_state, op)? } let result = in_ops::json_array_flatten(&mut plugin_state)?; diff --git a/rust/routee-compass/src/app/compass/compass_app_error.rs b/rust/routee-compass/src/app/compass/compass_app_error.rs index ba1c5336..b28255d7 100644 --- a/rust/routee-compass/src/app/compass/compass_app_error.rs +++ b/rust/routee-compass/src/app/compass/compass_app_error.rs @@ -1,49 +1,88 @@ -use super::{ - compass_input_field::CompassInputField, - config::compass_configuration_error::CompassConfigurationError, +use super::config::compass_configuration_error::CompassConfigurationError; +use crate::plugin::{ + input::InputPluginError, output::OutputPluginError, plugin_error::PluginError, }; -use crate::plugin::plugin_error::PluginError; use config::ConfigError; use routee_compass_core::{ algorithm::search::search_error::SearchError, model::{ - frontier::frontier_model_error::FrontierModelError, road_network::graph_error::GraphError, - state::state_error::StateError, traversal::traversal_model_error::TraversalModelError, + access::access_model_error::AccessModelError, cost::cost_model_error::CostModelError, + frontier::frontier_model_error::FrontierModelError, network::network_error::NetworkError, + state::state_model_error::StateModelError, + termination::termination_model_error::TerminationModelError, + traversal::traversal_model_error::TraversalModelError, }, }; #[derive(thiserror::Error, Debug)] pub enum CompassAppError { + // TOP-LEVEL FAILURES + // errors observed during build and run of compass that are not due to the search modules + #[error("failure building compass app: {0}")] + BuildFailure(String), + #[error("failure while running app: {0}")] + CompassFailure(String), + #[error("internal error: {0}")] + InternalError(String), + #[error("error accessing shared read-only dataset: {0}")] + ReadOnlyPoisonError(String), + + // TRANSPARENT MODULE FAILURES + // failures from these modules are detailed enough to get surfaced directly to the user #[error(transparent)] - SearchError(#[from] SearchError), + ConfigFailure(#[from] ConfigError), #[error(transparent)] - FrontierModelError(#[from] FrontierModelError), + CompassConfigurationError(#[from] CompassConfigurationError), #[error(transparent)] - TraversalModelError(#[from] TraversalModelError), + SearchFailure(#[from] SearchError), #[error(transparent)] PluginError(#[from] PluginError), #[error(transparent)] - IOError(#[from] std::io::Error), + InputPluginFailure(#[from] InputPluginError), #[error(transparent)] - CodecError(#[from] serde_json::Error), - #[error(transparent)] - ConfigError(#[from] ConfigError), - #[error(transparent)] - GraphError(#[from] GraphError), - #[error(transparent)] - StateError(#[from] StateError), - #[error("Input file {0} missing")] - NoInputFile(String), - #[error(transparent)] - CompassConfigurationError(#[from] CompassConfigurationError), - #[error("a ux component caused a failure: {0}")] - UXError(String), - #[error("internal error: {0}")] - InternalError(String), - #[error("app input JSON missing field: {0}")] - MissingInputField(CompassInputField), - #[error("error accessing shared read-only dataset: {0}")] - ReadOnlyPoisonError(String), - #[error("error decoding input:\n{0}")] - InvalidInput(String), + OutputPluginFailure(#[from] OutputPluginError), + + // CONTEXTUALIZED MODULE FAILURES + // failures from these modules are happening outside of the context of running the search, + // which is clarified for the user and may help direct where to look to solve the problem. + #[error("While interacting with the state model outside of the context of search, an error occurred. Source: {source}")] + StateFailure { + #[from] + source: StateModelError, + }, + #[error("While interacting with the network model outside of the context of search, an error occurred. Source: {source}")] + NetworkFailure { + #[from] + source: NetworkError, + }, + #[error("While interacting with the termination model outside of the context of search, an error occurred. Source: {source}")] + TerminationModelFailure { + #[from] + source: TerminationModelError, + }, + #[error("While interacting with the traversal model outside of the context of search, an error occurred. Source: {source}")] + TraversalModelFailure { + #[from] + source: TraversalModelError, + }, + #[error("While interacting with the access model outside of the context of search, an error occurred. Source: {source}")] + AccessModelFailure { + #[from] + source: AccessModelError, + }, + #[error("While interacting with the frontier model outside of the context of search, an error occurred. Source: {source}")] + FrontierModelFailure { + #[from] + source: FrontierModelError, + }, + #[error("While interacting with the cost model outside of the context of search, an error occurred. Source: {source}")] + CostFailure { + #[from] + source: CostModelError, + }, + #[error("failure due to JSON: {source}")] + JsonError { + #[from] + source: serde_json::Error, + }, } diff --git a/rust/routee-compass/src/app/compass/compass_app_ops.rs b/rust/routee-compass/src/app/compass/compass_app_ops.rs index 2bb2325f..df358a4d 100644 --- a/rust/routee-compass/src/app/compass/compass_app_ops.rs +++ b/rust/routee-compass/src/app/compass/compass_app_ops.rs @@ -36,8 +36,7 @@ pub fn read_config_from_file(config_path: &Path) -> Result<Config, CompassAppErr CompassInputField::ConfigInputFile.to_string(), conf_file_string, )? - .build() - .map_err(CompassAppError::ConfigError)?; + .build()?; Ok(config) } @@ -74,8 +73,7 @@ pub fn read_config_from_string( CompassInputField::ConfigInputFile.to_string(), original_file_path, )? - .build() - .map_err(CompassAppError::ConfigError)?; + .build()?; Ok(config) } diff --git a/rust/routee-compass/src/app/compass/compass_json_extensions.rs b/rust/routee-compass/src/app/compass/compass_json_extensions.rs index 3c0ad2a1..ee8211bb 100644 --- a/rust/routee-compass/src/app/compass/compass_json_extensions.rs +++ b/rust/routee-compass/src/app/compass/compass_json_extensions.rs @@ -19,11 +19,11 @@ impl CompassJsonExtensions for serde_json::Value { serde_json::Value::Array(vec) => Ok(vec.to_owned()), _ => { let msg = String::from("user JSON argument must be an object or an array"); - Err(CompassAppError::InvalidInput(msg)) + Err(CompassAppError::CompassFailure(msg)) } }, }, - _ => Err(CompassAppError::InvalidInput(String::from( + _ => Err(CompassAppError::CompassFailure(String::from( "expected object, object with queries, or array input", ))), } diff --git a/rust/routee-compass/src/app/compass/config/compass_configuration_error.rs b/rust/routee-compass/src/app/compass/config/compass_configuration_error.rs index 729e3203..4b8b7065 100644 --- a/rust/routee-compass/src/app/compass/config/compass_configuration_error.rs +++ b/rust/routee-compass/src/app/compass/config/compass_configuration_error.rs @@ -3,7 +3,7 @@ use config::ConfigError; use routee_compass_core::{ model::{ access::access_model_error::AccessModelError, - frontier::frontier_model_error::FrontierModelError, road_network::graph_error::GraphError, + frontier::frontier_model_error::FrontierModelError, network::network_error::NetworkError, traversal::traversal_model_error::TraversalModelError, }, util::{cache_policy::cache_error::CacheError, conversion::conversion_error::ConversionError}, @@ -53,7 +53,7 @@ pub enum CompassConfigurationError { #[error("{0}")] InsertError(String), #[error(transparent)] - GraphError(#[from] GraphError), + GraphError(#[from] NetworkError), #[error(transparent)] ConfigError(#[from] ConfigError), #[error(transparent)] diff --git a/rust/routee-compass/src/app/compass/config/frontier_model/combined/combined_model.rs b/rust/routee-compass/src/app/compass/config/frontier_model/combined/combined_model.rs index 9fda3248..75f5154d 100644 --- a/rust/routee-compass/src/app/compass/config/frontier_model/combined/combined_model.rs +++ b/rust/routee-compass/src/app/compass/config/frontier_model/combined/combined_model.rs @@ -1,6 +1,6 @@ use routee_compass_core::model::{ frontier::{frontier_model::FrontierModel, frontier_model_error::FrontierModelError}, - property::edge::Edge, + network::Edge, state::state_model::StateModel, traversal::state::state_variable::StateVar, }; diff --git a/rust/routee-compass/src/app/compass/config/frontier_model/road_class/road_class_model.rs b/rust/routee-compass/src/app/compass/config/frontier_model/road_class/road_class_model.rs index dd2ab068..abcf7a89 100644 --- a/rust/routee-compass/src/app/compass/config/frontier_model/road_class/road_class_model.rs +++ b/rust/routee-compass/src/app/compass/config/frontier_model/road_class/road_class_model.rs @@ -1,7 +1,7 @@ use super::road_class_service::RoadClassFrontierService; use routee_compass_core::model::{ frontier::{frontier_model::FrontierModel, frontier_model_error::FrontierModelError}, - property::edge::Edge, + network::Edge, state::state_model::StateModel, traversal::state::state_variable::StateVar, }; @@ -26,7 +26,12 @@ impl FrontierModel for RoadClassFrontierModel { .service .road_class_lookup .get(edge.edge_id.0) - .ok_or_else(\|\| FrontierModelError::MissingIndex(format!("{}", edge.edge_id))) + .ok_or_else(\|\| { + FrontierModelError::FrontierModelError(format!( + "edge id {} missing from frontier model file", + edge.edge_id + )) + }) .map(\|road_class\| road_classes.contains(road_class)), } } diff --git a/rust/routee-compass/src/app/compass/config/frontier_model/road_class/road_class_parser.rs b/rust/routee-compass/src/app/compass/config/frontier_model/road_class/road_class_parser.rs index 06df970d..dce3fe82 100644 --- a/rust/routee-compass/src/app/compass/config/frontier_model/road_class/road_class_parser.rs +++ b/rust/routee-compass/src/app/compass/config/frontier_model/road_class/road_class_parser.rs @@ -25,7 +25,7 @@ impl RoadClassParser { if self.mapping.is_empty() { // if we don't have a road class mapping then we fail let value_string = value.to_string(); - Err(CompassAppError::InvalidInput( + Err(CompassAppError::CompassFailure( formatdoc! {r#" Could not parse incoming query valid road_classes of {value_string} as an array of integers and this FrontierModel does not specify a mapping of string to integer. @@ -42,7 +42,7 @@ impl RoadClassParser { .map(\|s\| { self.mapping.get(s).cloned().ok_or_else(\|\| { let valid_mapping: Vec<String> = self.mapping.keys().cloned().collect(); - CompassAppError::InvalidInput( + CompassAppError::CompassFailure( format!("Could not find road class mapping for incoming value {}. Here are the road classes I have: {:?}", s, valid_mapping) ) }) diff --git a/rust/routee-compass/src/app/compass/config/frontier_model/turn_restrictions/turn_restriction_model.rs b/rust/routee-compass/src/app/compass/config/frontier_model/turn_restrictions/turn_restriction_model.rs index c719fd32..fbd489da 100644 --- a/rust/routee-compass/src/app/compass/config/frontier_model/turn_restrictions/turn_restriction_model.rs +++ b/rust/routee-compass/src/app/compass/config/frontier_model/turn_restrictions/turn_restriction_model.rs @@ -1,6 +1,6 @@ use routee_compass_core::model::{ frontier::{frontier_model::FrontierModel, frontier_model_error::FrontierModelError}, - property::edge::Edge, + network::Edge, state::state_model::StateModel, traversal::state::state_variable::StateVar, }; diff --git a/rust/routee-compass/src/app/compass/config/frontier_model/turn_restrictions/turn_restriction_service.rs b/rust/routee-compass/src/app/compass/config/frontier_model/turn_restrictions/turn_restriction_service.rs index 1953b3ef..4a7250a7 100644 --- a/rust/routee-compass/src/app/compass/config/frontier_model/turn_restrictions/turn_restriction_service.rs +++ b/rust/routee-compass/src/app/compass/config/frontier_model/turn_restrictions/turn_restriction_service.rs @@ -4,7 +4,7 @@ use routee_compass_core::model::{ frontier_model::FrontierModel, frontier_model_error::FrontierModelError, frontier_model_service::FrontierModelService, }, - road_network::edge_id::EdgeId, + network::edge_id::EdgeId, state::state_model::StateModel, }; use serde::Deserialize; diff --git a/rust/routee-compass/src/app/compass/config/frontier_model/vehicle_restrictions/vehicle_restriction_builder.rs b/rust/routee-compass/src/app/compass/config/frontier_model/vehicle_restrictions/vehicle_restriction_builder.rs index 683e4e56..ac36407b 100644 --- a/rust/routee-compass/src/app/compass/config/frontier_model/vehicle_restrictions/vehicle_restriction_builder.rs +++ b/rust/routee-compass/src/app/compass/config/frontier_model/vehicle_restrictions/vehicle_restriction_builder.rs @@ -8,7 +8,7 @@ use routee_compass_core::{ frontier_model_builder::FrontierModelBuilder, frontier_model_error::FrontierModelError, frontier_model_service::FrontierModelService, }, - road_network::edge_id::EdgeId, + network::edge_id::EdgeId, }, util::fs::read_utils, }; diff --git a/rust/routee-compass/src/app/compass/config/frontier_model/vehicle_restrictions/vehicle_restriction_model.rs b/rust/routee-compass/src/app/compass/config/frontier_model/vehicle_restrictions/vehicle_restriction_model.rs index 9ad3295f..77a6a43f 100644 --- a/rust/routee-compass/src/app/compass/config/frontier_model/vehicle_restrictions/vehicle_restriction_model.rs +++ b/rust/routee-compass/src/app/compass/config/frontier_model/vehicle_restrictions/vehicle_restriction_model.rs @@ -4,7 +4,7 @@ use super::{ }; use routee_compass_core::model::{ frontier::{frontier_model::FrontierModel, frontier_model_error::FrontierModelError}, - property::edge::Edge, + network::Edge, state::state_model::StateModel, traversal::state::state_variable::StateVar, }; diff --git a/rust/routee-compass/src/app/compass/config/frontier_model/vehicle_restrictions/vehicle_restriction_row.rs b/rust/routee-compass/src/app/compass/config/frontier_model/vehicle_restrictions/vehicle_restriction_row.rs index 428eaa33..834a5725 100644 --- a/rust/routee-compass/src/app/compass/config/frontier_model/vehicle_restrictions/vehicle_restriction_row.rs +++ b/rust/routee-compass/src/app/compass/config/frontier_model/vehicle_restrictions/vehicle_restriction_row.rs @@ -1,5 +1,5 @@ use routee_compass_core::model::{ - frontier::frontier_model_error::FrontierModelError, road_network::edge_id::EdgeId, + frontier::frontier_model_error::FrontierModelError, network::edge_id::EdgeId, }; use serde::Deserialize; diff --git a/rust/routee-compass/src/app/compass/config/frontier_model/vehicle_restrictions/vehicle_restriction_service.rs b/rust/routee-compass/src/app/compass/config/frontier_model/vehicle_restrictions/vehicle_restriction_service.rs index de97ab2c..d9092dd5 100644 --- a/rust/routee-compass/src/app/compass/config/frontier_model/vehicle_restrictions/vehicle_restriction_service.rs +++ b/rust/routee-compass/src/app/compass/config/frontier_model/vehicle_restrictions/vehicle_restriction_service.rs @@ -7,7 +7,7 @@ use routee_compass_core::model::{ frontier_model::FrontierModel, frontier_model_error::FrontierModelError, frontier_model_service::FrontierModelService, }, - road_network::edge_id::EdgeId, + network::edge_id::EdgeId, state::state_model::StateModel, }; use std::{collections::HashMap, sync::Arc}; diff --git a/rust/routee-compass/src/app/compass/config/graph_builder.rs b/rust/routee-compass/src/app/compass/config/graph_builder.rs index dc4f15aa..f74860c5 100644 --- a/rust/routee-compass/src/app/compass/config/graph_builder.rs +++ b/rust/routee-compass/src/app/compass/config/graph_builder.rs @@ -1,4 +1,4 @@ -use routee_compass_core::model::road_network::graph::Graph; +use routee_compass_core::model::network::graph::Graph; use crate::app::compass::config::compass_configuration_field::CompassConfigurationField; diff --git a/rust/routee-compass/src/app/compass/response/response_output_format_json.rs b/rust/routee-compass/src/app/compass/response/response_output_format_json.rs index d3baa8af..a3eb3dd8 100644 --- a/rust/routee-compass/src/app/compass/response/response_output_format_json.rs +++ b/rust/routee-compass/src/app/compass/response/response_output_format_json.rs @@ -21,9 +21,9 @@ pub fn format_response( newline_delimited: bool, ) -> Result<String, CompassAppError> { if newline_delimited { - serde_json::to_string(response).map_err(CompassAppError::CodecError) + serde_json::to_string(response).map_err(CompassAppError::from) } else { - let row = serde_json::to_string_pretty(response).map_err(CompassAppError::CodecError)?; + let row = serde_json::to_string_pretty(response).map_err(CompassAppError::from)?; Ok(row) } } diff --git a/rust/routee-compass/src/app/compass/response/response_output_policy.rs b/rust/routee-compass/src/app/compass/response/response_output_policy.rs index d79e2825..66699c25 100644 --- a/rust/routee-compass/src/app/compass/response/response_output_policy.rs +++ b/rust/routee-compass/src/app/compass/response/response_output_policy.rs @@ -44,7 +44,7 @@ impl ResponseOutputPolicy { let file_shareable = Arc::new(Mutex::new(file)); let iterations_per_flush = match file_flush_rate { - Some(rate) if rate <= 0 => Err(CompassAppError::InvalidInput(format!( + Some(rate) if rate <= 0 => Err(CompassAppError::CompassFailure(format!( "file policy iterations_per_flush must be positive, found {}", rate ))), diff --git a/rust/routee-compass/src/app/compass/response/response_sink.rs b/rust/routee-compass/src/app/compass/response/response_sink.rs index d829292d..08cb8396 100644 --- a/rust/routee-compass/src/app/compass/response/response_sink.rs +++ b/rust/routee-compass/src/app/compass/response/response_sink.rs @@ -25,7 +25,7 @@ impl ResponseSink { match self { ResponseSink::None => Ok(()), ResponseSink::File { - filename: _, + filename, file, format, delimiter: _, @@ -48,10 +48,20 @@ impl ResponseSink { })?; let output_row = format.format_response(response)?; - writeln!(file_attained, "{}", output_row).map_err(CompassAppError::IOError)?; + writeln!(file_attained, "{}", output_row).map_err(\|e\| { + CompassAppError::InternalError(format!( + "failure writing to {}: {}", + filename, e + )) + })?; it_attained += 1; if it_attained % iterations_per_flush == 0 { - file_attained.flush().map_err(CompassAppError::IOError)?; + file_attained.flush().map_err(\|e\| { + CompassAppError::InternalError(format!( + "failure flushing output to {}: {}", + filename, e + )) + })?; } Ok(()) @@ -87,7 +97,12 @@ impl ResponseSink { let final_contents = format .final_file_contents() .unwrap_or_else(\|\| String::from("")); - writeln!(file_attained, "{}", final_contents).map_err(CompassAppError::IOError)?; + writeln!(file_attained, "{}", final_contents).map_err(\|e\| { + CompassAppError::InternalError(format!( + "failure writing final contents to {}: {}", + filename, e + )) + })?; Ok(filename.clone()) } diff --git a/rust/routee-compass/src/app/compass/response/write_mode.rs b/rust/routee-compass/src/app/compass/response/write_mode.rs index 7e2cfb7e..4734f426 100644 --- a/rust/routee-compass/src/app/compass/response/write_mode.rs +++ b/rust/routee-compass/src/app/compass/response/write_mode.rs @@ -54,12 +54,21 @@ fn write_header(path: &Path, format: &ResponseOutputFormat) -> Result<(), Compas let header = format .initial_file_contents() .unwrap_or_else(\|\| String::from("")); - std::fs::write(path, header).map_err(CompassAppError::IOError) + std::fs::write(path, header).map_err(\|e\| { + CompassAppError::InternalError(format!( + "failure writing to {}: {}", + path.to_str().unwrap_or_default(), + e + )) + }) } fn open_append(path: &Path) -> Result<File, CompassAppError> { - OpenOptions::new() - .append(true) - .open(path) - .map_err(CompassAppError::IOError) + OpenOptions::new().append(true).open(path).map_err(\|e\| { + CompassAppError::InternalError(format!( + "failure opening file {} in append mode: {}", + path.to_str().unwrap_or_default(), + e + )) + }) } diff --git a/rust/routee-compass/src/app/geom/geom_app.rs b/rust/routee-compass/src/app/geom/geom_app.rs index 5a51dd1b..e1ad4983 100644 --- a/rust/routee-compass/src/app/geom/geom_app.rs +++ b/rust/routee-compass/src/app/geom/geom_app.rs @@ -24,14 +24,21 @@ impl TryFrom<&GeomAppConfig> for GeomApp { fn try_from(conf: &GeomAppConfig) -> Result<Self, Self::Error> { let count = fs_utils::line_count(conf.edge_file.clone(), fs_utils::is_gzip(&conf.edge_file)) - .map_err(CompassAppError::IOError)?; + .map_err(\|e\| { + CompassAppError::BuildFailure(format!( + "failure reading edge file {}: {}", + conf.edge_file, e + )) + })?; let mut pb = Bar::builder() .total(count) .animation("fillup") .desc("geometry file") .build() - .map_err(CompassAppError::UXError)?; + .map_err(\|e\| { + CompassAppError::InternalError(format!("could not build progress bar: {}", e)) + })?; let cb = Box::new(\|\| { let _ = pb.update(1); @@ -42,9 +49,13 @@ impl TryFrom<&GeomAppConfig> for GeomApp { Ok(result) }; - let geoms = read_utils::read_raw_file(&conf.edge_file, op, Some(cb)) - .map_err(CompassAppError::IOError)?; - println!(); + let geoms = read_utils::read_raw_file(&conf.edge_file, op, Some(cb)).map_err(\|e\| { + CompassAppError::BuildFailure(format!( + "failure reading edge file {}: {}", + conf.edge_file, e + )) + })?; + eprintln!(); let app = GeomApp { geoms }; Ok(app) } @@ -54,15 +65,21 @@ impl GeomApp { /// run the GeomApp. reads each line of a file, which is expected to be a number coorelating to /// some EdgeId. looks up the geometry for that EdgeId. pub fn run(&self, file: String) -> Result<Box<[LineString<f32>]>, CompassAppError> { - let count = fs_utils::line_count(file.clone(), fs_utils::is_gzip(&file)) - .map_err(CompassAppError::IOError)?; + let count = fs_utils::line_count(file.clone(), fs_utils::is_gzip(&file)).map_err(\|e\| { + CompassAppError::BuildFailure(format!( + "failure reading geometry index input file {}: {}", + file, e + )) + })?; let mut pb = Bar::builder() .total(count) .animation("fillup") .desc("edge id list") .build() - .map_err(CompassAppError::UXError)?; + .map_err(\|e\| { + CompassAppError::InternalError(format!("could not build progress bar: {}", e)) + })?; let cb = Box::new(\|\| { let _ = pb.update(1); @@ -81,8 +98,14 @@ impl GeomApp { result }; - let result: Box<[LineString<f32>]> = - read_utils::read_raw_file(&file, op, Some(cb)).map_err(CompassAppError::IOError)?; + let result: Box<[LineString<f32>]> = read_utils::read_raw_file(&file, op, Some(cb)) + .map_err(\|e\| { + CompassAppError::BuildFailure(format!( + "failure reading linestring file {}: {}", + file, e + )) + })?; + eprintln!(); Ok(result) } } diff --git a/rust/routee-compass/src/app/search/search_app.rs b/rust/routee-compass/src/app/search/search_app.rs index 834a22b5..5e99f621 100644 --- a/rust/routee-compass/src/app/search/search_app.rs +++ b/rust/routee-compass/src/app/search/search_app.rs @@ -5,7 +5,7 @@ use crate::{ config::cost_model::cost_model_service::CostModelService, search_orientation::SearchOrientation, }, - plugin::input::input_json_extensions::InputJsonExtensions, + plugin::{input::input_json_extensions::InputJsonExtensions, plugin_error::PluginError}, }; use chrono::Local; use routee_compass_core::{ @@ -16,7 +16,7 @@ use routee_compass_core::{ }, model::{ access::access_model_service::AccessModelService, - frontier::frontier_model_service::FrontierModelService, road_network::graph::Graph, + frontier::frontier_model_service::FrontierModelService, network::graph::Graph, state::state_model::StateModel, termination::termination_model::TerminationModel, traversal::traversal_model_service::TraversalModelService, }, @@ -112,35 +112,35 @@ impl SearchApp { &self, query: &serde_json::Value, ) -> Result<(SearchAlgorithmResult, SearchInstance), CompassAppError> { - let o = query - .get_origin_vertex() - .map_err(CompassAppError::PluginError)?; - let d = query - .get_destination_vertex() - .map_err(CompassAppError::PluginError)?; + let o = query.get_origin_vertex().map_err(\|e\| { + CompassAppError::PluginError(PluginError::InputPluginFailed { source: e }) + })?; + let d = query.get_destination_vertex().map_err(\|e\| { + CompassAppError::PluginError(PluginError::InputPluginFailed { source: e }) + })?; let search_instance = self.build_search_instance(query)?; self.search_algorithm .run_vertex_oriented(o, d, query, &Direction::Forward, &search_instance) .map(\|search_result\| (search_result, search_instance)) - .map_err(CompassAppError::SearchError) + .map_err(CompassAppError::SearchFailure) } pub fn run_edge_oriented( &self, query: &serde_json::Value, ) -> Result<(SearchAlgorithmResult, SearchInstance), CompassAppError> { - let o = query - .get_origin_edge() - .map_err(CompassAppError::PluginError)?; - let d_opt = query - .get_destination_edge() - .map_err(CompassAppError::PluginError)?; + let o = query.get_origin_edge().map_err(\|e\| { + CompassAppError::PluginError(PluginError::InputPluginFailed { source: e }) + })?; + let d_opt = query.get_destination_edge().map_err(\|e\| { + CompassAppError::PluginError(PluginError::InputPluginFailed { source: e }) + })?; let search_instance = self.build_search_instance(query)?; self.search_algorithm .run_edge_oriented(o, d_opt, query, &Direction::Forward, &search_instance) .map(\|search_result\| (search_result, search_instance)) - .map_err(CompassAppError::SearchError) + .map_err(CompassAppError::SearchFailure) } /// builds the assets that will run the search for this query instance. diff --git a/rust/routee-compass/src/app/search/search_app_graph_ops.rs b/rust/routee-compass/src/app/search/search_app_graph_ops.rs index 93438469..0d0802a1 100644 --- a/rust/routee-compass/src/app/search/search_app_graph_ops.rs +++ b/rust/routee-compass/src/app/search/search_app_graph_ops.rs @@ -2,51 +2,38 @@ use super::search_app::SearchApp; use crate::app::compass::compass_app_error::CompassAppError; use routee_compass_core::{ algorithm::search::direction::Direction, - model::road_network::{edge_id::EdgeId, vertex_id::VertexId}, + model::network::{edge_id::EdgeId, vertex_id::VertexId}, model::unit::{Distance, DistanceUnit}, }; pub trait SearchAppGraphOps { - fn get_edge_origin(&self, edge_id: EdgeId) -> Result<VertexId, CompassAppError>; - fn get_edge_destination(&self, edge_id: EdgeId) -> Result<VertexId, CompassAppError>; + fn get_edge_origin(&self, edge_id: &EdgeId) -> Result<VertexId, CompassAppError>; + fn get_edge_destination(&self, edge_id: &EdgeId) -> Result<VertexId, CompassAppError>; fn get_edge_distance( &self, - edge_id: EdgeId, + edge_id: &EdgeId, distance_unit: Option<DistanceUnit>, ) -> Result<Distance, CompassAppError>; - fn get_incident_edge_ids( - &self, - vertex_id: VertexId, - direction: Direction, - ) -> Result<Vec<EdgeId>, CompassAppError>; + fn get_incident_edge_ids(&self, vertex_id: &VertexId, direction: &Direction) -> Vec<EdgeId>; } impl SearchAppGraphOps for SearchApp { - fn get_edge_origin(&self, edge_id: EdgeId) -> Result<VertexId, CompassAppError> { - let edge = self - .directed_graph - .get_edge(edge_id) - .map_err(CompassAppError::GraphError)?; + fn get_edge_origin(&self, edge_id: &EdgeId) -> Result<VertexId, CompassAppError> { + let edge = self.directed_graph.get_edge(edge_id)?; Ok(edge.src_vertex_id) } - fn get_edge_destination(&self, edge_id: EdgeId) -> Result<VertexId, CompassAppError> { - let edge = self - .directed_graph - .get_edge(edge_id) - .map_err(CompassAppError::GraphError)?; + fn get_edge_destination(&self, edge_id: &EdgeId) -> Result<VertexId, CompassAppError> { + let edge = self.directed_graph.get_edge(edge_id)?; Ok(edge.dst_vertex_id) } fn get_edge_distance( &self, - edge_id: EdgeId, + edge_id: &EdgeId, distance_unit: Option<DistanceUnit>, ) -> Result<Distance, CompassAppError> { - let edge = self - .directed_graph - .get_edge(edge_id) - .map_err(CompassAppError::GraphError)?; + let edge = self.directed_graph.get_edge(edge_id)?; let result_base = edge.distance; let result = match distance_unit { Some(du) => DistanceUnit::Meters.convert(&result_base, &du), @@ -55,15 +42,7 @@ impl SearchAppGraphOps for SearchApp { Ok(result) } - fn get_incident_edge_ids( - &self, - vertex_id: VertexId, - direction: Direction, - ) -> Result<Vec<EdgeId>, CompassAppError> { - let incident_edges = self - .directed_graph - .incident_edges(vertex_id, direction) - .map_err(CompassAppError::GraphError)?; - Ok(incident_edges) + fn get_incident_edge_ids(&self, vertex_id: &VertexId, direction: &Direction) -> Vec<EdgeId> { + self.directed_graph.incident_edges(vertex_id, direction) } } diff --git a/rust/routee-compass/src/app/search/search_app_ops.rs b/rust/routee-compass/src/app/search/search_app_ops.rs index 7a2ac161..4791237b 100644 --- a/rust/routee-compass/src/app/search/search_app_ops.rs +++ b/rust/routee-compass/src/app/search/search_app_ops.rs @@ -3,7 +3,7 @@ use std::{collections::HashMap, sync::Arc}; use itertools::Itertools; use routee_compass_core::model::{ access::access_model::AccessModel, - state::{state_error::StateError, state_feature::StateFeature}, + state::{state_feature::StateFeature, state_model_error::StateModelError}, traversal::traversal_model::TraversalModel, }; @@ -22,7 +22,7 @@ pub fn collect_features( query: &serde_json::Value, traversal_model: Arc<dyn TraversalModel>, access_model: Arc<dyn AccessModel>, -) -> Result<Vec<(String, StateFeature)>, StateError> { +) -> Result<Vec<(String, StateFeature)>, StateModelError> { // prepare the set of features for this state model let model_features = traversal_model .state_features() @@ -34,17 +34,17 @@ pub fn collect_features( // state features. let user_features_option: Option<HashMap<String, StateFeature>> = query .get_config_serde_optional(&"state_features", &"query") - .map_err(\|e\| StateError::BuildError(e.to_string()))?; + .map_err(\|e\| StateModelError::BuildError(e.to_string()))?; let user_features = user_features_option .unwrap_or_default() .into_iter() .map(\|(name, feature)\| match model_features.get(&name) { None => { let fnames = model_features.keys().join(","); - Err(StateError::UnknownStateVariableName(name, fnames)) + Err(StateModelError::UnknownStateVariableName(name, fnames)) } Some(existing) if existing.get_feature_type() != feature.get_feature_type() => { - Err(StateError::UnexpectedFeatureType( + Err(StateModelError::UnexpectedFeatureType( existing.get_feature_type(), feature.get_feature_type(), )) diff --git a/rust/routee-compass/src/app/search/search_app_result.rs b/rust/routee-compass/src/app/search/search_app_result.rs index 3f71ccb2..c5461cee 100644 --- a/rust/routee-compass/src/app/search/search_app_result.rs +++ b/rust/routee-compass/src/app/search/search_app_result.rs @@ -2,7 +2,7 @@ use allocative::Allocative; use routee_compass_core::{ algorithm::search::{edge_traversal::EdgeTraversal, search_tree_branch::SearchTreeBranch}, - model::road_network::vertex_id::VertexId, + model::network::vertex_id::VertexId, }; use std::{collections::HashMap, time::Duration}; diff --git a/rust/routee-compass/src/plugin/input/default/debug/debug_plugin.rs b/rust/routee-compass/src/plugin/input/default/debug/debug_plugin.rs index 44a6abda..dbfd5daf 100644 --- a/rust/routee-compass/src/plugin/input/default/debug/debug_plugin.rs +++ b/rust/routee-compass/src/plugin/input/default/debug/debug_plugin.rs @@ -1,10 +1,11 @@ -use crate::plugin::{input::input_plugin::InputPlugin, plugin_error::PluginError}; +use crate::plugin::input::{input_plugin::InputPlugin, InputPluginError}; pub struct DebugInputPlugin {} impl InputPlugin for DebugInputPlugin { - fn process(&self, input: &mut serde_json::Value) -> Result<(), PluginError> { - let string = serde_json::to_string_pretty(input).map_err(PluginError::JsonError)?; + fn process(&self, input: &mut serde_json::Value) -> Result<(), InputPluginError> { + let string = serde_json::to_string_pretty(input) + .map_err(\|e\| InputPluginError::JsonError { source: e })?; println!("{}", string); Ok(()) } diff --git a/rust/routee-compass/src/plugin/input/default/edge_rtree/edge_rtree_input_plugin.rs b/rust/routee-compass/src/plugin/input/default/edge_rtree/edge_rtree_input_plugin.rs index eddfd61e..286e52c9 100644 --- a/rust/routee-compass/src/plugin/input/default/edge_rtree/edge_rtree_input_plugin.rs +++ b/rust/routee-compass/src/plugin/input/default/edge_rtree/edge_rtree_input_plugin.rs @@ -10,14 +10,13 @@ use crate::{ }, }, }, - plugin::{ - input::{input_json_extensions::InputJsonExtensions, input_plugin::InputPlugin}, - plugin_error::PluginError, + plugin::input::{ + input_json_extensions::InputJsonExtensions, input_plugin::InputPlugin, InputPluginError, }, }; use geo_types::Coord; use routee_compass_core::{ - model::road_network::edge_id::EdgeId, + model::network::edge_id::EdgeId, model::unit::{as_f64::AsF64, Distance, DistanceUnit, BASE_DISTANCE_UNIT}, util::{ fs::{read_decoders, read_utils}, @@ -49,9 +48,9 @@ pub struct EdgeRtreeInputPlugin { impl InputPlugin for EdgeRtreeInputPlugin { /// finds the nearest edge ids to the user-provided origin and destination coordinates. /// optionally restricts the search to a subset of road classes tagged by the user. - fn process(&self, query: &mut serde_json::Value) -> Result<(), PluginError> { + fn process(&self, query: &mut serde_json::Value) -> Result<(), InputPluginError> { let road_classes = self.road_class_parser.read_query(query).map_err(\|e\| { - PluginError::InputError(format!( + InputPluginError::InputPluginFailed(format!( "Unable to apply EdgeRtree Input Plugin due to:\n\n{}", e )) @@ -188,7 +187,7 @@ fn search( road_classes: &Option<HashSet<u8>>, vehicle_restrictions: &Option<HashMap<EdgeId, Vec<VehicleRestriction>>>, vehicle_parameters: &Option<VehicleParameters>, -) -> Result<Option<EdgeId>, PluginError> { +) -> Result<Option<EdgeId>, InputPluginError> { let point = geo::Point(coord); for (record, distance_meters) in rtree.nearest_neighbor_iter_with_distance_2(&point) { if !within_tolerance(tolerance, &distance_meters) { @@ -197,7 +196,7 @@ fn search( let valid_class = match (road_classes, road_class_lookup) { (Some(valid_classes), Some(lookup)) => { let this_class = lookup.get(record.edge_id.0).ok_or_else(\|\| { - PluginError::PluginFailed(format!( + InputPluginError::InputPluginFailed(format!( "edge rtree road class file missing edge {}", record.edge_id )) @@ -227,12 +226,15 @@ fn search( } /// helper to build a matching error response -fn matching_error(coord: &Coord<f32>, tolerance: Option<(Distance, DistanceUnit)>) -> PluginError { +fn matching_error( + coord: &Coord<f32>, + tolerance: Option<(Distance, DistanceUnit)>, +) -> InputPluginError { let mut message = format!("unable to match coordinate {:?} to network edge", coord); if let Some((dist, unit)) = tolerance { message.push_str(&format!(" within tolerance of {} {}", dist, unit)) }; - PluginError::PluginFailed(message) + InputPluginError::InputPluginFailed(message) } /// helper to test if some distance in meters is within the optionally-provided tolerance diff --git a/rust/routee-compass/src/plugin/input/default/edge_rtree/edge_rtree_record.rs b/rust/routee-compass/src/plugin/input/default/edge_rtree/edge_rtree_record.rs index 672bc36d..f0c67c62 100644 --- a/rust/routee-compass/src/plugin/input/default/edge_rtree/edge_rtree_record.rs +++ b/rust/routee-compass/src/plugin/input/default/edge_rtree/edge_rtree_record.rs @@ -1,5 +1,5 @@ use geo::{Centroid, LineString, Point}; -use routee_compass_core::model::road_network::edge_id::EdgeId; +use routee_compass_core::model::network::edge_id::EdgeId; use rstar::{PointDistance, RTreeObject, AABB}; pub struct EdgeRtreeRecord { diff --git a/rust/routee-compass/src/plugin/input/default/grid_search/plugin.rs b/rust/routee-compass/src/plugin/input/default/grid_search/plugin.rs index 9eb18f12..fb62246b 100644 --- a/rust/routee-compass/src/plugin/input/default/grid_search/plugin.rs +++ b/rust/routee-compass/src/plugin/input/default/grid_search/plugin.rs @@ -1,7 +1,7 @@ use crate::plugin::input::input_field::InputField; use crate::plugin::input::input_json_extensions::InputJsonExtensions; use crate::plugin::input::input_plugin::InputPlugin; -use crate::plugin::plugin_error::PluginError; +use crate::plugin::input::InputPluginError; use routee_compass_core::util::multiset::MultiSet; /// Builds an input plugin that duplicates queries if array-valued fields are present @@ -9,23 +9,23 @@ use routee_compass_core::util::multiset::MultiSet; pub struct GridSearchPlugin {} impl InputPlugin for GridSearchPlugin { - fn process(&self, input: &mut serde_json::Value) -> Result<(), PluginError> { + fn process(&self, input: &mut serde_json::Value) -> Result<(), InputPluginError> { match input.get_grid_search() { None => Ok(()), Some(grid_search_input) => { // prevent recursion due to nested grid search keys let recurses = serde_json::to_string(grid_search_input) - .map_err(PluginError::JsonError)? + .map_err(\|e\| InputPluginError::JsonError { source: e })? .contains("grid_search"); if recurses { - return Err(PluginError::PluginFailed(String::from( + return Err(InputPluginError::InputPluginFailed(String::from( "grid search section cannot contain the string 'grid_search'", ))); } - let map = grid_search_input - .as_object() - .ok_or_else(\|\| PluginError::UnexpectedQueryStructure(format!("{:?}", input)))?; + let map = grid_search_input.as_object().ok_or_else(\|\| { + InputPluginError::UnexpectedQueryStructure(format!("{:?}", input)) + })?; let mut keys: Vec<String> = vec![]; let mut multiset_input: Vec<Vec<serde_json::Value>> = vec![]; let mut multiset_indices: Vec<Vec<usize>> = vec![]; @@ -42,7 +42,9 @@ impl InputPlugin for GridSearchPlugin { // let remove_key = InputField::GridSearch.to_str(); let mut initial_map = input .as_object() - .ok_or_else(\|\| PluginError::UnexpectedQueryStructure(format!("{:?}", input)))? + .ok_or_else(\|\| { + InputPluginError::UnexpectedQueryStructure(format!("{:?}", input)) + })? .clone(); initial_map.remove(InputField::GridSearch.to_str()); let initial = serde_json::json!(initial_map); diff --git a/rust/routee-compass/src/plugin/input/default/inject/inject_builder.rs b/rust/routee-compass/src/plugin/input/default/inject/inject_builder.rs index 1949a805..b2c8a45a 100644 --- a/rust/routee-compass/src/plugin/input/default/inject/inject_builder.rs +++ b/rust/routee-compass/src/plugin/input/default/inject/inject_builder.rs @@ -20,7 +20,9 @@ impl InputPluginBuilder for InjectPluginBuilder { let value_string = parameters.get_config_string(&"value", &"inject")?; let format: InjectFormat = parameters.get_config_serde(&"format", &"inject")?; let value = format.to_json(&value_string)?; - let plugin = InjectInputPlugin::new(key, value); + let overwrite: Option<bool> = + parameters.get_config_serde_optional(&"overwrite", &"inject")?; + let plugin = InjectInputPlugin::new(key, value, overwrite); Ok(Arc::new(plugin)) } } diff --git a/rust/routee-compass/src/plugin/input/default/inject/inject_format.rs b/rust/routee-compass/src/plugin/input/default/inject/inject_format.rs index 0abdde00..ca16a714 100644 --- a/rust/routee-compass/src/plugin/input/default/inject/inject_format.rs +++ b/rust/routee-compass/src/plugin/input/default/inject/inject_format.rs @@ -1,11 +1,7 @@ +use crate::app::compass::config::compass_configuration_error::CompassConfigurationError; use routee_compass_core::util::serde::serde_ops; use serde::{Deserialize, Serialize}; -use crate::{ - app::compass::config::compass_configuration_error::CompassConfigurationError, - plugin::plugin_error::PluginError, -}; - #[derive(Debug, Serialize, Deserialize, Clone)] #[serde(rename_all = "snake_case")] pub enum InjectFormat { @@ -20,19 +16,19 @@ impl InjectFormat { InjectFormat::String => { let decode_result = serde_ops::string_deserialize(value); decode_result.map_err(\|e\| { - CompassConfigurationError::PluginError(PluginError::InputError(format!( + CompassConfigurationError::UserConfigurationError(format!( "could not deserialize inject value as string: {}", e - ))) + )) }) } InjectFormat::Json => { let result = serde_json::from_str(value); result.map_err(\|e\| { - CompassConfigurationError::PluginError(PluginError::InputError(format!( + CompassConfigurationError::UserConfigurationError(format!( "could not deserialize inject value as JSON: {}", e - ))) + )) }) } InjectFormat::Toml => todo!(), diff --git a/rust/routee-compass/src/plugin/input/default/inject/inject_plugin.rs b/rust/routee-compass/src/plugin/input/default/inject/inject_plugin.rs index 7aebc12d..596b3a58 100644 --- a/rust/routee-compass/src/plugin/input/default/inject/inject_plugin.rs +++ b/rust/routee-compass/src/plugin/input/default/inject/inject_plugin.rs @@ -1,18 +1,38 @@ -use crate::plugin::{input::input_plugin::InputPlugin, plugin_error::PluginError}; +use crate::plugin::input::{input_plugin::InputPlugin, InputPluginError}; pub struct InjectInputPlugin { key: String, value: serde_json::Value, + overwrite: bool, } impl InjectInputPlugin { - pub fn new(key: String, value: serde_json::Value) -> InjectInputPlugin { - InjectInputPlugin { key, value } + pub fn new( + key: String, + value: serde_json::Value, + overwrite: Option<bool>, + ) -> InjectInputPlugin { + InjectInputPlugin { + key, + value, + overwrite: overwrite.unwrap_or(true), + } } } impl InputPlugin for InjectInputPlugin { - fn process(&self, input: &mut serde_json::Value) -> Result<(), PluginError> { + fn process(&self, input: &mut serde_json::Value) -> Result<(), InputPluginError> { + if !self.overwrite { + if let Some(obj) = input.as_object() { + if obj.contains_key(&self.key) { + return Err(InputPluginError::InputPluginFailed(format!("inject plugin for key {} has a no-overwrite policy but encountered query with existing value", self.key))); + } + } else { + return Err(InputPluginError::UnexpectedQueryStructure(String::from( + "query is not a JSON object", + ))); + } + } input[self.key.clone()] = self.value.clone(); Ok(()) } diff --git a/rust/routee-compass/src/plugin/input/default/load_balancer/custom_weight_type.rs b/rust/routee-compass/src/plugin/input/default/load_balancer/custom_weight_type.rs index a0fae675..41db2230 100644 --- a/rust/routee-compass/src/plugin/input/default/load_balancer/custom_weight_type.rs +++ b/rust/routee-compass/src/plugin/input/default/load_balancer/custom_weight_type.rs @@ -4,12 +4,11 @@ use serde::{Deserialize, Serialize}; use crate::{ app::compass::config::config_json_extension::ConfigJsonExtensions, - plugin::{input::input_field::InputField, plugin_error::PluginError}, + plugin::input::{input_field::InputField, InputPluginError}, }; #[derive(Debug, Serialize, Deserialize, Clone)] #[serde(rename_all = "snake_case", tag = "type")] - pub enum CustomWeightType { /// a weight value found on each query which can be used directly as it is a numeric /// field. will use provided column_name or fall back to InputField::QueryWeightEstimate. @@ -27,7 +26,7 @@ pub enum CustomWeightType { } impl CustomWeightType { - pub fn get_weight(&self, query: &serde_json::Value) -> Result<f64, PluginError> { + pub fn get_weight(&self, query: &serde_json::Value) -> Result<f64, InputPluginError> { match self { CustomWeightType::Numeric { column_name } => { let col: String = column_name @@ -35,7 +34,12 @@ impl CustomWeightType { .unwrap_or(InputField::QueryWeightEstimate.to_string()); let value = query .get_config_f64(&col, &"custom_weight_type") - .map_err(\|_\| PluginError::ParseError(col, String::from("String")))?; + .map_err(\|_\| { + InputPluginError::QueryFieldHasInvalidType( + InputField::Custom(col), + String::from("String"), + ) + })?; Ok(value) } CustomWeightType::Categorical { @@ -48,11 +52,16 @@ impl CustomWeightType { .unwrap_or(InputField::QueryWeightEstimate.to_string()); let categorical_value = query .get_config_string(&col, &"custom_weight_type") - .map_err(\|_\| PluginError::ParseError(col.clone(), String::from("String")))?; + .map_err(\|_\| { + InputPluginError::QueryFieldHasInvalidType( + InputField::Custom(col), + String::from("String"), + ) + })?; match (mapping.get(&categorical_value), default) { (Some(result), _) => Ok(result), (None, Some(fallback)) => Ok(fallback), - _ => Err(PluginError::InputError(format!("load balancing categorical {} not found in mapping and no default was provided", categorical_value))), + _ => Err(InputPluginError::InputPluginFailed(format!("load balancing categorical {} not found in mapping and no default was provided", categorical_value))), } } } diff --git a/rust/routee-compass/src/plugin/input/default/load_balancer/plugin.rs b/rust/routee-compass/src/plugin/input/default/load_balancer/plugin.rs index 5bf587c2..56c3ae4d 100644 --- a/rust/routee-compass/src/plugin/input/default/load_balancer/plugin.rs +++ b/rust/routee-compass/src/plugin/input/default/load_balancer/plugin.rs @@ -1,14 +1,14 @@ use super::weight_heuristic::WeightHeuristic; use crate::plugin::input::input_json_extensions::InputJsonExtensions; use crate::plugin::input::input_plugin::InputPlugin; -use crate::plugin::plugin_error::PluginError; +use crate::plugin::input::InputPluginError; pub struct LoadBalancerPlugin { pub heuristic: WeightHeuristic, } impl InputPlugin for LoadBalancerPlugin { - fn process(&self, query: &mut serde_json::Value) -> Result<(), PluginError> { + fn process(&self, query: &mut serde_json::Value) -> Result<(), InputPluginError> { let w = self.heuristic.estimate_weight(query)?; let _updated = query.clone(); query.add_query_weight_estimate(w)?; diff --git a/rust/routee-compass/src/plugin/input/default/load_balancer/weight_heuristic.rs b/rust/routee-compass/src/plugin/input/default/load_balancer/weight_heuristic.rs index 77077fad..f3d90786 100644 --- a/rust/routee-compass/src/plugin/input/default/load_balancer/weight_heuristic.rs +++ b/rust/routee-compass/src/plugin/input/default/load_balancer/weight_heuristic.rs @@ -1,5 +1,5 @@ use super::custom_weight_type::CustomWeightType; -use crate::plugin::{input::input_json_extensions::InputJsonExtensions, plugin_error::PluginError}; +use crate::plugin::{input::input_json_extensions::InputJsonExtensions, input::InputPluginError}; use routee_compass_core::model::unit::{as_f64::AsF64, DistanceUnit}; use routee_compass_core::util::geo::haversine; use serde::{Deserialize, Serialize}; @@ -18,19 +18,19 @@ pub enum WeightHeuristic { } impl WeightHeuristic { - pub fn estimate_weight(&self, query: &serde_json::Value) -> Result<f64, PluginError> { + pub fn estimate_weight(&self, query: &serde_json::Value) -> Result<f64, InputPluginError> { match self { WeightHeuristic::Haversine => { let o = query.get_origin_coordinate()?; let d_option = query.get_destination_coordinate()?; match d_option { - None => Err(PluginError::InputError(String::from( + None => Err(InputPluginError::InputPluginFailed(String::from( "cannot estimate search size without destination coordinate", ))), Some(d) => haversine::coord_distance(&o, &d, DistanceUnit::Kilometers) .map(\|d\| d.as_f64()) .map_err(\|s\| { - PluginError::PluginFailed(format!( + InputPluginError::InputPluginFailed(format!( "failed calculating load balancing weight value due to {}", s )) diff --git a/rust/routee-compass/src/plugin/input/default/vertex_rtree/plugin.rs b/rust/routee-compass/src/plugin/input/default/vertex_rtree/plugin.rs index e08c5395..68c94535 100644 --- a/rust/routee-compass/src/plugin/input/default/vertex_rtree/plugin.rs +++ b/rust/routee-compass/src/plugin/input/default/vertex_rtree/plugin.rs @@ -1,12 +1,12 @@ use std::path::Path; -use crate::plugin::input::input_json_extensions::InputJsonExtensions; use crate::plugin::input::input_plugin::InputPlugin; +use crate::plugin::input::{input_json_extensions::InputJsonExtensions, InputPluginError}; use crate::plugin::plugin_error::PluginError; use geo::{coord, Coord}; use routee_compass_core::{ model::unit::{Distance, DistanceUnit, BASE_DISTANCE_UNIT}, - model::{property::vertex::Vertex, road_network::graph::Graph}, + model::{network::graph::Graph, network::Vertex}, util::{fs::read_utils, geo::haversine}, }; use rstar::{PointDistance, RTree, RTreeObject, AABB}; @@ -110,7 +110,9 @@ impl RTreePlugin { distance_unit: Option<DistanceUnit>, ) -> Result<Self, PluginError> { let vertices: Box<[Vertex]> = - read_utils::from_csv(&vertex_file, true, None).map_err(PluginError::CsvReadError)?; + read_utils::from_csv(&vertex_file, true, None).map_err(\|e\| { + InputPluginError::BuildFailed(format!("failure reading vertex file: {}", e)) + })?; let vertex_rtree = VertexRTree::new(vertices.to_vec()); let tolerance = match (tolerance_distance, distance_unit) { (None, None) => None, @@ -136,12 +138,12 @@ impl InputPlugin for RTreePlugin { /// /// * either vertex ids for the nearest coordinates to the the origin (and optionally destination), /// or, an error if not found or not within tolerance - fn process(&self, query: &mut serde_json::Value) -> Result<(), PluginError> { + fn process(&self, query: &mut serde_json::Value) -> Result<(), InputPluginError> { let src_coord = query.get_origin_coordinate()?; let dst_coord_option = query.get_destination_coordinate()?; let src_vertex = self.vertex_rtree.nearest_vertex(src_coord).ok_or_else(\|\| { - PluginError::PluginFailed(format!( + InputPluginError::InputPluginFailed(format!( "nearest vertex not found for origin coordinate {:?}", src_coord )) @@ -154,7 +156,7 @@ impl InputPlugin for RTreePlugin { None => {} Some(dst_coord) => { let dst_vertex = self.vertex_rtree.nearest_vertex(dst_coord).ok_or_else(\|\| { - PluginError::PluginFailed(format!( + InputPluginError::InputPluginFailed(format!( "nearest vertex not found for destination coordinate {:?}", dst_coord )) @@ -187,14 +189,14 @@ fn validate_tolerance( src: &Coord<f32>, dst: &Coord<f32>, tolerance: &Option<(Distance, DistanceUnit)>, -) -> Result<(), PluginError> { +) -> Result<(), InputPluginError> { match tolerance { Some((tolerance_distance, tolerance_distance_unit)) => { - let distance_meters = - haversine::coord_distance_meters(src, dst).map_err(PluginError::PluginFailed)?; + let distance_meters = haversine::coord_distance_meters(src, dst) + .map_err(InputPluginError::InputPluginFailed)?; let distance = DistanceUnit::Meters.convert(&distance_meters, tolerance_distance_unit); if &distance >= tolerance_distance { - Err(PluginError::PluginFailed( + Err(InputPluginError::InputPluginFailed( format!( "coord {:?} nearest vertex coord is {:?} which is {} {} away, exceeding the distance tolerance of {} {}", src, diff --git a/rust/routee-compass/src/plugin/input/input_field.rs b/rust/routee-compass/src/plugin/input/input_field.rs index cb475cca..69e2967c 100644 --- a/rust/routee-compass/src/plugin/input/input_field.rs +++ b/rust/routee-compass/src/plugin/input/input_field.rs @@ -1,5 +1,7 @@ +use serde::{Deserialize, Serialize}; use std::fmt::Display; +#[derive(Debug, Serialize, Deserialize)] pub enum InputField { OriginX, OriginY, @@ -11,10 +13,11 @@ pub enum InputField { DestinationEdge, GridSearch, QueryWeightEstimate, + Custom(String), } impl InputField { - pub fn to_str(&self) -> &'static str { + pub fn to_str(&self) -> &str { use InputField as I; match self { I::OriginX => "origin_x", @@ -27,6 +30,7 @@ impl InputField { I::DestinationEdge => "destination_edge", I::GridSearch => "grid_search", I::QueryWeightEstimate => "query_weight_estimate", + I::Custom(field) => field, } } } diff --git a/rust/routee-compass/src/plugin/input/input_json_extensions.rs b/rust/routee-compass/src/plugin/input/input_json_extensions.rs index b38d7770..48822dbd 100644 --- a/rust/routee-compass/src/plugin/input/input_json_extensions.rs +++ b/rust/routee-compass/src/plugin/input/input_json_extensions.rs @@ -1,70 +1,69 @@ -use super::input_field::InputField; -use crate::plugin::plugin_error::PluginError; +use crate::{app::compass::compass_app_error::CompassAppError, plugin::plugin_error::PluginError}; + +use super::{InputField, InputPluginError}; use geo; -use routee_compass_core::model::road_network::{edge_id::EdgeId, vertex_id::VertexId}; +use routee_compass_core::model::network::{edge_id::EdgeId, vertex_id::VertexId}; use serde_json::{self, json}; pub trait InputJsonExtensions { - fn get_origin_coordinate(&self) -> Result<geo::Coord<f32>, PluginError>; - fn get_destination_coordinate(&self) -> Result<Option<geo::Coord<f32>>, PluginError>; - fn add_origin_vertex(&mut self, vertex_id: VertexId) -> Result<(), PluginError>; - fn add_destination_vertex(&mut self, vertex_id: VertexId) -> Result<(), PluginError>; - fn add_origin_edge(&mut self, edge_id: EdgeId) -> Result<(), PluginError>; - fn add_destination_edge(&mut self, edge_id: EdgeId) -> Result<(), PluginError>; - fn get_origin_vertex(&self) -> Result<VertexId, PluginError>; - fn get_destination_vertex(&self) -> Result<Option<VertexId>, PluginError>; - fn get_origin_edge(&self) -> Result<EdgeId, PluginError>; - fn get_destination_edge(&self) -> Result<Option<EdgeId>, PluginError>; + fn get_origin_coordinate(&self) -> Result<geo::Coord<f32>, InputPluginError>; + fn get_destination_coordinate(&self) -> Result<Option<geo::Coord<f32>>, InputPluginError>; + fn add_origin_vertex(&mut self, vertex_id: VertexId) -> Result<(), InputPluginError>; + fn add_destination_vertex(&mut self, vertex_id: VertexId) -> Result<(), InputPluginError>; + fn add_origin_edge(&mut self, edge_id: EdgeId) -> Result<(), InputPluginError>; + fn add_destination_edge(&mut self, edge_id: EdgeId) -> Result<(), InputPluginError>; + fn get_origin_vertex(&self) -> Result<VertexId, InputPluginError>; + fn get_destination_vertex(&self) -> Result<Option<VertexId>, InputPluginError>; + fn get_origin_edge(&self) -> Result<EdgeId, InputPluginError>; + fn get_destination_edge(&self) -> Result<Option<EdgeId>, InputPluginError>; fn get_grid_search(&self) -> Option<&serde_json::Value>; - fn add_query_weight_estimate(&mut self, weight: f64) -> Result<(), PluginError>; - fn get_query_weight_estimate(&self) -> Result<Option<f64>, PluginError>; + fn add_query_weight_estimate(&mut self, weight: f64) -> Result<(), InputPluginError>; + fn get_query_weight_estimate(&self) -> Result<Option<f64>, CompassAppError>; } impl InputJsonExtensions for serde_json::Value { - fn get_origin_coordinate(&self) -> Result<geo::Coord<f32>, PluginError> { + fn get_origin_coordinate(&self) -> Result<geo::Coord<f32>, InputPluginError> { let origin_x = self - .get(InputField::OriginX.to_string()) - .ok_or_else(\|\| PluginError::MissingField(InputField::OriginX.to_string()))? + .get(InputField::OriginX.to_str()) + .ok_or(InputPluginError::MissingExpectedQueryField( + InputField::OriginX, + ))? .as_f64() .ok_or_else(\|\| { - PluginError::ParseError(InputField::OriginX.to_string(), String::from("f64")) + InputPluginError::QueryFieldHasInvalidType(InputField::OriginX, String::from("f64")) })?; let origin_y = self - .get(InputField::OriginY.to_string()) - .ok_or_else(\|\| PluginError::MissingField(InputField::OriginY.to_string()))? + .get(InputField::OriginY.to_str()) + .ok_or(InputPluginError::MissingExpectedQueryField( + InputField::OriginY, + ))? .as_f64() .ok_or_else(\|\| { - PluginError::ParseError(InputField::OriginY.to_string(), String::from("f64")) + InputPluginError::QueryFieldHasInvalidType(InputField::OriginY, String::from("f64")) })?; Ok(geo::Coord::from((origin_x as f32, origin_y as f32))) } - fn get_destination_coordinate(&self) -> Result<Option<geo::Coord<f32>>, PluginError> { - let x_field = InputField::DestinationX.to_string(); - let y_field = InputField::DestinationY.to_string(); - let x_opt = self.get(&x_field); - let y_opt = self.get(&y_field); + fn get_destination_coordinate(&self) -> Result<Option<geo::Coord<f32>>, InputPluginError> { + let x_field = InputField::DestinationX; + let y_field = InputField::DestinationY; + let x_opt = self.get(x_field.to_str()); + let y_opt = self.get(y_field.to_str()); match (x_opt, y_opt) { (None, None) => Ok(None), - (None, Some(_)) => Err(PluginError::MissingField(format!( - "{} provided without {}", - &y_field, &x_field - ))), - (Some(_), None) => Err(PluginError::MissingField(format!( - "{} provided without {}", - &x_field, &y_field - ))), + (None, Some(_)) => Err(InputPluginError::MissingQueryFieldPair(y_field, x_field)), + (Some(_), None) => Err(InputPluginError::MissingQueryFieldPair(x_field, y_field)), (Some(x_json), Some(y_json)) => { - let x = x_json - .as_f64() - .ok_or_else(\|\| PluginError::ParseError(x_field.clone(), String::from("f64")))?; - let y = y_json - .as_f64() - .ok_or_else(\|\| PluginError::ParseError(y_field.clone(), String::from("f64")))?; + let x = x_json.as_f64().ok_or_else(\|\| { + InputPluginError::QueryFieldHasInvalidType(x_field, String::from("f64")) + })?; + let y = y_json.as_f64().ok_or_else(\|\| { + InputPluginError::QueryFieldHasInvalidType(y_field, String::from("f64")) + })?; Ok(Some(geo::Coord::from((x as f32, y as f32)))) } } } - fn add_origin_vertex(&mut self, vertex_id: VertexId) -> Result<(), PluginError> { + fn add_origin_vertex(&mut self, vertex_id: VertexId) -> Result<(), InputPluginError> { match self { serde_json::Value::Object(map) => { map.insert( @@ -73,12 +72,12 @@ impl InputJsonExtensions for serde_json::Value { ); Ok(()) } - _ => Err(PluginError::InputError(String::from( + _ => Err(InputPluginError::UnexpectedQueryStructure(String::from( "InputQuery is not a JSON object", ))), } } - fn add_destination_vertex(&mut self, vertex_id: VertexId) -> Result<(), PluginError> { + fn add_destination_vertex(&mut self, vertex_id: VertexId) -> Result<(), InputPluginError> { match self { serde_json::Value::Object(map) => { map.insert( @@ -87,60 +86,73 @@ impl InputJsonExtensions for serde_json::Value { ); Ok(()) } - _ => Err(PluginError::InputError(String::from( + _ => Err(InputPluginError::UnexpectedQueryStructure(String::from( "InputQuery is not a JSON object", ))), } } - fn get_origin_vertex(&self) -> Result<VertexId, PluginError> { - self.get(InputField::OriginVertex.to_string()) - .ok_or_else(\|\| PluginError::MissingField(InputField::OriginVertex.to_string()))? + fn get_origin_vertex(&self) -> Result<VertexId, InputPluginError> { + self.get(InputField::OriginVertex.to_str()) + .ok_or(InputPluginError::MissingExpectedQueryField( + InputField::OriginVertex, + ))? .as_u64() .map(\|v\| VertexId(v as usize)) .ok_or_else(\|\| { - PluginError::ParseError(InputField::OriginVertex.to_string(), String::from("u64")) + InputPluginError::QueryFieldHasInvalidType( + InputField::OriginVertex, + String::from("u64"), + ) }) } - fn get_destination_vertex(&self) -> Result<Option<VertexId>, PluginError> { - match self.get(InputField::DestinationVertex.to_string()) { + fn get_destination_vertex(&self) -> Result<Option<VertexId>, InputPluginError> { + match self.get(InputField::DestinationVertex.to_str()) { None => Ok(None), Some(v) => v .as_u64() .map(\|v\| Some(VertexId(v as usize))) .ok_or_else(\|\| { - PluginError::ParseError( - InputField::DestinationVertex.to_string(), + InputPluginError::QueryFieldHasInvalidType( + InputField::DestinationVertex, String::from("u64"), ) }), } } - fn get_origin_edge(&self) -> Result<EdgeId, PluginError> { - self.get(InputField::OriginEdge.to_string()) - .ok_or_else(\|\| PluginError::MissingField(InputField::OriginEdge.to_string()))? + fn get_origin_edge(&self) -> Result<EdgeId, InputPluginError> { + self.get(InputField::OriginEdge.to_str()) + .ok_or(InputPluginError::MissingExpectedQueryField( + InputField::OriginEdge, + ))? .as_u64() .map(\|v\| EdgeId(v as usize)) .ok_or_else(\|\| { - PluginError::ParseError(InputField::OriginEdge.to_string(), String::from("u64")) + InputPluginError::QueryFieldHasInvalidType( + InputField::OriginEdge, + String::from("u64"), + ) }) } - fn get_destination_edge(&self) -> Result<Option<EdgeId>, PluginError> { - match self.get(InputField::DestinationEdge.to_string()) { + fn get_destination_edge(&self) -> Result<Option<EdgeId>, InputPluginError> { + match self.get(InputField::DestinationEdge.to_str()) { None => Ok(None), Some(v) => v.as_u64().map(\|v\| Some(EdgeId(v as usize))).ok_or_else(\|\| { - PluginError::ParseError(InputField::OriginEdge.to_string(), String::from("u64")) + InputPluginError::QueryFieldHasInvalidType( + InputField::OriginEdge, + String::from("u64"), + ) }), } } fn get_grid_search(&self) -> Option<&serde_json::Value> { - self.get(InputField::GridSearch.to_string()) + self.get(InputField::GridSearch.to_str()) } - fn add_origin_edge(&mut self, edge_id: EdgeId) -> Result<(), PluginError> { + fn add_origin_edge(&mut self, edge_id: EdgeId) -> Result<(), InputPluginError> { match self { serde_json::Value::Object(map) => { map.insert( @@ -149,13 +161,13 @@ impl InputJsonExtensions for serde_json::Value { ); Ok(()) } - _ => Err(PluginError::InputError(String::from( + _ => Err(InputPluginError::UnexpectedQueryStructure(String::from( "InputQuery is not a JSON object", ))), } } - fn add_destination_edge(&mut self, edge_id: EdgeId) -> Result<(), PluginError> { + fn add_destination_edge(&mut self, edge_id: EdgeId) -> Result<(), InputPluginError> { match self { serde_json::Value::Object(map) => { map.insert( @@ -164,32 +176,34 @@ impl InputJsonExtensions for serde_json::Value { ); Ok(()) } - _ => Err(PluginError::InputError(String::from( + _ => Err(InputPluginError::UnexpectedQueryStructure(String::from( "InputQuery is not a JSON object", ))), } } - fn add_query_weight_estimate(&mut self, weight: f64) -> Result<(), PluginError> { + fn add_query_weight_estimate(&mut self, weight: f64) -> Result<(), InputPluginError> { match self { serde_json::Value::Object(map) => { map.insert(InputField::QueryWeightEstimate.to_string(), json!(weight)); Ok(()) } - _ => Err(PluginError::InputError(String::from( + _ => Err(InputPluginError::UnexpectedQueryStructure(String::from( "InputQuery is not a JSON object", ))), } } - fn get_query_weight_estimate(&self) -> Result<Option<f64>, PluginError> { - match self.get(InputField::QueryWeightEstimate.to_string()) { + fn get_query_weight_estimate(&self) -> Result<Option<f64>, CompassAppError> { + match self.get(InputField::QueryWeightEstimate.to_str()) { None => Ok(None), Some(v) => v.as_f64().map(Some).ok_or_else(\|\| { - PluginError::ParseError( - InputField::QueryWeightEstimate.to_string(), + let ipe = InputPluginError::QueryFieldHasInvalidType( + InputField::QueryWeightEstimate, String::from("f64"), - ) + ); + let pe = PluginError::InputPluginFailed { source: ipe }; + CompassAppError::PluginError(pe) }), } } @@ -201,16 +215,16 @@ impl InputJsonExtensions for serde_json::Value { // value: &serde_json::Value, // field: InputField, // op: DecodeOp<T>, -// ) -> Result<T, PluginError> { +// ) -> Result<T, InputPluginError> { // let at_field = value.get(field.to_string()); // match at_field { -// None => Err(PluginError::MissingField(field.to_string())), -// Some(v) => op(v).ok_or_else( \|\|PluginError::ParseError(field.to_string(), ())), +// None => Err(InputPluginError::MissingField(field.to_string())), +// Some(v) => op(v).ok_or_else( \|\|InputPluginError::QueryFieldHasInvalidType(field.to_string(), ())), // }; // } -// fn get_f64(v: &serde_json::Value) -> Result<f64, PluginError> { +// fn get_f64(v: &serde_json::Value) -> Result<f64, InputPluginError> { // get_from_json(v, field, \|v\| v.as_f64()) -// v.as_f64().ok_or_else( \|\|PluginError::ParseError((), ()) +// v.as_f64().ok_or_else( \|\|InputPluginError::QueryFieldHasInvalidType((), ()) // } diff --git a/rust/routee-compass/src/plugin/input/input_plugin.rs b/rust/routee-compass/src/plugin/input/input_plugin.rs index 49cfb748..401b9dd9 100644 --- a/rust/routee-compass/src/plugin/input/input_plugin.rs +++ b/rust/routee-compass/src/plugin/input/input_plugin.rs @@ -1,4 +1,4 @@ -use crate::plugin::plugin_error::PluginError; +use super::InputPluginError; /// Performs some kind of pre-processing on a user query input. The input JSON is available /// to the plugin as a reference. The plugin produces a vector of zero to many JSON objects that will @@ -26,5 +26,5 @@ pub trait InputPlugin: Send + Sync { /// # Returns /// /// A `Vec` of JSON values to replace the input JSON, or an error - fn process(&self, input: &mut serde_json::Value) -> Result<(), PluginError>; + fn process(&self, input: &mut serde_json::Value) -> Result<(), InputPluginError>; } diff --git a/rust/routee-compass/src/plugin/input/input_plugin_error.rs b/rust/routee-compass/src/plugin/input/input_plugin_error.rs new file mode 100644 index 00000000..a0bb4bc9 --- /dev/null +++ b/rust/routee-compass/src/plugin/input/input_plugin_error.rs @@ -0,0 +1,24 @@ +use super::input_field::InputField; + +#[derive(thiserror::Error, Debug)] +pub enum InputPluginError { + #[error("failure building input plugin: {0}")] + BuildFailed(String), + #[error("required query field '{0}' not found")] + MissingExpectedQueryField(InputField), + #[error("{0} provided without {1}")] + MissingQueryFieldPair(InputField, InputField), + #[error("required query field '{0}' is not of type {1}")] + QueryFieldHasInvalidType(InputField, String), + #[error("expected query to be a json object '{{}}' but found {0}")] + UnexpectedQueryStructure(String), + #[error("plugin experienced JSON error: {source}")] + JsonError { + #[from] + source: serde_json::Error, + }, + #[error("failure running plugin: {0}")] + InputPluginFailed(String), + #[error("unexpected error: {0}")] + InternalError(String), +} diff --git a/rust/routee-compass/src/plugin/input/input_plugin_ops.rs b/rust/routee-compass/src/plugin/input/input_plugin_ops.rs index f2cceb96..a94eb008 100644 --- a/rust/routee-compass/src/plugin/input/input_plugin_ops.rs +++ b/rust/routee-compass/src/plugin/input/input_plugin_ops.rs @@ -1,8 +1,8 @@ -use std::rc::Rc; - -use crate::plugin::plugin_error::PluginError; use indoc::indoc; use serde_json::{json, Value}; +use std::rc::Rc; + +use super::InputPluginError; /// helper to return errors as JSON response objects which include the /// original request along with the error message @@ -62,13 +62,13 @@ pub fn package_invariant_error( } } -pub type ArrayOp<'a> = Rc<dyn Fn(&mut Value) -> Result<(), PluginError> + 'a>; +pub type InputArrayOp<'a> = Rc<dyn Fn(&mut Value) -> Result<(), InputPluginError> + 'a>; /// executes an operation on an input query. maintains the invariant that /// input queries should always remain wrapped in a top-level JSON Array /// so that we can perform operations like grid search, which transform a /// single query into multiple child queries. -pub fn json_array_op<'a>(query: &'a mut Value, op: ArrayOp<'a>) -> Result<(), Value> { +pub fn json_array_op<'a>(query: &'a mut Value, op: InputArrayOp<'a>) -> Result<(), Value> { match query { Value::Array(queries) => { for q in queries.iter_mut() { diff --git a/rust/routee-compass/src/plugin/input/mod.rs b/rust/routee-compass/src/plugin/input/mod.rs index 2dcabdb5..bd438a45 100644 --- a/rust/routee-compass/src/plugin/input/mod.rs +++ b/rust/routee-compass/src/plugin/input/mod.rs @@ -2,4 +2,8 @@ pub mod default; pub mod input_field; pub mod input_json_extensions; pub mod input_plugin; +pub mod input_plugin_error; pub mod input_plugin_ops; + +pub use input_field::InputField; +pub use input_plugin_error::InputPluginError; diff --git a/rust/routee-compass/src/plugin/output/default/summary/plugin.rs b/rust/routee-compass/src/plugin/output/default/summary/plugin.rs index c6375bea..4a66479f 100644 --- a/rust/routee-compass/src/plugin/output/default/summary/plugin.rs +++ b/rust/routee-compass/src/plugin/output/default/summary/plugin.rs @@ -2,7 +2,7 @@ use crate::app::{ compass::compass_app_error::CompassAppError, search::search_app_result::SearchAppResult, }; use crate::plugin::output::output_plugin::OutputPlugin; -use crate::plugin::plugin_error::PluginError; +use crate::plugin::output::OutputPluginError; use routee_compass_core::algorithm::search::search_instance::SearchInstance; use routee_compass_core::util::duration_extension::DurationExtension; use serde_json::{self, json}; @@ -16,7 +16,7 @@ impl OutputPlugin for SummaryOutputPlugin { &self, output: &mut serde_json::Value, search_result: &Result<(SearchAppResult, SearchInstance), CompassAppError>, - ) -> Result<(), PluginError> { + ) -> Result<(), OutputPluginError> { match search_result { Err(_e) => Ok(()), Ok((result, _)) => { diff --git a/rust/routee-compass/src/plugin/output/default/traversal/builder.rs b/rust/routee-compass/src/plugin/output/default/traversal/builder.rs index 4911472c..a5036d1a 100644 --- a/rust/routee-compass/src/plugin/output/default/traversal/builder.rs +++ b/rust/routee-compass/src/plugin/output/default/traversal/builder.rs @@ -6,7 +6,7 @@ use crate::{ builders::OutputPluginBuilder, compass_configuration_error::CompassConfigurationError, config_json_extension::ConfigJsonExtensions, }, - plugin::output::output_plugin::OutputPlugin, + plugin::{output::output_plugin::OutputPlugin, plugin_error::PluginError}, }; /// Builds a plugin that can generate traversal outputs. @@ -49,7 +49,8 @@ impl OutputPluginBuilder for TraversalPluginBuilder { let tree: Option<TraversalOutputFormat> = parameters.get_config_serde_optional(&"tree", &parent_key)?; - let geom_plugin = TraversalPlugin::from_file(&geometry_filename, route, tree)?; + let geom_plugin = TraversalPlugin::from_file(&geometry_filename, route, tree) + .map_err(\|e\| PluginError::OutputPluginFailed { source: e })?; Ok(Arc::new(geom_plugin)) } } diff --git a/rust/routee-compass/src/plugin/output/default/traversal/json_extensions.rs b/rust/routee-compass/src/plugin/output/default/traversal/json_extensions.rs index bd0d24fd..d3069ba5 100644 --- a/rust/routee-compass/src/plugin/output/default/traversal/json_extensions.rs +++ b/rust/routee-compass/src/plugin/output/default/traversal/json_extensions.rs @@ -1,7 +1,6 @@ +use crate::plugin::{input::InputField, output::OutputPluginError}; use std::fmt::Display; -use crate::plugin::plugin_error::PluginError; - pub enum TraversalJsonField { RouteOutput, TreeOutput, @@ -23,18 +22,22 @@ impl Display for TraversalJsonField { } pub trait TraversalJsonExtensions { - fn get_route_geometry_wkt(&self) -> Result<String, PluginError>; + fn get_route_geometry_wkt(&self) -> Result<String, OutputPluginError>; } impl TraversalJsonExtensions for serde_json::Value { - fn get_route_geometry_wkt(&self) -> Result<String, PluginError> { + fn get_route_geometry_wkt(&self) -> Result<String, OutputPluginError> { let geometry = self .get(TraversalJsonField::RouteOutput.as_str()) - .ok_or_else(\|\| PluginError::MissingField(TraversalJsonField::RouteOutput.to_string()))? - .as_str() .ok_or_else(\|\| { - PluginError::ParseError( + OutputPluginError::MissingExpectedQueryField(InputField::Custom( TraversalJsonField::RouteOutput.to_string(), + )) + })? + .as_str() + .ok_or_else(\|\| { + OutputPluginError::QueryFieldHasInvalidType( + InputField::Custom(TraversalJsonField::RouteOutput.to_string()), String::from("string"), ) })? diff --git a/rust/routee-compass/src/plugin/output/default/traversal/plugin.rs b/rust/routee-compass/src/plugin/output/default/traversal/plugin.rs index ef44ebb3..45529bc1 100644 --- a/rust/routee-compass/src/plugin/output/default/traversal/plugin.rs +++ b/rust/routee-compass/src/plugin/output/default/traversal/plugin.rs @@ -3,7 +3,7 @@ use super::traversal_output_format::TraversalOutputFormat; use crate::app::compass::compass_app_error::CompassAppError; use crate::app::search::search_app_result::SearchAppResult; use crate::plugin::output::output_plugin::OutputPlugin; -use crate::plugin::plugin_error::PluginError; +use crate::plugin::output::OutputPluginError; use geo::LineString; use kdam::Bar; use kdam::BarExt; @@ -28,9 +28,13 @@ impl TraversalPlugin { filename: &P, route: Option<TraversalOutputFormat>, tree: Option<TraversalOutputFormat>, - ) -> Result<TraversalPlugin, PluginError> { + ) -> Result<TraversalPlugin, OutputPluginError> { let count = fs_utils::line_count(filename, fs_utils::is_gzip(filename)).map_err(\|e\| { - PluginError::FileReadError(filename.as_ref().to_path_buf(), e.to_string()) + OutputPluginError::BuildFailed(format!( + "failure reading line count for file {}: {}", + filename.as_ref().to_str().unwrap_or_default(), + e + )) })?; let mut pb = Bar::builder() @@ -38,16 +42,20 @@ impl TraversalPlugin { .animation("fillup") .desc("geometry file") .build() - .map_err(PluginError::InternalError)?; + .map_err(OutputPluginError::InternalError)?; let cb = Box::new(\|\| { let _ = pb.update(1); }); let geoms = read_raw_file(filename, geo_io_utils::parse_wkt_linestring, Some(cb)).map_err(\|e\| { - PluginError::FileReadError(filename.as_ref().to_path_buf(), e.to_string()) + OutputPluginError::BuildFailed(format!( + "failure reading geometry file {}: {}", + filename.as_ref().to_str().unwrap_or_default(), + e + )) })?; - println!(); + eprintln!(); let route_key = TraversalJsonField::RouteOutput.to_string(); let tree_key = TraversalJsonField::TreeOutput.to_string(); @@ -66,7 +74,7 @@ impl OutputPlugin for TraversalPlugin { &self, output: &mut serde_json::Value, search_result: &Result<(SearchAppResult, SearchInstance), CompassAppError>, - ) -> Result<(), PluginError> { + ) -> Result<(), OutputPluginError> { match search_result { Err(_) => Ok(()), Ok((result, si)) => { @@ -80,7 +88,7 @@ impl OutputPlugin for TraversalPlugin { construct_route_output(route, si, &route_args, &self.geoms) }) .collect::<Result<Vec<_>, _>>() - .map_err(PluginError::PluginFailed)?; + .map_err(OutputPluginError::OutputPluginFailed)?; // vary the type of value stored at the route key. if there is // no route, store 'null'. if one, store an output object. if diff --git a/rust/routee-compass/src/plugin/output/default/traversal/traversal_ops.rs b/rust/routee-compass/src/plugin/output/default/traversal/traversal_ops.rs index 8960dffd..9d904090 100644 --- a/rust/routee-compass/src/plugin/output/default/traversal/traversal_ops.rs +++ b/rust/routee-compass/src/plugin/output/default/traversal/traversal_ops.rs @@ -1,30 +1,35 @@ -use crate::plugin::plugin_error::PluginError; +use crate::plugin::output::OutputPluginError; use geo::{LineString, MultiLineString, Point}; use geo_types::MultiPoint; use geojson::feature::Id; use geojson::{Feature, FeatureCollection}; use routee_compass_core::algorithm::search::edge_traversal::EdgeTraversal; use routee_compass_core::algorithm::search::search_tree_branch::SearchTreeBranch; -use routee_compass_core::model::road_network::vertex_id::VertexId; +use routee_compass_core::model::network::vertex_id::VertexId; use routee_compass_core::util::geo::geo_io_utils; use std::collections::HashMap; pub fn create_tree_geojson( tree: &HashMap<VertexId, SearchTreeBranch>, geoms: &[LineString<f32>], -) -> Result<serde_json::Value, PluginError> { +) -> Result<serde_json::Value, OutputPluginError> { let features = tree .values() .map(\|t\| { let row_result = geoms .get(t.edge_traversal.edge_id.0) .cloned() - .ok_or_else(\|\| PluginError::EdgeGeometryMissing(t.edge_traversal.edge_id)) + .ok_or_else(\|\| { + OutputPluginError::OutputPluginFailed(format!( + "geometry table missing edge id {}", + t.edge_traversal.edge_id + )) + }) .and_then(\|g\| create_geojson_feature(&t.edge_traversal, g)); row_result }) - .collect::<Result<Vec<_>, PluginError>>()?; + .collect::<Result<Vec<_>, OutputPluginError>>()?; // let result_json = serde_json::to_value(features)?;/ let feature_collection = FeatureCollection { bbox: None, @@ -38,19 +43,24 @@ pub fn create_tree_geojson( pub fn create_route_geojson( route: &[EdgeTraversal], geoms: &[LineString<f32>], -) -> Result<serde_json::Value, PluginError> { +) -> Result<serde_json::Value, OutputPluginError> { let features = route .iter() .map(\|t\| { let row_result = geoms .get(t.edge_id.0) .cloned() - .ok_or_else(\|\| PluginError::EdgeGeometryMissing(t.edge_id)) + .ok_or_else(\|\| { + OutputPluginError::OutputPluginFailed(format!( + "geometry table missing edge id {}", + t.edge_id + )) + }) .and_then(\|g\| create_geojson_feature(t, g)); row_result }) - .collect::<Result<Vec<_>, PluginError>>()?; + .collect::<Result<Vec<_>, OutputPluginError>>()?; // let result_json = serde_json::to_value(features)?;/ let feature_collection = FeatureCollection { bbox: None, @@ -64,14 +74,14 @@ pub fn create_route_geojson( pub fn create_geojson_feature( t: &EdgeTraversal, g: LineString<f32>, -) -> Result<Feature, PluginError> { +) -> Result<Feature, OutputPluginError> { let props = match serde_json::to_value(t).map(\|v\| v.as_object().cloned()) { - Ok(None) => Err(PluginError::InternalError(format!( + Ok(None) => Err(OutputPluginError::InternalError(format!( "serialized EdgeTraversal was not a JSON object for {}", t ))), Ok(Some(obj)) => Ok(obj), - Err(err) => Err(PluginError::JsonError(err)), + Err(err) => Err(OutputPluginError::JsonError { source: err }), }?; let id = Id::Number(serde_json::Number::from(t.edge_id.0)); @@ -89,24 +99,26 @@ pub fn create_geojson_feature( pub fn create_edge_geometry( edge: &EdgeTraversal, geoms: &[LineString<f32>], -) -> Result<LineString<f32>, PluginError> { - geoms - .get(edge.edge_id.0) - .cloned() - .ok_or_else(\|\| PluginError::EdgeGeometryMissing(edge.edge_id)) +) -> Result<LineString<f32>, OutputPluginError> { + geoms.get(edge.edge_id.0).cloned().ok_or_else(\|\| { + OutputPluginError::OutputPluginFailed(format!( + "geometry table missing edge id {}", + edge.edge_id + )) + }) } pub fn create_branch_geometry( branch: &SearchTreeBranch, geoms: &[LineString<f32>], -) -> Result<LineString<f32>, PluginError> { +) -> Result<LineString<f32>, OutputPluginError> { create_edge_geometry(&branch.edge_traversal, geoms) } pub fn create_route_linestring( route: &[EdgeTraversal], geoms: &[LineString<f32>], -) -> Result<LineString<f32>, PluginError> { +) -> Result<LineString<f32>, OutputPluginError> { let edge_ids = route .iter() .map(\|traversal\| traversal.edge_id) @@ -115,12 +127,15 @@ pub fn create_route_linestring( let edge_linestrings = edge_ids .iter() .map(\|eid\| { - let geom = geoms - .get(eid.0) - .ok_or_else(\|\| PluginError::EdgeGeometryMissing(eid)); + let geom = geoms.get(eid.0).ok_or_else(\|\| { + OutputPluginError::OutputPluginFailed(format!( + "geometry table missing edge id {}", + eid + )) + }); geom }) - .collect::<Result<Vec<&LineString<f32>>, PluginError>>()?; + .collect::<Result<Vec<&LineString<f32>>, OutputPluginError>>()?; let geometry = geo_io_utils::concat_linestrings(edge_linestrings); Ok(geometry) } @@ -128,7 +143,7 @@ pub fn create_route_linestring( pub fn create_tree_multilinestring( tree: &HashMap<VertexId, SearchTreeBranch>, geoms: &[LineString<f32>], -) -> Result<MultiLineString<f32>, PluginError> { +) -> Result<MultiLineString<f32>, OutputPluginError> { let edge_ids = tree .values() .map(\|traversal\| traversal.edge_traversal.edge_id) @@ -137,12 +152,15 @@ pub fn create_tree_multilinestring( let tree_linestrings = edge_ids .iter() .map(\|eid\| { - let geom = geoms - .get(eid.0) - .ok_or_else(\|\| PluginError::EdgeGeometryMissing(eid)); + let geom = geoms.get(eid.0).ok_or_else(\|\| { + OutputPluginError::OutputPluginFailed(format!( + "geometry table missing edge id {}", + eid + )) + }); geom.cloned() }) - .collect::<Result<Vec<LineString<f32>>, PluginError>>()?; + .collect::<Result<Vec<LineString<f32>>, OutputPluginError>>()?; let geometry = MultiLineString::new(tree_linestrings); Ok(geometry) } @@ -150,7 +168,7 @@ pub fn create_tree_multilinestring( pub fn create_tree_multipoint( tree: &HashMap<VertexId, SearchTreeBranch>, geoms: &[LineString<f64>], -) -> Result<MultiPoint, PluginError> { +) -> Result<MultiPoint, OutputPluginError> { let edge_ids = tree .values() .map(\|traversal\| traversal.edge_traversal.edge_id) @@ -161,10 +179,15 @@ pub fn create_tree_multipoint( .map(\|eid\| { let geom = geoms .get(eid.0) - .ok_or_else(\|\| PluginError::EdgeGeometryMissing(eid)) + .ok_or_else(\|\| { + OutputPluginError::OutputPluginFailed(format!( + "geometry table missing edge id {}", + eid + )) + }) .map(\|l\| { l.points().last().ok_or_else(\|\| { - PluginError::InputError(format!( + OutputPluginError::OutputPluginFailed(format!( "linestring is invalid for edge_id {}", eid )) @@ -177,7 +200,7 @@ pub fn create_tree_multipoint( Err(e) => Err(e), } }) - .collect::<Result<Vec<Point>, PluginError>>()?; + .collect::<Result<Vec<Point>, OutputPluginError>>()?; let geometry = MultiPoint::new(tree_destinations); Ok(geometry) } diff --git a/rust/routee-compass/src/plugin/output/default/traversal/traversal_output_format.rs b/rust/routee-compass/src/plugin/output/default/traversal/traversal_output_format.rs index bed69658..3be5b340 100644 --- a/rust/routee-compass/src/plugin/output/default/traversal/traversal_output_format.rs +++ b/rust/routee-compass/src/plugin/output/default/traversal/traversal_output_format.rs @@ -1,11 +1,11 @@ use std::collections::HashMap; use super::traversal_ops as ops; -use crate::plugin::plugin_error::PluginError; +use crate::plugin::output::OutputPluginError; use geo::{CoordFloat, Geometry, LineString}; use routee_compass_core::{ algorithm::search::{edge_traversal::EdgeTraversal, search_tree_branch::SearchTreeBranch}, - model::road_network::vertex_id::VertexId, + model::network::vertex_id::VertexId, }; use serde::{Deserialize, Serialize}; use wkt::ToWkt; @@ -30,7 +30,7 @@ impl TraversalOutputFormat { &self, route: &Vec<EdgeTraversal>, geoms: &[LineString<f32>], - ) -> Result<serde_json::Value, PluginError> { + ) -> Result<serde_json::Value, OutputPluginError> { match self { TraversalOutputFormat::Wkt => { let route_geometry = ops::create_route_linestring(route, geoms)?; @@ -64,7 +64,7 @@ impl TraversalOutputFormat { &self, tree: &HashMap<VertexId, SearchTreeBranch>, geoms: &[LineString<f32>], - ) -> Result<serde_json::Value, PluginError> { + ) -> Result<serde_json::Value, OutputPluginError> { match self { TraversalOutputFormat::Wkt => { let route_geometry = ops::create_tree_multilinestring(tree, geoms)?; @@ -99,9 +99,9 @@ impl TraversalOutputFormat { fn geometry_to_wkb_string<T: CoordFloat + Into<f64>>( geometry: &Geometry<T>, -) -> Result<String, PluginError> { +) -> Result<String, OutputPluginError> { let bytes = wkb::geom_to_wkb(geometry).map_err(\|e\| { - PluginError::PluginFailed(format!( + OutputPluginError::OutputPluginFailed(format!( "failed to generate wkb for geometry '{:?}' - {:?}", geometry, e )) @@ -122,9 +122,7 @@ mod test { use geo::{coord, LineString}; use routee_compass_core::{ algorithm::search::edge_traversal::EdgeTraversal, - model::{ - road_network::edge_id::EdgeId, traversal::state::state_variable::StateVar, unit::Cost, - }, + model::{network::edge_id::EdgeId, traversal::state::state_variable::StateVar, unit::Cost}, }; use std::time::Duration; diff --git a/rust/routee-compass/src/plugin/output/default/uuid/builder.rs b/rust/routee-compass/src/plugin/output/default/uuid/builder.rs index c20515e2..820d5c5c 100644 --- a/rust/routee-compass/src/plugin/output/default/uuid/builder.rs +++ b/rust/routee-compass/src/plugin/output/default/uuid/builder.rs @@ -5,7 +5,7 @@ use crate::{ builders::OutputPluginBuilder, compass_configuration_error::CompassConfigurationError, config_json_extension::ConfigJsonExtensions, }, - plugin::output::output_plugin::OutputPlugin, + plugin::{output::output_plugin::OutputPlugin, plugin_error::PluginError}, }; use super::plugin::UUIDOutputPlugin; @@ -19,8 +19,10 @@ impl OutputPluginBuilder for UUIDOutputPluginBuilder { ) -> Result<Arc<dyn OutputPlugin>, CompassConfigurationError> { let uuid_filename = parameters.get_config_path(&"uuid_input_file", &"uuid")?; - let uuid_plugin = UUIDOutputPlugin::from_file(&uuid_filename) - .map_err(CompassConfigurationError::PluginError)?; + let uuid_plugin = UUIDOutputPlugin::from_file(&uuid_filename).map_err(\|e\| { + let pe = PluginError::OutputPluginFailed { source: e }; + CompassConfigurationError::PluginError(pe) + })?; Ok(Arc::new(uuid_plugin)) } } diff --git a/rust/routee-compass/src/plugin/output/default/uuid/mod.rs b/rust/routee-compass/src/plugin/output/default/uuid/mod.rs index 668d9c48..fd557c7b 100644 --- a/rust/routee-compass/src/plugin/output/default/uuid/mod.rs +++ b/rust/routee-compass/src/plugin/output/default/uuid/mod.rs @@ -1,3 +1,3 @@ pub mod builder; -pub mod json_extensions; +pub mod output_json_extensions; pub mod plugin; diff --git a/rust/routee-compass/src/plugin/output/default/uuid/json_extensions.rs b/rust/routee-compass/src/plugin/output/default/uuid/output_json_extensions.rs similarity index 64% rename from rust/routee-compass/src/plugin/output/default/uuid/json_extensions.rs rename to rust/routee-compass/src/plugin/output/default/uuid/output_json_extensions.rs index 7943e8bc..9a671d12 100644 --- a/rust/routee-compass/src/plugin/output/default/uuid/json_extensions.rs +++ b/rust/routee-compass/src/plugin/output/default/uuid/output_json_extensions.rs @@ -1,8 +1,8 @@ use std::fmt::Display; -use routee_compass_core::model::road_network::vertex_id::VertexId; +use routee_compass_core::model::network::vertex_id::VertexId; -use crate::plugin::plugin_error::PluginError; +use crate::plugin::{input::InputField, output::OutputPluginError}; pub enum UUIDJsonField { Request, @@ -12,6 +12,12 @@ pub enum UUIDJsonField { DestinationVertexUUID, } +impl From<UUIDJsonField> for InputField { + fn from(value: UUIDJsonField) -> Self { + InputField::Custom(value.to_string()) + } +} + impl UUIDJsonField { pub fn as_str(&self) -> &'static str { match self { @@ -31,46 +37,52 @@ impl Display for UUIDJsonField { } pub trait UUIDJsonExtensions { - fn get_od_vertex_ids(&self) -> Result<(VertexId, VertexId), PluginError>; + fn get_od_vertex_ids(&self) -> Result<(VertexId, VertexId), OutputPluginError>; fn add_od_uuids( &mut self, origin_uuid: String, destination_uuid: String, - ) -> Result<(), PluginError>; + ) -> Result<(), OutputPluginError>; } impl UUIDJsonExtensions for serde_json::Value { - fn get_od_vertex_ids(&self) -> Result<(VertexId, VertexId), PluginError> { + fn get_od_vertex_ids(&self) -> Result<(VertexId, VertexId), OutputPluginError> { let request = self .get(UUIDJsonField::Request.as_str()) - .ok_or_else(\|\| PluginError::MissingField(UUIDJsonField::Request.to_string()))? + .ok_or_else(\|\| { + OutputPluginError::MissingExpectedQueryField(UUIDJsonField::Request.into()) + })? .as_object() .ok_or_else(\|\| { - PluginError::ParseError( - UUIDJsonField::Request.to_string(), + OutputPluginError::QueryFieldHasInvalidType( + UUIDJsonField::Request.into(), String::from("json object"), ) })?; let origin_vertex_id = request .get(&UUIDJsonField::OriginVertexId.to_string()) - .ok_or_else(\|\| PluginError::MissingField(UUIDJsonField::OriginVertexId.to_string()))? + .ok_or_else(\|\| { + OutputPluginError::MissingExpectedQueryField(UUIDJsonField::OriginVertexId.into()) + })? .as_u64() .ok_or_else(\|\| { - PluginError::ParseError( - UUIDJsonField::OriginVertexId.to_string(), + OutputPluginError::QueryFieldHasInvalidType( + UUIDJsonField::OriginVertexId.into(), String::from("u64"), ) })?; let destination_vertex_id = request .get(&UUIDJsonField::DestinationVertexId.to_string()) .ok_or_else(\|\| { - PluginError::MissingField(UUIDJsonField::DestinationVertexId.to_string()) + OutputPluginError::MissingExpectedQueryField( + UUIDJsonField::DestinationVertexId.into(), + ) })? .as_u64() .ok_or_else(\|\| { - PluginError::ParseError( - UUIDJsonField::DestinationVertexId.to_string(), + OutputPluginError::QueryFieldHasInvalidType( + UUIDJsonField::DestinationVertexId.into(), String::from("u64"), ) })?; @@ -83,14 +95,16 @@ impl UUIDJsonExtensions for serde_json::Value { &mut self, origin_uuid: String, destination_uuid: String, - ) -> Result<(), PluginError> { + ) -> Result<(), OutputPluginError> { let request = self .get_mut(UUIDJsonField::Request.as_str()) - .ok_or_else(\|\| PluginError::MissingField(UUIDJsonField::Request.to_string()))? + .ok_or_else(\|\| { + OutputPluginError::MissingExpectedQueryField(UUIDJsonField::Request.into()) + })? .as_object_mut() .ok_or_else(\|\| { - PluginError::ParseError( - UUIDJsonField::Request.to_string(), + OutputPluginError::QueryFieldHasInvalidType( + UUIDJsonField::Request.into(), String::from("json object"), ) })?; diff --git a/rust/routee-compass/src/plugin/output/default/uuid/plugin.rs b/rust/routee-compass/src/plugin/output/default/uuid/plugin.rs index 05623e94..be0ba1c6 100644 --- a/rust/routee-compass/src/plugin/output/default/uuid/plugin.rs +++ b/rust/routee-compass/src/plugin/output/default/uuid/plugin.rs @@ -1,8 +1,8 @@ -use super::json_extensions::UUIDJsonExtensions; +use super::output_json_extensions::UUIDJsonExtensions; use crate::app::compass::compass_app_error::CompassAppError; use crate::app::search::search_app_result::SearchAppResult; -use crate::plugin::output::default::uuid::json_extensions::UUIDJsonField; -use crate::plugin::{output::output_plugin::OutputPlugin, plugin_error::PluginError}; +use crate::plugin::output::default::uuid::output_json_extensions::UUIDJsonField; +use crate::plugin::output::{OutputPlugin, OutputPluginError}; use kdam::Bar; use kdam::BarExt; @@ -17,9 +17,13 @@ pub struct UUIDOutputPlugin { } impl UUIDOutputPlugin { - pub fn from_file<P: AsRef<Path>>(filename: &P) -> Result<UUIDOutputPlugin, PluginError> { + pub fn from_file<P: AsRef<Path>>(filename: &P) -> Result<UUIDOutputPlugin, OutputPluginError> { let count = fs_utils::line_count(filename, fs_utils::is_gzip(filename)).map_err(\|e\| { - PluginError::FileReadError(filename.as_ref().to_path_buf(), e.to_string()) + OutputPluginError::BuildFailed(format!( + "failure reading UUID file {}: {}", + filename.as_ref().to_str().unwrap_or_default(), + e + )) })?; let mut pb = Bar::builder() @@ -27,16 +31,20 @@ impl UUIDOutputPlugin { .animation("fillup") .desc("uuid file") .build() - .map_err(PluginError::InternalError)?; + .map_err(OutputPluginError::InternalError)?; let cb = Box::new(\|\| { let _ = pb.update(1); }); let uuids = read_raw_file(filename, \|_idx, row\| Ok(row), Some(cb)).map_err(\|e\| { - PluginError::FileReadError(filename.as_ref().to_path_buf(), e.to_string()) + OutputPluginError::BuildFailed(format!( + "failure reading UUID file {}: {}", + filename.as_ref().to_str().unwrap_or_default(), + e + )) })?; - println!(); + eprintln!(); let o_key = UUIDJsonField::OriginVertexUUID.to_string(); let d_key = UUIDJsonField::DestinationVertexUUID.to_string(); @@ -53,21 +61,27 @@ impl OutputPlugin for UUIDOutputPlugin { &self, output: &mut serde_json::Value, search_result: &Result<(SearchAppResult, SearchInstance), CompassAppError>, - ) -> Result<(), PluginError> { + ) -> Result<(), OutputPluginError> { match search_result { Err(_) => Ok(()), Ok(_) => { let (origin_vertex_id, destination_vertex_id) = output.get_od_vertex_ids()?; - let origin_uuid = self - .uuids - .get(origin_vertex_id.0) - .cloned() - .ok_or_else(\|\| PluginError::UUIDMissing(origin_vertex_id.0))?; + let origin_uuid = self.uuids.get(origin_vertex_id.0).cloned().ok_or_else(\|\| { + OutputPluginError::OutputPluginFailed(format!( + "UUID lookup table missing vertex index {}", + origin_vertex_id.0 + )) + })?; let destination_uuid = self .uuids .get(destination_vertex_id.0) .cloned() - .ok_or_else(\|\| PluginError::UUIDMissing(destination_vertex_id.0))?; + .ok_or_else(\|\| { + OutputPluginError::OutputPluginFailed(format!( + "UUID lookup table missing vertex index {}", + destination_vertex_id.0 + )) + })?; output[&self.o_key] = serde_json::Value::String(origin_uuid); output[&self.d_key] = serde_json::Value::String(destination_uuid); diff --git a/rust/routee-compass/src/plugin/output/mod.rs b/rust/routee-compass/src/plugin/output/mod.rs index acc6f390..28f0084a 100644 --- a/rust/routee-compass/src/plugin/output/mod.rs +++ b/rust/routee-compass/src/plugin/output/mod.rs @@ -1,3 +1,7 @@ pub mod default; pub mod output_plugin; +pub mod output_plugin_error; pub mod output_plugin_ops; + +pub use output_plugin::OutputPlugin; +pub use output_plugin_error::OutputPluginError; diff --git a/rust/routee-compass/src/plugin/output/output_plugin.rs b/rust/routee-compass/src/plugin/output/output_plugin.rs index de116ba5..263ed8b8 100644 --- a/rust/routee-compass/src/plugin/output/output_plugin.rs +++ b/rust/routee-compass/src/plugin/output/output_plugin.rs @@ -1,8 +1,7 @@ -use routee_compass_core::algorithm::search::search_instance::SearchInstance; - +use super::output_plugin_error::OutputPluginError; use crate::app::compass::compass_app_error::CompassAppError; use crate::app::search::search_app_result::SearchAppResult; -use crate::plugin::plugin_error::PluginError; +use routee_compass_core::algorithm::search::search_instance::SearchInstance; /// Performs some kind of post-processing on a search result. The result JSON is available /// to the plugin as a reference which was potentially modified upstream by another output @@ -40,5 +39,5 @@ pub trait OutputPlugin: Send + Sync { &self, output: &mut serde_json::Value, result: &Result<(SearchAppResult, SearchInstance), CompassAppError>, - ) -> Result<(), PluginError>; + ) -> Result<(), OutputPluginError>; } diff --git a/rust/routee-compass/src/plugin/output/output_plugin_error.rs b/rust/routee-compass/src/plugin/output/output_plugin_error.rs new file mode 100644 index 00000000..cb41a5e5 --- /dev/null +++ b/rust/routee-compass/src/plugin/output/output_plugin_error.rs @@ -0,0 +1,24 @@ +use crate::plugin::input::InputField; + +#[derive(thiserror::Error, Debug)] +pub enum OutputPluginError { + #[error("failure building input plugin: {0}")] + BuildFailed(String), + #[error("required query field '{0}' not found")] + MissingExpectedQueryField(InputField), + #[error("{0} provided without {1}")] + MissingQueryFieldPair(InputField, InputField), + #[error("required query field '{0}' is not of type {1}")] + QueryFieldHasInvalidType(InputField, String), + #[error("expected query to be a json object '{{}}' but found {0}")] + UnexpectedQueryStructure(String), + #[error("plugin experienced JSON error: {source}")] + JsonError { + #[from] + source: serde_json::Error, + }, + #[error("failure running plugin: {0}")] + OutputPluginFailed(String), + #[error("unexpected error: {0}")] + InternalError(String), +} diff --git a/rust/routee-compass/src/plugin/plugin_error.rs b/rust/routee-compass/src/plugin/plugin_error.rs index 786ef4b7..b7f7a588 100644 --- a/rust/routee-compass/src/plugin/plugin_error.rs +++ b/rust/routee-compass/src/plugin/plugin_error.rs @@ -1,37 +1,29 @@ -use std::path::PathBuf; - -use routee_compass_core::{ - algorithm::search::search_error::SearchError, model::road_network::edge_id::EdgeId, -}; +use super::input::InputPluginError; +use super::output::OutputPluginError; #[derive(thiserror::Error, Debug)] pub enum PluginError { - #[error("failed to parse {0} as {1}")] - ParseError(String, String), - #[error("missing field {0}")] - MissingField(String), - #[error("error with parsing query inputs:\n{0}")] - InputError(String), - #[error("error with building plugin")] - BuildError, - #[error("{0}")] - PluginFailed(String), - #[error("unable to read file {0} due to {1}")] - FileReadError(PathBuf, String), - #[error(transparent)] - JsonError(#[from] serde_json::Error), - #[error(transparent)] - GeoJsonError(#[from] geojson::Error), - #[error(transparent)] - CsvReadError(#[from] csv::Error), - #[error("geometry missing for edge id {0}")] - EdgeGeometryMissing(EdgeId), - #[error("uuid missing for edge id {0}")] - UUIDMissing(usize), - #[error(transparent)] - SearchError(#[from] SearchError), + #[error("failure building plugin: {0}")] + BuildFailed(String), + #[error("required query field '{0}' for plugin {1} not found")] + MissingExpectedQueryField(String, String), + #[error("failure running input plugin: {source}")] + InputPluginFailed { + #[from] + source: InputPluginError, + }, + #[error("failure running output plugin: {source}")] + OutputPluginFailed { + #[from] + source: OutputPluginError, + }, + #[error("plugin experienced JSON error: {source}")] + JsonError { + #[from] + source: serde_json::Error, + }, #[error("expected query to be a json object '{{}}' but found {0}")] UnexpectedQueryStructure(String), - #[error("unexpected error {0}")] + #[error("unexpected error: {0}")] InternalError(String), }	simplify error variants when we started _this journey_ we picked specified taxonomies for traversal model errors, frontier model errors, graph errors and the like, such as `TraversalModelError::MissingIdInTabularCostFunction`. but as we solidify RouteE Compass as a dependency of other downstream projects, those variants are not general enough. what is instead missing is a standardized taxonomy of error types. for example, the following variants (incomplete set): ```rust pub enum TraversalModelError { #[error("failure building traversal model from file {0}: {1}")] FileReadError(PathBuf, String), #[error("failure building traversal model: {0}")] BuildError(String), #[error("numeric error during calculation: {0}")] NumericError(String), #[error("index {0} for {1} not found on search state {2:?}")] StateVectorIndexOutOfBounds(usize, String, TraversalState), #[error("id {0} for id type {1} not found in tabular edge cost function {2}")] MissingIdInTabularCostFunction(String, String, String), #[error("prediction model failed with error {0}")] PredictionModel(String), } ``` could be reduced to something like ```rust pub enum TraversalModelError { #[error("failure during build for model {1} due to {0}")] BuildErrorForModel(String, String), #[error("failure during build for service {1} due to {0}")] BuildErrorForService(String, String), #[error("failure during build for builder {1} due to {0}")] BuildErrorForBuilder(String, String), #[error("runtime error for model {1} due to {0}")] RuntimeErrorForModel(String, String) } ``` no transparent errors should bubble to top of app attempting to run a compass app and i was getting this error: ``` [2024-04-08T22:49:11Z ERROR mep_routee_compass] invalid type: map, expected variant identifier ``` this was due to the fact that my configuration TOML's output policy had double braces instead of single: ```toml [[response_output_policy]] # should be [...] type = "file" ``` the error is vague because CompassAppError has a transparent ConfigError (from the config crate): ```rust #[error(transparent)] ConfigError(#[from] ConfigError), ``` Compass' top-level error types should not be transparent unless they do not require context. this should instead return a BuildError or similar where the developer includes the contextual information: ```rust #[error("error building component {0} of type {1}: {2}")] BuildError(String, String, String), ```		2024-12-02T17:21:28	[]	[]
NREL/routee-compass	NREL__routee-compass-168	9fa285c2ac27a9efcacba5658e7e9040c38ffb5d	diff --git a/docs/notebooks/open_street_maps_example.ipynb b/docs/notebooks/open_street_maps_example.ipynb index e73c5f8f..84831098 100644 --- a/docs/notebooks/open_street_maps_example.ipynb +++ b/docs/notebooks/open_street_maps_example.ipynb @@ -86,7 +86,21 @@ "output_type": "stream", "text": [ "processing graph topology and speeds\n", - "adding grade information\n", + "adding grade information\n" + ] + }, + { + "name": "stderr", + "output_type": "stream", + "text": [ + "/Users/rfitzger/anaconda3/envs/routee-rust/lib/python3.9/site-packages/osgeo/gdal.py:287: FutureWarning: Neither gdal.UseExceptions() nor gdal.DontUseExceptions() has been explicitly called. In GDAL 4.0, exceptions will be enabled by default.\n", + " warnings.warn(\n" + ] + }, + { + "name": "stdout", + "output_type": "stream", + "text": [ "processing vertices\n", "processing edges\n", "writing vertex files\n", @@ -119,7 +133,7 @@ }, { "cell_type": "code", - "execution_count": 4, + "execution_count": 5, "id": "edad6c15-c641-4010-b097-7d07aa86c1e7", "metadata": {}, "outputs": [ @@ -127,6 +141,9 @@ "name": "stdout", "output_type": "stream", "text": [ + "\n", + "\n", + "\n", "\n" ] }, @@ -134,16 +151,7 @@ "name": "stderr", "output_type": "stream", "text": [ - "uuid file: 100%\|██████████\| 16992/16992 [00:00<00:00, 6624988.00it/s]/s]" - ] - }, - { - "name": "stdout", - "output_type": "stream", - "text": [ - "\n", - "\n", - "\n" + "uuid file: 100%\|██████████\| 16991/16991 [00:00<00:00, 4182921.75it/s]/s]" ] } ], @@ -177,7 +185,7 @@ }, { "cell_type": "code", - "execution_count": 193, + "execution_count": 6, "id": "f01086d0-231f-4013-bc24-97f9c56670bd", "metadata": {}, "outputs": [], @@ -189,7 +197,7 @@ " \"destination_x\": -104.975360,\n", " \"destination_y\": 39.693005,\n", " \"model_name\": \"2016_TOYOTA_Camry_4cyl_2WD\",\n", - " \"vehicle_state_variable_rates\": {\n", + " \"vehicle_rates\": {\n", " \"distance\": {\"type\": \"factor\", \"factor\": 0.655},\n", " \"time\": {\"type\": \"factor\", \"factor\": 0.33},\n", " \"energy_liquid\": {\"type\": \"factor\", \"factor\": 3.0}, \n", @@ -198,7 +206,7 @@ " \"test_cases\": [\n", " {\n", " \"name\": \"least_time\",\n", - " \"state_variable_coefficients\": {\n", + " \"weights\": {\n", " \"distance\": 0,\n", " \"time\": 1,\n", " \"energy_liquid\": 0\n", @@ -206,7 +214,7 @@ " },\n", " {\n", " \"name\": \"least_energy\",\n", - " \"state_variable_coefficients\": {\n", + " \"weights\": {\n", " \"distance\": 0,\n", " \"time\": 0,\n", " \"energy_liquid\": 1\n", @@ -214,7 +222,7 @@ " },\n", " {\n", " \"name\": \"least_cost\",\n", - " \"state_variable_coefficients\": {\n", + " \"weights\": {\n", " \"distance\": 1,\n", " \"time\": 1,\n", " \"energy_liquid\": 1\n", @@ -241,7 +249,7 @@ }, { "cell_type": "code", - "execution_count": 176, + "execution_count": 7, "id": "c6e9361b-5ffd-4ec5-a09a-d51e46752ebf", "metadata": {}, "outputs": [ @@ -256,7 +264,7 @@ "name": "stderr", "output_type": "stream", "text": [ - "search: 100%\|██████████\| 3/3 [00:00<00:00, 58.36it/s]7.79it/s]" + "search: 100%\|██████████\| 3/3 [00:00<00:00, 35.65it/s]2.69it/s]" ] } ], @@ -277,7 +285,7 @@ }, { "cell_type": "code", - "execution_count": 177, + "execution_count": 7, "id": "2c51774c-acff-4569-a820-4be5ba5aa51b", "metadata": {}, "outputs": [], @@ -302,7 +310,7 @@ }, { "cell_type": "code", - "execution_count": 178, + "execution_count": 8, "id": "1825ee52", "metadata": {}, "outputs": [], @@ -325,7 +333,7 @@ }, { "cell_type": "code", - "execution_count": 179, + "execution_count": 9, "id": "961c910c-63ec-4c79-9522-19f24f151215", "metadata": {}, "outputs": [ @@ -336,8 +344,8 @@ "{\n", " \"distance\": 9.152878807061885,\n", " \"energy_electric\": 0.0,\n", - " \"energy_liquid\": 0.3200989780246874,\n", - " \"time\": 10.605213626907442\n", + " \"energy_liquid\": 0.32011129944536876,\n", + " \"time\": 10.905464892278664\n", "}\n" ] } @@ -356,7 +364,7 @@ }, { "cell_type": "code", - "execution_count": 180, + "execution_count": 10, "id": "804b8d54-43d8-4c57-889d-e23f947892ee", "metadata": {}, "outputs": [ @@ -401,7 +409,7 @@ }, { "cell_type": "code", - "execution_count": 181, + "execution_count": 11, "id": "d4501094", "metadata": {}, "outputs": [ @@ -412,9 +420,9 @@ "{\n", " \"distance\": 5.995135618625535,\n", " \"energy_electric\": 0.0,\n", - " \"energy_liquid\": 0.9602969340740621,\n", - " \"time\": 3.499720496879456,\n", - " \"total_cost\": 10.455153049579053\n", + " \"energy_liquid\": 0.9603338983361063,\n", + " \"time\": 3.598803414451959,\n", + " \"total_cost\": 10.5542729314136\n", "}\n" ] } @@ -437,7 +445,7 @@ }, { "cell_type": "code", - "execution_count": 182, + "execution_count": 12, "id": "788183c8", "metadata": {}, "outputs": [ @@ -508,7 +516,7 @@ }, { "cell_type": "code", - "execution_count": 183, + "execution_count": 13, "id": "e2c6cb39-49aa-4697-9785-fa839dc16a86", "metadata": {}, "outputs": [ @@ -519,15 +527,15 @@ "{\n", " \"distance\": 6.733433188572938,\n", " \"energy_electric\": 0.0,\n", - " \"energy_liquid\": 0.26646891776649895,\n", - " \"time\": 12.657830096916395\n", + " \"energy_liquid\": 0.2664853255906838,\n", + " \"time\": 12.777312852016575\n", "}\n", "{\n", " \"distance\": 4.4103987385152745,\n", " \"energy_electric\": 0.0,\n", - " \"energy_liquid\": 0.7994067532994968,\n", - " \"time\": 4.177083931982411,\n", - " \"total_cost\": 9.386889423797182\n", + " \"energy_liquid\": 0.7994559767720513,\n", + " \"time\": 4.21651324116547,\n", + " \"total_cost\": 9.426367956452795\n", "}\n" ] } @@ -547,7 +555,7 @@ }, { "cell_type": "code", - "execution_count": 184, + "execution_count": 14, "id": "6283c273", "metadata": {}, "outputs": [ @@ -556,9 +564,9 @@ "output_type": "stream", "text": [ " - distance: 2.42 miles further with time-optimal\n", - " - time: 2.05 minutes longer with energy-optimal\n", + " - time: 1.87 minutes longer with energy-optimal\n", " - energy: 0.05 gallons_gasoline more with time-optimal\n", - " - cost: $1.07 more with time-optimal\n" + " - cost: $1.13 more with time-optimal\n" ] } ], @@ -587,7 +595,7 @@ }, { "cell_type": "code", - "execution_count": 185, + "execution_count": 15, "id": "791e814f-333c-43f4-ae84-36b9be517548", "metadata": {}, "outputs": [ @@ -714,7 +722,7 @@ }, { "cell_type": "code", - "execution_count": 186, + "execution_count": 16, "id": "30257102-07ee-4338-9c8c-41676e1995ed", "metadata": {}, "outputs": [], @@ -732,7 +740,7 @@ }, { "cell_type": "code", - "execution_count": 187, + "execution_count": 17, "id": "1ac13aae-32fc-4c10-b888-6373b175655a", "metadata": {}, "outputs": [ @@ -766,7 +774,7 @@ " <meta name="viewport" content="width=device-width,\n", " initial-scale=1.0, maximum-scale=1.0, user-scalable=no" />\n", " <style>\n", - " #map_a5cb9760638531fd5b14e75a46b3b1f1 {\n", + " #map_b6cf708d54aa4599b756d3ce62516c9c {\n", " position: relative;\n", " width: 100.0%;\n", " height: 100.0%;\n", @@ -780,14 +788,14 @@ "<body>\n", " \n", " \n", - " <div class="folium-map" id="map_a5cb9760638531fd5b14e75a46b3b1f1" ></div>\n", + " <div class="folium-map" id="map_b6cf708d54aa4599b756d3ce62516c9c" ></div>\n", " \n", "</body>\n", "<script>\n", " \n", " \n", - " var map_a5cb9760638531fd5b14e75a46b3b1f1 = L.map(\n", - " "map_a5cb9760638531fd5b14e75a46b3b1f1",\n", + " var map_b6cf708d54aa4599b756d3ce62516c9c = L.map(\n", + " "map_b6cf708d54aa4599b756d3ce62516c9c",\n", " {\n", " center: [39.73823928833008, -105.01406860351562],\n", " crs: L.CRS.EPSG3857,\n", @@ -801,19 +809,19 @@ "\n", " \n", " \n", - " var tile_layer_9b8abbe2aac9b29cd6e7073b6627118d = L.tileLayer(\n", + " var tile_layer_91198d8453ecb8ddf13da3c43cd11146 = L.tileLayer(\n", " "https://{s}.tile.openstreetmap.org/{z}/{x}/{y}.png",\n", " {"attribution": "Data by \\u0026copy; \\u003ca target=\\"_blank\\" href=\\"http://openstreetmap.org\\"\\u003eOpenStreetMap\\u003c/a\\u003e, under \\u003ca target=\\"_blank\\" href=\\"http://www.openstreetmap.org/copyright\\"\\u003eODbL\\u003c/a\\u003e.", "detectRetina": false, "maxNativeZoom": 18, "maxZoom": 18, "minZoom": 0, "noWrap": false, "opacity": 1, "subdomains": "abc", "tms": false}\n", - " ).addTo(map_a5cb9760638531fd5b14e75a46b3b1f1);\n", + " ).addTo(map_b6cf708d54aa4599b756d3ce62516c9c);\n", " \n", " \n", - " var poly_line_b64ee0679bc07d7a55e8ab1075c0227e = L.polyline(\n", + " var poly_line_0633987bf2e82ded548a26dc464c4978 = L.polyline(\n", " [[39.77902603149414, -104.96916961669922], [39.77896499633789, -104.96910095214844], [39.77878189086914, -104.96887969970703], [39.778751373291016, -104.96884155273438], [39.778587341308594, -104.9686508178711], [39.778480529785156, -104.96852111816406], [39.77839660644531, -104.96841430664062], [39.77846908569336, -104.96832275390625], [39.77865982055664, -104.96807098388672], [39.778709411621094, -104.96791076660156], [39.77894592285156, -104.96761322021484], [39.77906799316406, -104.96745300292969], [39.77915954589844, -104.96732330322266], [39.7791862487793, -104.96728515625], [39.779232025146484, -104.96722412109375], [39.77926254272461, -104.96719360351562], [39.779293060302734, -104.9671630859375], [39.77932357788086, -104.96714782714844], [39.77938461303711, -104.96710968017578], [39.779441833496094, -104.96707916259766], [39.77951431274414, -104.96704864501953], [39.7796745300293, -104.96698760986328], [39.77983093261719, -104.9669418334961], [39.78031921386719, -104.96683502197266], [39.780364990234375, -104.9668197631836], [39.78050231933594, -104.966796875], [39.78049850463867, -104.96688079833984], [39.78049850463867, -104.96698760986328], [39.78049850463867, -104.96705627441406], [39.78049087524414, -104.96717834472656], [39.78046798706055, -104.96759033203125], [39.78045654296875, -104.96772003173828], [39.78042984008789, -104.96797180175781], [39.78041458129883, -104.96810913085938], [39.780372619628906, -104.96839904785156], [39.78034210205078, -104.96859741210938], [39.780296325683594, -104.96888732910156], [39.7802734375, -104.96903228759766], [39.78025436401367, -104.96917724609375], [39.780250549316406, -104.96920776367188], [39.780235290527344, -104.96934509277344], [39.78020095825195, -104.96965789794922], [39.78001403808594, -104.9715347290039], [39.77994155883789, -104.97315216064453], [39.77980422973633, -104.97496032714844], [39.7797737121582, -104.97691345214844], [39.7797737121582, -104.9772720336914], [39.7797737121582, -104.97753143310547], [39.7797737121582, -104.97764587402344], [39.779781341552734, -104.97801971435547], [39.77980041503906, -104.97874450683594], [39.77981948852539, -104.97929382324219], [39.77983093261719, -104.97946166992188], [39.77986526489258, -104.98001861572266], [39.77991485595703, -104.98080444335938], [39.779991149902344, -104.98214721679688], [39.78001403808594, -104.98335266113281], [39.7800178527832, -104.98450469970703], [39.78002166748047, -104.98558044433594], [39.780033111572266, -104.98602294921875], [39.7800407409668, -104.98613739013672], [39.780059814453125, -104.98633575439453], [39.780086517333984, -104.9865493774414], [39.780128479003906, -104.98677825927734], [39.780174255371094, -104.98695373535156], [39.78023910522461, -104.98715209960938], [39.780540466308594, -104.98787689208984], [39.78069305419922, -104.98828887939453], [39.78076934814453, -104.98860168457031], [39.780784606933594, -104.98871612548828], [39.780792236328125, -104.98896026611328], [39.78078079223633, -104.98912048339844], [39.780765533447266, -104.98928833007812], [39.78071975708008, -104.98950958251953], [39.78061294555664, -104.98975372314453], [39.78052520751953, -104.98993682861328], [39.78038787841797, -104.99011993408203], [39.78028869628906, -104.99024963378906], [39.780174255371094, -104.99034881591797], [39.77997589111328, -104.990478515625], [39.77995300292969, -104.99049377441406], [39.77978515625, -104.99056243896484], [39.77960205078125, -104.99060821533203], [39.7794189453125, -104.99060821533203], [39.77922439575195, -104.9905776977539], [39.778968811035156, -104.99048614501953], [39.77858352661133, -104.9903335571289], [39.778541564941406, -104.99031066894531], [39.77811050415039, -104.99011993408203], [39.777706146240234, -104.98995208740234], [39.7775764465332, -104.98990631103516], [39.77741241455078, -104.98985290527344], [39.77730941772461, -104.98983001708984], [39.7772102355957, -104.98980712890625], [39.777130126953125, -104.98979949951172], [39.77692794799805, -104.98979187011719], [39.77660369873047, -104.98979187011719], [39.77629089355469, -104.98980712890625], [39.77605056762695, -104.98982238769531], [39.77583694458008, -104.98986053466797], [39.77559280395508, -104.98991394042969], [39.7751579284668, -104.99003601074219], [39.774871826171875, -104.99012756347656], [39.77482986450195, -104.9901351928711], [39.774696350097656, -104.99018096923828], [39.774539947509766, -104.99022674560547], [39.774383544921875, -104.99026489257812], [39.77424621582031, -104.99028778076172], [39.77410888671875, -104.99029541015625], [39.773887634277344, -104.99030303955078], [39.77339553833008, -104.99031829833984], [39.773284912109375, -104.99031829833984], [39.77314376831055, -104.99032592773438], [39.773006439208984, -104.99034118652344], [39.77288055419922, -104.99036407470703], [39.77272033691406, -104.99040222167969], [39.77251434326172, -104.99046325683594], [39.77234649658203, -104.99051666259766], [39.77214431762695, -104.99059295654297], [39.77194595336914, -104.99068450927734], [39.77174377441406, -104.99079895019531], [39.77142333984375, -104.99098205566406], [39.77122116088867, -104.9911117553711], [39.77073287963867, -104.9914321899414], [39.7704963684082, -104.99164581298828], [39.770259857177734, -104.99187469482422], [39.77001190185547, -104.99214172363281], [39.769779205322266, -104.99241638183594], [39.76954650878906, -104.99272918701172], [39.76931381225586, -104.99305725097656], [39.76912307739258, -104.99334716796875], [39.76897048950195, -104.99359893798828], [39.768775939941406, -104.99393463134766], [39.768592834472656, -104.99432373046875], [39.76849365234375, -104.99454498291016], [39.76826477050781, -104.9951171875], [39.768192291259766, -104.99535369873047], [39.76802062988281, -104.99580383300781], [39.767581939697266, -104.99694061279297], [39.767154693603516, -104.99806213378906], [39.766963958740234, -104.9985580444336], [39.76682662963867, -104.9989013671875], [39.7667121887207, -104.99915313720703], [39.76656723022461, -104.99944305419922], [39.76641845703125, -104.99970245361328], [39.766273498535156, -104.99990844726562], [39.766109466552734, -105.00015258789062], [39.7659912109375, -105.00031280517578], [39.7658805847168, -105.0004653930664], [39.765750885009766, -105.00062561035156], [39.765567779541016, -105.00081634521484], [39.76536560058594, -105.00100708007812], [39.76518249511719, -105.00117492675781], [39.76498794555664, -105.00135803222656], [39.763919830322266, -105.002197265625], [39.76262283325195, -105.00331115722656], [39.76240539550781, -105.00350189208984], [39.762203216552734, -105.0036849975586], [39.76198959350586, -105.00389099121094], [39.76179885864258, -105.00407409667969], [39.7616081237793, -105.0042724609375], [39.761444091796875, -105.00445556640625], [39.761260986328125, -105.00466918945312], [39.76103973388672, -105.00492858886719], [39.76044845581055, -105.00575256347656], [39.759456634521484, -105.0072021484375], [39.75861740112305, -105.0083999633789], [39.758094787597656, -105.00910949707031], [39.75727844238281, -105.0101547241211], [39.75709915161133, -105.01038360595703], [39.756736755371094, -105.0108871459961], [39.75618362426758, -105.0116195678711], [39.75572967529297, -105.01225280761719], [39.75527572631836, -105.01285552978516], [39.75492858886719, -105.01334381103516], [39.7547492980957, -105.01359558105469], [39.75423812866211, -105.01427459716797], [39.75400924682617, -105.01457214355469], [39.753841400146484, -105.0147705078125], [39.75370788574219, -105.01492309570312], [39.75352096557617, -105.01513671875], [39.75336837768555, -105.01529693603516], [39.753173828125, -105.01548767089844], [39.75297164916992, -105.01567840576172], [39.752811431884766, -105.01581573486328], [39.75261688232422, -105.01597595214844], [39.75239944458008, -105.0161361694336], [39.75222396850586, -105.01627349853516], [39.752037048339844, -105.01639556884766], [39.751853942871094, -105.0165023803711], [39.75165939331055, -105.01661682128906], [39.75148391723633, -105.01670837402344], [39.75117874145508, -105.0168685913086], [39.75090026855469, -105.01699829101562], [39.75063705444336, -105.01710510253906], [39.74983215332031, -105.01744079589844], [39.74938201904297, -105.01763916015625], [39.74889373779297, -105.01783752441406], [39.748531341552734, -105.01799011230469], [39.74820327758789, -105.01811218261719], [39.74799728393555, -105.01818084716797], [39.74775695800781, -105.01825714111328], [39.74755096435547, -105.01830291748047], [39.74735641479492, -105.01834106445312], [39.747108459472656, -105.01836395263672], [39.74689865112305, -105.01837158203125], [39.74666976928711, -105.01836395263672], [39.74650955200195, -105.01835632324219], [39.74633026123047, -105.01832580566406], [39.746131896972656, -105.0182876586914], [39.74597930908203, -105.01824951171875], [39.74581527709961, -105.01819610595703], [39.745609283447266, -105.01811218261719], [39.74538040161133, -105.01800537109375], [39.74514389038086, -105.01786804199219], [39.744937896728516, -105.01773834228516], [39.744747161865234, -105.01759338378906], [39.74456787109375, -105.0174560546875], [39.74444580078125, -105.01734161376953], [39.743892669677734, -105.01679992675781], [39.74342727661133, -105.0163345336914], [39.743038177490234, -105.015869140625], [39.742897033691406, -105.01570892333984], [39.74272537231445, -105.0155258178711], [39.7425422668457, -105.01532745361328], [39.74235916137695, -105.01516723632812], [39.7421875, -105.01500701904297], [39.74201965332031, -105.0148696899414], [39.74187469482422, -105.01476287841797], [39.74171447753906, -105.0146484375], [39.74155807495117, -105.01456451416016], [39.741329193115234, -105.01445007324219], [39.74116134643555, -105.01438903808594], [39.74099349975586, -105.01432037353516], [39.74075698852539, -105.01425170898438], [39.7405891418457, -105.01421356201172], [39.740386962890625, -105.01416778564453], [39.740230560302734, -105.01415252685547], [39.74014663696289, -105.0141372680664], [39.73987579345703, -105.01412963867188], [39.73881530761719, -105.01406860351562], [39.73861312866211, -105.0140609741211], [39.73841094970703, -105.0140609741211], [39.73823928833008, -105.01406860351562], [39.73807907104492, -105.01408386230469], [39.7379035949707, -105.01409912109375], [39.73770523071289, -105.01412963867188], [39.737457275390625, -105.01416778564453], [39.737152099609375, -105.01428985595703], [39.73687744140625, -105.01436614990234], [39.7366828918457, -105.01441955566406], [39.73640823364258, -105.01451873779297], [39.73619079589844, -105.01460266113281], [39.73607635498047, -105.01464080810547], [39.735904693603516, -105.01471710205078], [39.73575973510742, -105.01478576660156], [39.73533630371094, -105.0149917602539], [39.73477554321289, -105.01528930664062], [39.734619140625, -105.01536560058594], [39.73448181152344, -105.01542663574219], [39.73434066772461, -105.01548767089844], [39.73418426513672, -105.01554870605469], [39.73397445678711, -105.015625], [39.733829498291016, -105.01566314697266], [39.73370361328125, -105.01570129394531], [39.73356246948242, -105.01573944091797], [39.733436584472656, -105.0157699584961], [39.73330307006836, -105.01579284667969], [39.73316955566406, -105.01582336425781], [39.73301315307617, -105.01583862304688], [39.73285675048828, -105.01585388183594], [39.73271942138672, -105.01586151123047], [39.732574462890625, -105.015869140625], [39.73243713378906, -105.01587677001953], [39.732303619384766, -105.01588439941406], [39.73218536376953, -105.01588439941406], [39.73207092285156, -105.01587677001953], [39.73195266723633, -105.015869140625], [39.73179244995117, -105.01585388183594], [39.73155212402344, -105.01580047607422], [39.73124313354492, -105.01573181152344], [39.731075286865234, -105.01568603515625], [39.7308464050293, -105.01561737060547], [39.730567932128906, -105.01553344726562], [39.730201721191406, -105.0154037475586], [39.729949951171875, -105.01531219482422], [39.72952651977539, -105.01518249511719], [39.729122161865234, -105.0150375366211], [39.728843688964844, -105.01493835449219], [39.72844314575195, -105.01477813720703], [39.72832107543945, -105.01473236083984], [39.728145599365234, -105.01466369628906], [39.727943420410156, -105.01458740234375], [39.727439880371094, -105.01441192626953], [39.72711181640625, -105.01427459716797], [39.72694778442383, -105.01421356201172], [39.726749420166016, -105.01412200927734], [39.726558685302734, -105.0140380859375], [39.726375579833984, -105.01394653320312], [39.72616195678711, -105.01382446289062], [39.72585678100586, -105.01362609863281], [39.725624084472656, -105.01348114013672], [39.72536087036133, -105.01329803466797], [39.72511672973633, -105.01311492919922], [39.72493362426758, -105.01297760009766], [39.7247200012207, -105.01280212402344], [39.72451400756836, -105.01262664794922], [39.72426986694336, -105.01239776611328], [39.72408676147461, -105.01221466064453], [39.723899841308594, -105.01201629638672], [39.72377395629883, -105.01188659667969], [39.72359848022461, -105.01168060302734], [39.72343826293945, -105.0114974975586], [39.72330093383789, -105.01133728027344], [39.723079681396484, -105.01106262207031], [39.7227897644043, -105.01066589355469], [39.72258377075195, -105.01038360595703], [39.722251892089844, -105.0099105834961], [39.7220573425293, -105.0096435546875], [39.721920013427734, -105.00946807861328], [39.721744537353516, -105.00924682617188], [39.721649169921875, -105.00914001464844], [39.7214241027832, -105.00888061523438], [39.72117233276367, -105.00861358642578], [39.720741271972656, -105.0081787109375], [39.720523834228516, -105.00794219970703], [39.7200813293457, -105.0074691772461], [39.71998977661133, -105.00737762451172], [39.71975326538086, -105.00711822509766], [39.71941375732422, -105.00675201416016], [39.719242095947266, -105.0065689086914], [39.719085693359375, -105.00640106201172], [39.718936920166016, -105.00623321533203], [39.718807220458984, -105.00608825683594], [39.718658447265625, -105.00591278076172], [39.71849060058594, -105.00570678710938], [39.7183723449707, -105.00556182861328], [39.71826171875, -105.00541687011719], [39.71812057495117, -105.00521850585938], [39.71792221069336, -105.00492858886719], [39.71759796142578, -105.00445556640625], [39.717166900634766, -105.0038070678711], [39.716880798339844, -105.00337982177734], [39.71671676635742, -105.00314331054688], [39.716552734375, -105.00292205810547], [39.7164306640625, -105.00276184082031], [39.71633529663086, -105.00264739990234], [39.71624755859375, -105.00254821777344], [39.71617889404297, -105.00247192382812], [39.71608352661133, -105.00237274169922], [39.715999603271484, -105.00228118896484], [39.71593475341797, -105.0022201538086], [39.715885162353516, -105.0021743774414], [39.715755462646484, -105.00206756591797], [39.715660095214844, -105.00198364257812], [39.715538024902344, -105.00189208984375], [39.71543884277344, -105.00181579589844], [39.71533203125, -105.00173950195312], [39.7152214050293, -105.00166320800781], [39.71499252319336, -105.00151824951172], [39.714324951171875, -105.0010757446289], [39.71399688720703, -105.00086212158203], [39.71346664428711, -105.00052642822266], [39.71318435668945, -105.00035095214844], [39.71266174316406, -105.00003051757812], [39.71223449707031, -104.9997787475586], [39.71159362792969, -104.99940490722656], [39.71141052246094, -104.99930572509766], [39.710933685302734, -104.99901580810547], [39.71067810058594, -104.99886322021484], [39.710357666015625, -104.99867248535156], [39.71013641357422, -104.99855041503906], [39.70988464355469, -104.99840545654297], [39.70966720581055, -104.99829864501953], [39.7094841003418, -104.99820709228516], [39.70940017700195, -104.99816131591797], [39.709136962890625, -104.99803924560547], [39.708702087402344, -104.99784088134766], [39.70827865600586, -104.99764251708984], [39.7078857421875, -104.99746704101562], [39.707275390625, -104.9971923828125], [39.70710372924805, -104.99711608886719], [39.7068977355957, -104.99701690673828], [39.706809997558594, -104.99697875976562], [39.706729888916016, -104.99693298339844], [39.70663070678711, -104.99688720703125], [39.706539154052734, -104.99683380126953], [39.70646667480469, -104.99678802490234], [39.70637130737305, -104.9967269897461], [39.70630645751953, -104.99668884277344], [39.70623779296875, -104.99664306640625], [39.70616912841797, -104.99658966064453], [39.706085205078125, -104.99652862548828], [39.70596694946289, -104.9964370727539], [39.705814361572266, -104.9963150024414], [39.7056999206543, -104.9962158203125], [39.70561218261719, -104.99613189697266], [39.70552062988281, -104.99604034423828], [39.705467224121094, -104.99598693847656], [39.70537567138672, -104.99588775634766], [39.705299377441406, -104.99580383300781], [39.705169677734375, -104.99565124511719], [39.70507049560547, -104.99552917480469], [39.704959869384766, -104.9953842163086], [39.704837799072266, -104.99520874023438], [39.70473098754883, -104.99505615234375], [39.704627990722656, -104.99488067626953], [39.704551696777344, -104.99474334716797], [39.704471588134766, -104.99459838867188], [39.70440673828125, -104.99447631835938], [39.7043342590332, -104.99433135986328], [39.704261779785156, -104.99417877197266], [39.70420455932617, -104.99404907226562], [39.70415115356445, -104.99392700195312], [39.704097747802734, -104.99378204345703], [39.70404815673828, -104.99363708496094], [39.70399475097656, -104.99347686767578], [39.70393753051758, -104.9933090209961], [39.70388412475586, -104.99313354492188], [39.703792572021484, -104.99284362792969], [39.70363235473633, -104.9923324584961], [39.703487396240234, -104.99188232421875], [39.703304290771484, -104.99127197265625], [39.703189849853516, -104.99092102050781], [39.70305633544922, -104.990478515625], [39.702980041503906, -104.99024200439453], [39.70287322998047, -104.989990234375], [39.70277404785156, -104.98977661132812], [39.70265197753906, -104.98954010009766], [39.70252990722656, -104.98933410644531], [39.70241928100586, -104.98914337158203], [39.70228576660156, -104.98894500732422], [39.702110290527344, -104.98870849609375], [39.70179748535156, -104.9883041381836], [39.70148849487305, -104.98790740966797], [39.700843811035156, -104.987060546875], [39.700706481933594, -104.98689270019531], [39.70027542114258, -104.98641967773438], [39.700077056884766, -104.9861831665039], [39.69994354248047, -104.98603057861328], [39.699771881103516, -104.98583984375], [39.69960021972656, -104.98567962646484], [39.699462890625, -104.98554229736328], [39.699344635009766, -104.98542022705078], [39.69913101196289, -104.98521423339844], [39.698974609375, -104.98504638671875], [39.69866180419922, -104.9847412109375], [39.6981086730957, -104.98410034179688], [39.69713592529297, -104.98299407958984], [39.696075439453125, -104.98184967041016], [39.69560623168945, -104.9813003540039], [39.69524383544922, -104.98087310791016], [39.69488525390625, -104.98043823242188], [39.694786071777344, -104.98033905029297], [39.6947021484375, -104.98025512695312], [39.69462585449219, -104.98019409179688], [39.6945915222168, -104.98016357421875], [39.69448471069336, -104.98009490966797], [39.69437789916992, -104.98002624511719], [39.6942024230957, -104.97992706298828], [39.69398498535156, -104.97977447509766], [39.69377899169922, -104.97953033447266], [39.69367599487305, -104.97943115234375], [39.69364929199219, -104.97940063476562], [39.69361114501953, -104.97936248779297], [39.69355010986328, -104.97931671142578], [39.6934814453125, -104.9792709350586], [39.69330978393555, -104.97918701171875], [39.69325637817383, -104.97914123535156], [39.693199157714844, -104.97909545898438], [39.693145751953125, -104.97903442382812], [39.693092346191406, -104.97897338867188], [39.69303512573242, -104.97886657714844], [39.69300079345703, -104.97877502441406], [39.69298553466797, -104.97871398925781], [39.69298553466797, -104.97850036621094], [39.69298553466797, -104.97827911376953], [39.69298553466797, -104.97792053222656], [39.6929931640625, -104.9774398803711], [39.6929931640625, -104.9773941040039], [39.6929931640625, -104.9772720336914], [39.692989349365234, -104.97704315185547], [39.692989349365234, -104.97694396972656], [39.692989349365234, -104.97685241699219], [39.69298553466797, -104.97636413574219], [39.6929817199707, -104.97577667236328]],\n", " {"bubblingMouseEvents": true, "color": "red", "dashArray": null, "dashOffset": null, "fill": false, "fillColor": "red", "fillOpacity": 0.2, "fillRule": "evenodd", "lineCap": "round", "lineJoin": "round", "noClip": false, "opacity": 0.8, "smoothFactor": 1.0, "stroke": true, "weight": 10}\n", - " ).addTo(map_a5cb9760638531fd5b14e75a46b3b1f1);\n", + " ).addTo(map_b6cf708d54aa4599b756d3ce62516c9c);\n", " \n", " \n", - " poly_line_b64ee0679bc07d7a55e8ab1075c0227e.bindTooltip(\n", + " poly_line_0633987bf2e82ded548a26dc464c4978.bindTooltip(\n", " `<div>\n", " Shortest Time\n", " </div>`,\n", @@ -821,19 +829,19 @@ " );\n", " \n", " \n", - " var marker_68f779ab1d817630f938af7a6c95a2b2 = L.marker(\n", + " var marker_51e8e84125a77a2fb180e6f323d27b85 = L.marker(\n", " [39.77902603149414, -104.96916961669922],\n", " {}\n", - " ).addTo(map_a5cb9760638531fd5b14e75a46b3b1f1);\n", + " ).addTo(map_b6cf708d54aa4599b756d3ce62516c9c);\n", " \n", " \n", - " var icon_cad02f21805c4a5539d3cc33a1463c3a = L.AwesomeMarkers.icon(\n", + " var icon_5fe7f896fc566755b249e564107262ca = L.AwesomeMarkers.icon(\n", " {"extraClasses": "fa-rotate-0", "icon": "circle", "iconColor": "white", "markerColor": "green", "prefix": "fa"}\n", " );\n", - " marker_68f779ab1d817630f938af7a6c95a2b2.setIcon(icon_cad02f21805c4a5539d3cc33a1463c3a);\n", + " marker_51e8e84125a77a2fb180e6f323d27b85.setIcon(icon_5fe7f896fc566755b249e564107262ca);\n", " \n", " \n", - " marker_68f779ab1d817630f938af7a6c95a2b2.bindTooltip(\n", + " marker_51e8e84125a77a2fb180e6f323d27b85.bindTooltip(\n", " `<div>\n", " Origin\n", " </div>`,\n", @@ -841,19 +849,19 @@ " );\n", " \n", " \n", - " var marker_a4d31af8d6a545d85c81795dd2d38ffe = L.marker(\n", + " var marker_b3dbf16a30f95c6a552145a7c117e057 = L.marker(\n", " [39.6929817199707, -104.97577667236328],\n", " {}\n", - " ).addTo(map_a5cb9760638531fd5b14e75a46b3b1f1);\n", + " ).addTo(map_b6cf708d54aa4599b756d3ce62516c9c);\n", " \n", " \n", - " var icon_2a7dbd406a91da668bcdaab591f6b9e1 = L.AwesomeMarkers.icon(\n", + " var icon_05c025ef9bca364192e682fa3b0133d4 = L.AwesomeMarkers.icon(\n", " {"extraClasses": "fa-rotate-0", "icon": "circle", "iconColor": "white", "markerColor": "red", "prefix": "fa"}\n", " );\n", - " marker_a4d31af8d6a545d85c81795dd2d38ffe.setIcon(icon_2a7dbd406a91da668bcdaab591f6b9e1);\n", + " marker_b3dbf16a30f95c6a552145a7c117e057.setIcon(icon_05c025ef9bca364192e682fa3b0133d4);\n", " \n", " \n", - " marker_a4d31af8d6a545d85c81795dd2d38ffe.bindTooltip(\n", + " marker_b3dbf16a30f95c6a552145a7c117e057.bindTooltip(\n", " `<div>\n", " Destination\n", " </div>`,\n", @@ -861,13 +869,13 @@ " );\n", " \n", " \n", - " var poly_line_812484b9eb27ef423783fc9826e4c79d = L.polyline(\n", + " var poly_line_b88057c0260a68bf7b601e9b657ca271 = L.polyline(\n", " [[39.77902603149414, -104.96916961669922], [39.77896499633789, -104.96910095214844], [39.77878189086914, -104.96887969970703], [39.778751373291016, -104.96884155273438], [39.778587341308594, -104.9686508178711], [39.778480529785156, -104.96852111816406], [39.77839660644531, -104.96841430664062], [39.778324127197266, -104.96851348876953], [39.778228759765625, -104.96863555908203], [39.778076171875, -104.96883392333984], [39.777626037597656, -104.96941375732422], [39.7774772644043, -104.96961212158203], [39.77703857421875, -104.97017669677734], [39.776737213134766, -104.97056579589844], [39.77666473388672, -104.97066497802734], [39.7766227722168, -104.97071838378906], [39.776371002197266, -104.97103881835938], [39.77608108520508, -104.9714126586914], [39.775978088378906, -104.97154998779297], [39.774810791015625, -104.97306060791016], [39.77475357055664, -104.97313690185547], [39.77467346191406, -104.97323608398438], [39.77459716796875, -104.97333526611328], [39.773738861083984, -104.97444915771484], [39.77363586425781, -104.97457885742188], [39.773353576660156, -104.97494506835938], [39.77303695678711, -104.97535705566406], [39.77297592163086, -104.9754409790039], [39.77290344238281, -104.97553253173828], [39.772544860839844, -104.97599029541016], [39.772335052490234, -104.97626495361328], [39.77207946777344, -104.97659301757812], [39.771976470947266, -104.97673034667969], [39.771705627441406, -104.97708129882812], [39.77125549316406, -104.97766876220703], [39.77039337158203, -104.9787826538086], [39.77033233642578, -104.9788589477539], [39.770263671875, -104.97895050048828], [39.769996643066406, -104.97929382324219], [39.76996994018555, -104.9793930053711], [39.76955032348633, -104.97993469238281], [39.769287109375, -104.98027801513672], [39.7691764831543, -104.98041534423828], [39.769039154052734, -104.98059844970703], [39.76881408691406, -104.98088836669922], [39.768733978271484, -104.98099517822266], [39.76864242553711, -104.98110961914062], [39.76858901977539, -104.9811782836914], [39.76840591430664, -104.98141479492188], [39.7678337097168, -104.98216247558594], [39.76764678955078, -104.9823989868164], [39.76701736450195, -104.98321533203125], [39.766876220703125, -104.9833984375], [39.766693115234375, -104.98363494873047], [39.766624450683594, -104.98372650146484], [39.76657485961914, -104.9837875366211], [39.76653289794922, -104.98384857177734], [39.766178131103516, -104.98429870605469], [39.766021728515625, -104.98450469970703], [39.76557159423828, -104.98507690429688], [39.76506423950195, -104.98573303222656], [39.76464080810547, -104.98627471923828], [39.764495849609375, -104.98646545410156], [39.76442337036133, -104.98655700683594], [39.764347076416016, -104.98664855957031], [39.763763427734375, -104.98738861083984], [39.76366424560547, -104.98750305175781], [39.763572692871094, -104.98760223388672], [39.76346969604492, -104.98770141601562], [39.76337814331055, -104.98778533935547], [39.76326370239258, -104.98786926269531], [39.76316452026367, -104.9879379272461], [39.76304626464844, -104.98800659179688], [39.762916564941406, -104.98807525634766], [39.762813568115234, -104.98812103271484], [39.76266860961914, -104.98816680908203], [39.76252746582031, -104.98819732666016], [39.762413024902344, -104.98820495605469], [39.762290954589844, -104.98821258544922], [39.76216125488281, -104.98821258544922], [39.762062072753906, -104.98820495605469], [39.76179885864258, -104.9881591796875], [39.76161575317383, -104.98812103271484], [39.76142501831055, -104.98808288574219], [39.76121520996094, -104.98804473876953], [39.75983428955078, -104.98776245117188], [39.75953674316406, -104.98770141601562], [39.75946807861328, -104.98763275146484], [39.75936508178711, -104.98761749267578], [39.7588996887207, -104.987548828125], [39.758426666259766, -104.98747253417969], [39.758270263671875, -104.9874496459961], [39.75812911987305, -104.98744201660156], [39.75796127319336, -104.98744201660156], [39.75767517089844, -104.9874267578125], [39.75761032104492, -104.9874267578125], [39.75736618041992, -104.98741912841797], [39.757179260253906, -104.98741149902344], [39.757102966308594, -104.98741149902344], [39.756961822509766, -104.98741149902344], [39.756805419921875, -104.9874038696289], [39.756404876708984, -104.9874038696289], [39.75601577758789, -104.9874038696289], [39.75578308105469, -104.98739624023438], [39.75563049316406, -104.98739624023438], [39.755496978759766, -104.98739624023438], [39.75507736206055, -104.98738861083984], [39.75468063354492, -104.98738861083984], [39.75453186035156, -104.98738098144531], [39.754398345947266, -104.98738098144531], [39.754364013671875, -104.98738098144531], [39.75428009033203, -104.98738861083984], [39.75388717651367, -104.98738861083984], [39.753604888916016, -104.98738861083984], [39.75343704223633, -104.98738861083984], [39.7532844543457, -104.98738861083984], [39.75323486328125, -104.98738861083984], [39.75295639038086, -104.98738861083984], [39.752933502197266, -104.98738861083984], [39.752662658691406, -104.98738861083984], [39.752262115478516, -104.98738861083984], [39.75224304199219, -104.98738098144531], [39.751827239990234, -104.98738098144531], [39.75173568725586, -104.98738098144531], [39.751583099365234, -104.98738098144531], [39.75148391723633, -104.98738098144531], [39.751014709472656, -104.98738098144531], [39.750850677490234, -104.98738098144531], [39.75082015991211, -104.98738098144531], [39.750736236572266, -104.98738098144531], [39.750389099121094, -104.98737335205078], [39.75023651123047, -104.98737335205078], [39.7501106262207, -104.98737335205078], [39.7497673034668, -104.98737335205078], [39.74958419799805, -104.98737335205078], [39.74951171875, -104.98737335205078], [39.749412536621094, -104.98737335205078], [39.74927520751953, -104.98738098144531], [39.749168395996094, -104.98738098144531], [39.74892044067383, -104.98738098144531], [39.74868392944336, -104.98738098144531], [39.74851989746094, -104.98738098144531], [39.74822235107422, -104.98738098144531], [39.74802017211914, -104.98738098144531], [39.7476692199707, -104.98738098144531], [39.74758529663086, -104.98738098144531], [39.74751281738281, -104.98738098144531], [39.7474365234375, -104.98738098144531], [39.747379302978516, -104.98738098144531], [39.74735641479492, -104.98738098144531], [39.74730682373047, -104.98738098144531], [39.74697494506836, -104.98738098144531], [39.746944427490234, -104.98738098144531], [39.74678039550781, -104.98738861083984], [39.74667739868164, -104.98738861083984], [39.74653244018555, -104.98738861083984], [39.746482849121094, -104.98738861083984], [39.74626541137695, -104.98738861083984], [39.74618148803711, -104.98738861083984], [39.74605178833008, -104.98738861083984], [39.745948791503906, -104.98738861083984], [39.74554443359375, -104.98738861083984], [39.745418548583984, -104.98738861083984], [39.74506378173828, -104.98739624023438], [39.74480438232422, -104.98739624023438], [39.74473571777344, -104.9874038696289], [39.74430847167969, -104.98739624023438], [39.743526458740234, -104.98739624023438], [39.74336242675781, -104.98739624023438], [39.74320602416992, -104.98739624023438], [39.74302291870117, -104.98739624023438], [39.74254608154297, -104.98739624023438], [39.74245834350586, -104.98739624023438], [39.74220275878906, -104.98739624023438], [39.74214172363281, -104.98738861083984], [39.74182891845703, -104.98738861083984], [39.74177932739258, -104.98738861083984], [39.7416877746582, -104.98738861083984], [39.74163818359375, -104.98738098144531], [39.74156188964844, -104.98738861083984], [39.74149703979492, -104.98738861083984], [39.741416931152344, -104.98738861083984], [39.74137496948242, -104.98738861083984], [39.74129867553711, -104.98738861083984], [39.7410888671875, -104.98738861083984], [39.74089050292969, -104.98739624023438], [39.740875244140625, -104.98739624023438], [39.74071502685547, -104.98739624023438], [39.740631103515625, -104.98739624023438], [39.740478515625, -104.98739624023438], [39.74021530151367, -104.9874038696289], [39.74014663696289, -104.9874038696289], [39.74000930786133, -104.9874038696289], [39.73992919921875, -104.9874038696289], [39.73912048339844, -104.98739624023438], [39.73855209350586, -104.98739624023438], [39.73845672607422, -104.98739624023438], [39.73835372924805, -104.98739624023438], [39.73801040649414, -104.98739624023438], [39.73710250854492, -104.98738861083984], [39.73695373535156, -104.98738861083984], [39.736900329589844, -104.98738861083984], [39.736846923828125, -104.98738861083984], [39.73676300048828, -104.98738861083984], [39.73644256591797, -104.98739624023438], [39.73640441894531, -104.98739624023438], [39.736244201660156, -104.98739624023438], [39.73619079589844, -104.98739624023438], [39.73557662963867, -104.98739624023438], [39.73536682128906, -104.98739624023438], [39.73527526855469, -104.98739624023438], [39.73518753051758, -104.98739624023438], [39.73509216308594, -104.98739624023438], [39.73466110229492, -104.98739624023438], [39.734039306640625, -104.9874038696289], [39.73394775390625, -104.98739624023438], [39.733848571777344, -104.9874038696289], [39.73375701904297, -104.9874038696289], [39.73365020751953, -104.9874038696289], [39.73337173461914, -104.9874038696289], [39.733306884765625, -104.9874038696289], [39.732810974121094, -104.9874038696289], [39.732444763183594, -104.98741149902344], [39.732147216796875, -104.98741149902344], [39.7320556640625, -104.98741149902344], [39.731964111328125, -104.98741149902344], [39.73185348510742, -104.98741149902344], [39.73183059692383, -104.98741149902344], [39.73168182373047, -104.98741149902344], [39.73122024536133, -104.98741912841797], [39.731178283691406, -104.98741912841797], [39.73095703125, -104.98741912841797], [39.73054885864258, -104.98741912841797], [39.73045349121094, -104.98741912841797], [39.73037338256836, -104.98741912841797], [39.730010986328125, -104.98741912841797], [39.729522705078125, -104.98741912841797], [39.72943115234375, -104.98741912841797], [39.72917175292969, -104.98741912841797], [39.72908020019531, -104.98741912841797], [39.7289924621582, -104.98741912841797], [39.728668212890625, -104.98741912841797], [39.72862243652344, -104.98741912841797], [39.72829055786133, -104.98741912841797], [39.727420806884766, -104.98741912841797], [39.727298736572266, -104.98741912841797], [39.72718048095703, -104.98741912841797], [39.72665786743164, -104.9874267578125], [39.72663497924805, -104.9874267578125], [39.726600646972656, -104.9874267578125], [39.726375579833984, -104.98743438720703], [39.726226806640625, -104.98744201660156], [39.726112365722656, -104.98744201660156], [39.72574996948242, -104.98745727539062], [39.72563934326172, -104.98746490478516], [39.72555923461914, -104.98746490478516], [39.725223541259766, -104.98747253417969], [39.7251091003418, -104.98747253417969], [39.724884033203125, -104.98748016357422], [39.72444152832031, -104.98748779296875], [39.72422409057617, -104.98748779296875], [39.72414779663086, -104.98748779296875], [39.72405242919922, -104.98748779296875], [39.72385025024414, -104.98748779296875], [39.72378921508789, -104.98748779296875], [39.72360610961914, -104.98748779296875], [39.72319793701172, -104.98750305175781], [39.723026275634766, -104.98751068115234], [39.722869873046875, -104.98751068115234], [39.72262191772461, -104.98751068115234], [39.72255325317383, -104.98751068115234], [39.72245788574219, -104.98751068115234], [39.72215270996094, -104.98750305175781], [39.72206497192383, -104.98750305175781], [39.72195816040039, -104.98749542236328], [39.7216911315918, -104.98748779296875], [39.72103500366211, -104.98748779296875], [39.720943450927734, -104.98748779296875], [39.720863342285156, -104.98748779296875], [39.71990966796875, -104.98748779296875], [39.71987533569336, -104.98748779296875], [39.719451904296875, -104.98748779296875], [39.7193603515625, -104.98748779296875], [39.71928787231445, -104.98748779296875], [39.71902847290039, -104.98748779296875], [39.71844482421875, -104.98748779296875], [39.71834945678711, -104.98748779296875], [39.7182731628418, -104.98748779296875], [39.71819305419922, -104.98748779296875], [39.71754455566406, -104.98748779296875], [39.717464447021484, -104.98748779296875], [39.7174186706543, -104.98748779296875], [39.716644287109375, -104.98748779296875], [39.71656799316406, -104.98748779296875], [39.71649932861328, -104.98748779296875], [39.716224670410156, -104.98749542236328], [39.71613311767578, -104.98749542236328], [39.71575927734375, -104.98750305175781], [39.71568298339844, -104.98750305175781], [39.715606689453125, -104.98750305175781], [39.71514892578125, -104.98751831054688], [39.714847564697266, -104.98753356933594], [39.71476364135742, -104.98753356933594], [39.71470260620117, -104.98753356933594], [39.713897705078125, -104.987548828125], [39.713294982910156, -104.987548828125], [39.71319580078125, -104.987548828125], [39.71306610107422, -104.98755645751953], [39.71297836303711, -104.98755645751953], [39.71291732788086, -104.98755645751953], [39.712486267089844, -104.98756408691406], [39.71209716796875, -104.9875717163086], [39.71201705932617, -104.98757934570312], [39.71174240112305, -104.98757934570312], [39.711578369140625, -104.98758697509766], [39.711280822753906, -104.98759460449219], [39.71114730834961, -104.98759460449219], [39.71101379394531, -104.98759460449219], [39.71085739135742, -104.98759460449219], [39.71025085449219, -104.98760986328125], [39.71002960205078, -104.98760986328125], [39.70989227294922, -104.98761749267578], [39.70948028564453, -104.98761749267578], [39.70942306518555, -104.98760986328125], [39.70934295654297, -104.98761749267578], [39.709259033203125, -104.98761749267578], [39.70914840698242, -104.98762512207031], [39.7091064453125, -104.98762512207031], [39.70846939086914, -104.9876480102539], [39.70835876464844, -104.9876480102539], [39.70762634277344, -104.9876480102539], [39.707542419433594, -104.9876480102539], [39.707462310791016, -104.98764038085938], [39.7066535949707, -104.98760223388672], [39.70616912841797, -104.98758697509766], [39.705841064453125, -104.9875717163086], [39.70573806762695, -104.98756408691406], [39.705665588378906, -104.98755645751953], [39.705406188964844, -104.98754119873047], [39.705223083496094, -104.98753356933594], [39.705101013183594, -104.9875259399414], [39.704872131347656, -104.9875259399414], [39.7047119140625, -104.9875259399414], [39.704627990722656, -104.98751831054688], [39.70394515991211, -104.98751068115234], [39.703880310058594, -104.98751068115234], [39.70368194580078, -104.98750305175781], [39.70335006713867, -104.98750305175781], [39.70302963256836, -104.98750305175781], [39.7028694152832, -104.98750305175781], [39.70270538330078, -104.98751831054688], [39.70259475708008, -104.9875259399414], [39.702491760253906, -104.98753356933594], [39.70227813720703, -104.987548828125], [39.7021369934082, -104.98755645751953], [39.70204162597656, -104.98756408691406], [39.700843811035156, -104.98760986328125], [39.700782775878906, -104.98749542236328], [39.70073318481445, -104.9874038696289], [39.700687408447266, -104.98731994628906], [39.700660705566406, -104.98727416992188], [39.70061111450195, -104.9871826171875], [39.70054626464844, -104.98707580566406], [39.70036315917969, -104.98675537109375], [39.70024490356445, -104.986572265625], [39.70008850097656, -104.98636627197266], [39.700042724609375, -104.98631286621094], [39.69866180419922, -104.9847412109375], [39.6981086730957, -104.98410034179688], [39.69713592529297, -104.98299407958984], [39.696075439453125, -104.98184967041016], [39.69560623168945, -104.9813003540039], [39.69524383544922, -104.98087310791016], [39.69488525390625, -104.98043823242188], [39.694786071777344, -104.98033905029297], [39.6947021484375, -104.98025512695312], [39.69462585449219, -104.98019409179688], [39.6945915222168, -104.98016357421875], [39.69448471069336, -104.98009490966797], [39.69437789916992, -104.98002624511719], [39.6942024230957, -104.97992706298828], [39.69398498535156, -104.97977447509766], [39.69377899169922, -104.97953033447266], [39.69367599487305, -104.97943115234375], [39.69364929199219, -104.97940063476562], [39.69361114501953, -104.97936248779297], [39.69355010986328, -104.97931671142578], [39.6934814453125, -104.9792709350586], [39.69330978393555, -104.97918701171875], [39.69325637817383, -104.97914123535156], [39.693199157714844, -104.97909545898438], [39.693145751953125, -104.97903442382812], [39.693092346191406, -104.97897338867188], [39.69303512573242, -104.97886657714844], [39.69300079345703, -104.97877502441406], [39.69298553466797, -104.97871398925781], [39.69298553466797, -104.97850036621094], [39.69298553466797, -104.97827911376953], [39.69298553466797, -104.97792053222656], [39.6929931640625, -104.9774398803711], [39.6929931640625, -104.9773941040039], [39.6929931640625, -104.9772720336914], [39.692989349365234, -104.97704315185547], [39.692989349365234, -104.97694396972656], [39.692989349365234, -104.97685241699219], [39.69298553466797, -104.97636413574219], [39.6929817199707, -104.97577667236328]],\n", " {"bubblingMouseEvents": true, "color": "green", "dashArray": null, "dashOffset": null, "fill": false, "fillColor": "green", "fillOpacity": 0.2, "fillRule": "evenodd", "lineCap": "round", "lineJoin": "round", "noClip": false, "opacity": 0.8, "smoothFactor": 1.0, "stroke": true, "weight": 10}\n", - " ).addTo(map_a5cb9760638531fd5b14e75a46b3b1f1);\n", + " ).addTo(map_b6cf708d54aa4599b756d3ce62516c9c);\n", " \n", " \n", - " poly_line_812484b9eb27ef423783fc9826e4c79d.bindTooltip(\n", + " poly_line_b88057c0260a68bf7b601e9b657ca271.bindTooltip(\n", " `<div>\n", " Least Energy\n", " </div>`,\n", @@ -875,19 +883,19 @@ " );\n", " \n", " \n", - " var marker_001ab4478d92ad31bfd4defff343e8ea = L.marker(\n", + " var marker_51449498af2992f67a03ee1a4743b9b3 = L.marker(\n", " [39.77902603149414, -104.96916961669922],\n", " {}\n", - " ).addTo(map_a5cb9760638531fd5b14e75a46b3b1f1);\n", + " ).addTo(map_b6cf708d54aa4599b756d3ce62516c9c);\n", " \n", " \n", - " var icon_2ab048e082a6330258a4a74b57a185dc = L.AwesomeMarkers.icon(\n", + " var icon_0d220eb467754e9f97fb970647ef051a = L.AwesomeMarkers.icon(\n", " {"extraClasses": "fa-rotate-0", "icon": "circle", "iconColor": "white", "markerColor": "green", "prefix": "fa"}\n", " );\n", - " marker_001ab4478d92ad31bfd4defff343e8ea.setIcon(icon_2ab048e082a6330258a4a74b57a185dc);\n", + " marker_51449498af2992f67a03ee1a4743b9b3.setIcon(icon_0d220eb467754e9f97fb970647ef051a);\n", " \n", " \n", - " marker_001ab4478d92ad31bfd4defff343e8ea.bindTooltip(\n", + " marker_51449498af2992f67a03ee1a4743b9b3.bindTooltip(\n", " `<div>\n", " Origin\n", " </div>`,\n", @@ -895,19 +903,19 @@ " );\n", " \n", " \n", - " var marker_f1c968e33b632da1bdd369a1c4c1d2eb = L.marker(\n", + " var marker_0844385ced7afed6079e777212cdc756 = L.marker(\n", " [39.6929817199707, -104.97577667236328],\n", " {}\n", - " ).addTo(map_a5cb9760638531fd5b14e75a46b3b1f1);\n", + " ).addTo(map_b6cf708d54aa4599b756d3ce62516c9c);\n", " \n", " \n", - " var icon_8497e7de14e7e225b487c736ec56ea06 = L.AwesomeMarkers.icon(\n", + " var icon_c89925817012d0ad601284ad1f049e9e = L.AwesomeMarkers.icon(\n", " {"extraClasses": "fa-rotate-0", "icon": "circle", "iconColor": "white", "markerColor": "red", "prefix": "fa"}\n", " );\n", - " marker_f1c968e33b632da1bdd369a1c4c1d2eb.setIcon(icon_8497e7de14e7e225b487c736ec56ea06);\n", + " marker_0844385ced7afed6079e777212cdc756.setIcon(icon_c89925817012d0ad601284ad1f049e9e);\n", " \n", " \n", - " marker_f1c968e33b632da1bdd369a1c4c1d2eb.bindTooltip(\n", + " marker_0844385ced7afed6079e777212cdc756.bindTooltip(\n", " `<div>\n", " Destination\n", " </div>`,\n", @@ -915,13 +923,13 @@ " );\n", " \n", " \n", - " var poly_line_6b1b8ea217704dede06972c2196f427f = L.polyline(\n", - " [[39.77902603149414, -104.96916961669922], [39.77896499633789, -104.96910095214844], [39.77878189086914, -104.96887969970703], [39.778751373291016, -104.96884155273438], [39.778587341308594, -104.9686508178711], [39.778480529785156, -104.96852111816406], [39.77839660644531, -104.96841430664062], [39.778324127197266, -104.96851348876953], [39.778228759765625, -104.96863555908203], [39.778076171875, -104.96883392333984], [39.777626037597656, -104.96941375732422], [39.7774772644043, -104.96961212158203], [39.77703857421875, -104.97017669677734], [39.776737213134766, -104.97056579589844], [39.77666473388672, -104.97066497802734], [39.7766227722168, -104.97071838378906], [39.776371002197266, -104.97103881835938], [39.77608108520508, -104.9714126586914], [39.775978088378906, -104.97154998779297], [39.774810791015625, -104.97306060791016], [39.77475357055664, -104.97313690185547], [39.77467346191406, -104.97323608398438], [39.77459716796875, -104.97333526611328], [39.773738861083984, -104.97444915771484], [39.77363586425781, -104.97457885742188], [39.773353576660156, -104.97494506835938], [39.77303695678711, -104.97535705566406], [39.77297592163086, -104.9754409790039], [39.77287673950195, -104.97531127929688], [39.77263259887695, -104.9749984741211], [39.77257537841797, -104.97492218017578], [39.77229690551758, -104.97456359863281], [39.77223587036133, -104.9744873046875], [39.77199935913086, -104.97418212890625], [39.77191162109375, -104.97406768798828], [39.77155685424805, -104.9736099243164], [39.771270751953125, -104.97322845458984], [39.7706184387207, -104.97236633300781], [39.77056884765625, -104.97230529785156], [39.77048110961914, -104.97219848632812], [39.77042770385742, -104.9721450805664], [39.7703742980957, -104.97209167480469], [39.77033233642578, -104.97205352783203], [39.770233154296875, -104.97198486328125], [39.770118713378906, -104.9721450805664], [39.76996994018555, -104.97232818603516], [39.76978302001953, -104.97257232666016], [39.76957321166992, -104.97283172607422], [39.76926803588867, -104.97325134277344], [39.76918411254883, -104.97335815429688], [39.76904296875, -104.97335815429688], [39.76840591430664, -104.97335052490234], [39.76824188232422, -104.97335052490234], [39.7681770324707, -104.97335052490234], [39.7680549621582, -104.97335052490234], [39.7677001953125, -104.97335052490234], [39.767547607421875, -104.97335052490234], [39.7670783996582, -104.97335052490234], [39.76704788208008, -104.97335052490234], [39.766971588134766, -104.97335052490234], [39.76690673828125, -104.97335052490234], [39.76670455932617, -104.97335052490234], [39.76658248901367, -104.97335052490234], [39.766502380371094, -104.97335052490234], [39.76637649536133, -104.97335052490234], [39.766136169433594, -104.97335052490234], [39.76605987548828, -104.97335052490234], [39.765865325927734, -104.97335052490234], [39.76577377319336, -104.97335052490234], [39.76567459106445, -104.97335052490234], [39.76523208618164, -104.97335052490234], [39.76495361328125, -104.97335052490234], [39.764835357666016, -104.97335052490234], [39.7646598815918, -104.97335052490234], [39.764591217041016, -104.97335052490234], [39.7645149230957, -104.97335052490234], [39.764427185058594, -104.97335052490234], [39.76421356201172, -104.97335052490234], [39.764034271240234, -104.97335052490234], [39.76380920410156, -104.97335052490234], [39.763580322265625, -104.97335052490234], [39.76325988769531, -104.97335052490234], [39.76316833496094, -104.97339630126953], [39.763118743896484, -104.97339630126953], [39.76306915283203, -104.97335052490234], [39.76280212402344, -104.97335815429688], [39.762454986572266, -104.97335815429688], [39.76237869262695, -104.97335815429688], [39.76214599609375, -104.97335052490234], [39.762054443359375, -104.97335052490234], [39.76194381713867, -104.97335052490234], [39.76179122924805, -104.97335052490234], [39.76169967651367, -104.97335052490234], [39.76160430908203, -104.97335052490234], [39.76118087768555, -104.97335052490234], [39.76103973388672, -104.97335052490234], [39.76084518432617, -104.97335052490234], [39.76076126098633, -104.97335052490234], [39.76066589355469, -104.97335052490234], [39.760440826416016, -104.97335052490234], [39.76033401489258, -104.97335052490234], [39.759613037109375, -104.97335052490234], [39.7595100402832, -104.97335052490234], [39.75939178466797, -104.97339630126953], [39.759151458740234, -104.97339630126953], [39.75912094116211, -104.97335052490234], [39.759010314941406, -104.97335052490234], [39.758827209472656, -104.97335052490234], [39.75831604003906, -104.97335052490234], [39.75823211669922, -104.97335052490234], [39.758174896240234, -104.97335052490234], [39.75764083862305, -104.97335815429688], [39.7572021484375, -104.97335815429688], [39.75708770751953, -104.97335815429688], [39.75699234008789, -104.97335815429688], [39.75688171386719, -104.97335815429688], [39.75673294067383, -104.97335815429688], [39.75627899169922, -104.97335815429688], [39.75574493408203, -104.97335815429688], [39.75540542602539, -104.97335815429688], [39.75531005859375, -104.97335815429688], [39.75495147705078, -104.97335815429688], [39.75469970703125, -104.97335815429688], [39.754547119140625, -104.97335815429688], [39.75442886352539, -104.97335815429688], [39.753578186035156, -104.97335815429688], [39.75333786010742, -104.97335815429688], [39.75325012207031, -104.97335815429688], [39.75314712524414, -104.97335815429688], [39.75208282470703, -104.97335815429688], [39.751991271972656, -104.97335815429688], [39.751895904541016, -104.97335815429688], [39.75139236450195, -104.9733657836914], [39.75082778930664, -104.9733657836914], [39.750732421875, -104.9733657836914], [39.75063705444336, -104.9733657836914], [39.750423431396484, -104.9733657836914], [39.750282287597656, -104.9733657836914], [39.75018310546875, -104.9733657836914], [39.749576568603516, -104.9733657836914], [39.74947738647461, -104.9733657836914], [39.74938201904297, -104.9733657836914], [39.74884033203125, -104.9733657836914], [39.74821853637695, -104.97337341308594], [39.74812698364258, -104.97337341308594], [39.748016357421875, -104.97337341308594], [39.747798919677734, -104.97337341308594], [39.74757766723633, -104.97337341308594], [39.74747085571289, -104.97337341308594], [39.74736785888672, -104.97337341308594], [39.74721908569336, -104.97337341308594], [39.747066497802734, -104.97337341308594], [39.746952056884766, -104.97337341308594], [39.746864318847656, -104.97337341308594], [39.74663162231445, -104.97337341308594], [39.746551513671875, -104.97338104248047], [39.746482849121094, -104.97338104248047], [39.7460823059082, -104.97338104248047], [39.74594497680664, -104.97338104248047], [39.74583435058594, -104.97338104248047], [39.74509048461914, -104.97340393066406], [39.745052337646484, -104.97340393066406], [39.745018005371094, -104.97344207763672], [39.744972229003906, -104.97344970703125], [39.744937896728516, -104.97344970703125], [39.744834899902344, -104.97339630126953], [39.744834899902344, -104.97355651855469], [39.744834899902344, -104.97415924072266], [39.74483871459961, -104.97454833984375], [39.74483871459961, -104.97476196289062], [39.74464416503906, -104.97476196289062], [39.744606018066406, -104.97476196289062], [39.74433517456055, -104.97476959228516], [39.74381637573242, -104.97476959228516], [39.74373245239258, -104.97476959228516], [39.743377685546875, -104.97477722167969], [39.74324417114258, -104.97477722167969], [39.743133544921875, -104.97477722167969], [39.7426872253418, -104.97477722167969], [39.74175262451172, -104.97477722167969], [39.74166488647461, -104.97477722167969], [39.74156951904297, -104.97477722167969], [39.7412109375, -104.97478485107422], [39.740718841552734, -104.97479248046875], [39.74055862426758, -104.97479248046875], [39.74032211303711, -104.97479248046875], [39.74012756347656, -104.97480010986328], [39.740020751953125, -104.97480010986328], [39.740020751953125, -104.9746322631836], [39.740020751953125, -104.97451782226562], [39.74002456665039, -104.97415924072266], [39.73991775512695, -104.97415924072266], [39.73970031738281, -104.97415924072266], [39.73955535888672, -104.97415161132812], [39.739315032958984, -104.97415161132812], [39.73896789550781, -104.97415161132812], [39.738609313964844, -104.97415161132812], [39.73850631713867, -104.9741439819336], [39.73841857910156, -104.9741439819336], [39.73833465576172, -104.9741439819336], [39.736942291259766, -104.9741439819336], [39.736881256103516, -104.9741439819336], [39.73681640625, -104.9741439819336], [39.73567581176758, -104.97412109375], [39.735252380371094, -104.97411346435547], [39.735164642333984, -104.97411346435547], [39.735069274902344, -104.97411346435547], [39.73379898071289, -104.97410583496094], [39.73370361328125, -104.97410583496094], [39.733612060546875, -104.97410583496094], [39.73240280151367, -104.97410583496094], [39.732357025146484, -104.97410583496094], [39.73214340209961, -104.97410583496094], [39.7320556640625, -104.97410583496094], [39.73196029663086, -104.97410583496094], [39.731422424316406, -104.97410583496094], [39.73128890991211, -104.97410583496094], [39.730960845947266, -104.97410583496094], [39.73088455200195, -104.97410583496094], [39.73067855834961, -104.97410583496094], [39.73054504394531, -104.97410583496094], [39.7304573059082, -104.97410583496094], [39.730369567871094, -104.97410583496094], [39.73027420043945, -104.97410583496094], [39.73003387451172, -104.97410583496094], [39.729976654052734, -104.97410583496094], [39.729896545410156, -104.97410583496094], [39.72964859008789, -104.97410583496094], [39.7294807434082, -104.97410583496094], [39.729434967041016, -104.97410583496094], [39.72917175292969, -104.97410583496094], [39.72907257080078, -104.97410583496094], [39.72898864746094, -104.97410583496094], [39.727378845214844, -104.9740982055664], [39.72727966308594, -104.9740982055664], [39.727195739746094, -104.9740982055664], [39.7259635925293, -104.9740982055664], [39.72566604614258, -104.97409057617188], [39.7255859375, -104.97409057617188], [39.72549819946289, -104.97409057617188], [39.725337982177734, -104.97409057617188], [39.7252082824707, -104.97409057617188], [39.724693298339844, -104.97409057617188], [39.72453689575195, -104.97409057617188], [39.72406768798828, -104.97408294677734], [39.72374725341797, -104.97408294677734], [39.72247314453125, -104.97408294677734], [39.72218322753906, -104.97408294677734], [39.72205352783203, -104.97408294677734], [39.721946716308594, -104.97408294677734], [39.72013473510742, -104.97406768798828], [39.719642639160156, -104.97406005859375], [39.71931457519531, -104.97406005859375], [39.71918869018555, -104.97406005859375], [39.718971252441406, -104.97405242919922], [39.71888732910156, -104.97404479980469], [39.71880340576172, -104.97403717041016], [39.71875, -104.97402954101562], [39.71870422363281, -104.9740219116211], [39.718650817871094, -104.97401428222656], [39.71857452392578, -104.97399139404297], [39.718505859375, -104.97396850585938], [39.71845245361328, -104.97393798828125], [39.71842956542969, -104.97392272949219], [39.71839141845703, -104.97390747070312], [39.71834945678711, -104.97386932373047], [39.718318939208984, -104.97382354736328], [39.71824264526367, -104.97368621826172], [39.71818923950195, -104.97358703613281], [39.71816635131836, -104.97354888916016], [39.71815490722656, -104.9735336303711], [39.71811294555664, -104.97348022460938], [39.71806335449219, -104.97343444824219], [39.71802520751953, -104.97341918945312], [39.717979431152344, -104.97340393066406], [39.717952728271484, -104.97339630126953], [39.71792221069336, -104.973388671875], [39.717899322509766, -104.973388671875], [39.71767807006836, -104.973388671875], [39.716922760009766, -104.973388671875], [39.71653366088867, -104.97339630126953], [39.71645736694336, -104.97339630126953], [39.71589660644531, -104.97339630126953], [39.71546936035156, -104.97341918945312], [39.71541976928711, -104.97341918945312], [39.71538543701172, -104.97342681884766], [39.71536636352539, -104.97342681884766], [39.71527862548828, -104.97342681884766], [39.7148551940918, -104.97344970703125], [39.71479415893555, -104.97344970703125], [39.71472930908203, -104.97345733642578], [39.714656829833984, -104.97344970703125], [39.71451950073242, -104.97344207763672], [39.71427917480469, -104.9734115600586], [39.712989807128906, -104.97340393066406], [39.71291732788086, -104.97340393066406], [39.71285629272461, -104.97340393066406], [39.71247482299805, -104.97340393066406], [39.711551666259766, -104.97340393066406], [39.71137237548828, -104.97340393066406], [39.711204528808594, -104.97340393066406], [39.71110534667969, -104.97340393066406], [39.711021423339844, -104.97340393066406], [39.7109489440918, -104.97340393066406], [39.710845947265625, -104.97340393066406], [39.709381103515625, -104.973388671875], [39.70929718017578, -104.973388671875], [39.707820892333984, -104.973388671875], [39.707557678222656, -104.973388671875], [39.70749282836914, -104.973388671875], [39.707420349121094, -104.973388671875], [39.70706558227539, -104.97338104248047], [39.70566940307617, -104.97337341308594], [39.70392990112305, -104.973388671875], [39.70386505126953, -104.973388671875], [39.70379638671875, -104.973388671875], [39.70212173461914, -104.97339630126953], [39.70206069946289, -104.97339630126953], [39.70030212402344, -104.97339630126953], [39.70021057128906, -104.97340393066406], [39.700138092041016, -104.97339630126953], [39.698429107666016, -104.97340393066406], [39.69660949707031, -104.9734115600586], [39.69625473022461, -104.97341918945312], [39.694740295410156, -104.97341918945312], [39.694149017333984, -104.97343444824219], [39.69339370727539, -104.97343444824219], [39.69325256347656, -104.97343444824219], [39.69306564331055, -104.97342681884766], [39.6929817199707, -104.97342681884766], [39.6929817199707, -104.97357177734375], [39.6929817199707, -104.97403717041016], [39.6929817199707, -104.974609375], [39.69298553466797, -104.97517395019531], [39.6929817199707, -104.97577667236328]],\n", + " var poly_line_cb6d42712bef935f49dac78318d38137 = L.polyline(\n", + " [[39.77902603149414, -104.96916961669922], [39.77896499633789, -104.96910095214844], [39.77878189086914, -104.96887969970703], [39.778751373291016, -104.96884155273438], [39.778587341308594, -104.9686508178711], [39.778480529785156, -104.96852111816406], [39.77839660644531, -104.96841430664062], [39.778324127197266, -104.96851348876953], [39.778228759765625, -104.96863555908203], [39.778076171875, -104.96883392333984], [39.777626037597656, -104.96941375732422], [39.7774772644043, -104.96961212158203], [39.77703857421875, -104.97017669677734], [39.776737213134766, -104.97056579589844], [39.77666473388672, -104.97066497802734], [39.7766227722168, -104.97071838378906], [39.776371002197266, -104.97103881835938], [39.77608108520508, -104.9714126586914], [39.775978088378906, -104.97154998779297], [39.774810791015625, -104.97306060791016], [39.77475357055664, -104.97313690185547], [39.77467346191406, -104.97323608398438], [39.77459716796875, -104.97333526611328], [39.773738861083984, -104.97444915771484], [39.77363586425781, -104.97457885742188], [39.773353576660156, -104.97494506835938], [39.77303695678711, -104.97535705566406], [39.77297592163086, -104.9754409790039], [39.77287673950195, -104.97531127929688], [39.77263259887695, -104.9749984741211], [39.77257537841797, -104.97492218017578], [39.77229690551758, -104.97456359863281], [39.77223587036133, -104.9744873046875], [39.77199935913086, -104.97418212890625], [39.77191162109375, -104.97406768798828], [39.77157974243164, -104.9736328125], [39.77155303955078, -104.9736328125], [39.77154541015625, -104.97361755371094], [39.7712516784668, -104.97325897216797], [39.77125549316406, -104.97321319580078], [39.77090072631836, -104.97274017333984], [39.7706184387207, -104.97236633300781], [39.77056884765625, -104.97230529785156], [39.77048110961914, -104.97219848632812], [39.77042770385742, -104.9721450805664], [39.7703742980957, -104.97209167480469], [39.77033233642578, -104.97205352783203], [39.770233154296875, -104.97198486328125], [39.770118713378906, -104.9721450805664], [39.76996994018555, -104.97232818603516], [39.76978302001953, -104.97257232666016], [39.76957321166992, -104.97283172607422], [39.76926803588867, -104.97325134277344], [39.76918411254883, -104.97335815429688], [39.76904296875, -104.97335815429688], [39.76840591430664, -104.97335052490234], [39.76824188232422, -104.97335052490234], [39.7681770324707, -104.97335052490234], [39.7680549621582, -104.97335052490234], [39.7677001953125, -104.97335052490234], [39.767547607421875, -104.97335052490234], [39.7670783996582, -104.97335052490234], [39.76704788208008, -104.97335052490234], [39.766971588134766, -104.97335052490234], [39.76690673828125, -104.97335052490234], [39.76670455932617, -104.97335052490234], [39.76658248901367, -104.97335052490234], [39.766502380371094, -104.97335052490234], [39.76637649536133, -104.97335052490234], [39.766136169433594, -104.97335052490234], [39.76605987548828, -104.97335052490234], [39.765865325927734, -104.97335052490234], [39.76577377319336, -104.97335052490234], [39.76567459106445, -104.97335052490234], [39.76523208618164, -104.97335052490234], [39.76495361328125, -104.97335052490234], [39.764835357666016, -104.97335052490234], [39.7646598815918, -104.97335052490234], [39.764591217041016, -104.97335052490234], [39.7645149230957, -104.97335052490234], [39.764427185058594, -104.97335052490234], [39.76421356201172, -104.97335052490234], [39.764034271240234, -104.97335052490234], [39.76380920410156, -104.97335052490234], [39.763580322265625, -104.97335052490234], [39.76325988769531, -104.97335052490234], [39.76316833496094, -104.97339630126953], [39.763118743896484, -104.97339630126953], [39.76306915283203, -104.97335052490234], [39.762874603271484, -104.97335815429688], [39.762454986572266, -104.97335815429688], [39.76237869262695, -104.97335815429688], [39.76214599609375, -104.97335052490234], [39.762054443359375, -104.97335052490234], [39.76194381713867, -104.97335052490234], [39.76179122924805, -104.97335052490234], [39.76169967651367, -104.97335052490234], [39.76160430908203, -104.97335052490234], [39.76118087768555, -104.97335052490234], [39.76103973388672, -104.97335052490234], [39.76084518432617, -104.97335052490234], [39.76076126098633, -104.97335052490234], [39.76066589355469, -104.97335052490234], [39.760440826416016, -104.97335052490234], [39.76033401489258, -104.97335052490234], [39.759613037109375, -104.97335052490234], [39.7595100402832, -104.97335052490234], [39.75939178466797, -104.97339630126953], [39.759151458740234, -104.97339630126953], [39.75912094116211, -104.97335052490234], [39.759010314941406, -104.97335052490234], [39.758827209472656, -104.97335052490234], [39.75831604003906, -104.97335052490234], [39.75823211669922, -104.97335052490234], [39.758174896240234, -104.97335052490234], [39.75764083862305, -104.97335815429688], [39.7572021484375, -104.97335815429688], [39.75708770751953, -104.97335815429688], [39.75699234008789, -104.97335815429688], [39.75688171386719, -104.97335815429688], [39.75673294067383, -104.97335815429688], [39.75627899169922, -104.97335815429688], [39.75574493408203, -104.97335815429688], [39.75540542602539, -104.97335815429688], [39.75531005859375, -104.97335815429688], [39.75495147705078, -104.97335815429688], [39.75469970703125, -104.97335815429688], [39.754547119140625, -104.97335815429688], [39.75442886352539, -104.97335815429688], [39.753578186035156, -104.97335815429688], [39.75333786010742, -104.97335815429688], [39.75325012207031, -104.97335815429688], [39.75314712524414, -104.97335815429688], [39.75208282470703, -104.97335815429688], [39.751991271972656, -104.97335815429688], [39.751895904541016, -104.97335815429688], [39.75139236450195, -104.9733657836914], [39.75082778930664, -104.9733657836914], [39.750732421875, -104.9733657836914], [39.75063705444336, -104.9733657836914], [39.750423431396484, -104.9733657836914], [39.750282287597656, -104.9733657836914], [39.75018310546875, -104.9733657836914], [39.749576568603516, -104.9733657836914], [39.74947738647461, -104.9733657836914], [39.74938201904297, -104.9733657836914], [39.74884033203125, -104.9733657836914], [39.74821853637695, -104.97337341308594], [39.74812698364258, -104.97337341308594], [39.748016357421875, -104.97337341308594], [39.747798919677734, -104.97337341308594], [39.74757766723633, -104.97337341308594], [39.74747085571289, -104.97337341308594], [39.74736785888672, -104.97337341308594], [39.74721908569336, -104.97337341308594], [39.747066497802734, -104.97337341308594], [39.746952056884766, -104.97337341308594], [39.746864318847656, -104.97337341308594], [39.74663162231445, -104.97337341308594], [39.746551513671875, -104.97338104248047], [39.746482849121094, -104.97338104248047], [39.7460823059082, -104.97338104248047], [39.74594497680664, -104.97338104248047], [39.74583435058594, -104.97338104248047], [39.74509048461914, -104.97340393066406], [39.745052337646484, -104.97340393066406], [39.745018005371094, -104.97344207763672], [39.744972229003906, -104.97344970703125], [39.744937896728516, -104.97344970703125], [39.744834899902344, -104.97339630126953], [39.744834899902344, -104.97355651855469], [39.744834899902344, -104.97415924072266], [39.74483871459961, -104.97454833984375], [39.74483871459961, -104.97476196289062], [39.74464416503906, -104.97476196289062], [39.744606018066406, -104.97476196289062], [39.74433517456055, -104.97476959228516], [39.74381637573242, -104.97476959228516], [39.74373245239258, -104.97476959228516], [39.743377685546875, -104.97477722167969], [39.74324417114258, -104.97477722167969], [39.743133544921875, -104.97477722167969], [39.7426872253418, -104.97477722167969], [39.74175262451172, -104.97477722167969], [39.74166488647461, -104.97477722167969], [39.74156951904297, -104.97477722167969], [39.7412109375, -104.97478485107422], [39.740718841552734, -104.97479248046875], [39.74055862426758, -104.97479248046875], [39.74032211303711, -104.97479248046875], [39.74012756347656, -104.97480010986328], [39.740020751953125, -104.97480010986328], [39.740020751953125, -104.9746322631836], [39.740020751953125, -104.97451782226562], [39.74002456665039, -104.97415924072266], [39.73991775512695, -104.97415924072266], [39.73970031738281, -104.97415924072266], [39.73955535888672, -104.97415161132812], [39.739315032958984, -104.97415161132812], [39.73896789550781, -104.97415161132812], [39.738609313964844, -104.97415161132812], [39.73850631713867, -104.9741439819336], [39.73841857910156, -104.9741439819336], [39.73833465576172, -104.9741439819336], [39.736942291259766, -104.9741439819336], [39.736881256103516, -104.9741439819336], [39.73681640625, -104.9741439819336], [39.73567581176758, -104.97412109375], [39.735252380371094, -104.97411346435547], [39.735164642333984, -104.97411346435547], [39.735069274902344, -104.97411346435547], [39.73379898071289, -104.97410583496094], [39.73370361328125, -104.97410583496094], [39.733612060546875, -104.97410583496094], [39.73240280151367, -104.97410583496094], [39.732357025146484, -104.97410583496094], [39.73214340209961, -104.97410583496094], [39.7320556640625, -104.97410583496094], [39.73196029663086, -104.97410583496094], [39.731422424316406, -104.97410583496094], [39.73128890991211, -104.97410583496094], [39.730960845947266, -104.97410583496094], [39.73088455200195, -104.97410583496094], [39.73067855834961, -104.97410583496094], [39.73054504394531, -104.97410583496094], [39.7304573059082, -104.97410583496094], [39.730369567871094, -104.97410583496094], [39.73027420043945, -104.97410583496094], [39.73003387451172, -104.97410583496094], [39.729976654052734, -104.97410583496094], [39.729896545410156, -104.97410583496094], [39.72964859008789, -104.97410583496094], [39.7294807434082, -104.97410583496094], [39.729434967041016, -104.97410583496094], [39.72917175292969, -104.97410583496094], [39.72907257080078, -104.97410583496094], [39.72898864746094, -104.97410583496094], [39.727378845214844, -104.9740982055664], [39.72727966308594, -104.9740982055664], [39.727195739746094, -104.9740982055664], [39.7259635925293, -104.9740982055664], [39.72566604614258, -104.97409057617188], [39.7255859375, -104.97409057617188], [39.72549819946289, -104.97409057617188], [39.725337982177734, -104.97409057617188], [39.7252082824707, -104.97409057617188], [39.724693298339844, -104.97409057617188], [39.72453689575195, -104.97409057617188], [39.72406768798828, -104.97408294677734], [39.72374725341797, -104.97408294677734], [39.72247314453125, -104.97408294677734], [39.72218322753906, -104.97408294677734], [39.72205352783203, -104.97408294677734], [39.721946716308594, -104.97408294677734], [39.72013473510742, -104.97406768798828], [39.719642639160156, -104.97406005859375], [39.71931457519531, -104.97406005859375], [39.71918869018555, -104.97406005859375], [39.718971252441406, -104.97405242919922], [39.71888732910156, -104.97404479980469], [39.71880340576172, -104.97403717041016], [39.71875, -104.97402954101562], [39.71870422363281, -104.9740219116211], [39.718650817871094, -104.97401428222656], [39.71857452392578, -104.97399139404297], [39.718505859375, -104.97396850585938], [39.71845245361328, -104.97393798828125], [39.71842956542969, -104.97392272949219], [39.71839141845703, -104.97390747070312], [39.71834945678711, -104.97386932373047], [39.718318939208984, -104.97382354736328], [39.71824264526367, -104.97368621826172], [39.71818923950195, -104.97358703613281], [39.71816635131836, -104.97354888916016], [39.71815490722656, -104.9735336303711], [39.71811294555664, -104.97348022460938], [39.71806335449219, -104.97343444824219], [39.71802520751953, -104.97341918945312], [39.717979431152344, -104.97340393066406], [39.717952728271484, -104.97339630126953], [39.71792221069336, -104.973388671875], [39.717899322509766, -104.973388671875], [39.71767807006836, -104.973388671875], [39.716922760009766, -104.973388671875], [39.71653366088867, -104.97339630126953], [39.71645736694336, -104.97339630126953], [39.71589660644531, -104.97339630126953], [39.71546936035156, -104.97341918945312], [39.71541976928711, -104.97341918945312], [39.71538543701172, -104.97342681884766], [39.71536636352539, -104.97342681884766], [39.71527862548828, -104.97342681884766], [39.7148551940918, -104.97344970703125], [39.71479415893555, -104.97344970703125], [39.71472930908203, -104.97345733642578], [39.714656829833984, -104.97344970703125], [39.71451950073242, -104.97344207763672], [39.71427917480469, -104.9734115600586], [39.712989807128906, -104.97340393066406], [39.71291732788086, -104.97340393066406], [39.71285629272461, -104.97340393066406], [39.71247482299805, -104.97340393066406], [39.711551666259766, -104.97340393066406], [39.71137237548828, -104.97340393066406], [39.711204528808594, -104.97340393066406], [39.71110534667969, -104.97340393066406], [39.711021423339844, -104.97340393066406], [39.7109489440918, -104.97340393066406], [39.710845947265625, -104.97340393066406], [39.709381103515625, -104.973388671875], [39.70929718017578, -104.973388671875], [39.707820892333984, -104.973388671875], [39.707557678222656, -104.973388671875], [39.70749282836914, -104.973388671875], [39.707420349121094, -104.973388671875], [39.70706558227539, -104.97338104248047], [39.70566940307617, -104.97337341308594], [39.70392990112305, -104.973388671875], [39.70386505126953, -104.973388671875], [39.70379638671875, -104.973388671875], [39.70212173461914, -104.97339630126953], [39.70206069946289, -104.97339630126953], [39.70030212402344, -104.97339630126953], [39.70021057128906, -104.97340393066406], [39.700138092041016, -104.97339630126953], [39.698429107666016, -104.97340393066406], [39.69660949707031, -104.9734115600586], [39.69625473022461, -104.97341918945312], [39.694740295410156, -104.97341918945312], [39.694149017333984, -104.97343444824219], [39.69339370727539, -104.97343444824219], [39.69325256347656, -104.97343444824219], [39.69306564331055, -104.97342681884766], [39.6929817199707, -104.97342681884766], [39.6929817199707, -104.97357177734375], [39.6929817199707, -104.97403717041016], [39.6929817199707, -104.974609375], [39.69298553466797, -104.97517395019531], [39.6929817199707, -104.97577667236328]],\n", " {"bubblingMouseEvents": true, "color": "blue", "dashArray": null, "dashOffset": null, "fill": false, "fillColor": "blue", "fillOpacity": 0.2, "fillRule": "evenodd", "lineCap": "round", "lineJoin": "round", "noClip": false, "opacity": 0.8, "smoothFactor": 1.0, "stroke": true, "weight": 10}\n", - " ).addTo(map_a5cb9760638531fd5b14e75a46b3b1f1);\n", + " ).addTo(map_b6cf708d54aa4599b756d3ce62516c9c);\n", " \n", " \n", - " poly_line_6b1b8ea217704dede06972c2196f427f.bindTooltip(\n", + " poly_line_cb6d42712bef935f49dac78318d38137.bindTooltip(\n", " `<div>\n", " Least Cost\n", " </div>`,\n", @@ -929,19 +937,19 @@ " );\n", " \n", " \n", - " var marker_aa49883cffaf0f523687a8cb8e017e8d = L.marker(\n", + " var marker_d7e8d51bdfa0b414359dba4b6850173d = L.marker(\n", " [39.77902603149414, -104.96916961669922],\n", " {}\n", - " ).addTo(map_a5cb9760638531fd5b14e75a46b3b1f1);\n", + " ).addTo(map_b6cf708d54aa4599b756d3ce62516c9c);\n", " \n", " \n", - " var icon_fbf3a814c59b3c7dc86707db6d25d53a = L.AwesomeMarkers.icon(\n", + " var icon_777c8c4fb41598cadb9bc5aa82e9fc8e = L.AwesomeMarkers.icon(\n", " {"extraClasses": "fa-rotate-0", "icon": "circle", "iconColor": "white", "markerColor": "green", "prefix": "fa"}\n", " );\n", - " marker_aa49883cffaf0f523687a8cb8e017e8d.setIcon(icon_fbf3a814c59b3c7dc86707db6d25d53a);\n", + " marker_d7e8d51bdfa0b414359dba4b6850173d.setIcon(icon_777c8c4fb41598cadb9bc5aa82e9fc8e);\n", " \n", " \n", - " marker_aa49883cffaf0f523687a8cb8e017e8d.bindTooltip(\n", + " marker_d7e8d51bdfa0b414359dba4b6850173d.bindTooltip(\n", " `<div>\n", " Origin\n", " </div>`,\n", @@ -949,19 +957,19 @@ " );\n", " \n", " \n", - " var marker_0652f42a9bda08b402dea3440ab7709e = L.marker(\n", + " var marker_cd94423e998833f6c8afd0bb042a8e37 = L.marker(\n", " [39.6929817199707, -104.97577667236328],\n", " {}\n", - " ).addTo(map_a5cb9760638531fd5b14e75a46b3b1f1);\n", + " ).addTo(map_b6cf708d54aa4599b756d3ce62516c9c);\n", " \n", " \n", - " var icon_2e33d13a04a5446fc4e8dc8f2295adc2 = L.AwesomeMarkers.icon(\n", + " var icon_67a4dd8906dac98fef7f4c548ce0a4fe = L.AwesomeMarkers.icon(\n", " {"extraClasses": "fa-rotate-0", "icon": "circle", "iconColor": "white", "markerColor": "red", "prefix": "fa"}\n", " );\n", - " marker_0652f42a9bda08b402dea3440ab7709e.setIcon(icon_2e33d13a04a5446fc4e8dc8f2295adc2);\n", + " marker_cd94423e998833f6c8afd0bb042a8e37.setIcon(icon_67a4dd8906dac98fef7f4c548ce0a4fe);\n", " \n", " \n", - " marker_0652f42a9bda08b402dea3440ab7709e.bindTooltip(\n", + " marker_cd94423e998833f6c8afd0bb042a8e37.bindTooltip(\n", " `<div>\n", " Destination\n", " </div>`,\n", @@ -972,10 +980,10 @@ "</html>\" style=\"position:absolute;width:100%;height:100%;left:0;top:0;border:none !important;\" allowfullscreen webkitallowfullscreen mozallowfullscreen></iframe></div></div>" ], "text/plain": [ - "<folium.folium.Map at 0x129306dd0>" + "<folium.folium.Map at 0x1b03568b0>" ] }, - "execution_count": 187, + "execution_count": 17, "metadata": {}, "output_type": "execute_result" } @@ -997,7 +1005,7 @@ }, { "cell_type": "code", - "execution_count": 188, + "execution_count": 18, "id": "aa1e6b11", "metadata": {}, "outputs": [ @@ -1031,7 +1039,7 @@ " <meta name="viewport" content="width=device-width,\n", " initial-scale=1.0, maximum-scale=1.0, user-scalable=no" />\n", " <style>\n", - " #map_63d14900899a70183befbf30d354febd {\n", + " #map_77bceaf15f8f92685ff07972bb265db5 {\n", " position: relative;\n", " width: 100.0%;\n", " height: 100.0%;\n", @@ -1045,14 +1053,14 @@ "<body>\n", " \n", " \n", - " <div class="folium-map" id="map_63d14900899a70183befbf30d354febd" ></div>\n", + " <div class="folium-map" id="map_77bceaf15f8f92685ff07972bb265db5" ></div>\n", " \n", "</body>\n", "<script>\n", " \n", " \n", - " var map_63d14900899a70183befbf30d354febd = L.map(\n", - " "map_63d14900899a70183befbf30d354febd",\n", + " var map_77bceaf15f8f92685ff07972bb265db5 = L.map(\n", + " "map_77bceaf15f8f92685ff07972bb265db5",\n", " {\n", " center: [39.73823928833008, -105.01406860351562],\n", " crs: L.CRS.EPSG3857,\n", @@ -1066,39 +1074,39 @@ "\n", " \n", " \n", - " var tile_layer_9092626b7ec75f73d91deb0b7833c4fb = L.tileLayer(\n", + " var tile_layer_63693c84a48315790e7524f1adf3fbe4 = L.tileLayer(\n", " "https://{s}.tile.openstreetmap.org/{z}/{x}/{y}.png",\n", " {"attribution": "Data by \\u0026copy; \\u003ca target=\\"_blank\\" href=\\"http://openstreetmap.org\\"\\u003eOpenStreetMap\\u003c/a\\u003e, under \\u003ca target=\\"_blank\\" href=\\"http://www.openstreetmap.org/copyright\\"\\u003eODbL\\u003c/a\\u003e.", "detectRetina": false, "maxNativeZoom": 18, "maxZoom": 18, "minZoom": 0, "noWrap": false, "opacity": 1, "subdomains": "abc", "tms": false}\n", - " ).addTo(map_63d14900899a70183befbf30d354febd);\n", + " ).addTo(map_77bceaf15f8f92685ff07972bb265db5);\n", " \n", " \n", - " var poly_line_a70438f127ac39c09a7541b9864285e7 = L.polyline(\n", + " var poly_line_3096e7d1c76ca60404f4c47b9117a47c = L.polyline(\n", " [[39.77902603149414, -104.96916961669922], [39.77896499633789, -104.96910095214844], [39.77878189086914, -104.96887969970703], [39.778751373291016, -104.96884155273438], [39.778587341308594, -104.9686508178711], [39.778480529785156, -104.96852111816406], [39.77839660644531, -104.96841430664062], [39.77846908569336, -104.96832275390625], [39.77865982055664, -104.96807098388672], [39.778709411621094, -104.96791076660156], [39.77894592285156, -104.96761322021484], [39.77906799316406, -104.96745300292969], [39.77915954589844, -104.96732330322266], [39.7791862487793, -104.96728515625], [39.779232025146484, -104.96722412109375], [39.77926254272461, -104.96719360351562], [39.779293060302734, -104.9671630859375], [39.77932357788086, -104.96714782714844], [39.77938461303711, -104.96710968017578], [39.779441833496094, -104.96707916259766], [39.77951431274414, -104.96704864501953], [39.7796745300293, -104.96698760986328], [39.77983093261719, -104.9669418334961], [39.78031921386719, -104.96683502197266], [39.780364990234375, -104.9668197631836], [39.78050231933594, -104.966796875], [39.78049850463867, -104.96688079833984], [39.78049850463867, -104.96698760986328], [39.78049850463867, -104.96705627441406], [39.78049087524414, -104.96717834472656], [39.78046798706055, -104.96759033203125], [39.78045654296875, -104.96772003173828], [39.78042984008789, -104.96797180175781], [39.78041458129883, -104.96810913085938], [39.780372619628906, -104.96839904785156], [39.78034210205078, -104.96859741210938], [39.780296325683594, -104.96888732910156], [39.7802734375, -104.96903228759766], [39.78025436401367, -104.96917724609375], [39.780250549316406, -104.96920776367188], [39.780235290527344, -104.96934509277344], [39.78020095825195, -104.96965789794922], [39.78001403808594, -104.9715347290039], [39.77994155883789, -104.97315216064453], [39.77980422973633, -104.97496032714844], [39.7797737121582, -104.97691345214844], [39.7797737121582, -104.9772720336914], [39.7797737121582, -104.97753143310547], [39.7797737121582, -104.97764587402344], [39.779781341552734, -104.97801971435547], [39.77980041503906, -104.97874450683594], [39.77981948852539, -104.97929382324219], [39.77983093261719, -104.97946166992188], [39.77986526489258, -104.98001861572266], [39.77991485595703, -104.98080444335938], [39.779991149902344, -104.98214721679688], [39.78001403808594, -104.98335266113281], [39.7800178527832, -104.98450469970703], [39.78002166748047, -104.98558044433594], [39.780033111572266, -104.98602294921875], [39.7800407409668, -104.98613739013672], [39.780059814453125, -104.98633575439453], [39.780086517333984, -104.9865493774414], [39.780128479003906, -104.98677825927734], [39.780174255371094, -104.98695373535156], [39.78023910522461, -104.98715209960938], [39.780540466308594, -104.98787689208984], [39.78069305419922, -104.98828887939453], [39.78076934814453, -104.98860168457031], [39.780784606933594, -104.98871612548828], [39.780792236328125, -104.98896026611328], [39.78078079223633, -104.98912048339844], [39.780765533447266, -104.98928833007812], [39.78071975708008, -104.98950958251953], [39.78061294555664, -104.98975372314453], [39.78052520751953, -104.98993682861328], [39.78038787841797, -104.99011993408203], [39.78028869628906, -104.99024963378906], [39.780174255371094, -104.99034881591797], [39.77997589111328, -104.990478515625], [39.77995300292969, -104.99049377441406], [39.77978515625, -104.99056243896484], [39.77960205078125, -104.99060821533203], [39.7794189453125, -104.99060821533203], [39.77922439575195, -104.9905776977539], [39.778968811035156, -104.99048614501953], [39.77858352661133, -104.9903335571289], [39.778541564941406, -104.99031066894531], [39.77811050415039, -104.99011993408203], [39.777706146240234, -104.98995208740234], [39.7775764465332, -104.98990631103516], [39.77741241455078, -104.98985290527344], [39.77730941772461, -104.98983001708984], [39.7772102355957, -104.98980712890625], [39.777130126953125, -104.98979949951172], [39.77692794799805, -104.98979187011719], [39.77660369873047, -104.98979187011719], [39.77629089355469, -104.98980712890625], [39.77605056762695, -104.98982238769531], [39.77583694458008, -104.98986053466797], [39.77559280395508, -104.98991394042969], [39.7751579284668, -104.99003601074219], [39.774871826171875, -104.99012756347656], [39.77482986450195, -104.9901351928711], [39.774696350097656, -104.99018096923828], [39.774539947509766, -104.99022674560547], [39.774383544921875, -104.99026489257812], [39.77424621582031, -104.99028778076172], [39.77410888671875, -104.99029541015625], [39.773887634277344, -104.99030303955078], [39.77339553833008, -104.99031829833984], [39.773284912109375, -104.99031829833984], [39.77314376831055, -104.99032592773438], [39.773006439208984, -104.99034118652344], [39.77288055419922, -104.99036407470703], [39.77272033691406, -104.99040222167969], [39.77251434326172, -104.99046325683594], [39.77234649658203, -104.99051666259766], [39.77214431762695, -104.99059295654297], [39.77194595336914, -104.99068450927734], [39.77174377441406, -104.99079895019531], [39.77142333984375, -104.99098205566406], [39.77122116088867, -104.9911117553711], [39.77073287963867, -104.9914321899414], [39.7704963684082, -104.99164581298828], [39.770259857177734, -104.99187469482422], [39.77001190185547, -104.99214172363281], [39.769779205322266, -104.99241638183594], [39.76954650878906, -104.99272918701172], [39.76931381225586, -104.99305725097656], [39.76912307739258, -104.99334716796875], [39.76897048950195, -104.99359893798828], [39.768775939941406, -104.99393463134766], [39.768592834472656, -104.99432373046875], [39.76849365234375, -104.99454498291016], [39.76826477050781, -104.9951171875], [39.768192291259766, -104.99535369873047], [39.76802062988281, -104.99580383300781], [39.767581939697266, -104.99694061279297], [39.767154693603516, -104.99806213378906], [39.766963958740234, -104.9985580444336], [39.76682662963867, -104.9989013671875], [39.7667121887207, -104.99915313720703], [39.76656723022461, -104.99944305419922], [39.76641845703125, -104.99970245361328], [39.766273498535156, -104.99990844726562], [39.766109466552734, -105.00015258789062], [39.7659912109375, -105.00031280517578], [39.7658805847168, -105.0004653930664], [39.765750885009766, -105.00062561035156], [39.765567779541016, -105.00081634521484], [39.76536560058594, -105.00100708007812], [39.76518249511719, -105.00117492675781], [39.76498794555664, -105.00135803222656], [39.763919830322266, -105.002197265625], [39.76262283325195, -105.00331115722656], [39.76240539550781, -105.00350189208984], [39.762203216552734, -105.0036849975586], [39.76198959350586, -105.00389099121094], [39.76179885864258, -105.00407409667969], [39.7616081237793, -105.0042724609375], [39.761444091796875, -105.00445556640625], [39.761260986328125, -105.00466918945312], [39.76103973388672, -105.00492858886719], [39.76044845581055, -105.00575256347656], [39.759456634521484, -105.0072021484375], [39.75861740112305, -105.0083999633789], [39.758094787597656, -105.00910949707031], [39.75727844238281, -105.0101547241211], [39.75709915161133, -105.01038360595703], [39.756736755371094, -105.0108871459961], [39.75618362426758, -105.0116195678711], [39.75572967529297, -105.01225280761719], [39.75527572631836, -105.01285552978516], [39.75492858886719, -105.01334381103516], [39.7547492980957, -105.01359558105469], [39.75423812866211, -105.01427459716797], [39.75400924682617, -105.01457214355469], [39.753841400146484, -105.0147705078125], [39.75370788574219, -105.01492309570312], [39.75352096557617, -105.01513671875], [39.75336837768555, -105.01529693603516], [39.753173828125, -105.01548767089844], [39.75297164916992, -105.01567840576172], [39.752811431884766, -105.01581573486328], [39.75261688232422, -105.01597595214844], [39.75239944458008, -105.0161361694336], [39.75222396850586, -105.01627349853516], [39.752037048339844, -105.01639556884766], [39.751853942871094, -105.0165023803711], [39.75165939331055, -105.01661682128906], [39.75148391723633, -105.01670837402344], [39.75117874145508, -105.0168685913086], [39.75090026855469, -105.01699829101562], [39.75063705444336, -105.01710510253906], [39.74983215332031, -105.01744079589844], [39.74938201904297, -105.01763916015625], [39.74889373779297, -105.01783752441406], [39.748531341552734, -105.01799011230469], [39.74820327758789, -105.01811218261719], [39.74799728393555, -105.01818084716797], [39.74775695800781, -105.01825714111328], [39.74755096435547, -105.01830291748047], [39.74735641479492, -105.01834106445312], [39.747108459472656, -105.01836395263672], [39.74689865112305, -105.01837158203125], [39.74666976928711, -105.01836395263672], [39.74650955200195, -105.01835632324219], [39.74633026123047, -105.01832580566406], [39.746131896972656, -105.0182876586914], [39.74597930908203, -105.01824951171875], [39.74581527709961, -105.01819610595703], [39.745609283447266, -105.01811218261719], [39.74538040161133, -105.01800537109375], [39.74514389038086, -105.01786804199219], [39.744937896728516, -105.01773834228516], [39.744747161865234, -105.01759338378906], [39.74456787109375, -105.0174560546875], [39.74444580078125, -105.01734161376953], [39.743892669677734, -105.01679992675781], [39.74342727661133, -105.0163345336914], [39.743038177490234, -105.015869140625], [39.742897033691406, -105.01570892333984], [39.74272537231445, -105.0155258178711], [39.7425422668457, -105.01532745361328], [39.74235916137695, -105.01516723632812], [39.7421875, -105.01500701904297], [39.74201965332031, -105.0148696899414], [39.74187469482422, -105.01476287841797], [39.74171447753906, -105.0146484375], [39.74155807495117, -105.01456451416016], [39.741329193115234, -105.01445007324219], [39.74116134643555, -105.01438903808594], [39.74099349975586, -105.01432037353516], [39.74075698852539, -105.01425170898438], [39.7405891418457, -105.01421356201172], [39.740386962890625, -105.01416778564453], [39.740230560302734, -105.01415252685547], [39.74014663696289, -105.0141372680664], [39.73987579345703, -105.01412963867188], [39.73881530761719, -105.01406860351562], [39.73861312866211, -105.0140609741211], [39.73841094970703, -105.0140609741211], [39.73823928833008, -105.01406860351562], [39.73807907104492, -105.01408386230469], [39.7379035949707, -105.01409912109375], [39.73770523071289, -105.01412963867188], [39.737457275390625, -105.01416778564453], [39.737152099609375, -105.01428985595703], [39.73687744140625, -105.01436614990234], [39.7366828918457, -105.01441955566406], [39.73640823364258, -105.01451873779297], [39.73619079589844, -105.01460266113281], [39.73607635498047, -105.01464080810547], [39.735904693603516, -105.01471710205078], [39.73575973510742, -105.01478576660156], [39.73533630371094, -105.0149917602539], [39.73477554321289, -105.01528930664062], [39.734619140625, -105.01536560058594], [39.73448181152344, -105.01542663574219], [39.73434066772461, -105.01548767089844], [39.73418426513672, -105.01554870605469], [39.73397445678711, -105.015625], [39.733829498291016, -105.01566314697266], [39.73370361328125, -105.01570129394531], [39.73356246948242, -105.01573944091797], [39.733436584472656, -105.0157699584961], [39.73330307006836, -105.01579284667969], [39.73316955566406, -105.01582336425781], [39.73301315307617, -105.01583862304688], [39.73285675048828, -105.01585388183594], [39.73271942138672, -105.01586151123047], [39.732574462890625, -105.015869140625], [39.73243713378906, -105.01587677001953], [39.732303619384766, -105.01588439941406], [39.73218536376953, -105.01588439941406], [39.73207092285156, -105.01587677001953], [39.73195266723633, -105.015869140625], [39.73179244995117, -105.01585388183594], [39.73155212402344, -105.01580047607422], [39.73124313354492, -105.01573181152344], [39.731075286865234, -105.01568603515625], [39.7308464050293, -105.01561737060547], [39.730567932128906, -105.01553344726562], [39.730201721191406, -105.0154037475586], [39.729949951171875, -105.01531219482422], [39.72952651977539, -105.01518249511719], [39.729122161865234, -105.0150375366211], [39.728843688964844, -105.01493835449219], [39.72844314575195, -105.01477813720703], [39.72832107543945, -105.01473236083984], [39.728145599365234, -105.01466369628906], [39.727943420410156, -105.01458740234375], [39.727439880371094, -105.01441192626953], [39.72711181640625, -105.01427459716797], [39.72694778442383, -105.01421356201172], [39.726749420166016, -105.01412200927734], [39.726558685302734, -105.0140380859375], [39.726375579833984, -105.01394653320312], [39.72616195678711, -105.01382446289062], [39.72585678100586, -105.01362609863281], [39.725624084472656, -105.01348114013672], [39.72536087036133, -105.01329803466797], [39.72511672973633, -105.01311492919922], [39.72493362426758, -105.01297760009766], [39.7247200012207, -105.01280212402344], [39.72451400756836, -105.01262664794922], [39.72426986694336, -105.01239776611328], [39.72408676147461, -105.01221466064453], [39.723899841308594, -105.01201629638672], [39.72377395629883, -105.01188659667969], [39.72359848022461, -105.01168060302734], [39.72343826293945, -105.0114974975586], [39.72330093383789, -105.01133728027344], [39.723079681396484, -105.01106262207031], [39.7227897644043, -105.01066589355469], [39.72258377075195, -105.01038360595703], [39.722251892089844, -105.0099105834961], [39.7220573425293, -105.0096435546875], [39.721920013427734, -105.00946807861328], [39.721744537353516, -105.00924682617188], [39.721649169921875, -105.00914001464844], [39.7214241027832, -105.00888061523438], [39.72117233276367, -105.00861358642578], [39.720741271972656, -105.0081787109375], [39.720523834228516, -105.00794219970703], [39.7200813293457, -105.0074691772461], [39.71998977661133, -105.00737762451172], [39.71975326538086, -105.00711822509766], [39.71941375732422, -105.00675201416016], [39.719242095947266, -105.0065689086914], [39.719085693359375, -105.00640106201172], [39.718936920166016, -105.00623321533203], [39.718807220458984, -105.00608825683594], [39.718658447265625, -105.00591278076172], [39.71849060058594, -105.00570678710938], [39.7183723449707, -105.00556182861328], [39.71826171875, -105.00541687011719], [39.71812057495117, -105.00521850585938], [39.71792221069336, -105.00492858886719], [39.71759796142578, -105.00445556640625], [39.717166900634766, -105.0038070678711], [39.716880798339844, -105.00337982177734], [39.71671676635742, -105.00314331054688], [39.716552734375, -105.00292205810547], [39.7164306640625, -105.00276184082031], [39.71633529663086, -105.00264739990234], [39.71624755859375, -105.00254821777344], [39.71617889404297, -105.00247192382812], [39.71608352661133, -105.00237274169922], [39.715999603271484, -105.00228118896484], [39.71593475341797, -105.0022201538086], [39.715885162353516, -105.0021743774414], [39.715755462646484, -105.00206756591797], [39.715660095214844, -105.00198364257812], [39.715538024902344, -105.00189208984375], [39.71543884277344, -105.00181579589844], [39.71533203125, -105.00173950195312], [39.7152214050293, -105.00166320800781], [39.71499252319336, -105.00151824951172], [39.714324951171875, -105.0010757446289], [39.71399688720703, -105.00086212158203], [39.71346664428711, -105.00052642822266], [39.71318435668945, -105.00035095214844], [39.71266174316406, -105.00003051757812], [39.71223449707031, -104.9997787475586], [39.71159362792969, -104.99940490722656], [39.71141052246094, -104.99930572509766], [39.710933685302734, -104.99901580810547], [39.71067810058594, -104.99886322021484], [39.710357666015625, -104.99867248535156], [39.71013641357422, -104.99855041503906], [39.70988464355469, -104.99840545654297], [39.70966720581055, -104.99829864501953], [39.7094841003418, -104.99820709228516], [39.70940017700195, -104.99816131591797], [39.709136962890625, -104.99803924560547], [39.708702087402344, -104.99784088134766], [39.70827865600586, -104.99764251708984], [39.7078857421875, -104.99746704101562], [39.707275390625, -104.9971923828125], [39.70710372924805, -104.99711608886719], [39.7068977355957, -104.99701690673828], [39.706809997558594, -104.99697875976562], [39.706729888916016, -104.99693298339844], [39.70663070678711, -104.99688720703125], [39.706539154052734, -104.99683380126953], [39.70646667480469, -104.99678802490234], [39.70637130737305, -104.9967269897461], [39.70630645751953, -104.99668884277344], [39.70623779296875, -104.99664306640625], [39.70616912841797, -104.99658966064453], [39.706085205078125, -104.99652862548828], [39.70596694946289, -104.9964370727539], [39.705814361572266, -104.9963150024414], [39.7056999206543, -104.9962158203125], [39.70561218261719, -104.99613189697266], [39.70552062988281, -104.99604034423828], [39.705467224121094, -104.99598693847656], [39.70537567138672, -104.99588775634766], [39.705299377441406, -104.99580383300781], [39.705169677734375, -104.99565124511719], [39.70507049560547, -104.99552917480469], [39.704959869384766, -104.9953842163086], [39.704837799072266, -104.99520874023438], [39.70473098754883, -104.99505615234375], [39.704627990722656, -104.99488067626953], [39.704551696777344, -104.99474334716797], [39.704471588134766, -104.99459838867188], [39.70440673828125, -104.99447631835938], [39.7043342590332, -104.99433135986328], [39.704261779785156, -104.99417877197266], [39.70420455932617, -104.99404907226562], [39.70415115356445, -104.99392700195312], [39.704097747802734, -104.99378204345703], [39.70404815673828, -104.99363708496094], [39.70399475097656, -104.99347686767578], [39.70393753051758, -104.9933090209961], [39.70388412475586, -104.99313354492188], [39.703792572021484, -104.99284362792969], [39.70363235473633, -104.9923324584961], [39.703487396240234, -104.99188232421875], [39.703304290771484, -104.99127197265625], [39.703189849853516, -104.99092102050781], [39.70305633544922, -104.990478515625], [39.702980041503906, -104.99024200439453], [39.70287322998047, -104.989990234375], [39.70277404785156, -104.98977661132812], [39.70265197753906, -104.98954010009766], [39.70252990722656, -104.98933410644531], [39.70241928100586, -104.98914337158203], [39.70228576660156, -104.98894500732422], [39.702110290527344, -104.98870849609375], [39.70179748535156, -104.9883041381836], [39.70148849487305, -104.98790740966797], [39.700843811035156, -104.987060546875], [39.700706481933594, -104.98689270019531], [39.70027542114258, -104.98641967773438], [39.700077056884766, -104.9861831665039], [39.69994354248047, -104.98603057861328], [39.699771881103516, -104.98583984375], [39.69960021972656, -104.98567962646484], [39.699462890625, -104.98554229736328], [39.699344635009766, -104.98542022705078], [39.69913101196289, -104.98521423339844], [39.698974609375, -104.98504638671875], [39.69866180419922, -104.9847412109375], [39.6981086730957, -104.98410034179688], [39.69713592529297, -104.98299407958984], [39.696075439453125, -104.98184967041016], [39.69560623168945, -104.9813003540039], [39.69524383544922, -104.98087310791016], [39.69488525390625, -104.98043823242188], [39.694786071777344, -104.98033905029297], [39.6947021484375, -104.98025512695312], [39.69462585449219, -104.98019409179688], [39.6945915222168, -104.98016357421875], [39.69448471069336, -104.98009490966797], [39.69437789916992, -104.98002624511719], [39.6942024230957, -104.97992706298828], [39.69398498535156, -104.97977447509766], [39.69377899169922, -104.97953033447266], [39.69367599487305, -104.97943115234375], [39.69364929199219, -104.97940063476562], [39.69361114501953, -104.97936248779297], [39.69355010986328, -104.97931671142578], [39.6934814453125, -104.9792709350586], [39.69330978393555, -104.97918701171875], [39.69325637817383, -104.97914123535156], [39.693199157714844, -104.97909545898438], [39.693145751953125, -104.97903442382812], [39.693092346191406, -104.97897338867188], [39.69303512573242, -104.97886657714844], [39.69300079345703, -104.97877502441406], [39.69298553466797, -104.97871398925781], [39.69298553466797, -104.97850036621094], [39.69298553466797, -104.97827911376953], [39.69298553466797, -104.97792053222656], [39.6929931640625, -104.9774398803711], [39.6929931640625, -104.9773941040039], [39.6929931640625, -104.9772720336914], [39.692989349365234, -104.97704315185547], [39.692989349365234, -104.97694396972656], [39.692989349365234, -104.97685241699219], [39.69298553466797, -104.97636413574219], [39.6929817199707, -104.97577667236328]],\n", " {"bubblingMouseEvents": true, "color": "#eff821", "dashArray": null, "dashOffset": null, "fill": false, "fillColor": "#eff821", "fillOpacity": 0.2, "fillRule": "evenodd", "lineCap": "round", "lineJoin": "round", "noClip": false, "opacity": 0.8, "smoothFactor": 1.0, "stroke": true, "weight": 10}\n", - " ).addTo(map_63d14900899a70183befbf30d354febd);\n", + " ).addTo(map_77bceaf15f8f92685ff07972bb265db5);\n", " \n", " \n", - " poly_line_a70438f127ac39c09a7541b9864285e7.bindTooltip(\n", + " poly_line_3096e7d1c76ca60404f4c47b9117a47c.bindTooltip(\n", " `<div>\n", - " 0.3200989780246874\n", + " 0.32011129944536876\n", " </div>`,\n", " {"sticky": true}\n", " );\n", " \n", " \n", - " var marker_32cb0a5c3e87861426d1338199f84c15 = L.marker(\n", + " var marker_b408369e32baf49a93b267bb1c26ab1e = L.marker(\n", " [39.77902603149414, -104.96916961669922],\n", " {}\n", - " ).addTo(map_63d14900899a70183befbf30d354febd);\n", + " ).addTo(map_77bceaf15f8f92685ff07972bb265db5);\n", " \n", " \n", - " var icon_125348ab18ac3db4155a4629ca19372d = L.AwesomeMarkers.icon(\n", + " var icon_6bdb7e8c90867a43a909c4b03b0f45ff = L.AwesomeMarkers.icon(\n", " {"extraClasses": "fa-rotate-0", "icon": "circle", "iconColor": "white", "markerColor": "green", "prefix": "fa"}\n", " );\n", - " marker_32cb0a5c3e87861426d1338199f84c15.setIcon(icon_125348ab18ac3db4155a4629ca19372d);\n", + " marker_b408369e32baf49a93b267bb1c26ab1e.setIcon(icon_6bdb7e8c90867a43a909c4b03b0f45ff);\n", " \n", " \n", - " marker_32cb0a5c3e87861426d1338199f84c15.bindTooltip(\n", + " marker_b408369e32baf49a93b267bb1c26ab1e.bindTooltip(\n", " `<div>\n", " Origin\n", " </div>`,\n", @@ -1106,19 +1114,19 @@ " );\n", " \n", " \n", - " var marker_1dd53c8b4ad0e849af3d27a233f7eb8d = L.marker(\n", + " var marker_5e58e74755fc0f7be486049c35336a3f = L.marker(\n", " [39.6929817199707, -104.97577667236328],\n", " {}\n", - " ).addTo(map_63d14900899a70183befbf30d354febd);\n", + " ).addTo(map_77bceaf15f8f92685ff07972bb265db5);\n", " \n", " \n", - " var icon_1a6efdb2b1623d9092fbc338dec339f1 = L.AwesomeMarkers.icon(\n", + " var icon_0a710de33c2e19f55db9d8a13e270ae9 = L.AwesomeMarkers.icon(\n", " {"extraClasses": "fa-rotate-0", "icon": "circle", "iconColor": "white", "markerColor": "red", "prefix": "fa"}\n", " );\n", - " marker_1dd53c8b4ad0e849af3d27a233f7eb8d.setIcon(icon_1a6efdb2b1623d9092fbc338dec339f1);\n", + " marker_5e58e74755fc0f7be486049c35336a3f.setIcon(icon_0a710de33c2e19f55db9d8a13e270ae9);\n", " \n", " \n", - " marker_1dd53c8b4ad0e849af3d27a233f7eb8d.bindTooltip(\n", + " marker_5e58e74755fc0f7be486049c35336a3f.bindTooltip(\n", " `<div>\n", " Destination\n", " </div>`,\n", @@ -1126,33 +1134,33 @@ " );\n", " \n", " \n", - " var poly_line_dc18b3d37d1d8229eeaee4b38b80d4ef = L.polyline(\n", - " [[39.77902603149414, -104.96916961669922], [39.77896499633789, -104.96910095214844], [39.77878189086914, -104.96887969970703], [39.778751373291016, -104.96884155273438], [39.778587341308594, -104.9686508178711], [39.778480529785156, -104.96852111816406], [39.77839660644531, -104.96841430664062], [39.778324127197266, -104.96851348876953], [39.778228759765625, -104.96863555908203], [39.778076171875, -104.96883392333984], [39.777626037597656, -104.96941375732422], [39.7774772644043, -104.96961212158203], [39.77703857421875, -104.97017669677734], [39.776737213134766, -104.97056579589844], [39.77666473388672, -104.97066497802734], [39.7766227722168, -104.97071838378906], [39.776371002197266, -104.97103881835938], [39.77608108520508, -104.9714126586914], [39.775978088378906, -104.97154998779297], [39.774810791015625, -104.97306060791016], [39.77475357055664, -104.97313690185547], [39.77467346191406, -104.97323608398438], [39.77459716796875, -104.97333526611328], [39.773738861083984, -104.97444915771484], [39.77363586425781, -104.97457885742188], [39.773353576660156, -104.97494506835938], [39.77303695678711, -104.97535705566406], [39.77297592163086, -104.9754409790039], [39.77287673950195, -104.97531127929688], [39.77263259887695, -104.9749984741211], [39.77257537841797, -104.97492218017578], [39.77229690551758, -104.97456359863281], [39.77223587036133, -104.9744873046875], [39.77199935913086, -104.97418212890625], [39.77191162109375, -104.97406768798828], [39.77155685424805, -104.9736099243164], [39.771270751953125, -104.97322845458984], [39.7706184387207, -104.97236633300781], [39.77056884765625, -104.97230529785156], [39.77048110961914, -104.97219848632812], [39.77042770385742, -104.9721450805664], [39.7703742980957, -104.97209167480469], [39.77033233642578, -104.97205352783203], [39.770233154296875, -104.97198486328125], [39.770118713378906, -104.9721450805664], [39.76996994018555, -104.97232818603516], [39.76978302001953, -104.97257232666016], [39.76957321166992, -104.97283172607422], [39.76926803588867, -104.97325134277344], [39.76918411254883, -104.97335815429688], [39.76904296875, -104.97335815429688], [39.76840591430664, -104.97335052490234], [39.76824188232422, -104.97335052490234], [39.7681770324707, -104.97335052490234], [39.7680549621582, -104.97335052490234], [39.7677001953125, -104.97335052490234], [39.767547607421875, -104.97335052490234], [39.7670783996582, -104.97335052490234], [39.76704788208008, -104.97335052490234], [39.766971588134766, -104.97335052490234], [39.76690673828125, -104.97335052490234], [39.76670455932617, -104.97335052490234], [39.76658248901367, -104.97335052490234], [39.766502380371094, -104.97335052490234], [39.76637649536133, -104.97335052490234], [39.766136169433594, -104.97335052490234], [39.76605987548828, -104.97335052490234], [39.765865325927734, -104.97335052490234], [39.76577377319336, -104.97335052490234], [39.76567459106445, -104.97335052490234], [39.76523208618164, -104.97335052490234], [39.76495361328125, -104.97335052490234], [39.764835357666016, -104.97335052490234], [39.7646598815918, -104.97335052490234], [39.764591217041016, -104.97335052490234], [39.7645149230957, -104.97335052490234], [39.764427185058594, -104.97335052490234], [39.76421356201172, -104.97335052490234], [39.764034271240234, -104.97335052490234], [39.76380920410156, -104.97335052490234], [39.763580322265625, -104.97335052490234], [39.76325988769531, -104.97335052490234], [39.76316833496094, -104.97339630126953], [39.763118743896484, -104.97339630126953], [39.76306915283203, -104.97335052490234], [39.76280212402344, -104.97335815429688], [39.762454986572266, -104.97335815429688], [39.76237869262695, -104.97335815429688], [39.76214599609375, -104.97335052490234], [39.762054443359375, -104.97335052490234], [39.76194381713867, -104.97335052490234], [39.76179122924805, -104.97335052490234], [39.76169967651367, -104.97335052490234], [39.76160430908203, -104.97335052490234], [39.76118087768555, -104.97335052490234], [39.76103973388672, -104.97335052490234], [39.76084518432617, -104.97335052490234], [39.76076126098633, -104.97335052490234], [39.76066589355469, -104.97335052490234], [39.760440826416016, -104.97335052490234], [39.76033401489258, -104.97335052490234], [39.759613037109375, -104.97335052490234], [39.7595100402832, -104.97335052490234], [39.75939178466797, -104.97339630126953], [39.759151458740234, -104.97339630126953], [39.75912094116211, -104.97335052490234], [39.759010314941406, -104.97335052490234], [39.758827209472656, -104.97335052490234], [39.75831604003906, -104.97335052490234], [39.75823211669922, -104.97335052490234], [39.758174896240234, -104.97335052490234], [39.75764083862305, -104.97335815429688], [39.7572021484375, -104.97335815429688], [39.75708770751953, -104.97335815429688], [39.75699234008789, -104.97335815429688], [39.75688171386719, -104.97335815429688], [39.75673294067383, -104.97335815429688], [39.75627899169922, -104.97335815429688], [39.75574493408203, -104.97335815429688], [39.75540542602539, -104.97335815429688], [39.75531005859375, -104.97335815429688], [39.75495147705078, -104.97335815429688], [39.75469970703125, -104.97335815429688], [39.754547119140625, -104.97335815429688], [39.75442886352539, -104.97335815429688], [39.753578186035156, -104.97335815429688], [39.75333786010742, -104.97335815429688], [39.75325012207031, -104.97335815429688], [39.75314712524414, -104.97335815429688], [39.75208282470703, -104.97335815429688], [39.751991271972656, -104.97335815429688], [39.751895904541016, -104.97335815429688], [39.75139236450195, -104.9733657836914], [39.75082778930664, -104.9733657836914], [39.750732421875, -104.9733657836914], [39.75063705444336, -104.9733657836914], [39.750423431396484, -104.9733657836914], [39.750282287597656, -104.9733657836914], [39.75018310546875, -104.9733657836914], [39.749576568603516, -104.9733657836914], [39.74947738647461, -104.9733657836914], [39.74938201904297, -104.9733657836914], [39.74884033203125, -104.9733657836914], [39.74821853637695, -104.97337341308594], [39.74812698364258, -104.97337341308594], [39.748016357421875, -104.97337341308594], [39.747798919677734, -104.97337341308594], [39.74757766723633, -104.97337341308594], [39.74747085571289, -104.97337341308594], [39.74736785888672, -104.97337341308594], [39.74721908569336, -104.97337341308594], [39.747066497802734, -104.97337341308594], [39.746952056884766, -104.97337341308594], [39.746864318847656, -104.97337341308594], [39.74663162231445, -104.97337341308594], [39.746551513671875, -104.97338104248047], [39.746482849121094, -104.97338104248047], [39.7460823059082, -104.97338104248047], [39.74594497680664, -104.97338104248047], [39.74583435058594, -104.97338104248047], [39.74509048461914, -104.97340393066406], [39.745052337646484, -104.97340393066406], [39.745018005371094, -104.97344207763672], [39.744972229003906, -104.97344970703125], [39.744937896728516, -104.97344970703125], [39.744834899902344, -104.97339630126953], [39.744834899902344, -104.97355651855469], [39.744834899902344, -104.97415924072266], [39.74483871459961, -104.97454833984375], [39.74483871459961, -104.97476196289062], [39.74464416503906, -104.97476196289062], [39.744606018066406, -104.97476196289062], [39.74433517456055, -104.97476959228516], [39.74381637573242, -104.97476959228516], [39.74373245239258, -104.97476959228516], [39.743377685546875, -104.97477722167969], [39.74324417114258, -104.97477722167969], [39.743133544921875, -104.97477722167969], [39.7426872253418, -104.97477722167969], [39.74175262451172, -104.97477722167969], [39.74166488647461, -104.97477722167969], [39.74156951904297, -104.97477722167969], [39.7412109375, -104.97478485107422], [39.740718841552734, -104.97479248046875], [39.74055862426758, -104.97479248046875], [39.74032211303711, -104.97479248046875], [39.74012756347656, -104.97480010986328], [39.740020751953125, -104.97480010986328], [39.740020751953125, -104.9746322631836], [39.740020751953125, -104.97451782226562], [39.74002456665039, -104.97415924072266], [39.73991775512695, -104.97415924072266], [39.73970031738281, -104.97415924072266], [39.73955535888672, -104.97415161132812], [39.739315032958984, -104.97415161132812], [39.73896789550781, -104.97415161132812], [39.738609313964844, -104.97415161132812], [39.73850631713867, -104.9741439819336], [39.73841857910156, -104.9741439819336], [39.73833465576172, -104.9741439819336], [39.736942291259766, -104.9741439819336], [39.736881256103516, -104.9741439819336], [39.73681640625, -104.9741439819336], [39.73567581176758, -104.97412109375], [39.735252380371094, -104.97411346435547], [39.735164642333984, -104.97411346435547], [39.735069274902344, -104.97411346435547], [39.73379898071289, -104.97410583496094], [39.73370361328125, -104.97410583496094], [39.733612060546875, -104.97410583496094], [39.73240280151367, -104.97410583496094], [39.732357025146484, -104.97410583496094], [39.73214340209961, -104.97410583496094], [39.7320556640625, -104.97410583496094], [39.73196029663086, -104.97410583496094], [39.731422424316406, -104.97410583496094], [39.73128890991211, -104.97410583496094], [39.730960845947266, -104.97410583496094], [39.73088455200195, -104.97410583496094], [39.73067855834961, -104.97410583496094], [39.73054504394531, -104.97410583496094], [39.7304573059082, -104.97410583496094], [39.730369567871094, -104.97410583496094], [39.73027420043945, -104.97410583496094], [39.73003387451172, -104.97410583496094], [39.729976654052734, -104.97410583496094], [39.729896545410156, -104.97410583496094], [39.72964859008789, -104.97410583496094], [39.7294807434082, -104.97410583496094], [39.729434967041016, -104.97410583496094], [39.72917175292969, -104.97410583496094], [39.72907257080078, -104.97410583496094], [39.72898864746094, -104.97410583496094], [39.727378845214844, -104.9740982055664], [39.72727966308594, -104.9740982055664], [39.727195739746094, -104.9740982055664], [39.7259635925293, -104.9740982055664], [39.72566604614258, -104.97409057617188], [39.7255859375, -104.97409057617188], [39.72549819946289, -104.97409057617188], [39.725337982177734, -104.97409057617188], [39.7252082824707, -104.97409057617188], [39.724693298339844, -104.97409057617188], [39.72453689575195, -104.97409057617188], [39.72406768798828, -104.97408294677734], [39.72374725341797, -104.97408294677734], [39.72247314453125, -104.97408294677734], [39.72218322753906, -104.97408294677734], [39.72205352783203, -104.97408294677734], [39.721946716308594, -104.97408294677734], [39.72013473510742, -104.97406768798828], [39.719642639160156, -104.97406005859375], [39.71931457519531, -104.97406005859375], [39.71918869018555, -104.97406005859375], [39.718971252441406, -104.97405242919922], [39.71888732910156, -104.97404479980469], [39.71880340576172, -104.97403717041016], [39.71875, -104.97402954101562], [39.71870422363281, -104.9740219116211], [39.718650817871094, -104.97401428222656], [39.71857452392578, -104.97399139404297], [39.718505859375, -104.97396850585938], [39.71845245361328, -104.97393798828125], [39.71842956542969, -104.97392272949219], [39.71839141845703, -104.97390747070312], [39.71834945678711, -104.97386932373047], [39.718318939208984, -104.97382354736328], [39.71824264526367, -104.97368621826172], [39.71818923950195, -104.97358703613281], [39.71816635131836, -104.97354888916016], [39.71815490722656, -104.9735336303711], [39.71811294555664, -104.97348022460938], [39.71806335449219, -104.97343444824219], [39.71802520751953, -104.97341918945312], [39.717979431152344, -104.97340393066406], [39.717952728271484, -104.97339630126953], [39.71792221069336, -104.973388671875], [39.717899322509766, -104.973388671875], [39.71767807006836, -104.973388671875], [39.716922760009766, -104.973388671875], [39.71653366088867, -104.97339630126953], [39.71645736694336, -104.97339630126953], [39.71589660644531, -104.97339630126953], [39.71546936035156, -104.97341918945312], [39.71541976928711, -104.97341918945312], [39.71538543701172, -104.97342681884766], [39.71536636352539, -104.97342681884766], [39.71527862548828, -104.97342681884766], [39.7148551940918, -104.97344970703125], [39.71479415893555, -104.97344970703125], [39.71472930908203, -104.97345733642578], [39.714656829833984, -104.97344970703125], [39.71451950073242, -104.97344207763672], [39.71427917480469, -104.9734115600586], [39.712989807128906, -104.97340393066406], [39.71291732788086, -104.97340393066406], [39.71285629272461, -104.97340393066406], [39.71247482299805, -104.97340393066406], [39.711551666259766, -104.97340393066406], [39.71137237548828, -104.97340393066406], [39.711204528808594, -104.97340393066406], [39.71110534667969, -104.97340393066406], [39.711021423339844, -104.97340393066406], [39.7109489440918, -104.97340393066406], [39.710845947265625, -104.97340393066406], [39.709381103515625, -104.973388671875], [39.70929718017578, -104.973388671875], [39.707820892333984, -104.973388671875], [39.707557678222656, -104.973388671875], [39.70749282836914, -104.973388671875], [39.707420349121094, -104.973388671875], [39.70706558227539, -104.97338104248047], [39.70566940307617, -104.97337341308594], [39.70392990112305, -104.973388671875], [39.70386505126953, -104.973388671875], [39.70379638671875, -104.973388671875], [39.70212173461914, -104.97339630126953], [39.70206069946289, -104.97339630126953], [39.70030212402344, -104.97339630126953], [39.70021057128906, -104.97340393066406], [39.700138092041016, -104.97339630126953], [39.698429107666016, -104.97340393066406], [39.69660949707031, -104.9734115600586], [39.69625473022461, -104.97341918945312], [39.694740295410156, -104.97341918945312], [39.694149017333984, -104.97343444824219], [39.69339370727539, -104.97343444824219], [39.69325256347656, -104.97343444824219], [39.69306564331055, -104.97342681884766], [39.6929817199707, -104.97342681884766], [39.6929817199707, -104.97357177734375], [39.6929817199707, -104.97403717041016], [39.6929817199707, -104.974609375], [39.69298553466797, -104.97517395019531], [39.6929817199707, -104.97577667236328]],\n", - " {"bubblingMouseEvents": true, "color": "#7e03a7", "dashArray": null, "dashOffset": null, "fill": false, "fillColor": "#7e03a7", "fillOpacity": 0.2, "fillRule": "evenodd", "lineCap": "round", "lineJoin": "round", "noClip": false, "opacity": 0.8, "smoothFactor": 1.0, "stroke": true, "weight": 10}\n", - " ).addTo(map_63d14900899a70183befbf30d354febd);\n", + " var poly_line_467195630d68d76e5593e7251fc8958f = L.polyline(\n", + " [[39.77902603149414, -104.96916961669922], [39.77896499633789, -104.96910095214844], [39.77878189086914, -104.96887969970703], [39.778751373291016, -104.96884155273438], [39.778587341308594, -104.9686508178711], [39.778480529785156, -104.96852111816406], [39.77839660644531, -104.96841430664062], [39.778324127197266, -104.96851348876953], [39.778228759765625, -104.96863555908203], [39.778076171875, -104.96883392333984], [39.777626037597656, -104.96941375732422], [39.7774772644043, -104.96961212158203], [39.77703857421875, -104.97017669677734], [39.776737213134766, -104.97056579589844], [39.77666473388672, -104.97066497802734], [39.7766227722168, -104.97071838378906], [39.776371002197266, -104.97103881835938], [39.77608108520508, -104.9714126586914], [39.775978088378906, -104.97154998779297], [39.774810791015625, -104.97306060791016], [39.77475357055664, -104.97313690185547], [39.77467346191406, -104.97323608398438], [39.77459716796875, -104.97333526611328], [39.773738861083984, -104.97444915771484], [39.77363586425781, -104.97457885742188], [39.773353576660156, -104.97494506835938], [39.77303695678711, -104.97535705566406], [39.77297592163086, -104.9754409790039], [39.77287673950195, -104.97531127929688], [39.77263259887695, -104.9749984741211], [39.77257537841797, -104.97492218017578], [39.77229690551758, -104.97456359863281], [39.77223587036133, -104.9744873046875], [39.77199935913086, -104.97418212890625], [39.77191162109375, -104.97406768798828], [39.77157974243164, -104.9736328125], [39.77155303955078, -104.9736328125], [39.77154541015625, -104.97361755371094], [39.7712516784668, -104.97325897216797], [39.77125549316406, -104.97321319580078], [39.77090072631836, -104.97274017333984], [39.7706184387207, -104.97236633300781], [39.77056884765625, -104.97230529785156], [39.77048110961914, -104.97219848632812], [39.77042770385742, -104.9721450805664], [39.7703742980957, -104.97209167480469], [39.77033233642578, -104.97205352783203], [39.770233154296875, -104.97198486328125], [39.770118713378906, -104.9721450805664], [39.76996994018555, -104.97232818603516], [39.76978302001953, -104.97257232666016], [39.76957321166992, -104.97283172607422], [39.76926803588867, -104.97325134277344], [39.76918411254883, -104.97335815429688], [39.76904296875, -104.97335815429688], [39.76840591430664, -104.97335052490234], [39.76824188232422, -104.97335052490234], [39.7681770324707, -104.97335052490234], [39.7680549621582, -104.97335052490234], [39.7677001953125, -104.97335052490234], [39.767547607421875, -104.97335052490234], [39.7670783996582, -104.97335052490234], [39.76704788208008, -104.97335052490234], [39.766971588134766, -104.97335052490234], [39.76690673828125, -104.97335052490234], [39.76670455932617, -104.97335052490234], [39.76658248901367, -104.97335052490234], [39.766502380371094, -104.97335052490234], [39.76637649536133, -104.97335052490234], [39.766136169433594, -104.97335052490234], [39.76605987548828, -104.97335052490234], [39.765865325927734, -104.97335052490234], [39.76577377319336, -104.97335052490234], [39.76567459106445, -104.97335052490234], [39.76523208618164, -104.97335052490234], [39.76495361328125, -104.97335052490234], [39.764835357666016, -104.97335052490234], [39.7646598815918, -104.97335052490234], [39.764591217041016, -104.97335052490234], [39.7645149230957, -104.97335052490234], [39.764427185058594, -104.97335052490234], [39.76421356201172, -104.97335052490234], [39.764034271240234, -104.97335052490234], [39.76380920410156, -104.97335052490234], [39.763580322265625, -104.97335052490234], [39.76325988769531, -104.97335052490234], [39.76316833496094, -104.97339630126953], [39.763118743896484, -104.97339630126953], [39.76306915283203, -104.97335052490234], [39.762874603271484, -104.97335815429688], [39.762454986572266, -104.97335815429688], [39.76237869262695, -104.97335815429688], [39.76214599609375, -104.97335052490234], [39.762054443359375, -104.97335052490234], [39.76194381713867, -104.97335052490234], [39.76179122924805, -104.97335052490234], [39.76169967651367, -104.97335052490234], [39.76160430908203, -104.97335052490234], [39.76118087768555, -104.97335052490234], [39.76103973388672, -104.97335052490234], [39.76084518432617, -104.97335052490234], [39.76076126098633, -104.97335052490234], [39.76066589355469, -104.97335052490234], [39.760440826416016, -104.97335052490234], [39.76033401489258, -104.97335052490234], [39.759613037109375, -104.97335052490234], [39.7595100402832, -104.97335052490234], [39.75939178466797, -104.97339630126953], [39.759151458740234, -104.97339630126953], [39.75912094116211, -104.97335052490234], [39.759010314941406, -104.97335052490234], [39.758827209472656, -104.97335052490234], [39.75831604003906, -104.97335052490234], [39.75823211669922, -104.97335052490234], [39.758174896240234, -104.97335052490234], [39.75764083862305, -104.97335815429688], [39.7572021484375, -104.97335815429688], [39.75708770751953, -104.97335815429688], [39.75699234008789, -104.97335815429688], [39.75688171386719, -104.97335815429688], [39.75673294067383, -104.97335815429688], [39.75627899169922, -104.97335815429688], [39.75574493408203, -104.97335815429688], [39.75540542602539, -104.97335815429688], [39.75531005859375, -104.97335815429688], [39.75495147705078, -104.97335815429688], [39.75469970703125, -104.97335815429688], [39.754547119140625, -104.97335815429688], [39.75442886352539, -104.97335815429688], [39.753578186035156, -104.97335815429688], [39.75333786010742, -104.97335815429688], [39.75325012207031, -104.97335815429688], [39.75314712524414, -104.97335815429688], [39.75208282470703, -104.97335815429688], [39.751991271972656, -104.97335815429688], [39.751895904541016, -104.97335815429688], [39.75139236450195, -104.9733657836914], [39.75082778930664, -104.9733657836914], [39.750732421875, -104.9733657836914], [39.75063705444336, -104.9733657836914], [39.750423431396484, -104.9733657836914], [39.750282287597656, -104.9733657836914], [39.75018310546875, -104.9733657836914], [39.749576568603516, -104.9733657836914], [39.74947738647461, -104.9733657836914], [39.74938201904297, -104.9733657836914], [39.74884033203125, -104.9733657836914], [39.74821853637695, -104.97337341308594], [39.74812698364258, -104.97337341308594], [39.748016357421875, -104.97337341308594], [39.747798919677734, -104.97337341308594], [39.74757766723633, -104.97337341308594], [39.74747085571289, -104.97337341308594], [39.74736785888672, -104.97337341308594], [39.74721908569336, -104.97337341308594], [39.747066497802734, -104.97337341308594], [39.746952056884766, -104.97337341308594], [39.746864318847656, -104.97337341308594], [39.74663162231445, -104.97337341308594], [39.746551513671875, -104.97338104248047], [39.746482849121094, -104.97338104248047], [39.7460823059082, -104.97338104248047], [39.74594497680664, -104.97338104248047], [39.74583435058594, -104.97338104248047], [39.74509048461914, -104.97340393066406], [39.745052337646484, -104.97340393066406], [39.745018005371094, -104.97344207763672], [39.744972229003906, -104.97344970703125], [39.744937896728516, -104.97344970703125], [39.744834899902344, -104.97339630126953], [39.744834899902344, -104.97355651855469], [39.744834899902344, -104.97415924072266], [39.74483871459961, -104.97454833984375], [39.74483871459961, -104.97476196289062], [39.74464416503906, -104.97476196289062], [39.744606018066406, -104.97476196289062], [39.74433517456055, -104.97476959228516], [39.74381637573242, -104.97476959228516], [39.74373245239258, -104.97476959228516], [39.743377685546875, -104.97477722167969], [39.74324417114258, -104.97477722167969], [39.743133544921875, -104.97477722167969], [39.7426872253418, -104.97477722167969], [39.74175262451172, -104.97477722167969], [39.74166488647461, -104.97477722167969], [39.74156951904297, -104.97477722167969], [39.7412109375, -104.97478485107422], [39.740718841552734, -104.97479248046875], [39.74055862426758, -104.97479248046875], [39.74032211303711, -104.97479248046875], [39.74012756347656, -104.97480010986328], [39.740020751953125, -104.97480010986328], [39.740020751953125, -104.9746322631836], [39.740020751953125, -104.97451782226562], [39.74002456665039, -104.97415924072266], [39.73991775512695, -104.97415924072266], [39.73970031738281, -104.97415924072266], [39.73955535888672, -104.97415161132812], [39.739315032958984, -104.97415161132812], [39.73896789550781, -104.97415161132812], [39.738609313964844, -104.97415161132812], [39.73850631713867, -104.9741439819336], [39.73841857910156, -104.9741439819336], [39.73833465576172, -104.9741439819336], [39.736942291259766, -104.9741439819336], [39.736881256103516, -104.9741439819336], [39.73681640625, -104.9741439819336], [39.73567581176758, -104.97412109375], [39.735252380371094, -104.97411346435547], [39.735164642333984, -104.97411346435547], [39.735069274902344, -104.97411346435547], [39.73379898071289, -104.97410583496094], [39.73370361328125, -104.97410583496094], [39.733612060546875, -104.97410583496094], [39.73240280151367, -104.97410583496094], [39.732357025146484, -104.97410583496094], [39.73214340209961, -104.97410583496094], [39.7320556640625, -104.97410583496094], [39.73196029663086, -104.97410583496094], [39.731422424316406, -104.97410583496094], [39.73128890991211, -104.97410583496094], [39.730960845947266, -104.97410583496094], [39.73088455200195, -104.97410583496094], [39.73067855834961, -104.97410583496094], [39.73054504394531, -104.97410583496094], [39.7304573059082, -104.97410583496094], [39.730369567871094, -104.97410583496094], [39.73027420043945, -104.97410583496094], [39.73003387451172, -104.97410583496094], [39.729976654052734, -104.97410583496094], [39.729896545410156, -104.97410583496094], [39.72964859008789, -104.97410583496094], [39.7294807434082, -104.97410583496094], [39.729434967041016, -104.97410583496094], [39.72917175292969, -104.97410583496094], [39.72907257080078, -104.97410583496094], [39.72898864746094, -104.97410583496094], [39.727378845214844, -104.9740982055664], [39.72727966308594, -104.9740982055664], [39.727195739746094, -104.9740982055664], [39.7259635925293, -104.9740982055664], [39.72566604614258, -104.97409057617188], [39.7255859375, -104.97409057617188], [39.72549819946289, -104.97409057617188], [39.725337982177734, -104.97409057617188], [39.7252082824707, -104.97409057617188], [39.724693298339844, -104.97409057617188], [39.72453689575195, -104.97409057617188], [39.72406768798828, -104.97408294677734], [39.72374725341797, -104.97408294677734], [39.72247314453125, -104.97408294677734], [39.72218322753906, -104.97408294677734], [39.72205352783203, -104.97408294677734], [39.721946716308594, -104.97408294677734], [39.72013473510742, -104.97406768798828], [39.719642639160156, -104.97406005859375], [39.71931457519531, -104.97406005859375], [39.71918869018555, -104.97406005859375], [39.718971252441406, -104.97405242919922], [39.71888732910156, -104.97404479980469], [39.71880340576172, -104.97403717041016], [39.71875, -104.97402954101562], [39.71870422363281, -104.9740219116211], [39.718650817871094, -104.97401428222656], [39.71857452392578, -104.97399139404297], [39.718505859375, -104.97396850585938], [39.71845245361328, -104.97393798828125], [39.71842956542969, -104.97392272949219], [39.71839141845703, -104.97390747070312], [39.71834945678711, -104.97386932373047], [39.718318939208984, -104.97382354736328], [39.71824264526367, -104.97368621826172], [39.71818923950195, -104.97358703613281], [39.71816635131836, -104.97354888916016], [39.71815490722656, -104.9735336303711], [39.71811294555664, -104.97348022460938], [39.71806335449219, -104.97343444824219], [39.71802520751953, -104.97341918945312], [39.717979431152344, -104.97340393066406], [39.717952728271484, -104.97339630126953], [39.71792221069336, -104.973388671875], [39.717899322509766, -104.973388671875], [39.71767807006836, -104.973388671875], [39.716922760009766, -104.973388671875], [39.71653366088867, -104.97339630126953], [39.71645736694336, -104.97339630126953], [39.71589660644531, -104.97339630126953], [39.71546936035156, -104.97341918945312], [39.71541976928711, -104.97341918945312], [39.71538543701172, -104.97342681884766], [39.71536636352539, -104.97342681884766], [39.71527862548828, -104.97342681884766], [39.7148551940918, -104.97344970703125], [39.71479415893555, -104.97344970703125], [39.71472930908203, -104.97345733642578], [39.714656829833984, -104.97344970703125], [39.71451950073242, -104.97344207763672], [39.71427917480469, -104.9734115600586], [39.712989807128906, -104.97340393066406], [39.71291732788086, -104.97340393066406], [39.71285629272461, -104.97340393066406], [39.71247482299805, -104.97340393066406], [39.711551666259766, -104.97340393066406], [39.71137237548828, -104.97340393066406], [39.711204528808594, -104.97340393066406], [39.71110534667969, -104.97340393066406], [39.711021423339844, -104.97340393066406], [39.7109489440918, -104.97340393066406], [39.710845947265625, -104.97340393066406], [39.709381103515625, -104.973388671875], [39.70929718017578, -104.973388671875], [39.707820892333984, -104.973388671875], [39.707557678222656, -104.973388671875], [39.70749282836914, -104.973388671875], [39.707420349121094, -104.973388671875], [39.70706558227539, -104.97338104248047], [39.70566940307617, -104.97337341308594], [39.70392990112305, -104.973388671875], [39.70386505126953, -104.973388671875], [39.70379638671875, -104.973388671875], [39.70212173461914, -104.97339630126953], [39.70206069946289, -104.97339630126953], [39.70030212402344, -104.97339630126953], [39.70021057128906, -104.97340393066406], [39.700138092041016, -104.97339630126953], [39.698429107666016, -104.97340393066406], [39.69660949707031, -104.9734115600586], [39.69625473022461, -104.97341918945312], [39.694740295410156, -104.97341918945312], [39.694149017333984, -104.97343444824219], [39.69339370727539, -104.97343444824219], [39.69325256347656, -104.97343444824219], [39.69306564331055, -104.97342681884766], [39.6929817199707, -104.97342681884766], [39.6929817199707, -104.97357177734375], [39.6929817199707, -104.97403717041016], [39.6929817199707, -104.974609375], [39.69298553466797, -104.97517395019531], [39.6929817199707, -104.97577667236328]],\n", + " {"bubblingMouseEvents": true, "color": "#8104a7", "dashArray": null, "dashOffset": null, "fill": false, "fillColor": "#8104a7", "fillOpacity": 0.2, "fillRule": "evenodd", "lineCap": "round", "lineJoin": "round", "noClip": false, "opacity": 0.8, "smoothFactor": 1.0, "stroke": true, "weight": 10}\n", + " ).addTo(map_77bceaf15f8f92685ff07972bb265db5);\n", " \n", " \n", - " poly_line_dc18b3d37d1d8229eeaee4b38b80d4ef.bindTooltip(\n", + " poly_line_467195630d68d76e5593e7251fc8958f.bindTooltip(\n", " `<div>\n", - " 0.28001659029997283\n", + " 0.280412929797509\n", " </div>`,\n", " {"sticky": true}\n", " );\n", " \n", " \n", - " var marker_3bdf654f47407dce4feca9fa68867609 = L.marker(\n", + " var marker_f8fb9f6234222553ddb96ee8b37c8ae1 = L.marker(\n", " [39.77902603149414, -104.96916961669922],\n", " {}\n", - " ).addTo(map_63d14900899a70183befbf30d354febd);\n", + " ).addTo(map_77bceaf15f8f92685ff07972bb265db5);\n", " \n", " \n", - " var icon_7f8da654b577358175378d695d3554c9 = L.AwesomeMarkers.icon(\n", + " var icon_727ad313e66aff1a5fd555ee887af284 = L.AwesomeMarkers.icon(\n", " {"extraClasses": "fa-rotate-0", "icon": "circle", "iconColor": "white", "markerColor": "green", "prefix": "fa"}\n", " );\n", - " marker_3bdf654f47407dce4feca9fa68867609.setIcon(icon_7f8da654b577358175378d695d3554c9);\n", + " marker_f8fb9f6234222553ddb96ee8b37c8ae1.setIcon(icon_727ad313e66aff1a5fd555ee887af284);\n", " \n", " \n", - " marker_3bdf654f47407dce4feca9fa68867609.bindTooltip(\n", + " marker_f8fb9f6234222553ddb96ee8b37c8ae1.bindTooltip(\n", " `<div>\n", " Origin\n", " </div>`,\n", @@ -1160,19 +1168,19 @@ " );\n", " \n", " \n", - " var marker_544ae1d7c714f4dfb886e68f5b2b7f9f = L.marker(\n", + " var marker_e81ff4662acbccd08209491db4657bd7 = L.marker(\n", " [39.6929817199707, -104.97577667236328],\n", " {}\n", - " ).addTo(map_63d14900899a70183befbf30d354febd);\n", + " ).addTo(map_77bceaf15f8f92685ff07972bb265db5);\n", " \n", " \n", - " var icon_2f6f228696d0f9e6b923b6575750ad35 = L.AwesomeMarkers.icon(\n", + " var icon_0081b7beb646617f32b06a6aaa6096ad = L.AwesomeMarkers.icon(\n", " {"extraClasses": "fa-rotate-0", "icon": "circle", "iconColor": "white", "markerColor": "red", "prefix": "fa"}\n", " );\n", - " marker_544ae1d7c714f4dfb886e68f5b2b7f9f.setIcon(icon_2f6f228696d0f9e6b923b6575750ad35);\n", + " marker_e81ff4662acbccd08209491db4657bd7.setIcon(icon_0081b7beb646617f32b06a6aaa6096ad);\n", " \n", " \n", - " marker_544ae1d7c714f4dfb886e68f5b2b7f9f.bindTooltip(\n", + " marker_e81ff4662acbccd08209491db4657bd7.bindTooltip(\n", " `<div>\n", " Destination\n", " </div>`,\n", @@ -1180,33 +1188,33 @@ " );\n", " \n", " \n", - " var poly_line_2d4942578ecd49eeb510a47f63676b70 = L.polyline(\n", + " var poly_line_26538eb8c516090da23fe9fc05784049 = L.polyline(\n", " [[39.77902603149414, -104.96916961669922], [39.77896499633789, -104.96910095214844], [39.77878189086914, -104.96887969970703], [39.778751373291016, -104.96884155273438], [39.778587341308594, -104.9686508178711], [39.778480529785156, -104.96852111816406], [39.77839660644531, -104.96841430664062], [39.778324127197266, -104.96851348876953], [39.778228759765625, -104.96863555908203], [39.778076171875, -104.96883392333984], [39.777626037597656, -104.96941375732422], [39.7774772644043, -104.96961212158203], [39.77703857421875, -104.97017669677734], [39.776737213134766, -104.97056579589844], [39.77666473388672, -104.97066497802734], [39.7766227722168, -104.97071838378906], [39.776371002197266, -104.97103881835938], [39.77608108520508, -104.9714126586914], [39.775978088378906, -104.97154998779297], [39.774810791015625, -104.97306060791016], [39.77475357055664, -104.97313690185547], [39.77467346191406, -104.97323608398438], [39.77459716796875, -104.97333526611328], [39.773738861083984, -104.97444915771484], [39.77363586425781, -104.97457885742188], [39.773353576660156, -104.97494506835938], [39.77303695678711, -104.97535705566406], [39.77297592163086, -104.9754409790039], [39.77290344238281, -104.97553253173828], [39.772544860839844, -104.97599029541016], [39.772335052490234, -104.97626495361328], [39.77207946777344, -104.97659301757812], [39.771976470947266, -104.97673034667969], [39.771705627441406, -104.97708129882812], [39.77125549316406, -104.97766876220703], [39.77039337158203, -104.9787826538086], [39.77033233642578, -104.9788589477539], [39.770263671875, -104.97895050048828], [39.769996643066406, -104.97929382324219], [39.76996994018555, -104.9793930053711], [39.76955032348633, -104.97993469238281], [39.769287109375, -104.98027801513672], [39.7691764831543, -104.98041534423828], [39.769039154052734, -104.98059844970703], [39.76881408691406, -104.98088836669922], [39.768733978271484, -104.98099517822266], [39.76864242553711, -104.98110961914062], [39.76858901977539, -104.9811782836914], [39.76840591430664, -104.98141479492188], [39.7678337097168, -104.98216247558594], [39.76764678955078, -104.9823989868164], [39.76701736450195, -104.98321533203125], [39.766876220703125, -104.9833984375], [39.766693115234375, -104.98363494873047], [39.766624450683594, -104.98372650146484], [39.76657485961914, -104.9837875366211], [39.76653289794922, -104.98384857177734], [39.766178131103516, -104.98429870605469], [39.766021728515625, -104.98450469970703], [39.76557159423828, -104.98507690429688], [39.76506423950195, -104.98573303222656], [39.76464080810547, -104.98627471923828], [39.764495849609375, -104.98646545410156], [39.76442337036133, -104.98655700683594], [39.764347076416016, -104.98664855957031], [39.763763427734375, -104.98738861083984], [39.76366424560547, -104.98750305175781], [39.763572692871094, -104.98760223388672], [39.76346969604492, -104.98770141601562], [39.76337814331055, -104.98778533935547], [39.76326370239258, -104.98786926269531], [39.76316452026367, -104.9879379272461], [39.76304626464844, -104.98800659179688], [39.762916564941406, -104.98807525634766], [39.762813568115234, -104.98812103271484], [39.76266860961914, -104.98816680908203], [39.76252746582031, -104.98819732666016], [39.762413024902344, -104.98820495605469], [39.762290954589844, -104.98821258544922], [39.76216125488281, -104.98821258544922], [39.762062072753906, -104.98820495605469], [39.76179885864258, -104.9881591796875], [39.76161575317383, -104.98812103271484], [39.76142501831055, -104.98808288574219], [39.76121520996094, -104.98804473876953], [39.75983428955078, -104.98776245117188], [39.75953674316406, -104.98770141601562], [39.75946807861328, -104.98763275146484], [39.75936508178711, -104.98761749267578], [39.7588996887207, -104.987548828125], [39.758426666259766, -104.98747253417969], [39.758270263671875, -104.9874496459961], [39.75812911987305, -104.98744201660156], [39.75796127319336, -104.98744201660156], [39.75767517089844, -104.9874267578125], [39.75761032104492, -104.9874267578125], [39.75736618041992, -104.98741912841797], [39.757179260253906, -104.98741149902344], [39.757102966308594, -104.98741149902344], [39.756961822509766, -104.98741149902344], [39.756805419921875, -104.9874038696289], [39.756404876708984, -104.9874038696289], [39.75601577758789, -104.9874038696289], [39.75578308105469, -104.98739624023438], [39.75563049316406, -104.98739624023438], [39.755496978759766, -104.98739624023438], [39.75507736206055, -104.98738861083984], [39.75468063354492, -104.98738861083984], [39.75453186035156, -104.98738098144531], [39.754398345947266, -104.98738098144531], [39.754364013671875, -104.98738098144531], [39.75428009033203, -104.98738861083984], [39.75388717651367, -104.98738861083984], [39.753604888916016, -104.98738861083984], [39.75343704223633, -104.98738861083984], [39.7532844543457, -104.98738861083984], [39.75323486328125, -104.98738861083984], [39.75295639038086, -104.98738861083984], [39.752933502197266, -104.98738861083984], [39.752662658691406, -104.98738861083984], [39.752262115478516, -104.98738861083984], [39.75224304199219, -104.98738098144531], [39.751827239990234, -104.98738098144531], [39.75173568725586, -104.98738098144531], [39.751583099365234, -104.98738098144531], [39.75148391723633, -104.98738098144531], [39.751014709472656, -104.98738098144531], [39.750850677490234, -104.98738098144531], [39.75082015991211, -104.98738098144531], [39.750736236572266, -104.98738098144531], [39.750389099121094, -104.98737335205078], [39.75023651123047, -104.98737335205078], [39.7501106262207, -104.98737335205078], [39.7497673034668, -104.98737335205078], [39.74958419799805, -104.98737335205078], [39.74951171875, -104.98737335205078], [39.749412536621094, -104.98737335205078], [39.74927520751953, -104.98738098144531], [39.749168395996094, -104.98738098144531], [39.74892044067383, -104.98738098144531], [39.74868392944336, -104.98738098144531], [39.74851989746094, -104.98738098144531], [39.74822235107422, -104.98738098144531], [39.74802017211914, -104.98738098144531], [39.7476692199707, -104.98738098144531], [39.74758529663086, -104.98738098144531], [39.74751281738281, -104.98738098144531], [39.7474365234375, -104.98738098144531], [39.747379302978516, -104.98738098144531], [39.74735641479492, -104.98738098144531], [39.74730682373047, -104.98738098144531], [39.74697494506836, -104.98738098144531], [39.746944427490234, -104.98738098144531], [39.74678039550781, -104.98738861083984], [39.74667739868164, -104.98738861083984], [39.74653244018555, -104.98738861083984], [39.746482849121094, -104.98738861083984], [39.74626541137695, -104.98738861083984], [39.74618148803711, -104.98738861083984], [39.74605178833008, -104.98738861083984], [39.745948791503906, -104.98738861083984], [39.74554443359375, -104.98738861083984], [39.745418548583984, -104.98738861083984], [39.74506378173828, -104.98739624023438], [39.74480438232422, -104.98739624023438], [39.74473571777344, -104.9874038696289], [39.74430847167969, -104.98739624023438], [39.743526458740234, -104.98739624023438], [39.74336242675781, -104.98739624023438], [39.74320602416992, -104.98739624023438], [39.74302291870117, -104.98739624023438], [39.74254608154297, -104.98739624023438], [39.74245834350586, -104.98739624023438], [39.74220275878906, -104.98739624023438], [39.74214172363281, -104.98738861083984], [39.74182891845703, -104.98738861083984], [39.74177932739258, -104.98738861083984], [39.7416877746582, -104.98738861083984], [39.74163818359375, -104.98738098144531], [39.74156188964844, -104.98738861083984], [39.74149703979492, -104.98738861083984], [39.741416931152344, -104.98738861083984], [39.74137496948242, -104.98738861083984], [39.74129867553711, -104.98738861083984], [39.7410888671875, -104.98738861083984], [39.74089050292969, -104.98739624023438], [39.740875244140625, -104.98739624023438], [39.74071502685547, -104.98739624023438], [39.740631103515625, -104.98739624023438], [39.740478515625, -104.98739624023438], [39.74021530151367, -104.9874038696289], [39.74014663696289, -104.9874038696289], [39.74000930786133, -104.9874038696289], [39.73992919921875, -104.9874038696289], [39.73912048339844, -104.98739624023438], [39.73855209350586, -104.98739624023438], [39.73845672607422, -104.98739624023438], [39.73835372924805, -104.98739624023438], [39.73801040649414, -104.98739624023438], [39.73710250854492, -104.98738861083984], [39.73695373535156, -104.98738861083984], [39.736900329589844, -104.98738861083984], [39.736846923828125, -104.98738861083984], [39.73676300048828, -104.98738861083984], [39.73644256591797, -104.98739624023438], [39.73640441894531, -104.98739624023438], [39.736244201660156, -104.98739624023438], [39.73619079589844, -104.98739624023438], [39.73557662963867, -104.98739624023438], [39.73536682128906, -104.98739624023438], [39.73527526855469, -104.98739624023438], [39.73518753051758, -104.98739624023438], [39.73509216308594, -104.98739624023438], [39.73466110229492, -104.98739624023438], [39.734039306640625, -104.9874038696289], [39.73394775390625, -104.98739624023438], [39.733848571777344, -104.9874038696289], [39.73375701904297, -104.9874038696289], [39.73365020751953, -104.9874038696289], [39.73337173461914, -104.9874038696289], [39.733306884765625, -104.9874038696289], [39.732810974121094, -104.9874038696289], [39.732444763183594, -104.98741149902344], [39.732147216796875, -104.98741149902344], [39.7320556640625, -104.98741149902344], [39.731964111328125, -104.98741149902344], [39.73185348510742, -104.98741149902344], [39.73183059692383, -104.98741149902344], [39.73168182373047, -104.98741149902344], [39.73122024536133, -104.98741912841797], [39.731178283691406, -104.98741912841797], [39.73095703125, -104.98741912841797], [39.73054885864258, -104.98741912841797], [39.73045349121094, -104.98741912841797], [39.73037338256836, -104.98741912841797], [39.730010986328125, -104.98741912841797], [39.729522705078125, -104.98741912841797], [39.72943115234375, -104.98741912841797], [39.72917175292969, -104.98741912841797], [39.72908020019531, -104.98741912841797], [39.7289924621582, -104.98741912841797], [39.728668212890625, -104.98741912841797], [39.72862243652344, -104.98741912841797], [39.72829055786133, -104.98741912841797], [39.727420806884766, -104.98741912841797], [39.727298736572266, -104.98741912841797], [39.72718048095703, -104.98741912841797], [39.72665786743164, -104.9874267578125], [39.72663497924805, -104.9874267578125], [39.726600646972656, -104.9874267578125], [39.726375579833984, -104.98743438720703], [39.726226806640625, -104.98744201660156], [39.726112365722656, -104.98744201660156], [39.72574996948242, -104.98745727539062], [39.72563934326172, -104.98746490478516], [39.72555923461914, -104.98746490478516], [39.725223541259766, -104.98747253417969], [39.7251091003418, -104.98747253417969], [39.724884033203125, -104.98748016357422], [39.72444152832031, -104.98748779296875], [39.72422409057617, -104.98748779296875], [39.72414779663086, -104.98748779296875], [39.72405242919922, -104.98748779296875], [39.72385025024414, -104.98748779296875], [39.72378921508789, -104.98748779296875], [39.72360610961914, -104.98748779296875], [39.72319793701172, -104.98750305175781], [39.723026275634766, -104.98751068115234], [39.722869873046875, -104.98751068115234], [39.72262191772461, -104.98751068115234], [39.72255325317383, -104.98751068115234], [39.72245788574219, -104.98751068115234], [39.72215270996094, -104.98750305175781], [39.72206497192383, -104.98750305175781], [39.72195816040039, -104.98749542236328], [39.7216911315918, -104.98748779296875], [39.72103500366211, -104.98748779296875], [39.720943450927734, -104.98748779296875], [39.720863342285156, -104.98748779296875], [39.71990966796875, -104.98748779296875], [39.71987533569336, -104.98748779296875], [39.719451904296875, -104.98748779296875], [39.7193603515625, -104.98748779296875], [39.71928787231445, -104.98748779296875], [39.71902847290039, -104.98748779296875], [39.71844482421875, -104.98748779296875], [39.71834945678711, -104.98748779296875], [39.7182731628418, -104.98748779296875], [39.71819305419922, -104.98748779296875], [39.71754455566406, -104.98748779296875], [39.717464447021484, -104.98748779296875], [39.7174186706543, -104.98748779296875], [39.716644287109375, -104.98748779296875], [39.71656799316406, -104.98748779296875], [39.71649932861328, -104.98748779296875], [39.716224670410156, -104.98749542236328], [39.71613311767578, -104.98749542236328], [39.71575927734375, -104.98750305175781], [39.71568298339844, -104.98750305175781], [39.715606689453125, -104.98750305175781], [39.71514892578125, -104.98751831054688], [39.714847564697266, -104.98753356933594], [39.71476364135742, -104.98753356933594], [39.71470260620117, -104.98753356933594], [39.713897705078125, -104.987548828125], [39.713294982910156, -104.987548828125], [39.71319580078125, -104.987548828125], [39.71306610107422, -104.98755645751953], [39.71297836303711, -104.98755645751953], [39.71291732788086, -104.98755645751953], [39.712486267089844, -104.98756408691406], [39.71209716796875, -104.9875717163086], [39.71201705932617, -104.98757934570312], [39.71174240112305, -104.98757934570312], [39.711578369140625, -104.98758697509766], [39.711280822753906, -104.98759460449219], [39.71114730834961, -104.98759460449219], [39.71101379394531, -104.98759460449219], [39.71085739135742, -104.98759460449219], [39.71025085449219, -104.98760986328125], [39.71002960205078, -104.98760986328125], [39.70989227294922, -104.98761749267578], [39.70948028564453, -104.98761749267578], [39.70942306518555, -104.98760986328125], [39.70934295654297, -104.98761749267578], [39.709259033203125, -104.98761749267578], [39.70914840698242, -104.98762512207031], [39.7091064453125, -104.98762512207031], [39.70846939086914, -104.9876480102539], [39.70835876464844, -104.9876480102539], [39.70762634277344, -104.9876480102539], [39.707542419433594, -104.9876480102539], [39.707462310791016, -104.98764038085938], [39.7066535949707, -104.98760223388672], [39.70616912841797, -104.98758697509766], [39.705841064453125, -104.9875717163086], [39.70573806762695, -104.98756408691406], [39.705665588378906, -104.98755645751953], [39.705406188964844, -104.98754119873047], [39.705223083496094, -104.98753356933594], [39.705101013183594, -104.9875259399414], [39.704872131347656, -104.9875259399414], [39.7047119140625, -104.9875259399414], [39.704627990722656, -104.98751831054688], [39.70394515991211, -104.98751068115234], [39.703880310058594, -104.98751068115234], [39.70368194580078, -104.98750305175781], [39.70335006713867, -104.98750305175781], [39.70302963256836, -104.98750305175781], [39.7028694152832, -104.98750305175781], [39.70270538330078, -104.98751831054688], [39.70259475708008, -104.9875259399414], [39.702491760253906, -104.98753356933594], [39.70227813720703, -104.987548828125], [39.7021369934082, -104.98755645751953], [39.70204162597656, -104.98756408691406], [39.700843811035156, -104.98760986328125], [39.700782775878906, -104.98749542236328], [39.70073318481445, -104.9874038696289], [39.700687408447266, -104.98731994628906], [39.700660705566406, -104.98727416992188], [39.70061111450195, -104.9871826171875], [39.70054626464844, -104.98707580566406], [39.70036315917969, -104.98675537109375], [39.70024490356445, -104.986572265625], [39.70008850097656, -104.98636627197266], [39.700042724609375, -104.98631286621094], [39.69866180419922, -104.9847412109375], [39.6981086730957, -104.98410034179688], [39.69713592529297, -104.98299407958984], [39.696075439453125, -104.98184967041016], [39.69560623168945, -104.9813003540039], [39.69524383544922, -104.98087310791016], [39.69488525390625, -104.98043823242188], [39.694786071777344, -104.98033905029297], [39.6947021484375, -104.98025512695312], [39.69462585449219, -104.98019409179688], [39.6945915222168, -104.98016357421875], [39.69448471069336, -104.98009490966797], [39.69437789916992, -104.98002624511719], [39.6942024230957, -104.97992706298828], [39.69398498535156, -104.97977447509766], [39.69377899169922, -104.97953033447266], [39.69367599487305, -104.97943115234375], [39.69364929199219, -104.97940063476562], [39.69361114501953, -104.97936248779297], [39.69355010986328, -104.97931671142578], [39.6934814453125, -104.9792709350586], [39.69330978393555, -104.97918701171875], [39.69325637817383, -104.97914123535156], [39.693199157714844, -104.97909545898438], [39.693145751953125, -104.97903442382812], [39.693092346191406, -104.97897338867188], [39.69303512573242, -104.97886657714844], [39.69300079345703, -104.97877502441406], [39.69298553466797, -104.97871398925781], [39.69298553466797, -104.97850036621094], [39.69298553466797, -104.97827911376953], [39.69298553466797, -104.97792053222656], [39.6929931640625, -104.9774398803711], [39.6929931640625, -104.9773941040039], [39.6929931640625, -104.9772720336914], [39.692989349365234, -104.97704315185547], [39.692989349365234, -104.97694396972656], [39.692989349365234, -104.97685241699219], [39.69298553466797, -104.97636413574219], [39.6929817199707, -104.97577667236328]],\n", " {"bubblingMouseEvents": true, "color": "#0c0786", "dashArray": null, "dashOffset": null, "fill": false, "fillColor": "#0c0786", "fillOpacity": 0.2, "fillRule": "evenodd", "lineCap": "round", "lineJoin": "round", "noClip": false, "opacity": 0.8, "smoothFactor": 1.0, "stroke": true, "weight": 10}\n", - " ).addTo(map_63d14900899a70183befbf30d354febd);\n", + " ).addTo(map_77bceaf15f8f92685ff07972bb265db5);\n", " \n", " \n", - " poly_line_2d4942578ecd49eeb510a47f63676b70.bindTooltip(\n", + " poly_line_26538eb8c516090da23fe9fc05784049.bindTooltip(\n", " `<div>\n", - " 0.26646891776649895\n", + " 0.2664853255906838\n", " </div>`,\n", " {"sticky": true}\n", " );\n", " \n", " \n", - " var marker_b9ee2d187851ba8e0a0ff1154ede17fc = L.marker(\n", + " var marker_50f1588a12c520bb262fc8c55a243907 = L.marker(\n", " [39.77902603149414, -104.96916961669922],\n", " {}\n", - " ).addTo(map_63d14900899a70183befbf30d354febd);\n", + " ).addTo(map_77bceaf15f8f92685ff07972bb265db5);\n", " \n", " \n", - " var icon_793a9186e619a2b12a3f732ea3a61bd5 = L.AwesomeMarkers.icon(\n", + " var icon_50d53abde7ef57cbeed74c98cc5cec60 = L.AwesomeMarkers.icon(\n", " {"extraClasses": "fa-rotate-0", "icon": "circle", "iconColor": "white", "markerColor": "green", "prefix": "fa"}\n", " );\n", - " marker_b9ee2d187851ba8e0a0ff1154ede17fc.setIcon(icon_793a9186e619a2b12a3f732ea3a61bd5);\n", + " marker_50f1588a12c520bb262fc8c55a243907.setIcon(icon_50d53abde7ef57cbeed74c98cc5cec60);\n", " \n", " \n", - " marker_b9ee2d187851ba8e0a0ff1154ede17fc.bindTooltip(\n", + " marker_50f1588a12c520bb262fc8c55a243907.bindTooltip(\n", " `<div>\n", " Origin\n", " </div>`,\n", @@ -1214,19 +1222,19 @@ " );\n", " \n", " \n", - " var marker_3d7b473615d4cde876195834b12934ae = L.marker(\n", + " var marker_b77a6be72ef94bcf92969d8a1c82f68f = L.marker(\n", " [39.6929817199707, -104.97577667236328],\n", " {}\n", - " ).addTo(map_63d14900899a70183befbf30d354febd);\n", + " ).addTo(map_77bceaf15f8f92685ff07972bb265db5);\n", " \n", " \n", - " var icon_c58caf8f919cb00c88bbd6461e14fecd = L.AwesomeMarkers.icon(\n", + " var icon_39a9568a303e01f77af42b73eab40e7f = L.AwesomeMarkers.icon(\n", " {"extraClasses": "fa-rotate-0", "icon": "circle", "iconColor": "white", "markerColor": "red", "prefix": "fa"}\n", " );\n", - " marker_3d7b473615d4cde876195834b12934ae.setIcon(icon_c58caf8f919cb00c88bbd6461e14fecd);\n", + " marker_b77a6be72ef94bcf92969d8a1c82f68f.setIcon(icon_39a9568a303e01f77af42b73eab40e7f);\n", " \n", " \n", - " marker_3d7b473615d4cde876195834b12934ae.bindTooltip(\n", + " marker_b77a6be72ef94bcf92969d8a1c82f68f.bindTooltip(\n", " `<div>\n", " Destination\n", " </div>`,\n", @@ -1237,10 +1245,10 @@ "</html>\" style=\"position:absolute;width:100%;height:100%;left:0;top:0;border:none !important;\" allowfullscreen webkitallowfullscreen mozallowfullscreen></iframe></div></div>" ], "text/plain": [ - "<folium.folium.Map at 0x1272b78d0>" + "<folium.folium.Map at 0x1a655fe20>" ] }, - "execution_count": 188, + "execution_count": 18, "metadata": {}, "output_type": "execute_result" } @@ -1488,7 +1496,7 @@ "name": "python", "nbconvert_exporter": "python", "pygments_lexer": "ipython3", - "version": "3.11.5" + "version": "3.9.15" } }, "nbformat": 4, diff --git a/python/nrel/routee/compass/resources/osm_default_distance.toml b/python/nrel/routee/compass/resources/osm_default_distance.toml index d93087b6..5e0fe982 100644 --- a/python/nrel/routee/compass/resources/osm_default_distance.toml +++ b/python/nrel/routee/compass/resources/osm_default_distance.toml @@ -30,10 +30,10 @@ sharp_left = 3.5 u_turn = 9.5 # based on 65.5 cents per mile 2023 IRS mileage rate, $/mile -[cost.vehicle_state_variable_rates.distance] +[cost.vehicle_rates.distance] type = "factor" factor = 0.655 -[cost.state_variable_coefficients] +[cost.weights] distance = 1 [plugin] diff --git a/python/nrel/routee/compass/resources/osm_default_energy.toml b/python/nrel/routee/compass/resources/osm_default_energy.toml index bf960dfc..6f602e33 100644 --- a/python/nrel/routee/compass/resources/osm_default_energy.toml +++ b/python/nrel/routee/compass/resources/osm_default_energy.toml @@ -24,26 +24,26 @@ output_plugins = [ { type = "uuid", uuid_input_file = "vertices-uuid-enumerated.txt.gz" }, ] # based on 65.5 cents per mile 2023 IRS mileage rate, $/mile -[cost.vehicle_state_variable_rates.distance] +[cost.vehicle_rates.distance] type = "factor" factor = 0.655 # based on $20/hr approximation of 2023 median hourly wages, $/second -[cost.vehicle_state_variable_rates.time] +[cost.vehicle_rates.time] type = "factor" factor = 0.333336 # based on AAA regular unleaded gas prices sampled 12/21/2023 -[cost.vehicle_state_variable_rates.energy_liquid] +[cost.vehicle_rates.energy_liquid] type = "factor" factor = 3.120 # based on $0.50/kWh approximation of DCFC charge rates, $/kWhtype = "factor" -[cost.vehicle_state_variable_rates.energy_electric] +[cost.vehicle_rates.energy_electric] type = "factor" factor = 0.50 -[cost.state_variable_coefficients] +[cost.weights] distance = 1 time = 1 energy_liquid = 1 diff --git a/python/nrel/routee/compass/resources/osm_default_speed.toml b/python/nrel/routee/compass/resources/osm_default_speed.toml index 3a3203bb..16a96e07 100644 --- a/python/nrel/routee/compass/resources/osm_default_speed.toml +++ b/python/nrel/routee/compass/resources/osm_default_speed.toml @@ -34,16 +34,16 @@ sharp_left = 3.5 u_turn = 9.5 # based on 65.5 cents per mile 2023 IRS mileage rate, $/mile -[cost.vehicle_state_variable_rates.distance] +[cost.vehicle_rates.distance] type = "factor" factor = 0.655 # based on $20/hr approximation of 2023 median hourly wages, $/second -[cost.vehicle_state_variable_rates.time] +[cost.vehicle_rates.time] type = "factor" factor = 0.333336 -[cost.state_variable_coefficients] +[cost.weights] distance = 1 time = 1 diff --git a/rust/routee-compass/Cargo.toml b/rust/routee-compass/Cargo.toml index 4f606c3c..c626e814 100644 --- a/rust/routee-compass/Cargo.toml +++ b/rust/routee-compass/Cargo.toml @@ -43,3 +43,4 @@ itertools = { workspace = true } ordered-float = { workspace = true } allocative = { workspace = true } indoc = { workspace = true } +ordered_hash_map = { version = "0.4.0", features = ["serde"] } diff --git a/rust/routee-compass/src/app/compass/compass_app.rs b/rust/routee-compass/src/app/compass/compass_app.rs index fa68339b..12a0bbcf 100644 --- a/rust/routee-compass/src/app/compass/compass_app.rs +++ b/rust/routee-compass/src/app/compass/compass_app.rs @@ -1,5 +1,5 @@ use super::response::response_output_policy::ResponseOutputPolicy; -use super::response::response_writer::ResponseWriter; +use super::response::response_sink::ResponseSink; use super::{ compass_app_ops as ops, config::compass_app_builder::CompassAppBuilder, search_orientation::SearchOrientation, @@ -507,7 +507,7 @@ pub fn run_batch_with_responses( search_orientation: &SearchOrientation, output_plugins: &[Arc<dyn OutputPlugin>], search_app: &SearchApp, - response_writer: &ResponseWriter, + response_writer: &ResponseSink, pb: Arc<Mutex<Bar>>, ) -> Result<Box<dyn Iterator<Item = Value>>, CompassAppError> { let run_query_result = load_balanced_inputs @@ -516,12 +516,12 @@ pub fn run_batch_with_responses( queries .iter() .map(\|q\| { - let response = + let mut response = run_single_query(q, search_orientation, output_plugins, search_app)?; if let Ok(mut pb_local) = pb.lock() { let _ = pb_local.update(1); } - response_writer.write_response(&response)?; + response_writer.write_response(&mut response)?; Ok(response) }) .collect::<Result<Vec<serde_json::Value>, CompassAppError>>() @@ -542,7 +542,7 @@ pub fn run_batch_without_responses( search_orientation: &SearchOrientation, output_plugins: &[Arc<dyn OutputPlugin>], search_app: &SearchApp, - response_writer: &ResponseWriter, + response_writer: &ResponseSink, pb: Arc<Mutex<Bar>>, ) -> Result<Box<dyn Iterator<Item = Value>>, CompassAppError> { // run the computations, discard values that do not trigger an error @@ -554,11 +554,12 @@ pub fn run_batch_without_responses( // within a for loop or for_each call, and map creates more allocations. open to other ideas! let initial: Result<(), CompassAppError> = Ok(()); let _ = queries.iter().fold(initial, \|_, q\| { - let response = run_single_query(q, search_orientation, output_plugins, search_app)?; + let mut response = + run_single_query(q, search_orientation, output_plugins, search_app)?; if let Ok(mut pb_local) = pb.lock() { let _ = pb_local.update(1); } - response_writer.write_response(&response)?; + response_writer.write_response(&mut response)?; Ok(()) }); Ok(()) diff --git a/rust/routee-compass/src/app/compass/config.default.toml b/rust/routee-compass/src/app/compass/config.default.toml index f2d74cba..2df35edf 100644 --- a/rust/routee-compass/src/app/compass/config.default.toml +++ b/rust/routee-compass/src/app/compass/config.default.toml @@ -2,12 +2,12 @@ parallelism = 2 search_orientation = "vertex" response_persistence_policy = "persist_response_in_memory" [response_output_policy] -type = "no_output" +type = "none" # # example file output policy # response_persistence_policy = "discard_response_from_memory" # [response_output_policy] -# type = "run_batch_file_sink" +# type = "file" # filename = "output.json" # format = { type = "json", newline_delimited = true } @@ -26,7 +26,7 @@ type = "no_access_model" [cost] cost_aggregation = "sum" -network_state_variable_rates = {} +network_rates = {} [frontier] type = "no_restriction" diff --git a/rust/routee-compass/src/app/compass/config/cost_model/cost_model_builder.rs b/rust/routee-compass/src/app/compass/config/cost_model/cost_model_builder.rs index 04a3c623..dfb2f841 100644 --- a/rust/routee-compass/src/app/compass/config/cost_model/cost_model_builder.rs +++ b/rust/routee-compass/src/app/compass/config/cost_model/cost_model_builder.rs @@ -19,19 +19,19 @@ impl CostModelBuilder { ) -> Result<CostModelService, CompassConfigurationError> { let parent_key = CompassConfigurationField::Cost.to_string(); let vehicle_rates: HashMap<String, VehicleCostRate> = - config.get_config_serde(&"vehicle_state_variable_rates", &parent_key)?; + config.get_config_serde(&"vehicle_rates", &parent_key)?; let network_rates: HashMap<String, NetworkCostRate> = - config.get_config_serde(&"network_state_variable_rates", &parent_key)?; + config.get_config_serde(&"network_rates", &parent_key)?; let coefficients: HashMap<String, f64> = - config.get_config_serde(&"state_variable_coefficients", &parent_key)?; + config.get_config_serde(&"weights", &parent_key)?; let cost_aggregation: CostAggregation = config.get_config_serde(&"cost_aggregation", &parent_key)?; let model = CostModelService { - vehicle_state_variable_rates: Arc::new(vehicle_rates), - network_state_variable_rates: Arc::new(network_rates), - state_variable_coefficients: Arc::new(coefficients), + vehicle_rates: Arc::new(vehicle_rates), + network_rates: Arc::new(network_rates), + weights: Arc::new(coefficients), cost_aggregation, }; Ok(model) diff --git a/rust/routee-compass/src/app/compass/config/cost_model/cost_model_service.rs b/rust/routee-compass/src/app/compass/config/cost_model/cost_model_service.rs index d605ce22..e863a3d1 100644 --- a/rust/routee-compass/src/app/compass/config/cost_model/cost_model_service.rs +++ b/rust/routee-compass/src/app/compass/config/cost_model/cost_model_service.rs @@ -13,9 +13,9 @@ use std::{ }; pub struct CostModelService { - pub vehicle_state_variable_rates: Arc<HashMap<String, VehicleCostRate>>, - pub network_state_variable_rates: Arc<HashMap<String, NetworkCostRate>>, - pub state_variable_coefficients: Arc<HashMap<String, f64>>, + pub vehicle_rates: Arc<HashMap<String, VehicleCostRate>>, + pub network_rates: Arc<HashMap<String, NetworkCostRate>>, + pub weights: Arc<HashMap<String, f64>>, pub cost_aggregation: CostAggregation, } @@ -52,12 +52,9 @@ impl CostModelService { // at minimum, we default to the "distance" traveled. // invariant: this hashmap dictates the list of keys for all subsequent CostModel hashmaps. let weights: Arc<HashMap<String, f64>> = query - .get_config_serde_optional::<HashMap<String, f64>>( - &"state_variable_coefficients", - &"cost_model", - )? + .get_config_serde_optional::<HashMap<String, f64>>(&"weights", &"cost_model")? .map(Arc::new) - .unwrap_or(self.state_variable_coefficients.clone()); + .unwrap_or(self.weights.clone()); // // union the requested state variables with those in the existing traversal model // // load only indices that appear in coefficients object @@ -88,12 +85,12 @@ impl CostModelService { // the user can append/replace rates from the query let vehicle_rates = query .get_config_serde_optional::<HashMap<String, VehicleCostRate>>( - &"vehicle_state_variable_rates", + &"vehicle_rates", &"cost_model", ) .map(\|opt_rates\| match opt_rates { Some(rates) => Arc::new(rates), - None => self.vehicle_state_variable_rates.clone(), + None => self.vehicle_rates.clone(), })?; let cost_aggregation: CostAggregation = query @@ -103,7 +100,7 @@ impl CostModelService { let model = CostModel::new( weights, vehicle_rates, - self.network_state_variable_rates.clone(), + self.network_rates.clone(), cost_aggregation, state_model, ) diff --git a/rust/routee-compass/src/app/compass/response/csv/csv_mapping.rs b/rust/routee-compass/src/app/compass/response/csv/csv_mapping.rs new file mode 100644 index 00000000..f7533590 --- /dev/null +++ b/rust/routee-compass/src/app/compass/response/csv/csv_mapping.rs @@ -0,0 +1,88 @@ +use itertools::Itertools; +use serde::{Deserialize, Serialize}; +use serde_json::json; + +#[derive(Serialize, Deserialize, Clone, Debug)] +#[serde(untagged)] +pub enum CsvMapping { + Path(String), + Sum { sum: Vec<Box<CsvMapping>> }, + Optional { optional: Box<CsvMapping> }, +} + +impl CsvMapping { + // const CSV_ERR_SEPARATOR: &'static str = "-"; + + // /// applies a user-specified JSON to CSV mapping. if the mapping fails, + // /// a helpful error message is returned. + pub fn apply_mapping(&self, json: &serde_json::Value) -> Result<serde_json::Value, String> { + match self { + CsvMapping::Path(p) => { + let split_path = p.split('.').collect_vec(); + traverse(json, &split_path) + } + CsvMapping::Sum { sum } => { + let (ok, err): (Vec<_>, Vec<_>) = + sum.iter().map(\|m\| m.apply_mapping(json)).partition_result(); + if !err.is_empty() { + let valid = if ok.is_empty() { + String::from("") + } else { + format!(" with these valid paths: {}", ok.iter().join(", ")) + }; + let invalid = err.join(", "); + let msg = format!("unable to sum invalid paths: {}{}", invalid, valid); + Err(msg) + } else { + let (nums, num_errs): (Vec<_>, Vec<_>) = ok + .iter() + .map(\|v\| match v { + serde_json::Value::Null => Ok(0.0), + serde_json::Value::Number(n) => { + n.as_f64().ok_or_else(\|\| format!("invalid number {}", v)) + } + _ => Err(format!("expected a number, found {}", v)), + }) + .partition_result(); + if !num_errs.is_empty() { + let valid = if nums.is_empty() { + String::from("") + } else { + format!(" with these valid numbers: {}", nums.iter().join(", ")) + }; + let invalid = num_errs.join(", "); + let msg = format!("unable to sum invalid numbers: {}{}", invalid, valid); + Err(msg) + } else { + Ok(json![nums.into_iter().sum::<f64>()]) + } + } + } + CsvMapping::Optional { optional } => match optional.apply_mapping(json).ok() { + Some(value) => Ok(value), + None => Ok(serde_json::Value::Null), + }, + } + } +} + +fn traverse(value: &serde_json::Value, path: &Vec<&str>) -> Result<serde_json::Value, String> { + let mut cursor = value; + let mut remaining = path.as_slice(); + while let Some(next) = remaining.first() { + match cursor.get(next) { + None => { + let path_str = path.iter().join("."); + return Err(format!( + "could not find object {} in path {}", + next, path_str + )); + } + Some(child) => { + cursor = child; + remaining = &remaining[1..]; + } + } + } + Ok(cursor.clone()) +} diff --git a/rust/routee-compass/src/app/compass/response/csv/mod.rs b/rust/routee-compass/src/app/compass/response/csv/mod.rs new file mode 100644 index 00000000..245298cc --- /dev/null +++ b/rust/routee-compass/src/app/compass/response/csv/mod.rs @@ -0,0 +1,1 @@ +pub mod csv_mapping; diff --git a/rust/routee-compass/src/app/compass/response/mod.rs b/rust/routee-compass/src/app/compass/response/mod.rs index 3f0300cf..208500b2 100644 --- a/rust/routee-compass/src/app/compass/response/mod.rs +++ b/rust/routee-compass/src/app/compass/response/mod.rs @@ -1,5 +1,6 @@ +pub mod csv; pub mod response_output_format; pub mod response_output_format_json; pub mod response_output_policy; pub mod response_persistence_policy; -pub mod response_writer; +pub mod response_sink; diff --git a/rust/routee-compass/src/app/compass/response/response_output_format.rs b/rust/routee-compass/src/app/compass/response/response_output_format.rs index 521b9e7a..1a683972 100644 --- a/rust/routee-compass/src/app/compass/response/response_output_format.rs +++ b/rust/routee-compass/src/app/compass/response/response_output_format.rs @@ -1,11 +1,24 @@ -use super::response_output_format_json as json_ops; +use super::{csv::csv_mapping::CsvMapping, response_output_format_json as json_ops}; use crate::app::compass::compass_app_error::CompassAppError; +use itertools::Itertools; +use ordered_hash_map::OrderedHashMap; use serde::{Deserialize, Serialize}; +use serde_json::json; +use std::collections::HashMap; -#[derive(Deserialize, Serialize, Debug, Clone, Copy)] +#[derive(Deserialize, Serialize, Debug, Clone)] #[serde(rename_all = "snake_case", tag = "type")] pub enum ResponseOutputFormat { + /// writes outputs to a JSON file as either ECMA-404 JSON or as newline-delimited + /// JSON. Json { newline_delimited: bool }, + /// writes outputs to a CSV file given the specified mapping. by default, column + /// order matches the order of keys in the map, unless "sorted" is true, in which + /// case the fields are sorted lexicagraphically. + Csv { + mapping: OrderedHashMap<String, CsvMapping>, + sorted: bool, + }, } impl ResponseOutputFormat { @@ -14,6 +27,14 @@ impl ResponseOutputFormat { ResponseOutputFormat::Json { newline_delimited } => { json_ops::initial_file_contents(newline_delimited) } + ResponseOutputFormat::Csv { mapping, sorted } => { + let header = if sorted { + mapping.keys().sorted().join(",") + } else { + mapping.keys().rev().join(",") + }; + Some(format!("{}\n", header)) + } } } @@ -22,14 +43,54 @@ impl ResponseOutputFormat { ResponseOutputFormat::Json { newline_delimited } => { json_ops::final_file_contents(newline_delimited) } + ResponseOutputFormat::Csv { + mapping: _, + sorted: _, + } => None, } } - pub fn format_response(&self, response: &serde_json::Value) -> Result<String, CompassAppError> { + pub fn format_response( + &self, + response: &mut serde_json::Value, + ) -> Result<String, CompassAppError> { match self { ResponseOutputFormat::Json { newline_delimited } => { json_ops::format_response(response, newline_delimited) } + ResponseOutputFormat::Csv { mapping, sorted } => { + let mut errors: HashMap<String, String> = HashMap::new(); + let row = if sorted { + mapping + .iter() + .sorted_by_key(\|(k, _)\| k) + .map(\|(k, v)\| match v.apply_mapping(response) { + Ok(cell) => cell.to_string(), + Err(msg) => { + errors.insert(k.clone(), msg); + String::from("") + } + }) + .join(",") + } else { + mapping + .iter() + .rev() + .map(\|(k, v)\| match v.apply_mapping(response) { + Ok(cell) => cell.to_string(), + Err(msg) => { + errors.insert(k.clone(), msg); + String::from("") + } + }) + .join(",") + }; + + if !errors.is_empty() { + response["error"] = json![{"csv": json![errors]}]; + } + Ok(row) + } } } @@ -38,6 +99,10 @@ impl ResponseOutputFormat { ResponseOutputFormat::Json { newline_delimited } => { json_ops::delimiter(newline_delimited) } + ResponseOutputFormat::Csv { + mapping: _, + sorted: _, + } => Some(String::from("\n")), } } } diff --git a/rust/routee-compass/src/app/compass/response/response_output_policy.rs b/rust/routee-compass/src/app/compass/response/response_output_policy.rs index cd64f8e7..305eef32 100644 --- a/rust/routee-compass/src/app/compass/response/response_output_policy.rs +++ b/rust/routee-compass/src/app/compass/response/response_output_policy.rs @@ -1,4 +1,4 @@ -use super::{response_output_format::ResponseOutputFormat, response_writer::ResponseWriter}; +use super::{response_output_format::ResponseOutputFormat, response_sink::ResponseSink}; use crate::app::compass::compass_app_error::CompassAppError; use serde::{Deserialize, Serialize}; use std::{ @@ -10,21 +10,24 @@ use std::{ #[derive(Deserialize, Serialize, Clone, Debug)] #[serde(rename_all = "snake_case", tag = "type")] pub enum ResponseOutputPolicy { - NoOutput, - RunBatchFileSink { + None, + File { filename: String, format: ResponseOutputFormat, }, + Combined { + policies: Vec<Box<ResponseOutputPolicy>>, + }, } impl ResponseOutputPolicy { /// creates an instance of a writer which writes responses to some destination. /// the act of building this writer may include writing initial content to some sink, /// such as a file header. - pub fn build(&self) -> Result<ResponseWriter, CompassAppError> { + pub fn build(&self) -> Result<ResponseSink, CompassAppError> { match self { - ResponseOutputPolicy::NoOutput => Ok(ResponseWriter::NoOutputWriter), - ResponseOutputPolicy::RunBatchFileSink { filename, format } => { + ResponseOutputPolicy::None => Ok(ResponseSink::None), + ResponseOutputPolicy::File { filename, format } => { let output_file_path = PathBuf::from(filename); // initialize the file @@ -39,13 +42,20 @@ impl ResponseOutputPolicy { // wrap the file in a mutex so we can share it between threads let file_shareable = Arc::new(Mutex::new(file)); - Ok(ResponseWriter::RunBatchFileSink { + Ok(ResponseSink::File { filename: filename.clone(), file: file_shareable, - format: *format, + format: format.clone(), delimiter: format.delimiter(), }) } + ResponseOutputPolicy::Combined { policies } => { + let policies = policies + .iter() + .map(\|p\| p.build().map(Box::new)) + .collect::<Result<Vec<_>, _>>()?; + Ok(ResponseSink::Combined(policies)) + } } } } diff --git a/rust/routee-compass/src/app/compass/response/response_writer.rs b/rust/routee-compass/src/app/compass/response/response_sink.rs similarity index 65% rename from rust/routee-compass/src/app/compass/response/response_writer.rs rename to rust/routee-compass/src/app/compass/response/response_sink.rs index 584c3205..bae4ad99 100644 --- a/rust/routee-compass/src/app/compass/response/response_writer.rs +++ b/rust/routee-compass/src/app/compass/response/response_sink.rs @@ -6,21 +6,23 @@ use std::{ sync::{Arc, Mutex}, }; -pub enum ResponseWriter { - NoOutputWriter, - RunBatchFileSink { +pub enum ResponseSink { + None, + File { filename: String, file: Arc<Mutex<File>>, format: ResponseOutputFormat, delimiter: Option<String>, }, + Combined(Vec<Box<ResponseSink>>), } -impl ResponseWriter { - pub fn write_response(&self, response: &serde_json::Value) -> Result<(), CompassAppError> { +impl ResponseSink { + /// uses a writer + pub fn write_response(&self, response: &mut serde_json::Value) -> Result<(), CompassAppError> { match self { - ResponseWriter::NoOutputWriter => Ok(()), - ResponseWriter::RunBatchFileSink { + ResponseSink::None => Ok(()), + ResponseSink::File { filename: _, file, format, @@ -34,25 +36,23 @@ impl ResponseWriter { )) })?; - // if write_delimiter { - // match delimiter { - // None => {} - // Some(delim) => writeln!(file_attained, "{}", delim) - // .map_err(CompassAppError::IOError)?, - // } - // } - let output_row = format.format_response(response)?; writeln!(file_attained, "{}", output_row).map_err(CompassAppError::IOError)?; Ok(()) } + ResponseSink::Combined(policies) => { + for policy in policies { + policy.write_response(response)?; + } + Ok(()) + } } } pub fn close(&self) -> Result<String, CompassAppError> { match self { - ResponseWriter::NoOutputWriter => Ok(String::from("")), - ResponseWriter::RunBatchFileSink { + ResponseSink::None => Ok(String::from("")), + ResponseSink::File { filename, file, format, @@ -73,6 +73,17 @@ impl ResponseWriter { Ok(filename.clone()) } + ResponseSink::Combined(policies) => { + let mut out_strs = vec![]; + for policy in policies { + let out_str = policy.close()?; + if !out_str.is_empty() { + out_strs.push(out_str); + } + } + + Ok(out_strs.join(",")) + } } } } diff --git a/rust/routee-compass/src/plugin/input/default/grid_search/plugin.rs b/rust/routee-compass/src/plugin/input/default/grid_search/plugin.rs index a21316e4..44f2317d 100644 --- a/rust/routee-compass/src/plugin/input/default/grid_search/plugin.rs +++ b/rust/routee-compass/src/plugin/input/default/grid_search/plugin.rs @@ -167,21 +167,21 @@ mod test { "grid_search":{ "model_name": ["2016_TOYOTA_Camry_4cyl_2WD","2017_CHEVROLET_Bolt"], "_ignore":[ - { "name":"d1", "state_variable_coefficients": { "distance":1, "time":0, "energy_electric":0 } }, - { "name":"t1", "state_variable_coefficients": { "distance":0, "time":1, "energy_electric":0 } }, - { "name":"e1", "state_variable_coefficients": { "distance":0, "time":0, "energy_electric":1 } } + { "name":"d1", "weights": { "distance":1, "time":0, "energy_electric":0 } }, + { "name":"t1", "weights": { "distance":0, "time":1, "energy_electric":0 } }, + { "name":"e1", "weights": { "distance":0, "time":0, "energy_electric":1 } } ] } }); let plugin = GridSearchPlugin {}; plugin.process(&mut input).unwrap(); let expected = vec![ - json![{"abc":123,"model_name":"2016_TOYOTA_Camry_4cyl_2WD","name":"d1","state_variable_coefficients":{"distance":1,"energy_electric":0,"time":0}}], - json![{"abc":123,"model_name":"2016_TOYOTA_Camry_4cyl_2WD","name":"t1","state_variable_coefficients":{"distance":0,"energy_electric":0,"time":1}}], - json![{"abc":123,"model_name":"2016_TOYOTA_Camry_4cyl_2WD","name":"e1","state_variable_coefficients":{"distance":0,"energy_electric":1,"time":0}}], - json![{"abc":123,"model_name":"2017_CHEVROLET_Bolt","name":"d1","state_variable_coefficients":{"distance":1,"energy_electric":0,"time":0}}], - json![{"abc":123,"model_name":"2017_CHEVROLET_Bolt","name":"t1","state_variable_coefficients":{"distance":0,"energy_electric":0,"time":1}}], - json![{"abc":123,"model_name":"2017_CHEVROLET_Bolt","name":"e1","state_variable_coefficients":{"distance":0,"energy_electric":1,"time":0}}], + json![{"abc":123,"model_name":"2016_TOYOTA_Camry_4cyl_2WD","name":"d1","weights":{"distance":1,"energy_electric":0,"time":0}}], + json![{"abc":123,"model_name":"2016_TOYOTA_Camry_4cyl_2WD","name":"t1","weights":{"distance":0,"energy_electric":0,"time":1}}], + json![{"abc":123,"model_name":"2016_TOYOTA_Camry_4cyl_2WD","name":"e1","weights":{"distance":0,"energy_electric":1,"time":0}}], + json![{"abc":123,"model_name":"2017_CHEVROLET_Bolt","name":"d1","weights":{"distance":1,"energy_electric":0,"time":0}}], + json![{"abc":123,"model_name":"2017_CHEVROLET_Bolt","name":"t1","weights":{"distance":0,"energy_electric":0,"time":1}}], + json![{"abc":123,"model_name":"2017_CHEVROLET_Bolt","name":"e1","weights":{"distance":0,"energy_electric":1,"time":0}}], ]; match input { serde_json::Value::Array(result) => assert_eq!(result, expected),	general CSV ResponseOutputPolicy a user should be able to specify a response output policy format targeting a CSV file. to make it general, a user should be allowed to specify the relevant fields to flatten onto the CSV row as a key/value mapping from destination keys to their dot-delimited JSON paths. for example: ```toml [response_output_policy] type = "run_batch_file_sink" filename = "mep_denver_output.csv" [response_output_policy.format] type = "csv" [response_output_policy.format.mapping] src_edge = "request.origin_edge_id" dst_edge = "request.destination_edge_id" cost = "cost.total_cost" ```		2024-03-28T16:59:48	[]	[]
NREL/routee-compass	NREL__routee-compass-97	834e23e5b02cea4f6a7bb5c40b5d42aba458a50b	diff --git a/rust/routee-compass/src/app/compass/config/compass_app_builder.rs b/rust/routee-compass/src/app/compass/config/compass_app_builder.rs index 19c3b489..24182b5f 100644 --- a/rust/routee-compass/src/app/compass/config/compass_app_builder.rs +++ b/rust/routee-compass/src/app/compass/config/compass_app_builder.rs @@ -16,6 +16,7 @@ use super::{ use crate::plugin::{ input::{ default::{ + debug::debug_builder::DebugInputPluginBuilder, edge_rtree::edge_rtree_input_plugin_builder::EdgeRtreeInputPluginBuilder, grid_search::builder::GridSearchBuilder, inject::inject_builder::InjectPluginBuilder, load_balancer::builder::LoadBalancerBuilder, vertex_rtree::builder::VertexRTreeBuilder, @@ -150,12 +151,14 @@ impl CompassAppBuilder { let edge_rtree: Rc<dyn InputPluginBuilder> = Rc::new(EdgeRtreeInputPluginBuilder {}); let load_balancer: Rc<dyn InputPluginBuilder> = Rc::new(LoadBalancerBuilder {}); let inject: Rc<dyn InputPluginBuilder> = Rc::new(InjectPluginBuilder {}); + let debug: Rc<dyn InputPluginBuilder> = Rc::new(DebugInputPluginBuilder {}); let input_plugin_builders = HashMap::from([ (String::from("grid_search"), grid_search), (String::from("vertex_rtree"), vertex_tree), (String::from("edge_rtree"), edge_rtree), (String::from("load_balancer"), load_balancer), (String::from("inject"), inject), + (String::from("debug"), debug), ]); // Output plugin builders diff --git a/rust/routee-compass/src/plugin/input/default/debug/debug_builder.rs b/rust/routee-compass/src/plugin/input/default/debug/debug_builder.rs new file mode 100644 index 00000000..f82dad99 --- /dev/null +++ b/rust/routee-compass/src/plugin/input/default/debug/debug_builder.rs @@ -0,0 +1,20 @@ +use crate::{ + app::compass::config::{ + builders::InputPluginBuilder, compass_configuration_error::CompassConfigurationError, + }, + plugin::input::input_plugin::InputPlugin, +}; +use std::sync::Arc; + +use super::debug_plugin::DebugInputPlugin; + +pub struct DebugInputPluginBuilder {} + +impl InputPluginBuilder for DebugInputPluginBuilder { + fn build( + &self, + _parameters: &serde_json::Value, + ) -> Result<Arc<dyn InputPlugin>, CompassConfigurationError> { + Ok(Arc::new(DebugInputPlugin {})) + } +} diff --git a/rust/routee-compass/src/plugin/input/default/debug/debug_plugin.rs b/rust/routee-compass/src/plugin/input/default/debug/debug_plugin.rs new file mode 100644 index 00000000..3f2393f1 --- /dev/null +++ b/rust/routee-compass/src/plugin/input/default/debug/debug_plugin.rs @@ -0,0 +1,11 @@ +use crate::plugin::{input::input_plugin::InputPlugin, plugin_error::PluginError}; + +pub struct DebugInputPlugin {} + +impl InputPlugin for DebugInputPlugin { + fn process(&self, input: &serde_json::Value) -> Result<Vec<serde_json::Value>, PluginError> { + let string = serde_json::to_string_pretty(input).map_err(PluginError::JsonError)?; + println!("{}", string); + Ok(vec![input.clone()]) + } +} diff --git a/rust/routee-compass/src/plugin/input/default/debug/mod.rs b/rust/routee-compass/src/plugin/input/default/debug/mod.rs new file mode 100644 index 00000000..fcde563f --- /dev/null +++ b/rust/routee-compass/src/plugin/input/default/debug/mod.rs @@ -0,0 +1,2 @@ +pub mod debug_builder; +pub mod debug_plugin; diff --git a/rust/routee-compass/src/plugin/input/default/grid_search/plugin.rs b/rust/routee-compass/src/plugin/input/default/grid_search/plugin.rs index 18704bbc..51e93d92 100644 --- a/rust/routee-compass/src/plugin/input/default/grid_search/plugin.rs +++ b/rust/routee-compass/src/plugin/input/default/grid_search/plugin.rs @@ -1,7 +1,7 @@ use crate::plugin::input::input_field::InputField; use crate::plugin::input::input_json_extensions::InputJsonExtensions; use crate::plugin::input::input_plugin::InputPlugin; -use crate::plugin::plugin_error::PluginError as E; +use crate::plugin::plugin_error::PluginError; use routee_compass_core::util::multiset::MultiSet; /// Builds an input plugin that duplicates queries if array-valued fields are present @@ -9,13 +9,27 @@ use routee_compass_core::util::multiset::MultiSet; pub struct GridSearchPlugin {} impl InputPlugin for GridSearchPlugin { - fn process(&self, input: &serde_json::Value) -> Result<Vec<serde_json::Value>, E> { + fn process(&self, input: &serde_json::Value) -> Result<Vec<serde_json::Value>, PluginError> { match input.get_grid_search() { None => Ok(vec![input.clone()]), Some(grid_search_input) => { - let map = grid_search_input - .as_object() - .ok_or(E::UnexpectedQueryStructure(format!("{:?}", input)))?; + // prevent recursion due to nested grid search keys + let recurses = serde_json::to_string(grid_search_input) + .map_err(PluginError::JsonError)? + .contains("grid_search"); + if recurses { + return Err(PluginError::PluginFailed(String::from( + "grid search section cannot contain the string 'grid_search'", + ))); + } + + let map = + grid_search_input + .as_object() + .ok_or(PluginError::UnexpectedQueryStructure(format!( + "{:?}", + input + )))?; let mut keys: Vec<String> = vec![]; let mut multiset_input: Vec<Vec<serde_json::Value>> = vec![]; let mut multiset_indices: Vec<Vec<usize>> = vec![]; @@ -32,7 +46,10 @@ impl InputPlugin for GridSearchPlugin { // let remove_key = InputField::GridSearch.to_str(); let mut initial_map = input .as_object() - .ok_or(E::UnexpectedQueryStructure(format!("{:?}", input)))? + .ok_or(PluginError::UnexpectedQueryStructure(format!( + "{:?}", + input + )))? .clone(); initial_map.remove(InputField::GridSearch.to_str()); let initial = serde_json::json!(initial_map); @@ -154,4 +171,51 @@ mod test { ]; assert_eq!(result, expected); } + + #[test] + fn test_nested() { + let input = serde_json::json!({ + "abc": 123, + "grid_search":{ + "model_name": ["2016_TOYOTA_Camry_4cyl_2WD","2017_CHEVROLET_Bolt"], + "_ignore":[ + { "name":"d1", "state_variable_coefficients": { "distance":1, "time":0, "energy_electric":0 } }, + { "name":"t1", "state_variable_coefficients": { "distance":0, "time":1, "energy_electric":0 } }, + { "name":"e1", "state_variable_coefficients": { "distance":0, "time":0, "energy_electric":1 } } + ] + } + }); + let plugin = GridSearchPlugin {}; + let result = plugin + .process(&input) + .unwrap() + .iter() + .map(serde_json::to_string) + .collect::<Result<Vec<String>, serde_json::Error>>() + .unwrap(); + let expected = vec![ + String::from("{\"abc\":123,\"model_name\":\"2016_TOYOTA_Camry_4cyl_2WD\",\"name\":\"d1\",\"state_variable_coefficients\":{\"distance\":1,\"energy_electric\":0,\"time\":0}}"), + String::from("{\"abc\":123,\"model_name\":\"2016_TOYOTA_Camry_4cyl_2WD\",\"name\":\"t1\",\"state_variable_coefficients\":{\"distance\":0,\"energy_electric\":0,\"time\":1}}"), + String::from("{\"abc\":123,\"model_name\":\"2016_TOYOTA_Camry_4cyl_2WD\",\"name\":\"e1\",\"state_variable_coefficients\":{\"distance\":0,\"energy_electric\":1,\"time\":0}}"), + String::from("{\"abc\":123,\"model_name\":\"2017_CHEVROLET_Bolt\",\"name\":\"d1\",\"state_variable_coefficients\":{\"distance\":1,\"energy_electric\":0,\"time\":0}}"), + String::from("{\"abc\":123,\"model_name\":\"2017_CHEVROLET_Bolt\",\"name\":\"t1\",\"state_variable_coefficients\":{\"distance\":0,\"energy_electric\":0,\"time\":1}}"), + String::from("{\"abc\":123,\"model_name\":\"2017_CHEVROLET_Bolt\",\"name\":\"e1\",\"state_variable_coefficients\":{\"distance\":0,\"energy_electric\":1,\"time\":0}}"), + ]; + assert_eq!(result, expected); + } + + #[test] + pub fn test_handle_recursion() { + let input = serde_json::json!({ + "abc": 123, + "grid_search":{ + "grid_search": { + "foo": [ "a", "b" ] + } + } + }); + let plugin = GridSearchPlugin {}; + let result = plugin.process(&input); + assert!(result.is_err()); + } } diff --git a/rust/routee-compass/src/plugin/input/default/mod.rs b/rust/routee-compass/src/plugin/input/default/mod.rs index 37ed26c7..a6d22176 100644 --- a/rust/routee-compass/src/plugin/input/default/mod.rs +++ b/rust/routee-compass/src/plugin/input/default/mod.rs @@ -1,3 +1,4 @@ +pub mod debug; pub mod edge_rtree; pub mod grid_search; pub mod inject; diff --git a/rust/routee-compass/src/plugin/input/default/vertex_rtree/plugin.rs b/rust/routee-compass/src/plugin/input/default/vertex_rtree/plugin.rs index 6c2516db..6e3f7ae8 100644 --- a/rust/routee-compass/src/plugin/input/default/vertex_rtree/plugin.rs +++ b/rust/routee-compass/src/plugin/input/default/vertex_rtree/plugin.rs @@ -233,6 +233,7 @@ mod test { .join("plugin") .join("input") .join("default") + .join("vertex_rtree") .join("test") .join("rtree_vertices.csv"); @@ -241,6 +242,7 @@ mod test { .join("plugin") .join("input") .join("default") + .join("vertex_rtree") .join("test") .join("rtree_query.json"); let query_str = fs::read_to_string(query_filepath).unwrap();	endless loop when "grid_search" contains "grid_search" by mistake i injected a value at key "grid_search" which contained a value "grid_search" within it: ```json "grid_search": { "grid_search": { "_details": [ {}, {}, {}, {} ], } } ``` this causes an infinite loop. endless loop when "grid_search" contains "grid_search" by mistake i injected a value at key "grid_search" which contained a value "grid_search" within it: ```json "grid_search": { "grid_search": { "_details": [ {}, {}, {}, {} ], } } ``` this causes an infinite loop.		2024-01-03T23:17:11	[]	[]
NREL/routee-compass	NREL__routee-compass-33	442f9a671677b2e59e1067d7efc23dfc9cf2e604	diff --git a/docs/notebooks/open_street_maps_example.ipynb b/docs/notebooks/open_street_maps_example.ipynb index 505fbf27..75979a9d 100644 --- a/docs/notebooks/open_street_maps_example.ipynb +++ b/docs/notebooks/open_street_maps_example.ipynb @@ -75,16 +75,7 @@ "execution_count": 3, "id": "2a99e267-0ba5-462f-83e4-e8428701597c", "metadata": {}, - "outputs": [ - { - "name": "stderr", - "output_type": "stream", - "text": [ - "/Users/nreinick/dev/miniconda3/envs/test/lib/python3.12/site-packages/osgeo/gdal.py:287: FutureWarning: Neither gdal.UseExceptions() nor gdal.DontUseExceptions() has been explicitly called. In GDAL 4.0, exceptions will be enabled by default.\n", - " warnings.warn(\n" - ] - } - ], + "outputs": [], "source": [ "generate_compass_dataset(g, output_directory=\"golden_co\", add_grade=True) " ] @@ -115,9 +106,6 @@ "name": "stdout", "output_type": "stream", "text": [ - "\n", - "\n", - "\n", "\n" ] }, @@ -125,7 +113,16 @@ "name": "stderr", "output_type": "stream", "text": [ - "uuid file: 100%\|██████████\| 746/746 [00:00<00:00, 3760516.50it/s]it/s]" + "uuid file: 100%\|██████████\| 746/746 [00:00<00:00, 4123301.00it/s]it/s]" + ] + }, + { + "name": "stdout", + "output_type": "stream", + "text": [ + "\n", + "\n", + "\n" ] } ], @@ -246,7 +243,7 @@ }, { "cell_type": "code", - "execution_count": 9, + "execution_count": 10, "id": "961c910c-63ec-4c79-9522-19f24f151215", "metadata": {}, "outputs": [ @@ -255,13 +252,13 @@ "text/plain": [ "{'distance': 4.23738036063421,\n", " 'distance_unit': 'miles',\n", - " 'energy': 0.14906186986507128,\n", - " 'energy_unit': 'gallons_gasoline',\n", " 'time': 8.501927726263151,\n", - " 'time_unit': 'minutes'}" + " 'time_unit': 'minutes',\n", + " 'vehicle': {'energy': 4.23738036063421},\n", + " 'vehicle_info': {'energy_unit': 'gallons_gasoline'}}" ] }, - "execution_count": 9, + "execution_count": 10, "metadata": {}, "output_type": "execute_result" } @@ -272,7 +269,7 @@ }, { "cell_type": "code", - "execution_count": 10, + "execution_count": 11, "id": "e2c6cb39-49aa-4697-9785-fa839dc16a86", "metadata": {}, "outputs": [ @@ -281,13 +278,13 @@ "text/plain": [ "{'distance': 4.013268489898491,\n", " 'distance_unit': 'miles',\n", - " 'energy': 0.14412252939511852,\n", - " 'energy_unit': 'gallons_gasoline',\n", " 'time': 8.978411394482656,\n", - " 'time_unit': 'minutes'}" + " 'time_unit': 'minutes',\n", + " 'vehicle': {'energy': 4.013268489898491},\n", + " 'vehicle_info': {'energy_unit': 'gallons_gasoline'}}" ] }, - "execution_count": 10, + "execution_count": 11, "metadata": {}, "output_type": "execute_result" } @@ -308,17 +305,17 @@ }, { "cell_type": "code", - "execution_count": 11, + "execution_count": 12, "id": "791e814f-333c-43f4-ae84-36b9be517548", "metadata": {}, "outputs": [ { "data": { "text/plain": [ - "dict_keys(['cost', 'request', 'route', 'route_edge_count', 'route_runtime', 'search_executed_time', 'search_runtime', 'total_runtime', 'traversal_summary', 'tree', 'tree_edge_count'])" + "dict_keys(['cost', 'request', 'route', 'route_edge_count', 'route_runtime', 'search_executed_time', 'search_runtime', 'total_runtime', 'traversal_summary', 'tree_edge_count'])" ] }, - "execution_count": 11, + "execution_count": 12, "metadata": {}, "output_type": "execute_result" } @@ -339,7 +336,7 @@ }, { "cell_type": "code", - "execution_count": 12, + "execution_count": 13, "id": "30257102-07ee-4338-9c8c-41676e1995ed", "metadata": {}, "outputs": [], @@ -357,7 +354,7 @@ }, { "cell_type": "code", - "execution_count": 13, + "execution_count": 14, "id": "1ac13aae-32fc-4c10-b888-6373b175655a", "metadata": {}, "outputs": [ @@ -378,7 +375,7 @@ " <style>html, body {width: 100%;height: 100%;margin: 0;padding: 0;}</style>\n", " <style>#map {position:absolute;top:0;bottom:0;right:0;left:0;}</style>\n", " <script src="https://cdn.jsdelivr.net/npm/[email protected]/dist/leaflet.js"></script>\n", - " <script src="https://code.jquery.com/jquery-3.7.1.min.js"></script>\n", + " <script src="https://code.jquery.com/jquery-1.12.4.min.js"></script>\n", " <script src="https://cdn.jsdelivr.net/npm/[email protected]/dist/js/bootstrap.bundle.min.js"></script>\n", " <script src="https://cdnjs.cloudflare.com/ajax/libs/Leaflet.awesome-markers/2.0.2/leaflet.awesome-markers.js"></script>\n", " <link rel="stylesheet" href="https://cdn.jsdelivr.net/npm/[email protected]/dist/leaflet.css"/>\n", @@ -391,7 +388,7 @@ " <meta name="viewport" content="width=device-width,\n", " initial-scale=1.0, maximum-scale=1.0, user-scalable=no" />\n", " <style>\n", - " #map_a2ba13c8f965257bcc81a5b9f1664801 {\n", + " #map_c7345d2ba3ee51ff5955d8b9d1dc20b6 {\n", " position: relative;\n", " width: 100.0%;\n", " height: 100.0%;\n", @@ -405,14 +402,14 @@ "<body>\n", " \n", " \n", - " <div class="folium-map" id="map_a2ba13c8f965257bcc81a5b9f1664801" ></div>\n", + " <div class="folium-map" id="map_c7345d2ba3ee51ff5955d8b9d1dc20b6" ></div>\n", " \n", "</body>\n", "<script>\n", " \n", " \n", - " var map_a2ba13c8f965257bcc81a5b9f1664801 = L.map(\n", - " "map_a2ba13c8f965257bcc81a5b9f1664801",\n", + " var map_c7345d2ba3ee51ff5955d8b9d1dc20b6 = L.map(\n", + " "map_c7345d2ba3ee51ff5955d8b9d1dc20b6",\n", " {\n", " center: [39.7465774, -105.2216276],\n", " crs: L.CRS.EPSG3857,\n", @@ -426,21 +423,19 @@ "\n", " \n", " \n", - " var tile_layer_cd7720600dfcd5d06923fef393e6de53 = L.tileLayer(\n", + " var tile_layer_cbd0b24e6b90b7467b78acb978f031a6 = L.tileLayer(\n", " "https://{s}.tile.openstreetmap.org/{z}/{x}/{y}.png",\n", " {"attribution": "Data by \\u0026copy; \\u003ca target=\\"_blank\\" href=\\"http://openstreetmap.org\\"\\u003eOpenStreetMap\\u003c/a\\u003e, under \\u003ca target=\\"_blank\\" href=\\"http://www.openstreetmap.org/copyright\\"\\u003eODbL\\u003c/a\\u003e.", "detectRetina": false, "maxNativeZoom": 18, "maxZoom": 18, "minZoom": 0, "noWrap": false, "opacity": 1, "subdomains": "abc", "tms": false}\n", - " );\n", + " ).addTo(map_c7345d2ba3ee51ff5955d8b9d1dc20b6);\n", " \n", " \n", - " tile_layer_cd7720600dfcd5d06923fef393e6de53.addTo(map_a2ba13c8f965257bcc81a5b9f1664801);\n", - " \n", - " var poly_line_ad1a043197792f1a02bcbfd31ef088d3 = L.polyline(\n", + " var poly_line_f65a6830adc436b28dd0922803db2edb = L.polyline(\n", " [[39.7273069, -105.1994169], [39.7272989, -105.1994535], [39.7272848, -105.1994868], [39.7272593, -105.1995218], [39.7272272, -105.1995461], [39.7271912, -105.1995576], [39.7271688, -105.199558], [39.7271461, -105.1995532], [39.7271142, -105.1995365], [39.7270868, -105.1995096], [39.7270657, -105.1994741], [39.7270519, -105.1994296], [39.7270483, -105.1993816], [39.727052, -105.1993539], [39.7270571, -105.1993247], [39.7270719, -105.1992916], [39.7270973, -105.1992566], [39.7271293, -105.1992323], [39.7271651, -105.1992207], [39.7271854, -105.1992226], [39.7272113, -105.199227], [39.7273496, -105.1991687], [39.7274185, -105.1991465], [39.7275326, -105.1990928], [39.7276828, -105.1990602], [39.7277455, -105.1990518], [39.7279702, -105.1990232], [39.7281558, -105.1990357], [39.7284384, -105.1990893], [39.7286639, -105.199177], [39.7288222, -105.1992773], [39.7289214, -105.199385], [39.7290292, -105.1995543], [39.7290506, -105.1995879], [39.7292157, -105.1998155], [39.7293094, -105.1998945], [39.7294079, -105.1999604], [39.7295437, -105.200014], [39.7296299, -105.2000721], [39.7297292, -105.2000757], [39.7297135, -105.2007081], [39.7297192, -105.2013106], [39.7298153, -105.2032041], [39.7298256, -105.2033784], [39.7298664, -105.2035708], [39.7298795, -105.203649], [39.7299267, -105.2038558], [39.7301082, -105.2042608], [39.7304911, -105.2048068], [39.7308741, -105.2053111], [39.7310676, -105.2055985], [39.7311933, -105.205824], [39.7312069, -105.2058543], [39.7313075, -105.2061158], [39.731348, -105.206248], [39.731386, -105.2064213], [39.7314237, -105.2067336], [39.731429, -105.2070565], [39.7314226, -105.207224], [39.7314076, -105.2074036], [39.7313611, -105.2077675], [39.7313265, -105.2079124], [39.7312831, -105.20806], [39.7312292, -105.2082086], [39.7311706, -105.208344], [39.7310965, -105.2084777], [39.7310049, -105.208616], [39.7309234, -105.2087235], [39.7308212, -105.2088303], [39.7306227, -105.2089819], [39.7304801, -105.2090736], [39.7303681, -105.2091988], [39.7303327, -105.2092597], [39.7303114, -105.2093181], [39.730299, -105.20939], [39.7302939, -105.2095126], [39.7303105, -105.2096804], [39.7304414, -105.2099267], [39.7305001, -105.2100344], [39.7305871, -105.2101842], [39.7307404, -105.2104481], [39.7309234, -105.2107229], [39.7310934, -105.2109639], [39.731238, -105.211145], [39.7314101, -105.2113464], [39.7315783, -105.2115336], [39.7317541, -105.2117064], [39.7319796, -105.2119089], [39.732359, -105.2122393], [39.7329187, -105.2127184], [39.7334557, -105.213188], [39.7338968, -105.2135676], [39.7342777, -105.2138968], [39.7347139, -105.2142786], [39.7354652, -105.2149301], [39.7356319, -105.2150733], [39.7358844, -105.2152902], [39.7363901, -105.2157275], [39.7369129, -105.2161859], [39.7377243, -105.2168831], [39.7390882, -105.2180682], [39.7397882, -105.2186718], [39.7403043, -105.2191078], [39.7405645, -105.2193336], [39.7406767, -105.2194225], [39.7415803, -105.2202142], [39.7416486, -105.2202727], [39.7417047, -105.2203151], [39.7421502, -105.2206932], [39.7425816, -105.221062], [39.7432687, -105.2215398], [39.7435797, -105.2217631], [39.7436721, -105.2218299], [39.7437366, -105.2218764], [39.7438794, -105.2219759], [39.7439546, -105.2220253], [39.7440082, -105.2220586], [39.7440819, -105.2221021], [39.7441377, -105.2221322], [39.7441808, -105.2221541], [39.7442796, -105.2222043], [39.7443357, -105.2222286], [39.7444062, -105.222259], [39.7444657, -105.2222871], [39.7444962, -105.2223029], [39.7445695, -105.2223431], [39.7446343, -105.2223808], [39.744683, -105.2224104], [39.7447215, -105.2224349], [39.7447658, -105.2224654], [39.744813, -105.222499], [39.7448547, -105.2225315], [39.7448881, -105.2225586], [39.7449328, -105.2225965], [39.7449815, -105.2226391], [39.7451542, -105.2227925], [39.7454784, -105.2230438], [39.7455034, -105.2230623], [39.7455159, -105.2230713], [39.7455403, -105.2230861], [39.7455548, -105.2230941], [39.7455766, -105.2231014], [39.7456018, -105.223106], [39.7456176, -105.2231071], [39.7456337, -105.223107], [39.7456638, -105.2231045], [39.7456814, -105.2231011], [39.7456983, -105.2230945], [39.7457224, -105.2230821], [39.7457338, -105.2230745], [39.7457588, -105.2230543], [39.7457709, -105.223043], [39.7457834, -105.2230284], [39.7458035, -105.2230027], [39.7458905, -105.2228787], [39.7460068, -105.2226624], [39.7461589, -105.2223894], [39.7463617, -105.2220126], [39.7464427, -105.2218672], [39.746482, -105.2217968], [39.7465202, -105.2217322], [39.7465523, -105.2216771], [39.7465774, -105.2216276], [39.7465928, -105.2215823], [39.7466004, -105.2215466], [39.7466042, -105.2215101], [39.7466044, -105.2214695], [39.746601, -105.2214368], [39.7465881, -105.2213862], [39.7465804, -105.2213567], [39.7465774, -105.2213321], [39.7465774, -105.2213048], [39.746581, -105.2212762], [39.7465874, -105.2212518], [39.7465978, -105.2212266], [39.7466125, -105.2212024], [39.7466266, -105.2211858], [39.7466485, -105.2211678], [39.7466708, -105.2211566], [39.746693, -105.2211511], [39.746715, -105.2211506], [39.7467203, -105.2211517], [39.7467798, -105.2211327], [39.7468101, -105.2211204], [39.7468347, -105.2211032], [39.7468673, -105.2210676], [39.7468863, -105.2210389], [39.7469079, -105.2209995], [39.7469752, -105.2208699], [39.7471073, -105.220655], [39.7471371, -105.2206129], [39.7471594, -105.2205845], [39.7472061, -105.2205376], [39.7473376, -105.2202962], [39.747357, -105.2202177], [39.7473682, -105.2201868], [39.7473848, -105.2201519], [39.7475089, -105.2199249], [39.7475861, -105.2197973], [39.7476254, -105.2197401], [39.7476917, -105.2196525], [39.7477724, -105.2195069], [39.7478875, -105.219299], [39.7479219, -105.2192372], [39.7479518, -105.219179], [39.7480957, -105.2189201], [39.7481337, -105.2188515], [39.7482567, -105.2186294], [39.7484851, -105.2182271], [39.7485167, -105.218164], [39.7485605, -105.2180881], [39.7490516, -105.217166], [39.7490955, -105.2170866], [39.7491263, -105.2170387], [39.7493736, -105.2165801], [39.7496841, -105.2160128], [39.7497482, -105.2160733], [39.7504593, -105.216753], [39.7505084, -105.2168007], [39.7505733, -105.2168567], [39.7509525, -105.2172063], [39.7512593, -105.2174884], [39.7513304, -105.2175506], [39.7513869, -105.217604], [39.7517458, -105.2179367], [39.7520945, -105.21826], [39.7521462, -105.2183047], [39.7522126, -105.218369], [39.7523012, -105.2184523], [39.7525052, -105.218642], [39.7527145, -105.2188366], [39.7528959, -105.2189993], [39.7529601, -105.2190649], [39.7530308, -105.219128], [39.7537033, -105.2197718], [39.753764, -105.2198302], [39.7538299, -105.2198915], [39.7542346, -105.2202638], [39.7545279, -105.2205433], [39.7545847, -105.2206031], [39.7546623, -105.2206728], [39.7553326, -105.2212752], [39.755407, -105.221342], [39.7554908, -105.2214211], [39.7561483, -105.2220415], [39.7562254, -105.2221143], [39.7563006, -105.2221853], [39.7566193, -105.222483], [39.7566315, -105.2224957], [39.7569475, -105.2227913], [39.7570561, -105.222895], [39.7574792, -105.2232693], [39.7575579, -105.2233409], [39.7576236, -105.2234019], [39.7578565, -105.2236199], [39.7579703, -105.2237258], [39.7580027, -105.2237561], [39.7583001, -105.2240544], [39.7583591, -105.2241084], [39.7587249, -105.2244446], [39.7588057, -105.22452], [39.7591609, -105.2248516], [39.7595089, -105.2251741], [39.7596887, -105.2253408], [39.7597554, -105.2254024], [39.7598802, -105.2255173], [39.7598939, -105.2255305], [39.7599658, -105.2255988], [39.7600186, -105.2256498], [39.7604008, -105.2259941], [39.7605066, -105.2260931], [39.7605924, -105.226174], [39.760667, -105.2262427], [39.7607946, -105.2263631], [39.7608753, -105.2264373], [39.7615829, -105.2271044], [39.7616266, -105.2271481], [39.7616788, -105.2272008], [39.762203, -105.2276943], [39.7623411, -105.2278299], [39.7630405, -105.2284565], [39.7632824, -105.2286685], [39.7637358, -105.2290679], [39.7640137, -105.229309], [39.7640954, -105.229381], [39.7641743, -105.229462], [39.76436, -105.2296495], [39.7645376, -105.2298658], [39.764824, -105.2301918], [39.7652051, -105.2307044], [39.7654711, -105.2310802], [39.7654801, -105.2310934], [39.7655308, -105.231168], [39.7655792, -105.231232], [39.7655918, -105.2312486], [39.7657719, -105.2315035], [39.7658973, -105.2316751], [39.7659962, -105.2318039], [39.7661018, -105.231924], [39.7662469, -105.2320614], [39.7664515, -105.2322588], [39.7671442, -105.2329283], [39.7673559, -105.2331244], [39.7674414, -105.2332053], [39.7675186, -105.2332845], [39.7675756, -105.2333522], [39.7676291, -105.2334243], [39.7676969, -105.2335369], [39.7677304, -105.233602], [39.767761, -105.2336649], [39.7677908, -105.233739], [39.7678172, -105.233812], [39.7678551, -105.2339422], [39.7678736, -105.2340294], [39.7678904, -105.2341257], [39.7679032, -105.2342329], [39.7679095, -105.2343359], [39.7679068, -105.2344027], [39.7679029, -105.234499], [39.7678875, -105.2346626], [39.7678651, -105.2347791], [39.7678384, -105.2348898], [39.7678006, -105.2350283], [39.7677517, -105.2352031], [39.7677119, -105.2353468]],\n", " {"bubblingMouseEvents": true, "color": "blue", "dashArray": null, "dashOffset": null, "fill": false, "fillColor": "blue", "fillOpacity": 0.2, "fillRule": "evenodd", "lineCap": "round", "lineJoin": "round", "noClip": false, "opacity": 0.8, "smoothFactor": 1.0, "stroke": true, "weight": 10}\n", - " ).addTo(map_a2ba13c8f965257bcc81a5b9f1664801);\n", + " ).addTo(map_c7345d2ba3ee51ff5955d8b9d1dc20b6);\n", " \n", " \n", - " poly_line_ad1a043197792f1a02bcbfd31ef088d3.bindTooltip(\n", + " poly_line_f65a6830adc436b28dd0922803db2edb.bindTooltip(\n", " `<div>\n", " Shortest Time\n", " </div>`,\n", @@ -448,19 +443,19 @@ " );\n", " \n", " \n", - " var marker_9aec00c6a27264acf91afd3f0996a66d = L.marker(\n", + " var marker_c509908f7e3039cb10134666c7e7930d = L.marker(\n", " [39.7273069, -105.1994169],\n", " {}\n", - " ).addTo(map_a2ba13c8f965257bcc81a5b9f1664801);\n", + " ).addTo(map_c7345d2ba3ee51ff5955d8b9d1dc20b6);\n", " \n", " \n", - " var icon_ddb004cd6627984db241463133fc8206 = L.AwesomeMarkers.icon(\n", + " var icon_fce1a97d514cb3efe68f0ab61c47881a = L.AwesomeMarkers.icon(\n", " {"extraClasses": "fa-rotate-0", "icon": "circle", "iconColor": "white", "markerColor": "green", "prefix": "fa"}\n", " );\n", - " marker_9aec00c6a27264acf91afd3f0996a66d.setIcon(icon_ddb004cd6627984db241463133fc8206);\n", + " marker_c509908f7e3039cb10134666c7e7930d.setIcon(icon_fce1a97d514cb3efe68f0ab61c47881a);\n", " \n", " \n", - " marker_9aec00c6a27264acf91afd3f0996a66d.bindTooltip(\n", + " marker_c509908f7e3039cb10134666c7e7930d.bindTooltip(\n", " `<div>\n", " Origin\n", " </div>`,\n", @@ -468,19 +463,19 @@ " );\n", " \n", " \n", - " var marker_2400252a991122b76a0f550d44f2471a = L.marker(\n", + " var marker_53fa08b69a274c2485a6a5bfe29337f8 = L.marker(\n", " [39.7677119, -105.2353468],\n", " {}\n", - " ).addTo(map_a2ba13c8f965257bcc81a5b9f1664801);\n", + " ).addTo(map_c7345d2ba3ee51ff5955d8b9d1dc20b6);\n", " \n", " \n", - " var icon_5275c70cca7a2265a72fb4c4f9fdd51d = L.AwesomeMarkers.icon(\n", + " var icon_a80a162478d0b051a78633e6cf82d5c2 = L.AwesomeMarkers.icon(\n", " {"extraClasses": "fa-rotate-0", "icon": "circle", "iconColor": "white", "markerColor": "red", "prefix": "fa"}\n", " );\n", - " marker_2400252a991122b76a0f550d44f2471a.setIcon(icon_5275c70cca7a2265a72fb4c4f9fdd51d);\n", + " marker_53fa08b69a274c2485a6a5bfe29337f8.setIcon(icon_a80a162478d0b051a78633e6cf82d5c2);\n", " \n", " \n", - " marker_2400252a991122b76a0f550d44f2471a.bindTooltip(\n", + " marker_53fa08b69a274c2485a6a5bfe29337f8.bindTooltip(\n", " `<div>\n", " Destination\n", " </div>`,\n", @@ -488,13 +483,13 @@ " );\n", " \n", " \n", - " var poly_line_f8164f878668d4d431bf9490b1c0c98b = L.polyline(\n", + " var poly_line_f7dfde4d70684478e2ba63e72b94ca5b = L.polyline(\n", " [[39.7273069, -105.1994169], [39.7272989, -105.1994535], [39.7272848, -105.1994868], [39.7272593, -105.1995218], [39.7272272, -105.1995461], [39.7271912, -105.1995576], [39.7271688, -105.199558], [39.7271461, -105.1995532], [39.7271142, -105.1995365], [39.7270868, -105.1995096], [39.7270657, -105.1994741], [39.7270519, -105.1994296], [39.7270483, -105.1993816], [39.727052, -105.1993539], [39.7270571, -105.1993247], [39.7270719, -105.1992916], [39.7270973, -105.1992566], [39.7271293, -105.1992323], [39.7271651, -105.1992207], [39.7271854, -105.1992226], [39.7272113, -105.199227], [39.7273496, -105.1991687], [39.7274185, -105.1991465], [39.7275326, -105.1990928], [39.7276828, -105.1990602], [39.7277455, -105.1990518], [39.7279702, -105.1990232], [39.7281558, -105.1990357], [39.7284384, -105.1990893], [39.7286639, -105.199177], [39.7288222, -105.1992773], [39.7289214, -105.199385], [39.7290292, -105.1995543], [39.7290506, -105.1995879], [39.7292157, -105.1998155], [39.7293094, -105.1998945], [39.7294079, -105.1999604], [39.7295437, -105.200014], [39.7296299, -105.2000721], [39.7296469, -105.1996789], [39.7296738, -105.1987284], [39.7296906, -105.1986794], [39.7297244, -105.1985856], [39.7297272, -105.1984193], [39.7299723, -105.1984198], [39.7302117, -105.1984227], [39.7303235, -105.1984285], [39.7304434, -105.198443], [39.7305651, -105.1984732], [39.7307182, -105.1985167], [39.7308487, -105.1985623], [39.7309909, -105.1986311], [39.731127, -105.1987144], [39.7312292, -105.1987897], [39.7313292, -105.1988731], [39.7314292, -105.1989611], [39.731524, -105.1990554], [39.7316089, -105.1991466], [39.7317085, -105.1992639], [39.7318575, -105.1994488], [39.7319522, -105.1995697], [39.7320423, -105.1996716], [39.7321372, -105.1997578], [39.7322615, -105.1998506], [39.7323697, -105.1999163], [39.7324712, -105.1999637], [39.7325898, -105.2000081], [39.732689, -105.2000309], [39.7327987, -105.2000627], [39.732945, -105.2001049], [39.7330635, -105.2001374], [39.7331838, -105.2001914], [39.7332899, -105.2002528], [39.7334312, -105.200363], [39.7335257, -105.2004544], [39.7335523, -105.2004835], [39.7336132, -105.2005503], [39.7336902, -105.2006555], [39.733767, -105.2007557], [39.7338411, -105.2008491], [39.7339359, -105.2009412], [39.7340357, -105.2010154], [39.7340742, -105.201042], [39.7341117, -105.2010678], [39.7341542, -105.2010901], [39.734211, -105.2011198], [39.7343173, -105.2011712], [39.7344056, -105.2012021], [39.7345704, -105.2012376], [39.7347726, -105.2012753], [39.7351329, -105.2013311], [39.7352203, -105.2013441], [39.7353243, -105.201352], [39.7354415, -105.2013441], [39.7355491, -105.2013268], [39.7357022, -105.2012901], [39.7358395, -105.201259], [39.7362429, -105.2011675], [39.7367157, -105.2010639], [39.7367857, -105.2010539], [39.7368531, -105.2010442], [39.7369645, -105.2010356], [39.7370755, -105.201044], [39.7371895, -105.2010632], [39.7373179, -105.201099], [39.737714, -105.2012295], [39.7379567, -105.2013009], [39.7384451, -105.2014404], [39.7386872, -105.2015167], [39.7395717, -105.2017905], [39.7397153, -105.2018369], [39.7398269, -105.2018673], [39.7399356, -105.2018843], [39.7400618, -105.2018961], [39.740184, -105.2018932], [39.7403133, -105.2018823], [39.7404288, -105.2018645], [39.7405518, -105.2018387], [39.740665, -105.2018073], [39.7407597, -105.2017713], [39.7408728, -105.2017196], [39.740992, -105.2016551], [39.7411268, -105.2015625], [39.7412599, -105.2014489], [39.7413718, -105.2013453], [39.7414758, -105.2012326], [39.741608, -105.2010691], [39.7416901, -105.200964], [39.7417903, -105.2008396], [39.7417994, -105.200826], [39.7419078, -105.2006985], [39.7419027, -105.2006189], [39.7419122, -105.200549], [39.7419496, -105.2004812], [39.7419894, -105.20044], [39.742047, -105.2004033], [39.7420816, -105.2004], [39.7421282, -105.2004131], [39.7421839, -105.200448], [39.7422234, -105.2004924], [39.7422448, -105.2005578], [39.7422566, -105.2006096], [39.7422556, -105.2006625], [39.7422421, -105.2007231], [39.7422712, -105.2008198], [39.7423227, -105.200901], [39.7428069, -105.2016549], [39.7429048, -105.2018114], [39.7429131, -105.2018245], [39.7431503, -105.2022035], [39.7434374, -105.2026477], [39.7435807, -105.2028659], [39.7437705, -105.2031843], [39.7441255, -105.2037193], [39.7444536, -105.2042289], [39.7445369, -105.204369], [39.7448087, -105.2048085], [39.7450389, -105.205197], [39.745164, -105.2054397], [39.7452617, -105.205642], [39.745359, -105.2058753], [39.7454413, -105.2060693], [39.7454698, -105.2061365], [39.7455586, -105.2063498], [39.7456394, -105.2066197], [39.7457343, -105.2068984], [39.7458096, -105.2071619], [39.7458432, -105.2072893], [39.7459057, -105.2075162], [39.7459547, -105.2077463], [39.7460104, -105.2080066], [39.7460312, -105.2081149], [39.7460508, -105.2082709], [39.7460836, -105.2084787], [39.7461092, -105.2086125], [39.7461165, -105.2087078], [39.7461128, -105.2088019], [39.7461078, -105.2089177], [39.7461163, -105.208995], [39.7461377, -105.2090672], [39.7461588, -105.2091147], [39.7462509, -105.2092159], [39.746281, -105.2092382], [39.7462978, -105.209256], [39.7463242, -105.209292], [39.7463458, -105.2093473], [39.746361, -105.2094251], [39.7463921, -105.2095768], [39.7464353, -105.2097133], [39.7465001, -105.2098285], [39.7467097, -105.21004], [39.7467814, -105.2101094], [39.7473357, -105.2106047], [39.7474554, -105.2107116], [39.7475615, -105.2108064], [39.7475958, -105.2108371], [39.747653, -105.2108907], [39.7481502, -105.2113567], [39.7483361, -105.2115309], [39.7483726, -105.2115651], [39.7484183, -105.2116079], [39.7484718, -105.2116584], [39.7487212, -105.2118951], [39.749238, -105.2123721], [39.7492814, -105.2124135], [39.7500059, -105.2130759], [39.7500454, -105.2131129], [39.7503812, -105.2134256], [39.7504627, -105.2135015], [39.7508219, -105.2138362], [39.7508792, -105.2138895], [39.7509348, -105.2139426], [39.7511093, -105.2141093], [39.7516271, -105.2146038], [39.7516716, -105.2146463], [39.751739, -105.2147085], [39.7524829, -105.2153953], [39.7525012, -105.2154123], [39.7532517, -105.2161101], [39.7533052, -105.2161598], [39.7540716, -105.2168674], [39.7541308, -105.2169221], [39.7544795, -105.2172508], [39.7546281, -105.2174522], [39.7546745, -105.2175037], [39.7548394, -105.2176869], [39.7549112, -105.2177871], [39.7548521, -105.2179433], [39.7546476, -105.2183047], [39.7544379, -105.2186805], [39.7544293, -105.218688], [39.7543723, -105.2187394], [39.7544376, -105.2188002], [39.7544872, -105.2188464], [39.7547934, -105.2191313], [39.7551071, -105.2194077], [39.7551368, -105.2194339], [39.755202, -105.2194913], [39.7552559, -105.2195411], [39.755615, -105.2198733], [39.755696, -105.2199479], [39.7559351, -105.2201693], [39.7559559, -105.220191], [39.7560109, -105.2202486], [39.7560703, -105.2203182], [39.756257, -105.2205368], [39.7562813, -105.2205598], [39.7564057, -105.2206774], [39.7565461, -105.2208058], [39.7567417, -105.2209442], [39.7566986, -105.2211001], [39.756638, -105.2213112], [39.7565732, -105.2214643], [39.7565209, -105.2215678], [39.7564595, -105.2216814], [39.7562914, -105.2219923], [39.7562254, -105.2221143], [39.7563006, -105.2221853], [39.7566193, -105.222483], [39.7566315, -105.2224957], [39.7569475, -105.2227913], [39.7570561, -105.222895], [39.7574792, -105.2232693], [39.7575579, -105.2233409], [39.7576236, -105.2234019], [39.7578565, -105.2236199], [39.7579703, -105.2237258], [39.7580027, -105.2237561], [39.7583001, -105.2240544], [39.7583591, -105.2241084], [39.7587249, -105.2244446], [39.7588057, -105.22452], [39.7591609, -105.2248516], [39.7595089, -105.2251741], [39.7596887, -105.2253408], [39.7597554, -105.2254024], [39.7598802, -105.2255173], [39.7598939, -105.2255305], [39.7599658, -105.2255988], [39.7600186, -105.2256498], [39.7604008, -105.2259941], [39.7605066, -105.2260931], [39.7605924, -105.226174], [39.760667, -105.2262427], [39.7607946, -105.2263631], [39.7608753, -105.2264373], [39.7615829, -105.2271044], [39.7616266, -105.2271481], [39.7616788, -105.2272008], [39.762203, -105.2276943], [39.7623411, -105.2278299], [39.7630405, -105.2284565], [39.7632824, -105.2286685], [39.7637358, -105.2290679], [39.7640137, -105.229309], [39.7640954, -105.229381], [39.7641743, -105.229462], [39.76436, -105.2296495], [39.7645376, -105.2298658], [39.764824, -105.2301918], [39.7652051, -105.2307044], [39.7654711, -105.2310802], [39.7654801, -105.2310934], [39.7655308, -105.231168], [39.7655792, -105.231232], [39.7655918, -105.2312486], [39.7657719, -105.2315035], [39.7658973, -105.2316751], [39.7659962, -105.2318039], [39.7661018, -105.231924], [39.7662469, -105.2320614], [39.7664515, -105.2322588], [39.7671442, -105.2329283], [39.7673559, -105.2331244], [39.7674414, -105.2332053], [39.7675186, -105.2332845], [39.7675756, -105.2333522], [39.7676291, -105.2334243], [39.7676969, -105.2335369], [39.7677304, -105.233602], [39.767761, -105.2336649], [39.7677908, -105.233739], [39.7678172, -105.233812], [39.7678551, -105.2339422], [39.7678736, -105.2340294], [39.7678904, -105.2341257], [39.7679032, -105.2342329], [39.7679095, -105.2343359], [39.7679068, -105.2344027], [39.7679029, -105.234499], [39.7678875, -105.2346626], [39.7678651, -105.2347791], [39.7678384, -105.2348898], [39.7678006, -105.2350283], [39.7677517, -105.2352031], [39.7677119, -105.2353468]],\n", " {"bubblingMouseEvents": true, "color": "green", "dashArray": null, "dashOffset": null, "fill": false, "fillColor": "green", "fillOpacity": 0.2, "fillRule": "evenodd", "lineCap": "round", "lineJoin": "round", "noClip": false, "opacity": 0.8, "smoothFactor": 1.0, "stroke": true, "weight": 10}\n", - " ).addTo(map_a2ba13c8f965257bcc81a5b9f1664801);\n", + " ).addTo(map_c7345d2ba3ee51ff5955d8b9d1dc20b6);\n", " \n", " \n", - " poly_line_f8164f878668d4d431bf9490b1c0c98b.bindTooltip(\n", + " poly_line_f7dfde4d70684478e2ba63e72b94ca5b.bindTooltip(\n", " `<div>\n", " Least Energy\n", " </div>`,\n", @@ -502,19 +497,19 @@ " );\n", " \n", " \n", - " var marker_e7903ecfe2687cdf35e58bb6187e05c3 = L.marker(\n", + " var marker_a8c28fbe383145e2e368ea5d04e0e725 = L.marker(\n", " [39.7273069, -105.1994169],\n", " {}\n", - " ).addTo(map_a2ba13c8f965257bcc81a5b9f1664801);\n", + " ).addTo(map_c7345d2ba3ee51ff5955d8b9d1dc20b6);\n", " \n", " \n", - " var icon_d00a95d2fc383df1ee01c11c1d9015a2 = L.AwesomeMarkers.icon(\n", + " var icon_f4b068a33dbb5d6e0551beafff0c091c = L.AwesomeMarkers.icon(\n", " {"extraClasses": "fa-rotate-0", "icon": "circle", "iconColor": "white", "markerColor": "green", "prefix": "fa"}\n", " );\n", - " marker_e7903ecfe2687cdf35e58bb6187e05c3.setIcon(icon_d00a95d2fc383df1ee01c11c1d9015a2);\n", + " marker_a8c28fbe383145e2e368ea5d04e0e725.setIcon(icon_f4b068a33dbb5d6e0551beafff0c091c);\n", " \n", " \n", - " marker_e7903ecfe2687cdf35e58bb6187e05c3.bindTooltip(\n", + " marker_a8c28fbe383145e2e368ea5d04e0e725.bindTooltip(\n", " `<div>\n", " Origin\n", " </div>`,\n", @@ -522,19 +517,19 @@ " );\n", " \n", " \n", - " var marker_a8605646854f3779e76608978a213f99 = L.marker(\n", + " var marker_896695b507afb28930fcb5a332f47cb2 = L.marker(\n", " [39.7677119, -105.2353468],\n", " {}\n", - " ).addTo(map_a2ba13c8f965257bcc81a5b9f1664801);\n", + " ).addTo(map_c7345d2ba3ee51ff5955d8b9d1dc20b6);\n", " \n", " \n", - " var icon_7e6ec847709d3163e63c9f64c56b3953 = L.AwesomeMarkers.icon(\n", + " var icon_3f03d1531d14f55d0f27e168b4b4da80 = L.AwesomeMarkers.icon(\n", " {"extraClasses": "fa-rotate-0", "icon": "circle", "iconColor": "white", "markerColor": "red", "prefix": "fa"}\n", " );\n", - " marker_a8605646854f3779e76608978a213f99.setIcon(icon_7e6ec847709d3163e63c9f64c56b3953);\n", + " marker_896695b507afb28930fcb5a332f47cb2.setIcon(icon_3f03d1531d14f55d0f27e168b4b4da80);\n", " \n", " \n", - " marker_a8605646854f3779e76608978a213f99.bindTooltip(\n", + " marker_896695b507afb28930fcb5a332f47cb2.bindTooltip(\n", " `<div>\n", " Destination\n", " </div>`,\n", @@ -545,10 +540,10 @@ "</html>\" style=\"position:absolute;width:100%;height:100%;left:0;top:0;border:none !important;\" allowfullscreen webkitallowfullscreen mozallowfullscreen></iframe></div></div>" ], "text/plain": [ - "<folium.folium.Map at 0x1a3fa3560>" + "<folium.folium.Map at 0x12c5a1dd0>" ] }, - "execution_count": 13, + "execution_count": 14, "metadata": {}, "output_type": "execute_result" } @@ -569,7 +564,7 @@ }, { "cell_type": "code", - "execution_count": 14, + "execution_count": 16, "id": "aa1e6b11", "metadata": {}, "outputs": [ @@ -590,7 +585,7 @@ " <style>html, body {width: 100%;height: 100%;margin: 0;padding: 0;}</style>\n", " <style>#map {position:absolute;top:0;bottom:0;right:0;left:0;}</style>\n", " <script src="https://cdn.jsdelivr.net/npm/[email protected]/dist/leaflet.js"></script>\n", - " <script src="https://code.jquery.com/jquery-3.7.1.min.js"></script>\n", + " <script src="https://code.jquery.com/jquery-1.12.4.min.js"></script>\n", " <script src="https://cdn.jsdelivr.net/npm/[email protected]/dist/js/bootstrap.bundle.min.js"></script>\n", " <script src="https://cdnjs.cloudflare.com/ajax/libs/Leaflet.awesome-markers/2.0.2/leaflet.awesome-markers.js"></script>\n", " <link rel="stylesheet" href="https://cdn.jsdelivr.net/npm/[email protected]/dist/leaflet.css"/>\n", @@ -603,7 +598,7 @@ " <meta name="viewport" content="width=device-width,\n", " initial-scale=1.0, maximum-scale=1.0, user-scalable=no" />\n", " <style>\n", - " #map_0a69fb0ffb4fdf07d938c59697d2d5b7 {\n", + " #map_08d0a92c80cee353042d813e9d897e9a {\n", " position: relative;\n", " width: 100.0%;\n", " height: 100.0%;\n", @@ -617,14 +612,14 @@ "<body>\n", " \n", " \n", - " <div class="folium-map" id="map_0a69fb0ffb4fdf07d938c59697d2d5b7" ></div>\n", + " <div class="folium-map" id="map_08d0a92c80cee353042d813e9d897e9a" ></div>\n", " \n", "</body>\n", "<script>\n", " \n", " \n", - " var map_0a69fb0ffb4fdf07d938c59697d2d5b7 = L.map(\n", - " "map_0a69fb0ffb4fdf07d938c59697d2d5b7",\n", + " var map_08d0a92c80cee353042d813e9d897e9a = L.map(\n", + " "map_08d0a92c80cee353042d813e9d897e9a",\n", " {\n", " center: [39.7465774, -105.2216276],\n", " crs: L.CRS.EPSG3857,\n", @@ -638,41 +633,39 @@ "\n", " \n", " \n", - " var tile_layer_bf7e6951595d6bd28e928637e222904d = L.tileLayer(\n", + " var tile_layer_3d8120759aee037fb6485ddcaae7eb08 = L.tileLayer(\n", " "https://{s}.tile.openstreetmap.org/{z}/{x}/{y}.png",\n", " {"attribution": "Data by \\u0026copy; \\u003ca target=\\"_blank\\" href=\\"http://openstreetmap.org\\"\\u003eOpenStreetMap\\u003c/a\\u003e, under \\u003ca target=\\"_blank\\" href=\\"http://www.openstreetmap.org/copyright\\"\\u003eODbL\\u003c/a\\u003e.", "detectRetina": false, "maxNativeZoom": 18, "maxZoom": 18, "minZoom": 0, "noWrap": false, "opacity": 1, "subdomains": "abc", "tms": false}\n", - " );\n", + " ).addTo(map_08d0a92c80cee353042d813e9d897e9a);\n", " \n", " \n", - " tile_layer_bf7e6951595d6bd28e928637e222904d.addTo(map_0a69fb0ffb4fdf07d938c59697d2d5b7);\n", - " \n", - " var poly_line_b2f52dbae2df737185eacbb7f7f1814f = L.polyline(\n", + " var poly_line_fce8b3c76b03f40e0dbe8d7422abf0e5 = L.polyline(\n", " [[39.7273069, -105.1994169], [39.7272989, -105.1994535], [39.7272848, -105.1994868], [39.7272593, -105.1995218], [39.7272272, -105.1995461], [39.7271912, -105.1995576], [39.7271688, -105.199558], [39.7271461, -105.1995532], [39.7271142, -105.1995365], [39.7270868, -105.1995096], [39.7270657, -105.1994741], [39.7270519, -105.1994296], [39.7270483, -105.1993816], [39.727052, -105.1993539], [39.7270571, -105.1993247], [39.7270719, -105.1992916], [39.7270973, -105.1992566], [39.7271293, -105.1992323], [39.7271651, -105.1992207], [39.7271854, -105.1992226], [39.7272113, -105.199227], [39.7273496, -105.1991687], [39.7274185, -105.1991465], [39.7275326, -105.1990928], [39.7276828, -105.1990602], [39.7277455, -105.1990518], [39.7279702, -105.1990232], [39.7281558, -105.1990357], [39.7284384, -105.1990893], [39.7286639, -105.199177], [39.7288222, -105.1992773], [39.7289214, -105.199385], [39.7290292, -105.1995543], [39.7290506, -105.1995879], [39.7292157, -105.1998155], [39.7293094, -105.1998945], [39.7294079, -105.1999604], [39.7295437, -105.200014], [39.7296299, -105.2000721], [39.7297292, -105.2000757], [39.7297135, -105.2007081], [39.7297192, -105.2013106], [39.7298153, -105.2032041], [39.7298256, -105.2033784], [39.7298664, -105.2035708], [39.7298795, -105.203649], [39.7299267, -105.2038558], [39.7301082, -105.2042608], [39.7304911, -105.2048068], [39.7308741, -105.2053111], [39.7310676, -105.2055985], [39.7311933, -105.205824], [39.7312069, -105.2058543], [39.7313075, -105.2061158], [39.731348, -105.206248], [39.731386, -105.2064213], [39.7314237, -105.2067336], [39.731429, -105.2070565], [39.7314226, -105.207224], [39.7314076, -105.2074036], [39.7313611, -105.2077675], [39.7313265, -105.2079124], [39.7312831, -105.20806], [39.7312292, -105.2082086], [39.7311706, -105.208344], [39.7310965, -105.2084777], [39.7310049, -105.208616], [39.7309234, -105.2087235], [39.7308212, -105.2088303], [39.7306227, -105.2089819], [39.7304801, -105.2090736], [39.7303681, -105.2091988], [39.7303327, -105.2092597], [39.7303114, -105.2093181], [39.730299, -105.20939], [39.7302939, -105.2095126], [39.7303105, -105.2096804], [39.7304414, -105.2099267], [39.7305001, -105.2100344], [39.7305871, -105.2101842], [39.7307404, -105.2104481], [39.7309234, -105.2107229], [39.7310934, -105.2109639], [39.731238, -105.211145], [39.7314101, -105.2113464], [39.7315783, -105.2115336], [39.7317541, -105.2117064], [39.7319796, -105.2119089], [39.732359, -105.2122393], [39.7329187, -105.2127184], [39.7334557, -105.213188], [39.7338968, -105.2135676], [39.7342777, -105.2138968], [39.7347139, -105.2142786], [39.7354652, -105.2149301], [39.7356319, -105.2150733], [39.7358844, -105.2152902], [39.7363901, -105.2157275], [39.7369129, -105.2161859], [39.7377243, -105.2168831], [39.7390882, -105.2180682], [39.7397882, -105.2186718], [39.7403043, -105.2191078], [39.7405645, -105.2193336], [39.7406767, -105.2194225], [39.7415803, -105.2202142], [39.7416486, -105.2202727], [39.7417047, -105.2203151], [39.7421502, -105.2206932], [39.7425816, -105.221062], [39.7432687, -105.2215398], [39.7435797, -105.2217631], [39.7436721, -105.2218299], [39.7437366, -105.2218764], [39.7438794, -105.2219759], [39.7439546, -105.2220253], [39.7440082, -105.2220586], [39.7440819, -105.2221021], [39.7441377, -105.2221322], [39.7441808, -105.2221541], [39.7442796, -105.2222043], [39.7443357, -105.2222286], [39.7444062, -105.222259], [39.7444657, -105.2222871], [39.7444962, -105.2223029], [39.7445695, -105.2223431], [39.7446343, -105.2223808], [39.744683, -105.2224104], [39.7447215, -105.2224349], [39.7447658, -105.2224654], [39.744813, -105.222499], [39.7448547, -105.2225315], [39.7448881, -105.2225586], [39.7449328, -105.2225965], [39.7449815, -105.2226391], [39.7451542, -105.2227925], [39.7454784, -105.2230438], [39.7455034, -105.2230623], [39.7455159, -105.2230713], [39.7455403, -105.2230861], [39.7455548, -105.2230941], [39.7455766, -105.2231014], [39.7456018, -105.223106], [39.7456176, -105.2231071], [39.7456337, -105.223107], [39.7456638, -105.2231045], [39.7456814, -105.2231011], [39.7456983, -105.2230945], [39.7457224, -105.2230821], [39.7457338, -105.2230745], [39.7457588, -105.2230543], [39.7457709, -105.223043], [39.7457834, -105.2230284], [39.7458035, -105.2230027], [39.7458905, -105.2228787], [39.7460068, -105.2226624], [39.7461589, -105.2223894], [39.7463617, -105.2220126], [39.7464427, -105.2218672], [39.746482, -105.2217968], [39.7465202, -105.2217322], [39.7465523, -105.2216771], [39.7465774, -105.2216276], [39.7465928, -105.2215823], [39.7466004, -105.2215466], [39.7466042, -105.2215101], [39.7466044, -105.2214695], [39.746601, -105.2214368], [39.7465881, -105.2213862], [39.7465804, -105.2213567], [39.7465774, -105.2213321], [39.7465774, -105.2213048], [39.746581, -105.2212762], [39.7465874, -105.2212518], [39.7465978, -105.2212266], [39.7466125, -105.2212024], [39.7466266, -105.2211858], [39.7466485, -105.2211678], [39.7466708, -105.2211566], [39.746693, -105.2211511], [39.746715, -105.2211506], [39.7467203, -105.2211517], [39.7467798, -105.2211327], [39.7468101, -105.2211204], [39.7468347, -105.2211032], [39.7468673, -105.2210676], [39.7468863, -105.2210389], [39.7469079, -105.2209995], [39.7469752, -105.2208699], [39.7471073, -105.220655], [39.7471371, -105.2206129], [39.7471594, -105.2205845], [39.7472061, -105.2205376], [39.7473376, -105.2202962], [39.747357, -105.2202177], [39.7473682, -105.2201868], [39.7473848, -105.2201519], [39.7475089, -105.2199249], [39.7475861, -105.2197973], [39.7476254, -105.2197401], [39.7476917, -105.2196525], [39.7477724, -105.2195069], [39.7478875, -105.219299], [39.7479219, -105.2192372], [39.7479518, -105.219179], [39.7480957, -105.2189201], [39.7481337, -105.2188515], [39.7482567, -105.2186294], [39.7484851, -105.2182271], [39.7485167, -105.218164], [39.7485605, -105.2180881], [39.7490516, -105.217166], [39.7490955, -105.2170866], [39.7491263, -105.2170387], [39.7493736, -105.2165801], [39.7496841, -105.2160128], [39.7497482, -105.2160733], [39.7504593, -105.216753], [39.7505084, -105.2168007], [39.7505733, -105.2168567], [39.7509525, -105.2172063], [39.7512593, -105.2174884], [39.7513304, -105.2175506], [39.7513869, -105.217604], [39.7517458, -105.2179367], [39.7520945, -105.21826], [39.7521462, -105.2183047], [39.7522126, -105.218369], [39.7523012, -105.2184523], [39.7525052, -105.218642], [39.7527145, -105.2188366], [39.7528959, -105.2189993], [39.7529601, -105.2190649], [39.7530308, -105.219128], [39.7537033, -105.2197718], [39.753764, -105.2198302], [39.7538299, -105.2198915], [39.7542346, -105.2202638], [39.7545279, -105.2205433], [39.7545847, -105.2206031], [39.7546623, -105.2206728], [39.7553326, -105.2212752], [39.755407, -105.221342], [39.7554908, -105.2214211], [39.7561483, -105.2220415], [39.7562254, -105.2221143], [39.7563006, -105.2221853], [39.7566193, -105.222483], [39.7566315, -105.2224957], [39.7569475, -105.2227913], [39.7570561, -105.222895], [39.7574792, -105.2232693], [39.7575579, -105.2233409], [39.7576236, -105.2234019], [39.7578565, -105.2236199], [39.7579703, -105.2237258], [39.7580027, -105.2237561], [39.7583001, -105.2240544], [39.7583591, -105.2241084], [39.7587249, -105.2244446], [39.7588057, -105.22452], [39.7591609, -105.2248516], [39.7595089, -105.2251741], [39.7596887, -105.2253408], [39.7597554, -105.2254024], [39.7598802, -105.2255173], [39.7598939, -105.2255305], [39.7599658, -105.2255988], [39.7600186, -105.2256498], [39.7604008, -105.2259941], [39.7605066, -105.2260931], [39.7605924, -105.226174], [39.760667, -105.2262427], [39.7607946, -105.2263631], [39.7608753, -105.2264373], [39.7615829, -105.2271044], [39.7616266, -105.2271481], [39.7616788, -105.2272008], [39.762203, -105.2276943], [39.7623411, -105.2278299], [39.7630405, -105.2284565], [39.7632824, -105.2286685], [39.7637358, -105.2290679], [39.7640137, -105.229309], [39.7640954, -105.229381], [39.7641743, -105.229462], [39.76436, -105.2296495], [39.7645376, -105.2298658], [39.764824, -105.2301918], [39.7652051, -105.2307044], [39.7654711, -105.2310802], [39.7654801, -105.2310934], [39.7655308, -105.231168], [39.7655792, -105.231232], [39.7655918, -105.2312486], [39.7657719, -105.2315035], [39.7658973, -105.2316751], [39.7659962, -105.2318039], [39.7661018, -105.231924], [39.7662469, -105.2320614], [39.7664515, -105.2322588], [39.7671442, -105.2329283], [39.7673559, -105.2331244], [39.7674414, -105.2332053], [39.7675186, -105.2332845], [39.7675756, -105.2333522], [39.7676291, -105.2334243], [39.7676969, -105.2335369], [39.7677304, -105.233602], [39.767761, -105.2336649], [39.7677908, -105.233739], [39.7678172, -105.233812], [39.7678551, -105.2339422], [39.7678736, -105.2340294], [39.7678904, -105.2341257], [39.7679032, -105.2342329], [39.7679095, -105.2343359], [39.7679068, -105.2344027], [39.7679029, -105.234499], [39.7678875, -105.2346626], [39.7678651, -105.2347791], [39.7678384, -105.2348898], [39.7678006, -105.2350283], [39.7677517, -105.2352031], [39.7677119, -105.2353468]],\n", " {"bubblingMouseEvents": true, "color": "#eff821", "dashArray": null, "dashOffset": null, "fill": false, "fillColor": "#eff821", "fillOpacity": 0.2, "fillRule": "evenodd", "lineCap": "round", "lineJoin": "round", "noClip": false, "opacity": 0.8, "smoothFactor": 1.0, "stroke": true, "weight": 10}\n", - " ).addTo(map_0a69fb0ffb4fdf07d938c59697d2d5b7);\n", + " ).addTo(map_08d0a92c80cee353042d813e9d897e9a);\n", " \n", " \n", - " poly_line_b2f52dbae2df737185eacbb7f7f1814f.bindTooltip(\n", + " poly_line_fce8b3c76b03f40e0dbe8d7422abf0e5.bindTooltip(\n", " `<div>\n", - " 0.14906186986507128\n", + " 4.23738036063421\n", " </div>`,\n", " {"sticky": true}\n", " );\n", " \n", " \n", - " var marker_594f011ae899e7390103390e1b1c4e44 = L.marker(\n", + " var marker_57a9fae5fd1181475e8678e53c2e738d = L.marker(\n", " [39.7273069, -105.1994169],\n", " {}\n", - " ).addTo(map_0a69fb0ffb4fdf07d938c59697d2d5b7);\n", + " ).addTo(map_08d0a92c80cee353042d813e9d897e9a);\n", " \n", " \n", - " var icon_bc3742f29d0bbe9069e73d45bf225cb0 = L.AwesomeMarkers.icon(\n", + " var icon_b7f64c69eda0aa3787f9d028a487e2d0 = L.AwesomeMarkers.icon(\n", " {"extraClasses": "fa-rotate-0", "icon": "circle", "iconColor": "white", "markerColor": "green", "prefix": "fa"}\n", " );\n", - " marker_594f011ae899e7390103390e1b1c4e44.setIcon(icon_bc3742f29d0bbe9069e73d45bf225cb0);\n", + " marker_57a9fae5fd1181475e8678e53c2e738d.setIcon(icon_b7f64c69eda0aa3787f9d028a487e2d0);\n", " \n", " \n", - " marker_594f011ae899e7390103390e1b1c4e44.bindTooltip(\n", + " marker_57a9fae5fd1181475e8678e53c2e738d.bindTooltip(\n", " `<div>\n", " Origin\n", " </div>`,\n", @@ -680,19 +673,19 @@ " );\n", " \n", " \n", - " var marker_12b03e286e45c752ef3ec596908d4c51 = L.marker(\n", + " var marker_f2ad9efdc5dfb0e30a8de14f167f2b0e = L.marker(\n", " [39.7677119, -105.2353468],\n", " {}\n", - " ).addTo(map_0a69fb0ffb4fdf07d938c59697d2d5b7);\n", + " ).addTo(map_08d0a92c80cee353042d813e9d897e9a);\n", " \n", " \n", - " var icon_c9a062431c29df954115c641bf7beb58 = L.AwesomeMarkers.icon(\n", + " var icon_6b5edaf0623d1591b034bea1de6b720e = L.AwesomeMarkers.icon(\n", " {"extraClasses": "fa-rotate-0", "icon": "circle", "iconColor": "white", "markerColor": "red", "prefix": "fa"}\n", " );\n", - " marker_12b03e286e45c752ef3ec596908d4c51.setIcon(icon_c9a062431c29df954115c641bf7beb58);\n", + " marker_f2ad9efdc5dfb0e30a8de14f167f2b0e.setIcon(icon_6b5edaf0623d1591b034bea1de6b720e);\n", " \n", " \n", - " marker_12b03e286e45c752ef3ec596908d4c51.bindTooltip(\n", + " marker_f2ad9efdc5dfb0e30a8de14f167f2b0e.bindTooltip(\n", " `<div>\n", " Destination\n", " </div>`,\n", @@ -700,33 +693,33 @@ " );\n", " \n", " \n", - " var poly_line_a48dac49b22022311b057855def06976 = L.polyline(\n", + " var poly_line_56d979cdb52e71e5e1c0fdd7c6f9b67b = L.polyline(\n", " [[39.7273069, -105.1994169], [39.7272989, -105.1994535], [39.7272848, -105.1994868], [39.7272593, -105.1995218], [39.7272272, -105.1995461], [39.7271912, -105.1995576], [39.7271688, -105.199558], [39.7271461, -105.1995532], [39.7271142, -105.1995365], [39.7270868, -105.1995096], [39.7270657, -105.1994741], [39.7270519, -105.1994296], [39.7270483, -105.1993816], [39.727052, -105.1993539], [39.7270571, -105.1993247], [39.7270719, -105.1992916], [39.7270973, -105.1992566], [39.7271293, -105.1992323], [39.7271651, -105.1992207], [39.7271854, -105.1992226], [39.7272113, -105.199227], [39.7273496, -105.1991687], [39.7274185, -105.1991465], [39.7275326, -105.1990928], [39.7276828, -105.1990602], [39.7277455, -105.1990518], [39.7279702, -105.1990232], [39.7281558, -105.1990357], [39.7284384, -105.1990893], [39.7286639, -105.199177], [39.7288222, -105.1992773], [39.7289214, -105.199385], [39.7290292, -105.1995543], [39.7290506, -105.1995879], [39.7292157, -105.1998155], [39.7293094, -105.1998945], [39.7294079, -105.1999604], [39.7295437, -105.200014], [39.7296299, -105.2000721], [39.7296469, -105.1996789], [39.7296738, -105.1987284], [39.7296906, -105.1986794], [39.7297244, -105.1985856], [39.7297272, -105.1984193], [39.7299723, -105.1984198], [39.7302117, -105.1984227], [39.7303235, -105.1984285], [39.7304434, -105.198443], [39.7305651, -105.1984732], [39.7307182, -105.1985167], [39.7308487, -105.1985623], [39.7309909, -105.1986311], [39.731127, -105.1987144], [39.7312292, -105.1987897], [39.7313292, -105.1988731], [39.7314292, -105.1989611], [39.731524, -105.1990554], [39.7316089, -105.1991466], [39.7317085, -105.1992639], [39.7318575, -105.1994488], [39.7319522, -105.1995697], [39.7320423, -105.1996716], [39.7321372, -105.1997578], [39.7322615, -105.1998506], [39.7323697, -105.1999163], [39.7324712, -105.1999637], [39.7325898, -105.2000081], [39.732689, -105.2000309], [39.7327987, -105.2000627], [39.732945, -105.2001049], [39.7330635, -105.2001374], [39.7331838, -105.2001914], [39.7332899, -105.2002528], [39.7334312, -105.200363], [39.7335257, -105.2004544], [39.7335523, -105.2004835], [39.7336132, -105.2005503], [39.7336902, -105.2006555], [39.733767, -105.2007557], [39.7338411, -105.2008491], [39.7339359, -105.2009412], [39.7340357, -105.2010154], [39.7340742, -105.201042], [39.7341117, -105.2010678], [39.7341542, -105.2010901], [39.734211, -105.2011198], [39.7343173, -105.2011712], [39.7344056, -105.2012021], [39.7345704, -105.2012376], [39.7347726, -105.2012753], [39.7351329, -105.2013311], [39.7352203, -105.2013441], [39.7353243, -105.201352], [39.7354415, -105.2013441], [39.7355491, -105.2013268], [39.7357022, -105.2012901], [39.7358395, -105.201259], [39.7362429, -105.2011675], [39.7367157, -105.2010639], [39.7367857, -105.2010539], [39.7368531, -105.2010442], [39.7369645, -105.2010356], [39.7370755, -105.201044], [39.7371895, -105.2010632], [39.7373179, -105.201099], [39.737714, -105.2012295], [39.7379567, -105.2013009], [39.7384451, -105.2014404], [39.7386872, -105.2015167], [39.7395717, -105.2017905], [39.7397153, -105.2018369], [39.7398269, -105.2018673], [39.7399356, -105.2018843], [39.7400618, -105.2018961], [39.740184, -105.2018932], [39.7403133, -105.2018823], [39.7404288, -105.2018645], [39.7405518, -105.2018387], [39.740665, -105.2018073], [39.7407597, -105.2017713], [39.7408728, -105.2017196], [39.740992, -105.2016551], [39.7411268, -105.2015625], [39.7412599, -105.2014489], [39.7413718, -105.2013453], [39.7414758, -105.2012326], [39.741608, -105.2010691], [39.7416901, -105.200964], [39.7417903, -105.2008396], [39.7417994, -105.200826], [39.7419078, -105.2006985], [39.7419027, -105.2006189], [39.7419122, -105.200549], [39.7419496, -105.2004812], [39.7419894, -105.20044], [39.742047, -105.2004033], [39.7420816, -105.2004], [39.7421282, -105.2004131], [39.7421839, -105.200448], [39.7422234, -105.2004924], [39.7422448, -105.2005578], [39.7422566, -105.2006096], [39.7422556, -105.2006625], [39.7422421, -105.2007231], [39.7422712, -105.2008198], [39.7423227, -105.200901], [39.7428069, -105.2016549], [39.7429048, -105.2018114], [39.7429131, -105.2018245], [39.7431503, -105.2022035], [39.7434374, -105.2026477], [39.7435807, -105.2028659], [39.7437705, -105.2031843], [39.7441255, -105.2037193], [39.7444536, -105.2042289], [39.7445369, -105.204369], [39.7448087, -105.2048085], [39.7450389, -105.205197], [39.745164, -105.2054397], [39.7452617, -105.205642], [39.745359, -105.2058753], [39.7454413, -105.2060693], [39.7454698, -105.2061365], [39.7455586, -105.2063498], [39.7456394, -105.2066197], [39.7457343, -105.2068984], [39.7458096, -105.2071619], [39.7458432, -105.2072893], [39.7459057, -105.2075162], [39.7459547, -105.2077463], [39.7460104, -105.2080066], [39.7460312, -105.2081149], [39.7460508, -105.2082709], [39.7460836, -105.2084787], [39.7461092, -105.2086125], [39.7461165, -105.2087078], [39.7461128, -105.2088019], [39.7461078, -105.2089177], [39.7461163, -105.208995], [39.7461377, -105.2090672], [39.7461588, -105.2091147], [39.7462509, -105.2092159], [39.746281, -105.2092382], [39.7462978, -105.209256], [39.7463242, -105.209292], [39.7463458, -105.2093473], [39.746361, -105.2094251], [39.7463921, -105.2095768], [39.7464353, -105.2097133], [39.7465001, -105.2098285], [39.7467097, -105.21004], [39.7467814, -105.2101094], [39.7473357, -105.2106047], [39.7474554, -105.2107116], [39.7475615, -105.2108064], [39.7475958, -105.2108371], [39.747653, -105.2108907], [39.7481502, -105.2113567], [39.7483361, -105.2115309], [39.7483726, -105.2115651], [39.7484183, -105.2116079], [39.7484718, -105.2116584], [39.7487212, -105.2118951], [39.749238, -105.2123721], [39.7492814, -105.2124135], [39.7500059, -105.2130759], [39.7500454, -105.2131129], [39.7503812, -105.2134256], [39.7504627, -105.2135015], [39.7508219, -105.2138362], [39.7508792, -105.2138895], [39.7509348, -105.2139426], [39.7511093, -105.2141093], [39.7516271, -105.2146038], [39.7516716, -105.2146463], [39.751739, -105.2147085], [39.7524829, -105.2153953], [39.7525012, -105.2154123], [39.7532517, -105.2161101], [39.7533052, -105.2161598], [39.7540716, -105.2168674], [39.7541308, -105.2169221], [39.7544795, -105.2172508], [39.7546281, -105.2174522], [39.7546745, -105.2175037], [39.7548394, -105.2176869], [39.7549112, -105.2177871], [39.7548521, -105.2179433], [39.7546476, -105.2183047], [39.7544379, -105.2186805], [39.7544293, -105.218688], [39.7543723, -105.2187394], [39.7544376, -105.2188002], [39.7544872, -105.2188464], [39.7547934, -105.2191313], [39.7551071, -105.2194077], [39.7551368, -105.2194339], [39.755202, -105.2194913], [39.7552559, -105.2195411], [39.755615, -105.2198733], [39.755696, -105.2199479], [39.7559351, -105.2201693], [39.7559559, -105.220191], [39.7560109, -105.2202486], [39.7560703, -105.2203182], [39.756257, -105.2205368], [39.7562813, -105.2205598], [39.7564057, -105.2206774], [39.7565461, -105.2208058], [39.7567417, -105.2209442], [39.7566986, -105.2211001], [39.756638, -105.2213112], [39.7565732, -105.2214643], [39.7565209, -105.2215678], [39.7564595, -105.2216814], [39.7562914, -105.2219923], [39.7562254, -105.2221143], [39.7563006, -105.2221853], [39.7566193, -105.222483], [39.7566315, -105.2224957], [39.7569475, -105.2227913], [39.7570561, -105.222895], [39.7574792, -105.2232693], [39.7575579, -105.2233409], [39.7576236, -105.2234019], [39.7578565, -105.2236199], [39.7579703, -105.2237258], [39.7580027, -105.2237561], [39.7583001, -105.2240544], [39.7583591, -105.2241084], [39.7587249, -105.2244446], [39.7588057, -105.22452], [39.7591609, -105.2248516], [39.7595089, -105.2251741], [39.7596887, -105.2253408], [39.7597554, -105.2254024], [39.7598802, -105.2255173], [39.7598939, -105.2255305], [39.7599658, -105.2255988], [39.7600186, -105.2256498], [39.7604008, -105.2259941], [39.7605066, -105.2260931], [39.7605924, -105.226174], [39.760667, -105.2262427], [39.7607946, -105.2263631], [39.7608753, -105.2264373], [39.7615829, -105.2271044], [39.7616266, -105.2271481], [39.7616788, -105.2272008], [39.762203, -105.2276943], [39.7623411, -105.2278299], [39.7630405, -105.2284565], [39.7632824, -105.2286685], [39.7637358, -105.2290679], [39.7640137, -105.229309], [39.7640954, -105.229381], [39.7641743, -105.229462], [39.76436, -105.2296495], [39.7645376, -105.2298658], [39.764824, -105.2301918], [39.7652051, -105.2307044], [39.7654711, -105.2310802], [39.7654801, -105.2310934], [39.7655308, -105.231168], [39.7655792, -105.231232], [39.7655918, -105.2312486], [39.7657719, -105.2315035], [39.7658973, -105.2316751], [39.7659962, -105.2318039], [39.7661018, -105.231924], [39.7662469, -105.2320614], [39.7664515, -105.2322588], [39.7671442, -105.2329283], [39.7673559, -105.2331244], [39.7674414, -105.2332053], [39.7675186, -105.2332845], [39.7675756, -105.2333522], [39.7676291, -105.2334243], [39.7676969, -105.2335369], [39.7677304, -105.233602], [39.767761, -105.2336649], [39.7677908, -105.233739], [39.7678172, -105.233812], [39.7678551, -105.2339422], [39.7678736, -105.2340294], [39.7678904, -105.2341257], [39.7679032, -105.2342329], [39.7679095, -105.2343359], [39.7679068, -105.2344027], [39.7679029, -105.234499], [39.7678875, -105.2346626], [39.7678651, -105.2347791], [39.7678384, -105.2348898], [39.7678006, -105.2350283], [39.7677517, -105.2352031], [39.7677119, -105.2353468]],\n", " {"bubblingMouseEvents": true, "color": "#0c0786", "dashArray": null, "dashOffset": null, "fill": false, "fillColor": "#0c0786", "fillOpacity": 0.2, "fillRule": "evenodd", "lineCap": "round", "lineJoin": "round", "noClip": false, "opacity": 0.8, "smoothFactor": 1.0, "stroke": true, "weight": 10}\n", - " ).addTo(map_0a69fb0ffb4fdf07d938c59697d2d5b7);\n", + " ).addTo(map_08d0a92c80cee353042d813e9d897e9a);\n", " \n", " \n", - " poly_line_a48dac49b22022311b057855def06976.bindTooltip(\n", + " poly_line_56d979cdb52e71e5e1c0fdd7c6f9b67b.bindTooltip(\n", " `<div>\n", - " 0.14412252939511852\n", + " 4.013268489898491\n", " </div>`,\n", " {"sticky": true}\n", " );\n", " \n", " \n", - " var marker_a42455ffdb4b3173ce10e967b7fe4e0d = L.marker(\n", + " var marker_8223aabc91c0f1ea02ec2ea5181c8453 = L.marker(\n", " [39.7273069, -105.1994169],\n", " {}\n", - " ).addTo(map_0a69fb0ffb4fdf07d938c59697d2d5b7);\n", + " ).addTo(map_08d0a92c80cee353042d813e9d897e9a);\n", " \n", " \n", - " var icon_182dc1421ab527e4215256615066eba2 = L.AwesomeMarkers.icon(\n", + " var icon_efff58254cff016e5572e3aa608b8d64 = L.AwesomeMarkers.icon(\n", " {"extraClasses": "fa-rotate-0", "icon": "circle", "iconColor": "white", "markerColor": "green", "prefix": "fa"}\n", " );\n", - " marker_a42455ffdb4b3173ce10e967b7fe4e0d.setIcon(icon_182dc1421ab527e4215256615066eba2);\n", + " marker_8223aabc91c0f1ea02ec2ea5181c8453.setIcon(icon_efff58254cff016e5572e3aa608b8d64);\n", " \n", " \n", - " marker_a42455ffdb4b3173ce10e967b7fe4e0d.bindTooltip(\n", + " marker_8223aabc91c0f1ea02ec2ea5181c8453.bindTooltip(\n", " `<div>\n", " Origin\n", " </div>`,\n", @@ -734,19 +727,19 @@ " );\n", " \n", " \n", - " var marker_3f60358a3318165e8ed01991c2970fe3 = L.marker(\n", + " var marker_e3fa3a2726000cb73f319b6758aa89a8 = L.marker(\n", " [39.7677119, -105.2353468],\n", " {}\n", - " ).addTo(map_0a69fb0ffb4fdf07d938c59697d2d5b7);\n", + " ).addTo(map_08d0a92c80cee353042d813e9d897e9a);\n", " \n", " \n", - " var icon_7a556a87128e1f5ad26d35ebcac3cf6e = L.AwesomeMarkers.icon(\n", + " var icon_1d3ff4766d7167f6aae206479f62d543 = L.AwesomeMarkers.icon(\n", " {"extraClasses": "fa-rotate-0", "icon": "circle", "iconColor": "white", "markerColor": "red", "prefix": "fa"}\n", " );\n", - " marker_3f60358a3318165e8ed01991c2970fe3.setIcon(icon_7a556a87128e1f5ad26d35ebcac3cf6e);\n", + " marker_e3fa3a2726000cb73f319b6758aa89a8.setIcon(icon_1d3ff4766d7167f6aae206479f62d543);\n", " \n", " \n", - " marker_3f60358a3318165e8ed01991c2970fe3.bindTooltip(\n", + " marker_e3fa3a2726000cb73f319b6758aa89a8.bindTooltip(\n", " `<div>\n", " Destination\n", " </div>`,\n", @@ -757,16 +750,16 @@ "</html>\" style=\"position:absolute;width:100%;height:100%;left:0;top:0;border:none !important;\" allowfullscreen webkitallowfullscreen mozallowfullscreen></iframe></div></div>" ], "text/plain": [ - "<folium.folium.Map at 0x1a3b215b0>" + "<folium.folium.Map at 0x12ca7f290>" ] }, - "execution_count": 14, + "execution_count": 16, "metadata": {}, "output_type": "execute_result" } ], "source": [ - "m = plot_routes_folium(results, value_fn=lambda r: r[\"traversal_summary\"][\"energy\"], color_map=\"plasma\")\n", + "m = plot_routes_folium(results, value_fn=lambda r: r[\"traversal_summary\"][\"vehicle\"][\"energy\"], color_map=\"plasma\")\n", "m" ] }, diff --git a/python/nrel/routee/compass/resources/models/2016_BMW_i3_REx_PHEV_Charge_Depleting.bin b/python/nrel/routee/compass/resources/models/2016_BMW_i3_REx_PHEV_Charge_Depleting.bin new file mode 100644 index 00000000..d6d6b993 Binary files /dev/null and b/python/nrel/routee/compass/resources/models/2016_BMW_i3_REx_PHEV_Charge_Depleting.bin differ diff --git a/python/nrel/routee/compass/resources/models/2016_BMW_i3_REx_PHEV_Charge_Sustaining.bin b/python/nrel/routee/compass/resources/models/2016_BMW_i3_REx_PHEV_Charge_Sustaining.bin new file mode 100644 index 00000000..373891f8 Binary files /dev/null and b/python/nrel/routee/compass/resources/models/2016_BMW_i3_REx_PHEV_Charge_Sustaining.bin differ diff --git a/python/nrel/routee/compass/resources/models/2016_CHEVROLET_Volt_Charge_Depleting.bin b/python/nrel/routee/compass/resources/models/2016_CHEVROLET_Volt_Charge_Depleting.bin new file mode 100644 index 00000000..32927154 Binary files /dev/null and b/python/nrel/routee/compass/resources/models/2016_CHEVROLET_Volt_Charge_Depleting.bin differ diff --git a/python/nrel/routee/compass/resources/models/2016_CHEVROLET_Volt_Charge_Sustaining.bin b/python/nrel/routee/compass/resources/models/2016_CHEVROLET_Volt_Charge_Sustaining.bin new file mode 100644 index 00000000..5745af3d Binary files /dev/null and b/python/nrel/routee/compass/resources/models/2016_CHEVROLET_Volt_Charge_Sustaining.bin differ diff --git a/python/nrel/routee/compass/resources/models/2016_FORD_C-MAX_(PHEV)_Charge_Depleting.bin b/python/nrel/routee/compass/resources/models/2016_FORD_C-MAX_(PHEV)_Charge_Depleting.bin new file mode 100644 index 00000000..a7bc8059 Binary files /dev/null and b/python/nrel/routee/compass/resources/models/2016_FORD_C-MAX_(PHEV)_Charge_Depleting.bin differ diff --git a/python/nrel/routee/compass/resources/models/2016_FORD_C-MAX_(PHEV)_Charge_Sustaining.bin b/python/nrel/routee/compass/resources/models/2016_FORD_C-MAX_(PHEV)_Charge_Sustaining.bin new file mode 100644 index 00000000..e283a16d Binary files /dev/null and b/python/nrel/routee/compass/resources/models/2016_FORD_C-MAX_(PHEV)_Charge_Sustaining.bin differ diff --git a/python/nrel/routee/compass/resources/models/2016_HYUNDAI_Sonata_PHEV_Charge_Depleting.bin b/python/nrel/routee/compass/resources/models/2016_HYUNDAI_Sonata_PHEV_Charge_Depleting.bin new file mode 100644 index 00000000..91446c6e Binary files /dev/null and b/python/nrel/routee/compass/resources/models/2016_HYUNDAI_Sonata_PHEV_Charge_Depleting.bin differ diff --git a/python/nrel/routee/compass/resources/models/2016_HYUNDAI_Sonata_PHEV_Charge_Sustaining.bin b/python/nrel/routee/compass/resources/models/2016_HYUNDAI_Sonata_PHEV_Charge_Sustaining.bin new file mode 100644 index 00000000..a47c1085 Binary files /dev/null and b/python/nrel/routee/compass/resources/models/2016_HYUNDAI_Sonata_PHEV_Charge_Sustaining.bin differ diff --git a/python/nrel/routee/compass/resources/models/2017_Prius_Prime_Charge_Depleting.bin b/python/nrel/routee/compass/resources/models/2017_Prius_Prime_Charge_Depleting.bin new file mode 100644 index 00000000..a55bd3eb Binary files /dev/null and b/python/nrel/routee/compass/resources/models/2017_Prius_Prime_Charge_Depleting.bin differ diff --git a/python/nrel/routee/compass/resources/models/2017_Prius_Prime_Charge_Sustaining.bin b/python/nrel/routee/compass/resources/models/2017_Prius_Prime_Charge_Sustaining.bin new file mode 100644 index 00000000..ed388588 Binary files /dev/null and b/python/nrel/routee/compass/resources/models/2017_Prius_Prime_Charge_Sustaining.bin differ diff --git a/python/nrel/routee/compass/resources/osm_default_energy.toml b/python/nrel/routee/compass/resources/osm_default_energy.toml index 7dae2fdb..a055ce36 100644 --- a/python/nrel/routee/compass/resources/osm_default_energy.toml +++ b/python/nrel/routee/compass/resources/osm_default_energy.toml @@ -17,14 +17,15 @@ output_plugins = [ ] [traversal] -type = "speed_grade_energy_model" +type = "energy_model" speed_table_input_file = "edges-posted-speed-enumerated.txt.gz" speed_table_speed_unit = "kilometers_per_hour" output_time_unit = "minutes" output_distance_unit = "miles" -[[traversal.energy_models]] +[[traversal.vehicles]] name = "2012_Ford_Focus" +type = "single_fuel" model_input_file = "models/2012_Ford_Focus.bin" model_type = "smartcore" speed_unit = "miles_per_hour" @@ -33,8 +34,9 @@ energy_rate_unit = "gallons_gasoline_per_mile" ideal_energy_rate = 0.02857143 real_world_energy_adjustment = 1.166 -[[traversal.energy_models]] +[[traversal.vehicles]] name = "2012_Ford_Fusion" +type = "single_fuel" model_input_file = "models/2012_Ford_Fusion.bin" model_type = "smartcore" speed_unit = "miles_per_hour" @@ -43,8 +45,9 @@ energy_rate_unit = "gallons_gasoline_per_mile" ideal_energy_rate = 0.02857143 real_world_energy_adjustment = 1.166 -[[traversal.energy_models]] +[[traversal.vehicles]] name = "2016_AUDI_A3_4cyl_2WD" +type = "single_fuel" model_input_file = "models/2016_AUDI_A3_4cyl_2WD.bin" model_type = "smartcore" speed_unit = "miles_per_hour" @@ -53,8 +56,9 @@ energy_rate_unit = "gallons_gasoline_per_mile" ideal_energy_rate = 0.02857143 real_world_energy_adjustment = 1.166 -[[traversal.energy_models]] +[[traversal.vehicles]] name = "2016_BMW_328d_4cyl_2WD" +type = "single_fuel" model_input_file = "models/2016_BMW_328d_4cyl_2WD.bin" model_type = "smartcore" speed_unit = "miles_per_hour" @@ -63,8 +67,9 @@ energy_rate_unit = "gallons_gasoline_per_mile" ideal_energy_rate = 0.02857143 real_world_energy_adjustment = 1.166 -[[traversal.energy_models]] +[[traversal.vehicles]] name = "2016_CHEVROLET_Malibu_4cyl_2WD" +type = "single_fuel" model_input_file = "models/2016_CHEVROLET_Malibu_4cyl_2WD.bin" model_type = "smartcore" speed_unit = "miles_per_hour" @@ -73,8 +78,9 @@ energy_rate_unit = "gallons_gasoline_per_mile" ideal_energy_rate = 0.02857143 real_world_energy_adjustment = 1.166 -[[traversal.energy_models]] +[[traversal.vehicles]] name = "2016_CHEVROLET_Spark_EV" +type = "single_fuel" model_input_file = "models/2016_CHEVROLET_Spark_EV.bin" model_type = "smartcore" speed_unit = "miles_per_hour" @@ -83,8 +89,9 @@ energy_rate_unit = "kilowatt_hours_per_mile" ideal_energy_rate = 0.2 real_world_energy_adjustment = 1.3958 -[[traversal.energy_models]] +[[traversal.vehicles]] name = "2016_FORD_C-MAX_HEV" +type = "single_fuel" model_input_file = "models/2016_FORD_C-MAX_HEV.bin" model_type = "smartcore" speed_unit = "miles_per_hour" @@ -93,8 +100,9 @@ energy_rate_unit = "gallons_gasoline_per_mile" ideal_energy_rate = 0.02 real_world_energy_adjustment = 1.1252 -[[traversal.energy_models]] +[[traversal.vehicles]] name = "2016_FORD_Escape_4cyl_2WD" +type = "single_fuel" model_input_file = "models/2016_FORD_Escape_4cyl_2WD.bin" model_type = "smartcore" speed_unit = "miles_per_hour" @@ -103,8 +111,9 @@ energy_rate_unit = "gallons_gasoline_per_mile" ideal_energy_rate = 0.02857143 real_world_energy_adjustment = 1.166 -[[traversal.energy_models]] +[[traversal.vehicles]] name = "2016_FORD_Explorer_4cyl_2WD" +type = "single_fuel" model_input_file = "models/2016_FORD_Explorer_4cyl_2WD.bin" model_type = "smartcore" speed_unit = "miles_per_hour" @@ -113,8 +122,9 @@ energy_rate_unit = "gallons_gasoline_per_mile" ideal_energy_rate = 0.02857143 real_world_energy_adjustment = 1.166 -[[traversal.energy_models]] +[[traversal.vehicles]] name = "2016_HYUNDAI_Elantra_4cyl_2WD" +type = "single_fuel" model_input_file = "models/2016_HYUNDAI_Elantra_4cyl_2WD.bin" model_type = "smartcore" speed_unit = "miles_per_hour" @@ -123,8 +133,9 @@ energy_rate_unit = "gallons_gasoline_per_mile" ideal_energy_rate = 0.02857143 real_world_energy_adjustment = 1.166 -[[traversal.energy_models]] +[[traversal.vehicles]] name = "2016_KIA_Optima_Hybrid" +type = "single_fuel" model_input_file = "models/2016_KIA_Optima_Hybrid.bin" model_type = "smartcore" speed_unit = "miles_per_hour" @@ -133,8 +144,9 @@ energy_rate_unit = "gallons_gasoline_per_mile" ideal_energy_rate = 0.02 real_world_energy_adjustment = 1.1252 -[[traversal.energy_models]] +[[traversal.vehicles]] name = "2016_Leaf_24_kWh" +type = "single_fuel" model_input_file = "models/2016_Leaf_24_kWh.bin" model_type = "smartcore" speed_unit = "miles_per_hour" @@ -143,8 +155,9 @@ energy_rate_unit = "kilowatt_hours_per_mile" ideal_energy_rate = 0.2 real_world_energy_adjustment = 1.3958 -[[traversal.energy_models]] +[[traversal.vehicles]] name = "2016_MITSUBISHI_i-MiEV" +type = "single_fuel" model_input_file = "models/2016_MITSUBISHI_i-MiEV.bin" model_type = "smartcore" speed_unit = "miles_per_hour" @@ -153,8 +166,9 @@ energy_rate_unit = "kilowatt_hours_per_mile" ideal_energy_rate = 0.2 real_world_energy_adjustment = 1.3958 -[[traversal.energy_models]] +[[traversal.vehicles]] name = "2016_Nissan_Leaf_30_kWh" +type = "single_fuel" model_input_file = "models/2016_Nissan_Leaf_30_kWh.bin" model_type = "smartcore" speed_unit = "miles_per_hour" @@ -163,8 +177,9 @@ energy_rate_unit = "kilowatt_hours_per_mile" ideal_energy_rate = 0.2 real_world_energy_adjustment = 1.3958 -[[traversal.energy_models]] +[[traversal.vehicles]] name = "2016_TESLA_Model_S60_2WD" +type = "single_fuel" model_input_file = "models/2016_TESLA_Model_S60_2WD.bin" model_type = "smartcore" speed_unit = "miles_per_hour" @@ -173,8 +188,9 @@ energy_rate_unit = "kilowatt_hours_per_mile" ideal_energy_rate = 0.2 real_world_energy_adjustment = 1.3958 -[[traversal.energy_models]] +[[traversal.vehicles]] name = "2016_TOYOTA_Camry_4cyl_2WD" +type = "single_fuel" model_input_file = "models/2016_TOYOTA_Camry_4cyl_2WD.bin" model_type = "smartcore" speed_unit = "miles_per_hour" @@ -183,8 +199,9 @@ energy_rate_unit = "gallons_gasoline_per_mile" ideal_energy_rate = 0.02857143 real_world_energy_adjustment = 1.166 -[[traversal.energy_models]] +[[traversal.vehicles]] name = "2016_TOYOTA_Corolla_4cyl_2WD" +type = "single_fuel" model_input_file = "models/2016_TOYOTA_Corolla_4cyl_2WD.bin" model_type = "smartcore" speed_unit = "miles_per_hour" @@ -193,8 +210,9 @@ energy_rate_unit = "gallons_gasoline_per_mile" ideal_energy_rate = 0.02857143 real_world_energy_adjustment = 1.166 -[[traversal.energy_models]] +[[traversal.vehicles]] name = "2016_TOYOTA_Highlander_Hybrid" +type = "single_fuel" model_input_file = "models/2016_TOYOTA_Highlander_Hybrid.bin" model_type = "smartcore" speed_unit = "miles_per_hour" @@ -203,8 +221,9 @@ energy_rate_unit = "gallons_gasoline_per_mile" ideal_energy_rate = 0.02 real_world_energy_adjustment = 1.1252 -[[traversal.energy_models]] +[[traversal.vehicles]] name = "2016_Toyota_Prius_Two_FWD" +type = "single_fuel" model_input_file = "models/2016_Toyota_Prius_Two_FWD.bin" model_type = "smartcore" speed_unit = "miles_per_hour" @@ -213,8 +232,9 @@ energy_rate_unit = "gallons_gasoline_per_mile" ideal_energy_rate = 0.02 real_world_energy_adjustment = 1.1252 -[[traversal.energy_models]] +[[traversal.vehicles]] name = "2017_CHEVROLET_Bolt" +type = "single_fuel" model_input_file = "models/2017_CHEVROLET_Bolt.bin" model_type = "smartcore" speed_unit = "miles_per_hour" @@ -223,8 +243,9 @@ energy_rate_unit = "kilowatt_hours_per_mile" ideal_energy_rate = 0.2 real_world_energy_adjustment = 1.3958 -[[traversal.energy_models]] +[[traversal.vehicles]] name = "2017_Maruti_Dzire_VDI" +type = "single_fuel" model_input_file = "models/2017_Maruti_Dzire_VDI.bin" model_type = "smartcore" speed_unit = "miles_per_hour" @@ -233,8 +254,9 @@ energy_rate_unit = "gallons_gasoline_per_mile" ideal_energy_rate = 0.02857143 real_world_energy_adjustment = 1.166 -[[traversal.energy_models]] +[[traversal.vehicles]] name = "2017_Toyota_Highlander_3.5_L" +type = "single_fuel" model_input_file = "models/2017_Toyota_Highlander_3.5_L.bin" model_type = "smartcore" speed_unit = "miles_per_hour" @@ -243,8 +265,9 @@ energy_rate_unit = "gallons_gasoline_per_mile" ideal_energy_rate = 0.02857143 real_world_energy_adjustment = 1.166 -[[traversal.energy_models]] +[[traversal.vehicles]] name = "2020_Chevrolet_Colorado_2WD_Diesel" +type = "single_fuel" model_input_file = "models/2020_Chevrolet_Colorado_2WD_Diesel.bin" model_type = "smartcore" speed_unit = "miles_per_hour" @@ -253,8 +276,9 @@ energy_rate_unit = "gallons_gasoline_per_mile" ideal_energy_rate = 0.02857143 real_world_energy_adjustment = 1.166 -[[traversal.energy_models]] +[[traversal.vehicles]] name = "2020_VW_Golf_1.5TSI" +type = "single_fuel" model_input_file = "models/2020_VW_Golf_1.5TSI.bin" model_type = "smartcore" speed_unit = "miles_per_hour" @@ -263,8 +287,9 @@ energy_rate_unit = "gallons_gasoline_per_mile" ideal_energy_rate = 0.02857143 real_world_energy_adjustment = 1.166 -[[traversal.energy_models]] +[[traversal.vehicles]] name = "2020_VW_Golf_2.0TDI" +type = "single_fuel" model_input_file = "models/2020_VW_Golf_2.0TDI.bin" model_type = "smartcore" speed_unit = "miles_per_hour" @@ -273,8 +298,9 @@ energy_rate_unit = "gallons_gasoline_per_mile" ideal_energy_rate = 0.02857143 real_world_energy_adjustment = 1.166 -[[traversal.energy_models]] +[[traversal.vehicles]] name = "2021_Fiat_Panda_Mild_Hybrid" +type = "single_fuel" model_input_file = "models/2021_Fiat_Panda_Mild_Hybrid.bin" model_type = "smartcore" speed_unit = "miles_per_hour" @@ -283,8 +309,9 @@ energy_rate_unit = "gallons_gasoline_per_mile" ideal_energy_rate = 0.02857143 real_world_energy_adjustment = 1.166 -[[traversal.energy_models]] +[[traversal.vehicles]] name = "2021_Peugot_3008" +type = "single_fuel" model_input_file = "models/2021_Peugot_3008.bin" model_type = "smartcore" speed_unit = "miles_per_hour" @@ -293,8 +320,9 @@ energy_rate_unit = "gallons_gasoline_per_mile" ideal_energy_rate = 0.02857143 real_world_energy_adjustment = 1.166 -[[traversal.energy_models]] +[[traversal.vehicles]] name = "2022_Ford_F-150_Lightning_4WD" +type = "single_fuel" model_input_file = "models/2022_Ford_F-150_Lightning_4WD.bin" model_type = "smartcore" speed_unit = "miles_per_hour" @@ -303,8 +331,9 @@ energy_rate_unit = "kilowatt_hours_per_mile" ideal_energy_rate = 0.2 real_world_energy_adjustment = 1.3958 -[[traversal.energy_models]] +[[traversal.vehicles]] name = "2022_Renault_Zoe_ZE50_R135" +type = "single_fuel" model_input_file = "models/2022_Renault_Zoe_ZE50_R135.bin" model_type = "smartcore" speed_unit = "miles_per_hour" @@ -313,8 +342,9 @@ energy_rate_unit = "kilowatt_hours_per_mile" ideal_energy_rate = 0.2 real_world_energy_adjustment = 1.3958 -[[traversal.energy_models]] +[[traversal.vehicles]] name = "2022_Tesla_Model_3_RWD" +type = "single_fuel" model_input_file = "models/2022_Tesla_Model_3_RWD.bin" model_type = "smartcore" speed_unit = "miles_per_hour" @@ -323,8 +353,9 @@ energy_rate_unit = "kilowatt_hours_per_mile" ideal_energy_rate = 0.2 real_world_energy_adjustment = 1.3958 -[[traversal.energy_models]] +[[traversal.vehicles]] name = "2022_Tesla_Model_Y_RWD" +type = "single_fuel" model_input_file = "models/2022_Tesla_Model_Y_RWD.bin" model_type = "smartcore" speed_unit = "miles_per_hour" @@ -333,8 +364,9 @@ energy_rate_unit = "kilowatt_hours_per_mile" ideal_energy_rate = 0.2 real_world_energy_adjustment = 1.3958 -[[traversal.energy_models]] +[[traversal.vehicles]] name = "2022_Toyota_Yaris_Hybrid_Mid" +type = "single_fuel" model_input_file = "models/2022_Toyota_Yaris_Hybrid_Mid.bin" model_type = "smartcore" speed_unit = "miles_per_hour" @@ -343,8 +375,9 @@ energy_rate_unit = "gallons_gasoline_per_mile" ideal_energy_rate = 0.02 real_world_energy_adjustment = 1.1252 -[[traversal.energy_models]] +[[traversal.vehicles]] name = "2022_Volvo_XC40_Recharge_twin" +type = "single_fuel" model_input_file = "models/2022_Volvo_XC40_Recharge_twin.bin" model_type = "smartcore" speed_unit = "miles_per_hour" @@ -353,8 +386,9 @@ energy_rate_unit = "kilowatt_hours_per_mile" ideal_energy_rate = 0.2 real_world_energy_adjustment = 1.3958 -[[traversal.energy_models]] +[[traversal.vehicles]] name = "2023_Mitsubishi_Pajero_Sport" +type = "single_fuel" model_input_file = "models/2023_Mitsubishi_Pajero_Sport.bin" model_type = "smartcore" speed_unit = "miles_per_hour" @@ -362,3 +396,123 @@ grade_unit = "decimal" energy_rate_unit = "gallons_gasoline_per_mile" ideal_energy_rate = 0.02857143 real_world_energy_adjustment = 1.166 + +[[traversal.vehicles]] +type = "dual_fuel" +name = "2016_BMW_i3_REx_PHEV" +battery_capacity = 12 +battery_capacity_unit = "kilowatt_hours" +[traversal.vehicles.charge_depleting] +name = "2016_BMW_i3_REx_PHEV_Charge_Depleting" +model_input_file = "models/2016_BMW_i3_REx_PHEV_Charge_Depleting.bin" +model_type = "smartcore" +speed_unit = "miles_per_hour" +grade_unit = "decimal" +energy_rate_unit = "kilowatt_hours_per_mile" +ideal_energy_rate = 0.2 +real_world_energy_adjustment = 1.3958 +[traversal.vehicles.charge_sustaining] +name = "2016_BMW_i3_REx_PHEV_Charge_Sustaining" +model_input_file = "models/2016_BMW_i3_REx_PHEV_Charge_Sustaining.bin" +model_type = "smartcore" +speed_unit = "miles_per_hour" +grade_unit = "decimal" +energy_rate_unit = "gallons_gasoline_per_mile" +ideal_energy_rate = 0.02 +real_world_energy_adjustment = 1.1252 + +[[traversal.vehicles]] +type = "dual_fuel" +name = "2016_CHEVROLET_Volt" +battery_capacity = 12 +battery_capacity_unit = "kilowatt_hours" +[traversal.vehicles.charge_depleting] +name = "2016_CHEVROLET_Volt_Charge_Depleting" +model_input_file = "models/2016_CHEVROLET_Volt_Charge_Depleting.bin" +model_type = "smartcore" +speed_unit = "miles_per_hour" +grade_unit = "decimal" +energy_rate_unit = "kilowatt_hours_per_mile" +ideal_energy_rate = 0.2 +real_world_energy_adjustment = 1.3958 +[traversal.vehicles.charge_sustaining] +name = "2016_CHEVROLET_Volt_Charge_Sustaining" +model_input_file = "models/2016_CHEVROLET_Volt_Charge_Sustaining.bin" +model_type = "smartcore" +speed_unit = "miles_per_hour" +grade_unit = "decimal" +energy_rate_unit = "gallons_gasoline_per_mile" +ideal_energy_rate = 0.02 +real_world_energy_adjustment = 1.1252 + +[[traversal.vehicles]] +type = "dual_fuel" +name = "2016_FORD_C-MAX_(PHEV)" +battery_capacity = 7.6 +battery_capacity_unit = "kilowatt_hours" +[traversal.vehicles.charge_depleting] +name = "2016_FORD_C-MAX_(PHEV)_Charge_Depleting" +model_input_file = "models/2016_FORD_C-MAX_(PHEV)_Charge_Depleting.bin" +model_type = "smartcore" +speed_unit = "miles_per_hour" +grade_unit = "decimal" +energy_rate_unit = "kilowatt_hours_per_mile" +ideal_energy_rate = 0.2 +real_world_energy_adjustment = 1.3958 +[traversal.vehicles.charge_sustaining] +name = "2016_FORD_C-MAX_(PHEV)_Charge_Sustaining" +model_input_file = "models/2016_FORD_C-MAX_(PHEV)_Charge_Sustaining.bin" +model_type = "smartcore" +speed_unit = "miles_per_hour" +grade_unit = "decimal" +energy_rate_unit = "gallons_gasoline_per_mile" +ideal_energy_rate = 0.02 +real_world_energy_adjustment = 1.1252 + +[[traversal.vehicles]] +type = "dual_fuel" +name = "2016_HYUNDAI_Sonata_PHEV" +battery_capacity = 9.8 +battery_capacity_unit = "kilowatt_hours" +[traversal.vehicles.charge_depleting] +name = "2016_HYUNDAI_Sonata_PHEV_Charge_Depleting" +model_input_file = "models/2016_HYUNDAI_Sonata_PHEV_Charge_Depleting.bin" +model_type = "smartcore" +speed_unit = "miles_per_hour" +grade_unit = "decimal" +energy_rate_unit = "kilowatt_hours_per_mile" +ideal_energy_rate = 0.2 +real_world_energy_adjustment = 1.3958 +[traversal.vehicles.charge_sustaining] +name = "2016_HYUNDAI_Sonata_PHEV_Charge_Sustaining" +model_input_file = "models/2016_HYUNDAI_Sonata_PHEV_Charge_Sustaining.bin" +model_type = "smartcore" +speed_unit = "miles_per_hour" +grade_unit = "decimal" +energy_rate_unit = "gallons_gasoline_per_mile" +ideal_energy_rate = 0.02 +real_world_energy_adjustment = 1.1252 + +[[traversal.vehicles]] +type = "dual_fuel" +name = "2017_Prius_Prime" +battery_capacity = 8 +battery_capacity_unit = "kilowatt_hours" +[traversal.vehicles.charge_depleting] +name = "2017_Prius_Prime_Charge_Depleting" +model_input_file = "models/2017_Prius_Prime_Charge_Depleting.bin" +model_type = "smartcore" +speed_unit = "miles_per_hour" +grade_unit = "decimal" +energy_rate_unit = "kilowatt_hours_per_mile" +ideal_energy_rate = 0.2 +real_world_energy_adjustment = 1.3958 +[traversal.vehicles.charge_sustaining] +name = "2017_Prius_Prime_Charge_Sustaining" +model_input_file = "models/2017_Prius_Prime_Charge_Sustaining.bin" +model_type = "smartcore" +speed_unit = "miles_per_hour" +grade_unit = "decimal" +energy_rate_unit = "gallons_gasoline_per_mile" +ideal_energy_rate = 0.02 +real_world_energy_adjustment = 1.1252 \ No newline at end of file diff --git a/rust/routee-compass-core/src/util/unit/energy_unit.rs b/rust/routee-compass-core/src/util/unit/energy_unit.rs index 3370a01e..1f718219 100644 --- a/rust/routee-compass-core/src/util/unit/energy_unit.rs +++ b/rust/routee-compass-core/src/util/unit/energy_unit.rs @@ -1,13 +1,25 @@ use serde::{Deserialize, Serialize}; -#[derive(Debug, Serialize, Deserialize, Clone, PartialEq)] +use super::Energy; + +#[derive(Debug, Serialize, Deserialize, Clone, PartialEq, Copy)] #[serde(rename_all = "snake_case")] pub enum EnergyUnit { GallonsGasoline, KilowattHours, } -impl EnergyUnit {} +impl EnergyUnit { + pub fn convert(&self, value: Energy, target: EnergyUnit) -> Energy { + use EnergyUnit as S; + match (self, target) { + (S::GallonsGasoline, S::GallonsGasoline) => value, + (S::GallonsGasoline, S::KilowattHours) => value * 33.41, + (S::KilowattHours, S::GallonsGasoline) => value * 0.0299, + (S::KilowattHours, S::KilowattHours) => value, + } + } +} impl std::fmt::Display for EnergyUnit { fn fmt(&self, f: &mut std::fmt::Formatter<'_>) -> std::fmt::Result { diff --git a/rust/routee-compass-powertrain/src/doc.md b/rust/routee-compass-powertrain/src/doc.md index 0ea5e76c..cbd066dc 100644 --- a/rust/routee-compass-powertrain/src/doc.md +++ b/rust/routee-compass-powertrain/src/doc.md @@ -11,7 +11,7 @@ RouteE Powertrain models are trained and exported via the RouteE Powertrain util ### Model Runtimes There are two underlying model runtimes available, [smartcore](https://smartcorelib.org/) and [ort](https://github.com/pykeio/ort) (for [ONNX](https://onnx.ai/) models). -By default, this crate is loaded with ONNX deactivated. +By default, this crate is loaded with ONNX deactivated. To activate the ONNX feature, pass the `onnx` feature flag during compilation. Runtime kernels for 3 common OSs have been provided in the onnx-runtime directory within this crate. For more information on cargo features, see The Cargo Book chapter on [Features](https://doc.rust-lang.org/cargo/reference/features.html). @@ -25,29 +25,32 @@ An example traversal model configuration that uses this crate may look like this ```toml [traversal] -type = "speed_grade_energy_model" -model_type = "smartcore" +type = "energy_model" speed_table_input_file = "edges-posted-speed-enumerated.txt.gz" -energy_model_input_file = "2016_TOYOTA_Camry_4cyl_2WD.bin" -ideal_energy_rate = 0.02857142857 speed_table_speed_unit = "kilometers_per_hour" -energy_model_speed_unit = "miles_per_hour" -energy_model_grade_unit = "decimal" -energy_model_energy_rate_unit = "gallons_gasoline_per_mile" output_time_unit = "minutes" output_distance_unit = "miles" -grade_table_input_file = "edges-grade-enumerated.txt.gz" -grade_table_grade_unit = "decimal" + +[[traversal.vehicles]] +name = "2012_Ford_Focus" +type = "single_fuel" +model_input_file = "models/2012_Ford_Focus.bin" +model_type = "smartcore" +speed_unit = "miles_per_hour" +grade_unit = "decimal" +energy_rate_unit = "gallons_gasoline_per_mile" +ideal_energy_rate = 0.02857143 +real_world_energy_adjustment = 1.166 ``` -This TOML section is deserialized into JSON and passed as arguments to the SpeedGradeEnergyModelBuilder which in turn loads the [SpeedGradeModelService] in this crate. -This in turn builds the [SpeedGradeModel]. +This TOML section is deserialized into JSON and passed as arguments to the EnergyModelBuilder which in turn loads the [EnergyModelService] in this crate. +This in turn builds the [EnergyTraversalModel]. ### Search -TraversalModels in this crate will add energy estimation to road network search, and will differ in their dependencies and evaluation procedures. +TraversalModels in this crate will add energy estimation to road network search, and will differ in their dependencies and evaluation procedures. -#### SpeedGradeModel +#### EnergyTraversalModel ##### Dependencies @@ -62,7 +65,7 @@ TraversalModels in this crate will add energy estimation to road network search, 2. compute travel time as `distance / speed` for this edge 3. lookup grade for this edge in table; if missing, use `0.0` 4. perform inference to retrieve energy rate from speed and grade values -5. compute energy as `energy_rate * distance * real_world_adjustment_factor` for this edge +5. compute energy as `energy_rate * distance * real_world_adjustment_factor` for this edge 6. compute link cost as `(energy * energy_cost_coefficient) + (time * (1 - energy_cost_coefficient))` ### Real-World Adjustment Factors @@ -70,13 +73,10 @@ TraversalModels in this crate will add energy estimation to road network search, In addition to calculating the energy based on a RouteE Powertrain output, an adjustment factor should be applied to capture real-world effects of running a powertrain in an environment. As a result of NREL research, some recommended values for this adjustment are: -powertrain type \| factor ---- \| --- -combustion vehicle (CV) \| 1.1660 -hybrid vehicle (HV) \| 1.1252 -electric vehicle (EV) \| 1.3958 +\| powertrain type \| factor \| +\| ----------------------- \| ------ \| +\| combustion vehicle (CV) \| 1.1660 \| +\| hybrid vehicle (HV) \| 1.1252 \| +\| electric vehicle (EV) \| 1.3958 \| A factor of 1.0 equates to 100% of the original energy value. - -[SpeedGradeModelService]: crate::routee::speed_grade_model_service::SpeedGradeModelService -[SpeedGradeModelService]: crate::routee::speed_grade_model::SpeedGradeModel \ No newline at end of file diff --git a/rust/routee-compass-powertrain/src/routee/energy_model_ops.rs b/rust/routee-compass-powertrain/src/routee/energy_model_ops.rs new file mode 100644 index 00000000..52ddc9b4 --- /dev/null +++ b/rust/routee-compass-powertrain/src/routee/energy_model_ops.rs @@ -0,0 +1,26 @@ +use routee_compass_core::{ + model::{road_network::edge_id::EdgeId, traversal::traversal_model_error::TraversalModelError}, + util::unit::Grade, +}; + +pub const ZERO_ENERGY: f64 = 1e-9; + +/// look up the grade from the grade table +pub fn get_grade( + grade_table: &Option<Box<[Grade]>>, + edge_id: EdgeId, +) -> Result<Grade, TraversalModelError> { + match grade_table { + None => Ok(Grade::ZERO), + Some(gt) => { + let grade: &Grade = gt.get(edge_id.as_usize()).ok_or( + TraversalModelError::MissingIdInTabularCostFunction( + format!("{}", edge_id), + String::from("EdgeId"), + String::from("grade table"), + ), + )?; + Ok(grade) + } + } +} diff --git a/rust/routee-compass-powertrain/src/routee/speed_grade_energy_model_service.rs b/rust/routee-compass-powertrain/src/routee/energy_model_service.rs similarity index 80% rename from rust/routee-compass-powertrain/src/routee/speed_grade_energy_model_service.rs rename to rust/routee-compass-powertrain/src/routee/energy_model_service.rs index eccb29d2..fce8d601 100644 --- a/rust/routee-compass-powertrain/src/routee/speed_grade_energy_model_service.rs +++ b/rust/routee-compass-powertrain/src/routee/energy_model_service.rs @@ -1,5 +1,3 @@ -use super::prediction_model::SpeedGradePredictionModelRecord; -use super::speed_grade_energy_model::SpeedGradeEnergyModel; use routee_compass_core::model::traversal::default::speed_lookup_model::get_max_speed; use routee_compass_core::model::traversal::traversal_model::TraversalModel; use routee_compass_core::model::traversal::traversal_model_error::TraversalModelError; @@ -11,8 +9,11 @@ use std::collections::HashMap; use std::path::Path; use std::sync::Arc; +use super::energy_traversal_model::EnergyTraversalModel; +use super::vehicle::VehicleType; + #[derive(Clone)] -pub struct SpeedGradeEnergyModelService { +pub struct EnergyModelService { pub speed_table: Arc<Box<[Speed]>>, pub speeds_table_speed_unit: SpeedUnit, pub max_speed: Speed, @@ -20,21 +21,10 @@ pub struct SpeedGradeEnergyModelService { pub grade_table_grade_unit: GradeUnit, pub output_time_unit: TimeUnit, pub output_distance_unit: DistanceUnit, - pub energy_model_library: Arc<HashMap<String, Arc<SpeedGradePredictionModelRecord>>>, -} - -impl TraversalModelService for SpeedGradeEnergyModelService { - fn build( - &self, - query: &serde_json::Value, - ) -> Result<Arc<dyn TraversalModel>, TraversalModelError> { - let arc_self = Arc::new(self.clone()); - let m = SpeedGradeEnergyModel::try_from((arc_self, query))?; - Ok(Arc::new(m)) - } + pub vehicle_library: HashMap<String, Arc<dyn VehicleType>>, } -impl SpeedGradeEnergyModelService { +impl EnergyModelService { pub fn new<P: AsRef<Path>>( speed_table_path: &P, speeds_table_speed_unit: SpeedUnit, @@ -42,7 +32,7 @@ impl SpeedGradeEnergyModelService { grade_table_grade_unit_option: Option<GradeUnit>, output_time_unit_option: Option<TimeUnit>, output_distance_unit_option: Option<DistanceUnit>, - energy_model_library: HashMap<String, Arc<SpeedGradePredictionModelRecord>>, + vehicle_library: HashMap<String, Arc<dyn VehicleType>>, ) -> Result<Self, TraversalModelError> { let output_time_unit = output_time_unit_option.unwrap_or(BASE_TIME_UNIT); let output_distance_unit = output_distance_unit_option.unwrap_or(BASE_DISTANCE_UNIT); @@ -74,7 +64,7 @@ impl SpeedGradeEnergyModelService { let max_speed = get_max_speed(&speed_table)?; - Ok(SpeedGradeEnergyModelService { + Ok(EnergyModelService { speed_table, speeds_table_speed_unit, max_speed, @@ -82,7 +72,18 @@ impl SpeedGradeEnergyModelService { grade_table_grade_unit, output_time_unit, output_distance_unit, - energy_model_library: Arc::new(energy_model_library), + vehicle_library, }) } } + +impl TraversalModelService for EnergyModelService { + fn build( + &self, + parameters: &serde_json::Value, + ) -> Result<Arc<dyn TraversalModel>, TraversalModelError> { + let arc_self = Arc::new(self.clone()); + let model = EnergyTraversalModel::try_from((arc_self, parameters))?; + Ok(Arc::new(model)) + } +} diff --git a/rust/routee-compass-powertrain/src/routee/speed_grade_energy_model.rs b/rust/routee-compass-powertrain/src/routee/energy_traversal_model.rs similarity index 69% rename from rust/routee-compass-powertrain/src/routee/speed_grade_energy_model.rs rename to rust/routee-compass-powertrain/src/routee/energy_traversal_model.rs index 52d45757..a43dc69b 100644 --- a/rust/routee-compass-powertrain/src/routee/speed_grade_energy_model.rs +++ b/rust/routee-compass-powertrain/src/routee/energy_traversal_model.rs @@ -1,9 +1,12 @@ -use super::prediction_model::SpeedGradePredictionModelRecord; -use super::speed_grade_energy_model_service::SpeedGradeEnergyModelService; +use crate::routee::energy_model_ops::ZERO_ENERGY; + +use super::energy_model_ops::get_grade; +use super::energy_model_service::EnergyModelService; +use super::vehicle::vehicle_type::{VehicleState, VehicleType}; + use routee_compass_core::model::cost::Cost; use routee_compass_core::model::property::edge::Edge; use routee_compass_core::model::property::vertex::Vertex; -use routee_compass_core::model::road_network::edge_id::EdgeId; use routee_compass_core::model::traversal::default::speed_lookup_model::get_speed; use routee_compass_core::model::traversal::state::state_variable::StateVar; use routee_compass_core::model::traversal::state::traversal_state::TraversalState; @@ -15,18 +18,22 @@ use routee_compass_core::util::unit::as_f64::AsF64; use routee_compass_core::util::unit::; use std::sync::Arc; -const ZERO_ENERGY: f64 = 1e-9; - -pub struct SpeedGradeEnergyModel { - pub service: Arc<SpeedGradeEnergyModelService>, - pub model_record: Arc<SpeedGradePredictionModelRecord>, +pub struct EnergyTraversalModel { + pub service: Arc<EnergyModelService>, + pub vehicle: Arc<dyn VehicleType>, pub energy_cost_coefficient: f64, } -impl TraversalModel for SpeedGradeEnergyModel { +impl TraversalModel for EnergyTraversalModel { fn initial_state(&self) -> TraversalState { - // distance, time, energy - vec![StateVar(0.0), StateVar(0.0), StateVar(0.0)] + // distance, time + let mut initial_state = vec![StateVar(0.0), StateVar(0.0)]; + + // vehicle state gets slots 2..n + let vehicle_state = self.vehicle.initial_state(); + initial_state.extend(vehicle_state); + + initial_state } /// estimate the cost of traveling between two vertices. /// given a distance estimate, @@ -47,12 +54,9 @@ impl TraversalModel for SpeedGradeEnergyModel { return Ok(Cost::ZERO); } - let (energy, _energy_unit) = Energy::create( - self.model_record.ideal_energy_rate, - self.model_record.energy_rate_unit, - distance, - self.service.output_distance_unit, - )?; + let (energy, _energy_unit) = self + .vehicle + .best_case_energy((distance, self.service.output_distance_unit))?; let time: Time = Time::create( self.service.max_speed, @@ -85,29 +89,24 @@ impl TraversalModel for SpeedGradeEnergyModel { self.service.output_time_unit.clone(), )?; - let (energy_rate, _energy_rate_unit) = self.model_record.prediction_model.predict( - speed, - self.service.speeds_table_speed_unit, - grade, - self.service.grade_table_grade_unit, + let energy_result = self.vehicle.consume_energy( + (speed, self.service.speeds_table_speed_unit), + (grade, self.service.grade_table_grade_unit), + (distance, self.service.output_distance_unit), + get_vehicle_state_from_state(state), )?; - let energy_rate_real_world = energy_rate * self.model_record.real_world_energy_adjustment; - - let (mut energy, _energy_unit) = Energy::create( - energy_rate_real_world, - self.model_record.energy_rate_unit, - distance, - self.service.output_distance_unit, - )?; + let mut energy = energy_result.energy; + // for now we need to truncate the energy at zero until we can handle these being negative if energy.as_f64() < 0.0 { energy = Energy::new(ZERO_ENERGY); log::debug!("negative energy encountered, setting to 1e-9"); } let total_cost = create_cost(energy, time, self.energy_cost_coefficient); - let updated_state = update_state(state, distance, time, energy); + + let updated_state = update_state(state, distance, time, energy_result.updated_state); let result = TraversalResult { total_cost, updated_state, @@ -118,30 +117,30 @@ impl TraversalModel for SpeedGradeEnergyModel { fn serialize_state(&self, state: &TraversalState) -> serde_json::Value { let distance = get_distance_from_state(state); let time = get_time_from_state(state); - let energy = get_energy_from_state(state); + let vehicle_state = get_vehicle_state_from_state(state); + let vehicle_state_summary = self.vehicle.serialize_state(vehicle_state); serde_json::json!({ "distance": distance, "time": time, - "energy": energy, + "vehicle": vehicle_state_summary, }) } - fn serialize_state_info(&self, _state: &TraversalState) -> serde_json::Value { - let energy_unit = self.model_record.energy_rate_unit.associated_energy_unit(); + fn serialize_state_info(&self, state: &TraversalState) -> serde_json::Value { + let vehicle_state = get_vehicle_state_from_state(state); + let vehicle_state_info = self.vehicle.serialize_state_info(vehicle_state); serde_json::json!({ "distance_unit": self.service.output_distance_unit, "time_unit": self.service.output_time_unit, - "energy_unit": energy_unit, + "vehicle_info": vehicle_state_info, }) } } -impl TryFrom<(Arc<SpeedGradeEnergyModelService>, &serde_json::Value)> for SpeedGradeEnergyModel { +impl TryFrom<(Arc<EnergyModelService>, &serde_json::Value)> for EnergyTraversalModel { type Error = TraversalModelError; - fn try_from( - input: (Arc<SpeedGradeEnergyModelService>, &serde_json::Value), - ) -> Result<Self, Self::Error> { + fn try_from(input: (Arc<EnergyModelService>, &serde_json::Value)) -> Result<Self, Self::Error> { let (service, conf) = input; let energy_cost_coefficient = match conf.get(String::from("energy_cost_coefficient")) { @@ -162,7 +161,6 @@ impl TryFrom<(Arc<SpeedGradeEnergyModelService>, &serde_json::Value)> for SpeedG } } }; - let prediction_model_name = conf .get("model_name".to_string()) .ok_or(TraversalModelError::BuildError( @@ -174,20 +172,21 @@ impl TryFrom<(Arc<SpeedGradeEnergyModelService>, &serde_json::Value)> for SpeedG ))? .to_string(); - let model_record = match service.energy_model_library.get(&prediction_model_name) { + let vehicle = match service.vehicle_library.get(&prediction_model_name) { None => { - let model_names: Vec<&String> = service.energy_model_library.keys().collect(); - return Err(TraversalModelError::BuildError(format!( - "No energy model found with model_name = '{}', try one of: {:?}", + let model_names: Vec<&String> = service.vehicle_library.keys().collect(); + Err(TraversalModelError::BuildError(format!( + "No vehicle found with model_name = '{}', try one of: {:?}", prediction_model_name, model_names - ))); + ))) } - Some(mr) => mr.clone(), - }; + Some(mr) => Ok(mr.clone()), + }? + .update_from_query(conf)?; - Ok(SpeedGradeEnergyModel { + Ok(EnergyTraversalModel { service, - model_record, + vehicle, energy_cost_coefficient, }) } @@ -206,12 +205,13 @@ fn update_state( state: &TraversalState, distance: Distance, time: Time, - energy: Energy, + vehicle_state: VehicleState, ) -> TraversalState { - let mut updated_state = state.clone(); - updated_state[0] = state[0] + distance.into(); - updated_state[1] = state[1] + time.into(); - updated_state[2] = state[2] + energy.into(); + let mut updated_state = Vec::new(); + + updated_state.push(state[0] + distance.into()); + updated_state.push(state[1] + time.into()); + updated_state.extend(vehicle_state); updated_state } @@ -223,33 +223,16 @@ fn get_time_from_state(state: &TraversalState) -> Time { Time::new(state[1].0) } -fn get_energy_from_state(state: &TraversalState) -> Energy { - Energy::new(state[2].0) -} - -/// look up the grade from the grade table -pub fn get_grade( - grade_table: &Option<Box<[Grade]>>, - edge_id: EdgeId, -) -> Result<Grade, TraversalModelError> { - match grade_table { - None => Ok(Grade::ZERO), - Some(gt) => { - let grade: &Grade = gt.get(edge_id.as_usize()).ok_or( - TraversalModelError::MissingIdInTabularCostFunction( - format!("{}", edge_id), - String::from("EdgeId"), - String::from("grade table"), - ), - )?; - Ok(grade) - } - } +fn get_vehicle_state_from_state(state: &TraversalState) -> &[StateVar] { + &state[2..] } #[cfg(test)] mod tests { - use crate::routee::model_type::ModelType; + use crate::routee::{ + prediction::load_prediction_model, prediction::model_type::ModelType, + vehicle::default::single_fuel_vehicle::SingleFuelVehicle, + }; use super::; use geo::coord; @@ -288,7 +271,7 @@ mod tests { distance: Distance::new(100.0), } } - let model_record = SpeedGradePredictionModelRecord::new( + let model_record = load_prediction_model( "Toyota_Camry".to_string(), &model_file_path, ModelType::Smartcore, @@ -299,10 +282,13 @@ mod tests { None, ) .unwrap(); - let mut model_library = HashMap::new(); - model_library.insert("Toyota_Camry".to_string(), Arc::new(model_record)); - let service = SpeedGradeEnergyModelService::new( + let camry = SingleFuelVehicle::new("Toyota_Camry".to_string(), model_record).unwrap(); + + let mut model_library: HashMap<String, Arc<dyn VehicleType>> = HashMap::new(); + model_library.insert("Toyota_Camry".to_string(), Arc::new(camry)); + + let service = EnergyModelService::new( &speed_file_path, SpeedUnit::KilometersPerHour, &Some(grade_file_path), @@ -317,7 +303,7 @@ mod tests { "model_name": "Toyota_Camry", "energy_cost_coefficient": 0.5, }); - let model = SpeedGradeEnergyModel::try_from((arc_service, &conf)).unwrap(); + let model = EnergyTraversalModel::try_from((arc_service, &conf)).unwrap(); let initial = model.initial_state(); let e1 = mock_edge(0); // 100 meters @ 10kph should take 36 seconds ((0.1/10) * 3600) diff --git a/rust/routee-compass-powertrain/src/routee/mod.rs b/rust/routee-compass-powertrain/src/routee/mod.rs index 3b546c57..e3eb32d0 100644 --- a/rust/routee-compass-powertrain/src/routee/mod.rs +++ b/rust/routee-compass-powertrain/src/routee/mod.rs @@ -1,8 +1,5 @@ -pub mod model_type; -pub mod prediction_model; -pub mod smartcore; -pub mod speed_grade_energy_model; -pub mod speed_grade_energy_model_service; - -#[cfg(feature = "onnx")] -pub mod onnx; +pub mod energy_model_ops; +pub mod energy_model_service; +pub mod energy_traversal_model; +pub mod prediction; +pub mod vehicle; diff --git a/rust/routee-compass-powertrain/src/routee/prediction/mod.rs b/rust/routee-compass-powertrain/src/routee/prediction/mod.rs new file mode 100644 index 00000000..05a8a2a4 --- /dev/null +++ b/rust/routee-compass-powertrain/src/routee/prediction/mod.rs @@ -0,0 +1,12 @@ +pub mod model_type; +pub mod prediction_model; +pub mod prediction_model_ops; +pub mod prediction_model_record; +pub mod smartcore; + +#[cfg(feature = "onnx")] +pub mod onnx; + +pub use prediction_model::PredictionModel; +pub use prediction_model_ops::load_prediction_model; +pub use prediction_model_record::PredictionModelRecord; diff --git a/rust/routee-compass-powertrain/src/routee/model_type.rs b/rust/routee-compass-powertrain/src/routee/prediction/model_type.rs similarity index 91% rename from rust/routee-compass-powertrain/src/routee/model_type.rs rename to rust/routee-compass-powertrain/src/routee/prediction/model_type.rs index 1799c9f8..6d2bc210 100644 --- a/rust/routee-compass-powertrain/src/routee/model_type.rs +++ b/rust/routee-compass-powertrain/src/routee/prediction/model_type.rs @@ -1,5 +1,5 @@ use super::{ - prediction_model::SpeedGradePredictionModel, + prediction_model::PredictionModel, smartcore::smartcore_speed_grade_model::SmartcoreSpeedGradeModel, }; use routee_compass_core::{ @@ -34,9 +34,9 @@ impl ModelType { energy_model_speed_unit: SpeedUnit, energy_model_grade_unit: GradeUnit, energy_model_energy_rate_unit: EnergyRateUnit, - ) -> Result<Arc<dyn SpeedGradePredictionModel>, TraversalModelError> { + ) -> Result<Arc<dyn PredictionModel>, TraversalModelError> { // Load random forest binary file - let model: Arc<dyn SpeedGradePredictionModel> = match self { + let model: Arc<dyn PredictionModel> = match self { ModelType::Smartcore => Arc::new(SmartcoreSpeedGradeModel::new( energy_model_path.clone(), energy_model_speed_unit, diff --git a/rust/routee-compass-powertrain/src/routee/onnx/mod.rs b/rust/routee-compass-powertrain/src/routee/prediction/onnx/mod.rs similarity index 100% rename from rust/routee-compass-powertrain/src/routee/onnx/mod.rs rename to rust/routee-compass-powertrain/src/routee/prediction/onnx/mod.rs diff --git a/rust/routee-compass-powertrain/src/routee/onnx/onnx_speed_grade_model.rs b/rust/routee-compass-powertrain/src/routee/prediction/onnx/onnx_speed_grade_model.rs similarity index 87% rename from rust/routee-compass-powertrain/src/routee/onnx/onnx_speed_grade_model.rs rename to rust/routee-compass-powertrain/src/routee/prediction/onnx/onnx_speed_grade_model.rs index c155d925..9025914d 100644 --- a/rust/routee-compass-powertrain/src/routee/onnx/onnx_speed_grade_model.rs +++ b/rust/routee-compass-powertrain/src/routee/prediction/onnx/onnx_speed_grade_model.rs @@ -1,6 +1,6 @@ use std::path::Path; -use crate::routee::prediction_model::SpeedGradePredictionModel; +use crate::routee::prediction::prediction_model::PredictionModel; use ndarray::CowArray; use ort::{ tensor::OrtOwnedTensor, Environment, GraphOptimizationLevel, Session, SessionBuilder, Value, @@ -17,14 +17,14 @@ pub struct OnnxSpeedGradeModel { energy_rate_unit: EnergyRateUnit, } -impl SpeedGradePredictionModel for OnnxSpeedGradeModel { +impl PredictionModel for OnnxSpeedGradeModel { fn predict( &self, - speed: Speed, - speed_unit: SpeedUnit, - grade: Grade, - grade_unit: GradeUnit, + speed: (Speed, SpeedUnit), + grade: (Grade, GradeUnit), ) -> Result<(EnergyRate, EnergyRateUnit), TraversalModelError> { + let (speed, speed_unit) = speed; + let (grade, grade_unit) = grade; let speed_value: f32 = speed_unit.convert(speed, self.speed_unit).as_f64() as f32; let grade_value: f32 = grade_unit.convert(grade, self.grade_unit).as_f64() as f32; let array = ndarray::Array1::from(vec![speed_value, grade_value]) @@ -89,8 +89,7 @@ mod test { use std::path::PathBuf; use crate::routee::{ - onnx::onnx_speed_grade_model::OnnxSpeedGradeModel, - prediction_model::SpeedGradePredictionModel, + prediction::onnx::onnx_speed_grade_model::OnnxSpeedGradeModel, prediction::PredictionModel, }; use rayon::prelude::; use routee_compass_core::{ @@ -106,13 +105,10 @@ mod test { .join("src") .join("routee") .join("test") - .join("Toyota_Camry.onnx") - .to_str() - .unwrap() - .into(); - let model: Box<dyn SpeedGradePredictionModel> = Box::new( + .join("Toyota_Camry.onnx"); + let model: Box<dyn PredictionModel> = Box::new( OnnxSpeedGradeModel::new( - model_file_path, + &model_file_path, SpeedUnit::MilesPerHour, GradeUnit::Decimal, routee_compass_core::util::unit::EnergyRateUnit::GallonsGasolinePerMile, @@ -132,7 +128,7 @@ mod test { let results = inputs .par_iter() .map(\|(speed, speed_unit, grade, grade_unit)\| { - model.predict(speed, speed_unit, grade, grade_unit) + model.predict((speed, speed_unit), (grade, grade_unit)) }) .collect::<Vec<Result<(EnergyRate, EnergyRateUnit), TraversalModelError>>>(); @@ -141,7 +137,10 @@ mod test { // assert that all the results are the same let (expected_er, expected_eru) = model - .predict(input_speed, input_speed_unit, input_grade, input_grade_unit) + .predict( + (input_speed, input_speed_unit), + (input_grade, input_grade_unit), + ) .unwrap(); assert!(results.iter().all(\|r\| match r { Err(e) => panic!("{}", e), diff --git a/rust/routee-compass-powertrain/src/routee/prediction/prediction_model.rs b/rust/routee-compass-powertrain/src/routee/prediction/prediction_model.rs new file mode 100644 index 00000000..166365cc --- /dev/null +++ b/rust/routee-compass-powertrain/src/routee/prediction/prediction_model.rs @@ -0,0 +1,12 @@ +use routee_compass_core::{ + model::traversal::traversal_model_error::TraversalModelError, + util::unit::{EnergyRate, EnergyRateUnit, Grade, GradeUnit, Speed, SpeedUnit}, +}; + +pub trait PredictionModel: Send + Sync { + fn predict( + &self, + speed: (Speed, SpeedUnit), + grade: (Grade, GradeUnit), + ) -> Result<(EnergyRate, EnergyRateUnit), TraversalModelError>; +} diff --git a/rust/routee-compass-powertrain/src/routee/prediction/prediction_model_ops.rs b/rust/routee-compass-powertrain/src/routee/prediction/prediction_model_ops.rs new file mode 100644 index 00000000..f825f724 --- /dev/null +++ b/rust/routee-compass-powertrain/src/routee/prediction/prediction_model_ops.rs @@ -0,0 +1,106 @@ +use std::{path::Path, sync::Arc}; + +use routee_compass_core::{ + model::traversal::traversal_model_error::TraversalModelError, + util::unit::{EnergyRate, EnergyRateUnit, Grade, GradeUnit, Speed, SpeedUnit}, +}; + +use super::{ + model_type::ModelType, smartcore::smartcore_speed_grade_model::SmartcoreSpeedGradeModel, + PredictionModel, PredictionModelRecord, +}; + +#[cfg(feature = "onnx")] +use crate::routee::prediction::onnx::onnx_speed_grade_model::OnnxSpeedGradeModel; + +#[allow(clippy::too_many_arguments)] +pub fn load_prediction_model<P: AsRef<Path>>( + model_name: String, + model_path: &P, + model_type: ModelType, + speed_unit: SpeedUnit, + grade_unit: GradeUnit, + energy_rate_unit: EnergyRateUnit, + ideal_energy_rate_option: Option<EnergyRate>, + real_world_energy_adjustment_option: Option<f64>, +) -> Result<PredictionModelRecord, TraversalModelError> { + let prediction_model: Arc<dyn PredictionModel> = match model_type { + ModelType::Smartcore => { + let model = SmartcoreSpeedGradeModel::new( + model_path, + speed_unit, + grade_unit, + energy_rate_unit, + )?; + Arc::new(model) + } + ModelType::Onnx => { + #[cfg(feature = "onnx")] + { + let model = + OnnxSpeedGradeModel::new(model_path, speed_unit, grade_unit, energy_rate_unit)?; + Arc::new(model) + } + #[cfg(not(feature = "onnx"))] + { + return Err(TraversalModelError::BuildError( + "Cannot build Onnx model without `onnx` feature enabled for compass-powertrain" + .to_string(), + )); + } + } + }; + let ideal_energy_rate = match ideal_energy_rate_option { + None => find_min_energy_rate(&prediction_model, &energy_rate_unit)?, + Some(ier) => ier, + }; + + let real_world_energy_adjustment = real_world_energy_adjustment_option.unwrap_or(1.0); + + Ok(PredictionModelRecord { + name: model_name, + prediction_model, + model_type, + speed_unit, + grade_unit, + energy_rate_unit, + ideal_energy_rate, + real_world_energy_adjustment, + }) +} + +/// sweep a fixed set of speed and grade values to find the minimum energy per mile rate from the incoming rf model +pub fn find_min_energy_rate( + model: &Arc<dyn PredictionModel>, + energy_model_energy_rate_unit: &EnergyRateUnit, +) -> Result<EnergyRate, TraversalModelError> { + // sweep a fixed set of speed and grade values to find the minimum energy per mile rate from the incoming rf model + let mut minimum_energy_rate = EnergyRate::new(f64::MAX); + let start_time = std::time::Instant::now(); + + let grade = Grade::ZERO; + for speed_i32 in 20..80 { + let speed = Speed::new(speed_i32 as f64); + let (energy_rate, _) = model + .predict( + (speed, SpeedUnit::MilesPerHour), + (grade, GradeUnit::Percent), + ) + .map_err(\|e\| TraversalModelError::PredictionModel(e.to_string()))?; + if energy_rate < minimum_energy_rate { + minimum_energy_rate = energy_rate; + } + } + + let end_time = std::time::Instant::now(); + let search_time = end_time - start_time; + + log::debug!( + "found minimum energy: {}/{} in {} milliseconds", + minimum_energy_rate, + energy_model_energy_rate_unit, + search_time.as_millis() + ); + + Ok(minimum_energy_rate) +} diff --git a/rust/routee-compass-powertrain/src/routee/prediction/prediction_model_record.rs b/rust/routee-compass-powertrain/src/routee/prediction/prediction_model_record.rs new file mode 100644 index 00000000..1e97490b --- /dev/null +++ b/rust/routee-compass-powertrain/src/routee/prediction/prediction_model_record.rs @@ -0,0 +1,44 @@ +use std::sync::Arc; + +use routee_compass_core::{ + model::traversal::traversal_model_error::TraversalModelError, + util::unit::{ + Distance, DistanceUnit, Energy, EnergyRate, EnergyRateUnit, EnergyUnit, Grade, GradeUnit, + Speed, SpeedUnit, + }, +}; + +use super::{model_type::ModelType, PredictionModel}; +/// A struct to hold the prediction model and associated metadata +pub struct PredictionModelRecord { + pub name: String, + pub prediction_model: Arc<dyn PredictionModel>, + pub model_type: ModelType, + pub speed_unit: SpeedUnit, + pub grade_unit: GradeUnit, + pub energy_rate_unit: EnergyRateUnit, + pub ideal_energy_rate: EnergyRate, + pub real_world_energy_adjustment: f64, +} + +impl PredictionModelRecord { + pub fn predict( + &self, + speed: (Speed, SpeedUnit), + grade: (Grade, GradeUnit), + distance: (Distance, DistanceUnit), + ) -> Result<(Energy, EnergyUnit), TraversalModelError> { + let (distance, distance_unit) = distance; + let (energy_rate, _energy_rate_unit) = self.prediction_model.predict(speed, grade)?; + + let energy_rate_real_world = energy_rate self.real_world_energy_adjustment; + + let (energy, energy_unit) = Energy::create( + energy_rate_real_world, + self.energy_rate_unit, + distance, + distance_unit, + )?; + Ok((energy, energy_unit)) + } +} diff --git a/rust/routee-compass-powertrain/src/routee/smartcore/mod.rs b/rust/routee-compass-powertrain/src/routee/prediction/smartcore/mod.rs similarity index 100% rename from rust/routee-compass-powertrain/src/routee/smartcore/mod.rs rename to rust/routee-compass-powertrain/src/routee/prediction/smartcore/mod.rs diff --git a/rust/routee-compass-powertrain/src/routee/smartcore/smartcore_speed_grade_model.rs b/rust/routee-compass-powertrain/src/routee/prediction/smartcore/smartcore_speed_grade_model.rs similarity index 88% rename from rust/routee-compass-powertrain/src/routee/smartcore/smartcore_speed_grade_model.rs rename to rust/routee-compass-powertrain/src/routee/prediction/smartcore/smartcore_speed_grade_model.rs index 6ce8d916..dede8ef4 100644 --- a/rust/routee-compass-powertrain/src/routee/smartcore/smartcore_speed_grade_model.rs +++ b/rust/routee-compass-powertrain/src/routee/prediction/smartcore/smartcore_speed_grade_model.rs @@ -1,6 +1,6 @@ use std::path::Path; -use crate::routee::prediction_model::SpeedGradePredictionModel; +use crate::routee::prediction::prediction_model::PredictionModel; use routee_compass_core::{ model::traversal::traversal_model_error::TraversalModelError, util::unit::{as_f64::AsF64, EnergyRate, EnergyRateUnit, Grade, GradeUnit, Speed, SpeedUnit}, @@ -16,14 +16,14 @@ pub struct SmartcoreSpeedGradeModel { energy_rate_unit: EnergyRateUnit, } -impl SpeedGradePredictionModel for SmartcoreSpeedGradeModel { +impl PredictionModel for SmartcoreSpeedGradeModel { fn predict( &self, - speed: Speed, - speed_unit: SpeedUnit, - grade: Grade, - grade_unit: GradeUnit, + speed: (Speed, SpeedUnit), + grade: (Grade, GradeUnit), ) -> Result<(EnergyRate, EnergyRateUnit), TraversalModelError> { + let (speed, speed_unit) = speed; + let (grade, grade_unit) = grade; let speed_value = speed_unit.convert(speed, self.speed_unit).as_f64(); let grade_value = grade_unit.convert(grade, self.grade_unit).as_f64(); let x = DenseMatrix::from_2d_vec(&vec![vec![speed_value, grade_value]]); diff --git a/rust/routee-compass-powertrain/src/routee/prediction_model.rs b/rust/routee-compass-powertrain/src/routee/prediction_model.rs deleted file mode 100644 index dcc7f18e..00000000 --- a/rust/routee-compass-powertrain/src/routee/prediction_model.rs +++ /dev/null @@ -1,132 +0,0 @@ -use std::{path::Path, sync::Arc}; - -use routee_compass_core::{ - model::traversal::traversal_model_error::TraversalModelError, - util::unit::{EnergyRate, EnergyRateUnit, Grade, GradeUnit, Speed, SpeedUnit}, -}; - -use super::{ - model_type::ModelType, smartcore::smartcore_speed_grade_model::SmartcoreSpeedGradeModel, -}; - -#[cfg(feature = "onnx")] -use crate::routee::onnx::onnx_speed_grade_model::OnnxSpeedGradeModel; - -pub struct SpeedGradePredictionModelRecord { - pub name: String, - pub prediction_model: Arc<dyn SpeedGradePredictionModel>, - pub model_type: ModelType, - pub speed_unit: SpeedUnit, - pub grade_unit: GradeUnit, - pub energy_rate_unit: EnergyRateUnit, - pub ideal_energy_rate: EnergyRate, - pub real_world_energy_adjustment: f64, -} - -impl SpeedGradePredictionModelRecord { - #[allow(clippy::too_many_arguments)] - pub fn new<P: AsRef<Path>>( - name: String, - model_path: &P, - model_type: ModelType, - speed_unit: SpeedUnit, - grade_unit: GradeUnit, - energy_rate_unit: EnergyRateUnit, - ideal_energy_rate_option: Option<EnergyRate>, - real_world_energy_adjustment_option: Option<f64>, - ) -> Result<Self, TraversalModelError> { - // Load random forest binary file - let prediction_model: Arc<dyn SpeedGradePredictionModel> = match model_type { - ModelType::Smartcore => { - let model = SmartcoreSpeedGradeModel::new( - model_path, - speed_unit, - grade_unit, - energy_rate_unit, - )?; - Arc::new(model) - } - ModelType::Onnx => { - #[cfg(feature = "onnx")] - { - let model = OnnxSpeedGradeModel::new( - model_path, - speed_unit, - grade_unit, - energy_rate_unit, - )?; - Arc::new(model) - } - #[cfg(not(feature = "onnx"))] - { - return Err( - TraversalModelError::BuildError( - "Cannot build Onnx model without `onnx` feature enabled for compass-powertrain" - .to_string(), - ) - ); - } - } - }; - let ideal_energy_rate = match ideal_energy_rate_option { - None => find_min_energy_rate(&prediction_model, &energy_rate_unit)?, - Some(ier) => ier, - }; - - let real_world_energy_adjustment = real_world_energy_adjustment_option.unwrap_or(1.0); - - Ok(Self { - name, - prediction_model, - model_type, - speed_unit, - grade_unit, - energy_rate_unit, - ideal_energy_rate, - real_world_energy_adjustment, - }) - } -} - -pub trait SpeedGradePredictionModel: Send + Sync { - fn predict( - &self, - speed: Speed, - speed_unit: SpeedUnit, - grade: Grade, - grade_unit: GradeUnit, - ) -> Result<(EnergyRate, EnergyRateUnit), TraversalModelError>; -} - -/// sweep a fixed set of speed and grade values to find the minimum energy per mile rate from the incoming rf model -pub fn find_min_energy_rate( - model: &Arc<dyn SpeedGradePredictionModel>, - energy_model_energy_rate_unit: &EnergyRateUnit, -) -> Result<EnergyRate, TraversalModelError> { - // sweep a fixed set of speed and grade values to find the minimum energy per mile rate from the incoming rf model - let mut minimum_energy_rate = EnergyRate::new(f64::MAX); - let start_time = std::time::Instant::now(); - - let grade = Grade::ZERO; - for speed_i32 in 20..80 { - let speed = Speed::new(speed_i32 as f64); - let (energy_rate, _) = model - .predict(speed, SpeedUnit::MilesPerHour, grade, GradeUnit::Percent) - .map_err(\|e\| TraversalModelError::PredictionModel(e.to_string()))?; - if energy_rate < minimum_energy_rate { - minimum_energy_rate = energy_rate; - } - } - - let end_time = std::time::Instant::now(); - let search_time = end_time - start_time; - - log::debug!( - "found minimum energy: {}/{} in {} milliseconds", - minimum_energy_rate, - energy_model_energy_rate_unit, - search_time.as_millis() - ); - - Ok(minimum_energy_rate) -} diff --git a/rust/routee-compass-powertrain/src/routee/vehicle/default/dual_fuel_vehicle.rs b/rust/routee-compass-powertrain/src/routee/vehicle/default/dual_fuel_vehicle.rs new file mode 100644 index 00000000..33b41057 --- /dev/null +++ b/rust/routee-compass-powertrain/src/routee/vehicle/default/dual_fuel_vehicle.rs @@ -0,0 +1,365 @@ +use std::sync::Arc; + +use routee_compass_core::{ + model::traversal::{ + state::{state_variable::StateVar, traversal_state::TraversalState}, + traversal_model_error::TraversalModelError, + }, + util::unit::{ + as_f64::AsF64, Distance, DistanceUnit, Energy, EnergyUnit, Grade, GradeUnit, Speed, + SpeedUnit, + }, +}; + +use crate::routee::{ + prediction::PredictionModelRecord, + vehicle::{vehicle_type::VehicleType, VehicleEnergyResult}, +}; + +pub struct DualFuelVehicle { + pub name: String, + pub charge_sustain_model: Arc<PredictionModelRecord>, + pub charge_depleting_model: Arc<PredictionModelRecord>, + pub battery_capacity: Energy, + pub starting_battery_energy: Energy, + pub battery_energy_unit: EnergyUnit, +} + +impl DualFuelVehicle { + pub fn new( + name: String, + charge_sustain_model: PredictionModelRecord, + charge_depleting_model: PredictionModelRecord, + battery_capacity: Energy, + starting_battery_energy: Energy, + battery_energy_unit: EnergyUnit, + ) -> Result<Self, TraversalModelError> { + Ok(Self { + name, + charge_sustain_model: Arc::new(charge_sustain_model), + charge_depleting_model: Arc::new(charge_depleting_model), + battery_capacity, + starting_battery_energy, + battery_energy_unit, + }) + } +} + +impl VehicleType for DualFuelVehicle { + fn name(&self) -> String { + self.name.clone() + } + fn initial_state(&self) -> TraversalState { + vec![ + StateVar(0.0), // accumulated electrical energy + StateVar(0.0), // accumulated gasoline energy + StateVar(self.starting_battery_energy.as_f64()), // battery energy remaining + ] + } + fn best_case_energy( + &self, + distance: (Distance, DistanceUnit), + ) -> Result<(Energy, EnergyUnit), TraversalModelError> { + let (distance, distance_unit) = distance; + + // assume lowest energy cost scenario for a PHEV is to just use the battery + let energy = Energy::create( + self.charge_depleting_model.ideal_energy_rate, + self.charge_depleting_model.energy_rate_unit, + distance, + distance_unit, + )?; + Ok(energy) + } + fn consume_energy( + &self, + speed: (Speed, SpeedUnit), + grade: (Grade, GradeUnit), + distance: (Distance, DistanceUnit), + state: &[StateVar], + ) -> Result<VehicleEnergyResult, TraversalModelError> { + let battery_soc_percentage = get_battery_soc_percent(self, state); + + let (electrical_energy, electrical_energy_unit, gasoline_energy, gasoline_energy_unit) = + get_phev_energy(self, battery_soc_percentage, speed, grade, distance)?; + + // convert both energy sources to kWh + let electrical_energy_kwh = + electrical_energy_unit.convert(electrical_energy, EnergyUnit::KilowattHours); + let gasoline_energy_kwh = + gasoline_energy_unit.convert(gasoline_energy, EnergyUnit::KilowattHours); + let total_energy_kwh = electrical_energy_kwh + gasoline_energy_kwh; + + let updated_state = update_state(state, electrical_energy, gasoline_energy); + + Ok(VehicleEnergyResult { + energy: total_energy_kwh, + energy_unit: EnergyUnit::KilowattHours, + updated_state, + }) + } + fn serialize_state(&self, state: &[StateVar]) -> serde_json::Value { + let battery_energy = get_electrical_energy_from_state(state); + let gasoline_energy = get_gasoline_energy_from_state(state); + let battery_soc_percent = get_battery_soc_percent(self, state); + serde_json::json!({ + "battery_energy": battery_energy.as_f64(), + "fuel_energy": gasoline_energy.as_f64(), + "battery_soc_percent": battery_soc_percent, + }) + } + + fn serialize_state_info(&self, _state: &[StateVar]) -> serde_json::Value { + let battery_energy_unit = self.battery_energy_unit; + let fuel_energy_unit = self + .charge_sustain_model + .energy_rate_unit + .associated_energy_unit(); + serde_json::json!({ + "battery_energy_unit": battery_energy_unit.to_string(), + "fuel_energy_unit": fuel_energy_unit.to_string(), + }) + } + + fn update_from_query( + &self, + query: &serde_json::Value, + ) -> Result<Arc<dyn VehicleType>, TraversalModelError> { + let starting_soc_percent = query + .get("starting_soc_percent".to_string()) + .ok_or(TraversalModelError::BuildError( + "No 'starting_soc_percent' key provided in query".to_string(), + ))? + .as_f64() + .ok_or(TraversalModelError::BuildError( + "Expected 'starting_soc_percent' value to be numeric".to_string(), + ))?; + if !(0.0..=100.0).contains(&starting_soc_percent) { + return Err(TraversalModelError::BuildError( + "Expected 'starting_soc_percent' value to be between 0 and 100".to_string(), + )); + } + let starting_battery_energy = self.battery_capacity * (starting_soc_percent / 100.0); + + let new_phev = DualFuelVehicle { + name: self.name.clone(), + charge_sustain_model: self.charge_sustain_model.clone(), + charge_depleting_model: self.charge_depleting_model.clone(), + battery_capacity: self.battery_capacity, + starting_battery_energy, + battery_energy_unit: self.battery_energy_unit, + }; + + Ok(Arc::new(new_phev)) + } +} + +fn update_state( + state: &[StateVar], + electrical_energy: Energy, + gasoline_energy: Energy, +) -> TraversalState { + let mut updated_state = Vec::with_capacity(state.len()); + + // accumulated electrical energy + updated_state.push(state[0] + electrical_energy.into()); + + // accumulated fuel energy + updated_state.push(state[1] + gasoline_energy.into()); + + // remaining battery energy + let current_battery_energy = get_remaining_battery_energy_from_state(state); + let new_battery_energy = (current_battery_energy - electrical_energy).max(Energy::new(0.0)); + updated_state.push(new_battery_energy.into()); + + updated_state +} +fn get_electrical_energy_from_state(state: &[StateVar]) -> Energy { + Energy::new(state[0].0) +} + +fn get_gasoline_energy_from_state(state: &[StateVar]) -> Energy { + Energy::new(state[1].0) +} + +fn get_remaining_battery_energy_from_state(state: &[StateVar]) -> Energy { + Energy::new(state[2].0) +} + +fn get_battery_soc_percent(vehicle: &DualFuelVehicle, state: &[StateVar]) -> f64 { + let battery_energy_unit = vehicle.battery_energy_unit; + + let battery_capacity_kwh = + battery_energy_unit.convert(vehicle.battery_capacity, EnergyUnit::KilowattHours); + + let remaining_battery_energy = get_remaining_battery_energy_from_state(state); + + let remaining_battery_energy_kwh = + battery_energy_unit.convert(remaining_battery_energy, EnergyUnit::KilowattHours); + + (remaining_battery_energy_kwh.as_f64() / battery_capacity_kwh.as_f64()) * 100.0 +} + +/// Compute the energy for the PHEV by converting gasoline to kWh. +/// This uses a simplified operation in which we assume that if the battery +/// SOC is greater than zero we can just operate on battery to traverse a link. +/// This is not entirely realistic as it's possible to arrive at a link with +/// 0.001% SOC and still need to use gasoline to traverse the link. +/// +/// In the future we could make this more sophisticated by calculating +/// the energy required to traverse the link using the battery and then +/// finding the point at which we would have to switch to gasoline +/// +/// Returns a tuple of (electrical_energy, electrical_energy_unit, gasoline_energy, gasoline_energy_unit) +fn get_phev_energy( + vehicle: &DualFuelVehicle, + battery_soc_percent: f64, + speed: (Speed, SpeedUnit), + grade: (Grade, GradeUnit), + distance: (Distance, DistanceUnit), +) -> Result<(Energy, EnergyUnit, Energy, EnergyUnit), TraversalModelError> { + let electrical_energy_unit = vehicle + .charge_depleting_model + .energy_rate_unit + .associated_energy_unit(); + let gasoline_energy_unit = vehicle + .charge_sustain_model + .energy_rate_unit + .associated_energy_unit(); + + if battery_soc_percent > 0.0 { + // assume we can just use the battery + let (electrical_energy, electrical_energy_unit) = vehicle + .charge_depleting_model + .predict(speed, grade, distance)?; + Ok(( + electrical_energy, + electrical_energy_unit, + Energy::new(0.0), + gasoline_energy_unit, + )) + } else { + // just use the gasoline engine + let (gasoline_energy, gasoline_energy_unit) = vehicle + .charge_sustain_model + .predict(speed, grade, distance)?; + Ok(( + Energy::new(0.0), + electrical_energy_unit, + gasoline_energy, + gasoline_energy_unit, + )) + } +} + +#[cfg(test)] +mod tests { + use routee_compass_core::util::unit::{EnergyRate, EnergyRateUnit}; + + use crate::routee::{prediction::load_prediction_model, prediction::model_type::ModelType}; + + use super::; + + use std::path::PathBuf; + + fn mock_vehicle() -> DualFuelVehicle { + let charge_sustain_model_file_path = PathBuf::from(env!("CARGO_MANIFEST_DIR")) + .join("src") + .join("routee") + .join("test") + .join("2016_CHEVROLET_Volt_Charge_Sustaining.bin"); + let charge_depleting_model_file_path = PathBuf::from(env!("CARGO_MANIFEST_DIR")) + .join("src") + .join("routee") + .join("test") + .join("2016_CHEVROLET_Volt_Charge_Depleting.bin"); + + let charge_sustain_model_record = load_prediction_model( + "Chevy_Volt_Charge_Sustaining".to_string(), + &charge_sustain_model_file_path, + ModelType::Smartcore, + SpeedUnit::MilesPerHour, + GradeUnit::Decimal, + EnergyRateUnit::GallonsGasolinePerMile, + Some(EnergyRate::new(0.02)), + Some(1.1252), + ) + .unwrap(); + let charge_depleting_model_record = load_prediction_model( + "Chevy_Volt_Charge_Depleting".to_string(), + &charge_depleting_model_file_path, + ModelType::Smartcore, + SpeedUnit::MilesPerHour, + GradeUnit::Decimal, + EnergyRateUnit::KilowattHoursPerMile, + Some(EnergyRate::new(0.2)), + Some(1.3958), + ) + .unwrap(); + + DualFuelVehicle::new( + "Chevy_Volt".to_string(), + charge_sustain_model_record, + charge_depleting_model_record, + Energy::new(12.0), + Energy::new(12.0), + EnergyUnit::KilowattHours, + ) + .unwrap() + } + + #[test] + fn test_phev_energy_model_just_electric() { + let vehicle = mock_vehicle(); + let initial = vehicle.initial_state(); + + // starting at 100% SOC, we should be able to traverse 1000 meters + // without using any gasoline + let distance = (Distance::new(1000.0), DistanceUnit::Meters); + let speed = (Speed::new(60.0), SpeedUnit::MilesPerHour); + let grade = (Grade::new(0.0), GradeUnit::Decimal); + + let result = vehicle + .consume_energy(speed, grade, distance, &initial) + .unwrap(); + + let gasoline_energy = get_gasoline_energy_from_state(&result.updated_state); + assert!(gasoline_energy.as_f64() < 1e-9); + + let electrical_energy = get_electrical_energy_from_state(&result.updated_state); + assert!(electrical_energy.as_f64() > 0.0); + + let battery_percent_soc = get_battery_soc_percent(&vehicle, &result.updated_state); + assert!(battery_percent_soc < 100.0); + } + + #[test] + fn test_phev_energy_model_gas_and_electric() { + let vehicle = mock_vehicle(); + let initial = vehicle.initial_state(); + + // now let's traverse a really long link to deplete the battery + let distance = (Distance::new(100.0), DistanceUnit::Miles); + let speed = (Speed::new(60.0), SpeedUnit::MilesPerHour); + let grade = (Grade::new(0.0), GradeUnit::Decimal); + + let result = vehicle + .consume_energy(speed, grade, distance, &initial) + .unwrap(); + + let electrical_energy = get_electrical_energy_from_state(&result.updated_state); + let battery_percent_soc = get_battery_soc_percent(&vehicle, &result.updated_state); + + assert!(electrical_energy.as_f64() > 0.0); + assert!(battery_percent_soc < 1e-9); + + // and then traverse the same distance but this time we should only use gasoline energy + let result2 = vehicle + .consume_energy(speed, grade, distance, &result.updated_state) + .unwrap(); + + let gasoline_energy = get_gasoline_energy_from_state(&result2.updated_state); + + assert!(gasoline_energy.as_f64() > 0.0); + } +} diff --git a/rust/routee-compass-powertrain/src/routee/vehicle/default/mod.rs b/rust/routee-compass-powertrain/src/routee/vehicle/default/mod.rs new file mode 100644 index 00000000..daa93c3e --- /dev/null +++ b/rust/routee-compass-powertrain/src/routee/vehicle/default/mod.rs @@ -0,0 +1,2 @@ +pub mod dual_fuel_vehicle; +pub mod single_fuel_vehicle; diff --git a/rust/routee-compass-powertrain/src/routee/vehicle/default/single_fuel_vehicle.rs b/rust/routee-compass-powertrain/src/routee/vehicle/default/single_fuel_vehicle.rs new file mode 100644 index 00000000..d270e8a7 --- /dev/null +++ b/rust/routee-compass-powertrain/src/routee/vehicle/default/single_fuel_vehicle.rs @@ -0,0 +1,113 @@ +use std::sync::Arc; + +use routee_compass_core::{ + model::traversal::{ + state::state_variable::StateVar, traversal_model_error::TraversalModelError, + }, + util::unit::{ + as_f64::AsF64, Distance, DistanceUnit, Energy, EnergyUnit, Grade, GradeUnit, Speed, + SpeedUnit, + }, +}; + +use crate::routee::{ + prediction::PredictionModelRecord, + vehicle::{VehicleEnergyResult, VehicleState, VehicleType}, +}; + +pub struct SingleFuelVehicle { + pub name: String, + pub prediction_model_record: Arc<PredictionModelRecord>, +} + +impl SingleFuelVehicle { + pub fn new( + name: String, + prediction_model_record: PredictionModelRecord, + ) -> Result<Self, TraversalModelError> { + Ok(Self { + name, + prediction_model_record: Arc::new(prediction_model_record), + }) + } +} + +impl VehicleType for SingleFuelVehicle { + fn name(&self) -> String { + self.name.clone() + } + fn initial_state(&self) -> VehicleState { + // accumulated energy + vec![StateVar(0.0)] + } + fn best_case_energy( + &self, + distance: (Distance, DistanceUnit), + ) -> Result<(Energy, EnergyUnit), TraversalModelError> { + let (distance, distance_unit) = distance; + let energy = Energy::create( + self.prediction_model_record.ideal_energy_rate, + self.prediction_model_record.energy_rate_unit, + distance, + distance_unit, + )?; + Ok(energy) + } + fn consume_energy( + &self, + speed: (Speed, SpeedUnit), + grade: (Grade, GradeUnit), + distance: (Distance, DistanceUnit), + state: &[StateVar], + ) -> Result<VehicleEnergyResult, TraversalModelError> { + let (energy, energy_unit) = self + .prediction_model_record + .predict(speed, grade, distance)?; + + let updated_state = update_state(state, energy); + + Ok(VehicleEnergyResult { + energy, + energy_unit, + updated_state, + }) + } + fn serialize_state(&self, state: &[StateVar]) -> serde_json::Value { + let energy = get_energy_from_state(state); + serde_json::json!({ + "energy": energy.as_f64(), + }) + } + + fn serialize_state_info(&self, _state: &[StateVar]) -> serde_json::Value { + let energy_unit = self + .prediction_model_record + .energy_rate_unit + .associated_energy_unit(); + serde_json::json!({ + "energy_unit": energy_unit.to_string(), + }) + } + + fn update_from_query( + &self, + _query: &serde_json::Value, + ) -> Result<Arc<dyn VehicleType>, TraversalModelError> { + // just return a clone of self + Ok(Arc::new(SingleFuelVehicle { + name: self.name.clone(), + prediction_model_record: self.prediction_model_record.clone(), + })) + } +} + +fn update_state(state: &[StateVar], energy: Energy) -> VehicleState { + let mut new_state = Vec::with_capacity(state.len()); + new_state.push(state[0] + energy.into()); + new_state +} + +fn get_energy_from_state(state: &[StateVar]) -> Energy { + let energy = state[0].0; + Energy::new(energy) +} diff --git a/rust/routee-compass-powertrain/src/routee/vehicle/mod.rs b/rust/routee-compass-powertrain/src/routee/vehicle/mod.rs new file mode 100644 index 00000000..ef315eec --- /dev/null +++ b/rust/routee-compass-powertrain/src/routee/vehicle/mod.rs @@ -0,0 +1,6 @@ +pub mod default; +pub mod vehicle_energy_result; +pub mod vehicle_type; + +pub use vehicle_energy_result::VehicleEnergyResult; +pub use vehicle_type::{VehicleState, VehicleType}; diff --git a/rust/routee-compass-powertrain/src/routee/vehicle/vehicle_energy_result.rs b/rust/routee-compass-powertrain/src/routee/vehicle/vehicle_energy_result.rs new file mode 100644 index 00000000..f4fcb9f1 --- /dev/null +++ b/rust/routee-compass-powertrain/src/routee/vehicle/vehicle_energy_result.rs @@ -0,0 +1,9 @@ +use routee_compass_core::util::unit::{Energy, EnergyUnit}; + +use super::VehicleState; + +pub struct VehicleEnergyResult { + pub energy: Energy, + pub energy_unit: EnergyUnit, + pub updated_state: VehicleState, +} diff --git a/rust/routee-compass-powertrain/src/routee/vehicle/vehicle_type.rs b/rust/routee-compass-powertrain/src/routee/vehicle/vehicle_type.rs new file mode 100644 index 00000000..4d2ec107 --- /dev/null +++ b/rust/routee-compass-powertrain/src/routee/vehicle/vehicle_type.rs @@ -0,0 +1,82 @@ +use std::sync::Arc; + +use routee_compass_core::{ + model::traversal::{ + state::state_variable::StateVar, traversal_model_error::TraversalModelError, + }, + util::unit::{Distance, DistanceUnit, Energy, EnergyUnit, Grade, GradeUnit, Speed, SpeedUnit}, +}; + +use super::VehicleEnergyResult; + +pub type VehicleState = Vec<StateVar>; + +/// A Vehicle Type represents a class of vehicles with a specific operating model. +pub trait VehicleType: Send + Sync { + /// Return the name of the vehicle type + fn name(&self) -> String; + + /// Return the energy required to travel a certain distance at a certain speed and grade. + /// + /// Arguments: + /// `speed` - The speed at which the vehicle is traveling + /// * `grade` - The grade of the road + /// * `distance` - The distance traveled + /// * `state` - The state of the vehicle + /// + /// Returns: + /// * `VehicleEnergyResult` - The energy required + fn consume_energy( + &self, + speed: (Speed, SpeedUnit), + grade: (Grade, GradeUnit), + distance: (Distance, DistanceUnit), + state: &[StateVar], + ) -> Result<VehicleEnergyResult, TraversalModelError>; + + /// Return the best case scenario for traveling a certain distance. + /// This is used in the a-star algorithm as a distance heuristic. + /// + /// Arguments: + /// * `distance` - The distance traveled + /// + /// Returns: + /// * `Energy` - The 'best case' energy required to travel the distance + fn best_case_energy( + &self, + distance: (Distance, DistanceUnit), + ) -> Result<(Energy, EnergyUnit), TraversalModelError>; + + /// Return the initial state of the vehicle + fn initial_state(&self) -> VehicleState; + + /// Serialize the state of the vehicle into JSON + /// + /// Arguments: + /// * `state` - The state of the vehicle + /// + /// Returns: + /// * `serde_json::Value` - The serialized state + fn serialize_state(&self, state: &[StateVar]) -> serde_json::Value; + + /// Serialize any supplemental state information (like units) into JSON + /// + /// Arguments: + /// * `state` - The state of the vehicle + /// + /// Returns: + /// * `serde_json::Value` - The serialized state information + fn serialize_state_info(&self, state: &[StateVar]) -> serde_json::Value; + + /// Give the vehicle a chance to update itself from the incoming query + /// + /// Arguments: + /// * `query` - The incoming query + /// + /// Returns: + /// * `Arc<dyn VehicleType>` - The updated vehicle type + fn update_from_query( + &self, + query: &serde_json::Value, + ) -> Result<Arc<dyn VehicleType>, TraversalModelError>; +} diff --git a/rust/routee-compass/src/app/compass/config/compass_app_builder.rs b/rust/routee-compass/src/app/compass/config/compass_app_builder.rs index bbe894c3..329f3a7c 100644 --- a/rust/routee-compass/src/app/compass/config/compass_app_builder.rs +++ b/rust/routee-compass/src/app/compass/config/compass_app_builder.rs @@ -7,8 +7,7 @@ use super::{ no_restriction_builder::NoRestrictionBuilder, road_class_builder::RoadClassBuilder, }, traversal_model::{ - distance_builder::DistanceBuilder, - speed_grade_energy_model_builder::SpeedGradeEnergyModelBuilder, + distance_builder::DistanceBuilder, energy_model_builder::EnergyModelBuilder, speed_lookup_builder::SpeedLookupBuilder, }, }; @@ -113,11 +112,11 @@ impl CompassAppBuilder { // Traversal model builders let dist: Box<dyn TraversalModelBuilder> = Box::new(DistanceBuilder {}); let velo: Box<dyn TraversalModelBuilder> = Box::new(SpeedLookupBuilder {}); - let smartcore: Box<dyn TraversalModelBuilder> = Box::new(SpeedGradeEnergyModelBuilder {}); + let energy_model: Box<dyn TraversalModelBuilder> = Box::new(EnergyModelBuilder {}); let tm_builders: HashMap<String, Box<dyn TraversalModelBuilder>> = HashMap::from([ (String::from("distance"), dist), (String::from("speed_table"), velo), - (String::from("speed_grade_energy_model"), smartcore), + (String::from("energy_model"), energy_model), ]); // Frontier model builders diff --git a/rust/routee-compass/src/app/compass/config/compass_configuration_field.rs b/rust/routee-compass/src/app/compass/config/compass_configuration_field.rs index e0d0a80d..09c248ec 100644 --- a/rust/routee-compass/src/app/compass/config/compass_configuration_field.rs +++ b/rust/routee-compass/src/app/compass/config/compass_configuration_field.rs @@ -13,6 +13,8 @@ pub enum CompassConfigurationField { Parallelism, QueryTimeoutMs, IncludeTree, + ChargeDepleting, + ChargeSustaining, } impl CompassConfigurationField { @@ -29,6 +31,8 @@ impl CompassConfigurationField { CompassConfigurationField::Plugins => "plugin", CompassConfigurationField::InputPlugins => "input_plugins", CompassConfigurationField::OutputPlugins => "output_plugins", + CompassConfigurationField::ChargeDepleting => "charge_depleting", + CompassConfigurationField::ChargeSustaining => "charge_sustaining", } } } diff --git a/rust/routee-compass/src/app/compass/config/traversal_model/energy_model_builder.rs b/rust/routee-compass/src/app/compass/config/traversal_model/energy_model_builder.rs new file mode 100644 index 00000000..c20ba155 --- /dev/null +++ b/rust/routee-compass/src/app/compass/config/traversal_model/energy_model_builder.rs @@ -0,0 +1,96 @@ +use std::collections::HashMap; +use std::sync::Arc; + +use crate::app::compass::config::compass_configuration_field::CompassConfigurationField; +use crate::app::compass::config::config_json_extension::ConfigJsonExtensions; +use routee_compass_core::model::traversal::traversal_model_builder::TraversalModelBuilder; +use routee_compass_core::model::traversal::traversal_model_error::TraversalModelError; +use routee_compass_core::model::traversal::traversal_model_service::TraversalModelService; +use routee_compass_core::util::unit::{DistanceUnit, GradeUnit, SpeedUnit, TimeUnit}; +use routee_compass_powertrain::routee::energy_model_service::EnergyModelService; + +use super::energy_model_vehicle_builders::VehicleBuilder; + +pub struct EnergyModelBuilder {} + +impl TraversalModelBuilder for EnergyModelBuilder { + fn build( + &self, + params: &serde_json::Value, + ) -> Result<Arc<dyn TraversalModelService>, TraversalModelError> { + let traversal_key = CompassConfigurationField::Traversal.to_string(); + + let speed_table_path = params + .get_config_path( + String::from("speed_table_input_file"), + traversal_key.clone(), + ) + .map_err(\|e\| TraversalModelError::BuildError(e.to_string()))?; + let speed_table_speed_unit = params + .get_config_serde::<SpeedUnit>( + String::from("speed_table_speed_unit"), + traversal_key.clone(), + ) + .map_err(\|e\| TraversalModelError::BuildError(e.to_string()))?; + + let grade_table_path = params + .get_config_path_optional( + String::from("grade_table_input_file"), + traversal_key.clone(), + ) + .map_err(\|e\| TraversalModelError::BuildError(e.to_string()))?; + let grade_table_grade_unit = params + .get_config_serde_optional::<GradeUnit>( + String::from("graph_grade_unit"), + traversal_key.clone(), + ) + .map_err(\|e\| TraversalModelError::BuildError(e.to_string()))?; + + let vehicle_configs = params + .get_config_array("vehicles".to_string(), traversal_key.clone()) + .map_err(\|e\| TraversalModelError::BuildError(e.to_string()))?; + + let mut vehicle_library = HashMap::new(); + + for vehicle_config in vehicle_configs { + let vehicle_type = vehicle_config + .get_config_string(String::from("type"), traversal_key.clone()) + .map_err(\|e\| TraversalModelError::BuildError(e.to_string()))?; + let vehicle_builder = VehicleBuilder::from_string(vehicle_type).map_err(\|e\| { + TraversalModelError::BuildError(format!( + "Error building vehicle builder: {}", + e.to_string() + )) + })?; + let vehicle = vehicle_builder + .build(&vehicle_config) + .map_err(\|e\| TraversalModelError::BuildError(e.to_string()))?; + vehicle_library.insert(vehicle.name(), vehicle); + } + + let output_time_unit_option = params + .get_config_serde_optional::<TimeUnit>( + String::from("output_time_unit"), + traversal_key.clone(), + ) + .map_err(\|e\| TraversalModelError::BuildError(e.to_string()))?; + let output_distance_unit_option = params + .get_config_serde_optional::<DistanceUnit>( + String::from("output_distance_unit"), + traversal_key.clone(), + ) + .map_err(\|e\| TraversalModelError::BuildError(e.to_string()))?; + + let service = EnergyModelService::new( + &speed_table_path, + speed_table_speed_unit, + &grade_table_path, + grade_table_grade_unit, + output_time_unit_option, + output_distance_unit_option, + vehicle_library, + )?; + + Ok(Arc::new(service)) + } +} diff --git a/rust/routee-compass/src/app/compass/config/traversal_model/energy_model_vehicle_builders.rs b/rust/routee-compass/src/app/compass/config/traversal_model/energy_model_vehicle_builders.rs new file mode 100644 index 00000000..d199502c --- /dev/null +++ b/rust/routee-compass/src/app/compass/config/traversal_model/energy_model_vehicle_builders.rs @@ -0,0 +1,136 @@ +use std::sync::Arc; + +use routee_compass_core::util::unit::{ + Energy, EnergyRate, EnergyRateUnit, EnergyUnit, GradeUnit, SpeedUnit, +}; +use routee_compass_powertrain::routee::{ + prediction::{load_prediction_model, model_type::ModelType, PredictionModelRecord}, + vehicle::{ + default::{dual_fuel_vehicle::DualFuelVehicle, single_fuel_vehicle::SingleFuelVehicle}, + VehicleType, + }, +}; + +use crate::app::compass::config::{ + compass_configuration_error::CompassConfigurationError, + compass_configuration_field::CompassConfigurationField, + config_json_extension::ConfigJsonExtensions, +}; + +pub enum VehicleBuilder { + SingleFuel, + DualFuel, +} + +impl VehicleBuilder { + pub fn from_string(vehicle_type: String) -> Result<VehicleBuilder, CompassConfigurationError> { + match vehicle_type.as_str() { + "single_fuel" => Ok(VehicleBuilder::SingleFuel), + "dual_fuel" => Ok(VehicleBuilder::DualFuel), + _ => Err(CompassConfigurationError::ExpectedFieldWithType( + "vehicle.type".to_string(), + "string".to_string(), + )), + } + } + pub fn build( + &self, + parameters: &serde_json::Value, + ) -> Result<Arc<dyn VehicleType>, CompassConfigurationError> { + match self { + VehicleBuilder::SingleFuel => build_conventional(parameters), + VehicleBuilder::DualFuel => build_plugin_hybrid(parameters), + } + } +} + +fn build_conventional( + parameters: &serde_json::Value, +) -> Result<Arc<dyn VehicleType>, CompassConfigurationError> { + let vehicle_key = String::from("single_fuel"); + let name = parameters.get_config_string(String::from("name"), vehicle_key.clone())?; + + let model_record = get_model_record_from_params(parameters, name.clone())?; + + let vehicle = SingleFuelVehicle::new(name, model_record)?; + + Ok(Arc::new(vehicle)) +} + +fn build_plugin_hybrid( + parameters: &serde_json::Value, +) -> Result<Arc<dyn VehicleType>, CompassConfigurationError> { + let vehicle_key = String::from("dual_fuel"); + let name = parameters.get_config_string(String::from("name"), vehicle_key.clone())?; + + let charge_depleting_params = + parameters.get_config_section(CompassConfigurationField::ChargeDepleting)?; + + let charge_depleting_record = get_model_record_from_params( + &charge_depleting_params, + format!("charge_depleting: {}", name.clone()), + )?; + let charge_sustain_params = + parameters.get_config_section(CompassConfigurationField::ChargeSustaining)?; + + let charge_sustain_record = get_model_record_from_params( + &charge_sustain_params, + format!("charge_sustain: {}", name.clone()), + )?; + + let battery_capacity = parameters + .get_config_serde::<Energy>(String::from("battery_capacity"), vehicle_key.clone())?; + let battery_energy_unit = parameters.get_config_serde::<EnergyUnit>( + String::from("battery_capacity_unit"), + vehicle_key.clone(), + )?; + let starting_battery_energy = battery_capacity; + let phev = DualFuelVehicle::new( + name, + charge_sustain_record, + charge_depleting_record, + battery_capacity, + starting_battery_energy, + battery_energy_unit, + )?; + Ok(Arc::new(phev)) +} + +fn get_model_record_from_params( + parameters: &serde_json::Value, + parent_key: String, +) -> Result<PredictionModelRecord, CompassConfigurationError> { + let name = parameters.get_config_string(String::from("name"), parent_key.clone())?; + let model_path = + parameters.get_config_path(String::from("model_input_file"), parent_key.clone())?; + let model_type = + parameters.get_config_serde::<ModelType>(String::from("model_type"), parent_key.clone())?; + let speed_unit = + parameters.get_config_serde::<SpeedUnit>(String::from("speed_unit"), parent_key.clone())?; + let ideal_energy_rate_option = parameters.get_config_serde_optional::<EnergyRate>( + String::from("ideal_energy_rate"), + parent_key.clone(), + )?; + let grade_unit = + parameters.get_config_serde::<GradeUnit>(String::from("grade_unit"), parent_key.clone())?; + + let energy_rate_unit = parameters + .get_config_serde::<EnergyRateUnit>(String::from("energy_rate_unit"), parent_key.clone())?; + let real_world_energy_adjustment_option = parameters.get_config_serde_optional::<f64>( + String::from("real_world_energy_adjustment"), + parent_key.clone(), + )?; + + let model_record = load_prediction_model( + name.clone(), + &model_path, + model_type, + speed_unit, + grade_unit, + energy_rate_unit, + ideal_energy_rate_option, + real_world_energy_adjustment_option, + )?; + + Ok(model_record) +} diff --git a/rust/routee-compass/src/app/compass/config/traversal_model/mod.rs b/rust/routee-compass/src/app/compass/config/traversal_model/mod.rs index 857fe071..4eace7aa 100644 --- a/rust/routee-compass/src/app/compass/config/traversal_model/mod.rs +++ b/rust/routee-compass/src/app/compass/config/traversal_model/mod.rs @@ -1,3 +1,4 @@ pub mod distance_builder; -pub mod speed_grade_energy_model_builder; +pub mod energy_model_builder; +pub mod energy_model_vehicle_builders; pub mod speed_lookup_builder; diff --git a/rust/routee-compass/src/app/compass/config/traversal_model/speed_grade_energy_model_builder.rs b/rust/routee-compass/src/app/compass/config/traversal_model/speed_grade_energy_model_builder.rs deleted file mode 100644 index d5b1c33b..00000000 --- a/rust/routee-compass/src/app/compass/config/traversal_model/speed_grade_energy_model_builder.rs +++ /dev/null @@ -1,138 +0,0 @@ -use std::collections::HashMap; -use std::sync::Arc; - -use crate::app::compass::config::compass_configuration_error::CompassConfigurationError; -use crate::app::compass::config::compass_configuration_field::CompassConfigurationField; -use crate::app::compass::config::config_json_extension::ConfigJsonExtensions; -use routee_compass_core::model::traversal::traversal_model_builder::TraversalModelBuilder; -use routee_compass_core::model::traversal::traversal_model_error::TraversalModelError; -use routee_compass_core::model::traversal::traversal_model_service::TraversalModelService; -use routee_compass_core::util::unit::{ - DistanceUnit, EnergyRate, EnergyRateUnit, GradeUnit, SpeedUnit, TimeUnit, -}; -use routee_compass_powertrain::routee::model_type::ModelType; -use routee_compass_powertrain::routee::prediction_model::SpeedGradePredictionModelRecord; -use routee_compass_powertrain::routee::speed_grade_energy_model_service::SpeedGradeEnergyModelService; - -pub struct SpeedGradeEnergyModelBuilder {} - -impl TraversalModelBuilder for SpeedGradeEnergyModelBuilder { - fn build( - &self, - params: &serde_json::Value, - ) -> Result<Arc<dyn TraversalModelService>, TraversalModelError> { - let traversal_key = CompassConfigurationField::Traversal.to_string(); - - let speed_table_path = params - .get_config_path( - String::from("speed_table_input_file"), - traversal_key.clone(), - ) - .map_err(\|e\| TraversalModelError::BuildError(e.to_string()))?; - let speed_table_speed_unit = params - .get_config_serde::<SpeedUnit>( - String::from("speed_table_speed_unit"), - traversal_key.clone(), - ) - .map_err(\|e\| TraversalModelError::BuildError(e.to_string()))?; - - let grade_table_path = params - .get_config_path_optional( - String::from("grade_table_input_file"), - traversal_key.clone(), - ) - .map_err(\|e\| TraversalModelError::BuildError(e.to_string()))?; - let grade_table_grade_unit = params - .get_config_serde_optional::<GradeUnit>( - String::from("graph_grade_unit"), - traversal_key.clone(), - ) - .map_err(\|e\| TraversalModelError::BuildError(e.to_string()))?; - - let energy_model_configs = params - .get_config_array("energy_models".to_string(), traversal_key.clone()) - .map_err(\|e\| TraversalModelError::BuildError(e.to_string()))?; - - let mut energy_model_library = HashMap::new(); - - for energy_model_config in energy_model_configs { - let name = energy_model_config - .get_config_string(String::from("name"), traversal_key.clone()) - .map_err(\|e\| TraversalModelError::BuildError(e.to_string()))?; - let model_path = energy_model_config - .get_config_path(String::from("model_input_file"), traversal_key.clone()) - .map_err(\|e\| TraversalModelError::BuildError(e.to_string()))?; - let model_type = energy_model_config - .get_config_serde::<ModelType>(String::from("model_type"), traversal_key.clone()) - .map_err(\|e\| TraversalModelError::BuildError(e.to_string()))?; - let speed_unit = energy_model_config - .get_config_serde::<SpeedUnit>(String::from("speed_unit"), traversal_key.clone()) - .map_err(\|e\| TraversalModelError::BuildError(e.to_string()))?; - let ideal_energy_rate_option = energy_model_config - .get_config_serde_optional::<EnergyRate>( - String::from("ideal_energy_rate"), - traversal_key.clone(), - ) - .map_err(\|e\| TraversalModelError::BuildError(e.to_string()))?; - let grade_unit = energy_model_config - .get_config_serde::<GradeUnit>(String::from("grade_unit"), traversal_key.clone()) - .map_err(\|e\| TraversalModelError::BuildError(e.to_string()))?; - - let energy_rate_unit = energy_model_config - .get_config_serde::<EnergyRateUnit>( - String::from("energy_rate_unit"), - traversal_key.clone(), - ) - .map_err(\|e\| TraversalModelError::BuildError(e.to_string()))?; - let real_world_energy_adjustment_option = params - .get_config_serde_optional::<f64>( - String::from("real_world_energy_adjustment"), - traversal_key.clone(), - ) - .map_err(\|e\| TraversalModelError::BuildError(e.to_string()))?; - - let model_record = SpeedGradePredictionModelRecord::new( - name.clone(), - &model_path, - model_type, - speed_unit, - grade_unit, - energy_rate_unit, - ideal_energy_rate_option, - real_world_energy_adjustment_option, - ) - .map_err(\|e\| TraversalModelError::BuildError(e.to_string()))?; - energy_model_library.insert(name, Arc::new(model_record)); - } - - let output_time_unit_option = params - .get_config_serde_optional::<TimeUnit>( - String::from("output_time_unit"), - traversal_key.clone(), - ) - .map_err(\|e\| TraversalModelError::BuildError(e.to_string()))?; - let output_distance_unit_option = params - .get_config_serde_optional::<DistanceUnit>( - String::from("output_distance_unit"), - traversal_key.clone(), - ) - .map_err(\|e\| TraversalModelError::BuildError(e.to_string()))?; - - let service = SpeedGradeEnergyModelService::new( - &speed_table_path, - speed_table_speed_unit, - &grade_table_path, - grade_table_grade_unit, - output_time_unit_option, - output_distance_unit_option, - energy_model_library, - ) - .map_err(CompassConfigurationError::TraversalModelError) - .map_err(\|e\| TraversalModelError::BuildError(e.to_string()))?; - // let service = SpeedGradeEnergyModelService { - // service: inner_service, - // }; - let result: Arc<dyn TraversalModelService> = Arc::new(service); - Ok(result) - } -}	PHEV implementation Our current energy traversal model supports conventional vehicles and battery electric vehicles but not plugin hybrid electric (PHEV) vehicles and would require a bit of a modification to do so. One primary limitation is that we're not keeping track of vehicle state of charge. This is important for a PHEV as it generally dictates what fuel type is going to be used. For example, if the state of charge is 100% and the route can be completed with battery energy only, we would report back 0 gasoline usage and X electrical energy usage. On the other hand, if the state of charge starts at 10% and the route cannot be completed using the battery energy, we might report back some electrical energy usage and some gasoline usage when the vehicle switched over to a different source of energy. Routee-Powertrain currently models PHEVs by creating two separate models, a charge sustaining (battery and gasoline) model and a charge depleting (just battery) model. To implement this as a traversal model we could build a simple PHEV that would use the charge depleting RouteE Powertrain model until the battery ran out of energy and then switch to the charge sustaining model for the remainder of the trip. This would require the user to pass in a state of charge parameter in the query and would require us to keep track of battery energy in our state variable. We would also need to define the battery capacity at the config level.	This also presents some difficulty with our existing implementation of applying an energy cost coefficient directly to the energy since the cost could be in variable units depending on the type of fuel being used. We could start by just converting the energy to a single unit (like converting gallons of gasoline to estimated equivalent kwh) but the best solution is probably to apply a $ value to each unit.	2023-11-16T22:08:28	[]	[]
NREL/routee-compass	NREL__routee-compass-8	2d65ee947b27536b0f97fb7be813bb4f3dd9dfb2	diff --git a/.cargo/config.toml b/.cargo/config.toml index ca43966b..4d1815a7 100644 --- a/.cargo/config.toml +++ b/.cargo/config.toml @@ -3,7 +3,7 @@ check-revoke = false [env] ORT_STRATEGY = { value = "system" } -ORT_LIB_LOCATION = { value = "rust/compass-powertrain/onnx-runtime/v1.15.1/osx-x86_64", relative = true } +ORT_LIB_LOCATION = { value = "rust/routee-compass-powertrain/onnx-runtime/v1.15.1/osx-x86_64", relative = true } [target.x86_64-apple-darwin] rustflags = ["-C", "link-arg=-undefined", "-C", "link-arg=dynamic_lookup"] diff --git a/docs/config.md b/docs/config.md index 33310934..e325d29c 100644 --- a/docs/config.md +++ b/docs/config.md @@ -13,9 +13,9 @@ parallelism = 2 # the parameters for the underlying road network graph [graph] # a file containing all the graph edges and their adjacencies -edge_list_csv = "edges-compass.csv.gz" +edge_list_file = "edges-compass.csv.gz" # a file containing all the graph verticies -vertex_list_csv = "vertices-compass.csv.gz" +vertex_list_file = "vertices-compass.csv.gz" # if verbose is true, you'll see more information when loading the graph verbose = true @@ -26,9 +26,9 @@ type = "speed_grade_energy_model" # what underlying machine learn framework to use [smartcore \| onnx] model_type = "smartcore" # speeds for each edge in the graph -speed_table_path = "edges-posted-speed-enumerated.txt.gz" +speed_table_file = "edges-posted-speed-enumerated.txt.gz" # the file for the routee-powertrain model -energy_model_path = "2016_TOYOTA_Camry_4cyl_2WD.bin" +energy_model_file = "2016_TOYOTA_Camry_4cyl_2WD.bin" # the "best case" energy rate for this particular vehicle (something like highway mpg) that's used in the a-star algorithm ideal_energy_rate = 0.02857142857 # the units of the above speed table @@ -103,7 +103,7 @@ The speed table traversal model uses a speed lookup table to compute the fastest ```toml [traversal] type = "speed_table" -speed_table_path = "edges-posted-speed-enumerated.txt.gz" +speed_table_file = "edges-posted-speed-enumerated.txt.gz" speed_unit = "kilometers_per_hour" output_distance_unit = "miles" output_time_unit = "minutes" @@ -117,10 +117,10 @@ The speed grade energy model computes energy (with a routee-powertrain vehicle m [traversal] type = "speed_grade_energy_model" model_type = "smartcore" -speed_table_path = "data/tomtom_metro_denver_network/edges-free-flow-speed-enumerated.txt.gz" -grade_table_path = "data/tomtom_metro_denver_network/edges-grade-enumerated.txt.gz" +speed_table_file = "data/tomtom_metro_denver_network/edges-free-flow-speed-enumerated.txt.gz" +grade_table_file = "data/tomtom_metro_denver_network/edges-grade-enumerated.txt.gz" grade_table_grade_unit = "decimal" -energy_model_path = "data/2016_TOYOTA_Camry_4cyl_2WD.bin" +energy_model_file = "data/2016_TOYOTA_Camry_4cyl_2WD.bin" ideal_energy_rate = 0.02857142857 speed_table_speed_unit = "kilometers_per_hour" energy_model_speed_unit = "miles_per_hour" diff --git a/docs/notebooks/open_street_maps_example.ipynb b/docs/notebooks/open_street_maps_example.ipynb index d19f81df..0641c4d7 100644 --- a/docs/notebooks/open_street_maps_example.ipynb +++ b/docs/notebooks/open_street_maps_example.ipynb @@ -184,17 +184,6 @@ { "cell_type": "code", "execution_count": 5, - "id": "30c0da5c-fd6d-44f2-873c-1ce8b863cda8", - "metadata": {}, - "outputs": [], - "source": [ - "import os\n", - "os.chdir(\"golden_co\")" - ] - }, - { - "cell_type": "code", - "execution_count": 6, "id": "edad6c15-c641-4010-b097-7d07aa86c1e7", "metadata": {}, "outputs": [ @@ -212,12 +201,12 @@ "name": "stderr", "output_type": "stream", "text": [ - "uuid file: 100%\|██████████\| 746/746 [00:00<00:00, 4245946.00it/s]it/s]" + "uuid file: 100%\|██████████\| 746/746 [00:00<00:00, 3849865.50it/s]it/s]" ] } ], "source": [ - "app = CompassApp.from_config_file(\"osm_default_energy.toml\") " + "app = CompassApp.from_config_file(\"golden_co/osm_default_energy.toml\") " ] }, { @@ -232,7 +221,7 @@ }, { "cell_type": "code", - "execution_count": 7, + "execution_count": 6, "id": "f01086d0-231f-4013-bc24-97f9c56670bd", "metadata": {}, "outputs": [], @@ -252,7 +241,7 @@ }, { "cell_type": "code", - "execution_count": 8, + "execution_count": 7, "id": "c6e9361b-5ffd-4ec5-a09a-d51e46752ebf", "metadata": {}, "outputs": [], @@ -270,7 +259,7 @@ }, { "cell_type": "code", - "execution_count": 9, + "execution_count": 8, "id": "2c51774c-acff-4569-a820-4be5ba5aa51b", "metadata": {}, "outputs": [ @@ -280,7 +269,7 @@ "2" ] }, - "execution_count": 9, + "execution_count": 8, "metadata": {}, "output_type": "execute_result" } @@ -299,7 +288,7 @@ }, { "cell_type": "code", - "execution_count": 10, + "execution_count": 9, "id": "d075cb16-b0ae-463a-a2ce-f94042bc13db", "metadata": {}, "outputs": [], @@ -309,7 +298,7 @@ }, { "cell_type": "code", - "execution_count": 11, + "execution_count": 10, "id": "961c910c-63ec-4c79-9522-19f24f151215", "metadata": {}, "outputs": [ @@ -318,13 +307,13 @@ "text/plain": [ "{'distance': 4.23738036063421,\n", " 'distance_unit': 'miles',\n", - " 'energy': 0.18821565778485838,\n", + " 'energy': 0.18816139103031826,\n", " 'energy_unit': 'gallons_gasoline',\n", - " 'time': 8.501927726263151,\n", + " 'time': 8.486183959367098,\n", " 'time_unit': 'minutes'}" ] }, - "execution_count": 11, + "execution_count": 10, "metadata": {}, "output_type": "execute_result" } @@ -335,7 +324,7 @@ }, { "cell_type": "code", - "execution_count": 12, + "execution_count": 11, "id": "e2c6cb39-49aa-4697-9785-fa839dc16a86", "metadata": {}, "outputs": [ @@ -344,13 +333,13 @@ "text/plain": [ "{'distance': 4.017396519730842,\n", " 'distance_unit': 'miles',\n", - " 'energy': 0.18722374836334246,\n", + " 'energy': 0.1871862788506351,\n", " 'energy_unit': 'gallons_gasoline',\n", - " 'time': 8.987039191276414,\n", + " 'time': 8.619006658572934,\n", " 'time_unit': 'minutes'}" ] }, - "execution_count": 12, + "execution_count": 11, "metadata": {}, "output_type": "execute_result" } @@ -362,7 +351,7 @@ }, { "cell_type": "code", - "execution_count": 13, + "execution_count": 12, "id": "30257102-07ee-4338-9c8c-41676e1995ed", "metadata": {}, "outputs": [], @@ -384,7 +373,7 @@ }, { "cell_type": "code", - "execution_count": 14, + "execution_count": 13, "id": "d3907fe2-14c8-4904-942a-d09fd7d42ab2", "metadata": {}, "outputs": [], @@ -402,7 +391,7 @@ }, { "cell_type": "code", - "execution_count": 15, + "execution_count": 14, "id": "1ac13aae-32fc-4c10-b888-6373b175655a", "metadata": {}, "outputs": [ @@ -436,7 +425,7 @@ " <meta name="viewport" content="width=device-width,\n", " initial-scale=1.0, maximum-scale=1.0, user-scalable=no" />\n", " <style>\n", - " #map_5e6fdd78aee60d0d6315972837ed641b {\n", + " #map_f0bafdc10b40d238de9da7b18732b217 {\n", " position: relative;\n", " width: 100.0%;\n", " height: 100.0%;\n", @@ -450,14 +439,14 @@ "<body>\n", " \n", " \n", - " <div class="folium-map" id="map_5e6fdd78aee60d0d6315972837ed641b" ></div>\n", + " <div class="folium-map" id="map_f0bafdc10b40d238de9da7b18732b217" ></div>\n", " \n", "</body>\n", "<script>\n", " \n", " \n", - " var map_5e6fdd78aee60d0d6315972837ed641b = L.map(\n", - " "map_5e6fdd78aee60d0d6315972837ed641b",\n", + " var map_f0bafdc10b40d238de9da7b18732b217 = L.map(\n", + " "map_f0bafdc10b40d238de9da7b18732b217",\n", " {\n", " center: [39.7465774, -105.2216276],\n", " crs: L.CRS.EPSG3857,\n", @@ -471,19 +460,19 @@ "\n", " \n", " \n", - " var tile_layer_ee2775096a1be4380f9bd39c42ca4e8e = L.tileLayer(\n", + " var tile_layer_41ead211ec41bcf73e91c5f85331e723 = L.tileLayer(\n", " "https://{s}.tile.openstreetmap.org/{z}/{x}/{y}.png",\n", " {"attribution": "Data by \\u0026copy; \\u003ca target=\\"_blank\\" href=\\"http://openstreetmap.org\\"\\u003eOpenStreetMap\\u003c/a\\u003e, under \\u003ca target=\\"_blank\\" href=\\"http://www.openstreetmap.org/copyright\\"\\u003eODbL\\u003c/a\\u003e.", "detectRetina": false, "maxNativeZoom": 18, "maxZoom": 18, "minZoom": 0, "noWrap": false, "opacity": 1, "subdomains": "abc", "tms": false}\n", - " ).addTo(map_5e6fdd78aee60d0d6315972837ed641b);\n", + " ).addTo(map_f0bafdc10b40d238de9da7b18732b217);\n", " \n", " \n", - " var poly_line_6406764e3f6bcb21b9ed2e3ae7f61509 = L.polyline(\n", + " var poly_line_51128cf08c47e0b255fa936899f6e0e4 = L.polyline(\n", " [[39.7273069, -105.1994169], [39.7272989, -105.1994535], [39.7272848, -105.1994868], [39.7272593, -105.1995218], [39.7272272, -105.1995461], [39.7271912, -105.1995576], [39.7271688, -105.199558], [39.7271461, -105.1995532], [39.7271142, -105.1995365], [39.7270868, -105.1995096], [39.7270657, -105.1994741], [39.7270519, -105.1994296], [39.7270483, -105.1993816], [39.727052, -105.1993539], [39.7270571, -105.1993247], [39.7270719, -105.1992916], [39.7270973, -105.1992566], [39.7271293, -105.1992323], [39.7271651, -105.1992207], [39.7271854, -105.1992226], [39.7272113, -105.199227], [39.7273496, -105.1991687], [39.7274185, -105.1991465], [39.7275326, -105.1990928], [39.7276828, -105.1990602], [39.7277455, -105.1990518], [39.7279702, -105.1990232], [39.7281558, -105.1990357], [39.7284384, -105.1990893], [39.7286639, -105.199177], [39.7288222, -105.1992773], [39.7289214, -105.199385], [39.7290292, -105.1995543], [39.7290506, -105.1995879], [39.7292157, -105.1998155], [39.7293094, -105.1998945], [39.7294079, -105.1999604], [39.7295437, -105.200014], [39.7296299, -105.2000721], [39.7297292, -105.2000757], [39.7297135, -105.2007081], [39.7297192, -105.2013106], [39.7298153, -105.2032041], [39.7298256, -105.2033784], [39.7298664, -105.2035708], [39.7298795, -105.203649], [39.7299267, -105.2038558], [39.7301082, -105.2042608], [39.7304911, -105.2048068], [39.7308741, -105.2053111], [39.7310676, -105.2055985], [39.7311933, -105.205824], [39.7312069, -105.2058543], [39.7313075, -105.2061158], [39.731348, -105.206248], [39.731386, -105.2064213], [39.7314237, -105.2067336], [39.731429, -105.2070565], [39.7314226, -105.207224], [39.7314076, -105.2074036], [39.7313611, -105.2077675], [39.7313265, -105.2079124], [39.7312831, -105.20806], [39.7312292, -105.2082086], [39.7311706, -105.208344], [39.7310965, -105.2084777], [39.7310049, -105.208616], [39.7309234, -105.2087235], [39.7308212, -105.2088303], [39.7306227, -105.2089819], [39.7304801, -105.2090736], [39.7303681, -105.2091988], [39.7303327, -105.2092597], [39.7303114, -105.2093181], [39.730299, -105.20939], [39.7302939, -105.2095126], [39.7303105, -105.2096804], [39.7304414, -105.2099267], [39.7305001, -105.2100344], [39.7305871, -105.2101842], [39.7307404, -105.2104481], [39.7309234, -105.2107229], [39.7310934, -105.2109639], [39.731238, -105.211145], [39.7314101, -105.2113464], [39.7315783, -105.2115336], [39.7317541, -105.2117064], [39.7319796, -105.2119089], [39.732359, -105.2122393], [39.7329187, -105.2127184], [39.7334557, -105.213188], [39.7338968, -105.2135676], [39.7342777, -105.2138968], [39.7347139, -105.2142786], [39.7354652, -105.2149301], [39.7356319, -105.2150733], [39.7358844, -105.2152902], [39.7363901, -105.2157275], [39.7369129, -105.2161859], [39.7377243, -105.2168831], [39.7390882, -105.2180682], [39.7397882, -105.2186718], [39.7403043, -105.2191078], [39.7405645, -105.2193336], [39.7406767, -105.2194225], [39.7415803, -105.2202142], [39.7416486, -105.2202727], [39.7417047, -105.2203151], [39.7421502, -105.2206932], [39.7425816, -105.221062], [39.7432687, -105.2215398], [39.7435797, -105.2217631], [39.7436721, -105.2218299], [39.7437366, -105.2218764], [39.7438794, -105.2219759], [39.7439546, -105.2220253], [39.7440082, -105.2220586], [39.7440819, -105.2221021], [39.7441377, -105.2221322], [39.7441808, -105.2221541], [39.7442796, -105.2222043], [39.7443357, -105.2222286], [39.7444062, -105.222259], [39.7444657, -105.2222871], [39.7444962, -105.2223029], [39.7445695, -105.2223431], [39.7446343, -105.2223808], [39.744683, -105.2224104], [39.7447215, -105.2224349], [39.7447658, -105.2224654], [39.744813, -105.222499], [39.7448547, -105.2225315], [39.7448881, -105.2225586], [39.7449328, -105.2225965], [39.7449815, -105.2226391], [39.7451542, -105.2227925], [39.7454784, -105.2230438], [39.7455034, -105.2230623], [39.7455159, -105.2230713], [39.7455403, -105.2230861], [39.7455548, -105.2230941], [39.7455766, -105.2231014], [39.7456018, -105.223106], [39.7456176, -105.2231071], [39.7456337, -105.223107], [39.7456638, -105.2231045], [39.7456814, -105.2231011], [39.7456983, -105.2230945], [39.7457224, -105.2230821], [39.7457338, -105.2230745], [39.7457588, -105.2230543], [39.7457709, -105.223043], [39.7457834, -105.2230284], [39.7458035, -105.2230027], [39.7458905, -105.2228787], [39.7460068, -105.2226624], [39.7461589, -105.2223894], [39.7463617, -105.2220126], [39.7464427, -105.2218672], [39.746482, -105.2217968], [39.7465202, -105.2217322], [39.7465523, -105.2216771], [39.7465774, -105.2216276], [39.7465928, -105.2215823], [39.7466004, -105.2215466], [39.7466042, -105.2215101], [39.7466044, -105.2214695], [39.746601, -105.2214368], [39.7465881, -105.2213862], [39.7465804, -105.2213567], [39.7465774, -105.2213321], [39.7465774, -105.2213048], [39.746581, -105.2212762], [39.7465874, -105.2212518], [39.7465978, -105.2212266], [39.7466125, -105.2212024], [39.7466266, -105.2211858], [39.7466485, -105.2211678], [39.7466708, -105.2211566], [39.746693, -105.2211511], [39.746715, -105.2211506], [39.7467203, -105.2211517], [39.7467798, -105.2211327], [39.7468101, -105.2211204], [39.7468347, -105.2211032], [39.7468673, -105.2210676], [39.7468863, -105.2210389], [39.7469079, -105.2209995], [39.7469752, -105.2208699], [39.7471073, -105.220655], [39.7471371, -105.2206129], [39.7471594, -105.2205845], [39.7472061, -105.2205376], [39.7473376, -105.2202962], [39.747357, -105.2202177], [39.7473682, -105.2201868], [39.7473848, -105.2201519], [39.7475089, -105.2199249], [39.7475861, -105.2197973], [39.7476254, -105.2197401], [39.7476917, -105.2196525], [39.7477724, -105.2195069], [39.7478875, -105.219299], [39.7479219, -105.2192372], [39.7479518, -105.219179], [39.7480957, -105.2189201], [39.7481337, -105.2188515], [39.7482567, -105.2186294], [39.7484851, -105.2182271], [39.7485167, -105.218164], [39.7485605, -105.2180881], [39.7490516, -105.217166], [39.7490955, -105.2170866], [39.7491263, -105.2170387], [39.7493736, -105.2165801], [39.7496841, -105.2160128], [39.7497482, -105.2160733], [39.7504593, -105.216753], [39.7505084, -105.2168007], [39.7505733, -105.2168567], [39.7509525, -105.2172063], [39.7512593, -105.2174884], [39.7513304, -105.2175506], [39.7513869, -105.217604], [39.7517458, -105.2179367], [39.7520945, -105.21826], [39.7521462, -105.2183047], [39.7522126, -105.218369], [39.7523012, -105.2184523], [39.7525052, -105.218642], [39.7527145, -105.2188366], [39.7528959, -105.2189993], [39.7529601, -105.2190649], [39.7530308, -105.219128], [39.7537033, -105.2197718], [39.753764, -105.2198302], [39.7538299, -105.2198915], [39.7542346, -105.2202638], [39.7545279, -105.2205433], [39.7545847, -105.2206031], [39.7546623, -105.2206728], [39.7553326, -105.2212752], [39.755407, -105.221342], [39.7554908, -105.2214211], [39.7561483, -105.2220415], [39.7562254, -105.2221143], [39.7563006, -105.2221853], [39.7566193, -105.222483], [39.7566315, -105.2224957], [39.7569475, -105.2227913], [39.7570561, -105.222895], [39.7574792, -105.2232693], [39.7575579, -105.2233409], [39.7576236, -105.2234019], [39.7578565, -105.2236199], [39.7579703, -105.2237258], [39.7580027, -105.2237561], [39.7583001, -105.2240544], [39.7583591, -105.2241084], [39.7587249, -105.2244446], [39.7588057, -105.22452], [39.7591609, -105.2248516], [39.7595089, -105.2251741], [39.7596887, -105.2253408], [39.7597554, -105.2254024], [39.7598802, -105.2255173], [39.7598939, -105.2255305], [39.7599658, -105.2255988], [39.7600186, -105.2256498], [39.7604008, -105.2259941], [39.7605066, -105.2260931], [39.7605924, -105.226174], [39.760667, -105.2262427], [39.7607946, -105.2263631], [39.7608753, -105.2264373], [39.7615829, -105.2271044], [39.7616266, -105.2271481], [39.7616788, -105.2272008], [39.762203, -105.2276943], [39.7623411, -105.2278299], [39.7630405, -105.2284565], [39.7632824, -105.2286685], [39.7637358, -105.2290679], [39.7640137, -105.229309], [39.7640954, -105.229381], [39.7641743, -105.229462], [39.76436, -105.2296495], [39.7645376, -105.2298658], [39.764824, -105.2301918], [39.7652051, -105.2307044], [39.7654711, -105.2310802], [39.7654801, -105.2310934], [39.7655308, -105.231168], [39.7655792, -105.231232], [39.7655918, -105.2312486], [39.7657719, -105.2315035], [39.7658973, -105.2316751], [39.7659962, -105.2318039], [39.7661018, -105.231924], [39.7662469, -105.2320614], [39.7664515, -105.2322588], [39.7671442, -105.2329283], [39.7673559, -105.2331244], [39.7674414, -105.2332053], [39.7675186, -105.2332845], [39.7675756, -105.2333522], [39.7676291, -105.2334243], [39.7676969, -105.2335369], [39.7677304, -105.233602], [39.767761, -105.2336649], [39.7677908, -105.233739], [39.7678172, -105.233812], [39.7678551, -105.2339422], [39.7678736, -105.2340294], [39.7678904, -105.2341257], [39.7679032, -105.2342329], [39.7679095, -105.2343359], [39.7679068, -105.2344027], [39.7679029, -105.234499], [39.7678875, -105.2346626], [39.7678651, -105.2347791], [39.7678384, -105.2348898], [39.7678006, -105.2350283], [39.7677517, -105.2352031], [39.7677119, -105.2353468]],\n", " {"bubblingMouseEvents": true, "color": "blue", "dashArray": null, "dashOffset": null, "fill": false, "fillColor": "blue", "fillOpacity": 0.2, "fillRule": "evenodd", "lineCap": "round", "lineJoin": "round", "noClip": false, "opacity": 0.8, "smoothFactor": 1.0, "stroke": true, "weight": 10}\n", - " ).addTo(map_5e6fdd78aee60d0d6315972837ed641b);\n", + " ).addTo(map_f0bafdc10b40d238de9da7b18732b217);\n", " \n", " \n", - " poly_line_6406764e3f6bcb21b9ed2e3ae7f61509.bindTooltip(\n", + " poly_line_51128cf08c47e0b255fa936899f6e0e4.bindTooltip(\n", " `<div>\n", " Shortest Time\n", " </div>`,\n", @@ -491,19 +480,19 @@ " );\n", " \n", " \n", - " var marker_6902ee7feb3c91f411bef91fffcafda8 = L.marker(\n", + " var marker_eb6728cbb45319dfc3741bebb57bec6e = L.marker(\n", " [39.7273069, -105.1994169],\n", " {}\n", - " ).addTo(map_5e6fdd78aee60d0d6315972837ed641b);\n", + " ).addTo(map_f0bafdc10b40d238de9da7b18732b217);\n", " \n", " \n", - " var icon_45c70091cf67f498c42e94b48f45100f = L.AwesomeMarkers.icon(\n", + " var icon_c34923765e4fff2b62cddf1bb4a18bea = L.AwesomeMarkers.icon(\n", " {"extraClasses": "fa-rotate-0", "icon": "circle", "iconColor": "white", "markerColor": "green", "prefix": "fa"}\n", " );\n", - " marker_6902ee7feb3c91f411bef91fffcafda8.setIcon(icon_45c70091cf67f498c42e94b48f45100f);\n", + " marker_eb6728cbb45319dfc3741bebb57bec6e.setIcon(icon_c34923765e4fff2b62cddf1bb4a18bea);\n", " \n", " \n", - " marker_6902ee7feb3c91f411bef91fffcafda8.bindTooltip(\n", + " marker_eb6728cbb45319dfc3741bebb57bec6e.bindTooltip(\n", " `<div>\n", " Origin\n", " </div>`,\n", @@ -511,19 +500,19 @@ " );\n", " \n", " \n", - " var marker_8561d2d910a8be3bc5d2cfa6a30c6063 = L.marker(\n", + " var marker_4ca91c0844bc6f79cdde370a45e8cf3c = L.marker(\n", " [39.7677119, -105.2353468],\n", " {}\n", - " ).addTo(map_5e6fdd78aee60d0d6315972837ed641b);\n", + " ).addTo(map_f0bafdc10b40d238de9da7b18732b217);\n", " \n", " \n", - " var icon_fea81095d90d74dad83b121c4e124834 = L.AwesomeMarkers.icon(\n", + " var icon_83eb5299c606253f41a628839749ec1f = L.AwesomeMarkers.icon(\n", " {"extraClasses": "fa-rotate-0", "icon": "circle", "iconColor": "white", "markerColor": "red", "prefix": "fa"}\n", " );\n", - " marker_8561d2d910a8be3bc5d2cfa6a30c6063.setIcon(icon_fea81095d90d74dad83b121c4e124834);\n", + " marker_4ca91c0844bc6f79cdde370a45e8cf3c.setIcon(icon_83eb5299c606253f41a628839749ec1f);\n", " \n", " \n", - " marker_8561d2d910a8be3bc5d2cfa6a30c6063.bindTooltip(\n", + " marker_4ca91c0844bc6f79cdde370a45e8cf3c.bindTooltip(\n", " `<div>\n", " Destination\n", " </div>`,\n", @@ -531,13 +520,13 @@ " );\n", " \n", " \n", - " var poly_line_3349ea569e76e0f0fe319322fd08afcb = L.polyline(\n", + " var poly_line_95e6364ccd9609756d15b7d7194b0d25 = L.polyline(\n", " [[39.7273069, -105.1994169], [39.7272989, -105.1994535], [39.7272848, -105.1994868], [39.7272593, -105.1995218], [39.7272272, -105.1995461], [39.7271912, -105.1995576], [39.7271688, -105.199558], [39.7271461, -105.1995532], [39.7271142, -105.1995365], [39.7270868, -105.1995096], [39.7270657, -105.1994741], [39.7270519, -105.1994296], [39.7270483, -105.1993816], [39.727052, -105.1993539], [39.7270571, -105.1993247], [39.7270719, -105.1992916], [39.7270973, -105.1992566], [39.7271293, -105.1992323], [39.7271651, -105.1992207], [39.7271854, -105.1992226], [39.7272113, -105.199227], [39.7273496, -105.1991687], [39.7274185, -105.1991465], [39.7275326, -105.1990928], [39.7276828, -105.1990602], [39.7277455, -105.1990518], [39.7279702, -105.1990232], [39.7281558, -105.1990357], [39.7284384, -105.1990893], [39.7286639, -105.199177], [39.7288222, -105.1992773], [39.7289214, -105.199385], [39.7290292, -105.1995543], [39.7290506, -105.1995879], [39.7292157, -105.1998155], [39.7293094, -105.1998945], [39.7294079, -105.1999604], [39.7295437, -105.200014], [39.7296299, -105.2000721], [39.7296469, -105.1996789], [39.7296738, -105.1987284], [39.7296906, -105.1986794], [39.7297244, -105.1985856], [39.7297272, -105.1984193], [39.7299723, -105.1984198], [39.7302117, -105.1984227], [39.7303235, -105.1984285], [39.7304434, -105.198443], [39.7305651, -105.1984732], [39.7307182, -105.1985167], [39.7308487, -105.1985623], [39.7309909, -105.1986311], [39.731127, -105.1987144], [39.7312292, -105.1987897], [39.7313292, -105.1988731], [39.7314292, -105.1989611], [39.731524, -105.1990554], [39.7316089, -105.1991466], [39.7317085, -105.1992639], [39.7318575, -105.1994488], [39.7319522, -105.1995697], [39.7320423, -105.1996716], [39.7321372, -105.1997578], [39.7322615, -105.1998506], [39.7323697, -105.1999163], [39.7324712, -105.1999637], [39.7325898, -105.2000081], [39.732689, -105.2000309], [39.7327987, -105.2000627], [39.732945, -105.2001049], [39.7330635, -105.2001374], [39.7331838, -105.2001914], [39.7332899, -105.2002528], [39.7334312, -105.200363], [39.7335257, -105.2004544], [39.7335523, -105.2004835], [39.7336132, -105.2005503], [39.7336902, -105.2006555], [39.733767, -105.2007557], [39.7338411, -105.2008491], [39.7339359, -105.2009412], [39.7340357, -105.2010154], [39.7340742, -105.201042], [39.7341117, -105.2010678], [39.7341542, -105.2010901], [39.734211, -105.2011198], [39.7343173, -105.2011712], [39.7344056, -105.2012021], [39.7345704, -105.2012376], [39.7347726, -105.2012753], [39.7351329, -105.2013311], [39.7352203, -105.2013441], [39.7353243, -105.201352], [39.7354415, -105.2013441], [39.7355491, -105.2013268], [39.7357022, -105.2012901], [39.7358395, -105.201259], [39.7362429, -105.2011675], [39.7367157, -105.2010639], [39.7367857, -105.2010539], [39.7368531, -105.2010442], [39.7369645, -105.2010356], [39.7370755, -105.201044], [39.7371895, -105.2010632], [39.7373179, -105.201099], [39.737714, -105.2012295], [39.7379567, -105.2013009], [39.7384451, -105.2014404], [39.7386872, -105.2015167], [39.7395717, -105.2017905], [39.7397153, -105.2018369], [39.7398269, -105.2018673], [39.7399356, -105.2018843], [39.7400618, -105.2018961], [39.740184, -105.2018932], [39.7403133, -105.2018823], [39.7404288, -105.2018645], [39.7405518, -105.2018387], [39.740665, -105.2018073], [39.7407597, -105.2017713], [39.7408728, -105.2017196], [39.740992, -105.2016551], [39.7411268, -105.2015625], [39.7412599, -105.2014489], [39.7413718, -105.2013453], [39.7414758, -105.2012326], [39.741608, -105.2010691], [39.7416901, -105.200964], [39.7417903, -105.2008396], [39.7417994, -105.200826], [39.7418968, -105.200686], [39.7418878, -105.200623], [39.7418917, -105.2005592], [39.7419081, -105.2004988], [39.7419393, -105.2004416], [39.7419818, -105.2003982], [39.7420319, -105.2003723], [39.7420864, -105.2003661], [39.7421399, -105.2003805], [39.742188, -105.2004144], [39.7422255, -105.2004628], [39.742251, -105.2005229], [39.7422628, -105.2005898], [39.7422597, -105.2006584], [39.7422421, -105.2007231], [39.7422712, -105.2008198], [39.7423227, -105.200901], [39.7428069, -105.2016549], [39.7429048, -105.2018114], [39.7429131, -105.2018245], [39.7431503, -105.2022035], [39.7434374, -105.2026477], [39.7435807, -105.2028659], [39.7437705, -105.2031843], [39.7441255, -105.2037193], [39.7444536, -105.2042289], [39.7445369, -105.204369], [39.7448087, -105.2048085], [39.7450389, -105.205197], [39.745164, -105.2054397], [39.7452617, -105.205642], [39.745359, -105.2058753], [39.7454413, -105.2060693], [39.7454698, -105.2061365], [39.7455586, -105.2063498], [39.7456394, -105.2066197], [39.7457343, -105.2068984], [39.7458096, -105.2071619], [39.7458432, -105.2072893], [39.7459057, -105.2075162], [39.7459547, -105.2077463], [39.7460104, -105.2080066], [39.7460312, -105.2081149], [39.7460508, -105.2082709], [39.7460836, -105.2084787], [39.7461092, -105.2086125], [39.7461165, -105.2087078], [39.7461128, -105.2088019], [39.7461078, -105.2089177], [39.7461163, -105.208995], [39.7461377, -105.2090672], [39.7461588, -105.2091147], [39.7462509, -105.2092159], [39.746281, -105.2092382], [39.7462978, -105.209256], [39.7463242, -105.209292], [39.7463458, -105.2093473], [39.746361, -105.2094251], [39.7463921, -105.2095768], [39.7464353, -105.2097133], [39.7465001, -105.2098285], [39.7467097, -105.21004], [39.7467814, -105.2101094], [39.7473357, -105.2106047], [39.7474554, -105.2107116], [39.7475615, -105.2108064], [39.7475958, -105.2108371], [39.747653, -105.2108907], [39.7481502, -105.2113567], [39.7483361, -105.2115309], [39.7483726, -105.2115651], [39.7484183, -105.2116079], [39.7484718, -105.2116584], [39.7487212, -105.2118951], [39.749238, -105.2123721], [39.7492814, -105.2124135], [39.7500059, -105.2130759], [39.7500454, -105.2131129], [39.7503812, -105.2134256], [39.7504627, -105.2135015], [39.7508219, -105.2138362], [39.7508792, -105.2138895], [39.7509348, -105.2139426], [39.7511093, -105.2141093], [39.7516271, -105.2146038], [39.7516716, -105.2146463], [39.751739, -105.2147085], [39.7524829, -105.2153953], [39.7525012, -105.2154123], [39.7532517, -105.2161101], [39.7533052, -105.2161598], [39.7540716, -105.2168674], [39.7541308, -105.2169221], [39.7544795, -105.2172508], [39.7546281, -105.2174522], [39.7546745, -105.2175037], [39.7548394, -105.2176869], [39.7549112, -105.2177871], [39.7548521, -105.2179433], [39.7546476, -105.2183047], [39.7544379, -105.2186805], [39.7544293, -105.218688], [39.7543723, -105.2187394], [39.7544376, -105.2188002], [39.7544872, -105.2188464], [39.7547934, -105.2191313], [39.7551071, -105.2194077], [39.7551368, -105.2194339], [39.755202, -105.2194913], [39.7552559, -105.2195411], [39.755615, -105.2198733], [39.755696, -105.2199479], [39.7559351, -105.2201693], [39.7559559, -105.220191], [39.7560109, -105.2202486], [39.7560703, -105.2203182], [39.756257, -105.2205368], [39.7562813, -105.2205598], [39.7564057, -105.2206774], [39.7565461, -105.2208058], [39.7567417, -105.2209442], [39.7566986, -105.2211001], [39.756638, -105.2213112], [39.7565732, -105.2214643], [39.7565209, -105.2215678], [39.7564595, -105.2216814], [39.7562914, -105.2219923], [39.7562254, -105.2221143], [39.7563006, -105.2221853], [39.7566193, -105.222483], [39.7566315, -105.2224957], [39.7569475, -105.2227913], [39.7570561, -105.222895], [39.7574792, -105.2232693], [39.7575579, -105.2233409], [39.7576236, -105.2234019], [39.7578565, -105.2236199], [39.7579703, -105.2237258], [39.7580027, -105.2237561], [39.7583001, -105.2240544], [39.7583591, -105.2241084], [39.7587249, -105.2244446], [39.7588057, -105.22452], [39.7591609, -105.2248516], [39.7595089, -105.2251741], [39.7596887, -105.2253408], [39.7597554, -105.2254024], [39.7598802, -105.2255173], [39.7598939, -105.2255305], [39.7599658, -105.2255988], [39.7600186, -105.2256498], [39.7604008, -105.2259941], [39.7605066, -105.2260931], [39.7605924, -105.226174], [39.760667, -105.2262427], [39.7607946, -105.2263631], [39.7608753, -105.2264373], [39.7615829, -105.2271044], [39.7616266, -105.2271481], [39.7616788, -105.2272008], [39.762203, -105.2276943], [39.7623411, -105.2278299], [39.7630405, -105.2284565], [39.7632824, -105.2286685], [39.7637358, -105.2290679], [39.7640137, -105.229309], [39.7640954, -105.229381], [39.7641743, -105.229462], [39.76436, -105.2296495], [39.7645376, -105.2298658], [39.764824, -105.2301918], [39.7652051, -105.2307044], [39.7654711, -105.2310802], [39.7654801, -105.2310934], [39.7655308, -105.231168], [39.7655792, -105.231232], [39.7655918, -105.2312486], [39.7657719, -105.2315035], [39.7658973, -105.2316751], [39.7659962, -105.2318039], [39.7661018, -105.231924], [39.7662469, -105.2320614], [39.7664515, -105.2322588], [39.7671442, -105.2329283], [39.7673559, -105.2331244], [39.7674414, -105.2332053], [39.7675186, -105.2332845], [39.7675756, -105.2333522], [39.7676291, -105.2334243], [39.7676969, -105.2335369], [39.7677304, -105.233602], [39.767761, -105.2336649], [39.7677908, -105.233739], [39.7678172, -105.233812], [39.7678551, -105.2339422], [39.7678736, -105.2340294], [39.7678904, -105.2341257], [39.7679032, -105.2342329], [39.7679095, -105.2343359], [39.7679068, -105.2344027], [39.7679029, -105.234499], [39.7678875, -105.2346626], [39.7678651, -105.2347791], [39.7678384, -105.2348898], [39.7678006, -105.2350283], [39.7677517, -105.2352031], [39.7677119, -105.2353468]],\n", " {"bubblingMouseEvents": true, "color": "green", "dashArray": null, "dashOffset": null, "fill": false, "fillColor": "green", "fillOpacity": 0.2, "fillRule": "evenodd", "lineCap": "round", "lineJoin": "round", "noClip": false, "opacity": 0.8, "smoothFactor": 1.0, "stroke": true, "weight": 10}\n", - " ).addTo(map_5e6fdd78aee60d0d6315972837ed641b);\n", + " ).addTo(map_f0bafdc10b40d238de9da7b18732b217);\n", " \n", " \n", - " poly_line_3349ea569e76e0f0fe319322fd08afcb.bindTooltip(\n", + " poly_line_95e6364ccd9609756d15b7d7194b0d25.bindTooltip(\n", " `<div>\n", " Least Energy\n", " </div>`,\n", @@ -545,19 +534,19 @@ " );\n", " \n", " \n", - " var marker_b9a3acafc0bb939f9f5fb5f2fa8ba7f7 = L.marker(\n", + " var marker_b7c7d5ee54034a20d3fee0fe3d1733e2 = L.marker(\n", " [39.7273069, -105.1994169],\n", " {}\n", - " ).addTo(map_5e6fdd78aee60d0d6315972837ed641b);\n", + " ).addTo(map_f0bafdc10b40d238de9da7b18732b217);\n", " \n", " \n", - " var icon_d6912a1b5e7e079949d11e32e00dea90 = L.AwesomeMarkers.icon(\n", + " var icon_a0c3146e5134d59292b0627e0c924a1f = L.AwesomeMarkers.icon(\n", " {"extraClasses": "fa-rotate-0", "icon": "circle", "iconColor": "white", "markerColor": "green", "prefix": "fa"}\n", " );\n", - " marker_b9a3acafc0bb939f9f5fb5f2fa8ba7f7.setIcon(icon_d6912a1b5e7e079949d11e32e00dea90);\n", + " marker_b7c7d5ee54034a20d3fee0fe3d1733e2.setIcon(icon_a0c3146e5134d59292b0627e0c924a1f);\n", " \n", " \n", - " marker_b9a3acafc0bb939f9f5fb5f2fa8ba7f7.bindTooltip(\n", + " marker_b7c7d5ee54034a20d3fee0fe3d1733e2.bindTooltip(\n", " `<div>\n", " Origin\n", " </div>`,\n", @@ -565,19 +554,19 @@ " );\n", " \n", " \n", - " var marker_d09760fd1b167457ff8d82085d575f1e = L.marker(\n", + " var marker_6fa879a02d34f1b588c3db5c0a5b6446 = L.marker(\n", " [39.7677119, -105.2353468],\n", " {}\n", - " ).addTo(map_5e6fdd78aee60d0d6315972837ed641b);\n", + " ).addTo(map_f0bafdc10b40d238de9da7b18732b217);\n", " \n", " \n", - " var icon_068d667c897f021e6c99fb3531ba48e3 = L.AwesomeMarkers.icon(\n", + " var icon_e58bcdf28679f4df3d0e2476c62b2048 = L.AwesomeMarkers.icon(\n", " {"extraClasses": "fa-rotate-0", "icon": "circle", "iconColor": "white", "markerColor": "red", "prefix": "fa"}\n", " );\n", - " marker_d09760fd1b167457ff8d82085d575f1e.setIcon(icon_068d667c897f021e6c99fb3531ba48e3);\n", + " marker_6fa879a02d34f1b588c3db5c0a5b6446.setIcon(icon_e58bcdf28679f4df3d0e2476c62b2048);\n", " \n", " \n", - " marker_d09760fd1b167457ff8d82085d575f1e.bindTooltip(\n", + " marker_6fa879a02d34f1b588c3db5c0a5b6446.bindTooltip(\n", " `<div>\n", " Destination\n", " </div>`,\n", @@ -588,10 +577,10 @@ "</html>\" style=\"position:absolute;width:100%;height:100%;left:0;top:0;border:none !important;\" allowfullscreen webkitallowfullscreen mozallowfullscreen></iframe></div></div>" ], "text/plain": [ - "<folium.folium.Map at 0x123135d10>" + "<folium.folium.Map at 0x1231d5ad0>" ] }, - "execution_count": 15, + "execution_count": 14, "metadata": {}, "output_type": "execute_result" } @@ -612,7 +601,7 @@ }, { "cell_type": "code", - "execution_count": 16, + "execution_count": 15, "id": "aa1e6b11", "metadata": {}, "outputs": [ @@ -646,7 +635,7 @@ " <meta name="viewport" content="width=device-width,\n", " initial-scale=1.0, maximum-scale=1.0, user-scalable=no" />\n", " <style>\n", - " #map_f5e9fdbd4de71b8ef0064aeb6f947a41 {\n", + " #map_ad8e071bae664cc7b1a973d4509a33a8 {\n", " position: relative;\n", " width: 100.0%;\n", " height: 100.0%;\n", @@ -660,14 +649,14 @@ "<body>\n", " \n", " \n", - " <div class="folium-map" id="map_f5e9fdbd4de71b8ef0064aeb6f947a41" ></div>\n", + " <div class="folium-map" id="map_ad8e071bae664cc7b1a973d4509a33a8" ></div>\n", " \n", "</body>\n", "<script>\n", " \n", " \n", - " var map_f5e9fdbd4de71b8ef0064aeb6f947a41 = L.map(\n", - " "map_f5e9fdbd4de71b8ef0064aeb6f947a41",\n", + " var map_ad8e071bae664cc7b1a973d4509a33a8 = L.map(\n", + " "map_ad8e071bae664cc7b1a973d4509a33a8",\n", " {\n", " center: [39.7465774, -105.2216276],\n", " crs: L.CRS.EPSG3857,\n", @@ -681,39 +670,39 @@ "\n", " \n", " \n", - " var tile_layer_b22ff300046905100f9b524af60a2514 = L.tileLayer(\n", + " var tile_layer_5afdab62ec2b889ffe53761579c5a24a = L.tileLayer(\n", " "https://{s}.tile.openstreetmap.org/{z}/{x}/{y}.png",\n", " {"attribution": "Data by \\u0026copy; \\u003ca target=\\"_blank\\" href=\\"http://openstreetmap.org\\"\\u003eOpenStreetMap\\u003c/a\\u003e, under \\u003ca target=\\"_blank\\" href=\\"http://www.openstreetmap.org/copyright\\"\\u003eODbL\\u003c/a\\u003e.", "detectRetina": false, "maxNativeZoom": 18, "maxZoom": 18, "minZoom": 0, "noWrap": false, "opacity": 1, "subdomains": "abc", "tms": false}\n", - " ).addTo(map_f5e9fdbd4de71b8ef0064aeb6f947a41);\n", + " ).addTo(map_ad8e071bae664cc7b1a973d4509a33a8);\n", " \n", " \n", - " var poly_line_8ee9a6fcbe409c7f61afa98464585f6a = L.polyline(\n", + " var poly_line_da49ac63cdfeea189953099ec23dbaf5 = L.polyline(\n", " [[39.7273069, -105.1994169], [39.7272989, -105.1994535], [39.7272848, -105.1994868], [39.7272593, -105.1995218], [39.7272272, -105.1995461], [39.7271912, -105.1995576], [39.7271688, -105.199558], [39.7271461, -105.1995532], [39.7271142, -105.1995365], [39.7270868, -105.1995096], [39.7270657, -105.1994741], [39.7270519, -105.1994296], [39.7270483, -105.1993816], [39.727052, -105.1993539], [39.7270571, -105.1993247], [39.7270719, -105.1992916], [39.7270973, -105.1992566], [39.7271293, -105.1992323], [39.7271651, -105.1992207], [39.7271854, -105.1992226], [39.7272113, -105.199227], [39.7273496, -105.1991687], [39.7274185, -105.1991465], [39.7275326, -105.1990928], [39.7276828, -105.1990602], [39.7277455, -105.1990518], [39.7279702, -105.1990232], [39.7281558, -105.1990357], [39.7284384, -105.1990893], [39.7286639, -105.199177], [39.7288222, -105.1992773], [39.7289214, -105.199385], [39.7290292, -105.1995543], [39.7290506, -105.1995879], [39.7292157, -105.1998155], [39.7293094, -105.1998945], [39.7294079, -105.1999604], [39.7295437, -105.200014], [39.7296299, -105.2000721], [39.7297292, -105.2000757], [39.7297135, -105.2007081], [39.7297192, -105.2013106], [39.7298153, -105.2032041], [39.7298256, -105.2033784], [39.7298664, -105.2035708], [39.7298795, -105.203649], [39.7299267, -105.2038558], [39.7301082, -105.2042608], [39.7304911, -105.2048068], [39.7308741, -105.2053111], [39.7310676, -105.2055985], [39.7311933, -105.205824], [39.7312069, -105.2058543], [39.7313075, -105.2061158], [39.731348, -105.206248], [39.731386, -105.2064213], [39.7314237, -105.2067336], [39.731429, -105.2070565], [39.7314226, -105.207224], [39.7314076, -105.2074036], [39.7313611, -105.2077675], [39.7313265, -105.2079124], [39.7312831, -105.20806], [39.7312292, -105.2082086], [39.7311706, -105.208344], [39.7310965, -105.2084777], [39.7310049, -105.208616], [39.7309234, -105.2087235], [39.7308212, -105.2088303], [39.7306227, -105.2089819], [39.7304801, -105.2090736], [39.7303681, -105.2091988], [39.7303327, -105.2092597], [39.7303114, -105.2093181], [39.730299, -105.20939], [39.7302939, -105.2095126], [39.7303105, -105.2096804], [39.7304414, -105.2099267], [39.7305001, -105.2100344], [39.7305871, -105.2101842], [39.7307404, -105.2104481], [39.7309234, -105.2107229], [39.7310934, -105.2109639], [39.731238, -105.211145], [39.7314101, -105.2113464], [39.7315783, -105.2115336], [39.7317541, -105.2117064], [39.7319796, -105.2119089], [39.732359, -105.2122393], [39.7329187, -105.2127184], [39.7334557, -105.213188], [39.7338968, -105.2135676], [39.7342777, -105.2138968], [39.7347139, -105.2142786], [39.7354652, -105.2149301], [39.7356319, -105.2150733], [39.7358844, -105.2152902], [39.7363901, -105.2157275], [39.7369129, -105.2161859], [39.7377243, -105.2168831], [39.7390882, -105.2180682], [39.7397882, -105.2186718], [39.7403043, -105.2191078], [39.7405645, -105.2193336], [39.7406767, -105.2194225], [39.7415803, -105.2202142], [39.7416486, -105.2202727], [39.7417047, -105.2203151], [39.7421502, -105.2206932], [39.7425816, -105.221062], [39.7432687, -105.2215398], [39.7435797, -105.2217631], [39.7436721, -105.2218299], [39.7437366, -105.2218764], [39.7438794, -105.2219759], [39.7439546, -105.2220253], [39.7440082, -105.2220586], [39.7440819, -105.2221021], [39.7441377, -105.2221322], [39.7441808, -105.2221541], [39.7442796, -105.2222043], [39.7443357, -105.2222286], [39.7444062, -105.222259], [39.7444657, -105.2222871], [39.7444962, -105.2223029], [39.7445695, -105.2223431], [39.7446343, -105.2223808], [39.744683, -105.2224104], [39.7447215, -105.2224349], [39.7447658, -105.2224654], [39.744813, -105.222499], [39.7448547, -105.2225315], [39.7448881, -105.2225586], [39.7449328, -105.2225965], [39.7449815, -105.2226391], [39.7451542, -105.2227925], [39.7454784, -105.2230438], [39.7455034, -105.2230623], [39.7455159, -105.2230713], [39.7455403, -105.2230861], [39.7455548, -105.2230941], [39.7455766, -105.2231014], [39.7456018, -105.223106], [39.7456176, -105.2231071], [39.7456337, -105.223107], [39.7456638, -105.2231045], [39.7456814, -105.2231011], [39.7456983, -105.2230945], [39.7457224, -105.2230821], [39.7457338, -105.2230745], [39.7457588, -105.2230543], [39.7457709, -105.223043], [39.7457834, -105.2230284], [39.7458035, -105.2230027], [39.7458905, -105.2228787], [39.7460068, -105.2226624], [39.7461589, -105.2223894], [39.7463617, -105.2220126], [39.7464427, -105.2218672], [39.746482, -105.2217968], [39.7465202, -105.2217322], [39.7465523, -105.2216771], [39.7465774, -105.2216276], [39.7465928, -105.2215823], [39.7466004, -105.2215466], [39.7466042, -105.2215101], [39.7466044, -105.2214695], [39.746601, -105.2214368], [39.7465881, -105.2213862], [39.7465804, -105.2213567], [39.7465774, -105.2213321], [39.7465774, -105.2213048], [39.746581, -105.2212762], [39.7465874, -105.2212518], [39.7465978, -105.2212266], [39.7466125, -105.2212024], [39.7466266, -105.2211858], [39.7466485, -105.2211678], [39.7466708, -105.2211566], [39.746693, -105.2211511], [39.746715, -105.2211506], [39.7467203, -105.2211517], [39.7467798, -105.2211327], [39.7468101, -105.2211204], [39.7468347, -105.2211032], [39.7468673, -105.2210676], [39.7468863, -105.2210389], [39.7469079, -105.2209995], [39.7469752, -105.2208699], [39.7471073, -105.220655], [39.7471371, -105.2206129], [39.7471594, -105.2205845], [39.7472061, -105.2205376], [39.7473376, -105.2202962], [39.747357, -105.2202177], [39.7473682, -105.2201868], [39.7473848, -105.2201519], [39.7475089, -105.2199249], [39.7475861, -105.2197973], [39.7476254, -105.2197401], [39.7476917, -105.2196525], [39.7477724, -105.2195069], [39.7478875, -105.219299], [39.7479219, -105.2192372], [39.7479518, -105.219179], [39.7480957, -105.2189201], [39.7481337, -105.2188515], [39.7482567, -105.2186294], [39.7484851, -105.2182271], [39.7485167, -105.218164], [39.7485605, -105.2180881], [39.7490516, -105.217166], [39.7490955, -105.2170866], [39.7491263, -105.2170387], [39.7493736, -105.2165801], [39.7496841, -105.2160128], [39.7497482, -105.2160733], [39.7504593, -105.216753], [39.7505084, -105.2168007], [39.7505733, -105.2168567], [39.7509525, -105.2172063], [39.7512593, -105.2174884], [39.7513304, -105.2175506], [39.7513869, -105.217604], [39.7517458, -105.2179367], [39.7520945, -105.21826], [39.7521462, -105.2183047], [39.7522126, -105.218369], [39.7523012, -105.2184523], [39.7525052, -105.218642], [39.7527145, -105.2188366], [39.7528959, -105.2189993], [39.7529601, -105.2190649], [39.7530308, -105.219128], [39.7537033, -105.2197718], [39.753764, -105.2198302], [39.7538299, -105.2198915], [39.7542346, -105.2202638], [39.7545279, -105.2205433], [39.7545847, -105.2206031], [39.7546623, -105.2206728], [39.7553326, -105.2212752], [39.755407, -105.221342], [39.7554908, -105.2214211], [39.7561483, -105.2220415], [39.7562254, -105.2221143], [39.7563006, -105.2221853], [39.7566193, -105.222483], [39.7566315, -105.2224957], [39.7569475, -105.2227913], [39.7570561, -105.222895], [39.7574792, -105.2232693], [39.7575579, -105.2233409], [39.7576236, -105.2234019], [39.7578565, -105.2236199], [39.7579703, -105.2237258], [39.7580027, -105.2237561], [39.7583001, -105.2240544], [39.7583591, -105.2241084], [39.7587249, -105.2244446], [39.7588057, -105.22452], [39.7591609, -105.2248516], [39.7595089, -105.2251741], [39.7596887, -105.2253408], [39.7597554, -105.2254024], [39.7598802, -105.2255173], [39.7598939, -105.2255305], [39.7599658, -105.2255988], [39.7600186, -105.2256498], [39.7604008, -105.2259941], [39.7605066, -105.2260931], [39.7605924, -105.226174], [39.760667, -105.2262427], [39.7607946, -105.2263631], [39.7608753, -105.2264373], [39.7615829, -105.2271044], [39.7616266, -105.2271481], [39.7616788, -105.2272008], [39.762203, -105.2276943], [39.7623411, -105.2278299], [39.7630405, -105.2284565], [39.7632824, -105.2286685], [39.7637358, -105.2290679], [39.7640137, -105.229309], [39.7640954, -105.229381], [39.7641743, -105.229462], [39.76436, -105.2296495], [39.7645376, -105.2298658], [39.764824, -105.2301918], [39.7652051, -105.2307044], [39.7654711, -105.2310802], [39.7654801, -105.2310934], [39.7655308, -105.231168], [39.7655792, -105.231232], [39.7655918, -105.2312486], [39.7657719, -105.2315035], [39.7658973, -105.2316751], [39.7659962, -105.2318039], [39.7661018, -105.231924], [39.7662469, -105.2320614], [39.7664515, -105.2322588], [39.7671442, -105.2329283], [39.7673559, -105.2331244], [39.7674414, -105.2332053], [39.7675186, -105.2332845], [39.7675756, -105.2333522], [39.7676291, -105.2334243], [39.7676969, -105.2335369], [39.7677304, -105.233602], [39.767761, -105.2336649], [39.7677908, -105.233739], [39.7678172, -105.233812], [39.7678551, -105.2339422], [39.7678736, -105.2340294], [39.7678904, -105.2341257], [39.7679032, -105.2342329], [39.7679095, -105.2343359], [39.7679068, -105.2344027], [39.7679029, -105.234499], [39.7678875, -105.2346626], [39.7678651, -105.2347791], [39.7678384, -105.2348898], [39.7678006, -105.2350283], [39.7677517, -105.2352031], [39.7677119, -105.2353468]],\n", " {"bubblingMouseEvents": true, "color": "#eff821", "dashArray": null, "dashOffset": null, "fill": false, "fillColor": "#eff821", "fillOpacity": 0.2, "fillRule": "evenodd", "lineCap": "round", "lineJoin": "round", "noClip": false, "opacity": 0.8, "smoothFactor": 1.0, "stroke": true, "weight": 10}\n", - " ).addTo(map_f5e9fdbd4de71b8ef0064aeb6f947a41);\n", + " ).addTo(map_ad8e071bae664cc7b1a973d4509a33a8);\n", " \n", " \n", - " poly_line_8ee9a6fcbe409c7f61afa98464585f6a.bindTooltip(\n", + " poly_line_da49ac63cdfeea189953099ec23dbaf5.bindTooltip(\n", " `<div>\n", - " 0.18821565778485838\n", + " 0.18816139103031826\n", " </div>`,\n", " {"sticky": true}\n", " );\n", " \n", " \n", - " var marker_f42f7bb5b3e19865927ef3411c0a0a48 = L.marker(\n", + " var marker_e384d848ce36e7fbaacb0ebe3c380791 = L.marker(\n", " [39.7273069, -105.1994169],\n", " {}\n", - " ).addTo(map_f5e9fdbd4de71b8ef0064aeb6f947a41);\n", + " ).addTo(map_ad8e071bae664cc7b1a973d4509a33a8);\n", " \n", " \n", - " var icon_15aeb0dc4825e560514b0f26c8c5cb20 = L.AwesomeMarkers.icon(\n", + " var icon_b55345fc80c9f7480e986bedc3b62179 = L.AwesomeMarkers.icon(\n", " {"extraClasses": "fa-rotate-0", "icon": "circle", "iconColor": "white", "markerColor": "green", "prefix": "fa"}\n", " );\n", - " marker_f42f7bb5b3e19865927ef3411c0a0a48.setIcon(icon_15aeb0dc4825e560514b0f26c8c5cb20);\n", + " marker_e384d848ce36e7fbaacb0ebe3c380791.setIcon(icon_b55345fc80c9f7480e986bedc3b62179);\n", " \n", " \n", - " marker_f42f7bb5b3e19865927ef3411c0a0a48.bindTooltip(\n", + " marker_e384d848ce36e7fbaacb0ebe3c380791.bindTooltip(\n", " `<div>\n", " Origin\n", " </div>`,\n", @@ -721,19 +710,19 @@ " );\n", " \n", " \n", - " var marker_bd64e10e467bfaa6054456bc65c65dcf = L.marker(\n", + " var marker_0e442f1cd248b85e7b27087a06524f49 = L.marker(\n", " [39.7677119, -105.2353468],\n", " {}\n", - " ).addTo(map_f5e9fdbd4de71b8ef0064aeb6f947a41);\n", + " ).addTo(map_ad8e071bae664cc7b1a973d4509a33a8);\n", " \n", " \n", - " var icon_3e1f9de6460d5c0512d0986e97601baa = L.AwesomeMarkers.icon(\n", + " var icon_dcd2bb1e268f16dded3a453f250931ef = L.AwesomeMarkers.icon(\n", " {"extraClasses": "fa-rotate-0", "icon": "circle", "iconColor": "white", "markerColor": "red", "prefix": "fa"}\n", " );\n", - " marker_bd64e10e467bfaa6054456bc65c65dcf.setIcon(icon_3e1f9de6460d5c0512d0986e97601baa);\n", + " marker_0e442f1cd248b85e7b27087a06524f49.setIcon(icon_dcd2bb1e268f16dded3a453f250931ef);\n", " \n", " \n", - " marker_bd64e10e467bfaa6054456bc65c65dcf.bindTooltip(\n", + " marker_0e442f1cd248b85e7b27087a06524f49.bindTooltip(\n", " `<div>\n", " Destination\n", " </div>`,\n", @@ -741,33 +730,33 @@ " );\n", " \n", " \n", - " var poly_line_63ad4d4e5561017b7e7fd40a459bf578 = L.polyline(\n", + " var poly_line_453a4ea787ffe3c7d6510177b14e1d10 = L.polyline(\n", " [[39.7273069, -105.1994169], [39.7272989, -105.1994535], [39.7272848, -105.1994868], [39.7272593, -105.1995218], [39.7272272, -105.1995461], [39.7271912, -105.1995576], [39.7271688, -105.199558], [39.7271461, -105.1995532], [39.7271142, -105.1995365], [39.7270868, -105.1995096], [39.7270657, -105.1994741], [39.7270519, -105.1994296], [39.7270483, -105.1993816], [39.727052, -105.1993539], [39.7270571, -105.1993247], [39.7270719, -105.1992916], [39.7270973, -105.1992566], [39.7271293, -105.1992323], [39.7271651, -105.1992207], [39.7271854, -105.1992226], [39.7272113, -105.199227], [39.7273496, -105.1991687], [39.7274185, -105.1991465], [39.7275326, -105.1990928], [39.7276828, -105.1990602], [39.7277455, -105.1990518], [39.7279702, -105.1990232], [39.7281558, -105.1990357], [39.7284384, -105.1990893], [39.7286639, -105.199177], [39.7288222, -105.1992773], [39.7289214, -105.199385], [39.7290292, -105.1995543], [39.7290506, -105.1995879], [39.7292157, -105.1998155], [39.7293094, -105.1998945], [39.7294079, -105.1999604], [39.7295437, -105.200014], [39.7296299, -105.2000721], [39.7296469, -105.1996789], [39.7296738, -105.1987284], [39.7296906, -105.1986794], [39.7297244, -105.1985856], [39.7297272, -105.1984193], [39.7299723, -105.1984198], [39.7302117, -105.1984227], [39.7303235, -105.1984285], [39.7304434, -105.198443], [39.7305651, -105.1984732], [39.7307182, -105.1985167], [39.7308487, -105.1985623], [39.7309909, -105.1986311], [39.731127, -105.1987144], [39.7312292, -105.1987897], [39.7313292, -105.1988731], [39.7314292, -105.1989611], [39.731524, -105.1990554], [39.7316089, -105.1991466], [39.7317085, -105.1992639], [39.7318575, -105.1994488], [39.7319522, -105.1995697], [39.7320423, -105.1996716], [39.7321372, -105.1997578], [39.7322615, -105.1998506], [39.7323697, -105.1999163], [39.7324712, -105.1999637], [39.7325898, -105.2000081], [39.732689, -105.2000309], [39.7327987, -105.2000627], [39.732945, -105.2001049], [39.7330635, -105.2001374], [39.7331838, -105.2001914], [39.7332899, -105.2002528], [39.7334312, -105.200363], [39.7335257, -105.2004544], [39.7335523, -105.2004835], [39.7336132, -105.2005503], [39.7336902, -105.2006555], [39.733767, -105.2007557], [39.7338411, -105.2008491], [39.7339359, -105.2009412], [39.7340357, -105.2010154], [39.7340742, -105.201042], [39.7341117, -105.2010678], [39.7341542, -105.2010901], [39.734211, -105.2011198], [39.7343173, -105.2011712], [39.7344056, -105.2012021], [39.7345704, -105.2012376], [39.7347726, -105.2012753], [39.7351329, -105.2013311], [39.7352203, -105.2013441], [39.7353243, -105.201352], [39.7354415, -105.2013441], [39.7355491, -105.2013268], [39.7357022, -105.2012901], [39.7358395, -105.201259], [39.7362429, -105.2011675], [39.7367157, -105.2010639], [39.7367857, -105.2010539], [39.7368531, -105.2010442], [39.7369645, -105.2010356], [39.7370755, -105.201044], [39.7371895, -105.2010632], [39.7373179, -105.201099], [39.737714, -105.2012295], [39.7379567, -105.2013009], [39.7384451, -105.2014404], [39.7386872, -105.2015167], [39.7395717, -105.2017905], [39.7397153, -105.2018369], [39.7398269, -105.2018673], [39.7399356, -105.2018843], [39.7400618, -105.2018961], [39.740184, -105.2018932], [39.7403133, -105.2018823], [39.7404288, -105.2018645], [39.7405518, -105.2018387], [39.740665, -105.2018073], [39.7407597, -105.2017713], [39.7408728, -105.2017196], [39.740992, -105.2016551], [39.7411268, -105.2015625], [39.7412599, -105.2014489], [39.7413718, -105.2013453], [39.7414758, -105.2012326], [39.741608, -105.2010691], [39.7416901, -105.200964], [39.7417903, -105.2008396], [39.7417994, -105.200826], [39.7418968, -105.200686], [39.7418878, -105.200623], [39.7418917, -105.2005592], [39.7419081, -105.2004988], [39.7419393, -105.2004416], [39.7419818, -105.2003982], [39.7420319, -105.2003723], [39.7420864, -105.2003661], [39.7421399, -105.2003805], [39.742188, -105.2004144], [39.7422255, -105.2004628], [39.742251, -105.2005229], [39.7422628, -105.2005898], [39.7422597, -105.2006584], [39.7422421, -105.2007231], [39.7422712, -105.2008198], [39.7423227, -105.200901], [39.7428069, -105.2016549], [39.7429048, -105.2018114], [39.7429131, -105.2018245], [39.7431503, -105.2022035], [39.7434374, -105.2026477], [39.7435807, -105.2028659], [39.7437705, -105.2031843], [39.7441255, -105.2037193], [39.7444536, -105.2042289], [39.7445369, -105.204369], [39.7448087, -105.2048085], [39.7450389, -105.205197], [39.745164, -105.2054397], [39.7452617, -105.205642], [39.745359, -105.2058753], [39.7454413, -105.2060693], [39.7454698, -105.2061365], [39.7455586, -105.2063498], [39.7456394, -105.2066197], [39.7457343, -105.2068984], [39.7458096, -105.2071619], [39.7458432, -105.2072893], [39.7459057, -105.2075162], [39.7459547, -105.2077463], [39.7460104, -105.2080066], [39.7460312, -105.2081149], [39.7460508, -105.2082709], [39.7460836, -105.2084787], [39.7461092, -105.2086125], [39.7461165, -105.2087078], [39.7461128, -105.2088019], [39.7461078, -105.2089177], [39.7461163, -105.208995], [39.7461377, -105.2090672], [39.7461588, -105.2091147], [39.7462509, -105.2092159], [39.746281, -105.2092382], [39.7462978, -105.209256], [39.7463242, -105.209292], [39.7463458, -105.2093473], [39.746361, -105.2094251], [39.7463921, -105.2095768], [39.7464353, -105.2097133], [39.7465001, -105.2098285], [39.7467097, -105.21004], [39.7467814, -105.2101094], [39.7473357, -105.2106047], [39.7474554, -105.2107116], [39.7475615, -105.2108064], [39.7475958, -105.2108371], [39.747653, -105.2108907], [39.7481502, -105.2113567], [39.7483361, -105.2115309], [39.7483726, -105.2115651], [39.7484183, -105.2116079], [39.7484718, -105.2116584], [39.7487212, -105.2118951], [39.749238, -105.2123721], [39.7492814, -105.2124135], [39.7500059, -105.2130759], [39.7500454, -105.2131129], [39.7503812, -105.2134256], [39.7504627, -105.2135015], [39.7508219, -105.2138362], [39.7508792, -105.2138895], [39.7509348, -105.2139426], [39.7511093, -105.2141093], [39.7516271, -105.2146038], [39.7516716, -105.2146463], [39.751739, -105.2147085], [39.7524829, -105.2153953], [39.7525012, -105.2154123], [39.7532517, -105.2161101], [39.7533052, -105.2161598], [39.7540716, -105.2168674], [39.7541308, -105.2169221], [39.7544795, -105.2172508], [39.7546281, -105.2174522], [39.7546745, -105.2175037], [39.7548394, -105.2176869], [39.7549112, -105.2177871], [39.7548521, -105.2179433], [39.7546476, -105.2183047], [39.7544379, -105.2186805], [39.7544293, -105.218688], [39.7543723, -105.2187394], [39.7544376, -105.2188002], [39.7544872, -105.2188464], [39.7547934, -105.2191313], [39.7551071, -105.2194077], [39.7551368, -105.2194339], [39.755202, -105.2194913], [39.7552559, -105.2195411], [39.755615, -105.2198733], [39.755696, -105.2199479], [39.7559351, -105.2201693], [39.7559559, -105.220191], [39.7560109, -105.2202486], [39.7560703, -105.2203182], [39.756257, -105.2205368], [39.7562813, -105.2205598], [39.7564057, -105.2206774], [39.7565461, -105.2208058], [39.7567417, -105.2209442], [39.7566986, -105.2211001], [39.756638, -105.2213112], [39.7565732, -105.2214643], [39.7565209, -105.2215678], [39.7564595, -105.2216814], [39.7562914, -105.2219923], [39.7562254, -105.2221143], [39.7563006, -105.2221853], [39.7566193, -105.222483], [39.7566315, -105.2224957], [39.7569475, -105.2227913], [39.7570561, -105.222895], [39.7574792, -105.2232693], [39.7575579, -105.2233409], [39.7576236, -105.2234019], [39.7578565, -105.2236199], [39.7579703, -105.2237258], [39.7580027, -105.2237561], [39.7583001, -105.2240544], [39.7583591, -105.2241084], [39.7587249, -105.2244446], [39.7588057, -105.22452], [39.7591609, -105.2248516], [39.7595089, -105.2251741], [39.7596887, -105.2253408], [39.7597554, -105.2254024], [39.7598802, -105.2255173], [39.7598939, -105.2255305], [39.7599658, -105.2255988], [39.7600186, -105.2256498], [39.7604008, -105.2259941], [39.7605066, -105.2260931], [39.7605924, -105.226174], [39.760667, -105.2262427], [39.7607946, -105.2263631], [39.7608753, -105.2264373], [39.7615829, -105.2271044], [39.7616266, -105.2271481], [39.7616788, -105.2272008], [39.762203, -105.2276943], [39.7623411, -105.2278299], [39.7630405, -105.2284565], [39.7632824, -105.2286685], [39.7637358, -105.2290679], [39.7640137, -105.229309], [39.7640954, -105.229381], [39.7641743, -105.229462], [39.76436, -105.2296495], [39.7645376, -105.2298658], [39.764824, -105.2301918], [39.7652051, -105.2307044], [39.7654711, -105.2310802], [39.7654801, -105.2310934], [39.7655308, -105.231168], [39.7655792, -105.231232], [39.7655918, -105.2312486], [39.7657719, -105.2315035], [39.7658973, -105.2316751], [39.7659962, -105.2318039], [39.7661018, -105.231924], [39.7662469, -105.2320614], [39.7664515, -105.2322588], [39.7671442, -105.2329283], [39.7673559, -105.2331244], [39.7674414, -105.2332053], [39.7675186, -105.2332845], [39.7675756, -105.2333522], [39.7676291, -105.2334243], [39.7676969, -105.2335369], [39.7677304, -105.233602], [39.767761, -105.2336649], [39.7677908, -105.233739], [39.7678172, -105.233812], [39.7678551, -105.2339422], [39.7678736, -105.2340294], [39.7678904, -105.2341257], [39.7679032, -105.2342329], [39.7679095, -105.2343359], [39.7679068, -105.2344027], [39.7679029, -105.234499], [39.7678875, -105.2346626], [39.7678651, -105.2347791], [39.7678384, -105.2348898], [39.7678006, -105.2350283], [39.7677517, -105.2352031], [39.7677119, -105.2353468]],\n", " {"bubblingMouseEvents": true, "color": "#0c0786", "dashArray": null, "dashOffset": null, "fill": false, "fillColor": "#0c0786", "fillOpacity": 0.2, "fillRule": "evenodd", "lineCap": "round", "lineJoin": "round", "noClip": false, "opacity": 0.8, "smoothFactor": 1.0, "stroke": true, "weight": 10}\n", - " ).addTo(map_f5e9fdbd4de71b8ef0064aeb6f947a41);\n", + " ).addTo(map_ad8e071bae664cc7b1a973d4509a33a8);\n", " \n", " \n", - " poly_line_63ad4d4e5561017b7e7fd40a459bf578.bindTooltip(\n", + " poly_line_453a4ea787ffe3c7d6510177b14e1d10.bindTooltip(\n", " `<div>\n", - " 0.18722374836334246\n", + " 0.1871862788506351\n", " </div>`,\n", " {"sticky": true}\n", " );\n", " \n", " \n", - " var marker_57adc3ad6e7d8c7bd278d30d1df431de = L.marker(\n", + " var marker_e0b5ef061c2c427ae01d059e6d72c777 = L.marker(\n", " [39.7273069, -105.1994169],\n", " {}\n", - " ).addTo(map_f5e9fdbd4de71b8ef0064aeb6f947a41);\n", + " ).addTo(map_ad8e071bae664cc7b1a973d4509a33a8);\n", " \n", " \n", - " var icon_917c456f73b69b0b53fa8d8048d5f344 = L.AwesomeMarkers.icon(\n", + " var icon_5b733c3811ecc918d30b6cdbb7c63b27 = L.AwesomeMarkers.icon(\n", " {"extraClasses": "fa-rotate-0", "icon": "circle", "iconColor": "white", "markerColor": "green", "prefix": "fa"}\n", " );\n", - " marker_57adc3ad6e7d8c7bd278d30d1df431de.setIcon(icon_917c456f73b69b0b53fa8d8048d5f344);\n", + " marker_e0b5ef061c2c427ae01d059e6d72c777.setIcon(icon_5b733c3811ecc918d30b6cdbb7c63b27);\n", " \n", " \n", - " marker_57adc3ad6e7d8c7bd278d30d1df431de.bindTooltip(\n", + " marker_e0b5ef061c2c427ae01d059e6d72c777.bindTooltip(\n", " `<div>\n", " Origin\n", " </div>`,\n", @@ -775,19 +764,19 @@ " );\n", " \n", " \n", - " var marker_9f05f804eab779baf28827016a73a4e7 = L.marker(\n", + " var marker_ea7244601f3a5c7f675da2105a2d5354 = L.marker(\n", " [39.7677119, -105.2353468],\n", " {}\n", - " ).addTo(map_f5e9fdbd4de71b8ef0064aeb6f947a41);\n", + " ).addTo(map_ad8e071bae664cc7b1a973d4509a33a8);\n", " \n", " \n", - " var icon_9c0d129fcb8005579733c0f5eae70f74 = L.AwesomeMarkers.icon(\n", + " var icon_2266673043479453c0c2882dd6b522ed = L.AwesomeMarkers.icon(\n", " {"extraClasses": "fa-rotate-0", "icon": "circle", "iconColor": "white", "markerColor": "red", "prefix": "fa"}\n", " );\n", - " marker_9f05f804eab779baf28827016a73a4e7.setIcon(icon_9c0d129fcb8005579733c0f5eae70f74);\n", + " marker_ea7244601f3a5c7f675da2105a2d5354.setIcon(icon_2266673043479453c0c2882dd6b522ed);\n", " \n", " \n", - " marker_9f05f804eab779baf28827016a73a4e7.bindTooltip(\n", + " marker_ea7244601f3a5c7f675da2105a2d5354.bindTooltip(\n", " `<div>\n", " Destination\n", " </div>`,\n", @@ -798,10 +787,10 @@ "</html>\" style=\"position:absolute;width:100%;height:100%;left:0;top:0;border:none !important;\" allowfullscreen webkitallowfullscreen mozallowfullscreen></iframe></div></div>" ], "text/plain": [ - "<folium.folium.Map at 0x124205e50>" + "<folium.folium.Map at 0x1231d71d0>" ] }, - "execution_count": 16, + "execution_count": 15, "metadata": {}, "output_type": "execute_result" } diff --git a/python/nrel/routee/compass/io/generate_dataset.py b/python/nrel/routee/compass/io/generate_dataset.py index f3047028..cc58bfac 100644 --- a/python/nrel/routee/compass/io/generate_dataset.py +++ b/python/nrel/routee/compass/io/generate_dataset.py @@ -179,7 +179,7 @@ def replace_id(vertex_uuid): if filename == "osm_default_energy.toml": if add_grade: init_toml["traversal"][ - "grade_table_path" + "grade_table_file" ] = "edges-grade-enumerated.txt.gz" init_toml["traversal"]["grade_table_grade_unit"] = "decimal" @@ -196,7 +196,7 @@ def replace_id(vertex_uuid): init_toml["traversal"][ "real_world_energy_adjustment" ] = adjustment_factor - init_toml["traversal"]["energy_model_path"] = f"{energy_model}.bin" + init_toml["traversal"]["energy_model_file"] = f"{energy_model}.bin" energy_model_energy_rate_unit = ENERGY_OUTPUT_UNITS.get( powertrain_type diff --git a/python/nrel/routee/compass/resources/osm_default_distance.toml b/python/nrel/routee/compass/resources/osm_default_distance.toml index ee84bb3f..c9d6d012 100644 --- a/python/nrel/routee/compass/resources/osm_default_distance.toml +++ b/python/nrel/routee/compass/resources/osm_default_distance.toml @@ -1,8 +1,8 @@ parallelism = 2 [graph] -edge_list_csv = "edges-compass.csv.gz" -vertex_list_csv = "vertices-compass.csv.gz" +edge_list_file = "edges-compass.csv.gz" +vertex_list_file = "vertices-compass.csv.gz" verbose = true [traversal] diff --git a/python/nrel/routee/compass/resources/osm_default_energy.toml b/python/nrel/routee/compass/resources/osm_default_energy.toml index ba67777b..47e90ca2 100644 --- a/python/nrel/routee/compass/resources/osm_default_energy.toml +++ b/python/nrel/routee/compass/resources/osm_default_energy.toml @@ -1,14 +1,14 @@ parallelism = 2 [graph] -edge_list_csv = "edges-compass.csv.gz" -vertex_list_csv = "vertices-compass.csv.gz" +edge_list_file = "edges-compass.csv.gz" +vertex_list_file = "vertices-compass.csv.gz" verbose = true [traversal] type = "speed_grade_energy_model" model_type = "smartcore" -speed_table_path = "edges-posted-speed-enumerated.txt.gz" +speed_table_file = "edges-posted-speed-enumerated.txt.gz" speed_table_speed_unit = "kilometers_per_hour" energy_model_speed_unit = "miles_per_hour" energy_model_grade_unit = "decimal" diff --git a/python/nrel/routee/compass/resources/osm_default_speed.toml b/python/nrel/routee/compass/resources/osm_default_speed.toml index 245465dd..8a5279ca 100644 --- a/python/nrel/routee/compass/resources/osm_default_speed.toml +++ b/python/nrel/routee/compass/resources/osm_default_speed.toml @@ -1,13 +1,13 @@ parallelism = 2 [graph] -edge_list_csv = "edges-compass.csv.gz" -vertex_list_csv = "vertices-compass.csv.gz" +edge_list_file = "edges-compass.csv.gz" +vertex_list_file = "vertices-compass.csv.gz" verbose = true [traversal] type = "speed_table" -speed_table_path = "edges-posted-speed-enumerated.txt.gz" +speed_table_file = "edges-posted-speed-enumerated.txt.gz" speed_unit = "kilometers_per_hour" output_distance_unit = "miles" output_time_unit = "minutes" diff --git a/rust/routee-compass-core/src/algorithm/component/scc.rs b/rust/routee-compass-core/src/algorithm/component/scc.rs index 9a1fa8c8..d657c0b8 100644 --- a/rust/routee-compass-core/src/algorithm/component/scc.rs +++ b/rust/routee-compass-core/src/algorithm/component/scc.rs @@ -148,26 +148,50 @@ fn largest_strongly_connected_component(graph: &Graph) -> Result<Vec<VertexId>, #[cfg(test)] mod tests { use super::; - use crate::model::graph::graph_config::GraphConfig; - use std::path::PathBuf; + use crate::model::property::{edge::Edge, vertex::Vertex}; + use std::collections::HashMap; fn build_mock_graph() -> Graph { - // A test graph with 2 strongly connected components - let test_path = PathBuf::from(env!("CARGO_MANIFEST_DIR")) - .join("src") - .join("algorithm") - .join("component") - .join("test"); - let edge_list_csv = test_path.join("edges.csv").to_str().unwrap().to_string(); - let vertex_list_csv = test_path.join("vertices.csv").to_str().unwrap().to_string(); - let graph_conf = GraphConfig { - edge_list_csv, - vertex_list_csv, - n_edges: None, - n_vertices: None, - verbose: false, + let vertices = vec![ + Vertex::new(0, 0.0, 0.0), + Vertex::new(1, 1.0, 1.0), + Vertex::new(2, 2.0, 2.0), + Vertex::new(3, 3.0, 3.0), + Vertex::new(4, 4.0, 4.0), + ]; + + let edges = vec![ + Edge::new(0, 0, 1, 10.0), + Edge::new(1, 1, 0, 10.0), + Edge::new(2, 1, 2, 10.0), + Edge::new(3, 2, 1, 10.0), + Edge::new(4, 2, 3, 10.0), + Edge::new(5, 3, 2, 10.0), + Edge::new(6, 3, 0, 10.0), + Edge::new(7, 0, 3, 10.0), + Edge::new(8, 0, 2, 10.0), + Edge::new(9, 1, 3, 10.0), + Edge::new(10, 2, 0, 10.0), + Edge::new(11, 3, 1, 10.0), + Edge::new(12, 4, 4, 10.0), + ]; + + // Create the adjacency and reverse adjacency lists. + let mut adj = vec![HashMap::new(); vertices.len()]; + let mut rev = vec![HashMap::new(); vertices.len()]; + + for edge in &edges { + adj[edge.src_vertex_id.0].insert(edge.edge_id, edge.dst_vertex_id); + rev[edge.dst_vertex_id.0].insert(edge.edge_id, edge.src_vertex_id); + } + + // Construct the Graph instance. + let graph = Graph { + adj, + rev, + edges, + vertices, }; - let graph = Graph::try_from(&graph_conf).unwrap(); graph } diff --git a/rust/routee-compass-core/src/algorithm/search/a_star/a_star.rs b/rust/routee-compass-core/src/algorithm/search/a_star/a_star.rs index c57b2082..7835875c 100644 --- a/rust/routee-compass-core/src/algorithm/search/a_star/a_star.rs +++ b/rust/routee-compass-core/src/algorithm/search/a_star/a_star.rs @@ -330,13 +330,50 @@ mod tests { use crate::algorithm::search::backtrack::vertex_oriented_route; use crate::model::frontier::default::no_restriction; use crate::model::graph::graph::Graph; - use crate::model::graph::graph_config::GraphConfig; + use crate::model::property::edge::Edge; + use crate::model::property::vertex::Vertex; use crate::model::traversal::default::distance::DistanceModel; use crate::model::traversal::traversal_model::TraversalModel; use crate::util::unit::DistanceUnit; use crate::{model::graph::edge_id::EdgeId, util::read_only_lock::DriverReadOnlyLock}; use rayon::prelude::; + fn build_mock_graph() -> Graph { + let vertices = vec![ + Vertex::new(0, 0.0, 0.0), + Vertex::new(1, 0.0, 0.0), + Vertex::new(2, 0.0, 0.0), + Vertex::new(3, 0.0, 0.0), + ]; + + let edges = vec![ + Edge::new(0, 0, 1, 10.0), + Edge::new(1, 1, 0, 10.0), + Edge::new(2, 1, 2, 2.0), + Edge::new(3, 2, 1, 2.0), + Edge::new(4, 2, 3, 1.0), + Edge::new(5, 3, 2, 1.0), + Edge::new(6, 3, 0, 2.0), + Edge::new(7, 0, 3, 2.0), + ]; + + let mut adj = vec![HashMap::new(); vertices.len()]; + let mut rev = vec![HashMap::new(); vertices.len()]; + + for edge in &edges { + adj[edge.src_vertex_id.0].insert(edge.edge_id, edge.dst_vertex_id); + rev[edge.dst_vertex_id.0].insert(edge.edge_id, edge.src_vertex_id); + } + + let graph = Graph { + adj, + rev, + edges, + vertices, + }; + graph + } + #[test] fn test_e2e_queries() { // simple box world that exists in a non-euclidean plane that stretches @@ -386,22 +423,7 @@ mod tests { // setup the graph, traversal model, and a* heuristic to be shared across the queries in parallel // these live in the "driver" process and are passed as read-only memory to each executor process - let test_path = PathBuf::from(env!("CARGO_MANIFEST_DIR")) - .join("src") - .join("algorithm") - .join("search") - .join("a_star") - .join("test"); - let edge_list_csv = test_path.join("edges.csv").to_str().unwrap().to_string(); - let vertex_list_csv = test_path.join("vertices.csv").to_str().unwrap().to_string(); - let graph_conf = GraphConfig { - edge_list_csv, - vertex_list_csv, - n_edges: None, - n_vertices: None, - verbose: false, - }; - let graph = Graph::try_from(&graph_conf).unwrap(); + let graph = build_mock_graph(); let driver_dg = Arc::new(DriverReadOnlyLock::new(graph)); let no_restriction: Box<dyn FrontierModel> = Box::new(no_restriction::NoRestriction {}); diff --git a/rust/routee-compass-core/src/model/graph/edge_loader.rs b/rust/routee-compass-core/src/model/graph/edge_loader.rs index 7f8bbc15..164818e4 100644 --- a/rust/routee-compass-core/src/model/graph/edge_loader.rs +++ b/rust/routee-compass-core/src/model/graph/edge_loader.rs @@ -1,4 +1,3 @@ -use super::graph_config::GraphConfig; use crate::{ model::{ graph::graph_error::GraphError, @@ -9,7 +8,10 @@ use crate::{ }; use kdam::Bar; use kdam::BarExt; -use std::collections::{HashMap, HashSet}; +use std::{ + collections::{HashMap, HashSet}, + path::PathBuf, +}; pub struct EdgeLoader { pub edges: Vec<Edge>, @@ -17,13 +19,13 @@ pub struct EdgeLoader { pub rev: Vec<HashMap<EdgeId, VertexId>>, } -pub struct EdgeLoaderConfig<'a> { - pub config: &'a GraphConfig, +pub struct EdgeLoaderConfig { + pub edge_list_csv: PathBuf, pub n_edges: usize, pub n_vertices: usize, } -impl<'a> TryFrom<EdgeLoaderConfig<'a>> for EdgeLoader { +impl TryFrom<EdgeLoaderConfig> for EdgeLoader { type Error = GraphError; fn try_from(c: EdgeLoaderConfig) -> Result<Self, Self::Error> { @@ -62,8 +64,7 @@ impl<'a> TryFrom<EdgeLoaderConfig<'a>> for EdgeLoader { pb.update(1); }); - let edges = - read_utils::vec_from_csv(&c.config.edge_list_csv, true, Some(c.n_edges), Some(cb))?; + let edges = read_utils::vec_from_csv(&c.edge_list_csv, true, Some(c.n_edges), Some(cb))?; print!("\n"); let result = EdgeLoader { edges, adj, rev }; diff --git a/rust/routee-compass-core/src/model/graph/graph.rs b/rust/routee-compass-core/src/model/graph/graph.rs index a4162183..75508359 100644 --- a/rust/routee-compass-core/src/model/graph/graph.rs +++ b/rust/routee-compass-core/src/model/graph/graph.rs @@ -1,14 +1,12 @@ use crate::algorithm::search::direction::Direction; -use crate::model::graph::edge_loader::EdgeLoader; -use crate::model::graph::edge_loader::EdgeLoaderConfig; -use crate::model::graph::graph_config::GraphConfig; use crate::model::graph::graph_error::GraphError; -use crate::model::graph::vertex_loader::VertexLoaderConfig; use crate::model::graph::{edge_id::EdgeId, vertex_id::VertexId}; use crate::model::property::edge::Edge; use crate::model::property::vertex::Vertex; -use log::info; use std::collections::HashMap; +use std::path::Path; + +use super::graph_loader::graph_from_files; /// Road network topology represented as an adjacency list. /// The `EdgeId` and `VertexId` values correspond to edge and @@ -28,6 +26,8 @@ use std::collections::HashMap; /// collections will prefer chained iterators. A few will collect /// into Vecs because of error handling or lifetimes, but those cases will only produce a /// smaller subset of the source data. + +#[derive(Debug)] pub struct Graph { pub adj: Vec<HashMap<EdgeId, VertexId>>, pub rev: Vec<HashMap<EdgeId, VertexId>>, @@ -36,6 +36,32 @@ pub struct Graph { } impl Graph { + /// Build a `Graph` from a pair of CSV files. + /// You can also pass in the number of edges and vertices to avoid + /// scanning the input files and potentially building vectors that + /// have more memory than needed. + /// + /// # Arguments + /// + /// * `edge_list_csv` - path to the CSV file containing edge attributes + /// * `vertex_list_csv` - path to the CSV file containing vertex attributes + /// * `n_edges` - number of edges in the graph + /// * `n_vertices` - number of vertices in the graph + /// * `verbose` - whether to print progress information to the console + /// + /// # Returns + /// + /// A graph instance, or an error if an IO error occurred. + /// + pub fn from_files<P: AsRef<Path>>( + edge_list_csv: &P, + vertex_list_csv: &P, + n_edges: Option<usize>, + n_vertices: Option<usize>, + verbose: Option<bool>, + ) -> Result<Graph, GraphError> { + graph_from_files(edge_list_csv, vertex_list_csv, n_edges, n_vertices, verbose) + } /// number of edges in the Graph pub fn n_edges(&self) -> usize { self.edges.len() @@ -285,56 +311,3 @@ impl Graph { .collect() } } - -impl TryFrom<&GraphConfig> for Graph { - type Error = GraphError; - - /// tries to build a Graph from a GraphConfig. - /// - /// for both edge and vertex lists, we assume all ids can be used as indices - /// to an array data structure. to find the size of each array, we pass once - /// through each file to count the number of rows (minus header) of the CSV. - /// then we can build a Vec once and insert rows as we decode them without - /// a sort. - /// - /// # Arguments - /// - /// * `config` - configuration object for building a `Graph` instance - /// - /// # Returns - /// - /// A graph instance, or an error if an IO error occurred. - fn try_from(config: &GraphConfig) -> Result<Self, Self::Error> { - info!("checking file length of edge and vertex input files"); - let (n_edges, n_vertices) = config.read_file_sizes()?; - info!( - "creating data structures to hold {} edges, {} vertices", - n_edges, n_vertices - ); - - info!("reading edge list"); - - let e_conf = EdgeLoaderConfig { - config: &config, - n_edges, - n_vertices, - }; - let e_result = EdgeLoader::try_from(e_conf)?; - - info!("reading vertex list"); - let v_conf = VertexLoaderConfig { - config: &config, - n_vertices, - }; - let vertices: Vec<Vertex> = v_conf.try_into()?; - - let graph = Graph { - adj: e_result.adj, - rev: e_result.rev, - edges: e_result.edges, - vertices, - }; - - Ok(graph) - } -} diff --git a/rust/routee-compass-core/src/model/graph/graph_config.rs b/rust/routee-compass-core/src/model/graph/graph_config.rs deleted file mode 100644 index 559fd25c..00000000 --- a/rust/routee-compass-core/src/model/graph/graph_config.rs +++ /dev/null @@ -1,59 +0,0 @@ -use crate::model::graph::graph_error::GraphError; -use crate::util::fs::fs_utils::line_count; -use log::warn; -use serde::{Deserialize, Serialize}; - -#[derive(Debug, Deserialize, Serialize)] -pub struct GraphConfig { - pub edge_list_csv: String, - pub vertex_list_csv: String, - pub n_edges: Option<usize>, - pub n_vertices: Option<usize>, - pub verbose: bool, -} - -impl GraphConfig { - pub fn read_file_sizes(&self) -> Result<(usize, usize), GraphError> { - let n_edges = self - .get_n_edges() - .map_err(\|e\| GraphError::IOError { source: e })?; - let n_vertices = self - .get_n_vertices() - .map_err(\|e\| GraphError::IOError { source: e })?; - if n_edges < 1 { - return Err(GraphError::EmptyFileSource { - filename: self.edge_list_csv.clone(), - }); - } - if n_vertices < 1 { - return Err(GraphError::EmptyFileSource { - filename: self.vertex_list_csv.clone(), - }); - } - Ok((n_edges, n_vertices)) - } - - pub fn get_n_edges(&self) -> std::io::Result<usize> { - match self.n_edges { - Some(n) => Ok(n), - None => { - warn!("edge list size not provided, scanning input to determine size"); - let is_gzip = self.edge_list_csv.ends_with(".gz"); - let n = line_count(self.edge_list_csv.clone(), is_gzip)?; - Ok(n - 1) // drop count of header line - } - } - } - - pub fn get_n_vertices(&self) -> std::io::Result<usize> { - match self.n_vertices { - Some(n) => Ok(n), - None => { - warn!("vertex list size not provided, scanning input to determine size"); - let is_gzip = self.vertex_list_csv.ends_with(".gz"); - let n = line_count(self.vertex_list_csv.clone(), is_gzip)?; - Ok(n - 1) // drop count of header line - } - } - } -} diff --git a/rust/routee-compass-core/src/model/graph/graph_error.rs b/rust/routee-compass-core/src/model/graph/graph_error.rs index fa161884..70652ab4 100644 --- a/rust/routee-compass-core/src/model/graph/graph_error.rs +++ b/rust/routee-compass-core/src/model/graph/graph_error.rs @@ -1,3 +1,5 @@ +use std::path::PathBuf; + use crate::model::graph::{edge_id::EdgeId, vertex_id::VertexId}; #[derive(thiserror::Error, Debug)] @@ -17,7 +19,7 @@ pub enum GraphError { #[error("error in test setup")] TestError, #[error("{filename} file source was empty")] - EmptyFileSource { filename: String }, + EmptyFileSource { filename: PathBuf }, #[error("failure reading TomTom graph: {source}")] IOError { #[from] diff --git a/rust/routee-compass-core/src/model/graph/graph_loader.rs b/rust/routee-compass-core/src/model/graph/graph_loader.rs new file mode 100644 index 00000000..5bd0fa95 --- /dev/null +++ b/rust/routee-compass-core/src/model/graph/graph_loader.rs @@ -0,0 +1,97 @@ +use std::path::Path; + +use log::warn; + +use crate::{model::property::vertex::Vertex, util::fs::fs_utils::line_count}; + +use super::{ + edge_loader::{EdgeLoader, EdgeLoaderConfig}, + graph::Graph, + graph_error::GraphError, + vertex_loader::VertexLoaderConfig, +}; + +pub fn graph_from_files<P: AsRef<Path>>( + edge_list_csv: &P, + vertex_list_csv: &P, + n_edges: Option<usize>, + n_vertices: Option<usize>, + verbose: Option<bool>, +) -> Result<Graph, GraphError> { + let verbose = verbose.unwrap_or(false); + let n_edges = match n_edges { + Some(n) => n, + None => { + if verbose { + warn!("edge list size not provided, scanning input to determine size"); + } + get_n_edges(&edge_list_csv)? + } + }; + + let n_vertices = match n_vertices { + Some(n) => n, + None => { + if verbose { + warn!("vertex list size not provided, scanning input to determine size"); + } + get_n_vertices(&vertex_list_csv)? + } + }; + let e_conf = EdgeLoaderConfig { + edge_list_csv: edge_list_csv.as_ref().to_path_buf(), + n_edges, + n_vertices, + }; + + let e_result = EdgeLoader::try_from(e_conf)?; + + let v_conf = VertexLoaderConfig { + vertex_list_csv: vertex_list_csv.as_ref().to_path_buf(), + n_vertices, + }; + + let vertices: Vec<Vertex> = v_conf.try_into()?; + + let graph = Graph { + adj: e_result.adj, + rev: e_result.rev, + edges: e_result.edges, + vertices, + }; + + Ok(graph) +} + +fn get_n_edges<P: AsRef<Path>>(edge_list_csv: &P) -> Result<usize, GraphError> { + // check if the extension is .gz + let is_gzip = edge_list_csv + .as_ref() + .to_path_buf() + .extension() + .map(\|ext\| ext.to_str() == Some("gz")) + .unwrap_or(false); + let n = line_count(edge_list_csv.clone(), is_gzip)?; + if n < 1 { + return Err(GraphError::EmptyFileSource { + filename: edge_list_csv.as_ref().to_path_buf(), + }); + } + Ok(n - 1) // drop count of header line +} + +fn get_n_vertices<P: AsRef<Path>>(vertex_list_csv: &P) -> Result<usize, GraphError> { + let is_gzip = vertex_list_csv + .as_ref() + .to_path_buf() + .extension() + .map(\|ext\| ext.to_str() == Some("gz")) + .unwrap_or(false); + let n = line_count(vertex_list_csv.clone(), is_gzip)?; + if n < 1 { + return Err(GraphError::EmptyFileSource { + filename: vertex_list_csv.as_ref().to_path_buf(), + }); + } + Ok(n - 1) // drop count of header line +} diff --git a/rust/routee-compass-core/src/model/graph/mod.rs b/rust/routee-compass-core/src/model/graph/mod.rs index 50fd24da..2e41655a 100644 --- a/rust/routee-compass-core/src/model/graph/mod.rs +++ b/rust/routee-compass-core/src/model/graph/mod.rs @@ -1,7 +1,7 @@ pub mod edge_id; pub mod edge_loader; pub mod graph; -pub mod graph_config; pub mod graph_error; +pub mod graph_loader; pub mod vertex_id; pub mod vertex_loader; diff --git a/rust/routee-compass-core/src/model/graph/vertex_loader.rs b/rust/routee-compass-core/src/model/graph/vertex_loader.rs index a181743d..83734cee 100644 --- a/rust/routee-compass-core/src/model/graph/vertex_loader.rs +++ b/rust/routee-compass-core/src/model/graph/vertex_loader.rs @@ -1,18 +1,19 @@ -use crate::model::graph::graph_config::GraphConfig; +use std::path::PathBuf; + use crate::model::graph::graph_error::GraphError; use crate::model::property::vertex::Vertex; use crate::util::fs::read_utils; use kdam::{Bar, BarExt}; -pub struct VertexLoaderConfig<'a> { - pub config: &'a GraphConfig, +pub struct VertexLoaderConfig { + pub vertex_list_csv: PathBuf, pub n_vertices: usize, } -impl<'a> TryFrom<VertexLoaderConfig<'a>> for Vec<Vertex> { +impl TryFrom<VertexLoaderConfig> for Vec<Vertex> { type Error = GraphError; - fn try_from(conf: VertexLoaderConfig<'a>) -> Result<Self, Self::Error> { + fn try_from(conf: VertexLoaderConfig) -> Result<Self, Self::Error> { let mut processed: usize = 0; let mut pb = Bar::builder() .total(conf.n_vertices) @@ -25,12 +26,8 @@ impl<'a> TryFrom<VertexLoaderConfig<'a>> for Vec<Vertex> { pb.update(1); processed = processed + 1; }); - let result: Vec<Vertex> = read_utils::vec_from_csv( - &conf.config.vertex_list_csv, - true, - Some(conf.n_vertices), - Some(cb), - )?; + let result: Vec<Vertex> = + read_utils::vec_from_csv(&conf.vertex_list_csv, true, Some(conf.n_vertices), Some(cb))?; print!("\n"); Ok(result) diff --git a/rust/routee-compass-core/src/model/property/edge.rs b/rust/routee-compass-core/src/model/property/edge.rs index d02a78b1..cbc5a2a4 100644 --- a/rust/routee-compass-core/src/model/property/edge.rs +++ b/rust/routee-compass-core/src/model/property/edge.rs @@ -13,6 +13,17 @@ pub struct Edge { pub distance: Distance, } +impl Edge { + pub fn new(edge_id: usize, src_vertex_id: usize, dst_vertex_id: usize, distance: f64) -> Self { + Self { + edge_id: EdgeId(edge_id), + src_vertex_id: VertexId(src_vertex_id), + dst_vertex_id: VertexId(dst_vertex_id), + distance: Distance::new(distance), + } + } +} + impl Default for Edge { fn default() -> Self { Edge { diff --git a/rust/routee-compass-core/src/model/traversal/default/speed_lookup_model.rs b/rust/routee-compass-core/src/model/traversal/default/speed_lookup_model.rs index 03eb0bc7..6b547753 100644 --- a/rust/routee-compass-core/src/model/traversal/default/speed_lookup_model.rs +++ b/rust/routee-compass-core/src/model/traversal/default/speed_lookup_model.rs @@ -1,7 +1,9 @@ +use std::path::{Path, PathBuf}; + use crate::model::graph::edge_id::EdgeId; use crate::util::fs::read_decoders; use crate::util::geo::haversine; -use crate::util::unit::{Distance, DistanceUnit, Grade, BASE_SPEED_UNIT}; +use crate::util::unit::{Distance, DistanceUnit}; use crate::util::unit::{SpeedUnit, Time, TimeUnit, BASE_DISTANCE_UNIT, BASE_TIME_UNIT}; use crate::{ model::{ @@ -26,15 +28,20 @@ pub struct SpeedLookupModel { } impl SpeedLookupModel { - pub fn new( - speed_table_path: &String, + pub fn new<P: AsRef<Path>>( + speed_table_path: &P, speed_unit: SpeedUnit, output_distance_unit_opt: Option<DistanceUnit>, output_time_unit_opt: Option<TimeUnit>, ) -> Result<SpeedLookupModel, TraversalModelError> { let speed_table: Vec<Speed> = - read_utils::read_raw_file(speed_table_path, read_decoders::default, None).map_err( - \|e\| TraversalModelError::FileReadError(speed_table_path.clone(), e.to_string()), + read_utils::read_raw_file(&speed_table_path, read_decoders::default, None).map_err( + \|e\| { + TraversalModelError::FileReadError( + speed_table_path.as_ref().to_path_buf(), + e.to_string(), + ) + }, )?; let max_speed = get_max_speed(&speed_table)?; let output_time_unit = output_time_unit_opt.unwrap_or(BASE_TIME_UNIT); @@ -200,7 +207,7 @@ mod tests { distance: Distance::new(100.0), }; } - fn filepath() -> String { + fn filepath() -> PathBuf { let filepath = PathBuf::from(env!("CARGO_MANIFEST_DIR")) .join("src") .join("model") @@ -208,8 +215,7 @@ mod tests { .join("default") .join("test") .join("velocities.txt"); - let filename = filepath.to_str().unwrap().to_owned(); - return filename; + filepath } fn approx_eq(a: f64, b: f64, error: f64) { diff --git a/rust/routee-compass-core/src/model/traversal/traversal_model_error.rs b/rust/routee-compass-core/src/model/traversal/traversal_model_error.rs index 15ff4e9d..4f3d043e 100644 --- a/rust/routee-compass-core/src/model/traversal/traversal_model_error.rs +++ b/rust/routee-compass-core/src/model/traversal/traversal_model_error.rs @@ -1,10 +1,12 @@ +use std::path::PathBuf; + use super::state::traversal_state::TraversalState; use crate::util::unit::UnitError; #[derive(thiserror::Error, Debug)] pub enum TraversalModelError { #[error("failure building traversal model from file {0}: {1}")] - FileReadError(String, String), + FileReadError(PathBuf, String), #[error("failure building traversal model: {0}")] BuildError(String), #[error("numeric error during calculation: {0}")] diff --git a/rust/routee-compass-powertrain/README.md b/rust/routee-compass-powertrain/README.md index 6c6ccac2..5f87f34f 100644 --- a/rust/routee-compass-powertrain/README.md +++ b/rust/routee-compass-powertrain/README.md @@ -27,8 +27,8 @@ An example traversal model configuration that uses this crate may look like this [traversal] type = "speed_grade_energy_model" model_type = "smartcore" -speed_table_path = "edges-posted-speed-enumerated.txt.gz" -energy_model_path = "2016_TOYOTA_Camry_4cyl_2WD.bin" +speed_table_file = "edges-posted-speed-enumerated.txt.gz" +energy_model_file = "2016_TOYOTA_Camry_4cyl_2WD.bin" ideal_energy_rate = 0.02857142857 speed_table_speed_unit = "kilometers_per_hour" energy_model_speed_unit = "miles_per_hour" @@ -36,7 +36,7 @@ energy_model_grade_unit = "decimal" energy_model_energy_rate_unit = "gallons_gasoline_per_mile" output_time_unit = "minutes" output_distance_unit = "miles" -grade_table_path = "edges-grade-enumerated.txt.gz" +grade_table_file = "edges-grade-enumerated.txt.gz" grade_table_grade_unit = "decimal" ``` diff --git a/rust/routee-compass-powertrain/src/routee/model_type.rs b/rust/routee-compass-powertrain/src/routee/model_type.rs index 31da2782..c2d69236 100644 --- a/rust/routee-compass-powertrain/src/routee/model_type.rs +++ b/rust/routee-compass-powertrain/src/routee/model_type.rs @@ -1,4 +1,7 @@ -use std::sync::Arc; +use std::{ + path::{Path, PathBuf}, + sync::Arc, +}; use routee_compass_core::{ model::traversal::traversal_model_error::TraversalModelError, @@ -30,9 +33,9 @@ impl std::fmt::Display for ModelType { impl ModelType { /// builds a speed grade energy prediction model - pub fn build( + pub fn build<P: AsRef<Path>>( &self, - energy_model_path: String, + energy_model_path: &P, energy_model_speed_unit: SpeedUnit, energy_model_grade_unit: GradeUnit, energy_model_energy_rate_unit: EnergyRateUnit, diff --git a/rust/routee-compass-powertrain/src/routee/smartcore/smartcore_speed_grade_model.rs b/rust/routee-compass-powertrain/src/routee/smartcore/smartcore_speed_grade_model.rs index 42e6a8b1..58a0dd2d 100644 --- a/rust/routee-compass-powertrain/src/routee/smartcore/smartcore_speed_grade_model.rs +++ b/rust/routee-compass-powertrain/src/routee/smartcore/smartcore_speed_grade_model.rs @@ -1,3 +1,5 @@ +use std::path::Path; + use crate::routee::prediction_model::SpeedGradePredictionModel; use routee_compass_core::{ model::traversal::traversal_model_error::TraversalModelError, @@ -36,19 +38,25 @@ impl SpeedGradePredictionModel for SmartcoreSpeedGradeModel { } impl SmartcoreSpeedGradeModel { - pub fn new( - routee_model_path: String, + pub fn new<P: AsRef<Path>>( + routee_model_path: &P, speed_unit: SpeedUnit, grade_unit: GradeUnit, energy_rate_unit: EnergyRateUnit, ) -> Result<Self, TraversalModelError> { // Load random forest binary file let rf_binary = std::fs::read(routee_model_path.clone()).map_err(\|e\| { - TraversalModelError::FileReadError(routee_model_path.clone(), e.to_string()) + TraversalModelError::FileReadError( + routee_model_path.as_ref().to_path_buf(), + e.to_string(), + ) })?; let rf: RandomForestRegressor<f64, f64, DenseMatrix<f64>, Vec<f64>> = bincode::deserialize(&rf_binary).map_err(\|e\| { - TraversalModelError::FileReadError(routee_model_path.clone(), e.to_string()) + TraversalModelError::FileReadError( + routee_model_path.as_ref().to_path_buf(), + e.to_string(), + ) })?; Ok(SmartcoreSpeedGradeModel { rf, diff --git a/rust/routee-compass-powertrain/src/routee/speed_grade_model.rs b/rust/routee-compass-powertrain/src/routee/speed_grade_model.rs index 2870e8ef..8930c5cb 100644 --- a/rust/routee-compass-powertrain/src/routee/speed_grade_model.rs +++ b/rust/routee-compass-powertrain/src/routee/speed_grade_model.rs @@ -255,9 +255,6 @@ mod tests { .join("routee") .join("test") .join("Toyota_Camry.bin"); - let speed_file_name = speed_file_path.to_str().unwrap(); - let grade_file_name = grade_file_path.to_str().unwrap(); - let model_file_name = model_file_path.to_str().unwrap(); let v = Vertex { vertex_id: VertexId(0), coordinate: coord! {x: -86.67, y: 36.12}, @@ -270,15 +267,12 @@ mod tests { distance: Distance::new(100.0), }; } - let speed_file = String::from(speed_file_name); - let grade_file = String::from(grade_file_name); - let routee_model_path = String::from(model_file_name); let service = SpeedGradeModelService::new( - speed_file, + &speed_file_path, SpeedUnit::KilometersPerHour, + &Some(grade_file_path), None, - None, - routee_model_path, + &model_file_path, ModelType::Smartcore, None, SpeedUnit::MilesPerHour, diff --git a/rust/routee-compass-powertrain/src/routee/speed_grade_model_service.rs b/rust/routee-compass-powertrain/src/routee/speed_grade_model_service.rs index bfa890ef..266dc51d 100644 --- a/rust/routee-compass-powertrain/src/routee/speed_grade_model_service.rs +++ b/rust/routee-compass-powertrain/src/routee/speed_grade_model_service.rs @@ -7,6 +7,8 @@ use routee_compass_core::model::traversal::traversal_model_error::TraversalModel use routee_compass_core::util::fs::read_decoders; use routee_compass_core::util::fs::read_utils; use routee_compass_core::util::unit::; +use std::path::Path; +use std::path::PathBuf; use std::sync::Arc; #[cfg(feature = "onnx")] @@ -30,12 +32,12 @@ pub struct SpeedGradeModelService { } impl SpeedGradeModelService { - pub fn new( - speed_table_path: String, + pub fn new<P: AsRef<Path>>( + speed_table_path: &P, speeds_table_speed_unit: SpeedUnit, - grade_table_path_option: Option<String>, + grade_table_path_option: &Option<P>, grade_table_grade_unit_option: Option<GradeUnit>, - energy_model_path: String, + energy_model_path: &P, model_type: ModelType, ideal_energy_rate_option: Option<EnergyRate>, energy_model_speed_unit: SpeedUnit, @@ -63,14 +65,22 @@ impl SpeedGradeModelService { // load speeds table let speed_table: Arc<Vec<Speed>> = Arc::new( read_utils::read_raw_file(&speed_table_path, read_decoders::default, None).map_err( - \|e\| TraversalModelError::FileReadError(speed_table_path.clone(), e.to_string()), + \|e\| { + TraversalModelError::FileReadError( + speed_table_path.as_ref().to_path_buf(), + e.to_string(), + ) + }, )?, ); let grade_table: Arc<Option<Vec<Grade>>> = match grade_table_path_option { Some(gtp) => Arc::new(Some( read_utils::read_raw_file(&gtp, read_decoders::default, None).map_err(\|e\| { - TraversalModelError::FileReadError(speed_table_path.clone(), e.to_string()) + TraversalModelError::FileReadError( + speed_table_path.as_ref().to_path_buf(), + e.to_string(), + ) })?, )), None => Arc::new(None), diff --git a/rust/routee-compass-py/src/app_wrapper.rs b/rust/routee-compass-py/src/app_wrapper.rs index 5d4a1663..c365c44b 100644 --- a/rust/routee-compass-py/src/app_wrapper.rs +++ b/rust/routee-compass-py/src/app_wrapper.rs @@ -1,4 +1,4 @@ -use std::path::PathBuf; +use std::path::Path; use pyo3::{exceptions::PyException, prelude::, types::PyType}; use routee_compass::app::compass::compass_app::CompassApp; @@ -12,7 +12,7 @@ pub struct CompassAppWrapper { impl CompassAppWrapper { #[classmethod] pub fn _from_config_file(_cls: &PyType, config_file: String) -> PyResult<Self> { - let config_path = PathBuf::from(config_file.clone()); + let config_path = Path::new(&config_file); let routee_compass = CompassApp::try_from(config_path).map_err(\|e\| { PyException::new_err(format!( "Could not create CompassApp from config file {}: {}", diff --git a/rust/routee-compass/src/app/compass/compass_app.rs b/rust/routee-compass/src/app/compass/compass_app.rs index 84391770..b1941729 100644 --- a/rust/routee-compass/src/app/compass/compass_app.rs +++ b/rust/routee-compass/src/app/compass/compass_app.rs @@ -2,10 +2,13 @@ use super::config::compass_app_builder::CompassAppBuilder; use crate::{ app::{ compass::{ - compass_configuration_field::CompassConfigurationField, - config::termination_model_builder::TerminationModelBuilder, + compass_app_error::CompassAppError, + config::{ + compass_configuration_field::CompassConfigurationField, + config_json_extension::ConfigJsonExtensions, graph_builder::DefaultGraphBuilder, + termination_model_builder::TerminationModelBuilder, + }, }, - compass_app_error::CompassAppError, search::{search_app::SearchApp, search_app_result::SearchAppResult}, }, plugin::{ @@ -18,14 +21,12 @@ use config::Config; use itertools::{Either, Itertools}; use rayon::{current_num_threads, prelude::}; use routee_compass_core::{ - algorithm::search::search_algorithm::SearchAlgorithm, - model::{ - cost::cost::Cost, - graph::{graph::Graph, graph_config::GraphConfig}, - }, + algorithm::search::search_algorithm::SearchAlgorithm, model::cost::cost::Cost, util::duration_extension::DurationExtension, }; -use std::path::PathBuf; +use std::path::{Path, PathBuf}; + +pub const CONFIG_FILE_KEY: &str = "config_file"; /// Instance of RouteE Compass as an application. /// When constructed, it holds @@ -42,7 +43,7 @@ pub struct CompassApp { pub parallelism: usize, } -impl TryFrom<PathBuf> for CompassApp { +impl TryFrom<&Path> for CompassApp { type Error = CompassAppError; /// Builds a CompassApp from a configuration filepath, using the default CompassAppBuilder. @@ -57,18 +58,30 @@ impl TryFrom<PathBuf> for CompassApp { /// # Returns /// /// an instance of [`CompassApp`], or an error if load failed. - fn try_from(conf_file: PathBuf) -> Result<Self, Self::Error> { + fn try_from(conf_file: &Path) -> Result<Self, Self::Error> { let default_config = config::File::from_str( include_str!("config.default.toml"), config::FileFormat::Toml, ); + // We want to store the location of where the config file + // was found so we can use it later to resolve relative paths + let conf_file_string = conf_file + .to_str() + .ok_or(CompassAppError::InternalError( + "Could not parse incoming config file path".to_string(), + ))? + .to_string(); + let config = Config::builder() .add_source(default_config) .add_source(config::File::from(conf_file)) + .set_override(CONFIG_FILE_KEY, conf_file_string)? .build() .map_err(CompassAppError::ConfigError)?; - log::info!("Config: {:?}", config); + + log::debug!("Loaded config: {:?}", config); + let builder = CompassAppBuilder::default(); let compass_app = CompassApp::try_from((&config, &builder))?; return Ok(compass_app); @@ -98,14 +111,22 @@ impl TryFrom<(&Config, &CompassAppBuilder)> for CompassApp { fn try_from(pair: (&Config, &CompassAppBuilder)) -> Result<Self, Self::Error> { let (config, builder) = pair; - let alg_params = - config.get::<serde_json::Value>(CompassConfigurationField::Algorithm.to_str())?; + // Get the root config path so we can resolve paths relative + // to where the config file is located. + let root_config_path = config.get::<PathBuf>(CONFIG_FILE_KEY)?; + + let config_json = config + .clone() + .try_deserialize::<serde_json::Value>()? + .normalize_file_paths(&root_config_path)?; + + let alg_params = config_json.get_config_section(CompassConfigurationField::Algorithm)?; let search_algorithm = SearchAlgorithm::try_from(&alg_params)?; // build traversal model let traversal_start = Local::now(); let traversal_params = - config.get::<serde_json::Value>(CompassConfigurationField::Traversal.to_str())?; + config_json.get_config_section(CompassConfigurationField::Traversal)?; let traversal_model_service = builder.build_traversal_model_service(&traversal_params)?; let traversal_duration = (Local::now() - traversal_start) .to_std() @@ -118,7 +139,7 @@ impl TryFrom<(&Config, &CompassAppBuilder)> for CompassApp { // build frontier model let frontier_start = Local::now(); let frontier_params = - config.get::<serde_json::Value>(CompassConfigurationField::Frontier.to_str())?; + config_json.get_config_section(CompassConfigurationField::Frontier)?; let frontier_model = builder.build_frontier_model(frontier_params)?; let frontier_duration = (Local::now() - frontier_start) .to_std() @@ -130,15 +151,13 @@ impl TryFrom<(&Config, &CompassAppBuilder)> for CompassApp { // build termination model let termination_model_json = - config.get::<serde_json::Value>(CompassConfigurationField::Termination.to_str())?; + config_json.get_config_section(CompassConfigurationField::Termination)?; let termination_model = TerminationModelBuilder::build(&termination_model_json)?; // build graph let graph_start = Local::now(); - let graph_conf = &config - .get::<GraphConfig>(CompassConfigurationField::Graph.to_str()) - .map_err(CompassAppError::ConfigError)?; - let graph = Graph::try_from(graph_conf)?; + let graph_params = config_json.get_config_section(CompassConfigurationField::Graph)?; + let graph = DefaultGraphBuilder::build(&graph_params)?; let graph_duration = (Local::now() - graph_start) .to_std() .map_err(\|e\| CompassAppError::InternalError(e.to_string()))?; @@ -165,11 +184,10 @@ impl TryFrom<(&Config, &CompassAppBuilder)> for CompassApp { // build plugins let plugins_start = Local::now(); - let plugins_config = - config.get::<serde_json::Value>(CompassConfigurationField::Plugins.to_str())?; + let plugins_config = config_json.get_config_section(CompassConfigurationField::Plugins)?; - let input_plugins = builder.build_input_plugins(plugins_config.clone())?; - let output_plugins = builder.build_output_plugins(plugins_config.clone())?; + let input_plugins = builder.build_input_plugins(&plugins_config)?; + let output_plugins = builder.build_output_plugins(&plugins_config)?; let plugins_duration = to_std(Local::now() - plugins_start)?; log::info!( @@ -423,8 +441,8 @@ mod tests { .join("speeds_test") .join("speeds_debug.toml"); - let app = CompassApp::try_from(conf_file_test) - .or(CompassApp::try_from(conf_file_debug)) + let app = CompassApp::try_from(conf_file_test.as_path()) + .or(CompassApp::try_from(conf_file_debug.as_path())) .unwrap(); let query = serde_json::json!({ "origin_vertex": 0, diff --git a/rust/routee-compass/src/app/compass_app_error.rs b/rust/routee-compass/src/app/compass/compass_app_error.rs similarity index 98% rename from rust/routee-compass/src/app/compass_app_error.rs rename to rust/routee-compass/src/app/compass/compass_app_error.rs index ed0ae0ba..ba73aa2d 100644 --- a/rust/routee-compass/src/app/compass_app_error.rs +++ b/rust/routee-compass/src/app/compass/compass_app_error.rs @@ -1,4 +1,4 @@ -use super::compass::{ +use super::{ compass_input_field::CompassInputField, config::compass_configuration_error::CompassConfigurationError, }; diff --git a/rust/routee-compass/src/app/compass/compass_json_extensions.rs b/rust/routee-compass/src/app/compass/compass_json_extensions.rs index 1b93fa99..3c0ad2a1 100644 --- a/rust/routee-compass/src/app/compass/compass_json_extensions.rs +++ b/rust/routee-compass/src/app/compass/compass_json_extensions.rs @@ -1,5 +1,5 @@ use super::compass_input_field::CompassInputField; -use crate::app::compass_app_error::CompassAppError; +use crate::app::compass::compass_app_error::CompassAppError; pub trait CompassJsonExtensions { fn get_queries(&self) -> Result<Vec<serde_json::Value>, CompassAppError>; diff --git a/rust/routee-compass/src/app/compass/config.default.toml b/rust/routee-compass/src/app/compass/config.default.toml index 538edbf0..d1f4ab37 100644 --- a/rust/routee-compass/src/app/compass/config.default.toml +++ b/rust/routee-compass/src/app/compass/config.default.toml @@ -1,8 +1,6 @@ parallelism = 2 [graph] -edge_list_csv = "edges.csv.gz" -vertex_list_csv = "vertices.csv.gz" verbose = true [algorithm] @@ -20,5 +18,5 @@ limit = "00:02:00" frequency = 100_000 [plugin] -input_plugins = [{ type = "vertex_rtree", vertices_file = "vertices.csv.gz" }] +input_plugins = [] output_plugins = [] diff --git a/rust/routee-compass/src/app/compass/config/builders.rs b/rust/routee-compass/src/app/compass/config/builders.rs index 4c7c06d3..c5d24896 100644 --- a/rust/routee-compass/src/app/compass/config/builders.rs +++ b/rust/routee-compass/src/app/compass/config/builders.rs @@ -1,7 +1,8 @@ use super::compass_configuration_error::CompassConfigurationError; use crate::plugin::{input::input_plugin::InputPlugin, output::output_plugin::OutputPlugin}; use routee_compass_core::model::{ - frontier::frontier_model::FrontierModel, traversal::traversal_model::TraversalModel, + frontier::frontier_model::FrontierModel, graph::graph::Graph, + traversal::traversal_model::TraversalModel, }; use std::sync::Arc; @@ -11,7 +12,7 @@ use std::sync::Arc; /// A [`TraversalModelBuilder`] instance should be an empty struct that implements /// this trait. pub trait TraversalModelBuilder { - /// Builds a [`TraversalModelService`] from JSON configuration. + /// Builds a [`TraversalModelService`] from configuration. /// /// # Arguments /// diff --git a/rust/routee-compass/src/app/compass/config/compass_app_builder.rs b/rust/routee-compass/src/app/compass/config/compass_app_builder.rs index 76fc91eb..ecc7f08b 100644 --- a/rust/routee-compass/src/app/compass/config/compass_app_builder.rs +++ b/rust/routee-compass/src/app/compass/config/compass_app_builder.rs @@ -4,6 +4,7 @@ use super::{ TraversalModelService, }, compass_configuration_error::CompassConfigurationError, + compass_configuration_field::CompassConfigurationField, frontier_model::{ no_restriction_builder::NoRestrictionBuilder, road_class_builder::RoadClassBuilder, }, @@ -13,23 +14,18 @@ use super::{ speed_lookup_builder::SpeedLookupBuilder, }, }; -use crate::{ - app::compass::compass_configuration_field::CompassConfigurationField, - plugin::{ - input::{ - default::{ - grid_search::builder::GridSearchBuilder, rtree::builder::VertexRTreeBuilder, - }, - input_plugin::InputPlugin, - }, - output::{ - default::{ - edgeidlist::builder::EdgeIdListOutputPluginBuilder, - summary::builder::SummaryOutputPluginBuilder, - traversal::builder::TraversalPluginBuilder, uuid::builder::UUIDOutputPluginBuilder, - }, - output_plugin::OutputPlugin, +use crate::plugin::{ + input::{ + default::{grid_search::builder::GridSearchBuilder, rtree::builder::VertexRTreeBuilder}, + input_plugin::InputPlugin, + }, + output::{ + default::{ + edgeidlist::builder::EdgeIdListOutputPluginBuilder, + summary::builder::SummaryOutputPluginBuilder, + traversal::builder::TraversalPluginBuilder, uuid::builder::UUIDOutputPluginBuilder, }, + output_plugin::OutputPlugin, }, }; use routee_compass_core::model::frontier::frontier_model::FrontierModel; @@ -198,14 +194,14 @@ impl CompassAppBuilder { pub fn build_input_plugins( &self, - config: serde_json::Value, + config: &serde_json::Value, ) -> Result<Vec<Box<dyn InputPlugin>>, CompassConfigurationError> { - let input_plugins_obj = config.get("input_plugins").ok_or( - CompassConfigurationError::ExpectedFieldForComponent( + let input_plugins_obj = config + .get(CompassConfigurationField::InputPlugins.to_str()) + .ok_or(CompassConfigurationError::ExpectedFieldForComponent( CompassConfigurationField::InputPlugins.to_string(), - String::from("input_plugins"), - ), - )?; + CompassConfigurationField::Plugins.to_string(), + ))?; let input_plugins = input_plugins_obj.as_array().ok_or( CompassConfigurationError::ExpectedFieldWithType( String::from("input_plugins"), @@ -247,14 +243,14 @@ impl CompassAppBuilder { pub fn build_output_plugins( &self, - config: serde_json::Value, + config: &serde_json::Value, ) -> Result<Vec<Box<dyn OutputPlugin>>, CompassConfigurationError> { - let output_plugins_obj = config.get("output_plugins").ok_or( - CompassConfigurationError::ExpectedFieldForComponent( + let output_plugins_obj = config + .get(CompassConfigurationField::OutputPlugins.to_str()) + .ok_or(CompassConfigurationError::ExpectedFieldForComponent( CompassConfigurationField::OutputPlugins.to_string(), - String::from("output_plugins"), - ), - )?; + CompassConfigurationField::Plugins.to_string(), + ))?; let output_plugins = output_plugins_obj.as_array().ok_or( CompassConfigurationError::ExpectedFieldWithType( String::from("output_plugins"), diff --git a/rust/routee-compass/src/app/compass/config/compass_configuration_error.rs b/rust/routee-compass/src/app/compass/config/compass_configuration_error.rs index aaa627c3..0f46547d 100644 --- a/rust/routee-compass/src/app/compass/config/compass_configuration_error.rs +++ b/rust/routee-compass/src/app/compass/config/compass_configuration_error.rs @@ -1,5 +1,8 @@ +use config::ConfigError; use routee_compass_core::{ - model::traversal::traversal_model_error::TraversalModelError, + model::{ + graph::graph_error::GraphError, traversal::traversal_model_error::TraversalModelError, + }, util::conversion::conversion_error::ConversionError, }; @@ -15,6 +18,31 @@ pub enum CompassConfigurationError { ExpectedFieldWithTypeUnrecognized(String, String, String), #[error("unknown module {0} for component {1} provided by configuration")] UnknownModelNameForComponent(String, String), + #[error( + r#" + File '{0}' was not found. + This file came from field '{1}' for component '{2}'. + + First, make sure this file path is either relative to your config file, + or, is provided as an absolute path. + + Second, make sure the file exists. + + Third, make sure the config key ends with '_file' which is a schema requirement + for the CompassApp config. + "# + )] + FileNotFoundForComponent(String, String, String), + #[error("could not normalize incoming file {0}")] + FileNormalizationError(String), + #[error("Could not find incoming configuration file, tried {0} and {1}. Make sure the file exists and that the config key ends with '_file'")] + FileNormalizationNotFound(String, String), + #[error("{0}")] + InsertError(String), + #[error(transparent)] + GraphError(#[from] GraphError), + #[error(transparent)] + ConfigError(#[from] ConfigError), #[error(transparent)] IoError(#[from] std::io::Error), #[error(transparent)] diff --git a/rust/routee-compass/src/app/compass/compass_configuration_field.rs b/rust/routee-compass/src/app/compass/config/compass_configuration_field.rs similarity index 100% rename from rust/routee-compass/src/app/compass/compass_configuration_field.rs rename to rust/routee-compass/src/app/compass/config/compass_configuration_field.rs diff --git a/rust/routee-compass/src/app/compass/config/config_json_extension.rs b/rust/routee-compass/src/app/compass/config/config_json_extension.rs index 56b09948..9668f671 100644 --- a/rust/routee-compass/src/app/compass/config/config_json_extension.rs +++ b/rust/routee-compass/src/app/compass/config/config_json_extension.rs @@ -1,8 +1,26 @@ use super::compass_configuration_error::CompassConfigurationError; +use super::compass_configuration_field::CompassConfigurationField; use serde::de; -use std::str::FromStr; +use std::{ + path::{Path, PathBuf}, + str::FromStr, +}; pub trait ConfigJsonExtensions { + fn get_config_section( + &self, + section: CompassConfigurationField, + ) -> Result<serde_json::Value, CompassConfigurationError>; + fn get_config_path( + &self, + key: String, + parent_key: String, + ) -> Result<PathBuf, CompassConfigurationError>; + fn get_config_path_optional( + &self, + key: String, + parent_key: String, + ) -> Result<Option<PathBuf>, CompassConfigurationError>; fn get_config_string( &self, key: String, @@ -33,9 +51,60 @@ pub trait ConfigJsonExtensions { key: String, parent_key: String, ) -> Result<Option<T>, CompassConfigurationError>; + fn normalize_file_paths( + &self, + root_config_path: &Path, + ) -> Result<serde_json::Value, CompassConfigurationError>; } impl ConfigJsonExtensions for serde_json::Value { + fn get_config_section( + &self, + section: CompassConfigurationField, + ) -> Result<serde_json::Value, CompassConfigurationError> { + let section = self + .get(section.to_str()) + .ok_or(CompassConfigurationError::ExpectedFieldForComponent( + section.to_string(), + String::from(""), + ))? + .clone(); + + Ok(section) + } + fn get_config_path_optional( + &self, + key: String, + parent_key: String, + ) -> Result<Option<PathBuf>, CompassConfigurationError> { + match self.get(key.clone()) { + None => Ok(None), + Some(_) => { + let config_path = self.get_config_path(key, parent_key)?; + Ok(Some(config_path)) + } + } + } + fn get_config_path( + &self, + key: String, + parent_key: String, + ) -> Result<PathBuf, CompassConfigurationError> { + let path_string = self.get_config_string(key.clone(), parent_key.clone())?; + let path = PathBuf::from(&path_string); + + // if file can be found, just return it + if path.is_file() { + return Ok(path); + } else { + // can't find the file + return Err(CompassConfigurationError::FileNotFoundForComponent( + path_string, + key, + parent_key, + )); + } + } fn get_config_string( &self, key: String, @@ -154,4 +223,79 @@ impl ConfigJsonExtensions for serde_json::Value { } } } + + /// This function is used to normalize file paths in the configuration JSON object. + /// Incoming file paths can be in one of three locations: + /// + /// 1. Absolute path + /// 2. Relative path to the config file itself + /// 3. Relative path to where the application is being run + /// + /// This function scans each key value pair in the config and for any key that + /// ends with `_file`, it will attempt to normalize the path such that the application + /// can find the file regardless of where it is being executed. + /// + /// Arguments: + /// + /// * `root_config_path` - The path to the where the config file is located. + /// + /// Returns: + /// + /// * `Result<serde_json::Value, CompassConfigurationError>` - The JSON object with normalized paths. + fn normalize_file_paths( + &self, + root_config_path: &Path, + ) -> Result<serde_json::Value, CompassConfigurationError> { + match self { + serde_json::Value::String(path_string) => { + let path = Path::new(path_string); + + // no need to modify if the file exists + if path.is_file() { + return Ok(serde_json::Value::String(path_string.clone())); + } + + // next we try adding the root config path and see if that exists + let root_config_parent = match root_config_path.parent() { + Some(parent) => parent, + None => Path::new(""), + }; + let new_path = root_config_parent.join(&path); + let new_path_string = new_path + .to_str() + .ok_or(CompassConfigurationError::FileNormalizationError( + path_string.clone(), + ))? + .to_string(); + if new_path.is_file() { + Ok(serde_json::Value::String(new_path_string)) + } else { + // if we can't find the file in either location, we throw an error + Err(CompassConfigurationError::FileNormalizationNotFound( + path_string.clone(), + new_path_string, + )) + } + } + serde_json::Value::Object(obj) => { + let mut new_obj = serde_json::map::Map::new(); + for (key, value) in obj.iter() { + if key.ends_with("_file") \|\| value.is_object() \|\| value.is_array() { + new_obj.insert(key.clone(), value.normalize_file_paths(root_config_path)?); + } else { + new_obj.insert(key.clone(), value.clone()); + } + } + Ok(serde_json::Value::Object(new_obj)) + } + serde_json::Value::Array(arr) => { + let mut new_arr = Vec::new(); + for value in arr.iter() { + new_arr.push(value.normalize_file_paths(root_config_path)?); + } + Ok(serde_json::Value::Array(new_arr)) + } + _ => Ok(self.clone()), + } + } } diff --git a/rust/routee-compass/src/app/compass/config/frontier_model/road_class_builder.rs b/rust/routee-compass/src/app/compass/config/frontier_model/road_class_builder.rs index 6a018e0d..4cfe56b3 100644 --- a/rust/routee-compass/src/app/compass/config/frontier_model/road_class_builder.rs +++ b/rust/routee-compass/src/app/compass/config/frontier_model/road_class_builder.rs @@ -1,9 +1,7 @@ -use crate::app::compass::{ +use crate::app::compass::config::{ + builders::FrontierModelBuilder, compass_configuration_error::CompassConfigurationError, compass_configuration_field::CompassConfigurationField, - config::{ - builders::FrontierModelBuilder, compass_configuration_error::CompassConfigurationError, - config_json_extension::ConfigJsonExtensions, - }, + config_json_extension::ConfigJsonExtensions, }; use routee_compass_core::{ model::frontier::{default::road_class::RoadClassFrontierModel, frontier_model::FrontierModel}, @@ -23,7 +21,7 @@ impl FrontierModelBuilder for RoadClassBuilder { let valid_road_class_key = String::from("valid_road_classes"); let road_class_file = - parameters.get_config_string(road_class_file_key.clone(), frontier_key.clone())?; + parameters.get_config_path(road_class_file_key.clone(), frontier_key.clone())?; let road_classes_json = parameters .get(valid_road_class_key.clone()) diff --git a/rust/routee-compass/src/app/compass/config/graph_builder.rs b/rust/routee-compass/src/app/compass/config/graph_builder.rs new file mode 100644 index 00000000..1ddc01a3 --- /dev/null +++ b/rust/routee-compass/src/app/compass/config/graph_builder.rs @@ -0,0 +1,51 @@ +use routee_compass_core::model::graph::graph::Graph; + +use crate::app::compass::config::compass_configuration_field::CompassConfigurationField; + +use super::{ + compass_configuration_error::CompassConfigurationError, + config_json_extension::ConfigJsonExtensions, +}; + +pub struct DefaultGraphBuilder {} + +impl DefaultGraphBuilder { + /// tries to build a Graph from a JSON object. + /// + /// for both edge and vertex lists, we assume all ids can be used as indices + /// to an array data structure. to find the size of each array, we pass once + /// through each file to count the number of rows (minus header) of the CSV. + /// then we can build a Vec once and insert rows as we decode them without + /// a sort. + /// + /// # Arguments + /// + /// * `params` - configuration JSON object for building a `Graph` instance + /// + /// # Returns + /// + /// A graph instance, or an error if an IO error occurred. + pub fn build(params: &serde_json::Value) -> Result<Graph, CompassConfigurationError> { + let graph_key = CompassConfigurationField::Graph.to_string(); + let edge_list_csv = + params.get_config_path(String::from("edge_list_file"), graph_key.clone())?; + let vertex_list_csv = + params.get_config_path(String::from("vertex_list_file"), graph_key.clone())?; + let n_edges = + params.get_config_serde_optional(String::from("n_edges"), graph_key.clone())?; + let n_vertices = + params.get_config_serde_optional(String::from("n_vertices"), graph_key.clone())?; + let verbose: Option<bool> = + params.get_config_serde_optional(String::from("verbose"), graph_key.clone())?; + + let graph = Graph::from_files( + &edge_list_csv, + &vertex_list_csv, + n_edges, + n_vertices, + verbose, + )?; + + Ok(graph) + } +} diff --git a/rust/routee-compass/src/app/compass/config/mod.rs b/rust/routee-compass/src/app/compass/config/mod.rs index 6142bc6b..c3b300ef 100644 --- a/rust/routee-compass/src/app/compass/config/mod.rs +++ b/rust/routee-compass/src/app/compass/config/mod.rs @@ -1,7 +1,9 @@ pub mod builders; pub mod compass_app_builder; pub mod compass_configuration_error; +pub mod compass_configuration_field; pub mod config_json_extension; pub mod frontier_model; +pub mod graph_builder; pub mod termination_model_builder; pub mod traversal_model; diff --git a/rust/routee-compass/src/app/compass/config/termination_model_builder.rs b/rust/routee-compass/src/app/compass/config/termination_model_builder.rs index 9f392a8d..73415295 100644 --- a/rust/routee-compass/src/app/compass/config/termination_model_builder.rs +++ b/rust/routee-compass/src/app/compass/config/termination_model_builder.rs @@ -1,5 +1,5 @@ use super::compass_configuration_error::CompassConfigurationError; -use crate::app::compass::compass_configuration_field::CompassConfigurationField; +use crate::app::compass::config::compass_configuration_field::CompassConfigurationField; use log; use routee_compass_core::model::termination::termination_model::TerminationModel; use routee_compass_core::util::conversion::duration_extension::DurationExtension; diff --git a/rust/routee-compass/src/app/compass/config/traversal_model/distance_builder.rs b/rust/routee-compass/src/app/compass/config/traversal_model/distance_builder.rs index e6b31906..0ca1f9a8 100644 --- a/rust/routee-compass/src/app/compass/config/traversal_model/distance_builder.rs +++ b/rust/routee-compass/src/app/compass/config/traversal_model/distance_builder.rs @@ -1,10 +1,8 @@ use crate::app::compass::config::config_json_extension::ConfigJsonExtensions; -use crate::app::compass::{ +use crate::app::compass::config::{ + builders::{TraversalModelBuilder, TraversalModelService}, + compass_configuration_error::CompassConfigurationError, compass_configuration_field::CompassConfigurationField, - config::{ - builders::{TraversalModelBuilder, TraversalModelService}, - compass_configuration_error::CompassConfigurationError, - }, }; use routee_compass_core::model::traversal::traversal_model::TraversalModel; use routee_compass_core::util::unit::BASE_DISTANCE_UNIT; diff --git a/rust/routee-compass/src/app/compass/config/traversal_model/speed_grade_energy_model_builder.rs b/rust/routee-compass/src/app/compass/config/traversal_model/speed_grade_energy_model_builder.rs index 23393eb1..f7f17127 100644 --- a/rust/routee-compass/src/app/compass/config/traversal_model/speed_grade_energy_model_builder.rs +++ b/rust/routee-compass/src/app/compass/config/traversal_model/speed_grade_energy_model_builder.rs @@ -1,7 +1,7 @@ use std::sync::Arc; -use crate::app::compass::compass_configuration_field::CompassConfigurationField; use crate::app::compass::config::builders::TraversalModelService; +use crate::app::compass::config::compass_configuration_field::CompassConfigurationField; use crate::app::compass::config::config_json_extension::ConfigJsonExtensions; use crate::app::compass::config::{ builders::TraversalModelBuilder, compass_configuration_error::CompassConfigurationError, @@ -28,23 +28,21 @@ impl TraversalModelBuilder for SpeedGradeEnergyModelBuilder { let traversal_key = CompassConfigurationField::Traversal.to_string(); let speed_table_path = - params.get_config_string(String::from("speed_table_path"), traversal_key.clone())?; + params.get_config_path(String::from("speed_table_file"), traversal_key.clone())?; let speed_table_speed_unit = params.get_config_serde::<SpeedUnit>( String::from("speed_table_speed_unit"), traversal_key.clone(), )?; - let grade_table_path = params.get_config_serde_optional::<String>( - String::from("grade_table_path"), - traversal_key.clone(), - )?; + let grade_table_path = params + .get_config_path_optional(String::from("grade_table_file"), traversal_key.clone())?; let grade_table_grade_unit = params.get_config_serde_optional::<GradeUnit>( String::from("graph_grade_unit"), traversal_key.clone(), )?; let energy_model_path = - params.get_config_string(String::from("energy_model_path"), traversal_key.clone())?; + params.get_config_path(String::from("energy_model_file"), traversal_key.clone())?; let model_type = params .get_config_serde::<ModelType>(String::from("model_type"), traversal_key.clone())?; let energy_model_speed_unit = params.get_config_serde::<SpeedUnit>( @@ -80,11 +78,11 @@ impl TraversalModelBuilder for SpeedGradeEnergyModelBuilder { )?; let inner_service = SpeedGradeModelService::new( - speed_table_path, + &speed_table_path, speed_table_speed_unit, - grade_table_path, + &grade_table_path, grade_table_grade_unit, - energy_model_path, + &energy_model_path, model_type, ideal_energy_rate_option, energy_model_speed_unit, diff --git a/rust/routee-compass/src/app/compass/config/traversal_model/speed_lookup_builder.rs b/rust/routee-compass/src/app/compass/config/traversal_model/speed_lookup_builder.rs index 18fa13ce..eaf30d18 100644 --- a/rust/routee-compass/src/app/compass/config/traversal_model/speed_lookup_builder.rs +++ b/rust/routee-compass/src/app/compass/config/traversal_model/speed_lookup_builder.rs @@ -1,5 +1,5 @@ -use crate::app::compass::compass_configuration_field::CompassConfigurationField; use crate::app::compass::config::builders::TraversalModelService; +use crate::app::compass::config::compass_configuration_field::CompassConfigurationField; use crate::app::compass::config::config_json_extension::ConfigJsonExtensions; use crate::app::compass::config::{ builders::TraversalModelBuilder, compass_configuration_error::CompassConfigurationError, @@ -23,7 +23,7 @@ impl TraversalModelBuilder for SpeedLookupBuilder { let traversal_key = CompassConfigurationField::Traversal.to_string(); // todo: optional output time unit let filename = - params.get_config_string(String::from("speed_table_path"), traversal_key.clone())?; + params.get_config_path(String::from("speed_table_file"), traversal_key.clone())?; let speed_unit = params .get_config_serde::<SpeedUnit>(String::from("speed_unit"), traversal_key.clone())?; let distance_unit = params.get_config_serde_optional::<DistanceUnit>( diff --git a/rust/routee-compass/src/app/compass/mod.rs b/rust/routee-compass/src/app/compass/mod.rs index e3b2835f..cdb5ee53 100644 --- a/rust/routee-compass/src/app/compass/mod.rs +++ b/rust/routee-compass/src/app/compass/mod.rs @@ -1,6 +1,6 @@ pub mod compass_app; pub mod compass_app_args; -pub mod compass_configuration_field; +pub mod compass_app_error; pub mod compass_input_field; pub mod compass_json_extensions; pub mod config; diff --git a/rust/routee-compass/src/app/geom/geom_app.rs b/rust/routee-compass/src/app/geom/geom_app.rs index 731d9eae..37b3cd83 100644 --- a/rust/routee-compass/src/app/geom/geom_app.rs +++ b/rust/routee-compass/src/app/geom/geom_app.rs @@ -1,4 +1,4 @@ -use crate::app::compass_app_error::CompassAppError; +use crate::app::compass::compass_app_error::CompassAppError; use crate::plugin::output::default::traversal::utils::parse_linestring; use geo::LineString; use kdam::Bar; diff --git a/rust/routee-compass/src/app/mod.rs b/rust/routee-compass/src/app/mod.rs index b7d62691..ae855b0d 100644 --- a/rust/routee-compass/src/app/mod.rs +++ b/rust/routee-compass/src/app/mod.rs @@ -1,4 +1,3 @@ pub mod compass; -pub mod compass_app_error; pub mod geom; pub mod search; diff --git a/rust/routee-compass/src/app/search/search_app.rs b/rust/routee-compass/src/app/search/search_app.rs index d37019f2..d2d5780a 100644 --- a/rust/routee-compass/src/app/search/search_app.rs +++ b/rust/routee-compass/src/app/search/search_app.rs @@ -1,6 +1,9 @@ use super::search_app_result::SearchAppResult; use crate::{ - app::{compass::config::builders::TraversalModelService, compass_app_error::CompassAppError}, + app::{ + compass::compass_app_error::CompassAppError, + compass::config::builders::TraversalModelService, + }, plugin::input::input_json_extensions::InputJsonExtensions, }; use chrono::Local; diff --git a/rust/routee-compass/src/main.rs b/rust/routee-compass/src/main.rs index 9f9b5c47..899a1d6f 100644 --- a/rust/routee-compass/src/main.rs +++ b/rust/routee-compass/src/main.rs @@ -1,9 +1,9 @@ use clap::Parser; -use log::info; +use log::{error, info}; use routee_compass::app::compass::compass_app::CompassApp; use routee_compass::app::compass::compass_app_args::CompassAppArgs; +use routee_compass::app::compass::compass_app_error::CompassAppError; use routee_compass::app::compass::compass_json_extensions::CompassJsonExtensions; -use routee_compass::app::compass_app_error::CompassAppError; use std::error::Error; use std::fs::File; use std::io::BufReader; @@ -18,7 +18,13 @@ fn main() -> Result<(), Box<dyn Error>> { "No configuration file specified".to_string(), ))?; - let compass_app = CompassApp::try_from(conf_file)?; + let compass_app = match CompassApp::try_from(conf_file.as_path()) { + Ok(app) => app, + Err(e) => { + error!("Could not build CompassApp from config file: {}", e); + return Err(Box::new(e)); + } + }; // read user file containing JSON query/queries let query_file = File::open(args.query_file.clone()).map_err(\|_e\| { diff --git a/rust/routee-compass/src/plugin/input/default/rtree/builder.rs b/rust/routee-compass/src/plugin/input/default/rtree/builder.rs index 6c110850..491a5c43 100644 --- a/rust/routee-compass/src/plugin/input/default/rtree/builder.rs +++ b/rust/routee-compass/src/plugin/input/default/rtree/builder.rs @@ -1,8 +1,7 @@ -use std::path::PathBuf; - use crate::{ app::compass::config::{ builders::InputPluginBuilder, compass_configuration_error::CompassConfigurationError, + config_json_extension::ConfigJsonExtensions, }, plugin::input::input_plugin::InputPlugin, }; @@ -17,19 +16,10 @@ impl InputPluginBuilder for VertexRTreeBuilder { parameters: &serde_json::Value, ) -> Result<Box<dyn InputPlugin>, CompassConfigurationError> { let vertex_filename_key = String::from("vertices_file"); - let vertex_filename = parameters - .get(&vertex_filename_key) - .ok_or(CompassConfigurationError::ExpectedFieldForComponent( - vertex_filename_key.clone(), - String::from("Vertex RTree Input Plugin"), - ))? - .as_str() - .map(String::from) - .ok_or(CompassConfigurationError::ExpectedFieldWithType( - vertex_filename_key.clone(), - String::from("String"), - ))?; - let vertex_path = PathBuf::from(vertex_filename); + let vertex_path = parameters.get_config_path( + vertex_filename_key, + String::from("Vertex RTree Input Plugin"), + )?; let rtree = RTreePlugin::from_file(&vertex_path).map_err(CompassConfigurationError::PluginError)?; let m: Box<dyn InputPlugin> = Box::new(rtree); diff --git a/rust/routee-compass/src/plugin/input/default/rtree/plugin.rs b/rust/routee-compass/src/plugin/input/default/rtree/plugin.rs index 499d2c8c..c37bc082 100644 --- a/rust/routee-compass/src/plugin/input/default/rtree/plugin.rs +++ b/rust/routee-compass/src/plugin/input/default/rtree/plugin.rs @@ -1,4 +1,4 @@ -use std::path::Path; +use std::path::{Path, PathBuf}; use crate::plugin::input::input_json_extensions::InputJsonExtensions; use crate::plugin::input::input_plugin::InputPlugin; diff --git a/rust/routee-compass/src/plugin/output/default/traversal/builder.rs b/rust/routee-compass/src/plugin/output/default/traversal/builder.rs index d2dd2596..be82bb4d 100644 --- a/rust/routee-compass/src/plugin/output/default/traversal/builder.rs +++ b/rust/routee-compass/src/plugin/output/default/traversal/builder.rs @@ -45,7 +45,7 @@ impl OutputPluginBuilder for TraversalPluginBuilder { let tree_geometry_key = String::from("tree"); let geometry_filename = - parameters.get_config_string(geometry_filename_key, parent_key.clone())?; + parameters.get_config_path(geometry_filename_key, parent_key.clone())?; let route: Option<TraversalOutputFormat> = parameters.get_config_serde_optional(route_geometry_key, parent_key.clone())?; let tree: Option<TraversalOutputFormat> = diff --git a/rust/routee-compass/src/plugin/output/default/traversal/plugin.rs b/rust/routee-compass/src/plugin/output/default/traversal/plugin.rs index af6d4d31..4a784919 100644 --- a/rust/routee-compass/src/plugin/output/default/traversal/plugin.rs +++ b/rust/routee-compass/src/plugin/output/default/traversal/plugin.rs @@ -1,3 +1,6 @@ +use std::path::Path; +use std::path::PathBuf; + use super::json_extensions::TraversalJsonField; use super::traversal_output_format::TraversalOutputFormat; use super::utils::parse_linestring; @@ -18,15 +21,15 @@ pub struct TraversalPlugin { } impl TraversalPlugin { - pub fn from_file( - filename: &String, + pub fn from_file<P: AsRef<Path>>( + filename: &P, route: Option<TraversalOutputFormat>, tree: Option<TraversalOutputFormat>, ) -> Result<TraversalPlugin, PluginError> { let count = fs_utils::line_count(filename.clone(), fs_utils::is_gzip(&filename)).map_err(\|e\| { PluginError::FileReadError { - filename: filename.clone(), + filename: filename.as_ref().to_path_buf(), message: e.to_string(), } })?; @@ -43,7 +46,7 @@ impl TraversalPlugin { }); let geoms = read_raw_file(&filename, parse_linestring, Some(cb)).map_err(\|e\| { PluginError::FileReadError { - filename: filename.clone(), + filename: filename.as_ref().to_path_buf(), message: e.to_string(), } })?; @@ -171,7 +174,7 @@ mod tests { route_runtime: Duration::ZERO, total_runtime: Duration::ZERO, }; - let filename = mock_geometry_file().to_str().unwrap().to_string(); + let filename = mock_geometry_file(); let route_geometry = true; let tree_geometry = false; let geom_plugin = diff --git a/rust/routee-compass/src/plugin/output/default/uuid/builder.rs b/rust/routee-compass/src/plugin/output/default/uuid/builder.rs index fc6ed2a1..8281be3c 100644 --- a/rust/routee-compass/src/plugin/output/default/uuid/builder.rs +++ b/rust/routee-compass/src/plugin/output/default/uuid/builder.rs @@ -1,6 +1,7 @@ use crate::{ app::compass::config::{ builders::OutputPluginBuilder, compass_configuration_error::CompassConfigurationError, + config_json_extension::ConfigJsonExtensions, }, plugin::output::output_plugin::OutputPlugin, }; @@ -15,18 +16,7 @@ impl OutputPluginBuilder for UUIDOutputPluginBuilder { parameters: &serde_json::Value, ) -> Result<Box<dyn OutputPlugin>, CompassConfigurationError> { let uuid_filename_key = String::from("uuid_file"); - let uuid_filename = parameters - .get(&uuid_filename_key) - .ok_or(CompassConfigurationError::ExpectedFieldForComponent( - uuid_filename_key.clone(), - String::from("UUID Output Plugin"), - ))? - .as_str() - .map(String::from) - .ok_or(CompassConfigurationError::ExpectedFieldWithType( - uuid_filename_key.clone(), - String::from("String"), - ))?; + let uuid_filename = parameters.get_config_path(uuid_filename_key, String::from("uuid"))?; let uuid_plugin = UUIDOutputPlugin::from_file(&uuid_filename) .map_err(CompassConfigurationError::PluginError)?; diff --git a/rust/routee-compass/src/plugin/output/default/uuid/plugin.rs b/rust/routee-compass/src/plugin/output/default/uuid/plugin.rs index 40a7341d..9954e999 100644 --- a/rust/routee-compass/src/plugin/output/default/uuid/plugin.rs +++ b/rust/routee-compass/src/plugin/output/default/uuid/plugin.rs @@ -1,3 +1,5 @@ +use std::path::{Path, PathBuf}; + use super::json_extensions::UUIDJsonExtensions; use crate::app::search::search_app_result::SearchAppResult; @@ -12,11 +14,11 @@ pub struct UUIDOutputPlugin { } impl UUIDOutputPlugin { - pub fn from_file(filename: &String) -> Result<UUIDOutputPlugin, PluginError> { + pub fn from_file<P: AsRef<Path>>(filename: &P) -> Result<UUIDOutputPlugin, PluginError> { let count = fs_utils::line_count(filename.clone(), fs_utils::is_gzip(&filename)).map_err(\|e\| { PluginError::FileReadError { - filename: filename.clone(), + filename: filename.as_ref().to_path_buf(), message: e.to_string(), } })?; @@ -34,7 +36,7 @@ impl UUIDOutputPlugin { let uuids = read_raw_file(&filename, \|_idx, row\| Ok(row), Some(cb)).map_err(\|e\| { PluginError::FileReadError { - filename: filename.clone(), + filename: filename.as_ref().to_path_buf(), message: e.to_string(), } })?; diff --git a/rust/routee-compass/src/plugin/plugin_error.rs b/rust/routee-compass/src/plugin/plugin_error.rs index 62200727..62a89fd7 100644 --- a/rust/routee-compass/src/plugin/plugin_error.rs +++ b/rust/routee-compass/src/plugin/plugin_error.rs @@ -1,3 +1,5 @@ +use std::path::PathBuf; + use geo::Coord; use routee_compass_core::{ algorithm::search::search_error::SearchError, model::graph::edge_id::EdgeId, @@ -16,7 +18,7 @@ pub enum PluginError { #[error("nearest vertex not found for coord {0:?}")] NearestVertexNotFound(Coord), #[error("unable to read file {filename} due to {message}")] - FileReadError { filename: String, message: String }, + FileReadError { filename: PathBuf, message: String }, #[error(transparent)] JsonError(#[from] serde_json::Error), #[error(transparent)]	Standardize paths types Right now we have a mix of different ways to represent our paths in the application, for example: - `PathBuf` - `&Path` - `P: AsRef<Path>` We should spend some time to review and attempt to standardize these where it makes sense to do so.		2023-10-27T20:56:54	[]	[]
aerleon/aerleon	aerleon__aerleon-335	fb514ba28a0d535fe914d4dd602a180b69b041bb	diff --git a/aerleon/lib/speedway.py b/aerleon/lib/speedway.py index 4a0c1835..632b7d0b 100644 --- a/aerleon/lib/speedway.py +++ b/aerleon/lib/speedway.py @@ -43,6 +43,6 @@ class Speedway(iptables.Iptables): _RENDER_PREFIX = '*filter' _RENDER_SUFFIX = 'COMMIT' - _DEFAULTACTION_FORMAT = ':%s %s' + _DEFAULTACTION_FORMAT = ':%s %s [0:0]' _TERM = Term	Speedway does not include packet/byte count in its output Normally the iptables-save command will provide details on how many packets and bytes have gone through the specific filter. The syntax is like ``` :OUTPUT DROP [0:0] -N O_deny-all -A O_deny-all -p tcp -j DROP -A O_deny-all -p udp -j DROP COMMIT ``` where the counters are specifically `[0:0]` This is not included in our output. This may cause some unknown incompatibility we have not heard of or it might be fine. In either case any parser that is working off outputs from `iptables-save` will expect to see this in our output.		2023-08-07T03:07:22	[]	[]
aerleon/aerleon	aerleon__aerleon-334	7ba9f2b08f3fea764defe38e3c2b6fb71cfc0e55	diff --git a/aerleon/lib/juniper.py b/aerleon/lib/juniper.py index 13f00d23..1132946d 100644 --- a/aerleon/lib/juniper.py +++ b/aerleon/lib/juniper.py @@ -16,7 +16,6 @@ """Juniper JCL generator.""" - from typing import Dict, List, Set, Tuple, Union from absl import logging @@ -486,17 +485,25 @@ def __str__(self): # protocol if self.term.protocol: + # Juniper has deprecated the use of icmpv6 in favor of icmp6. + # https://github.com/aerleon/aerleon/issues/323 + protocol = self.term.protocol[:] + if 'icmpv6' in self.term.protocol: + loc = protocol.index('icmpv6') + protocol[loc] = 'icmp6' # both are supported on JunOS, but only icmp6 is supported # on SRX loopback stateless filter - config.Append(family_keywords['protocol'] + ' ' + self._Group(self.term.protocol)) + config.Append(family_keywords['protocol'] + ' ' + self._Group(protocol)) # protocol if self.term.protocol_except: # same as above + protocol_except = self.term.protocol_except[:] + if 'icmpv6' in self.term.protocol_except: + loc = protocol_except.index('icmpv6') + protocol_except[loc] = 'icmp6' config.Append( - family_keywords['protocol-except'] - + ' ' - + self._Group(self.term.protocol_except) + family_keywords['protocol-except'] + ' ' + self._Group(protocol_except) ) # port	Minor regression on juniper generator After upgrading from Capirca to Aerleon, the generated `juniper` policies use `icmpv6` instead of `icmp6` ```diff [edit firewall family inet6 filter loopback6 term allow_ok_icmp6 from] - next-header icmp6; + next-header icmpv6; ``` Both work but according to Juniper's [doc](https://www.juniper.net/documentation/us/en/software/junos/routing-policy/topics/concept/firewall-filter-match-conditions-for-ipv6-traffic.html): > Note: next-header icmp6 and next-header icmpv6 match conditions perform the same function. next-header icmp6 is the preferred option. next-header icmpv6 is hidden in the Junos OS CLI. So there is a risk that `icmpv6` becomes decommissioned at some point and breaks.	Does your policy look like ``` - name: foo protocol-except: icmp6 action: accept ``` or ``` - name: bar protocol-except: icmpv6 action: accept ``` The code as it exists now should be outputting verbatim what ever the protocol-except is. I believe this was the same in Capirca too. Thanks, I tested it with https://gerrit.wikimedia.org/r/c/operations/homer/public/+/945550 which is the only location in our policies where "icmpv6" is present, but unfortunately this fails with: > aerleon.lib.aclgenerator.UnsupportedFilterError: Protocol(s) ['icmp6'] are not supported.	2023-08-05T05:55:02	[]	[]
aerleon/aerleon	aerleon__aerleon-235	71d1b70b6301283c66adada532410abef21f508a	diff --git a/policies/includes/arista_terms_include.yaml b/policies/includes/arista_terms_include.yaml index af9358b7..5eede3ca 100644 --- a/policies/includes/arista_terms_include.yaml +++ b/policies/includes/arista_terms_include.yaml @@ -10,6 +10,14 @@ terms: counter: wonky-prots-loopback action: accept + - name: accept-traceroute-exc + comment: "allow inbound traceroute from any source." + destination-port: TRACEROUTE + protocol: udp + counter: inbound-traceroute + action: accept + platform-exclude: arista_tp + - name: accept-traceroute comment: "allow inbound traceroute from any source." destination-port: TRACEROUTE diff --git a/policies/pol/sample_arista_tp.pol b/policies/pol/sample_arista_tp.pol index 04cabf27..b39a5c55 100644 --- a/policies/pol/sample_arista_tp.pol +++ b/policies/pol/sample_arista_tp.pol @@ -29,7 +29,6 @@ term accept-traceroute-exc { protocol:: udp counter:: inbound-traceroute action:: accept - expiration:: 2001-12-31 platform-exclude:: arista_tp }	Move term platform check to ACLGenerator Move term platform check to ACLGenerator	# [Codecov](https://codecov.io/gh/aerleon/aerleon/pull/218?src=pr&el=h1&utm_medium=referral&utm_source=github&utm_content=comment&utm_campaign=pr+comments&utm_term=aerleon) Report Base: 90.61% // Head: 91.14% // Increases project coverage by `+0.52%` :tada: > Coverage data is based on head [(`85ae7b2`)](https://codecov.io/gh/aerleon/aerleon/pull/218?src=pr&el=desc&utm_medium=referral&utm_source=github&utm_content=comment&utm_campaign=pr+comments&utm_term=aerleon) compared to base [(`6b3be9d`)](https://codecov.io/gh/aerleon/aerleon/commit/6b3be9dc015f0def70a953ecc396367532c23b2d?el=desc&utm_medium=referral&utm_source=github&utm_content=comment&utm_campaign=pr+comments&utm_term=aerleon). > Patch coverage: 98.43% of modified lines in pull request are covered. <details><summary>Additional details and impacted files</summary> ```diff @@ Coverage Diff @@ ## main #218 +/- ## ========================================== + Coverage 90.61% 91.14% +0.52% ========================================== Files 95 95 Lines 21960 21742 -218 Branches 3382 3265 -117 ========================================== - Hits 19900 19816 -84 + Misses 1340 1256 -84 + Partials 720 670 -50 ``` \| Flag \| Coverage Δ \| \| \|---\|---\|---\| \| tests \| `91.12% <98.43%> (+0.52%)` \| :arrow_up: \| Flags with carried forward coverage won't be shown. [Click here](https://docs.codecov.io/docs/carryforward-flags?utm_medium=referral&utm_source=github&utm_content=comment&utm_campaign=pr+comments&utm_term=aerleon#carryforward-flags-in-the-pull-request-comment) to find out more. \| [Impacted Files](https://codecov.io/gh/aerleon/aerleon/pull/218?src=pr&el=tree&utm_medium=referral&utm_source=github&utm_content=comment&utm_campaign=pr+comments&utm_term=aerleon) \| Coverage Δ \| \| \|---\|---\|---\| \| [aerleon/lib/arista\_tp.py](https://codecov.io/gh/aerleon/aerleon/pull/218?src=pr&el=tree&utm_medium=referral&utm_source=github&utm_content=comment&utm_campaign=pr+comments&utm_term=aerleon#diff-YWVybGVvbi9saWIvYXJpc3RhX3RwLnB5) \| `93.66% <ø> (+1.06%)` \| :arrow_up: \| \| [aerleon/lib/cisco.py](https://codecov.io/gh/aerleon/aerleon/pull/218?src=pr&el=tree&utm_medium=referral&utm_source=github&utm_content=comment&utm_campaign=pr+comments&utm_term=aerleon#diff-YWVybGVvbi9saWIvY2lzY28ucHk=) \| `89.07% <ø> (+3.13%)` \| :arrow_up: \| \| [aerleon/lib/gcp\_hf.py](https://codecov.io/gh/aerleon/aerleon/pull/218?src=pr&el=tree&utm_medium=referral&utm_source=github&utm_content=comment&utm_campaign=pr+comments&utm_term=aerleon#diff-YWVybGVvbi9saWIvZ2NwX2hmLnB5) \| `95.60% <ø> (+1.06%)` \| :arrow_up: \| \| [aerleon/lib/iptables.py](https://codecov.io/gh/aerleon/aerleon/pull/218?src=pr&el=tree&utm_medium=referral&utm_source=github&utm_content=comment&utm_campaign=pr+comments&utm_term=aerleon#diff-YWVybGVvbi9saWIvaXB0YWJsZXMucHk=) \| `90.52% <ø> (+0.86%)` \| :arrow_up: \| \| [aerleon/lib/juniper.py](https://codecov.io/gh/aerleon/aerleon/pull/218?src=pr&el=tree&utm_medium=referral&utm_source=github&utm_content=comment&utm_campaign=pr+comments&utm_term=aerleon#diff-YWVybGVvbi9saWIvanVuaXBlci5weQ==) \| `86.80% <ø> (+0.97%)` \| :arrow_up: \| \| [aerleon/lib/junipersrx.py](https://codecov.io/gh/aerleon/aerleon/pull/218?src=pr&el=tree&utm_medium=referral&utm_source=github&utm_content=comment&utm_campaign=pr+comments&utm_term=aerleon#diff-YWVybGVvbi9saWIvanVuaXBlcnNyeC5weQ==) \| `90.10% <ø> (+0.68%)` \| :arrow_up: \| \| [aerleon/lib/nsxv.py](https://codecov.io/gh/aerleon/aerleon/pull/218?src=pr&el=tree&utm_medium=referral&utm_source=github&utm_content=comment&utm_campaign=pr+comments&utm_term=aerleon#diff-YWVybGVvbi9saWIvbnN4di5weQ==) \| `88.35% <ø> (+3.33%)` \| :arrow_up: \| \| [aerleon/lib/packetfilter.py](https://codecov.io/gh/aerleon/aerleon/pull/218?src=pr&el=tree&utm_medium=referral&utm_source=github&utm_content=comment&utm_campaign=pr+comments&utm_term=aerleon#diff-YWVybGVvbi9saWIvcGFja2V0ZmlsdGVyLnB5) \| `83.09% <ø> (+1.49%)` \| :arrow_up: \| \| [aerleon/lib/pcap.py](https://codecov.io/gh/aerleon/aerleon/pull/218?src=pr&el=tree&utm_medium=referral&utm_source=github&utm_content=comment&utm_campaign=pr+comments&utm_term=aerleon#diff-YWVybGVvbi9saWIvcGNhcC5weQ==) \| `71.93% <ø> (+1.32%)` \| :arrow_up: \| \| [aerleon/lib/windows.py](https://codecov.io/gh/aerleon/aerleon/pull/218?src=pr&el=tree&utm_medium=referral&utm_source=github&utm_content=comment&utm_campaign=pr+comments&utm_term=aerleon#diff-YWVybGVvbi9saWIvd2luZG93cy5weQ==) \| `74.64% <ø> (+2.03%)` \| :arrow_up: \| \| ... and [16 more](https://codecov.io/gh/aerleon/aerleon/pull/218?src=pr&el=tree-more&utm_medium=referral&utm_source=github&utm_content=comment&utm_campaign=pr+comments&utm_term=aerleon) \| \| Help us with your feedback. Take ten seconds to tell us [how you rate us](https://about.codecov.io/nps?utm_medium=referral&utm_source=github&utm_content=comment&utm_campaign=pr+comments&utm_term=aerleon). Have a feature suggestion? [Share it here.](https://app.codecov.io/gh/feedback/?utm_medium=referral&utm_source=github&utm_content=comment&utm_campaign=pr+comments&utm_term=aerleon) </details> [:umbrella: View full report at Codecov](https://codecov.io/gh/aerleon/aerleon/pull/218?src=pr&el=continue&utm_medium=referral&utm_source=github&utm_content=comment&utm_campaign=pr+comments&utm_term=aerleon). :loudspeaker: Do you have feedback about the report comment? [Let us know in this issue](https://about.codecov.io/codecov-pr-comment-feedback/?utm_medium=referral&utm_source=github&utm_content=comment&utm_campaign=pr+comments&utm_term=aerleon). # [Codecov](https://codecov.io/gh/aerleon/aerleon/pull/218?src=pr&el=h1&utm_medium=referral&utm_source=github&utm_content=comment&utm_campaign=pr+comments&utm_term=aerleon) Report Base: 90.61% // Head: 91.14% // Increases project coverage by `+0.52%` :tada: > Coverage data is based on head [(`85ae7b2`)](https://codecov.io/gh/aerleon/aerleon/pull/218?src=pr&el=desc&utm_medium=referral&utm_source=github&utm_content=comment&utm_campaign=pr+comments&utm_term=aerleon) compared to base [(`6b3be9d`)](https://codecov.io/gh/aerleon/aerleon/commit/6b3be9dc015f0def70a953ecc396367532c23b2d?el=desc&utm_medium=referral&utm_source=github&utm_content=comment&utm_campaign=pr+comments&utm_term=aerleon). > Patch coverage: 98.43% of modified lines in pull request are covered. <details><summary>Additional details and impacted files</summary> ```diff @@ Coverage Diff @@ ## main #218 +/- ## ========================================== + Coverage 90.61% 91.14% +0.52% ========================================== Files 95 95 Lines 21960 21742 -218 Branches 3382 3265 -117 ========================================== - Hits 19900 19816 -84 + Misses 1340 1256 -84 + Partials 720 670 -50 ``` \| Flag \| Coverage Δ \| \| \|---\|---\|---\| \| tests \| `91.12% <98.43%> (+0.52%)` \| :arrow_up: \| Flags with carried forward coverage won't be shown. [Click here](https://docs.codecov.io/docs/carryforward-flags?utm_medium=referral&utm_source=github&utm_content=comment&utm_campaign=pr+comments&utm_term=aerleon#carryforward-flags-in-the-pull-request-comment) to find out more. \| [Impacted Files](https://codecov.io/gh/aerleon/aerleon/pull/218?src=pr&el=tree&utm_medium=referral&utm_source=github&utm_content=comment&utm_campaign=pr+comments&utm_term=aerleon) \| Coverage Δ \| \| \|---\|---\|---\| \| [aerleon/lib/arista\_tp.py](https://codecov.io/gh/aerleon/aerleon/pull/218?src=pr&el=tree&utm_medium=referral&utm_source=github&utm_content=comment&utm_campaign=pr+comments&utm_term=aerleon#diff-YWVybGVvbi9saWIvYXJpc3RhX3RwLnB5) \| `93.66% <ø> (+1.06%)` \| :arrow_up: \| \| [aerleon/lib/cisco.py](https://codecov.io/gh/aerleon/aerleon/pull/218?src=pr&el=tree&utm_medium=referral&utm_source=github&utm_content=comment&utm_campaign=pr+comments&utm_term=aerleon#diff-YWVybGVvbi9saWIvY2lzY28ucHk=) \| `89.07% <ø> (+3.13%)` \| :arrow_up: \| \| [aerleon/lib/gcp\_hf.py](https://codecov.io/gh/aerleon/aerleon/pull/218?src=pr&el=tree&utm_medium=referral&utm_source=github&utm_content=comment&utm_campaign=pr+comments&utm_term=aerleon#diff-YWVybGVvbi9saWIvZ2NwX2hmLnB5) \| `95.60% <ø> (+1.06%)` \| :arrow_up: \| \| [aerleon/lib/iptables.py](https://codecov.io/gh/aerleon/aerleon/pull/218?src=pr&el=tree&utm_medium=referral&utm_source=github&utm_content=comment&utm_campaign=pr+comments&utm_term=aerleon#diff-YWVybGVvbi9saWIvaXB0YWJsZXMucHk=) \| `90.52% <ø> (+0.86%)` \| :arrow_up: \| \| [aerleon/lib/juniper.py](https://codecov.io/gh/aerleon/aerleon/pull/218?src=pr&el=tree&utm_medium=referral&utm_source=github&utm_content=comment&utm_campaign=pr+comments&utm_term=aerleon#diff-YWVybGVvbi9saWIvanVuaXBlci5weQ==) \| `86.80% <ø> (+0.97%)` \| :arrow_up: \| \| [aerleon/lib/junipersrx.py](https://codecov.io/gh/aerleon/aerleon/pull/218?src=pr&el=tree&utm_medium=referral&utm_source=github&utm_content=comment&utm_campaign=pr+comments&utm_term=aerleon#diff-YWVybGVvbi9saWIvanVuaXBlcnNyeC5weQ==) \| `90.10% <ø> (+0.68%)` \| :arrow_up: \| \| [aerleon/lib/nsxv.py](https://codecov.io/gh/aerleon/aerleon/pull/218?src=pr&el=tree&utm_medium=referral&utm_source=github&utm_content=comment&utm_campaign=pr+comments&utm_term=aerleon#diff-YWVybGVvbi9saWIvbnN4di5weQ==) \| `88.35% <ø> (+3.33%)` \| :arrow_up: \| \| [aerleon/lib/packetfilter.py](https://codecov.io/gh/aerleon/aerleon/pull/218?src=pr&el=tree&utm_medium=referral&utm_source=github&utm_content=comment&utm_campaign=pr+comments&utm_term=aerleon#diff-YWVybGVvbi9saWIvcGFja2V0ZmlsdGVyLnB5) \| `83.09% <ø> (+1.49%)` \| :arrow_up: \| \| [aerleon/lib/pcap.py](https://codecov.io/gh/aerleon/aerleon/pull/218?src=pr&el=tree&utm_medium=referral&utm_source=github&utm_content=comment&utm_campaign=pr+comments&utm_term=aerleon#diff-YWVybGVvbi9saWIvcGNhcC5weQ==) \| `71.93% <ø> (+1.32%)` \| :arrow_up: \| \| [aerleon/lib/windows.py](https://codecov.io/gh/aerleon/aerleon/pull/218?src=pr&el=tree&utm_medium=referral&utm_source=github&utm_content=comment&utm_campaign=pr+comments&utm_term=aerleon#diff-YWVybGVvbi9saWIvd2luZG93cy5weQ==) \| `74.64% <ø> (+2.03%)` \| :arrow_up: \| \| ... and [16 more](https://codecov.io/gh/aerleon/aerleon/pull/218?src=pr&el=tree-more&utm_medium=referral&utm_source=github&utm_content=comment&utm_campaign=pr+comments&utm_term=aerleon) \| \| Help us with your feedback. Take ten seconds to tell us [how you rate us](https://about.codecov.io/nps?utm_medium=referral&utm_source=github&utm_content=comment&utm_campaign=pr+comments&utm_term=aerleon). Have a feature suggestion? [Share it here.](https://app.codecov.io/gh/feedback/?utm_medium=referral&utm_source=github&utm_content=comment&utm_campaign=pr+comments&utm_term=aerleon) </details> [:umbrella: View full report at Codecov](https://codecov.io/gh/aerleon/aerleon/pull/218?src=pr&el=continue&utm_medium=referral&utm_source=github&utm_content=comment&utm_campaign=pr+comments&utm_term=aerleon). :loudspeaker: Do you have feedback about the report comment? [Let us know in this issue](https://about.codecov.io/codecov-pr-comment-feedback/?utm_medium=referral&utm_source=github&utm_content=comment&utm_campaign=pr+comments&utm_term=aerleon).	2023-02-03T09:10:15	[]	[]
aerleon/aerleon	aerleon__aerleon-205	0c764e8e94f078b604da1d831ee2312807e8264b	diff --git a/aerleon/lib/aclgenerator.py b/aerleon/lib/aclgenerator.py index b1c95a7c..04573f5e 100644 --- a/aerleon/lib/aclgenerator.py +++ b/aerleon/lib/aclgenerator.py @@ -17,11 +17,13 @@ """ACL Generator base class.""" import copy +import datetime import hashlib -import logging import re import string +from absl import logging + from aerleon.lib import policy @@ -303,6 +305,8 @@ def __init__(self, pol, exp_info): all_err = [] all_warn = [] for header, terms in pol.filters: + terms[:] = self._FilteredTerms(header, terms, exp_info) + # Verify valid keywords # error on unsupported optional keywords that could result # in dangerous or unexpected results @@ -541,6 +545,31 @@ def HexDigest(self, name, truncation_length=None): name_bytes = name.encode('UTF-8') return hashlib.sha256(name_bytes).hexdigest()[:truncation_length] + def _FilteredTerms(self, header, terms, exp_info): + new_terms = [] + filter_name = header.FilterName(self._PLATFORM) + current_date = datetime.datetime.utcnow().date() + exp_info_date = current_date + datetime.timedelta(weeks=exp_info) + + for term in terms: + if term.expiration: + if term.expiration <= current_date: + logging.warning( + 'WARNING: Term %s in policy %s is expired and will not be rendered.', + term.name, + filter_name, + ) + continue + elif term.expiration <= exp_info_date: + logging.info( + "INFO: Term %s in policy %s expires in less than two weeks.", + term.name, + filter_name, + ) + + new_terms.append(term) + + return new_terms def ProtocolNameToNumber(protocols, proto_to_num, name_to_num_map): """Convert a protocol name to a numeric value. diff --git a/aerleon/lib/arista_tp.py b/aerleon/lib/arista_tp.py index 27988fbb..bcf35d97 100644 --- a/aerleon/lib/arista_tp.py +++ b/aerleon/lib/arista_tp.py @@ -16,7 +16,6 @@ """arista traffic-policy generator.""" import copy -import datetime import re from absl import logging @@ -244,7 +243,7 @@ def __str__(self): if not has_match_criteria and not is_default_term: # this term doesn't match on anything and isn't a default-term logging.warning( - "WARNING: term %s has no valid match criteria and " "will not be rendered.", + "WARNING: term %s has no valid match criteria and will not be rendered.", self.term.name, ) return "" @@ -718,9 +717,6 @@ def _TranslatePolicy(self, pol, exp_info): self.arista_traffic_policies = [] af_map_txt = {"inet": "ipv4", "inet6": "ipv6"} - current_date = datetime.datetime.utcnow().date() - exp_info_date = current_date + datetime.timedelta(weeks=exp_info) - for header, terms in pol.filters: filter_options = header.FilterOptions(self._PLATFORM) filter_name = header.FilterName(self._PLATFORM) @@ -778,22 +774,6 @@ def _TranslatePolicy(self, pol, exp_info): if not term: continue - if term.expiration: - if term.expiration <= exp_info_date: - logging.info( - "INFO: term %s in policy %s expires " "in less than two weeks.", - term.name, - filter_name, - ) - if term.expiration <= current_date: - logging.warning( - "WARNING: term %s in policy %s is expired and " - "will not be rendered.", - term.name, - filter_name, - ) - continue - # emit warnings for unsupported options / terms if term.option: unsupported_opts = [] diff --git a/aerleon/lib/aruba.py b/aerleon/lib/aruba.py index 49b2edb6..1eec765d 100644 --- a/aerleon/lib/aruba.py +++ b/aerleon/lib/aruba.py @@ -16,10 +16,10 @@ """Aruba generator.""" -import datetime -import logging from typing import List, Tuple +from absl import logging + from aerleon.lib import aclgenerator _COMMENT_MARKER = '#' @@ -271,9 +271,6 @@ def _BuildTokens(self): def _TranslatePolicy(self, pol, exp_info): self.aruba_policies = [] - current_date = datetime.datetime.utcnow().date() - exp_info_date = current_date + datetime.timedelta(weeks=exp_info) - for header, terms in pol.filters: filter_name = header.FilterName(self._PLATFORM) filter_options = header.FilterOptions(self._PLATFORM) @@ -288,23 +285,6 @@ def _TranslatePolicy(self, pol, exp_info): new_terms = [] for term in terms: - if term.expiration: - if term.expiration <= exp_info_date: - logging.info( - 'INFO: Term %s in policy %s expires ' 'in less than two weeks.', - term.name, - filter_name, - ) - - if term.expiration <= current_date: - logging.warning( - 'WARNING: Term %s in policy %s is expired and ' - 'will not be rendered.', - term.name, - filter_name, - ) - continue - new_terms.append(Term(term, filter_type, verbose)) self.aruba_policies.append((filter_name, new_terms, filter_type)) diff --git a/aerleon/lib/cisco.py b/aerleon/lib/cisco.py index 7d049c2c..26cefbb8 100644 --- a/aerleon/lib/cisco.py +++ b/aerleon/lib/cisco.py @@ -16,7 +16,6 @@ """Cisco generator.""" -import datetime import ipaddress from typing import Union, cast @@ -991,9 +990,6 @@ def _BuildTokens(self): def _TranslatePolicy(self, pol, exp_info): self.cisco_policies = [] - current_date = datetime.datetime.utcnow().date() - exp_info_date = current_date + datetime.timedelta(weeks=exp_info) - # a mixed filter outputs both ipv4 and ipv6 acls in the same output file good_filters = ['extended', 'standard', 'object-group', 'inet6', 'mixed', 'enable_dsmo'] @@ -1060,22 +1056,6 @@ def _TranslatePolicy(self, pol, exp_info): if term.restrict_address_family and term.restrict_address_family != af: continue - if term.expiration: - if term.expiration <= exp_info_date: - logging.info( - 'INFO: Term %s in policy %s expires ' 'in less than two weeks.', - term.name, - filter_name, - ) - if term.expiration <= current_date: - logging.warning( - 'WARNING: Term %s in policy %s is expired and ' - 'will not be rendered.', - term.name, - filter_name, - ) - continue - # render terms based on filter type if next_filter == 'standard': # keep track of sequence numbers across terms diff --git a/aerleon/lib/ciscoasa.py b/aerleon/lib/ciscoasa.py index b3b75d8c..8f7651e5 100644 --- a/aerleon/lib/ciscoasa.py +++ b/aerleon/lib/ciscoasa.py @@ -16,12 +16,12 @@ """Cisco ASA renderer.""" -import datetime import ipaddress -import logging import re from typing import cast +from absl import logging + from aerleon.lib import aclgenerator, cisco, nacaddr, summarizer _ACTION_TABLE = { @@ -335,9 +335,6 @@ def _BuildTokens(self): def _TranslatePolicy(self, pol, exp_info): self.ciscoasa_policies = [] - current_date = datetime.date.today() - exp_info_date = current_date + datetime.timedelta(weeks=exp_info) - for header, terms in self.policy.filters: filter_name = header.FilterName(self._PLATFORM) filter_options = header.FilterOptions(self._PLATFORM) @@ -346,22 +343,6 @@ def _TranslatePolicy(self, pol, exp_info): new_terms = [] # now add the terms for term in terms: - if term.expiration: - if term.expiration <= exp_info_date: - logging.info( - 'INFO: Term %s in policy %s expires ' 'in less than two weeks.', - term.name, - filter_name, - ) - if term.expiration <= current_date: - logging.warning( - 'WARNING: Term %s in policy %s is expired and ' - 'will not be rendered.', - term.name, - filter_name, - ) - continue - new_terms.append(str(Term(term, filter_name, enable_dsmo=enable_dsmo))) self.ciscoasa_policies.append((header, filter_name, new_terms)) diff --git a/aerleon/lib/cloudarmor.py b/aerleon/lib/cloudarmor.py index 56124009..96e3d840 100644 --- a/aerleon/lib/cloudarmor.py +++ b/aerleon/lib/cloudarmor.py @@ -9,7 +9,8 @@ import copy import json -import logging + +from absl import logging from aerleon.lib import aclgenerator diff --git a/aerleon/lib/demo.py b/aerleon/lib/demo.py index 914fb347..0a1a6195 100644 --- a/aerleon/lib/demo.py +++ b/aerleon/lib/demo.py @@ -15,8 +15,7 @@ # """Demo generator for Aerleon.""" -import datetime -import logging +from absl import logging from aerleon.lib import aclgenerator @@ -192,8 +191,6 @@ class Demo(aclgenerator.ACLGenerator): ) def _TranslatePolicy(self, pol, exp_info): - current_date = datetime.date.today() - exp_info_date = current_date + datetime.timedelta(weeks=exp_info) self.demo_policies = [] for header, terms in pol.filters: filter_options = header.FilterOptions('demo') @@ -209,21 +206,6 @@ def _TranslatePolicy(self, pol, exp_info): if term.name in term_names: raise DemoFilterError('Duplicate term name') term_names.add(term.name) - if term.expiration: - if term.expiration <= exp_info_date: - logging.info( - 'INFO: Term %s in policy %s expires ' 'in less than two weeks.', - term.name, - filter_name, - ) - if term.expiration <= current_date: - logging.warning( - 'WARNING: Term %s in policy %s is expired and ' - 'will not be rendered.', - term.name, - filter_name, - ) - continue new_terms.append(Term(term, filter_type)) self.demo_policies.append( (header, filter_name, filter_type, interface_specific, new_terms) diff --git a/aerleon/lib/gce.py b/aerleon/lib/gce.py index 3243dda8..5ea1f4ed 100644 --- a/aerleon/lib/gce.py +++ b/aerleon/lib/gce.py @@ -22,13 +22,13 @@ """ import copy -import datetime import ipaddress import json -import logging import re from typing import Any, Dict +from absl import logging + from aerleon.lib import gcp, nacaddr @@ -444,9 +444,6 @@ def _TranslatePolicy(self, pol, exp_info): total_attribute_count = 0 total_rule_count = 0 - current_date = datetime.datetime.utcnow().date() - exp_info_date = current_date + datetime.timedelta(weeks=exp_info) - for header, terms in pol.filters: filter_options = header.FilterOptions(self._PLATFORM) filter_name = header.FilterName(self._PLATFORM) @@ -479,6 +476,9 @@ def _TranslatePolicy(self, pol, exp_info): else: logging.warning('GCE filter does not specify a network.') + if not terms: + continue + term_names = set() if IsDefaultDeny(terms[-1]): terms[-1].protocol = ['all'] @@ -519,21 +519,6 @@ def _TranslatePolicy(self, pol, exp_info): term_names.add(term.name) term.direction = direction - if term.expiration: - if term.expiration <= exp_info_date: - logging.info( - 'INFO: Term %s in policy %s expires ' 'in less than two weeks.', - term.name, - filter_name, - ) - if term.expiration <= current_date: - logging.warning( - 'WARNING: Term %s in policy %s is expired and ' - 'will not be rendered.', - term.name, - filter_name, - ) - continue if term.option: raise GceFirewallError('GCE firewall does not support term options.') diff --git a/aerleon/lib/iptables.py b/aerleon/lib/iptables.py index 0a811ad6..3a189a1f 100644 --- a/aerleon/lib/iptables.py +++ b/aerleon/lib/iptables.py @@ -16,7 +16,6 @@ """Iptables generator.""" -import datetime import re from string import Template # pylint: disable=g-importing-member @@ -769,9 +768,6 @@ def _WarnIfCustomTarget(self, target): def _TranslatePolicy(self, pol, exp_info): """Translate a policy from objects into strings.""" - current_date = datetime.datetime.utcnow().date() - exp_info_date = current_date + datetime.timedelta(weeks=exp_info) - default_action = None good_default_actions = ['ACCEPT', 'DROP'] good_afs = ['inet', 'inet6'] @@ -869,22 +865,6 @@ def _TranslatePolicy(self, pol, exp_info): if not term: continue - if term.expiration: - if term.expiration <= exp_info_date: - logging.info( - 'INFO: Term %s in policy %s expires ' 'in less than two weeks.', - term.name, - filter_name, - ) - if term.expiration <= current_date: - logging.warning( - 'WARNING: Term %s in policy %s is expired and ' - 'will not be rendered.', - term.name, - filter_name, - ) - continue - new_terms.append( self._TERM( term, diff --git a/aerleon/lib/juniper.py b/aerleon/lib/juniper.py index 61f2f897..3fd85de1 100644 --- a/aerleon/lib/juniper.py +++ b/aerleon/lib/juniper.py @@ -16,7 +16,6 @@ """Juniper JCL generator.""" -import datetime from absl import logging @@ -981,9 +980,6 @@ def _BuildTokens(self): def _TranslatePolicy(self, pol, exp_info): self.juniper_policies = [] - current_date = datetime.datetime.utcnow().date() - exp_info_date = current_date + datetime.timedelta(weeks=exp_info) - for header, terms in pol.filters: filter_options = header.FilterOptions(self._PLATFORM) filter_name = header.FilterName(self._PLATFORM) @@ -1061,21 +1057,6 @@ def _TranslatePolicy(self, pol, exp_info): if not term: continue - if term.expiration: - if term.expiration <= exp_info_date: - logging.info( - 'INFO: Term %s in policy %s expires ' 'in less than two weeks.', - term.name, - filter_name, - ) - if term.expiration <= current_date: - logging.warning( - 'WARNING: Term %s in policy %s is expired and ' - 'will not be rendered.', - term.name, - filter_name, - ) - continue if 'is-fragment' in term.option and filter_type == 'inet6': raise JuniperFragmentInV6Error( 'The term %s uses "is-fragment" but ' 'is a v6 policy.' % term.name diff --git a/aerleon/lib/junipermsmpc.py b/aerleon/lib/junipermsmpc.py index d697643c..006e2fa2 100644 --- a/aerleon/lib/junipermsmpc.py +++ b/aerleon/lib/junipermsmpc.py @@ -15,8 +15,7 @@ # """Juniper MS-MPC generator for Aerleon.""" -import datetime -import logging +from absl import logging from aerleon.lib import aclgenerator, juniper, nacaddr @@ -513,8 +512,6 @@ def _GenerateApplications(self, filter_name): return [] def _TranslatePolicy(self, pol, exp_info): - current_date = datetime.date.today() - exp_info_date = current_date + datetime.timedelta(weeks=exp_info) self.junipermsmpc_policies = [] for header, terms in pol.filters: filter_options = header.FilterOptions(self._PLATFORM) @@ -597,21 +594,6 @@ def _TranslatePolicy(self, pol, exp_info): if term.name in term_names: raise JuniperMSMPCFilterError('Duplicate term name') term_names.add(term.name) - if term.expiration: - if term.expiration <= exp_info_date: - logging.info( - 'INFO: Term %s in policy %s expires ' 'in less than two weeks.', - term.name, - filter_name, - ) - if term.expiration <= current_date: - logging.warning( - 'WARNING: Term %s in policy %s is expired and ' - 'will not be rendered.', - term.name, - filter_name, - ) - continue new_term = Term(term, filter_type, noverbose, filter_name, apply_groups) new_terms.append(new_term) diff --git a/aerleon/lib/junipersrx.py b/aerleon/lib/junipersrx.py index faa39db9..c4a02a9a 100644 --- a/aerleon/lib/junipersrx.py +++ b/aerleon/lib/junipersrx.py @@ -20,7 +20,6 @@ import collections import copy -import datetime import heapq import ipaddress import itertools @@ -361,9 +360,6 @@ def _TranslatePolicy(self, pol, exp_info): ConflictingApplicationSetsError: When two duplicate named terms have conflicting application entries """ - current_date = datetime.datetime.utcnow().date() - exp_info_date = current_date + datetime.timedelta(weeks=exp_info) - for header, terms in pol.filters: filter_options = header.FilterOptions(self._PLATFORM) @@ -499,23 +495,6 @@ def _TranslatePolicy(self, pol, exp_info): raise SRXDuplicateTermError('You have a duplicate term: %s' % term.name) term_dup_check.add(term.name) - if term.expiration: - if term.expiration <= exp_info_date: - logging.info( - 'INFO: Term %s in policy %s>%s expires ' 'in less than two weeks.', - term.name, - self.from_zone, - self.to_zone, - ) - if term.expiration <= current_date: - logging.warning( - 'WARNING: Term %s in policy %s>%s is expired.', - term.name, - self.from_zone, - self.to_zone, - ) - continue - # SRX address books leverage network token names for IPs. # When excluding addresses, we lose those distinct names so we need # to create a new unique name based off the term name before excluding. diff --git a/aerleon/lib/k8s.py b/aerleon/lib/k8s.py index 932f7cca..df7f04ad 100644 --- a/aerleon/lib/k8s.py +++ b/aerleon/lib/k8s.py @@ -21,11 +21,10 @@ """ import copy -import datetime -import logging import re import yaml +from absl import logging from aerleon.lib import aclgenerator @@ -307,9 +306,6 @@ def _TranslatePolicy(self, pol, exp_info): self.network_policies = [] total_rule_count = 0 - current_date = datetime.datetime.utcnow().date() - exp_info_date = current_date + datetime.timedelta(weeks=exp_info) - for header, terms in pol.filters: filter_options = header.FilterOptions(self._PLATFORM) filter_name = header.FilterName(self._PLATFORM) @@ -341,21 +337,6 @@ def _TranslatePolicy(self, pol, exp_info): term_names.add(term.name) term.direction = direction - if term.expiration: - if term.expiration <= current_date: - logging.warning( - 'WARNING: Term %s in policy %s is expired and ' - 'will not be rendered.', - term.name, - filter_name, - ) - continue - if term.expiration <= exp_info_date: - logging.info( - 'INFO: Term %s in policy %s expires ' 'in less than two weeks.', - term.name, - filter_name, - ) if term.option: raise K8sNetworkPolicyError( 'Kubernetes NetworkPolicy does not support term options.' diff --git a/aerleon/lib/nftables.py b/aerleon/lib/nftables.py index 01dfb6ac..865958c9 100644 --- a/aerleon/lib/nftables.py +++ b/aerleon/lib/nftables.py @@ -17,8 +17,8 @@ import collections import copy -import datetime -import logging + +from absl import logging from aerleon.lib import aclgenerator, nacaddr @@ -641,8 +641,6 @@ def _TranslatePolicy(self, pol, exp_info): pol_counter = 0 - current_date = datetime.date.today() - exp_info_date = current_date + datetime.timedelta(weeks=exp_info) for header, terms in pol.filters: filter_options = header.FilterOptions('nftables') ( @@ -683,21 +681,6 @@ def _TranslatePolicy(self, pol, exp_info): term.name, ) continue - if term.expiration: - if term.expiration <= exp_info_date: - logging.info( - 'INFO: Term %s in policy %s expires ' 'in less than two weeks.', - term.name, - nf_af, - ) - if term.expiration <= current_date: - logging.warning( - 'WARNING: Term %s in policy %s is expired and ' - 'will not be rendered.', - term.name, - nf_af, - ) - continue # Handle address excludes before building nft address book dict. for i in term.source_address_exclude: term.source_address = nacaddr.RemoveAddressFromList(term.source_address, i) diff --git a/aerleon/lib/nsxv.py b/aerleon/lib/nsxv.py index 7f252feb..587b735f 100644 --- a/aerleon/lib/nsxv.py +++ b/aerleon/lib/nsxv.py @@ -16,7 +16,6 @@ """Nsxv generator.""" -import datetime import re import xml @@ -522,9 +521,6 @@ def _BuildTokens(self): def _TranslatePolicy(self, pol, exp_info): self.nsxv_policies = [] - current_date = datetime.datetime.utcnow().date() - exp_info_date = current_date + datetime.timedelta(weeks=exp_info) - for header, terms in pol.filters: filter_options = header.FilterOptions(self._PLATFORM) if len(filter_options) >= 2: @@ -544,21 +540,6 @@ def _TranslatePolicy(self, pol, exp_info): raise NsxvDuplicateTermError('There are multiple terms named: %s' % term.name) term_names.add(term.name) - if term.expiration: - if term.expiration <= exp_info_date: - logging.info( - 'INFO: Term %s in policy %s expires ' 'in less than two weeks.', - term.name, - filter_name, - ) - if term.expiration <= current_date: - logging.warning( - 'WARNING: Term %s in policy %s is expired and ' - 'will not be rendered.', - term.name, - filter_name, - ) - continue # Get the mapped action value # If there is no mapped action value term is not rendered mapped_action = _ACTION_TABLE.get(str(term.action[0])) diff --git a/aerleon/lib/openconfig.py b/aerleon/lib/openconfig.py index 54e661ab..e4616fdc 100644 --- a/aerleon/lib/openconfig.py +++ b/aerleon/lib/openconfig.py @@ -21,14 +21,14 @@ """ import copy -import datetime import ipaddress import json -import logging import re from collections import defaultdict from typing import Any, Dict +from absl import logging + from aerleon.lib import aclgenerator @@ -208,9 +208,6 @@ def _TranslatePolicy(self, pol, exp_info): self.oc_policies = [] total_rule_count = 0 - current_date = datetime.datetime.utcnow().date() - exp_info_date = current_date + datetime.timedelta(weeks=exp_info) - for header, terms in pol.filters: filter_options = header.FilterOptions(self._PLATFORM) filter_name = header.FilterName(self._PLATFORM) @@ -227,21 +224,6 @@ def _TranslatePolicy(self, pol, exp_info): for term in terms: - if term.expiration: - if term.expiration <= exp_info_date: - logging.info( - 'INFO: Term %s in policy %s expires ' 'in less than two weeks.', - term.name, - filter_name, - ) - if term.expiration <= current_date: - logging.warning( - 'WARNING: Term %s in policy %s is expired and ' - 'will not be rendered.', - term.name, - filter_name, - ) - continue # TODO(b/196430344): Add support for options such as # established/rst/first-fragment if term.option: diff --git a/aerleon/lib/packetfilter.py b/aerleon/lib/packetfilter.py index 6737c291..f0316539 100644 --- a/aerleon/lib/packetfilter.py +++ b/aerleon/lib/packetfilter.py @@ -18,7 +18,6 @@ import collections import copy -import datetime from typing import cast from absl import logging @@ -458,9 +457,6 @@ def _TranslatePolicy(self, pol, exp_info): self.pf_policies = [] self.address_book = {} self.def_short_to_long = {} - current_date = datetime.datetime.utcnow().date() - exp_info_date = current_date + datetime.timedelta(weeks=exp_info) - good_afs = ['inet', 'inet6', 'mixed'] good_options = ['in', 'out', 'nostate'] all_protocols_stateful = True @@ -573,22 +569,6 @@ def _TranslatePolicy(self, pol, exp_info): if not term: continue - if term.expiration: - if term.expiration <= exp_info_date: - logging.info( - 'INFO: Term %s in policy %s expires ' 'in less than two weeks.', - term.name, - filter_name, - ) - if term.expiration <= current_date: - logging.warning( - 'WARNING: Term %s in policy %s is expired and ' - 'will not be rendered.', - term.name, - filter_name, - ) - continue - if term.destination_interface and term.source_interface: raise Error( "packetfilter only supports destination interface XOR source interface per term" diff --git a/aerleon/lib/paloaltofw.py b/aerleon/lib/paloaltofw.py index 3c3c14e6..e89917f6 100644 --- a/aerleon/lib/paloaltofw.py +++ b/aerleon/lib/paloaltofw.py @@ -16,12 +16,12 @@ import collections import copy -import datetime -import logging import re import xml.etree.ElementTree as etree from xml.dom import minidom +from absl import logging + from aerleon.lib import aclgenerator, addressbook, nacaddr, policy @@ -390,8 +390,6 @@ def _TranslatePolicy(self, pol, exp_info): PaloAltoFWUnsupportedProtocolError: The term contains unsupporter protocol name. """ - current_date = datetime.date.today() - exp_info_date = current_date + datetime.timedelta(weeks=exp_info) first_addr_obj = None address_book_dup_check = set() @@ -514,24 +512,6 @@ def _TranslatePolicy(self, pol, exp_info): % (term.name, ', '.join(term.pan_application), ', '.join(term.protocol)) ) - if term.expiration: - if term.expiration <= exp_info_date: - logging.info( - "INFO: Term %s in policy %s>%s expires " "in less than two weeks.", - term.name, - self.from_zone, - self.to_zone, - ) - if term.expiration <= current_date: - logging.warning( - "WARNING: Term %s in policy %s>%s is expired and " - "will not be rendered.", - term.name, - self.from_zone, - self.to_zone, - ) - continue - for i in term.source_address_exclude: term.source_address = nacaddr.RemoveAddressFromList(term.source_address, i) for i in term.destination_address_exclude: diff --git a/aerleon/lib/pcap.py b/aerleon/lib/pcap.py index 1f7c00a8..01fea9d4 100644 --- a/aerleon/lib/pcap.py +++ b/aerleon/lib/pcap.py @@ -28,7 +28,6 @@ Stolen liberally from packetfilter.py. """ -import datetime from absl import logging @@ -375,9 +374,6 @@ def _BuildTokens(self): def _TranslatePolicy(self, pol, exp_info): self.pcap_policies = [] - current_date = datetime.datetime.utcnow().date() - exp_info_date = current_date + datetime.timedelta(weeks=exp_info) - good_afs = ['inet', 'inet6', 'mixed'] good_options = ['in', 'out'] direction = '' @@ -433,22 +429,6 @@ def _TranslatePolicy(self, pol, exp_info): 'You have a duplicate term: %s' % term.name ) - if term.expiration: - if term.expiration <= exp_info_date: - logging.info( - 'INFO: Term %s in policy %s expires ' 'in less than two weeks.', - term.name, - filter_name, - ) - if term.expiration <= current_date: - logging.warning( - 'WARNING: Term %s in policy %s is expired and ' - 'will not be rendered.', - term.name, - filter_name, - ) - continue - if not term: continue diff --git a/aerleon/lib/policy.py b/aerleon/lib/policy.py index 3e133005..097c7757 100644 --- a/aerleon/lib/policy.py +++ b/aerleon/lib/policy.py @@ -1690,7 +1690,13 @@ def FilterName(self, platform): for target in self.target: if target.platform == platform: if target.options: - return target.options[0] + if platform in ['srx', 'paloalto']: + if len(target.options) >= 3: + return '%s>%s' % (target.options[1], target.options[3]) + else: + return None + else: + return target.options[0] return None def __str__(self): diff --git a/aerleon/lib/windows.py b/aerleon/lib/windows.py index 55dcb932..89172d63 100644 --- a/aerleon/lib/windows.py +++ b/aerleon/lib/windows.py @@ -16,7 +16,6 @@ """Generic Windows security policy generator; requires subclassing.""" -import datetime import string from absl import logging @@ -267,9 +266,6 @@ def _BuildTokens(self): def _TranslatePolicy(self, pol, exp_info): """Translate a policy from objects into strings.""" self.windows_policies = [] - current_date = datetime.datetime.utcnow().date() - exp_info_date = current_date + datetime.timedelta(weeks=exp_info) - default_action = None good_default_actions = ['permit', 'block'] good_options = [] @@ -339,21 +335,6 @@ def _TranslatePolicy(self, pol, exp_info): ) term_names.add(term.name) - if term.expiration: - if term.expiration <= exp_info_date: - logging.info( - 'INFO: Term %s in policy %s expires ' 'in less than two weeks.', - term.name, - filter_name, - ) - if term.expiration <= current_date: - logging.warning( - 'WARNING: Term %s in policy %s is expired and ' - 'will not be rendered.', - term.name, - filter_name, - ) - continue if 'established' in term.option or 'tcp-established' in term.option: continue new_terms.append(self._TERM(term, filter_name, default_action, filter_type))	Hoist "expiration" impl into core Currently the implementation of the 'expiration' property is handled through a block of copy-paste code in every generator. Should a generator author forget to paste in this code, expiration won't work. The core application should handle the expiration implementation and not even pass it through to the generator.		2023-01-26T07:38:34	[]	[]
aerleon/aerleon	aerleon__aerleon-192	44fba2be640eac226dc98a4af0996ce8d0ed2093	diff --git a/aerleon/lib/junipermsmpc.py b/aerleon/lib/junipermsmpc.py index eb338787..b6fbc1ee 100644 --- a/aerleon/lib/junipermsmpc.py +++ b/aerleon/lib/junipermsmpc.py @@ -34,6 +34,7 @@ class Term(juniper.Term): _PLATFORM = 'msmpc' _DEFAULT_INDENT = 20 + _NO_APPLY_GROUPS_INDENT = 12 _ACTIONS = {'accept': 'accept', 'deny': 'discard', 'reject': 'reject'} # msmpc supports a limited number of protocol names # https://www.juniper.net/documentation/us/en/software/junos/security-policies/topics/ref/statement/applications-edit-protocol.html @@ -55,12 +56,13 @@ class Term(juniper.Term): 'udp', ) - def __init__(self, term, term_type, noverbose, filter_name): + def __init__(self, term, term_type, noverbose, filter_name, apply_groups): enable_dsmo = False super().__init__(term, term_type, enable_dsmo, noverbose) self.term = term self.term_type = term_type self.noverbose = noverbose + self.apply_groups = apply_groups self.filter_name = filter_name for prot in self.term.protocol: @@ -81,7 +83,11 @@ def __str__(self): if self.enable_dsmo: raise NotImplementedError('enable_dsmo not implemented for msmpc') - ret_str = juniper.Config(indent=self._DEFAULT_INDENT) + if self.apply_groups: + indent = self._DEFAULT_INDENT + else: + indent = self._NO_APPLY_GROUPS_INDENT + ret_str = juniper.Config(indent=indent) # COMMENTS # this deals just fine with multi line comments, but we could probably @@ -516,11 +522,15 @@ def _TranslatePolicy(self, pol, exp_info): filter_options = header.FilterOptions(self._PLATFORM) filter_name = header.FilterName(self._PLATFORM) filter_options.remove(filter_name) + apply_groups = 'no-apply-groups' not in filter_options filter_direction = None filter_type = None noverbose = 'noverbose' in filter_options self.applications[filter_name] = [] + if not apply_groups: + filter_options.remove('no-apply-groups') + if noverbose: # noverbose is a strict boolean, remove it # prior to iterating through the other options @@ -604,7 +614,7 @@ def _TranslatePolicy(self, pol, exp_info): filter_name, ) continue - new_term = Term(term, filter_type, noverbose, filter_name) + new_term = Term(term, filter_type, noverbose, filter_name, apply_groups) new_terms.append(new_term) # Because MSMPC terms can contain inet and inet6 addresses. We have to @@ -660,7 +670,9 @@ def _TranslatePolicy(self, pol, exp_info): new_application_set['name'] = modified_term_name self.applications[filter_name].append(new_application_set) - self.junipermsmpc_policies.append((header, filter_name, filter_direction, new_terms)) + self.junipermsmpc_policies.append( + (header, filter_name, filter_direction, new_terms, apply_groups) + ) def _Group(self, group, lc=True): """If 1 item return it, else return [ item1 item2 ]. @@ -707,9 +719,16 @@ def _FormattedGroup(el, lc=True): def __str__(self): target = juniper.Config() - for (header, filter_name, filter_direction, terms) in self.junipermsmpc_policies: - target.Append('groups {') - target.Append('replace:') + for ( + header, + filter_name, + filter_direction, + terms, + apply_groups, + ) in self.junipermsmpc_policies: + if apply_groups: + target.Append('groups {') + target.Append('replace:') target.Append('/') # we want the acl to contain id and date tags, but p4 will expand @@ -725,7 +744,8 @@ def __str__(self): target.Append('* ' + line) target.Append('/') - target.Append('%s {' % filter_name) + if apply_groups: + target.Append('%s {' % filter_name) target.Append('services {') target.Append('stateful-firewall {') target.Append('rule %s {' % filter_name) @@ -739,9 +759,10 @@ def __str__(self): target.Append('}') # services { ... } for line in self._GenerateApplications(filter_name): target.Append(line) - target.Append('}') # filter_name { ... } - target.Append('}') # groups { ... } - target.Append('apply-groups %s;' % filter_name) + if apply_groups: + target.Append('}') # filter_name { ... } + target.Append('}') # groups { ... } + target.Append('apply-groups %s;' % filter_name) return str(target) + '\n' diff --git a/docs/reference/generators.md b/docs/reference/generators.md index bb9bf836..d27b4db6 100644 --- a/docs/reference/generators.md +++ b/docs/reference/generators.md @@ -820,7 +820,7 @@ The Juniper MSMPC header designation has the following format: ```yaml targets: - msmpc: [filter name] {inet\|inet6\|mixed} {noverbose} {ingress\|egress} + msmpc: [filter name] {inet\|inet6\|mixed} {noverbose} {ingress\|egress} [no-apply-groups] ``` _filter name_: defines the name of the Juniper msmpc filter. @@ -829,6 +829,7 @@ targets: * _noverbose_: omit additional term and address comments. * _ingress_: filter will be applied in the input direction. * _egress_: filter will be appliced in the output direction. +* _no-apply-groups_: generate configuration without `apply-groups` (optional) When inet4 or inet6 is specified, naming tokens with both IPv4 and IPv6 filters will be rendered using only the specified addresses.	[LEGACY ISSUE PR307] New flag: --[no]apply_groups to enable/disable apply-groups creation for Juniper MS-MPC	There is a PR for this pending. We should invite them to contribute here. My comments for that PR is that these are not appropriate as flags and should be options for the platform itself.	2023-01-23T10:18:58	[]	[]
ndif-team/nnsight	ndif-team__nnsight-179	da36d3cff8c6c4b490daf1e7451b3acb6010b3fd	diff --git a/src/nnsight/__init__.py b/src/nnsight/__init__.py index bde37c2e..a0dff340 100644 --- a/src/nnsight/__init__.py +++ b/src/nnsight/__init__.py @@ -33,10 +33,9 @@ for key, value in getmembers(math, isbuiltin): DEFAULT_PATCHER.add(Patch(math, proxy_wrapper(value), key)) -# TODO THis does not work. Because of accelerate also patching? because they are overloaded? -#DEFAULT_PATCHER.add(Patch(torch, proxy_wrapper(torch.zeros), "zeros")) -# DEFAULT_PATCHER.add(Patch(torch, proxy_wrapper(torch.ones), "ones")) -# DEFAULT_PATCHER.add(Patch(torch, proxy_wrapper(torch.rand), "rand")) +DEFAULT_PATCHER.add(Patch(torch, proxy_wrapper(torch.zeros), "zeros")) +DEFAULT_PATCHER.add(Patch(torch, proxy_wrapper(torch.ones), "ones")) +DEFAULT_PATCHER.add(Patch(torch, proxy_wrapper(torch.rand), "rand")) from torch._subclasses.fake_tensor import FakeTensor diff --git a/src/nnsight/intervention.py b/src/nnsight/intervention.py index e4655574..801930d9 100644 --- a/src/nnsight/intervention.py +++ b/src/nnsight/intervention.py @@ -107,6 +107,8 @@ def grad(self, value: Union[InterventionProxy, Any]) -> None: """ self.node.add(target="swap", args=[self.grad.node, value], value=True) + self.__dict__["_grad"] = None + def __call__(self, args, kwargs) -> Self: # We don't want to call backward on fake tensors @@ -148,7 +150,7 @@ def __setattr__( ) -> None: if key == "grad": - getattr(self.__class__, key).fset(self, value) + return getattr(self.__class__, key).fset(self, value) return super().__setattr__(key, value) diff --git a/src/nnsight/models/NNsightModel.py b/src/nnsight/models/NNsightModel.py index 32d32204..058772c4 100644 --- a/src/nnsight/models/NNsightModel.py +++ b/src/nnsight/models/NNsightModel.py @@ -66,9 +66,7 @@ def __init__( # Otherwise load from _load(...). if not self._custom_model: - # accelerate.init_empty_weights makes all parameters loaded on the 'meta' device. - # Also do .to('meta') because why not. - with accelerate.init_empty_weights(include_buffers=True): + with torch.device('meta'): self._model = self._load(self._model_key, args, **kwargs).to("meta") self._envoy = Envoy(self._model) diff --git a/src/nnsight/pydantics/format/types.py b/src/nnsight/pydantics/format/types.py index 6692006e..55ffa724 100644 --- a/src/nnsight/pydantics/format/types.py +++ b/src/nnsight/pydantics/format/types.py @@ -86,6 +86,14 @@ def compile(self, graph: Graph, nodes: Dict[str, NodeModel]) -> slice: self.step.compile(graph, nodes), ) +class EllipsisModel(BaseModel): + model_config = ConfigDict(arbitrary_types_allowed=True) + type_name: Literal["ELLIPSIS"] = "ELLIPSIS" + + def compile(self, graph: Graph, nodes: Dict[str, NodeModel]) -> type(...): # It will be better to use EllipsisType, but it requires python>=3.10 + return ... + + class ListModel(BaseModel): model_config = ConfigDict(arbitrary_types_allowed=True) @@ -165,6 +173,14 @@ def compile(self, graph: Graph, nodes: Dict[str, NodeModel]) -> FUNCTION: ), ] +EllipsisType = Annotated[ + type(...), # It will be better to use EllipsisType, but it requires python>=3.10 + AfterValidator( + lambda value: EllipsisModel() + ), +] + + ListType = Annotated[list, AfterValidator(lambda value: ListModel(values=value))] TupleType = Annotated[ @@ -201,6 +217,7 @@ def compile(self, graph: Graph, nodes: Dict[str, NodeModel]) -> FUNCTION: ListModel, TupleModel, DictModel, + EllipsisModel ], Field(discriminator="type_name"), ], @@ -212,5 +229,6 @@ def compile(self, graph: Graph, nodes: Dict[str, NodeModel]) -> FUNCTION: ListType, TupleType, DictType, + EllipsisType ], ] diff --git a/src/nnsight/tracing/Node.py b/src/nnsight/tracing/Node.py index f1df651c..d70afc6a 100644 --- a/src/nnsight/tracing/Node.py +++ b/src/nnsight/tracing/Node.py @@ -2,7 +2,8 @@ import inspect import weakref -from typing import TYPE_CHECKING, Any, Callable, Dict, List, Optional, Tuple, Union +from typing import (TYPE_CHECKING, Any, Callable, Dict, List, Optional, Tuple, + Union) import torch from torch._subclasses.fake_tensor import FakeTensor @@ -45,8 +46,6 @@ def __init__( args = list() if kwargs is None: kwargs = dict() - - args = list(args) self.name = name self.graph: "Graph" = graph @@ -64,7 +63,9 @@ def __init__( # Resolve values from completed tracer/runner contexts self.args = util.apply(self.args, lambda x: x.value if x.done() else x, Node) - self.kwargs = util.apply(self.kwargs, lambda x: x.value if x.done() else x, Node) + self.kwargs = util.apply( + self.kwargs, lambda x: x.value if x.done() else x, Node + ) # Add all arguments that are nodes to nodes dependencies # (unless the arg is already .done(), for when you want to apply things to proxies after model execution?) @@ -187,8 +188,8 @@ def prepare_inputs( Returns: Any: Prepared inputs. """ - - inputs = util.apply(inputs, lambda x : x, object) + + inputs = util.apply(inputs, lambda x: x, inspect._empty) def _value(node: Proxy \| Node): diff --git a/src/nnsight/tracing/Proxy.py b/src/nnsight/tracing/Proxy.py index bb0b4c08..3476755e 100755 --- a/src/nnsight/tracing/Proxy.py +++ b/src/nnsight/tracing/Proxy.py @@ -221,13 +221,19 @@ def __rtruediv__(self, other: Union[Proxy, Any]) -> Self: target=operator.truediv, args=[other, self.node], ) - + def __floordiv__(self, other: Union[Proxy, Any]) -> Self: return self.node.add( target=operator.floordiv, args=[self.node, other], ) + def __rfloordiv__(self, other: Union[Proxy, Any]) -> Self: + return self.node.add( + target=operator.floordiv, + args=[other, self.node], + ) + def __eq__(self, other: Union[Proxy, Any]) -> Self: return self.node.add(target=operator.eq, args=[self.node, other])	Meta device Loading parameters to the `'meta'` device & `torch.zeros()`, `torch.ones()`, and `torch.rand()` are now supported by nnsight via patching.		2024-07-23T20:11:40	[]	[]
ndif-team/nnsight	ndif-team__nnsight-81	24253c42dbf19cc6a13235adfced6ba70cc97073	diff --git a/.gitignore b/.gitignore index 78373ef0..1e311508 100755 --- a/.gitignore +++ b/.gitignore @@ -3,4 +3,5 @@ dist/ public/ .egg-info/ .installed.cfg +.vscode __pycache__/ diff --git a/.vscode/settings.json b/.vscode/settings.json deleted file mode 100644 index 45268e01..00000000 --- a/.vscode/settings.json +++ /dev/null @@ -1,5 +0,0 @@ -{ - "restructuredtext.syntaxHighlighting.disabled": true, - "restructuredtext.preview.docutils.disabled": true, - "esbonio.sphinx.confDir": "" -} \ No newline at end of file diff --git a/src/nnsight/intervention.py b/src/nnsight/intervention.py index 5977e68a..461e7908 100644 --- a/src/nnsight/intervention.py +++ b/src/nnsight/intervention.py @@ -162,6 +162,11 @@ def shape(self) -> Collection[torch.Size]: if not self.node.is_tracing(): return util.apply(self.value, lambda x: x.shape, torch.Tensor) + + # If we haven't scanned in a proxy_value, just return a proxy to get the attribute. + if self.node.proxy_value is None: + + return super().__getattr__('shape') return util.apply(self.node.proxy_value, lambda x: x.shape, torch.Tensor) @@ -176,6 +181,11 @@ def device(self) -> Collection[torch.device]: if not self.node.is_tracing(): return util.apply(self.value, lambda x: x.device, torch.Tensor) + + # If we haven't scanned in a proxy_value, just return a proxy to get the attribute. + if self.node.proxy_value is None: + + return super().__getattr__('device') return util.apply(self.node.proxy_value, lambda x: x.device, torch.Tensor) @@ -190,6 +200,11 @@ def dtype(self) -> Collection[torch.device]: if not self.node.is_tracing(): return util.apply(self.value, lambda x: x.dtype, torch.Tensor) + + # If we haven't scanned in a proxy_value, just return a proxy to get the attribute. + if self.node.proxy_value is None: + + return super().__getattr__('dtype') return util.apply(self.node.proxy_value, lambda x: x.dtype, torch.Tensor) diff --git a/src/nnsight/tracing/Graph.py b/src/nnsight/tracing/Graph.py index d8647434..184725b7 100644 --- a/src/nnsight/tracing/Graph.py +++ b/src/nnsight/tracing/Graph.py @@ -5,7 +5,7 @@ from typing import Any, Callable, Dict, List, Type, Union import torch -from torch._subclasses.fake_tensor import FakeTensorMode +from torch._subclasses.fake_tensor import FakeCopyMode, FakeTensorMode from torch.fx.experimental.symbolic_shapes import ShapeEnv from .. import util @@ -141,18 +141,13 @@ def add( allow_non_fake_inputs=True, shape_env=ShapeEnv(assume_static_by_default=True), ) as fake_mode: - - try: - + with FakeCopyMode(fake_mode): + value = target( Node.prepare_proxy_values(_args), **Node.prepare_proxy_values(_kwargs), ) - except RuntimeError: - - value = None - target_name = target if isinstance(target, str) else target.__name__ if target_name not in self.name_idx: diff --git a/src/nnsight/tracing/Node.py b/src/nnsight/tracing/Node.py index aff0dd77..97530e6e 100644 --- a/src/nnsight/tracing/Node.py +++ b/src/nnsight/tracing/Node.py @@ -4,6 +4,7 @@ from typing import TYPE_CHECKING, Any, Callable, Dict, List, Optional, Tuple, Union import torch +from torch._subclasses.fake_tensor import FakeTensor from .. import util from ..logger import logger @@ -59,6 +60,22 @@ def slice_to_value(arg: slice): if torch._GLOBAL_DEVICE_CONTEXT is not None else None ) + + if device is None: + + # Arguments might be tensors created outside of scanning. Also the model might be a 'meta' pre-dispatched version of the model. + # That means the tensors as args and the model are different devices but we dont want to have to have the users move tensors to 'meta' + # So only when theres a FakeTensor with device meta, we move other tensors also to meta. + + def get_device(tensor:torch.Tensor): + + nonlocal device + + if device is None and isinstance(tensor, FakeTensor) and tensor.device.type == 'meta': + + device = tensor.device.type + + util.apply(values, get_device, torch.Tensor) if device is not None:	If theres no scanned in value for a Proxy and .shape, .dtype, or .de… …vice is called on it, return a Proxy of the operation, not the actual attribute on the proxy_value		2024-03-01T20:06:29	[]	[]
ndif-team/nnsight	ndif-team__nnsight-77	0513025c6112b8b1512b5637a1cc7592f5730136	diff --git a/.DS_Store b/.DS_Store new file mode 100644 index 00000000..ecf0fd2f Binary files /dev/null and b/.DS_Store differ diff --git a/NNsight_v0_2.ipynb b/NNsight_v0_2.ipynb index 675ea875..b95e9a11 100644 --- a/NNsight_v0_2.ipynb +++ b/NNsight_v0_2.ipynb @@ -1,1191 +1,1627 @@ { - "nbformat": 4, - "nbformat_minor": 0, - "metadata": { - "colab": { - "provenance": [], - "toc_visible": true, - "include_colab_link": true + "cells": [ + { + "cell_type": "markdown", + "metadata": { + "colab_type": "text", + "id": "view-in-github" + }, + "source": [ + "<a href=\"https://colab.research.google.com/github/JadenFiotto-Kaufman/nnsight/blob/dev/NNsight_v0_2.ipynb\" target=\"_parent\"><img src=\"https://colab.research.google.com/assets/colab-badge.svg\" alt=\"Open In Colab\"/></a>" + ] }, - "kernelspec": { - "name": "python3", - "display_name": "Python 3" + { + "cell_type": "markdown", + "metadata": { + "id": "-Scfr942GwO4" + }, + "source": [ + "<img src=\"https://nnsight.net/_static/images/nnsight_logo.svg\" alt=\"drawing\" width=\"200\"/>\n", + "\n", + "# NNsight\n", + "## The API for a transparent science on black-box AI\n", + "\n", + "In this era of large-scale deep learning, the most interesting AI models are massive black boxes that are hard to run. Ordinary commercial inference service APIs let you interact with huge models, but they do not let you access model internals.\n", + "\n", + "The nnsight library is different: it gives you full access to all the neural network internals. When used together with a remote service like the [National Deep Inference Facility](https://thevisible.net/docs/NDIF-proposal.pdf) (NDIF), it lets you run complex experiments on huge open source models easily, with fully transparent access.\n", + "\n", + "Our team wants to enable entire labs and independent researchers alike, as we believe a large, passionate, and collaborative community will produce the next big insights on a profoundly important field.\n", + "\n", + "\n" + ] }, - "language_info": { - "name": "python" + { + "cell_type": "markdown", + "metadata": { + "id": "N1OemD2VGyZx" + }, + "source": [ + "# 1 First, let's start small\n", + "\n", + "\n", + "\n" + ] }, - "widgets": { - "application/vnd.jupyter.widget-state+json": { - "0a676b8f2dc34fed9731f19a03e4bfa1": { - "model_module": "@jupyter-widgets/controls", - "model_name": "HBoxModel", - "model_module_version": "1.5.0", - "state": { - "_dom_classes": [], - "_model_module": "@jupyter-widgets/controls", - "_model_module_version": "1.5.0", - "_model_name": "HBoxModel", - "_view_count": null, - "_view_module": "@jupyter-widgets/controls", - "_view_module_version": "1.5.0", - "_view_name": "HBoxView", - "box_style": "", - "children": [ - "IPY_MODEL_74358c341daa44faa41a92b0eb073615", - "IPY_MODEL_c40e17a994ed44d79ef54421f4772858", - "IPY_MODEL_431ffcb0dbb1419dbbb237a29842b905" - ], - "layout": "IPY_MODEL_c28c185a508e459b95edb29afedd9c79" - } - }, - "74358c341daa44faa41a92b0eb073615": { - "model_module": "@jupyter-widgets/controls", - "model_name": "HTMLModel", - "model_module_version": "1.5.0", - "state": { - "_dom_classes": [], - "_model_module": "@jupyter-widgets/controls", - "_model_module_version": "1.5.0", - "_model_name": "HTMLModel", - "_view_count": null, - "_view_module": "@jupyter-widgets/controls", - "_view_module_version": "1.5.0", - "_view_name": "HTMLView", - "description": "", - "description_tooltip": null, - "layout": "IPY_MODEL_7a0f3d2deeb14bbba7f0df208133a7e2", - "placeholder": "", - "style": "IPY_MODEL_94cdcf1fb0914dfe99c3c615e68b7ec8", - "value": "config.json: 100%" - } + { + "cell_type": "markdown", + "metadata": { + "id": "WyLMmhrAKTNM" + }, + "source": [ + "## The Tracing Context\n", + "\n", + "To demonstrate the core functionality and syntax of nnsight, we'll define and use a tiny two layer neural network." + ] + }, + { + "cell_type": "code", + "execution_count": 1, + "metadata": { + "id": "R0RJijD0eXwf" + }, + "outputs": [], + "source": [ + "# Install nnsight\n", + "!pip install nnsight\n", + "\n", + "from IPython.display import clear_output\n", + "\n", + "clear_output()" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "bgydw7i3HmIH" + }, + "source": [ + "Our little model here is composed of four sub-modules, two linear layers ('layer1', 'layer2'). We specify the sizes of each of these modules, and create some complementary example input." + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "id": "2pX2Wg8Ceo6N" + }, + "outputs": [], + "source": [ + "from collections import OrderedDict\n", + "import torch\n", + "\n", + "input_size = 5\n", + "hidden_dims = 10\n", + "output_size = 2\n", + "\n", + "net = torch.nn.Sequential(OrderedDict([\n", + " ('layer1', torch.nn.Linear(input_size, hidden_dims)),\n", + " ('layer2', torch.nn.Linear(hidden_dims, output_size)),\n", + "])).requires_grad_(False)\n", + "\n", + "input = torch.rand((1, input_size))" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "tIPa2h2pJIwl" + }, + "source": [ + "The core object of the nnsight package is `NNsight`. This wraps around a given pytorch model to enable the capabilites nnsight provides." + ] + }, + { + "cell_type": "code", + "execution_count": 2, + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/", + "height": 426, + "referenced_widgets": [ + "b3078a7a41c642839748d9864220cbde", + "3aa1aef302474a0cba55152e311acd40", + "dda1704b991a44dfb2b28191f307e081", + "5c0847c658d3489ca529d1afc8c96ed5", + "9fbbb0bf7d8e4c12939193791efc590f", + "ded9b4b26ce14ba0aae0b19e0274fcd0", + "47b21a1460f446d9911f2d06c9dd5bbd", + "21f5a81dd29549e69b5cfa2661371e35", + "5f608f4313d64e458314ace8237708b1", + "5266eb8eb83946e0acbda1362f3bd48a", + "8d7ef198c08b4a4ab49501e2204bdbb6" + ] }, - "c40e17a994ed44d79ef54421f4772858": { - "model_module": "@jupyter-widgets/controls", - "model_name": "FloatProgressModel", - "model_module_version": "1.5.0", - "state": { - "_dom_classes": [], - "_model_module": "@jupyter-widgets/controls", - "_model_module_version": "1.5.0", - "_model_name": "FloatProgressModel", - "_view_count": null, - "_view_module": "@jupyter-widgets/controls", - "_view_module_version": "1.5.0", - "_view_name": "ProgressView", - "bar_style": "success", - "description": "", - "description_tooltip": null, - "layout": "IPY_MODEL_f3e24aa6619e4185a3e2f53f2e473e45", - "max": 665, - "min": 0, - "orientation": "horizontal", - "style": "IPY_MODEL_b7be20bd738940658f710309729d8ccd", - "value": 665 - } + "id": "8H9R_ynTJI5y", + "outputId": "5d6b545d-750e-441a-fde9-2dc53e2f1f5e" + }, + "outputs": [ + { + "name": "stderr", + "output_type": "stream", + "text": [ + "The cache for model files in Transformers v4.22.0 has been updated. Migrating your old cache. This is a one-time only operation. You can interrupt this and resume the migration later on by calling `transformers.utils.move_cache()`.\n" + ] }, - "431ffcb0dbb1419dbbb237a29842b905": { - "model_module": "@jupyter-widgets/controls", - "model_name": "HTMLModel", - "model_module_version": "1.5.0", - "state": { - "_dom_classes": [], - "_model_module": "@jupyter-widgets/controls", - "_model_module_version": "1.5.0", - "_model_name": "HTMLModel", - "_view_count": null, - "_view_module": "@jupyter-widgets/controls", - "_view_module_version": "1.5.0", - "_view_name": "HTMLView", - "description": "", - "description_tooltip": null, - "layout": "IPY_MODEL_562a287082eb4882af864062ae13c91b", - "placeholder": "", - "style": "IPY_MODEL_11e99924bdf84739b3cfc7d676b797e5", - "value": " 665/665 [00:00<00:00, 12.8kB/s]" - } + { + "data": { + "application/vnd.jupyter.widget-view+json": { + "model_id": "b3078a7a41c642839748d9864220cbde", + "version_major": 2, + "version_minor": 0 + }, + "text/plain": [ + "0it [00:00, ?it/s]" + ] + }, + "metadata": {}, + "output_type": "display_data" }, - "c28c185a508e459b95edb29afedd9c79": { - "model_module": "@jupyter-widgets/base", - "model_name": "LayoutModel", - "model_module_version": "1.2.0", - "state": { - "_model_module": "@jupyter-widgets/base", - "_model_module_version": "1.2.0", - "_model_name": "LayoutModel", - "_view_count": null, - "_view_module": "@jupyter-widgets/base", - "_view_module_version": "1.2.0", - "_view_name": "LayoutView", - "align_content": null, - "align_items": null, - "align_self": null, - "border": null, - "bottom": null, - "display": null, - "flex": null, - "flex_flow": null, - "grid_area": null, - "grid_auto_columns": null, - "grid_auto_flow": null, - "grid_auto_rows": null, - "grid_column": null, - "grid_gap": null, - "grid_row": null, - "grid_template_areas": null, - "grid_template_columns": null, - "grid_template_rows": null, - "height": null, - "justify_content": null, - "justify_items": null, - "left": null, - "margin": null, - "max_height": null, - "max_width": null, - "min_height": null, - "min_width": null, - "object_fit": null, - "object_position": null, - "order": null, - "overflow": null, - "overflow_x": null, - "overflow_y": null, - "padding": null, - "right": null, - "top": null, - "visibility": null, - "width": null - } - }, - "7a0f3d2deeb14bbba7f0df208133a7e2": { - "model_module": "@jupyter-widgets/base", - "model_name": "LayoutModel", - "model_module_version": "1.2.0", - "state": { - "_model_module": "@jupyter-widgets/base", - "_model_module_version": "1.2.0", - "_model_name": "LayoutModel", - "_view_count": null, - "_view_module": "@jupyter-widgets/base", - "_view_module_version": "1.2.0", - "_view_name": "LayoutView", - "align_content": null, - "align_items": null, - "align_self": null, - "border": null, - "bottom": null, - "display": null, - "flex": null, - "flex_flow": null, - "grid_area": null, - "grid_auto_columns": null, - "grid_auto_flow": null, - "grid_auto_rows": null, - "grid_column": null, - "grid_gap": null, - "grid_row": null, - "grid_template_areas": null, - "grid_template_columns": null, - "grid_template_rows": null, - "height": null, - "justify_content": null, - "justify_items": null, - "left": null, - "margin": null, - "max_height": null, - "max_width": null, - "min_height": null, - "min_width": null, - "object_fit": null, - "object_position": null, - "order": null, - "overflow": null, - "overflow_x": null, - "overflow_y": null, - "padding": null, - "right": null, - "top": null, - "visibility": null, - "width": null - } - }, - "94cdcf1fb0914dfe99c3c615e68b7ec8": { - "model_module": "@jupyter-widgets/controls", - "model_name": "DescriptionStyleModel", - "model_module_version": "1.5.0", - "state": { - "_model_module": "@jupyter-widgets/controls", - "_model_module_version": "1.5.0", - "_model_name": "DescriptionStyleModel", - "_view_count": null, - "_view_module": "@jupyter-widgets/base", - "_view_module_version": "1.2.0", - "_view_name": "StyleView", - "description_width": "" - } - }, - "f3e24aa6619e4185a3e2f53f2e473e45": { - "model_module": "@jupyter-widgets/base", - "model_name": "LayoutModel", - "model_module_version": "1.2.0", - "state": { - "_model_module": "@jupyter-widgets/base", - "_model_module_version": "1.2.0", - "_model_name": "LayoutModel", - "_view_count": null, - "_view_module": "@jupyter-widgets/base", - "_view_module_version": "1.2.0", - "_view_name": "LayoutView", - "align_content": null, - "align_items": null, - "align_self": null, - "border": null, - "bottom": null, - "display": null, - "flex": null, - "flex_flow": null, - "grid_area": null, - "grid_auto_columns": null, - "grid_auto_flow": null, - "grid_auto_rows": null, - "grid_column": null, - "grid_gap": null, - "grid_row": null, - "grid_template_areas": null, - "grid_template_columns": null, - "grid_template_rows": null, - "height": null, - "justify_content": null, - "justify_items": null, - "left": null, - "margin": null, - "max_height": null, - "max_width": null, - "min_height": null, - "min_width": null, - "object_fit": null, - "object_position": null, - "order": null, - "overflow": null, - "overflow_x": null, - "overflow_y": null, - "padding": null, - "right": null, - "top": null, - "visibility": null, - "width": null - } - }, - "b7be20bd738940658f710309729d8ccd": { - "model_module": "@jupyter-widgets/controls", - "model_name": "ProgressStyleModel", - "model_module_version": "1.5.0", - "state": { - "_model_module": "@jupyter-widgets/controls", - "_model_module_version": "1.5.0", - "_model_name": "ProgressStyleModel", - "_view_count": null, - "_view_module": "@jupyter-widgets/base", - "_view_module_version": "1.2.0", - "_view_name": "StyleView", - "bar_color": null, - "description_width": "" - } - }, - "562a287082eb4882af864062ae13c91b": { - "model_module": "@jupyter-widgets/base", - "model_name": "LayoutModel", - "model_module_version": "1.2.0", - "state": { - "_model_module": "@jupyter-widgets/base", - "_model_module_version": "1.2.0", - "_model_name": "LayoutModel", - "_view_count": null, - "_view_module": "@jupyter-widgets/base", - "_view_module_version": "1.2.0", - "_view_name": "LayoutView", - "align_content": null, - "align_items": null, - "align_self": null, - "border": null, - "bottom": null, - "display": null, - "flex": null, - "flex_flow": null, - "grid_area": null, - "grid_auto_columns": null, - "grid_auto_flow": null, - "grid_auto_rows": null, - "grid_column": null, - "grid_gap": null, - "grid_row": null, - "grid_template_areas": null, - "grid_template_columns": null, - "grid_template_rows": null, - "height": null, - "justify_content": null, - "justify_items": null, - "left": null, - "margin": null, - "max_height": null, - "max_width": null, - "min_height": null, - "min_width": null, - "object_fit": null, - "object_position": null, - "order": null, - "overflow": null, - "overflow_x": null, - "overflow_y": null, - "padding": null, - "right": null, - "top": null, - "visibility": null, - "width": null - } - }, - "11e99924bdf84739b3cfc7d676b797e5": { - "model_module": "@jupyter-widgets/controls", - "model_name": "DescriptionStyleModel", - "model_module_version": "1.5.0", - "state": { - "_model_module": "@jupyter-widgets/controls", - "_model_module_version": "1.5.0", - "_model_name": "DescriptionStyleModel", - "_view_count": null, - "_view_module": "@jupyter-widgets/base", - "_view_module_version": "1.2.0", - "_view_name": "StyleView", - "description_width": "" - } - }, - "60fe026e908741209637604edfeb0563": { - "model_module": "@jupyter-widgets/controls", - "model_name": "HBoxModel", - "model_module_version": "1.5.0", - "state": { - "_dom_classes": [], - "_model_module": "@jupyter-widgets/controls", - "_model_module_version": "1.5.0", - "_model_name": "HBoxModel", - "_view_count": null, - "_view_module": "@jupyter-widgets/controls", - "_view_module_version": "1.5.0", - "_view_name": "HBoxView", - "box_style": "", - "children": [ - "IPY_MODEL_e7a73cfc3fa04425afc08368532ef0be", - "IPY_MODEL_d839da1c31a34659988e3420ba520725", - "IPY_MODEL_96a3b8c765374a5e98eaf713dd5e85d8" - ], - "layout": "IPY_MODEL_b49ce0d00e5b4df097b3a9fc4c0bd8e7" - } - }, - "e7a73cfc3fa04425afc08368532ef0be": { - "model_module": "@jupyter-widgets/controls", - "model_name": "HTMLModel", - "model_module_version": "1.5.0", - "state": { - "_dom_classes": [], - "_model_module": "@jupyter-widgets/controls", - "_model_module_version": "1.5.0", - "_model_name": "HTMLModel", - "_view_count": null, - "_view_module": "@jupyter-widgets/controls", - "_view_module_version": "1.5.0", - "_view_name": "HTMLView", - "description": "", - "description_tooltip": null, - "layout": "IPY_MODEL_b2748df9aff540df954f002310606a56", - "placeholder": "", - "style": "IPY_MODEL_f4caf2bc58704464b6ac7bf6da31b71d", - "value": "vocab.json: 100%" - } - }, - "d839da1c31a34659988e3420ba520725": { - "model_module": "@jupyter-widgets/controls", - "model_name": "FloatProgressModel", - "model_module_version": "1.5.0", - "state": { - "_dom_classes": [], - "_model_module": "@jupyter-widgets/controls", - "_model_module_version": "1.5.0", - "_model_name": "FloatProgressModel", - "_view_count": null, - "_view_module": "@jupyter-widgets/controls", - "_view_module_version": "1.5.0", - "_view_name": "ProgressView", - "bar_style": "success", - "description": "", - "description_tooltip": null, - "layout": "IPY_MODEL_3731a5747724403e86ef1abaf4fb389b", - "max": 1042301, - "min": 0, - "orientation": "horizontal", - "style": "IPY_MODEL_688bf676958a4050b73cc9597f5110de", - "value": 1042301 - } - }, - "96a3b8c765374a5e98eaf713dd5e85d8": { - "model_module": "@jupyter-widgets/controls", - "model_name": "HTMLModel", - "model_module_version": "1.5.0", - "state": { - "_dom_classes": [], - "_model_module": "@jupyter-widgets/controls", - "_model_module_version": "1.5.0", - "_model_name": "HTMLModel", - "_view_count": null, - "_view_module": "@jupyter-widgets/controls", - "_view_module_version": "1.5.0", - "_view_name": "HTMLView", - "description": "", - "description_tooltip": null, - "layout": "IPY_MODEL_8776d04bfd3748b69e10708466e9c132", - "placeholder": "", - "style": "IPY_MODEL_56490c20aef04fe4a63ba7c990e71e77", - "value": " 1.04M/1.04M [00:00<00:00, 3.16MB/s]" - } + { + "ename": "ImportError", + "evalue": "cannot import name 'NNsight' from 'nnsight' (/usr/local/lib/python3.10/dist-packages/nnsight/__init__.py)", + "output_type": "error", + "traceback": [ + "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m", + "\u001b[0;31mImportError\u001b[0m Traceback (most recent call last)", + "\u001b[0;32m<ipython-input-2-801ad166c2ee>\u001b[0m in \u001b[0;36m<cell line: 1>\u001b[0;34m()\u001b[0m\n\u001b[0;32m----> 1\u001b[0;31m \u001b[0;32mfrom\u001b[0m \u001b[0mnnsight\u001b[0m \u001b[0;32mimport\u001b[0m \u001b[0mNNsight\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m 2\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 3\u001b[0m \u001b[0mmodel\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mNNsight\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mnet\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n", + "\u001b[0;31mImportError\u001b[0m: cannot import name 'NNsight' from 'nnsight' (/usr/local/lib/python3.10/dist-packages/nnsight/__init__.py)", + "", + "\u001b[0;31m---------------------------------------------------------------------------\u001b[0;32m\nNOTE: If your import is failing due to a missing package, you can\nmanually install dependencies using either !pip or !apt.\n\nTo view examples of installing some common dependencies, click the\n\"Open Examples\" button below.\n\u001b[0;31m---------------------------------------------------------------------------\u001b[0m\n" + ] + } + ], + "source": [ + "from nnsight import NNsight\n", + "\n", + "model = NNsight(net)" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "2NfISlQ_Ilvp" + }, + "source": [ + "Printing a Pytorch model shows a named hierarchy of modules which is very useful when accessing sub-components directly. NNsight models work the same." + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/" }, - "b49ce0d00e5b4df097b3a9fc4c0bd8e7": { - "model_module": "@jupyter-widgets/base", - "model_name": "LayoutModel", - "model_module_version": "1.2.0", - "state": { - "_model_module": "@jupyter-widgets/base", - "_model_module_version": "1.2.0", - "_model_name": "LayoutModel", - "_view_count": null, - "_view_module": "@jupyter-widgets/base", - "_view_module_version": "1.2.0", - "_view_name": "LayoutView", - "align_content": null, - "align_items": null, - "align_self": null, - "border": null, - "bottom": null, - "display": null, - "flex": null, - "flex_flow": null, - "grid_area": null, - "grid_auto_columns": null, - "grid_auto_flow": null, - "grid_auto_rows": null, - "grid_column": null, - "grid_gap": null, - "grid_row": null, - "grid_template_areas": null, - "grid_template_columns": null, - "grid_template_rows": null, - "height": null, - "justify_content": null, - "justify_items": null, - "left": null, - "margin": null, - "max_height": null, - "max_width": null, - "min_height": null, - "min_width": null, - "object_fit": null, - "object_position": null, - "order": null, - "overflow": null, - "overflow_x": null, - "overflow_y": null, - "padding": null, - "right": null, - "top": null, - "visibility": null, - "width": null - } + "id": "bYtnbJHvlGZV", + "outputId": "5eb7c572-451c-43c6-956f-af224ae1867c" + }, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "Sequential(\n", + " (layer1): Linear(in_features=5, out_features=10, bias=True)\n", + " (layer2): Linear(in_features=10, out_features=2, bias=True)\n", + ")\n" + ] + } + ], + "source": [ + "print(model)" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "djC5kyJWLuUH" + }, + "source": [ + "Before we actually get to using the model we just created, let's talk about Python contexts.\n", + "\n", + "Python contexts define a scope using the `with` statement and are often used to create some object, or initiate some logic, that you later want to destroy or conclude.\n", + "\n", + "The most common application is opening files like the following example:\n", + "\n", + "```python\n", + "with open('myfile.txt', 'r') as file:\n", + " text = file.read()\n", + "```\n", + "\n", + "Python uses the `with` keyword to enter a context-like object. This object defines logic to be run at the start of the `with` block, as well as logic to be run when exiting. When using `with` for a file, entering the context opens the file and exiting the context closes it. Being within the context means we can read from the file. Simple enough! Now we can discuss how `nnsight` uses contexts to enable intuitive access into the internals of a neural network.\n" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "iNvuCOeyojcA" + }, + "source": [ + "The main tool with `nnsight` is a context for tracing.\n", + "\n", + "We enter the tracing context by calling `model.trace(<input>)` on an `NNsight` model, which defines how we want to run the model. Inside the context, we will be able to customize how the neural network runs. The model is actually run upon exiting the tracing context." + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "id": "qEXQ4auPSL-m" + }, + "outputs": [], + "source": [ + "with model.trace(input) as tracer:\n", + " pass" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "UZQsHinjqicJ" + }, + "source": [ + "But where's the output? To get that, we'll have to learn how to request it from within the tracing conext." + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "xMfBpYzDPMoB" + }, + "source": [ + "## Getting" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "B5_aFwFRv0ax" + }, + "source": [ + "Earlier, when we wrapped our little neural net with the `NNsight` class. This added a couple properties to each module in the model (including the root model itself). The two most important ones are `.input` and `.output`.\n", + "\n", + "```python\n", + "model.input\n", + "model.output\n", + "```\n", + "\n", + "The names are self explainatory. They correspond to the inputs and outputs of their respective modules during a forward pass of the model. We can use these attributes inside the `with` block.\n", + "\n", + "However, it is important to understand that the model is not executed until the end of the tracing context. How can we access inputs and outputs before the model is run? The trick is deferred execution.\n", + "\n", + "`.input` and `.output` are Proxies for the eventual inputs and outputs of a module. In other words, when you access `model.output` what you are communicating to `nnsight` is, \"When you compute the output of `model`, please grab it for me and put the value into its corresponding Proxy object's `.value` attribute.\" Let's try it:" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/", + "height": 303 }, - "b2748df9aff540df954f002310606a56": { - "model_module": "@jupyter-widgets/base", - "model_name": "LayoutModel", - "model_module_version": "1.2.0", - "state": { - "_model_module": "@jupyter-widgets/base", - "_model_module_version": "1.2.0", - "_model_name": "LayoutModel", - "_view_count": null, - "_view_module": "@jupyter-widgets/base", - "_view_module_version": "1.2.0", - "_view_name": "LayoutView", - "align_content": null, - "align_items": null, - "align_self": null, - "border": null, - "bottom": null, - "display": null, - "flex": null, - "flex_flow": null, - "grid_area": null, - "grid_auto_columns": null, - "grid_auto_flow": null, - "grid_auto_rows": null, - "grid_column": null, - "grid_gap": null, - "grid_row": null, - "grid_template_areas": null, - "grid_template_columns": null, - "grid_template_rows": null, - "height": null, - "justify_content": null, - "justify_items": null, - "left": null, - "margin": null, - "max_height": null, - "max_width": null, - "min_height": null, - "min_width": null, - "object_fit": null, - "object_position": null, - "order": null, - "overflow": null, - "overflow_x": null, - "overflow_y": null, - "padding": null, - "right": null, - "top": null, - "visibility": null, - "width": null - } + "id": "cYe8-r9ptGaG", + "outputId": "adf25efe-d2cf-4f65-9eaf-15b155762f4b" + }, + "outputs": [ + { + "ename": "ValueError", + "evalue": "Accessing Proxy value before it's been set.", + "output_type": "error", + "traceback": [ + "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m", + "\u001b[0;31mValueError\u001b[0m Traceback (most recent call last)", + "\u001b[0;32m<ipython-input-34-c7e0c74b12fa>\u001b[0m in \u001b[0;36m<cell line: 5>\u001b[0;34m()\u001b[0m\n\u001b[1;32m 3\u001b[0m \u001b[0moutput\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mmodel\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0moutput\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 4\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m----> 5\u001b[0;31m \u001b[0mprint\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0moutput\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mvalue\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m", + "\u001b[0;32m/usr/local/lib/python3.10/dist-packages/nnsight/tracing/Proxy.py\u001b[0m in \u001b[0;36mvalue\u001b[0;34m(self)\u001b[0m\n\u001b[1;32m 47\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 48\u001b[0m \u001b[0;32mif\u001b[0m \u001b[0;32mnot\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mnode\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mdone\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m---> 49\u001b[0;31m \u001b[0;32mraise\u001b[0m \u001b[0mValueError\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m\"Accessing Proxy value before it's been set.\"\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m 50\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 51\u001b[0m \u001b[0;32mreturn\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mnode\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mvalue\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n", + "\u001b[0;31mValueError\u001b[0m: Accessing Proxy value before it's been set." + ] + } + ], + "source": [ + "with model.trace(input) as tracer:\n", + "\n", + " output = model.output\n", + "\n", + "print(output.value)" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "sBz5qfwuR-6F" + }, + "source": [ + "Oh no an error! \"Accessing Proxy value before it's been set.\"\n", + "\n", + "Why doesn't our `output` have a `value`?\n", + "\n", + "Proxy objects will only have their value at the end of a context if we call `.save()` on them. This helps to reduce memory costs. Adding `.save()` fixes the error:" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/" }, - "f4caf2bc58704464b6ac7bf6da31b71d": { - "model_module": "@jupyter-widgets/controls", - "model_name": "DescriptionStyleModel", - "model_module_version": "1.5.0", - "state": { - "_model_module": "@jupyter-widgets/controls", - "_model_module_version": "1.5.0", - "_model_name": "DescriptionStyleModel", - "_view_count": null, - "_view_module": "@jupyter-widgets/base", - "_view_module_version": "1.2.0", - "_view_name": "StyleView", - "description_width": "" - } + "id": "y_bXRd5dvsBu", + "outputId": "a8e8985c-2661-46a8-975b-d43f67cfc162" + }, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "tensor([[ 0.1473, -0.1518]])\n" + ] + } + ], + "source": [ + "with model.trace(input) as tracer:\n", + "\n", + " output = model.output.save()\n", + "\n", + "print(output.value)" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "t5C0UZCwvrYn" + }, + "source": [ + "Success! We now have the model output. You just completed your first intervention using `nnsight`.\n", + "\n", + "\n", + "Each time you access a module's input or output, you create an intervention in the neural network's forward pass. Collectively these requests form the intervention graph. We call the process of executing it alongside the model's normal computation graph, interleaving.\n", + "\n", + "<details>\n", + "<summary>On Model output</summary>\n", + "\n", + "---\n", + "\n", + "If you don't need to access anything other than the final model output, you can call the tracing context with `trace=False` and not use it as a context:\n", + "\n", + "```python\n", + "output = model.trace(<inputs>, trace=False)\n", + "```\n", + "\n", + "---\n", + "\n", + "</details>\n", + "\n", + "Just like we saved the output of the model as a whole, we can save the output of any of its submodules. We use normal Python attribute syntax. We can discover how to access them by name by printing out the model:" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/" }, - "3731a5747724403e86ef1abaf4fb389b": { - "model_module": "@jupyter-widgets/base", - "model_name": "LayoutModel", - "model_module_version": "1.2.0", - "state": { - "_model_module": "@jupyter-widgets/base", - "_model_module_version": "1.2.0", - "_model_name": "LayoutModel", - "_view_count": null, - "_view_module": "@jupyter-widgets/base", - "_view_module_version": "1.2.0", - "_view_name": "LayoutView", - "align_content": null, - "align_items": null, - "align_self": null, - "border": null, - "bottom": null, - "display": null, - "flex": null, - "flex_flow": null, - "grid_area": null, - "grid_auto_columns": null, - "grid_auto_flow": null, - "grid_auto_rows": null, - "grid_column": null, - "grid_gap": null, - "grid_row": null, - "grid_template_areas": null, - "grid_template_columns": null, - "grid_template_rows": null, - "height": null, - "justify_content": null, - "justify_items": null, - "left": null, - "margin": null, - "max_height": null, - "max_width": null, - "min_height": null, - "min_width": null, - "object_fit": null, - "object_position": null, - "order": null, - "overflow": null, - "overflow_x": null, - "overflow_y": null, - "padding": null, - "right": null, - "top": null, - "visibility": null, - "width": null - } + "id": "0WcVUSP-0CJi", + "outputId": "7fd9d6ac-152d-4431-9a4d-0b60321638eb" + }, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "Sequential(\n", + " (layer1): Linear(in_features=5, out_features=10, bias=True)\n", + " (layer2): Linear(in_features=10, out_features=2, bias=True)\n", + ")\n" + ] + } + ], + "source": [ + "print(model)" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/" }, - "688bf676958a4050b73cc9597f5110de": { - "model_module": "@jupyter-widgets/controls", - "model_name": "ProgressStyleModel", - "model_module_version": "1.5.0", - "state": { - "_model_module": "@jupyter-widgets/controls", - "_model_module_version": "1.5.0", - "_model_name": "ProgressStyleModel", - "_view_count": null, - "_view_module": "@jupyter-widgets/base", - "_view_module_version": "1.2.0", - "_view_name": "StyleView", - "bar_color": null, - "description_width": "" - } + "id": "akWr-cNqy-9O", + "outputId": "68b9e071-c609-4b70-906d-2f2cd93b4495" + }, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "tensor([[ 0.0458, 0.5267, 0.7119, 0.4046, 0.2460, 0.7998, 0.4485, -0.2506,\n", + " 0.2968, -0.8834]])\n" + ] + } + ], + "source": [ + "with model.trace(input) as tracer:\n", + "\n", + " l1_output = model.layer1.output.save()\n", + "\n", + "print(l1_output.value)" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "85A-aP_03ht6" + }, + "source": [ + "Let's do the same for the input of layer2. While we're at it, let's also drop the `as tracer`, as we won't be needing the tracer object itself for a few sections:" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/" + }, + "id": "3EHEN38N3nXR", + "outputId": "5afd17a7-0bcc-4d1a-c131-7038bde9455d" + }, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "((tensor([[ 0.0458, 0.5267, 0.7119, 0.4046, 0.2460, 0.7998, 0.4485, -0.2506,\n", + " 0.2968, -0.8834]]),), {})\n" + ] + } + ], + "source": [ + "with model.trace(input):\n", + "\n", + " l2_input = model.layer2.input.save()\n", + "\n", + "print(l2_input.value)" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "jk-U8zi33Gi-" + }, + "source": [ + "<details>\n", + " <summary>On module inputs</summary>\n", + "\n", + " ---\n", + "\n", + " Notice how the value for `l2_input`, was not just a single tensor.\n", + " The type/shape of values from `.input` is in the form of:\n", + "\n", + " tuple(tuple(args), dictionary(kwargs))\n", + "\n", + " Where the first index of the tuple is itself a tuple of all positional arguments, and the second index is a dictionary of the keyword arguments.\n", + "\n", + " ---\n", + "\n", + "</details>\n", + "\n", + "\n", + "Now that we can access activations, we also want to do some post-processing on it. Let's find out which dimension of layer1's output has the highest value." + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "p7Xo_PHyPr4p" + }, + "source": [ + "## Functions, Methods, and Operations" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "ehSofWbx5DSx" + }, + "source": [ + "We could do this by calling `torch.argmax(...)` after the tracing context or we can just leverage the fact that `nnsight` handles functions and methods within the tracing context, by creating a Proxy request for it:" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/" }, - "8776d04bfd3748b69e10708466e9c132": { - "model_module": "@jupyter-widgets/base", - "model_name": "LayoutModel", - "model_module_version": "1.2.0", - "state": { - "_model_module": "@jupyter-widgets/base", - "_model_module_version": "1.2.0", - "_model_name": "LayoutModel", - "_view_count": null, - "_view_module": "@jupyter-widgets/base", - "_view_module_version": "1.2.0", - "_view_name": "LayoutView", - "align_content": null, - "align_items": null, - "align_self": null, - "border": null, - "bottom": null, - "display": null, - "flex": null, - "flex_flow": null, - "grid_area": null, - "grid_auto_columns": null, - "grid_auto_flow": null, - "grid_auto_rows": null, - "grid_column": null, - "grid_gap": null, - "grid_row": null, - "grid_template_areas": null, - "grid_template_columns": null, - "grid_template_rows": null, - "height": null, - "justify_content": null, - "justify_items": null, - "left": null, - "margin": null, - "max_height": null, - "max_width": null, - "min_height": null, - "min_width": null, - "object_fit": null, - "object_position": null, - "order": null, - "overflow": null, - "overflow_x": null, - "overflow_y": null, - "padding": null, - "right": null, - "top": null, - "visibility": null, - "width": null - } + "id": "c5XCiSZn2p3k", + "outputId": "97d34dee-7bb5-4735-8c53-6aac6377af9e" + }, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "tensor(5)\n" + ] + } + ], + "source": [ + "with model.trace(input):\n", + "\n", + " # Note we don't need to call .save() on the output,\n", + " # as we're only using its value within the tracing context.\n", + " l1_output = model.layer1.output\n", + "\n", + " l1_amax = torch.argmax(l1_output, dim=1).save()\n", + "\n", + "print(l1_amax[0])" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "BGPUJvWq_bOp" + }, + "source": [ + "Nice! That worked seamlessly, but hold on, how come we didn't need to call `.value[0]` on the result? In previous sections, we were just being explicit to get an understanding of Proxies and their value. In practice, however, `nnsight` knows that when outside of the tracing context we only care about the actual value, and so printing, indexing, and applying functions all immediately return and reflect the data in `.value`. So for the rest of the tutorial we won't use it.\n", + "\n", + "The same principles work for methods and operations as well:" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/" }, - "56490c20aef04fe4a63ba7c990e71e77": { - "model_module": "@jupyter-widgets/controls", - "model_name": "DescriptionStyleModel", - "model_module_version": "1.5.0", - "state": { - "_model_module": "@jupyter-widgets/controls", - "_model_module_version": "1.5.0", - "_model_name": "DescriptionStyleModel", - "_view_count": null, - "_view_module": "@jupyter-widgets/base", - "_view_module_version": "1.2.0", - "_view_name": "StyleView", - "description_width": "" - } + "id": "FIcOYeEjFuln", + "outputId": "d836abd3-b713-44c5-aedc-95bf85abf11a" + }, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "tensor(2.3416)\n" + ] + } + ], + "source": [ + "with model.trace(input):\n", + "\n", + " value = (model.layer1.output.sum() + model.layer2.output.sum()).save()\n", + "\n", + "print(value)" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "V0vNOJtJ2oFR" + }, + "source": [ + "By default, torch functiona, methods and all operators work with `nnsight`. We also enable the use of the `einops` library.\n", + "\n", + "So to recap, the above code block is saying to `nnsight`, \"Run the model with the given `input`. When the output of layer1 is computed, take its sum. Then do the same for layer2. Now that both of those are computed, add them and make sure not to delete this value as I wish to use it outside of the tracing context.\"\n", + "\n", + "Getting and analyzing the activations from various points in a model can be really insightful, and a number of ML techniques do exactly that. However, often times we not only want to view the computation of a model, but influence it as well." + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "m_5qH5gHPOT_" + }, + "source": [ + "## Setting" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "rgju-b_IOLlq" + }, + "source": [ + "To demonstrate the effect of editing the flow of information through the model, let's set the first dimension of the first layer's output to 0. `NNsight` makes this really easy using '=' operator:" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/" }, - "b8ad433cc3dd4a8689cb06655158d545": { - "model_module": "@jupyter-widgets/controls", - "model_name": "HBoxModel", - "model_module_version": "1.5.0", - "state": { - "_dom_classes": [], - "_model_module": "@jupyter-widgets/controls", - "_model_module_version": "1.5.0", - "_model_name": "HBoxModel", - "_view_count": null, - "_view_module": "@jupyter-widgets/controls", - "_view_module_version": "1.5.0", - "_view_name": "HBoxView", - "box_style": "", - "children": [ - "IPY_MODEL_e51620404d244a4e806eeaf3b96a474f", - "IPY_MODEL_e4568f72971749ac92cc5577cb9c4f9c", - "IPY_MODEL_1a1ada8b122d4502949c4c75861a82bf" - ], - "layout": "IPY_MODEL_18d4c2302f424bc18f9215f2d5b83393" - } + "id": "-6y2wzJqOz3a", + "outputId": "6274e5c5-c5fe-4e07-87c4-a119cfd09742" + }, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "Before: tensor([[ 0.0458, 0.5267, 0.7119, 0.4046, 0.2460, 0.7998, 0.4485, -0.2506,\n", + " 0.2968, -0.8834]])\n", + "After: tensor([[ 0.0000, 0.5267, 0.7119, 0.4046, 0.2460, 0.7998, 0.4485, -0.2506,\n", + " 0.2968, -0.8834]])\n" + ] + } + ], + "source": [ + "with model.trace(input):\n", + "\n", + " # Save the output before the edit to compare.\n", + " # Notice we apply .clone() before saving as the setting operation is in-place.\n", + " l1_output_before = model.layer1.output.clone().save()\n", + "\n", + " # Access the 0th index of the hidden state dimension and set it to 0.\n", + " model.layer1.output[:, 0] = 0\n", + "\n", + " # Save the output after to see our edit.\n", + " l1_output_after = model.layer1.output.save()\n", + "\n", + "print(\"Before:\", l1_output_before)\n", + "print(\"After:\", l1_output_after)" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "oZz9SMs3Y_iS" + }, + "source": [ + "Seems our change was reflected. Now the same for the last dimension:" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/", + "height": 363 }, - "e51620404d244a4e806eeaf3b96a474f": { - "model_module": "@jupyter-widgets/controls", - "model_name": "HTMLModel", - "model_module_version": "1.5.0", - "state": { - "_dom_classes": [], - "_model_module": "@jupyter-widgets/controls", - "_model_module_version": "1.5.0", - "_model_name": "HTMLModel", - "_view_count": null, - "_view_module": "@jupyter-widgets/controls", - "_view_module_version": "1.5.0", - "_view_name": "HTMLView", - "description": "", - "description_tooltip": null, - "layout": "IPY_MODEL_f28f557210b7402b8b2d0fd83f038dc1", - "placeholder": "", - "style": "IPY_MODEL_bc8190c700e949ebb6f874c8f8055161", - "value": "merges.txt: 100%" - } + "id": "qwlqvHFcxld2", + "outputId": "28ae7175-5af0-4bd0-f9f0-21680a4fd748" + }, + "outputs": [ + { + "ename": "IndexError", + "evalue": "index 10 is out of bounds for dimension 1 with size 10", + "output_type": "error", + "traceback": [ + "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m", + "\u001b[0;31mIndexError\u001b[0m Traceback (most recent call last)", + "\u001b[0;32m<ipython-input-42-a1e18ebd4137>\u001b[0m in \u001b[0;36m<cell line: 1>\u001b[0;34m()\u001b[0m\n\u001b[0;32m----> 1\u001b[0;31m \u001b[0;32mwith\u001b[0m \u001b[0mmodel\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mtrace\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0minput\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m 2\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 3\u001b[0m \u001b[0;31m# Save the output before the edit to compare.\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 4\u001b[0m \u001b[0;31m# Notice we apply .clone() before saving as the setting operation is in-place.\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 5\u001b[0m \u001b[0ml1_output_before\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mmodel\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mlayer1\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0moutput\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mclone\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0msave\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n", + "\u001b[0;32m/usr/local/lib/python3.10/dist-packages/nnsight/contexts/Runner.py\u001b[0m in \u001b[0;36m__exit__\u001b[0;34m(self, exc_type, exc_val, exc_tb)\u001b[0m\n\u001b[1;32m 40\u001b[0m \u001b[0;32mraise\u001b[0m \u001b[0mexc_val\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 41\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m---> 42\u001b[0;31m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_graph\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mtracing\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0;32mFalse\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m 43\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 44\u001b[0m \u001b[0;32mif\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mremote\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n", + "\u001b[0;32m<ipython-input-42-a1e18ebd4137>\u001b[0m in \u001b[0;36m<cell line: 1>\u001b[0;34m()\u001b[0m\n\u001b[1;32m 6\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 7\u001b[0m \u001b[0;31m# Access the last index of the hidden state dimension and set it to 0.\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m----> 8\u001b[0;31m \u001b[0mmodel\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mlayer1\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0moutput\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mhidden_dims\u001b[0m\u001b[0;34m]\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0;36m0\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m 9\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 10\u001b[0m \u001b[0;31m# Save the output after to see our edit.\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n", + "\u001b[0;32m/usr/local/lib/python3.10/dist-packages/nnsight/tracing/Proxy.py\u001b[0m in \u001b[0;36m__setitem__\u001b[0;34m(self, key, value)\u001b[0m\n\u001b[1;32m 90\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 91\u001b[0m \u001b[0;32mdef\u001b[0m \u001b[0m__setitem__\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mkey\u001b[0m\u001b[0;34m:\u001b[0m \u001b[0mUnion\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0mProxy\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mAny\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mvalue\u001b[0m\u001b[0;34m:\u001b[0m \u001b[0mUnion\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0mSelf\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mAny\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m)\u001b[0m \u001b[0;34m->\u001b[0m \u001b[0;32mNone\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m---> 92\u001b[0;31m self.node.graph.add(\n\u001b[0m\u001b[1;32m 93\u001b[0m \u001b[0mtarget\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0moperator\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0msetitem\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 94\u001b[0m \u001b[0margs\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mnode\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mkey\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mvalue\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n", + "\u001b[0;32m/usr/local/lib/python3.10/dist-packages/nnsight/tracing/Graph.py\u001b[0m in \u001b[0;36madd\u001b[0;34m(self, target, value, args, kwargs, name)\u001b[0m\n\u001b[1;32m 144\u001b[0m \u001b[0;32mtry\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 145\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 146\u001b[0;31m value = target(\n\u001b[0m\u001b[1;32m 147\u001b[0m \u001b[0;34m\u001b[0m\u001b[0mNode\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mprepare_proxy_values\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0m_args\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 148\u001b[0m \u001b[0;34m\u001b[0m\u001b[0mNode\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mprepare_proxy_values\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0m_kwargs\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n", + "\u001b[0;31mIndexError\u001b[0m: index 10 is out of bounds for dimension 1 with size 10" + ] + } + ], + "source": [ + "with model.trace(input):\n", + "\n", + " # Save the output before the edit to compare.\n", + " # Notice we apply .clone() before saving as the setting operation is in-place.\n", + " l1_output_before = model.layer1.output.clone().save()\n", + "\n", + " # Access the last index of the hidden state dimension and set it to 0.\n", + " model.layer1.output[:, hidden_dims] = 0\n", + "\n", + " # Save the output after to see our edit.\n", + " l1_output_after = model.layer1.output.save()\n", + "\n", + "print(\"Before:\", l1_output_before)\n", + "print(\"After:\", l1_output_after)" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "JCNR_Jkpxr8l" + }, + "source": [ + "Ah of course, we needed to index at `hidden_dims - 1` not `hidden_dims`. How did `nnsight` know there was this indexing error before leaving the tracing context?\n", + "\n", + "Earlier when discussing contexts in Python, we learned some logic happens upon entering, and some logic happens upon exiting. We know the model is actually run on exit, but what happens on enter? Our input IS actually run though the model, however under its own \"fake\" context. This means the input makes its way through all of the model operations, allowing `nnsight` to record the shapes and data types of module inputs and outputs! The operations are never executed using tensors with real values so it doesn't incur any memory costs. Then, when creating proxy requests like the setting one above, `nnsight` also attempts to execute the request on the \"fake\" values we recorded. Hence, it lets us know if our request is feasible before even running the model.\n", + "\n", + "<details>\n", + "<summary>On scanning</summary>\n", + "\n", + "---\n", + "\n", + "\"Scanning\" is what we call running \"fake\" inputs throught the model to collect information like shapes and types. \"Validating\" is what we call trying to execute your intervention proxies with \"fake\" inputs to see if they work. If you are doing anything in a loop where efficiency is important, you should turn off scanning and validating. You can turn off validating in `.trace(...)` like `.trace(..., validate=False)`. You can turn off scanning in `Tracer.invoke(...)` ([see the Batching section](#batching-id)) like `Tracer.invoke(..., scan=False)`\n", + "\n", + "---\n", + "\n", + "</details>\n", + "\n", + "Let's try again with the correct indexing, and view the shape of the output before leaving the tracing context:" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/" }, - "e4568f72971749ac92cc5577cb9c4f9c": { - "model_module": "@jupyter-widgets/controls", - "model_name": "FloatProgressModel", - "model_module_version": "1.5.0", - "state": { - "_dom_classes": [], - "_model_module": "@jupyter-widgets/controls", - "_model_module_version": "1.5.0", - "_model_name": "FloatProgressModel", - "_view_count": null, - "_view_module": "@jupyter-widgets/controls", - "_view_module_version": "1.5.0", - "_view_name": "ProgressView", - "bar_style": "success", - "description": "", - "description_tooltip": null, - "layout": "IPY_MODEL_ceebda1d07b545f5b605551c08e73be8", - "max": 456318, - "min": 0, - "orientation": "horizontal", - "style": "IPY_MODEL_6aa96b80e3b746b48f5c13bcafe1134e", - "value": 456318 - } + "id": "Wf8ugJKB2mru", + "outputId": "4eed6c19-cb8c-4deb-e0e1-25aa80959c64" + }, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "layer1 output shape: torch.Size([1, 10])\n", + "Before: tensor([[ 0.0458, 0.5267, 0.7119, 0.4046, 0.2460, 0.7998, 0.4485, -0.2506,\n", + " 0.2968, -0.8834]])\n", + "After: tensor([[ 0.0458, 0.5267, 0.7119, 0.4046, 0.2460, 0.7998, 0.4485, -0.2506,\n", + " 0.2968, 0.0000]])\n" + ] + } + ], + "source": [ + "with model.trace(input):\n", + "\n", + " # Save the output before the edit to compare.\n", + " # Notice we apply .clone() before saving as the setting operation is in-place.\n", + " l1_output_before = model.layer1.output.clone().save()\n", + "\n", + " print(f\"layer1 output shape: {model.layer1.output.shape}\")\n", + "\n", + " # Access the last index of the hidden state dimension and set it to 0.\n", + " model.layer1.output[:, hidden_dims - 1] = 0\n", + "\n", + " # Save the output after to see our edit.\n", + " l1_output_after = model.layer1.output.save()\n", + "\n", + "print(\"Before:\", l1_output_before)\n", + "print(\"After:\", l1_output_after)" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "9DH7PeL13-WU" + }, + "source": [ + "We can also just replace proxy inputs and outputs with tensors of the same shape and type. Let's use the shape information we have at our disposal to add noise to the output, and replace it with this new noised tensor:" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/" }, - "1a1ada8b122d4502949c4c75861a82bf": { - "model_module": "@jupyter-widgets/controls", - "model_name": "HTMLModel", - "model_module_version": "1.5.0", - "state": { - "_dom_classes": [], - "_model_module": "@jupyter-widgets/controls", - "_model_module_version": "1.5.0", - "_model_name": "HTMLModel", - "_view_count": null, - "_view_module": "@jupyter-widgets/controls", - "_view_module_version": "1.5.0", - "_view_name": "HTMLView", - "description": "", - "description_tooltip": null, - "layout": "IPY_MODEL_8de63414803f41c29e75242756cfd67f", - "placeholder": "", - "style": "IPY_MODEL_f693dfed91df492c986afbd4ff925f89", - "value": " 456k/456k [00:00<00:00, 1.87MB/s]" - } + "id": "AP5wRmz_4YE7", + "outputId": "4c30dc1d-232d-4900-bfdd-0c93a9b23ad7" + }, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "Before: tensor([[ 0.0458, 0.5267, 0.7119, 0.4046, 0.2460, 0.7998, 0.4485, -0.2506,\n", + " 0.2968, -0.8834]])\n", + "After: tensor([[ 0.0581, 0.5168, 0.6561, 0.4083, 0.2617, 0.7800, 0.4080, -0.2213,\n", + " 0.3394, -0.9187]])\n" + ] + } + ], + "source": [ + "with model.trace(input):\n", + "\n", + " # Save the output before the edit to compare.\n", + " # Notice we apply .clone() before saving as the setting operation is in-place.\n", + " l1_output_before = model.layer1.output.clone().save()\n", + "\n", + " # Create random noise with variance of .001\n", + " noise = (0.0010.5)torch.randn(l1_output_before.shape)\n", + "\n", + " # Add to original value and replace.\n", + " model.layer1.output = l1_output_before + noise\n", + "\n", + " # Save the output after to see our edit.\n", + " l1_output_after = model.layer1.output.save()\n", + "\n", + "print(\"Before:\", l1_output_before)\n", + "print(\"After:\", l1_output_after)" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "s016CelFP8sx" + }, + "source": [ + "## Gradients\n", + "\n", + "`NNsight` can also let you apply backprop and access gradients with respect to a loss. Like `.input` and `.output` on modules, `nnsight` also exposes `.grad` on Proxies themselves (assuming they are proxies of tensors):" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/" }, - "18d4c2302f424bc18f9215f2d5b83393": { - "model_module": "@jupyter-widgets/base", - "model_name": "LayoutModel", - "model_module_version": "1.2.0", - "state": { - "_model_module": "@jupyter-widgets/base", - "_model_module_version": "1.2.0", - "_model_name": "LayoutModel", - "_view_count": null, - "_view_module": "@jupyter-widgets/base", - "_view_module_version": "1.2.0", - "_view_name": "LayoutView", - "align_content": null, - "align_items": null, - "align_self": null, - "border": null, - "bottom": null, - "display": null, - "flex": null, - "flex_flow": null, - "grid_area": null, - "grid_auto_columns": null, - "grid_auto_flow": null, - "grid_auto_rows": null, - "grid_column": null, - "grid_gap": null, - "grid_row": null, - "grid_template_areas": null, - "grid_template_columns": null, - "grid_template_rows": null, - "height": null, - "justify_content": null, - "justify_items": null, - "left": null, - "margin": null, - "max_height": null, - "max_width": null, - "min_height": null, - "min_width": null, - "object_fit": null, - "object_position": null, - "order": null, - "overflow": null, - "overflow_x": null, - "overflow_y": null, - "padding": null, - "right": null, - "top": null, - "visibility": null, - "width": null - } + "id": "L-cYUNNyRDn1", + "outputId": "d03dafa1-ef1e-40f1-de28-4dc2f77fe4db" + }, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "Layer 1 output gradient: tensor([[ 0.4545, -0.0596, -0.2059, 0.4643, -0.4211, -0.2813, 0.2126, 0.5016,\n", + " -0.0126, -0.1564]])\n", + "Layer 2 output gradient: tensor([[1., 1.]])\n" + ] + } + ], + "source": [ + "with model.trace(input):\n", + "\n", + " # We need to explicitly have the tensor require grad\n", + " # as the model we defined earlier turned off requiring grad.\n", + " model.layer1.output.requires_grad = True\n", + "\n", + " # We call .grad on a tensor Proxy to communicate we want to store its gradient.\n", + " # We need to call .save() of course as .grad is its own Proxy.\n", + " layer1_output_grad = model.layer1.output.grad.save()\n", + " layer2_output_grad = model.layer2.output.grad.save()\n", + "\n", + " # Need a loss to propagate through the later modules in order to have a grad.\n", + " loss = model.output.sum()\n", + " loss.backward()\n", + "\n", + "print(\"Layer 1 output gradient:\", layer1_output_grad)\n", + "print(\"Layer 2 output gradient:\", layer2_output_grad)\n" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "1o0JaaYvWlHG" + }, + "source": [ + "All of the features we learned previously, also apply to `.grad`. In other words, we can apply operations to and edit the gradients. Let's zero the grad of `layer1` and double the grad of `layer2`." + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/" }, - "f28f557210b7402b8b2d0fd83f038dc1": { - "model_module": "@jupyter-widgets/base", - "model_name": "LayoutModel", - "model_module_version": "1.2.0", - "state": { - "_model_module": "@jupyter-widgets/base", - "_model_module_version": "1.2.0", - "_model_name": "LayoutModel", - "_view_count": null, - "_view_module": "@jupyter-widgets/base", - "_view_module_version": "1.2.0", - "_view_name": "LayoutView", - "align_content": null, - "align_items": null, - "align_self": null, - "border": null, - "bottom": null, - "display": null, - "flex": null, - "flex_flow": null, - "grid_area": null, - "grid_auto_columns": null, - "grid_auto_flow": null, - "grid_auto_rows": null, - "grid_column": null, - "grid_gap": null, - "grid_row": null, - "grid_template_areas": null, - "grid_template_columns": null, - "grid_template_rows": null, - "height": null, - "justify_content": null, - "justify_items": null, - "left": null, - "margin": null, - "max_height": null, - "max_width": null, - "min_height": null, - "min_width": null, - "object_fit": null, - "object_position": null, - "order": null, - "overflow": null, - "overflow_x": null, - "overflow_y": null, - "padding": null, - "right": null, - "top": null, - "visibility": null, - "width": null - } + "id": "9bFoaJpOWlRb", + "outputId": "1c026534-9ffe-40e7-d016-e4565ddde8ac" + }, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "Layer 1 output gradient: tensor([[0., 0., 0., 0., 0., 0., 0., 0., 0., 0.]])\n", + "Layer 2 output gradient: tensor([[2., 2.]])\n" + ] + } + ], + "source": [ + "with model.trace(input):\n", + "\n", + " # We need to explicitly have the tensor require grad\n", + " # as the model we defined earlier turned off requiring grad.\n", + " model.layer1.output.requires_grad = True\n", + "\n", + " model.layer1.output.grad[:] = 0\n", + " model.layer2.output.grad = model.layer2.output.grad.clone() * 2\n", + "\n", + " layer1_output_grad = model.layer1.output.grad.save()\n", + " layer2_output_grad = model.layer2.output.grad.save()\n", + "\n", + " # Need a loss to propagate through the later modules in order to have a grad.\n", + " loss = model.output.sum()\n", + " loss.backward()\n", + "\n", + "print(\"Layer 1 output gradient:\", layer1_output_grad)\n", + "print(\"Layer 2 output gradient:\", layer2_output_grad)" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "gvQ1nZgYQDG3" + }, + "source": [ + "# 2 Bigger" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "miQgUY4NAmDQ" + }, + "source": [ + "Now that we have the basics of `nnsight` under our belt, we can scale our model up and combine the techniques we've learned into more interesting experiments.\n", + "\n", + "The `NNsight` class is very bare bones. It wraps a pre-defined model and does no pre-processing on the inputs we enter. It's designed to be extended with more complex and powerful types of models and we're excited to see what can be done to leverage its features." + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "5TJDblHiQpp1" + }, + "source": [ + "## LanguageModel" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "9l9mZOY5HFH2" + }, + "source": [ + "`LanguageModel` is a subclass of `NNsight`. While we could define and create a model to pass in directly, `LanguageModel` includes special support for Huggingface language models, including automatically loading models from a Huggingface ID, and loading the model together with the appropriate tokenizer.\n", + "\n", + "Here is how you can use `LanguageModel` to load `GPT-2`:" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/", + "height": 769, + "referenced_widgets": [ + "0a676b8f2dc34fed9731f19a03e4bfa1", + "74358c341daa44faa41a92b0eb073615", + "c40e17a994ed44d79ef54421f4772858", + "431ffcb0dbb1419dbbb237a29842b905", + "c28c185a508e459b95edb29afedd9c79", + "7a0f3d2deeb14bbba7f0df208133a7e2", + "94cdcf1fb0914dfe99c3c615e68b7ec8", + "f3e24aa6619e4185a3e2f53f2e473e45", + "b7be20bd738940658f710309729d8ccd", + "562a287082eb4882af864062ae13c91b", + "11e99924bdf84739b3cfc7d676b797e5", + "60fe026e908741209637604edfeb0563", + "e7a73cfc3fa04425afc08368532ef0be", + "d839da1c31a34659988e3420ba520725", + "96a3b8c765374a5e98eaf713dd5e85d8", + "b49ce0d00e5b4df097b3a9fc4c0bd8e7", + "b2748df9aff540df954f002310606a56", + "f4caf2bc58704464b6ac7bf6da31b71d", + "3731a5747724403e86ef1abaf4fb389b", + "688bf676958a4050b73cc9597f5110de", + "8776d04bfd3748b69e10708466e9c132", + "56490c20aef04fe4a63ba7c990e71e77", + "b8ad433cc3dd4a8689cb06655158d545", + "e51620404d244a4e806eeaf3b96a474f", + "e4568f72971749ac92cc5577cb9c4f9c", + "1a1ada8b122d4502949c4c75861a82bf", + "18d4c2302f424bc18f9215f2d5b83393", + "f28f557210b7402b8b2d0fd83f038dc1", + "bc8190c700e949ebb6f874c8f8055161", + "ceebda1d07b545f5b605551c08e73be8", + "6aa96b80e3b746b48f5c13bcafe1134e", + "8de63414803f41c29e75242756cfd67f", + "f693dfed91df492c986afbd4ff925f89", + "cef2516aabc24f6d852079984afa8090", + "95d303d83d3348a0adfdf6b5cb94c41c", + "9373fcd054014760b23e3acc928da0a5", + "7e6b41ec535648b190bfc655d9e430dd", + "d0198fa98e5f412380f2ba3e9448aef3", + "eb0c489b913744a28304e5334bb66532", + "c80603555c634e338f16c0021f752420", + "0fddebfb42e847e08890d09f30161984", + "d07a0a49a2924092affc0e07bf0b5ef0", + "beb4231b49024d8a9bb7ead19647d99d", + "26d09aa072d64af0bf4df9a89309093a" + ] }, - "bc8190c700e949ebb6f874c8f8055161": { - "model_module": "@jupyter-widgets/controls", - "model_name": "DescriptionStyleModel", - "model_module_version": "1.5.0", - "state": { - "_model_module": "@jupyter-widgets/controls", - "_model_module_version": "1.5.0", - "_model_name": "DescriptionStyleModel", - "_view_count": null, - "_view_module": "@jupyter-widgets/base", - "_view_module_version": "1.2.0", - "_view_name": "StyleView", - "description_width": "" - } + "id": "1OD2z7d3HQJU", + "outputId": "7a53864c-6965-4c25-9668-4b0d56db9f6b" + }, + "outputs": [ + { + "name": "stderr", + "output_type": "stream", + "text": [ + "/usr/local/lib/python3.10/dist-packages/huggingface_hub/utils/_token.py:88: UserWarning: \n", + "The secret `HF_TOKEN` does not exist in your Colab secrets.\n", + "To authenticate with the Hugging Face Hub, create a token in your settings tab (https://huggingface.co/settings/tokens), set it as secret in your Google Colab and restart your session.\n", + "You will be able to reuse this secret in all of your notebooks.\n", + "Please note that authentication is recommended but still optional to access public models or datasets.\n", + " warnings.warn(\n" + ] }, - "ceebda1d07b545f5b605551c08e73be8": { - "model_module": "@jupyter-widgets/base", - "model_name": "LayoutModel", - "model_module_version": "1.2.0", - "state": { - "_model_module": "@jupyter-widgets/base", - "_model_module_version": "1.2.0", - "_model_name": "LayoutModel", - "_view_count": null, - "_view_module": "@jupyter-widgets/base", - "_view_module_version": "1.2.0", - "_view_name": "LayoutView", - "align_content": null, - "align_items": null, - "align_self": null, - "border": null, - "bottom": null, - "display": null, - "flex": null, - "flex_flow": null, - "grid_area": null, - "grid_auto_columns": null, - "grid_auto_flow": null, - "grid_auto_rows": null, - "grid_column": null, - "grid_gap": null, - "grid_row": null, - "grid_template_areas": null, - "grid_template_columns": null, - "grid_template_rows": null, - "height": null, - "justify_content": null, - "justify_items": null, - "left": null, - "margin": null, - "max_height": null, - "max_width": null, - "min_height": null, - "min_width": null, - "object_fit": null, - "object_position": null, - "order": null, - "overflow": null, - "overflow_x": null, - "overflow_y": null, - "padding": null, - "right": null, - "top": null, - "visibility": null, - "width": null - } + { + "data": { + "application/vnd.jupyter.widget-view+json": { + "model_id": "0a676b8f2dc34fed9731f19a03e4bfa1", + "version_major": 2, + "version_minor": 0 + }, + "text/plain": [ + "config.json: 0%\| \| 0.00/665 [00:00<?, ?B/s]" + ] + }, + "metadata": {}, + "output_type": "display_data" }, - "6aa96b80e3b746b48f5c13bcafe1134e": { - "model_module": "@jupyter-widgets/controls", - "model_name": "ProgressStyleModel", - "model_module_version": "1.5.0", - "state": { - "_model_module": "@jupyter-widgets/controls", - "_model_module_version": "1.5.0", - "_model_name": "ProgressStyleModel", - "_view_count": null, - "_view_module": "@jupyter-widgets/base", - "_view_module_version": "1.2.0", - "_view_name": "StyleView", - "bar_color": null, - "description_width": "" - } + { + "data": { + "application/vnd.jupyter.widget-view+json": { + "model_id": "60fe026e908741209637604edfeb0563", + "version_major": 2, + "version_minor": 0 + }, + "text/plain": [ + "vocab.json: 0%\| \| 0.00/1.04M [00:00<?, ?B/s]" + ] + }, + "metadata": {}, + "output_type": "display_data" }, - "8de63414803f41c29e75242756cfd67f": { - "model_module": "@jupyter-widgets/base", - "model_name": "LayoutModel", - "model_module_version": "1.2.0", - "state": { - "_model_module": "@jupyter-widgets/base", - "_model_module_version": "1.2.0", - "_model_name": "LayoutModel", - "_view_count": null, - "_view_module": "@jupyter-widgets/base", - "_view_module_version": "1.2.0", - "_view_name": "LayoutView", - "align_content": null, - "align_items": null, - "align_self": null, - "border": null, - "bottom": null, - "display": null, - "flex": null, - "flex_flow": null, - "grid_area": null, - "grid_auto_columns": null, - "grid_auto_flow": null, - "grid_auto_rows": null, - "grid_column": null, - "grid_gap": null, - "grid_row": null, - "grid_template_areas": null, - "grid_template_columns": null, - "grid_template_rows": null, - "height": null, - "justify_content": null, - "justify_items": null, - "left": null, - "margin": null, - "max_height": null, - "max_width": null, - "min_height": null, - "min_width": null, - "object_fit": null, - "object_position": null, - "order": null, - "overflow": null, - "overflow_x": null, - "overflow_y": null, - "padding": null, - "right": null, - "top": null, - "visibility": null, - "width": null - } + { + "data": { + "application/vnd.jupyter.widget-view+json": { + "model_id": "b8ad433cc3dd4a8689cb06655158d545", + "version_major": 2, + "version_minor": 0 + }, + "text/plain": [ + "merges.txt: 0%\| \| 0.00/456k [00:00<?, ?B/s]" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "application/vnd.jupyter.widget-view+json": { + "model_id": "cef2516aabc24f6d852079984afa8090", + "version_major": 2, + "version_minor": 0 + }, + "text/plain": [ + "tokenizer.json: 0%\| \| 0.00/1.36M [00:00<?, ?B/s]" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "name": "stdout", + "output_type": "stream", + "text": [ + "GPT2LMHeadModel(\n", + " (transformer): GPT2Model(\n", + " (wte): Embedding(50257, 768)\n", + " (wpe): Embedding(1024, 768)\n", + " (drop): Dropout(p=0.1, inplace=False)\n", + " (h): ModuleList(\n", + " (0-11): 12 x GPT2Block(\n", + " (ln_1): LayerNorm((768,), eps=1e-05, elementwise_affine=True)\n", + " (attn): GPT2Attention(\n", + " (c_attn): Conv1D()\n", + " (c_proj): Conv1D()\n", + " (attn_dropout): Dropout(p=0.1, inplace=False)\n", + " (resid_dropout): Dropout(p=0.1, inplace=False)\n", + " )\n", + " (ln_2): LayerNorm((768,), eps=1e-05, elementwise_affine=True)\n", + " (mlp): GPT2MLP(\n", + " (c_fc): Conv1D()\n", + " (c_proj): Conv1D()\n", + " (act): NewGELUActivation()\n", + " (dropout): Dropout(p=0.1, inplace=False)\n", + " )\n", + " )\n", + " )\n", + " (ln_f): LayerNorm((768,), eps=1e-05, elementwise_affine=True)\n", + " )\n", + " (lm_head): Linear(in_features=768, out_features=50257, bias=False)\n", + " (generator): WrapperModule()\n", + ")\n" + ] + } + ], + "source": [ + "from nnsight import LanguageModel\n", + "\n", + "model = LanguageModel('openai-community/gpt2', device_map=\"auto\")\n", + "\n", + "print(model)" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "Hw2VaAr_ezkj" + }, + "source": [ + "<details>\n", + "<summary>On Model Initialization</summary>\n", + "\n", + "---\n", + "\n", + "A few important things to note:\n", + "\n", + " Keyword arguments passed to the initialization of `LanguageModel` is forwarded to HuggingFace specific loading logic. In this case, `device_map` specifies which devices to use and its value `auto` indicates to evenly distribute it to all available GPUs (and cpu if no GPUs available). Other arguments can be found here: https://huggingface.co/docs/transformers/model_doc/auto#transformers.AutoModelForCausalLM\n", + "\n", + "\n", + " When we initialize `LanguageModel`, we aren't yet loading the parameters of the model into memory. We are actually loading a 'meta' version of the model which doesn't take up any memory, but still allows us to view and trace actions on it. After exiting the first tracing context, the model is then fully loaded into memory. To load into memory on initialization, you can pass `dispatch=True` into `LanguageModel` like `LanguageModel('openai-community/gpt2', device_map=\"auto\", dispatch=True)`.\n", + "\n", + "---\n", + "\n", + "</details>\n", + "\n", + "\n", + "Let's put together some of the features we applied to the small model, but now on `GPT-2`. Unlike `NNsight`, `LanguageModel` does define logic to pre-process inputs upon entering the tracing context. This makes interacting with the model simpler without having to directly access the tokenizer.\n", + "\n", + "In the following example, we ablate the value coming from the last layer's MLP module and decode the logits to see what token the model predicts without influence from that particular module:" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/" }, - "f693dfed91df492c986afbd4ff925f89": { - "model_module": "@jupyter-widgets/controls", - "model_name": "DescriptionStyleModel", - "model_module_version": "1.5.0", - "state": { - "_model_module": "@jupyter-widgets/controls", - "_model_module_version": "1.5.0", - "_model_name": "DescriptionStyleModel", - "_view_count": null, - "_view_module": "@jupyter-widgets/base", - "_view_module_version": "1.2.0", - "_view_name": "StyleView", - "description_width": "" - } + "id": "mLSapaMCgLNU", + "outputId": "192b8998-ecca-4f25-ecf2-5196c0d2d672" + }, + "outputs": [ + { + "name": "stderr", + "output_type": "stream", + "text": [ + "You're using a GPT2TokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n" + ] }, - "cef2516aabc24f6d852079984afa8090": { - "model_module": "@jupyter-widgets/controls", - "model_name": "HBoxModel", - "model_module_version": "1.5.0", - "state": { - "_dom_classes": [], - "_model_module": "@jupyter-widgets/controls", - "_model_module_version": "1.5.0", - "_model_name": "HBoxModel", - "_view_count": null, - "_view_module": "@jupyter-widgets/controls", - "_view_module_version": "1.5.0", - "_view_name": "HBoxView", - "box_style": "", - "children": [ - "IPY_MODEL_95d303d83d3348a0adfdf6b5cb94c41c", - "IPY_MODEL_9373fcd054014760b23e3acc928da0a5", - "IPY_MODEL_7e6b41ec535648b190bfc655d9e430dd" - ], - "layout": "IPY_MODEL_d0198fa98e5f412380f2ba3e9448aef3" - } + { + "name": "stdout", + "output_type": "stream", + "text": [ + "Token IDs: tensor([[ 262, 12, 417, 8765, 11, 257, 262, 3504, 338, 3576]])\n", + "Prediction: London\n" + ] + } + ], + "source": [ + "with model.trace('The Eiffel Tower is in the city of'):\n", + "\n", + " # Access the last layer using h[-1] as it's a ModuleList\n", + " # Access the first index of .output as that's where the hidden states are.\n", + " model.transformer.h[-1].mlp.output[0][:] = 0\n", + "\n", + " # Logits come out of model.lm_head and we apply argmax to get the predicted token ids.\n", + " token_ids = model.lm_head.output.argmax(dim=-1).save()\n", + "\n", + "print(\"Token IDs:\", token_ids)\n", + "\n", + "# Apply the tokenizer to decode the ids into words after the tracing context.\n", + "print(\"Prediction:\", model.tokenizer.decode(token_ids[0][-1]))\n" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "X2-HiCQhkv-B" + }, + "source": [ + "You just ran a little intervention on a much more complex model with a lot more parameters! An important piece of information we're missing though is what the prediction would look like without our ablation.\n", + "\n", + "Of course we could just run two tracing contexts and compare the outputs. This, however, would require two forward passes through the model. `NNsight` can do better than that." + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "vLjHnRyRPXjp" + }, + "source": [ + "<a name=\"batching-id\"></a>\n", + "## Batching" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "_Gc-Zr6NmEOn" + }, + "source": [ + "It's time to bring back the `Tracer` object we dropped before. See, when you call `.trace(...)` with some input, it's actually creating two different contexts behind the scenes. The second one is the invoker context. Being within this context just means that `.input` and `.output` should refer only to the input you've given invoke. Calling `.trace(...)` with some input just means there's only one input and therefore only one invoker context.\n", + "\n", + "We can call `.trace()` without input and call `Tracer.invoke(...)` to manually create the invoker context with our input. Now every subsequent time we call `.invoke(...)`, new interventions will only refer to the input in that particular invoke. When exiting the tracing context, the inputs from all of the invokers will be batched together, and they will be executed in one forward pass! So let's do the ablation experiment, and compute a 'control' output to compare to:\n", + "\n", + "<details>\n", + "<summary>On the invoker context</summary>\n", + "\n", + "---\n", + "\n", + "Note that when injecting data to only the relevant invoker interventions, `nnsight` tries, but can't guarantee, that it can narrow the data into the right batch idxs (in the case of an object as input or output). So there are cases where all invokes will get all of the data.\n", + "\n", + "Just like `.trace(...)` created a `Tracer` object, `.invoke(...)` creates an `Invoker` object. The `Invoker` object has post-processed inputs at `invoker.inputs`, which can be useful for seeing information about your input. If you are using `.trace(...)` with inputs, you can still access the invoker object at `tracer._invoker`.\n", + "\n", + "Keyword arguments given to `.invoke(..)` make its way to the input pre-processing. For example in `LanguageModel`, the keyword arguments are used to tokenize like `max_length` and `truncation`. If you need to pass in keyword arguments directly to one input `.trace(...)`, you can pass an `invoker_args` keyword argument that should be a dictionary of keyword arguments for the invoker. `.trace(..., invoker_args={...})`\n", + "\n", + "---\n", + "\n", + "</details>" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/" }, - "95d303d83d3348a0adfdf6b5cb94c41c": { - "model_module": "@jupyter-widgets/controls", - "model_name": "HTMLModel", - "model_module_version": "1.5.0", - "state": { - "_dom_classes": [], - "_model_module": "@jupyter-widgets/controls", - "_model_module_version": "1.5.0", - "_model_name": "HTMLModel", - "_view_count": null, - "_view_module": "@jupyter-widgets/controls", - "_view_module_version": "1.5.0", - "_view_name": "HTMLView", - "description": "", - "description_tooltip": null, - "layout": "IPY_MODEL_eb0c489b913744a28304e5334bb66532", - "placeholder": "", - "style": "IPY_MODEL_c80603555c634e338f16c0021f752420", - "value": "tokenizer.json: 100%" - } + "id": "Kdcq4oCNmEua", + "outputId": "e6d33c07-d22a-41af-b84a-91c9681c7524" + }, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "Original token IDs: tensor([[ 198, 12, 417, 8765, 318, 257, 262, 3504, 7372, 6342]])\n", + "Intervention token IDs: tensor([[ 262, 12, 417, 8765, 11, 257, 262, 3504, 338, 3576]])\n", + "Original prediction: Paris\n", + "Intervention prediction: London\n" + ] + } + ], + "source": [ + "with model.trace() as tracer:\n", + "\n", + " with tracer.invoke('The Eiffel Tower is in the city of'):\n", + "\n", + " # Ablate the last MLP for only this batch.\n", + " model.transformer.h[-1].mlp.output[0][:] = 0\n", + "\n", + " # Get the output for only the intervened on batch.\n", + " token_ids_intervention = model.lm_head.output.argmax(dim=-1).save()\n", + "\n", + " with tracer.invoke('The Eiffel Tower is in the city of'):\n", + "\n", + " # Get the output for only the original batch.\n", + " token_ids_original = model.lm_head.output.argmax(dim=-1).save()\n", + "\n", + "print(\"Original token IDs:\", token_ids_original)\n", + "print(\"Intervention token IDs:\", token_ids_intervention)\n", + "\n", + "print(\"Original prediction:\", model.tokenizer.decode(token_ids_original[0][-1]))\n", + "print(\"Intervention prediction:\", model.tokenizer.decode(token_ids_intervention[0][-1]))" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "8BsybgsK2Bbr" + }, + "source": [ + "So it did end up affecting what the model predicted. That's pretty neat!\n", + "\n", + "Another cool thing with multiple invokes is that the Proxies can interact between them. Here we transfer the word token embeddings from a real prompt into another placeholder prompt. Therefore the latter prompt produces the output of the former prompt:" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/" + }, + "id": "syKTI_KhpvCY", + "outputId": "b2238c5c-d2d8-47bf-b73a-730f1e03cf99" + }, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "Original prediction: _\n", + "Intervention prediction: Paris\n" + ] + } + ], + "source": [ + "with model.trace() as tracer:\n", + "\n", + " with tracer.invoke(\"The Eiffel Tower is in the city of\"):\n", + "\n", + " embeddings = model.transformer.wte.output\n", + "\n", + " with tracer.invoke(\"_ _ _ _ _ _ _ _ _ _\"):\n", + "\n", + " model.transformer.wte.output = embeddings\n", + "\n", + " token_ids_intervention = model.lm_head.output.argmax(dim=-1).save()\n", + "\n", + " with tracer.invoke(\"_ _ _ _ _ _ _ _ _ _\"):\n", + "\n", + " token_ids_original = model.lm_head.output.argmax(dim=-1).save()\n", + "\n", + "print(\"Original prediction:\", model.tokenizer.decode(token_ids_original[0][-1]))\n", + "print(\"Intervention prediction:\", model.tokenizer.decode(token_ids_intervention[0][-1]))" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "vvWA-CWqQtah" + }, + "source": [ + "## .next()\n", + "\n", + "Some HuggingFace models define methods to generate multiple outputs at a time. `LanguageModel` wraps that functionality to provide the same tracing features by using `.generate(...)` instead of `.trace(...)`. This calls the underlying model's `.generate` method. It passes the output through a `model.generator` module that we've added onto the model, allowing you to get the generate output at `model.generator.output`.\n", + "\n", + "In a case like this, the underlying model is called more than once; the modules of said model produce more than one output. Which iteration should a given `module.output` refer to? That's where `Module.next()` comes in.\n", + "\n", + "Each module has a call idx associated with it and `.next()` simply increments that attribute. At the time of execution, data is injected into the intervention graph only at the iteration that matches the call idx." + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/" }, - "9373fcd054014760b23e3acc928da0a5": { - "model_module": "@jupyter-widgets/controls", - "model_name": "FloatProgressModel", - "model_module_version": "1.5.0", - "state": { - "_dom_classes": [], - "_model_module": "@jupyter-widgets/controls", - "_model_module_version": "1.5.0", - "_model_name": "FloatProgressModel", - "_view_count": null, - "_view_module": "@jupyter-widgets/controls", - "_view_module_version": "1.5.0", - "_view_name": "ProgressView", - "bar_style": "success", - "description": "", - "description_tooltip": null, - "layout": "IPY_MODEL_0fddebfb42e847e08890d09f30161984", - "max": 1355256, - "min": 0, - "orientation": "horizontal", - "style": "IPY_MODEL_d07a0a49a2924092affc0e07bf0b5ef0", - "value": 1355256 - } + "id": "Yy5Z9NE1GkaN", + "outputId": "0b4c1bd3-a6f6-4ba1-9cb8-6019c222311f" + }, + "outputs": [ + { + "name": "stderr", + "output_type": "stream", + "text": [ + "Setting `pad_token_id` to `eos_token_id`:50256 for open-end generation.\n", + "A decoder-only architecture is being used, but right-padding was detected! For correct generation results, please set `padding_side='left'` when initializing the tokenizer.\n", + "Setting `pad_token_id` to `eos_token_id`:50256 for open-end generation.\n" + ] }, - "7e6b41ec535648b190bfc655d9e430dd": { + { + "name": "stdout", + "output_type": "stream", + "text": [ + "Prediction 1: Paris\n", + "Prediction 2: ,\n", + "Prediction 3: and\n", + "All token ids: tensor([[ 464, 412, 733, 417, 8765, 318, 287, 262, 1748, 286, 6342, 11,\n", + " 290]])\n", + "All prediction: ['The Eiffel Tower is in the city of Paris, and']\n" + ] + } + ], + "source": [ + "with model.generate(\"The Eiffel Tower is in the city of\", max_new_tokens=3):\n", + "\n", + " token_ids_1 = model.lm_head.output.argmax(dim=-1).save()\n", + "\n", + " token_ids_2 = model.lm_head.next().output.argmax(dim=-1).save()\n", + "\n", + " token_ids_3 = model.lm_head.next().output.argmax(dim=-1).save()\n", + "\n", + " output = model.generator.output.save()\n", + "\n", + "print(\"Prediction 1: \", model.tokenizer.decode(token_ids_1[0][-1]))\n", + "print(\"Prediction 2: \", model.tokenizer.decode(token_ids_2[0][-1]))\n", + "print(\"Prediction 3: \", model.tokenizer.decode(token_ids_3[0][-1]))\n", + "\n", + "print(\"All token ids: \", output)\n", + "\n", + "print(\"All prediction: \", model.tokenizer.batch_decode(output))" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "rpkX-LwBQZHo" + }, + "source": [ + "# 3 I thought you said huge models?" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "uCZR65VmILEb" + }, + "source": [ + "`NNsight` is only one part of our project to democratize access to AI internals. The other half is `NDIF` (National Deep Inference Facility).\n", + "\n", + "The interaction between the two is fairly straightforward. The `intervention graph` we create via the tracing context can be encoded into a custom json format and sent via an http request to the `NDIF` servers. `NDIF` then decodes the `intervention graph` and `interleaves` it alongside the specified model.\n", + "\n", + "To see which models are currently being hosted, check out the following status page: https://nnsight.net/status/" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "65Ks1LUvQaER" + }, + "source": [ + "## Remote execution" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "Aa-ZuVOFKS5k" + }, + "source": [ + "In its current state, `NDIF` requires you to recieve an API key. Therefore, to run the rest of this colab, you would need one of your own. To get one, simply join the [NDIF discord](https://discord.gg/6uFJmCSwW7) and introduce yourself on the `#introductions` channel. Then DM either @JadenFK or @caden and we'll create one for you.\n", + "\n", + "Once you have one, to register your api key with `nnsight`, do the following:" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "id": "ax1NWoS9MJeZ" + }, + "outputs": [], + "source": [ + "from nnsight import CONFIG\n", + "\n", + "CONFIG.set_default_api_key(\"<your api key here>\")" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "mvXLOh65MXmF" + }, + "source": [ + "This only needs to be run once as it will save this api key as the default in a config file along with the `nnsight` installation.\n", + "\n", + "To amp things up a few levels, let's demonstrate using `nnsight`'s tracing context with one of the larger open source language models, `Llama-2-70b`!" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "id": "Q1N04sJPZnJt" + }, + "outputs": [], + "source": [ + "import os\n", + "\n", + "# llama2 70b is a gated model and you need access via your huggingface token\n", + "os.environ['HF_TOKEN'] = \"<your huggingface token>\"\n", + "\n", + "# llama response object requires the version of transformers from github\n", + "!pip uninstall -y transformers\n", + "!pip install git+https://github.com/huggingface/transformers\n", + "\n", + "clear_output()" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "id": "oqVzjNoyNGc7" + }, + "outputs": [], + "source": [ + "# We'll never actually load the parameters so no need to specify a device_map.\n", + "model = LanguageModel(\"meta-llama/Llama-2-70b-hf\")\n", + "\n", + "# All we need to specify using NDIF vs executing locally is remote=True.\n", + "with model.trace('The Eiffel Tower is in the city of', remote=True) as runner:\n", + "\n", + " hidden_states = model.model.layers[-1].output.save()\n", + "\n", + " output = model.output.save()\n", + "\n", + "print(hidden_states)\n", + "\n", + "print(output['logits'])" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "IceVpnZcZ9F0" + }, + "source": [ + "It really is as simple as `remote=True`. All of the techniques we went through in earlier sections work just the same when running locally and remotely.\n", + "\n", + "Note that both `nnsight`, but especially `NDIF`, is in active development and therefore there may be caveats, changes, and errors to work through." + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "P3zRm-7VRRov" + }, + "source": [ + "# Getting Involved!" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "BnCc9xgjvxEP" + }, + "source": [ + "If you're interested in following updates to `nnsight`, contributing, giving feedback, or finding collaborators, please join the [NDIF discord](https://discord.gg/6uFJmCSwW7)!\n", + "\n", + "The [Mech Interp discord](https://discord.gg/km2RQBzaUn) is also a fantastic place to discuss all things mech interp with a really cool community.\n", + "\n", + "Our website [nnsight.net](https://nnsight.net/), has a bunch more tutorials detailing more complex interpretability techniques using `nnsight`. If you want to share any of the work you do using `nnsight`, let others know on either of the discords above and we might turn it into a tutorial on our website.\n", + "\n", + "💟\n", + "\n", + "\n", + "\n" + ] + } + ], + "metadata": { + "colab": { + "include_colab_link": true, + "provenance": [], + "toc_visible": true + }, + "kernelspec": { + "display_name": "Python 3", + "name": "python3" + }, + "language_info": { + "name": "python" + }, + "widgets": { + "application/vnd.jupyter.widget-state+json": { + "0a676b8f2dc34fed9731f19a03e4bfa1": { "model_module": "@jupyter-widgets/controls", - "model_name": "HTMLModel", "model_module_version": "1.5.0", + "model_name": "HBoxModel", "state": { "_dom_classes": [], "_model_module": "@jupyter-widgets/controls", "_model_module_version": "1.5.0", - "_model_name": "HTMLModel", + "_model_name": "HBoxModel", "_view_count": null, "_view_module": "@jupyter-widgets/controls", "_view_module_version": "1.5.0", - "_view_name": "HTMLView", - "description": "", - "description_tooltip": null, - "layout": "IPY_MODEL_beb4231b49024d8a9bb7ead19647d99d", - "placeholder": "", - "style": "IPY_MODEL_26d09aa072d64af0bf4df9a89309093a", - "value": " 1.36M/1.36M [00:00<00:00, 4.16MB/s]" + "_view_name": "HBoxView", + "box_style": "", + "children": [ + "IPY_MODEL_74358c341daa44faa41a92b0eb073615", + "IPY_MODEL_c40e17a994ed44d79ef54421f4772858", + "IPY_MODEL_431ffcb0dbb1419dbbb237a29842b905" + ], + "layout": "IPY_MODEL_c28c185a508e459b95edb29afedd9c79" } }, - "d0198fa98e5f412380f2ba3e9448aef3": { + "0fddebfb42e847e08890d09f30161984": { "model_module": "@jupyter-widgets/base", - "model_name": "LayoutModel", "model_module_version": "1.2.0", - "state": { - "_model_module": "@jupyter-widgets/base", - "_model_module_version": "1.2.0", - "_model_name": "LayoutModel", - "_view_count": null, - "_view_module": "@jupyter-widgets/base", - "_view_module_version": "1.2.0", - "_view_name": "LayoutView", - "align_content": null, - "align_items": null, - "align_self": null, - "border": null, - "bottom": null, - "display": null, - "flex": null, - "flex_flow": null, - "grid_area": null, - "grid_auto_columns": null, - "grid_auto_flow": null, - "grid_auto_rows": null, - "grid_column": null, - "grid_gap": null, - "grid_row": null, - "grid_template_areas": null, - "grid_template_columns": null, - "grid_template_rows": null, - "height": null, - "justify_content": null, - "justify_items": null, - "left": null, - "margin": null, - "max_height": null, - "max_width": null, - "min_height": null, - "min_width": null, - "object_fit": null, - "object_position": null, - "order": null, - "overflow": null, - "overflow_x": null, - "overflow_y": null, - "padding": null, - "right": null, - "top": null, - "visibility": null, - "width": null - } - }, - "eb0c489b913744a28304e5334bb66532": { - "model_module": "@jupyter-widgets/base", "model_name": "LayoutModel", - "model_module_version": "1.2.0", "state": { "_model_module": "@jupyter-widgets/base", "_model_module_version": "1.2.0", @@ -1234,10 +1670,10 @@ "width": null } }, - "c80603555c634e338f16c0021f752420": { + "11e99924bdf84739b3cfc7d676b797e5": { "model_module": "@jupyter-widgets/controls", - "model_name": "DescriptionStyleModel", "model_module_version": "1.5.0", + "model_name": "DescriptionStyleModel", "state": { "_model_module": "@jupyter-widgets/controls", "_model_module_version": "1.5.0", @@ -1249,10 +1685,10 @@ "description_width": "" } }, - "0fddebfb42e847e08890d09f30161984": { + "18d4c2302f424bc18f9215f2d5b83393": { "model_module": "@jupyter-widgets/base", - "model_name": "LayoutModel", "model_module_version": "1.2.0", + "model_name": "LayoutModel", "state": { "_model_module": "@jupyter-widgets/base", "_model_module_version": "1.2.0", @@ -1301,26 +1737,31 @@ "width": null } }, - "d07a0a49a2924092affc0e07bf0b5ef0": { + "1a1ada8b122d4502949c4c75861a82bf": { "model_module": "@jupyter-widgets/controls", - "model_name": "ProgressStyleModel", "model_module_version": "1.5.0", + "model_name": "HTMLModel", "state": { + "_dom_classes": [], "_model_module": "@jupyter-widgets/controls", "_model_module_version": "1.5.0", - "_model_name": "ProgressStyleModel", + "_model_name": "HTMLModel", "_view_count": null, - "_view_module": "@jupyter-widgets/base", - "_view_module_version": "1.2.0", - "_view_name": "StyleView", - "bar_color": null, - "description_width": "" + "_view_module": "@jupyter-widgets/controls", + "_view_module_version": "1.5.0", + "_view_name": "HTMLView", + "description": "", + "description_tooltip": null, + "layout": "IPY_MODEL_8de63414803f41c29e75242756cfd67f", + "placeholder": "", + "style": "IPY_MODEL_f693dfed91df492c986afbd4ff925f89", + "value": " 456k/456k [00:00<00:00, 1.87MB/s]" } }, - "beb4231b49024d8a9bb7ead19647d99d": { + "21f5a81dd29549e69b5cfa2661371e35": { "model_module": "@jupyter-widgets/base", - "model_name": "LayoutModel", "model_module_version": "1.2.0", + "model_name": "LayoutModel", "state": { "_model_module": "@jupyter-widgets/base", "_model_module_version": "1.2.0", @@ -1366,13 +1807,13 @@ "right": null, "top": null, "visibility": null, - "width": null + "width": "20px" } }, "26d09aa072d64af0bf4df9a89309093a": { "model_module": "@jupyter-widgets/controls", - "model_name": "DescriptionStyleModel", "model_module_version": "1.5.0", + "model_name": "DescriptionStyleModel", "state": { "_model_module": "@jupyter-widgets/controls", "_model_module_version": "1.5.0", @@ -1384,98 +1825,10 @@ "description_width": "" } }, - "b3078a7a41c642839748d9864220cbde": { - "model_module": "@jupyter-widgets/controls", - "model_name": "HBoxModel", - "model_module_version": "1.5.0", - "state": { - "_dom_classes": [], - "_model_module": "@jupyter-widgets/controls", - "_model_module_version": "1.5.0", - "_model_name": "HBoxModel", - "_view_count": null, - "_view_module": "@jupyter-widgets/controls", - "_view_module_version": "1.5.0", - "_view_name": "HBoxView", - "box_style": "", - "children": [ - "IPY_MODEL_3aa1aef302474a0cba55152e311acd40", - "IPY_MODEL_dda1704b991a44dfb2b28191f307e081", - "IPY_MODEL_5c0847c658d3489ca529d1afc8c96ed5" - ], - "layout": "IPY_MODEL_9fbbb0bf7d8e4c12939193791efc590f" - } - }, - "3aa1aef302474a0cba55152e311acd40": { - "model_module": "@jupyter-widgets/controls", - "model_name": "HTMLModel", - "model_module_version": "1.5.0", - "state": { - "_dom_classes": [], - "_model_module": "@jupyter-widgets/controls", - "_model_module_version": "1.5.0", - "_model_name": "HTMLModel", - "_view_count": null, - "_view_module": "@jupyter-widgets/controls", - "_view_module_version": "1.5.0", - "_view_name": "HTMLView", - "description": "", - "description_tooltip": null, - "layout": "IPY_MODEL_ded9b4b26ce14ba0aae0b19e0274fcd0", - "placeholder": "", - "style": "IPY_MODEL_47b21a1460f446d9911f2d06c9dd5bbd", - "value": "" - } - }, - "dda1704b991a44dfb2b28191f307e081": { - "model_module": "@jupyter-widgets/controls", - "model_name": "FloatProgressModel", - "model_module_version": "1.5.0", - "state": { - "_dom_classes": [], - "_model_module": "@jupyter-widgets/controls", - "_model_module_version": "1.5.0", - "_model_name": "FloatProgressModel", - "_view_count": null, - "_view_module": "@jupyter-widgets/controls", - "_view_module_version": "1.5.0", - "_view_name": "ProgressView", - "bar_style": "success", - "description": "", - "description_tooltip": null, - "layout": "IPY_MODEL_21f5a81dd29549e69b5cfa2661371e35", - "max": 1, - "min": 0, - "orientation": "horizontal", - "style": "IPY_MODEL_5f608f4313d64e458314ace8237708b1", - "value": 0 - } - }, - "5c0847c658d3489ca529d1afc8c96ed5": { - "model_module": "@jupyter-widgets/controls", - "model_name": "HTMLModel", - "model_module_version": "1.5.0", - "state": { - "_dom_classes": [], - "_model_module": "@jupyter-widgets/controls", - "_model_module_version": "1.5.0", - "_model_name": "HTMLModel", - "_view_count": null, - "_view_module": "@jupyter-widgets/controls", - "_view_module_version": "1.5.0", - "_view_name": "HTMLView", - "description": "", - "description_tooltip": null, - "layout": "IPY_MODEL_5266eb8eb83946e0acbda1362f3bd48a", - "placeholder": "", - "style": "IPY_MODEL_8d7ef198c08b4a4ab49501e2204bdbb6", - "value": " 0/0 [00:00<?, ?it/s]" - } - }, - "9fbbb0bf7d8e4c12939193791efc590f": { + "3731a5747724403e86ef1abaf4fb389b": { "model_module": "@jupyter-widgets/base", - "model_name": "LayoutModel", "model_module_version": "1.2.0", + "model_name": "LayoutModel", "state": { "_model_module": "@jupyter-widgets/base", "_model_module_version": "1.2.0", @@ -1524,62 +1877,52 @@ "width": null } }, - "ded9b4b26ce14ba0aae0b19e0274fcd0": { - "model_module": "@jupyter-widgets/base", - "model_name": "LayoutModel", - "model_module_version": "1.2.0", + "3aa1aef302474a0cba55152e311acd40": { + "model_module": "@jupyter-widgets/controls", + "model_module_version": "1.5.0", + "model_name": "HTMLModel", "state": { - "_model_module": "@jupyter-widgets/base", - "_model_module_version": "1.2.0", - "_model_name": "LayoutModel", - "_view_count": null, - "_view_module": "@jupyter-widgets/base", - "_view_module_version": "1.2.0", - "_view_name": "LayoutView", - "align_content": null, - "align_items": null, - "align_self": null, - "border": null, - "bottom": null, - "display": null, - "flex": null, - "flex_flow": null, - "grid_area": null, - "grid_auto_columns": null, - "grid_auto_flow": null, - "grid_auto_rows": null, - "grid_column": null, - "grid_gap": null, - "grid_row": null, - "grid_template_areas": null, - "grid_template_columns": null, - "grid_template_rows": null, - "height": null, - "justify_content": null, - "justify_items": null, - "left": null, - "margin": null, - "max_height": null, - "max_width": null, - "min_height": null, - "min_width": null, - "object_fit": null, - "object_position": null, - "order": null, - "overflow": null, - "overflow_x": null, - "overflow_y": null, - "padding": null, - "right": null, - "top": null, - "visibility": null, - "width": null + "_dom_classes": [], + "_model_module": "@jupyter-widgets/controls", + "_model_module_version": "1.5.0", + "_model_name": "HTMLModel", + "_view_count": null, + "_view_module": "@jupyter-widgets/controls", + "_view_module_version": "1.5.0", + "_view_name": "HTMLView", + "description": "", + "description_tooltip": null, + "layout": "IPY_MODEL_ded9b4b26ce14ba0aae0b19e0274fcd0", + "placeholder": "", + "style": "IPY_MODEL_47b21a1460f446d9911f2d06c9dd5bbd", + "value": "" + } + }, + "431ffcb0dbb1419dbbb237a29842b905": { + "model_module": "@jupyter-widgets/controls", + "model_module_version": "1.5.0", + "model_name": "HTMLModel", + "state": { + "_dom_classes": [], + "_model_module": "@jupyter-widgets/controls", + "_model_module_version": "1.5.0", + "_model_name": "HTMLModel", + "_view_count": null, + "_view_module": "@jupyter-widgets/controls", + "_view_module_version": "1.5.0", + "_view_name": "HTMLView", + "description": "", + "description_tooltip": null, + "layout": "IPY_MODEL_562a287082eb4882af864062ae13c91b", + "placeholder": "", + "style": "IPY_MODEL_11e99924bdf84739b3cfc7d676b797e5", + "value": " 665/665 [00:00<00:00, 12.8kB/s]" } }, "47b21a1460f446d9911f2d06c9dd5bbd": { "model_module": "@jupyter-widgets/controls", - "model_name": "DescriptionStyleModel", "model_module_version": "1.5.0", + "model_name": "DescriptionStyleModel", "state": { "_model_module": "@jupyter-widgets/controls", "_model_module_version": "1.5.0", @@ -1591,10 +1934,10 @@ "description_width": "" } }, - "21f5a81dd29549e69b5cfa2661371e35": { + "5266eb8eb83946e0acbda1362f3bd48a": { "model_module": "@jupyter-widgets/base", - "model_name": "LayoutModel", "model_module_version": "1.2.0", + "model_name": "LayoutModel", "state": { "_model_module": "@jupyter-widgets/base", "_model_module_version": "1.2.0", @@ -1640,29 +1983,13 @@ "right": null, "top": null, "visibility": null, - "width": "20px" - } - }, - "5f608f4313d64e458314ace8237708b1": { - "model_module": "@jupyter-widgets/controls", - "model_name": "ProgressStyleModel", - "model_module_version": "1.5.0", - "state": { - "_model_module": "@jupyter-widgets/controls", - "_model_module_version": "1.5.0", - "_model_name": "ProgressStyleModel", - "_view_count": null, - "_view_module": "@jupyter-widgets/base", - "_view_module_version": "1.2.0", - "_view_name": "StyleView", - "bar_color": null, - "description_width": "" + "width": null } }, - "5266eb8eb83946e0acbda1362f3bd48a": { + "562a287082eb4882af864062ae13c91b": { "model_module": "@jupyter-widgets/base", - "model_name": "LayoutModel", "model_module_version": "1.2.0", + "model_name": "LayoutModel", "state": { "_model_module": "@jupyter-widgets/base", "_model_module_version": "1.2.0", @@ -1711,10 +2038,10 @@ "width": null } }, - "8d7ef198c08b4a4ab49501e2204bdbb6": { + "56490c20aef04fe4a63ba7c990e71e77": { "model_module": "@jupyter-widgets/controls", - "model_name": "DescriptionStyleModel", "model_module_version": "1.5.0", + "model_name": "DescriptionStyleModel", "state": { "_model_module": "@jupyter-widgets/controls", "_model_module_version": "1.5.0", @@ -1725,1599 +2052,1263 @@ "_view_name": "StyleView", "description_width": "" } - } - } - } - }, - "cells": [ - { - "cell_type": "markdown", - "metadata": { - "id": "view-in-github", - "colab_type": "text" - }, - "source": [ - "<a href=\"https://colab.research.google.com/github/JadenFiotto-Kaufman/nnsight/blob/dev/NNsight_v0_2.ipynb\" target=\"_parent\"><img src=\"https://colab.research.google.com/assets/colab-badge.svg\" alt=\"Open In Colab\"/></a>" - ] - }, - { - "cell_type": "markdown", - "source": [ - "<img src=\"https://nnsight.net/_static/images/nnsight_logo.svg\" alt=\"drawing\" width=\"200\"/>\n", - "\n", - "# NNsight\n", - "## The API for a transparent science on black-box AI\n", - "\n", - "In this era of large-scale deep learning, the most interesting AI models are massive black boxes that are hard to run. Ordinary commercial inference service APIs let you interact with huge models, but they do not let you access model internals.\n", - "\n", - "The nnsight library is different: it gives you full access to all the neural network internals. When used together with a remote service like the [National Deep Inference Facility](https://thevisible.net/docs/NDIF-proposal.pdf) (NDIF), it lets you run complex experiments on huge open source models easily, with fully transparent access.\n", - "\n", - "Our team wants to enable entire labs and independent researchers alike, as we believe a large, passionate, and collaborative community will produce the next big insights on a profoundly important field.\n", - "\n", - "\n" - ], - "metadata": { - "id": "-Scfr942GwO4" - } - }, - { - "cell_type": "markdown", - "source": [ - "# 1 First, let's start small\n", - "\n", - "\n", - "\n" - ], - "metadata": { - "id": "N1OemD2VGyZx" - } - }, - { - "cell_type": "markdown", - "source": [ - "## The Tracing Context\n", - "\n", - "To demonstrate the core functionality and syntax of nnsight, we'll define and use a tiny two layer neural network." - ], - "metadata": { - "id": "WyLMmhrAKTNM" - } - }, - { - "cell_type": "code", - "execution_count": 1, - "metadata": { - "id": "R0RJijD0eXwf" - }, - "outputs": [], - "source": [ - "# Install nnsight\n", - "!pip install nnsight==0.2.3.dev0\n", - "\n", - "from IPython.display import clear_output\n", - "\n", - "clear_output()" - ] - }, - { - "cell_type": "markdown", - "source": [ - "Our little model here is composed of four sub-modules, two linear layers ('layer1', 'layer2'). We specify the sizes of each of these modules, and create some complementary example input." - ], - "metadata": { - "id": "bgydw7i3HmIH" - } - }, - { - "cell_type": "code", - "source": [ - "from collections import OrderedDict\n", - "import torch\n", - "\n", - "input_size = 5\n", - "hidden_dims = 10\n", - "output_size = 2\n", - "\n", - "net = torch.nn.Sequential(OrderedDict([\n", - " ('layer1', torch.nn.Linear(input_size, hidden_dims)),\n", - " ('layer2', torch.nn.Linear(hidden_dims, output_size)),\n", - "])).requires_grad_(False)\n", - "\n", - "input = torch.rand((1, input_size))" - ], - "metadata": { - "id": "2pX2Wg8Ceo6N" - }, - "execution_count": null, - "outputs": [] - }, - { - "cell_type": "markdown", - "source": [ - "The core object of the nnsight package is `NNsight`. This wraps around a given pytorch model to enable the capabilites nnsight provides." - ], - "metadata": { - "id": "tIPa2h2pJIwl" - } - }, - { - "cell_type": "code", - "source": [ - "from nnsight import NNsight\n", - "\n", - "model = NNsight(net)" - ], - "metadata": { - "id": "8H9R_ynTJI5y", - "outputId": "5d6b545d-750e-441a-fde9-2dc53e2f1f5e", - "colab": { - "base_uri": "https://localhost:8080/", - "height": 426, - "referenced_widgets": [ - "b3078a7a41c642839748d9864220cbde", - "3aa1aef302474a0cba55152e311acd40", - "dda1704b991a44dfb2b28191f307e081", - "5c0847c658d3489ca529d1afc8c96ed5", - "9fbbb0bf7d8e4c12939193791efc590f", - "ded9b4b26ce14ba0aae0b19e0274fcd0", - "47b21a1460f446d9911f2d06c9dd5bbd", - "21f5a81dd29549e69b5cfa2661371e35", - "5f608f4313d64e458314ace8237708b1", - "5266eb8eb83946e0acbda1362f3bd48a", - "8d7ef198c08b4a4ab49501e2204bdbb6" - ] - } - }, - "execution_count": 2, - "outputs": [ - { - "output_type": "stream", - "name": "stderr", - "text": [ - "The cache for model files in Transformers v4.22.0 has been updated. Migrating your old cache. This is a one-time only operation. You can interrupt this and resume the migration later on by calling `transformers.utils.move_cache()`.\n" - ] - }, - { - "output_type": "display_data", - "data": { - "text/plain": [ - "0it [00:00, ?it/s]" - ], - "application/vnd.jupyter.widget-view+json": { - "version_major": 2, - "version_minor": 0, - "model_id": "b3078a7a41c642839748d9864220cbde" - } - }, - "metadata": {} - }, - { - "output_type": "error", - "ename": "ImportError", - "evalue": "cannot import name 'NNsight' from 'nnsight' (/usr/local/lib/python3.10/dist-packages/nnsight/__init__.py)", - "traceback": [ - "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m", - "\u001b[0;31mImportError\u001b[0m Traceback (most recent call last)", - "\u001b[0;32m<ipython-input-2-801ad166c2ee>\u001b[0m in \u001b[0;36m<cell line: 1>\u001b[0;34m()\u001b[0m\n\u001b[0;32m----> 1\u001b[0;31m \u001b[0;32mfrom\u001b[0m \u001b[0mnnsight\u001b[0m \u001b[0;32mimport\u001b[0m \u001b[0mNNsight\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m 2\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 3\u001b[0m \u001b[0mmodel\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mNNsight\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mnet\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n", - "\u001b[0;31mImportError\u001b[0m: cannot import name 'NNsight' from 'nnsight' (/usr/local/lib/python3.10/dist-packages/nnsight/__init__.py)", - "", - "\u001b[0;31m---------------------------------------------------------------------------\u001b[0;32m\nNOTE: If your import is failing due to a missing package, you can\nmanually install dependencies using either !pip or !apt.\n\nTo view examples of installing some common dependencies, click the\n\"Open Examples\" button below.\n\u001b[0;31m---------------------------------------------------------------------------\u001b[0m\n" - ], - "errorDetails": { - "actions": [ - { - "action": "open_url", - "actionText": "Open Examples", - "url": "/notebooks/snippets/importing_libraries.ipynb" - } - ] - } - } - ] - }, - { - "cell_type": "markdown", - "source": [ - "Printing a Pytorch model shows a named hierarchy of modules which is very useful when accessing sub-components directly. NNsight models work the same." - ], - "metadata": { - "id": "2NfISlQ_Ilvp" - } - }, - { - "cell_type": "code", - "source": [ - "print(model)" - ], - "metadata": { - "id": "bYtnbJHvlGZV", - "colab": { - "base_uri": "https://localhost:8080/" }, - "outputId": "5eb7c572-451c-43c6-956f-af224ae1867c" - }, - "execution_count": null, - "outputs": [ - { - "output_type": "stream", - "name": "stdout", - "text": [ - "Sequential(\n", - " (layer1): Linear(in_features=5, out_features=10, bias=True)\n", - " (layer2): Linear(in_features=10, out_features=2, bias=True)\n", - ")\n" - ] - } - ] - }, - { - "cell_type": "markdown", - "source": [ - "Before we actually get to using the model we just created, let's talk about Python contexts.\n", - "\n", - "Python contexts define a scope using the `with` statement and are often used to create some object, or initiate some logic, that you later want to destroy or conclude.\n", - "\n", - "The most common application is opening files like the following example:\n", - "\n", - "```python\n", - "with open('myfile.txt', 'r') as file:\n", - " text = file.read()\n", - "```\n", - "\n", - "Python uses the `with` keyword to enter a context-like object. This object defines logic to be run at the start of the `with` block, as well as logic to be run when exiting. When using `with` for a file, entering the context opens the file and exiting the context closes it. Being within the context means we can read from the file. Simple enough! Now we can discuss how `nnsight` uses contexts to enable intuitive access into the internals of a neural network.\n" - ], - "metadata": { - "id": "djC5kyJWLuUH" - } - }, - { - "cell_type": "markdown", - "source": [ - "The main tool with `nnsight` is a context for tracing.\n", - "\n", - "We enter the tracing context by calling `model.trace(<input>)` on an `NNsight` model, which defines how we want to run the model. Inside the context, we will be able to customize how the neural network runs. The model is actually run upon exiting the tracing context." - ], - "metadata": { - "id": "iNvuCOeyojcA" - } - }, - { - "cell_type": "code", - "source": [ - "with model.trace(input) as tracer:\n", - " pass" - ], - "metadata": { - "id": "qEXQ4auPSL-m" - }, - "execution_count": null, - "outputs": [] - }, - { - "cell_type": "markdown", - "source": [ - "But where's the output? To get that, we'll have to learn how to request it from within the tracing conext." - ], - "metadata": { - "id": "UZQsHinjqicJ" - } - }, - { - "cell_type": "markdown", - "source": [ - "## Getting" - ], - "metadata": { - "id": "xMfBpYzDPMoB" - } - }, - { - "cell_type": "markdown", - "source": [ - "Earlier, when we wrapped our little neural net with the `NNsight` class. This added a couple properties to each module in the model (including the root model itself). The two most important ones are `.input` and `.output`.\n", - "\n", - "```python\n", - "model.input\n", - "model.output\n", - "```\n", - "\n", - "The names are self explainatory. They correspond to the inputs and outputs of their respective modules during a forward pass of the model. We can use these attributes inside the `with` block.\n", - "\n", - "However, it is important to understand that the model is not executed until the end of the tracing context. How can we access inputs and outputs before the model is run? The trick is deferred execution.\n", - "\n", - "`.input` and `.output` are Proxies for the eventual inputs and outputs of a module. In other words, when you access `model.output` what you are communicating to `nnsight` is, \"When you compute the output of `model`, please grab it for me and put the value into its corresponding Proxy object's `.value` attribute.\" Let's try it:" - ], - "metadata": { - "id": "B5_aFwFRv0ax" - } - }, - { - "cell_type": "code", - "source": [ - "with model.trace(input) as tracer:\n", - "\n", - " output = model.output\n", - "\n", - "print(output.value)" - ], - "metadata": { - "id": "cYe8-r9ptGaG", - "colab": { - "base_uri": "https://localhost:8080/", - "height": 303 + "5c0847c658d3489ca529d1afc8c96ed5": { + "model_module": "@jupyter-widgets/controls", + "model_module_version": "1.5.0", + "model_name": "HTMLModel", + "state": { + "_dom_classes": [], + "_model_module": "@jupyter-widgets/controls", + "_model_module_version": "1.5.0", + "_model_name": "HTMLModel", + "_view_count": null, + "_view_module": "@jupyter-widgets/controls", + "_view_module_version": "1.5.0", + "_view_name": "HTMLView", + "description": "", + "description_tooltip": null, + "layout": "IPY_MODEL_5266eb8eb83946e0acbda1362f3bd48a", + "placeholder": "", + "style": "IPY_MODEL_8d7ef198c08b4a4ab49501e2204bdbb6", + "value": " 0/0 [00:00<?, ?it/s]" + } }, - "outputId": "adf25efe-d2cf-4f65-9eaf-15b155762f4b" - }, - "execution_count": null, - "outputs": [ - { - "output_type": "error", - "ename": "ValueError", - "evalue": "Accessing Proxy value before it's been set.", - "traceback": [ - "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m", - "\u001b[0;31mValueError\u001b[0m Traceback (most recent call last)", - "\u001b[0;32m<ipython-input-34-c7e0c74b12fa>\u001b[0m in \u001b[0;36m<cell line: 5>\u001b[0;34m()\u001b[0m\n\u001b[1;32m 3\u001b[0m \u001b[0moutput\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mmodel\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0moutput\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 4\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m----> 5\u001b[0;31m \u001b[0mprint\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0moutput\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mvalue\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m", - "\u001b[0;32m/usr/local/lib/python3.10/dist-packages/nnsight/tracing/Proxy.py\u001b[0m in \u001b[0;36mvalue\u001b[0;34m(self)\u001b[0m\n\u001b[1;32m 47\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 48\u001b[0m \u001b[0;32mif\u001b[0m \u001b[0;32mnot\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mnode\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mdone\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m---> 49\u001b[0;31m \u001b[0;32mraise\u001b[0m \u001b[0mValueError\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m\"Accessing Proxy value before it's been set.\"\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m 50\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 51\u001b[0m \u001b[0;32mreturn\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mnode\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mvalue\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n", - "\u001b[0;31mValueError\u001b[0m: Accessing Proxy value before it's been set." - ] - } - ] - }, - { - "cell_type": "markdown", - "source": [ - "Oh no an error! \"Accessing Proxy value before it's been set.\"\n", - "\n", - "Why doesn't our `output` have a `value`?\n", - "\n", - "Proxy objects will only have their value at the end of a context if we call `.save()` on them. This helps to reduce memory costs. Adding `.save()` fixes the error:" - ], - "metadata": { - "id": "sBz5qfwuR-6F" - } - }, - { - "cell_type": "code", - "source": [ - "with model.trace(input) as tracer:\n", - "\n", - " output = model.output.save()\n", - "\n", - "print(output.value)" - ], - "metadata": { - "id": "y_bXRd5dvsBu", - "colab": { - "base_uri": "https://localhost:8080/" + "5f608f4313d64e458314ace8237708b1": { + "model_module": "@jupyter-widgets/controls", + "model_module_version": "1.5.0", + "model_name": "ProgressStyleModel", + "state": { + "_model_module": "@jupyter-widgets/controls", + "_model_module_version": "1.5.0", + "_model_name": "ProgressStyleModel", + "_view_count": null, + "_view_module": "@jupyter-widgets/base", + "_view_module_version": "1.2.0", + "_view_name": "StyleView", + "bar_color": null, + "description_width": "" + } }, - "outputId": "a8e8985c-2661-46a8-975b-d43f67cfc162" - }, - "execution_count": null, - "outputs": [ - { - "output_type": "stream", - "name": "stdout", - "text": [ - "tensor([[ 0.1473, -0.1518]])\n" - ] - } - ] - }, - { - "cell_type": "markdown", - "source": [ - "Success! We now have the model output. You just completed your first intervention using `nnsight`.\n", - "\n", - "\n", - "Each time you access a module's input or output, you create an intervention in the neural network's forward pass. Collectively these requests form the intervention graph. We call the process of executing it alongside the model's normal computation graph, interleaving.\n", - "\n", - "<details>\n", - "<summary>On Model output</summary>\n", - "\n", - "---\n", - "\n", - "If you don't need to access anything other than the final model output, you can call the tracing context with `trace=False` and not use it as a context:\n", - "\n", - "```python\n", - "output = model.trace(<inputs>, trace=False)\n", - "```\n", - "\n", - "---\n", - "\n", - "</details>\n", - "\n", - "Just like we saved the output of the model as a whole, we can save the output of any of its submodules. We use normal Python attribute syntax. We can discover how to access them by name by printing out the model:" - ], - "metadata": { - "id": "t5C0UZCwvrYn" - } - }, - { - "cell_type": "code", - "source": [ - "print(model)" - ], - "metadata": { - "colab": { - "base_uri": "https://localhost:8080/" + "60fe026e908741209637604edfeb0563": { + "model_module": "@jupyter-widgets/controls", + "model_module_version": "1.5.0", + "model_name": "HBoxModel", + "state": { + "_dom_classes": [], + "_model_module": "@jupyter-widgets/controls", + "_model_module_version": "1.5.0", + "_model_name": "HBoxModel", + "_view_count": null, + "_view_module": "@jupyter-widgets/controls", + "_view_module_version": "1.5.0", + "_view_name": "HBoxView", + "box_style": "", + "children": [ + "IPY_MODEL_e7a73cfc3fa04425afc08368532ef0be", + "IPY_MODEL_d839da1c31a34659988e3420ba520725", + "IPY_MODEL_96a3b8c765374a5e98eaf713dd5e85d8" + ], + "layout": "IPY_MODEL_b49ce0d00e5b4df097b3a9fc4c0bd8e7" + } }, - "id": "0WcVUSP-0CJi", - "outputId": "7fd9d6ac-152d-4431-9a4d-0b60321638eb" - }, - "execution_count": null, - "outputs": [ - { - "output_type": "stream", - "name": "stdout", - "text": [ - "Sequential(\n", - " (layer1): Linear(in_features=5, out_features=10, bias=True)\n", - " (layer2): Linear(in_features=10, out_features=2, bias=True)\n", - ")\n" - ] - } - ] - }, - { - "cell_type": "code", - "source": [ - "with model.trace(input) as tracer:\n", - "\n", - " l1_output = model.layer1.output.save()\n", - "\n", - "print(l1_output.value)" - ], - "metadata": { - "colab": { - "base_uri": "https://localhost:8080/" + "688bf676958a4050b73cc9597f5110de": { + "model_module": "@jupyter-widgets/controls", + "model_module_version": "1.5.0", + "model_name": "ProgressStyleModel", + "state": { + "_model_module": "@jupyter-widgets/controls", + "_model_module_version": "1.5.0", + "_model_name": "ProgressStyleModel", + "_view_count": null, + "_view_module": "@jupyter-widgets/base", + "_view_module_version": "1.2.0", + "_view_name": "StyleView", + "bar_color": null, + "description_width": "" + } }, - "id": "akWr-cNqy-9O", - "outputId": "68b9e071-c609-4b70-906d-2f2cd93b4495" - }, - "execution_count": null, - "outputs": [ - { - "output_type": "stream", - "name": "stdout", - "text": [ - "tensor([[ 0.0458, 0.5267, 0.7119, 0.4046, 0.2460, 0.7998, 0.4485, -0.2506,\n", - " 0.2968, -0.8834]])\n" - ] - } - ] - }, - { - "cell_type": "markdown", - "source": [ - "Let's do the same for the input of layer2. While we're at it, let's also drop the `as tracer`, as we won't be needing the tracer object itself for a few sections:" - ], - "metadata": { - "id": "85A-aP_03ht6" - } - }, - { - "cell_type": "code", - "source": [ - "with model.trace(input):\n", - "\n", - " l2_input = model.layer2.input.save()\n", - "\n", - "print(l2_input.value)" - ], - "metadata": { - "id": "3EHEN38N3nXR", - "colab": { - "base_uri": "https://localhost:8080/" + "6aa96b80e3b746b48f5c13bcafe1134e": { + "model_module": "@jupyter-widgets/controls", + "model_module_version": "1.5.0", + "model_name": "ProgressStyleModel", + "state": { + "_model_module": "@jupyter-widgets/controls", + "_model_module_version": "1.5.0", + "_model_name": "ProgressStyleModel", + "_view_count": null, + "_view_module": "@jupyter-widgets/base", + "_view_module_version": "1.2.0", + "_view_name": "StyleView", + "bar_color": null, + "description_width": "" + } }, - "outputId": "5afd17a7-0bcc-4d1a-c131-7038bde9455d" - }, - "execution_count": null, - "outputs": [ - { - "output_type": "stream", - "name": "stdout", - "text": [ - "((tensor([[ 0.0458, 0.5267, 0.7119, 0.4046, 0.2460, 0.7998, 0.4485, -0.2506,\n", - " 0.2968, -0.8834]]),), {})\n" - ] - } - ] - }, - { - "cell_type": "markdown", - "source": [ - "<details>\n", - " <summary>On module inputs</summary>\n", - "\n", - " ---\n", - "\n", - " Notice how the value for `l2_input`, was not just a single tensor.\n", - " The type/shape of values from `.input` is in the form of:\n", - "\n", - " tuple(tuple(args), dictionary(kwargs))\n", - "\n", - " Where the first index of the tuple is itself a tuple of all positional arguments, and the second index is a dictionary of the keyword arguments.\n", - "\n", - " ---\n", - "\n", - "</details>\n", - "\n", - "\n", - "Now that we can access activations, we also want to do some post-processing on it. Let's find out which dimension of layer1's output has the highest value." - ], - "metadata": { - "id": "jk-U8zi33Gi-" - } - }, - { - "cell_type": "markdown", - "source": [ - "## Functions, Methods, and Operations" - ], - "metadata": { - "id": "p7Xo_PHyPr4p" - } - }, - { - "cell_type": "markdown", - "source": [ - "We could do this by calling `torch.argmax(...)` after the tracing context or we can just leverage the fact that `nnsight` handles functions and methods within the tracing context, by creating a Proxy request for it:" - ], - "metadata": { - "id": "ehSofWbx5DSx" - } - }, - { - "cell_type": "code", - "source": [ - "with model.trace(input):\n", - "\n", - " # Note we don't need to call .save() on the output,\n", - " # as we're only using its value within the tracing context.\n", - " l1_output = model.layer1.output\n", - "\n", - " l1_amax = torch.argmax(l1_output, dim=1).save()\n", - "\n", - "print(l1_amax[0])" - ], - "metadata": { - "id": "c5XCiSZn2p3k", - "colab": { - "base_uri": "https://localhost:8080/" + "74358c341daa44faa41a92b0eb073615": { + "model_module": "@jupyter-widgets/controls", + "model_module_version": "1.5.0", + "model_name": "HTMLModel", + "state": { + "_dom_classes": [], + "_model_module": "@jupyter-widgets/controls", + "_model_module_version": "1.5.0", + "_model_name": "HTMLModel", + "_view_count": null, + "_view_module": "@jupyter-widgets/controls", + "_view_module_version": "1.5.0", + "_view_name": "HTMLView", + "description": "", + "description_tooltip": null, + "layout": "IPY_MODEL_7a0f3d2deeb14bbba7f0df208133a7e2", + "placeholder": "", + "style": "IPY_MODEL_94cdcf1fb0914dfe99c3c615e68b7ec8", + "value": "config.json: 100%" + } + }, + "7a0f3d2deeb14bbba7f0df208133a7e2": { + "model_module": "@jupyter-widgets/base", + "model_module_version": "1.2.0", + "model_name": "LayoutModel", + "state": { + "_model_module": "@jupyter-widgets/base", + "_model_module_version": "1.2.0", + "_model_name": "LayoutModel", + "_view_count": null, + "_view_module": "@jupyter-widgets/base", + "_view_module_version": "1.2.0", + "_view_name": "LayoutView", + "align_content": null, + "align_items": null, + "align_self": null, + "border": null, + "bottom": null, + "display": null, + "flex": null, + "flex_flow": null, + "grid_area": null, + "grid_auto_columns": null, + "grid_auto_flow": null, + "grid_auto_rows": null, + "grid_column": null, + "grid_gap": null, + "grid_row": null, + "grid_template_areas": null, + "grid_template_columns": null, + "grid_template_rows": null, + "height": null, + "justify_content": null, + "justify_items": null, + "left": null, + "margin": null, + "max_height": null, + "max_width": null, + "min_height": null, + "min_width": null, + "object_fit": null, + "object_position": null, + "order": null, + "overflow": null, + "overflow_x": null, + "overflow_y": null, + "padding": null, + "right": null, + "top": null, + "visibility": null, + "width": null + } + }, + "7e6b41ec535648b190bfc655d9e430dd": { + "model_module": "@jupyter-widgets/controls", + "model_module_version": "1.5.0", + "model_name": "HTMLModel", + "state": { + "_dom_classes": [], + "_model_module": "@jupyter-widgets/controls", + "_model_module_version": "1.5.0", + "_model_name": "HTMLModel", + "_view_count": null, + "_view_module": "@jupyter-widgets/controls", + "_view_module_version": "1.5.0", + "_view_name": "HTMLView", + "description": "", + "description_tooltip": null, + "layout": "IPY_MODEL_beb4231b49024d8a9bb7ead19647d99d", + "placeholder": "", + "style": "IPY_MODEL_26d09aa072d64af0bf4df9a89309093a", + "value": " 1.36M/1.36M [00:00<00:00, 4.16MB/s]" + } }, - "outputId": "97d34dee-7bb5-4735-8c53-6aac6377af9e" - }, - "execution_count": null, - "outputs": [ - { - "output_type": "stream", - "name": "stdout", - "text": [ - "tensor(5)\n" - ] - } - ] - }, - { - "cell_type": "markdown", - "source": [ - "Nice! That worked seamlessly, but hold on, how come we didn't need to call `.value[0]` on the result? In previous sections, we were just being explicit to get an understanding of Proxies and their value. In practice, however, `nnsight` knows that when outside of the tracing context we only care about the actual value, and so printing, indexing, and applying functions all immediately return and reflect the data in `.value`. So for the rest of the tutorial we won't use it.\n", - "\n", - "The same principles work for methods and operations as well:" - ], - "metadata": { - "id": "BGPUJvWq_bOp" - } - }, - { - "cell_type": "code", - "source": [ - "with model.trace(input):\n", - "\n", - " value = (model.layer1.output.sum() + model.layer2.output.sum()).save()\n", - "\n", - "print(value)" - ], - "metadata": { - "id": "FIcOYeEjFuln", - "colab": { - "base_uri": "https://localhost:8080/" + "8776d04bfd3748b69e10708466e9c132": { + "model_module": "@jupyter-widgets/base", + "model_module_version": "1.2.0", + "model_name": "LayoutModel", + "state": { + "_model_module": "@jupyter-widgets/base", + "_model_module_version": "1.2.0", + "_model_name": "LayoutModel", + "_view_count": null, + "_view_module": "@jupyter-widgets/base", + "_view_module_version": "1.2.0", + "_view_name": "LayoutView", + "align_content": null, + "align_items": null, + "align_self": null, + "border": null, + "bottom": null, + "display": null, + "flex": null, + "flex_flow": null, + "grid_area": null, + "grid_auto_columns": null, + "grid_auto_flow": null, + "grid_auto_rows": null, + "grid_column": null, + "grid_gap": null, + "grid_row": null, + "grid_template_areas": null, + "grid_template_columns": null, + "grid_template_rows": null, + "height": null, + "justify_content": null, + "justify_items": null, + "left": null, + "margin": null, + "max_height": null, + "max_width": null, + "min_height": null, + "min_width": null, + "object_fit": null, + "object_position": null, + "order": null, + "overflow": null, + "overflow_x": null, + "overflow_y": null, + "padding": null, + "right": null, + "top": null, + "visibility": null, + "width": null + } }, - "outputId": "d836abd3-b713-44c5-aedc-95bf85abf11a" - }, - "execution_count": null, - "outputs": [ - { - "output_type": "stream", - "name": "stdout", - "text": [ - "tensor(2.3416)\n" - ] - } - ] - }, - { - "cell_type": "markdown", - "source": [ - "By default, torch functiona, methods and all operators work with `nnsight`. We also enable the use of the `einops` library.\n", - "\n", - "So to recap, the above code block is saying to `nnsight`, \"Run the model with the given `input`. When the output of layer1 is computed, take its sum. Then do the same for layer2. Now that both of those are computed, add them and make sure not to delete this value as I wish to use it outside of the tracing context.\"\n", - "\n", - "Getting and analyzing the activations from various points in a model can be really insightful, and a number of ML techniques do exactly that. However, often times we not only want to view the computation of a model, but influence it as well." - ], - "metadata": { - "id": "V0vNOJtJ2oFR" - } - }, - { - "cell_type": "markdown", - "source": [ - "## Setting" - ], - "metadata": { - "id": "m_5qH5gHPOT_" - } - }, - { - "cell_type": "markdown", - "source": [ - "To demonstrate the effect of editing the flow of information through the model, let's set the first dimension of the first layer's output to 0. `NNsight` makes this really easy using '=' operator:" - ], - "metadata": { - "id": "rgju-b_IOLlq" - } - }, - { - "cell_type": "code", - "source": [ - "with model.trace(input):\n", - "\n", - " # Save the output before the edit to compare.\n", - " # Notice we apply .clone() before saving as the setting operation is in-place.\n", - " l1_output_before = model.layer1.output.clone().save()\n", - "\n", - " # Access the 0th index of the hidden state dimension and set it to 0.\n", - " model.layer1.output[:, 0] = 0\n", - "\n", - " # Save the output after to see our edit.\n", - " l1_output_after = model.layer1.output.save()\n", - "\n", - "print(\"Before:\", l1_output_before)\n", - "print(\"After:\", l1_output_after)" - ], - "metadata": { - "id": "-6y2wzJqOz3a", - "colab": { - "base_uri": "https://localhost:8080/" + "8d7ef198c08b4a4ab49501e2204bdbb6": { + "model_module": "@jupyter-widgets/controls", + "model_module_version": "1.5.0", + "model_name": "DescriptionStyleModel", + "state": { + "_model_module": "@jupyter-widgets/controls", + "_model_module_version": "1.5.0", + "_model_name": "DescriptionStyleModel", + "_view_count": null, + "_view_module": "@jupyter-widgets/base", + "_view_module_version": "1.2.0", + "_view_name": "StyleView", + "description_width": "" + } + }, + "8de63414803f41c29e75242756cfd67f": { + "model_module": "@jupyter-widgets/base", + "model_module_version": "1.2.0", + "model_name": "LayoutModel", + "state": { + "_model_module": "@jupyter-widgets/base", + "_model_module_version": "1.2.0", + "_model_name": "LayoutModel", + "_view_count": null, + "_view_module": "@jupyter-widgets/base", + "_view_module_version": "1.2.0", + "_view_name": "LayoutView", + "align_content": null, + "align_items": null, + "align_self": null, + "border": null, + "bottom": null, + "display": null, + "flex": null, + "flex_flow": null, + "grid_area": null, + "grid_auto_columns": null, + "grid_auto_flow": null, + "grid_auto_rows": null, + "grid_column": null, + "grid_gap": null, + "grid_row": null, + "grid_template_areas": null, + "grid_template_columns": null, + "grid_template_rows": null, + "height": null, + "justify_content": null, + "justify_items": null, + "left": null, + "margin": null, + "max_height": null, + "max_width": null, + "min_height": null, + "min_width": null, + "object_fit": null, + "object_position": null, + "order": null, + "overflow": null, + "overflow_x": null, + "overflow_y": null, + "padding": null, + "right": null, + "top": null, + "visibility": null, + "width": null + } + }, + "9373fcd054014760b23e3acc928da0a5": { + "model_module": "@jupyter-widgets/controls", + "model_module_version": "1.5.0", + "model_name": "FloatProgressModel", + "state": { + "_dom_classes": [], + "_model_module": "@jupyter-widgets/controls", + "_model_module_version": "1.5.0", + "_model_name": "FloatProgressModel", + "_view_count": null, + "_view_module": "@jupyter-widgets/controls", + "_view_module_version": "1.5.0", + "_view_name": "ProgressView", + "bar_style": "success", + "description": "", + "description_tooltip": null, + "layout": "IPY_MODEL_0fddebfb42e847e08890d09f30161984", + "max": 1355256, + "min": 0, + "orientation": "horizontal", + "style": "IPY_MODEL_d07a0a49a2924092affc0e07bf0b5ef0", + "value": 1355256 + } }, - "outputId": "6274e5c5-c5fe-4e07-87c4-a119cfd09742" - }, - "execution_count": null, - "outputs": [ - { - "output_type": "stream", - "name": "stdout", - "text": [ - "Before: tensor([[ 0.0458, 0.5267, 0.7119, 0.4046, 0.2460, 0.7998, 0.4485, -0.2506,\n", - " 0.2968, -0.8834]])\n", - "After: tensor([[ 0.0000, 0.5267, 0.7119, 0.4046, 0.2460, 0.7998, 0.4485, -0.2506,\n", - " 0.2968, -0.8834]])\n" - ] - } - ] - }, - { - "cell_type": "markdown", - "source": [ - "Seems our change was reflected. Now the same for the last dimension:" - ], - "metadata": { - "id": "oZz9SMs3Y_iS" - } - }, - { - "cell_type": "code", - "source": [ - "with model.trace(input):\n", - "\n", - " # Save the output before the edit to compare.\n", - " # Notice we apply .clone() before saving as the setting operation is in-place.\n", - " l1_output_before = model.layer1.output.clone().save()\n", - "\n", - " # Access the last index of the hidden state dimension and set it to 0.\n", - " model.layer1.output[:, hidden_dims] = 0\n", - "\n", - " # Save the output after to see our edit.\n", - " l1_output_after = model.layer1.output.save()\n", - "\n", - "print(\"Before:\", l1_output_before)\n", - "print(\"After:\", l1_output_after)" - ], - "metadata": { - "colab": { - "base_uri": "https://localhost:8080/", - "height": 363 + "94cdcf1fb0914dfe99c3c615e68b7ec8": { + "model_module": "@jupyter-widgets/controls", + "model_module_version": "1.5.0", + "model_name": "DescriptionStyleModel", + "state": { + "_model_module": "@jupyter-widgets/controls", + "_model_module_version": "1.5.0", + "_model_name": "DescriptionStyleModel", + "_view_count": null, + "_view_module": "@jupyter-widgets/base", + "_view_module_version": "1.2.0", + "_view_name": "StyleView", + "description_width": "" + } }, - "id": "qwlqvHFcxld2", - "outputId": "28ae7175-5af0-4bd0-f9f0-21680a4fd748" - }, - "execution_count": null, - "outputs": [ - { - "output_type": "error", - "ename": "IndexError", - "evalue": "index 10 is out of bounds for dimension 1 with size 10", - "traceback": [ - "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m", - "\u001b[0;31mIndexError\u001b[0m Traceback (most recent call last)", - "\u001b[0;32m<ipython-input-42-a1e18ebd4137>\u001b[0m in \u001b[0;36m<cell line: 1>\u001b[0;34m()\u001b[0m\n\u001b[0;32m----> 1\u001b[0;31m \u001b[0;32mwith\u001b[0m \u001b[0mmodel\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mtrace\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0minput\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m 2\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 3\u001b[0m \u001b[0;31m# Save the output before the edit to compare.\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 4\u001b[0m \u001b[0;31m# Notice we apply .clone() before saving as the setting operation is in-place.\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 5\u001b[0m \u001b[0ml1_output_before\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mmodel\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mlayer1\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0moutput\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mclone\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0msave\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n", - "\u001b[0;32m/usr/local/lib/python3.10/dist-packages/nnsight/contexts/Runner.py\u001b[0m in \u001b[0;36m__exit__\u001b[0;34m(self, exc_type, exc_val, exc_tb)\u001b[0m\n\u001b[1;32m 40\u001b[0m \u001b[0;32mraise\u001b[0m \u001b[0mexc_val\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 41\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m---> 42\u001b[0;31m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_graph\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mtracing\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0;32mFalse\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m 43\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 44\u001b[0m \u001b[0;32mif\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mremote\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n", - "\u001b[0;32m<ipython-input-42-a1e18ebd4137>\u001b[0m in \u001b[0;36m<cell line: 1>\u001b[0;34m()\u001b[0m\n\u001b[1;32m 6\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 7\u001b[0m \u001b[0;31m# Access the last index of the hidden state dimension and set it to 0.\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m----> 8\u001b[0;31m \u001b[0mmodel\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mlayer1\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0moutput\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mhidden_dims\u001b[0m\u001b[0;34m]\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0;36m0\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m 9\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 10\u001b[0m \u001b[0;31m# Save the output after to see our edit.\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n", - "\u001b[0;32m/usr/local/lib/python3.10/dist-packages/nnsight/tracing/Proxy.py\u001b[0m in \u001b[0;36m__setitem__\u001b[0;34m(self, key, value)\u001b[0m\n\u001b[1;32m 90\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 91\u001b[0m \u001b[0;32mdef\u001b[0m \u001b[0m__setitem__\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mkey\u001b[0m\u001b[0;34m:\u001b[0m \u001b[0mUnion\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0mProxy\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mAny\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mvalue\u001b[0m\u001b[0;34m:\u001b[0m \u001b[0mUnion\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0mSelf\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mAny\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m)\u001b[0m \u001b[0;34m->\u001b[0m \u001b[0;32mNone\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m---> 92\u001b[0;31m self.node.graph.add(\n\u001b[0m\u001b[1;32m 93\u001b[0m \u001b[0mtarget\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0moperator\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0msetitem\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 94\u001b[0m \u001b[0margs\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mnode\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mkey\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mvalue\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n", - "\u001b[0;32m/usr/local/lib/python3.10/dist-packages/nnsight/tracing/Graph.py\u001b[0m in \u001b[0;36madd\u001b[0;34m(self, target, value, args, kwargs, name)\u001b[0m\n\u001b[1;32m 144\u001b[0m \u001b[0;32mtry\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 145\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 146\u001b[0;31m value = target(\n\u001b[0m\u001b[1;32m 147\u001b[0m \u001b[0;34m\u001b[0m\u001b[0mNode\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mprepare_proxy_values\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0m_args\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 148\u001b[0m \u001b[0;34m\u001b[0m\u001b[0mNode\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mprepare_proxy_values\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0m_kwargs\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n", - "\u001b[0;31mIndexError\u001b[0m: index 10 is out of bounds for dimension 1 with size 10" - ] - } - ] - }, - { - "cell_type": "markdown", - "source": [ - "Ah of course, we needed to index at `hidden_dims - 1` not `hidden_dims`. How did `nnsight` know there was this indexing error before leaving the tracing context?\n", - "\n", - "Earlier when discussing contexts in Python, we learned some logic happens upon entering, and some logic happens upon exiting. We know the model is actually run on exit, but what happens on enter? Our input IS actually run though the model, however under its own \"fake\" context. This means the input makes its way through all of the model operations, allowing `nnsight` to record the shapes and data types of module inputs and outputs! The operations are never executed using tensors with real values so it doesn't incur any memory costs. Then, when creating proxy requests like the setting one above, `nnsight` also attempts to execute the request on the \"fake\" values we recorded. Hence, it lets us know if our request is feasible before even running the model.\n", - "\n", - "<details>\n", - "<summary>On scanning</summary>\n", - "\n", - "---\n", - "\n", - "\"Scanning\" is what we call running \"fake\" inputs throught the model to collect information like shapes and types. \"Validating\" is what we call trying to execute your intervention proxies with \"fake\" inputs to see if they work. If you are doing anything in a loop where efficiency is important, you should turn off scanning and validating. You can turn off validating in `.trace(...)` like `.trace(..., validate=False)`. You can turn off scanning in `Tracer.invoke(...)` ([see the Batching section](#batching-id)) like `Tracer.invoke(..., scan=False)`\n", - "\n", - "---\n", - "\n", - "</details>\n", - "\n", - "Let's try again with the correct indexing, and view the shape of the output before leaving the tracing context:" - ], - "metadata": { - "id": "JCNR_Jkpxr8l" - } - }, - { - "cell_type": "code", - "source": [ - "with model.trace(input):\n", - "\n", - " # Save the output before the edit to compare.\n", - " # Notice we apply .clone() before saving as the setting operation is in-place.\n", - " l1_output_before = model.layer1.output.clone().save()\n", - "\n", - " print(f\"layer1 output shape: {model.layer1.output.shape}\")\n", - "\n", - " # Access the last index of the hidden state dimension and set it to 0.\n", - " model.layer1.output[:, hidden_dims - 1] = 0\n", - "\n", - " # Save the output after to see our edit.\n", - " l1_output_after = model.layer1.output.save()\n", - "\n", - "print(\"Before:\", l1_output_before)\n", - "print(\"After:\", l1_output_after)" - ], - "metadata": { - "colab": { - "base_uri": "https://localhost:8080/" + "95d303d83d3348a0adfdf6b5cb94c41c": { + "model_module": "@jupyter-widgets/controls", + "model_module_version": "1.5.0", + "model_name": "HTMLModel", + "state": { + "_dom_classes": [], + "_model_module": "@jupyter-widgets/controls", + "_model_module_version": "1.5.0", + "_model_name": "HTMLModel", + "_view_count": null, + "_view_module": "@jupyter-widgets/controls", + "_view_module_version": "1.5.0", + "_view_name": "HTMLView", + "description": "", + "description_tooltip": null, + "layout": "IPY_MODEL_eb0c489b913744a28304e5334bb66532", + "placeholder": "", + "style": "IPY_MODEL_c80603555c634e338f16c0021f752420", + "value": "tokenizer.json: 100%" + } + }, + "96a3b8c765374a5e98eaf713dd5e85d8": { + "model_module": "@jupyter-widgets/controls", + "model_module_version": "1.5.0", + "model_name": "HTMLModel", + "state": { + "_dom_classes": [], + "_model_module": "@jupyter-widgets/controls", + "_model_module_version": "1.5.0", + "_model_name": "HTMLModel", + "_view_count": null, + "_view_module": "@jupyter-widgets/controls", + "_view_module_version": "1.5.0", + "_view_name": "HTMLView", + "description": "", + "description_tooltip": null, + "layout": "IPY_MODEL_8776d04bfd3748b69e10708466e9c132", + "placeholder": "", + "style": "IPY_MODEL_56490c20aef04fe4a63ba7c990e71e77", + "value": " 1.04M/1.04M [00:00<00:00, 3.16MB/s]" + } + }, + "9fbbb0bf7d8e4c12939193791efc590f": { + "model_module": "@jupyter-widgets/base", + "model_module_version": "1.2.0", + "model_name": "LayoutModel", + "state": { + "_model_module": "@jupyter-widgets/base", + "_model_module_version": "1.2.0", + "_model_name": "LayoutModel", + "_view_count": null, + "_view_module": "@jupyter-widgets/base", + "_view_module_version": "1.2.0", + "_view_name": "LayoutView", + "align_content": null, + "align_items": null, + "align_self": null, + "border": null, + "bottom": null, + "display": null, + "flex": null, + "flex_flow": null, + "grid_area": null, + "grid_auto_columns": null, + "grid_auto_flow": null, + "grid_auto_rows": null, + "grid_column": null, + "grid_gap": null, + "grid_row": null, + "grid_template_areas": null, + "grid_template_columns": null, + "grid_template_rows": null, + "height": null, + "justify_content": null, + "justify_items": null, + "left": null, + "margin": null, + "max_height": null, + "max_width": null, + "min_height": null, + "min_width": null, + "object_fit": null, + "object_position": null, + "order": null, + "overflow": null, + "overflow_x": null, + "overflow_y": null, + "padding": null, + "right": null, + "top": null, + "visibility": null, + "width": null + } + }, + "b2748df9aff540df954f002310606a56": { + "model_module": "@jupyter-widgets/base", + "model_module_version": "1.2.0", + "model_name": "LayoutModel", + "state": { + "_model_module": "@jupyter-widgets/base", + "_model_module_version": "1.2.0", + "_model_name": "LayoutModel", + "_view_count": null, + "_view_module": "@jupyter-widgets/base", + "_view_module_version": "1.2.0", + "_view_name": "LayoutView", + "align_content": null, + "align_items": null, + "align_self": null, + "border": null, + "bottom": null, + "display": null, + "flex": null, + "flex_flow": null, + "grid_area": null, + "grid_auto_columns": null, + "grid_auto_flow": null, + "grid_auto_rows": null, + "grid_column": null, + "grid_gap": null, + "grid_row": null, + "grid_template_areas": null, + "grid_template_columns": null, + "grid_template_rows": null, + "height": null, + "justify_content": null, + "justify_items": null, + "left": null, + "margin": null, + "max_height": null, + "max_width": null, + "min_height": null, + "min_width": null, + "object_fit": null, + "object_position": null, + "order": null, + "overflow": null, + "overflow_x": null, + "overflow_y": null, + "padding": null, + "right": null, + "top": null, + "visibility": null, + "width": null + } }, - "id": "Wf8ugJKB2mru", - "outputId": "4eed6c19-cb8c-4deb-e0e1-25aa80959c64" - }, - "execution_count": null, - "outputs": [ - { - "output_type": "stream", - "name": "stdout", - "text": [ - "layer1 output shape: torch.Size([1, 10])\n", - "Before: tensor([[ 0.0458, 0.5267, 0.7119, 0.4046, 0.2460, 0.7998, 0.4485, -0.2506,\n", - " 0.2968, -0.8834]])\n", - "After: tensor([[ 0.0458, 0.5267, 0.7119, 0.4046, 0.2460, 0.7998, 0.4485, -0.2506,\n", - " 0.2968, 0.0000]])\n" - ] - } - ] - }, - { - "cell_type": "markdown", - "source": [ - "We can also just replace proxy inputs and outputs with tensors of the same shape and type. Let's use the shape information we have at our disposal to add noise to the output, and replace it with this new noised tensor:" - ], - "metadata": { - "id": "9DH7PeL13-WU" - } - }, - { - "cell_type": "code", - "source": [ - "with model.trace(input):\n", - "\n", - " # Save the output before the edit to compare.\n", - " # Notice we apply .clone() before saving as the setting operation is in-place.\n", - " l1_output_before = model.layer1.output.clone().save()\n", - "\n", - " # Create random noise with variance of .001\n", - " noise = (0.0010.5)torch.randn(l1_output_before.shape)\n", - "\n", - " # Add to original value and replace.\n", - " model.layer1.output = l1_output_before + noise\n", - "\n", - " # Save the output after to see our edit.\n", - " l1_output_after = model.layer1.output.save()\n", - "\n", - "print(\"Before:\", l1_output_before)\n", - "print(\"After:\", l1_output_after)" - ], - "metadata": { - "colab": { - "base_uri": "https://localhost:8080/" + "b3078a7a41c642839748d9864220cbde": { + "model_module": "@jupyter-widgets/controls", + "model_module_version": "1.5.0", + "model_name": "HBoxModel", + "state": { + "_dom_classes": [], + "_model_module": "@jupyter-widgets/controls", + "_model_module_version": "1.5.0", + "_model_name": "HBoxModel", + "_view_count": null, + "_view_module": "@jupyter-widgets/controls", + "_view_module_version": "1.5.0", + "_view_name": "HBoxView", + "box_style": "", + "children": [ + "IPY_MODEL_3aa1aef302474a0cba55152e311acd40", + "IPY_MODEL_dda1704b991a44dfb2b28191f307e081", + "IPY_MODEL_5c0847c658d3489ca529d1afc8c96ed5" + ], + "layout": "IPY_MODEL_9fbbb0bf7d8e4c12939193791efc590f" + } }, - "id": "AP5wRmz_4YE7", - "outputId": "4c30dc1d-232d-4900-bfdd-0c93a9b23ad7" - }, - "execution_count": null, - "outputs": [ - { - "output_type": "stream", - "name": "stdout", - "text": [ - "Before: tensor([[ 0.0458, 0.5267, 0.7119, 0.4046, 0.2460, 0.7998, 0.4485, -0.2506,\n", - " 0.2968, -0.8834]])\n", - "After: tensor([[ 0.0581, 0.5168, 0.6561, 0.4083, 0.2617, 0.7800, 0.4080, -0.2213,\n", - " 0.3394, -0.9187]])\n" - ] - } - ] - }, - { - "cell_type": "markdown", - "source": [ - "## Gradients\n", - "\n", - "`NNsight` can also let you apply backprop and access gradients with respect to a loss. Like `.input` and `.output` on modules, `nnsight` also exposes `.grad` on Proxies themselves (assuming they are proxies of tensors):" - ], - "metadata": { - "id": "s016CelFP8sx" - } - }, - { - "cell_type": "code", - "source": [ - "with model.trace(input):\n", - "\n", - " # We need to explicitly have the tensor require grad\n", - " # as the model we defined earlier turned off requiring grad.\n", - " model.layer1.output.requires_grad = True\n", - "\n", - " # We call .grad on a tensor Proxy to communicate we want to store its gradient.\n", - " # We need to call .save() of course as .grad is its own Proxy.\n", - " layer1_output_grad = model.layer1.output.grad.save()\n", - " layer2_output_grad = model.layer2.output.grad.save()\n", - "\n", - " # Need a loss to propagate through the later modules in order to have a grad.\n", - " loss = model.output.sum()\n", - " loss.backward()\n", - "\n", - "print(\"Layer 1 output gradient:\", layer1_output_grad)\n", - "print(\"Layer 2 output gradient:\", layer2_output_grad)\n" - ], - "metadata": { - "colab": { - "base_uri": "https://localhost:8080/" + "b49ce0d00e5b4df097b3a9fc4c0bd8e7": { + "model_module": "@jupyter-widgets/base", + "model_module_version": "1.2.0", + "model_name": "LayoutModel", + "state": { + "_model_module": "@jupyter-widgets/base", + "_model_module_version": "1.2.0", + "_model_name": "LayoutModel", + "_view_count": null, + "_view_module": "@jupyter-widgets/base", + "_view_module_version": "1.2.0", + "_view_name": "LayoutView", + "align_content": null, + "align_items": null, + "align_self": null, + "border": null, + "bottom": null, + "display": null, + "flex": null, + "flex_flow": null, + "grid_area": null, + "grid_auto_columns": null, + "grid_auto_flow": null, + "grid_auto_rows": null, + "grid_column": null, + "grid_gap": null, + "grid_row": null, + "grid_template_areas": null, + "grid_template_columns": null, + "grid_template_rows": null, + "height": null, + "justify_content": null, + "justify_items": null, + "left": null, + "margin": null, + "max_height": null, + "max_width": null, + "min_height": null, + "min_width": null, + "object_fit": null, + "object_position": null, + "order": null, + "overflow": null, + "overflow_x": null, + "overflow_y": null, + "padding": null, + "right": null, + "top": null, + "visibility": null, + "width": null + } }, - "id": "L-cYUNNyRDn1", - "outputId": "d03dafa1-ef1e-40f1-de28-4dc2f77fe4db" - }, - "execution_count": null, - "outputs": [ - { - "output_type": "stream", - "name": "stdout", - "text": [ - "Layer 1 output gradient: tensor([[ 0.4545, -0.0596, -0.2059, 0.4643, -0.4211, -0.2813, 0.2126, 0.5016,\n", - " -0.0126, -0.1564]])\n", - "Layer 2 output gradient: tensor([[1., 1.]])\n" - ] - } - ] - }, - { - "cell_type": "markdown", - "source": [ - "All of the features we learned previously, also apply to `.grad`. In other words, we can apply operations to and edit the gradients. Let's zero the grad of `layer1` and double the grad of `layer2`." - ], - "metadata": { - "id": "1o0JaaYvWlHG" - } - }, - { - "cell_type": "code", - "source": [ - "with model.trace(input):\n", - "\n", - " # We need to explicitly have the tensor require grad\n", - " # as the model we defined earlier turned off requiring grad.\n", - " model.layer1.output.requires_grad = True\n", - "\n", - " model.layer1.output.grad[:] = 0\n", - " model.layer2.output.grad = model.layer2.output.grad.clone() * 2\n", - "\n", - " layer1_output_grad = model.layer1.output.grad.save()\n", - " layer2_output_grad = model.layer2.output.grad.save()\n", - "\n", - " # Need a loss to propagate through the later modules in order to have a grad.\n", - " loss = model.output.sum()\n", - " loss.backward()\n", - "\n", - "print(\"Layer 1 output gradient:\", layer1_output_grad)\n", - "print(\"Layer 2 output gradient:\", layer2_output_grad)" - ], - "metadata": { - "colab": { - "base_uri": "https://localhost:8080/" + "b7be20bd738940658f710309729d8ccd": { + "model_module": "@jupyter-widgets/controls", + "model_module_version": "1.5.0", + "model_name": "ProgressStyleModel", + "state": { + "_model_module": "@jupyter-widgets/controls", + "_model_module_version": "1.5.0", + "_model_name": "ProgressStyleModel", + "_view_count": null, + "_view_module": "@jupyter-widgets/base", + "_view_module_version": "1.2.0", + "_view_name": "StyleView", + "bar_color": null, + "description_width": "" + } }, - "id": "9bFoaJpOWlRb", - "outputId": "1c026534-9ffe-40e7-d016-e4565ddde8ac" - }, - "execution_count": null, - "outputs": [ - { - "output_type": "stream", - "name": "stdout", - "text": [ - "Layer 1 output gradient: tensor([[0., 0., 0., 0., 0., 0., 0., 0., 0., 0.]])\n", - "Layer 2 output gradient: tensor([[2., 2.]])\n" - ] - } - ] - }, - { - "cell_type": "markdown", - "source": [ - "# 2 Bigger" - ], - "metadata": { - "id": "gvQ1nZgYQDG3" - } - }, - { - "cell_type": "markdown", - "source": [ - "Now that we have the basics of `nnsight` under our belt, we can scale our model up and combine the techniques we've learned into more interesting experiments.\n", - "\n", - "The `NNsight` class is very bare bones. It wraps a pre-defined model and does no pre-processing on the inputs we enter. It's designed to be extended with more complex and powerful types of models and we're excited to see what can be done to leverage its features." - ], - "metadata": { - "id": "miQgUY4NAmDQ" - } - }, - { - "cell_type": "markdown", - "source": [ - "## LanguageModel" - ], - "metadata": { - "id": "5TJDblHiQpp1" - } - }, - { - "cell_type": "markdown", - "source": [ - "`LanguageModel` is a subclass of `NNsight`. While we could define and create a model to pass in directly, `LanguageModel` includes special support for Huggingface language models, including automatically loading models from a Huggingface ID, and loading the model together with the appropriate tokenizer.\n", - "\n", - "Here is how you can use `LanguageModel` to load `GPT-2`:" - ], - "metadata": { - "id": "9l9mZOY5HFH2" - } - }, - { - "cell_type": "code", - "source": [ - "from nnsight import LanguageModel\n", - "\n", - "model = LanguageModel('openai-community/gpt2', device_map=\"auto\")\n", - "\n", - "print(model)" - ], - "metadata": { - "id": "1OD2z7d3HQJU", - "colab": { - "base_uri": "https://localhost:8080/", - "height": 769, - "referenced_widgets": [ - "0a676b8f2dc34fed9731f19a03e4bfa1", - "74358c341daa44faa41a92b0eb073615", - "c40e17a994ed44d79ef54421f4772858", - "431ffcb0dbb1419dbbb237a29842b905", - "c28c185a508e459b95edb29afedd9c79", - "7a0f3d2deeb14bbba7f0df208133a7e2", - "94cdcf1fb0914dfe99c3c615e68b7ec8", - "f3e24aa6619e4185a3e2f53f2e473e45", - "b7be20bd738940658f710309729d8ccd", - "562a287082eb4882af864062ae13c91b", - "11e99924bdf84739b3cfc7d676b797e5", - "60fe026e908741209637604edfeb0563", - "e7a73cfc3fa04425afc08368532ef0be", - "d839da1c31a34659988e3420ba520725", - "96a3b8c765374a5e98eaf713dd5e85d8", - "b49ce0d00e5b4df097b3a9fc4c0bd8e7", - "b2748df9aff540df954f002310606a56", - "f4caf2bc58704464b6ac7bf6da31b71d", - "3731a5747724403e86ef1abaf4fb389b", - "688bf676958a4050b73cc9597f5110de", - "8776d04bfd3748b69e10708466e9c132", - "56490c20aef04fe4a63ba7c990e71e77", - "b8ad433cc3dd4a8689cb06655158d545", - "e51620404d244a4e806eeaf3b96a474f", - "e4568f72971749ac92cc5577cb9c4f9c", - "1a1ada8b122d4502949c4c75861a82bf", - "18d4c2302f424bc18f9215f2d5b83393", - "f28f557210b7402b8b2d0fd83f038dc1", - "bc8190c700e949ebb6f874c8f8055161", - "ceebda1d07b545f5b605551c08e73be8", - "6aa96b80e3b746b48f5c13bcafe1134e", - "8de63414803f41c29e75242756cfd67f", - "f693dfed91df492c986afbd4ff925f89", - "cef2516aabc24f6d852079984afa8090", - "95d303d83d3348a0adfdf6b5cb94c41c", - "9373fcd054014760b23e3acc928da0a5", - "7e6b41ec535648b190bfc655d9e430dd", - "d0198fa98e5f412380f2ba3e9448aef3", - "eb0c489b913744a28304e5334bb66532", - "c80603555c634e338f16c0021f752420", - "0fddebfb42e847e08890d09f30161984", - "d07a0a49a2924092affc0e07bf0b5ef0", - "beb4231b49024d8a9bb7ead19647d99d", - "26d09aa072d64af0bf4df9a89309093a" - ] + "b8ad433cc3dd4a8689cb06655158d545": { + "model_module": "@jupyter-widgets/controls", + "model_module_version": "1.5.0", + "model_name": "HBoxModel", + "state": { + "_dom_classes": [], + "_model_module": "@jupyter-widgets/controls", + "_model_module_version": "1.5.0", + "_model_name": "HBoxModel", + "_view_count": null, + "_view_module": "@jupyter-widgets/controls", + "_view_module_version": "1.5.0", + "_view_name": "HBoxView", + "box_style": "", + "children": [ + "IPY_MODEL_e51620404d244a4e806eeaf3b96a474f", + "IPY_MODEL_e4568f72971749ac92cc5577cb9c4f9c", + "IPY_MODEL_1a1ada8b122d4502949c4c75861a82bf" + ], + "layout": "IPY_MODEL_18d4c2302f424bc18f9215f2d5b83393" + } }, - "outputId": "7a53864c-6965-4c25-9668-4b0d56db9f6b" - }, - "execution_count": null, - "outputs": [ - { - "output_type": "stream", - "name": "stderr", - "text": [ - "/usr/local/lib/python3.10/dist-packages/huggingface_hub/utils/_token.py:88: UserWarning: \n", - "The secret `HF_TOKEN` does not exist in your Colab secrets.\n", - "To authenticate with the Hugging Face Hub, create a token in your settings tab (https://huggingface.co/settings/tokens), set it as secret in your Google Colab and restart your session.\n", - "You will be able to reuse this secret in all of your notebooks.\n", - "Please note that authentication is recommended but still optional to access public models or datasets.\n", - " warnings.warn(\n" - ] + "bc8190c700e949ebb6f874c8f8055161": { + "model_module": "@jupyter-widgets/controls", + "model_module_version": "1.5.0", + "model_name": "DescriptionStyleModel", + "state": { + "_model_module": "@jupyter-widgets/controls", + "_model_module_version": "1.5.0", + "_model_name": "DescriptionStyleModel", + "_view_count": null, + "_view_module": "@jupyter-widgets/base", + "_view_module_version": "1.2.0", + "_view_name": "StyleView", + "description_width": "" + } + }, + "beb4231b49024d8a9bb7ead19647d99d": { + "model_module": "@jupyter-widgets/base", + "model_module_version": "1.2.0", + "model_name": "LayoutModel", + "state": { + "_model_module": "@jupyter-widgets/base", + "_model_module_version": "1.2.0", + "_model_name": "LayoutModel", + "_view_count": null, + "_view_module": "@jupyter-widgets/base", + "_view_module_version": "1.2.0", + "_view_name": "LayoutView", + "align_content": null, + "align_items": null, + "align_self": null, + "border": null, + "bottom": null, + "display": null, + "flex": null, + "flex_flow": null, + "grid_area": null, + "grid_auto_columns": null, + "grid_auto_flow": null, + "grid_auto_rows": null, + "grid_column": null, + "grid_gap": null, + "grid_row": null, + "grid_template_areas": null, + "grid_template_columns": null, + "grid_template_rows": null, + "height": null, + "justify_content": null, + "justify_items": null, + "left": null, + "margin": null, + "max_height": null, + "max_width": null, + "min_height": null, + "min_width": null, + "object_fit": null, + "object_position": null, + "order": null, + "overflow": null, + "overflow_x": null, + "overflow_y": null, + "padding": null, + "right": null, + "top": null, + "visibility": null, + "width": null + } + }, + "c28c185a508e459b95edb29afedd9c79": { + "model_module": "@jupyter-widgets/base", + "model_module_version": "1.2.0", + "model_name": "LayoutModel", + "state": { + "_model_module": "@jupyter-widgets/base", + "_model_module_version": "1.2.0", + "_model_name": "LayoutModel", + "_view_count": null, + "_view_module": "@jupyter-widgets/base", + "_view_module_version": "1.2.0", + "_view_name": "LayoutView", + "align_content": null, + "align_items": null, + "align_self": null, + "border": null, + "bottom": null, + "display": null, + "flex": null, + "flex_flow": null, + "grid_area": null, + "grid_auto_columns": null, + "grid_auto_flow": null, + "grid_auto_rows": null, + "grid_column": null, + "grid_gap": null, + "grid_row": null, + "grid_template_areas": null, + "grid_template_columns": null, + "grid_template_rows": null, + "height": null, + "justify_content": null, + "justify_items": null, + "left": null, + "margin": null, + "max_height": null, + "max_width": null, + "min_height": null, + "min_width": null, + "object_fit": null, + "object_position": null, + "order": null, + "overflow": null, + "overflow_x": null, + "overflow_y": null, + "padding": null, + "right": null, + "top": null, + "visibility": null, + "width": null + } }, - { - "output_type": "display_data", - "data": { - "text/plain": [ - "config.json: 0%\| \| 0.00/665 [00:00<?, ?B/s]" - ], - "application/vnd.jupyter.widget-view+json": { - "version_major": 2, - "version_minor": 0, - "model_id": "0a676b8f2dc34fed9731f19a03e4bfa1" - } - }, - "metadata": {} + "c40e17a994ed44d79ef54421f4772858": { + "model_module": "@jupyter-widgets/controls", + "model_module_version": "1.5.0", + "model_name": "FloatProgressModel", + "state": { + "_dom_classes": [], + "_model_module": "@jupyter-widgets/controls", + "_model_module_version": "1.5.0", + "_model_name": "FloatProgressModel", + "_view_count": null, + "_view_module": "@jupyter-widgets/controls", + "_view_module_version": "1.5.0", + "_view_name": "ProgressView", + "bar_style": "success", + "description": "", + "description_tooltip": null, + "layout": "IPY_MODEL_f3e24aa6619e4185a3e2f53f2e473e45", + "max": 665, + "min": 0, + "orientation": "horizontal", + "style": "IPY_MODEL_b7be20bd738940658f710309729d8ccd", + "value": 665 + } }, - { - "output_type": "display_data", - "data": { - "text/plain": [ - "vocab.json: 0%\| \| 0.00/1.04M [00:00<?, ?B/s]" - ], - "application/vnd.jupyter.widget-view+json": { - "version_major": 2, - "version_minor": 0, - "model_id": "60fe026e908741209637604edfeb0563" - } - }, - "metadata": {} + "c80603555c634e338f16c0021f752420": { + "model_module": "@jupyter-widgets/controls", + "model_module_version": "1.5.0", + "model_name": "DescriptionStyleModel", + "state": { + "_model_module": "@jupyter-widgets/controls", + "_model_module_version": "1.5.0", + "_model_name": "DescriptionStyleModel", + "_view_count": null, + "_view_module": "@jupyter-widgets/base", + "_view_module_version": "1.2.0", + "_view_name": "StyleView", + "description_width": "" + } }, - { - "output_type": "display_data", - "data": { - "text/plain": [ - "merges.txt: 0%\| \| 0.00/456k [00:00<?, ?B/s]" - ], - "application/vnd.jupyter.widget-view+json": { - "version_major": 2, - "version_minor": 0, - "model_id": "b8ad433cc3dd4a8689cb06655158d545" - } - }, - "metadata": {} + "ceebda1d07b545f5b605551c08e73be8": { + "model_module": "@jupyter-widgets/base", + "model_module_version": "1.2.0", + "model_name": "LayoutModel", + "state": { + "_model_module": "@jupyter-widgets/base", + "_model_module_version": "1.2.0", + "_model_name": "LayoutModel", + "_view_count": null, + "_view_module": "@jupyter-widgets/base", + "_view_module_version": "1.2.0", + "_view_name": "LayoutView", + "align_content": null, + "align_items": null, + "align_self": null, + "border": null, + "bottom": null, + "display": null, + "flex": null, + "flex_flow": null, + "grid_area": null, + "grid_auto_columns": null, + "grid_auto_flow": null, + "grid_auto_rows": null, + "grid_column": null, + "grid_gap": null, + "grid_row": null, + "grid_template_areas": null, + "grid_template_columns": null, + "grid_template_rows": null, + "height": null, + "justify_content": null, + "justify_items": null, + "left": null, + "margin": null, + "max_height": null, + "max_width": null, + "min_height": null, + "min_width": null, + "object_fit": null, + "object_position": null, + "order": null, + "overflow": null, + "overflow_x": null, + "overflow_y": null, + "padding": null, + "right": null, + "top": null, + "visibility": null, + "width": null + } }, - { - "output_type": "display_data", - "data": { - "text/plain": [ - "tokenizer.json: 0%\| \| 0.00/1.36M [00:00<?, ?B/s]" + "cef2516aabc24f6d852079984afa8090": { + "model_module": "@jupyter-widgets/controls", + "model_module_version": "1.5.0", + "model_name": "HBoxModel", + "state": { + "_dom_classes": [], + "_model_module": "@jupyter-widgets/controls", + "_model_module_version": "1.5.0", + "_model_name": "HBoxModel", + "_view_count": null, + "_view_module": "@jupyter-widgets/controls", + "_view_module_version": "1.5.0", + "_view_name": "HBoxView", + "box_style": "", + "children": [ + "IPY_MODEL_95d303d83d3348a0adfdf6b5cb94c41c", + "IPY_MODEL_9373fcd054014760b23e3acc928da0a5", + "IPY_MODEL_7e6b41ec535648b190bfc655d9e430dd" ], - "application/vnd.jupyter.widget-view+json": { - "version_major": 2, - "version_minor": 0, - "model_id": "cef2516aabc24f6d852079984afa8090" - } - }, - "metadata": {} + "layout": "IPY_MODEL_d0198fa98e5f412380f2ba3e9448aef3" + } }, - { - "output_type": "stream", - "name": "stdout", - "text": [ - "GPT2LMHeadModel(\n", - " (transformer): GPT2Model(\n", - " (wte): Embedding(50257, 768)\n", - " (wpe): Embedding(1024, 768)\n", - " (drop): Dropout(p=0.1, inplace=False)\n", - " (h): ModuleList(\n", - " (0-11): 12 x GPT2Block(\n", - " (ln_1): LayerNorm((768,), eps=1e-05, elementwise_affine=True)\n", - " (attn): GPT2Attention(\n", - " (c_attn): Conv1D()\n", - " (c_proj): Conv1D()\n", - " (attn_dropout): Dropout(p=0.1, inplace=False)\n", - " (resid_dropout): Dropout(p=0.1, inplace=False)\n", - " )\n", - " (ln_2): LayerNorm((768,), eps=1e-05, elementwise_affine=True)\n", - " (mlp): GPT2MLP(\n", - " (c_fc): Conv1D()\n", - " (c_proj): Conv1D()\n", - " (act): NewGELUActivation()\n", - " (dropout): Dropout(p=0.1, inplace=False)\n", - " )\n", - " )\n", - " )\n", - " (ln_f): LayerNorm((768,), eps=1e-05, elementwise_affine=True)\n", - " )\n", - " (lm_head): Linear(in_features=768, out_features=50257, bias=False)\n", - " (generator): WrapperModule()\n", - ")\n" - ] - } - ] - }, - { - "cell_type": "markdown", - "source": [ - "<details>\n", - "<summary>On Model Initialization</summary>\n", - "\n", - "---\n", - "\n", - "A few important things to note:\n", - "\n", - " Keyword arguments passed to the initialization of `LanguageModel` is forwarded to HuggingFace specific loading logic. In this case, `device_map` specifies which devices to use and its value `auto` indicates to evenly distribute it to all available GPUs (and cpu if no GPUs available). Other arguments can be found here: https://huggingface.co/docs/transformers/model_doc/auto#transformers.AutoModelForCausalLM\n", - "\n", - "\n", - " When we initialize `LanguageModel`, we aren't yet loading the parameters of the model into memory. We are actually loading a 'meta' version of the model which doesn't take up any memory, but still allows us to view and trace actions on it. After exiting the first tracing context, the model is then fully loaded into memory. To load into memory on initialization, you can pass `dispatch=True` into `LanguageModel` like `LanguageModel('openai-community/gpt2', device_map=\"auto\", dispatch=True)`.\n", - "\n", - "---\n", - "\n", - "</details>\n", - "\n", - "\n", - "Let's put together some of the features we applied to the small model, but now on `GPT-2`. Unlike `NNsight`, `LanguageModel` does define logic to pre-process inputs upon entering the tracing context. This makes interacting with the model simpler without having to directly access the tokenizer.\n", - "\n", - "In the following example, we ablate the value coming from the last layer's MLP module and decode the logits to see what token the model predicts without influence from that particular module:" - ], - "metadata": { - "id": "Hw2VaAr_ezkj" - } - }, - { - "cell_type": "code", - "source": [ - "with model.trace('The Eiffel Tower is in the city of'):\n", - "\n", - " # Access the last layer using h[-1] as it's a ModuleList\n", - " # Access the first index of .output as that's where the hidden states are.\n", - " model.transformer.h[-1].mlp.output[0][:] = 0\n", - "\n", - " # Logits come out of model.lm_head and we apply argmax to get the predicted token ids.\n", - " token_ids = model.lm_head.output.argmax(dim=-1).save()\n", - "\n", - "print(\"Token IDs:\", token_ids)\n", - "\n", - "# Apply the tokenizer to decode the ids into words after the tracing context.\n", - "print(\"Prediction:\", model.tokenizer.decode(token_ids[0][-1]))\n" - ], - "metadata": { - "colab": { - "base_uri": "https://localhost:8080/" + "d0198fa98e5f412380f2ba3e9448aef3": { + "model_module": "@jupyter-widgets/base", + "model_module_version": "1.2.0", + "model_name": "LayoutModel", + "state": { + "_model_module": "@jupyter-widgets/base", + "_model_module_version": "1.2.0", + "_model_name": "LayoutModel", + "_view_count": null, + "_view_module": "@jupyter-widgets/base", + "_view_module_version": "1.2.0", + "_view_name": "LayoutView", + "align_content": null, + "align_items": null, + "align_self": null, + "border": null, + "bottom": null, + "display": null, + "flex": null, + "flex_flow": null, + "grid_area": null, + "grid_auto_columns": null, + "grid_auto_flow": null, + "grid_auto_rows": null, + "grid_column": null, + "grid_gap": null, + "grid_row": null, + "grid_template_areas": null, + "grid_template_columns": null, + "grid_template_rows": null, + "height": null, + "justify_content": null, + "justify_items": null, + "left": null, + "margin": null, + "max_height": null, + "max_width": null, + "min_height": null, + "min_width": null, + "object_fit": null, + "object_position": null, + "order": null, + "overflow": null, + "overflow_x": null, + "overflow_y": null, + "padding": null, + "right": null, + "top": null, + "visibility": null, + "width": null + } }, - "id": "mLSapaMCgLNU", - "outputId": "192b8998-ecca-4f25-ecf2-5196c0d2d672" - }, - "execution_count": null, - "outputs": [ - { - "output_type": "stream", - "name": "stderr", - "text": [ - "You're using a GPT2TokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n" - ] + "d07a0a49a2924092affc0e07bf0b5ef0": { + "model_module": "@jupyter-widgets/controls", + "model_module_version": "1.5.0", + "model_name": "ProgressStyleModel", + "state": { + "_model_module": "@jupyter-widgets/controls", + "_model_module_version": "1.5.0", + "_model_name": "ProgressStyleModel", + "_view_count": null, + "_view_module": "@jupyter-widgets/base", + "_view_module_version": "1.2.0", + "_view_name": "StyleView", + "bar_color": null, + "description_width": "" + } }, - { - "output_type": "stream", - "name": "stdout", - "text": [ - "Token IDs: tensor([[ 262, 12, 417, 8765, 11, 257, 262, 3504, 338, 3576]])\n", - "Prediction: London\n" - ] - } - ] - }, - { - "cell_type": "markdown", - "source": [ - "You just ran a little intervention on a much more complex model with a lot more parameters! An important piece of information we're missing though is what the prediction would look like without our ablation.\n", - "\n", - "Of course we could just run two tracing contexts and compare the outputs. This, however, would require two forward passes through the model. `NNsight` can do better than that." - ], - "metadata": { - "id": "X2-HiCQhkv-B" - } - }, - { - "cell_type": "markdown", - "source": [ - "<a name=\"batching-id\"></a>\n", - "## Batching" - ], - "metadata": { - "id": "vLjHnRyRPXjp" - } - }, - { - "cell_type": "markdown", - "source": [ - "It's time to bring back the `Tracer` object we dropped before. See, when you call `.trace(...)` with some input, it's actually creating two different contexts behind the scenes. The second one is the invoker context. Being within this context just means that `.input` and `.output` should refer only to the input you've given invoke. Calling `.trace(...)` with some input just means there's only one input and therefore only one invoker context.\n", - "\n", - "We can call `.trace()` without input and call `Tracer.invoke(...)` to manually create the invoker context with our input. Now every subsequent time we call `.invoke(...)`, new interventions will only refer to the input in that particular invoke. When exiting the tracing context, the inputs from all of the invokers will be batched together, and they will be executed in one forward pass! So let's do the ablation experiment, and compute a 'control' output to compare to:\n", - "\n", - "<details>\n", - "<summary>On the invoker context</summary>\n", - "\n", - "---\n", - "\n", - "Note that when injecting data to only the relevant invoker interventions, `nnsight` tries, but can't guarantee, that it can narrow the data into the right batch idxs (in the case of an object as input or output). So there are cases where all invokes will get all of the data.\n", - "\n", - "Just like `.trace(...)` created a `Tracer` object, `.invoke(...)` creates an `Invoker` object. The `Invoker` object has post-processed inputs at `invoker.inputs`, which can be useful for seeing information about your input. If you are using `.trace(...)` with inputs, you can still access the invoker object at `tracer._invoker`.\n", - "\n", - "Keyword arguments given to `.invoke(..)` make its way to the input pre-processing. For example in `LanguageModel`, the keyword arguments are used to tokenize like `max_length` and `truncation`. If you need to pass in keyword arguments directly to one input `.trace(...)`, you can pass an `invoker_args` keyword argument that should be a dictionary of keyword arguments for the invoker. `.trace(..., invoker_args={...})`\n", - "\n", - "---\n", - "\n", - "</details>" - ], - "metadata": { - "id": "_Gc-Zr6NmEOn" - } - }, - { - "cell_type": "code", - "source": [ - "with model.trace() as tracer:\n", - "\n", - " with tracer.invoke('The Eiffel Tower is in the city of'):\n", - "\n", - " # Ablate the last MLP for only this batch.\n", - " model.transformer.h[-1].mlp.output[0][:] = 0\n", - "\n", - " # Get the output for only the intervened on batch.\n", - " token_ids_intervention = model.lm_head.output.argmax(dim=-1).save()\n", - "\n", - " with tracer.invoke('The Eiffel Tower is in the city of'):\n", - "\n", - " # Get the output for only the original batch.\n", - " token_ids_original = model.lm_head.output.argmax(dim=-1).save()\n", - "\n", - "print(\"Original token IDs:\", token_ids_original)\n", - "print(\"Intervention token IDs:\", token_ids_intervention)\n", - "\n", - "print(\"Original prediction:\", model.tokenizer.decode(token_ids_original[0][-1]))\n", - "print(\"Intervention prediction:\", model.tokenizer.decode(token_ids_intervention[0][-1]))" - ], - "metadata": { - "colab": { - "base_uri": "https://localhost:8080/" + "d839da1c31a34659988e3420ba520725": { + "model_module": "@jupyter-widgets/controls", + "model_module_version": "1.5.0", + "model_name": "FloatProgressModel", + "state": { + "_dom_classes": [], + "_model_module": "@jupyter-widgets/controls", + "_model_module_version": "1.5.0", + "_model_name": "FloatProgressModel", + "_view_count": null, + "_view_module": "@jupyter-widgets/controls", + "_view_module_version": "1.5.0", + "_view_name": "ProgressView", + "bar_style": "success", + "description": "", + "description_tooltip": null, + "layout": "IPY_MODEL_3731a5747724403e86ef1abaf4fb389b", + "max": 1042301, + "min": 0, + "orientation": "horizontal", + "style": "IPY_MODEL_688bf676958a4050b73cc9597f5110de", + "value": 1042301 + } + }, + "dda1704b991a44dfb2b28191f307e081": { + "model_module": "@jupyter-widgets/controls", + "model_module_version": "1.5.0", + "model_name": "FloatProgressModel", + "state": { + "_dom_classes": [], + "_model_module": "@jupyter-widgets/controls", + "_model_module_version": "1.5.0", + "_model_name": "FloatProgressModel", + "_view_count": null, + "_view_module": "@jupyter-widgets/controls", + "_view_module_version": "1.5.0", + "_view_name": "ProgressView", + "bar_style": "success", + "description": "", + "description_tooltip": null, + "layout": "IPY_MODEL_21f5a81dd29549e69b5cfa2661371e35", + "max": 1, + "min": 0, + "orientation": "horizontal", + "style": "IPY_MODEL_5f608f4313d64e458314ace8237708b1", + "value": 0 + } + }, + "ded9b4b26ce14ba0aae0b19e0274fcd0": { + "model_module": "@jupyter-widgets/base", + "model_module_version": "1.2.0", + "model_name": "LayoutModel", + "state": { + "_model_module": "@jupyter-widgets/base", + "_model_module_version": "1.2.0", + "_model_name": "LayoutModel", + "_view_count": null, + "_view_module": "@jupyter-widgets/base", + "_view_module_version": "1.2.0", + "_view_name": "LayoutView", + "align_content": null, + "align_items": null, + "align_self": null, + "border": null, + "bottom": null, + "display": null, + "flex": null, + "flex_flow": null, + "grid_area": null, + "grid_auto_columns": null, + "grid_auto_flow": null, + "grid_auto_rows": null, + "grid_column": null, + "grid_gap": null, + "grid_row": null, + "grid_template_areas": null, + "grid_template_columns": null, + "grid_template_rows": null, + "height": null, + "justify_content": null, + "justify_items": null, + "left": null, + "margin": null, + "max_height": null, + "max_width": null, + "min_height": null, + "min_width": null, + "object_fit": null, + "object_position": null, + "order": null, + "overflow": null, + "overflow_x": null, + "overflow_y": null, + "padding": null, + "right": null, + "top": null, + "visibility": null, + "width": null + } + }, + "e4568f72971749ac92cc5577cb9c4f9c": { + "model_module": "@jupyter-widgets/controls", + "model_module_version": "1.5.0", + "model_name": "FloatProgressModel", + "state": { + "_dom_classes": [], + "_model_module": "@jupyter-widgets/controls", + "_model_module_version": "1.5.0", + "_model_name": "FloatProgressModel", + "_view_count": null, + "_view_module": "@jupyter-widgets/controls", + "_view_module_version": "1.5.0", + "_view_name": "ProgressView", + "bar_style": "success", + "description": "", + "description_tooltip": null, + "layout": "IPY_MODEL_ceebda1d07b545f5b605551c08e73be8", + "max": 456318, + "min": 0, + "orientation": "horizontal", + "style": "IPY_MODEL_6aa96b80e3b746b48f5c13bcafe1134e", + "value": 456318 + } }, - "id": "Kdcq4oCNmEua", - "outputId": "e6d33c07-d22a-41af-b84a-91c9681c7524" - }, - "execution_count": null, - "outputs": [ - { - "output_type": "stream", - "name": "stdout", - "text": [ - "Original token IDs: tensor([[ 198, 12, 417, 8765, 318, 257, 262, 3504, 7372, 6342]])\n", - "Intervention token IDs: tensor([[ 262, 12, 417, 8765, 11, 257, 262, 3504, 338, 3576]])\n", - "Original prediction: Paris\n", - "Intervention prediction: London\n" - ] - } - ] - }, - { - "cell_type": "markdown", - "source": [ - "So it did end up affecting what the model predicted. That's pretty neat!\n", - "\n", - "Another cool thing with multiple invokes is that the Proxies can interact between them. Here we transfer the word token embeddings from a real prompt into another placeholder prompt. Therefore the latter prompt produces the output of the former prompt:" - ], - "metadata": { - "id": "8BsybgsK2Bbr" - } - }, - { - "cell_type": "code", - "source": [ - "with model.trace() as tracer:\n", - "\n", - " with tracer.invoke(\"The Eiffel Tower is in the city of\"):\n", - "\n", - " embeddings = model.transformer.wte.output\n", - "\n", - " with tracer.invoke(\"_ _ _ _ _ _ _ _ _ _\"):\n", - "\n", - " model.transformer.wte.output = embeddings\n", - "\n", - " token_ids_intervention = model.lm_head.output.argmax(dim=-1).save()\n", - "\n", - " with tracer.invoke(\"_ _ _ _ _ _ _ _ _ _\"):\n", - "\n", - " token_ids_original = model.lm_head.output.argmax(dim=-1).save()\n", - "\n", - "print(\"Original prediction:\", model.tokenizer.decode(token_ids_original[0][-1]))\n", - "print(\"Intervention prediction:\", model.tokenizer.decode(token_ids_intervention[0][-1]))" - ], - "metadata": { - "colab": { - "base_uri": "https://localhost:8080/" + "e51620404d244a4e806eeaf3b96a474f": { + "model_module": "@jupyter-widgets/controls", + "model_module_version": "1.5.0", + "model_name": "HTMLModel", + "state": { + "_dom_classes": [], + "_model_module": "@jupyter-widgets/controls", + "_model_module_version": "1.5.0", + "_model_name": "HTMLModel", + "_view_count": null, + "_view_module": "@jupyter-widgets/controls", + "_view_module_version": "1.5.0", + "_view_name": "HTMLView", + "description": "", + "description_tooltip": null, + "layout": "IPY_MODEL_f28f557210b7402b8b2d0fd83f038dc1", + "placeholder": "", + "style": "IPY_MODEL_bc8190c700e949ebb6f874c8f8055161", + "value": "merges.txt: 100%" + } }, - "id": "syKTI_KhpvCY", - "outputId": "b2238c5c-d2d8-47bf-b73a-730f1e03cf99" - }, - "execution_count": null, - "outputs": [ - { - "output_type": "stream", - "name": "stdout", - "text": [ - "Original prediction: _\n", - "Intervention prediction: Paris\n" - ] - } - ] - }, - { - "cell_type": "markdown", - "source": [ - "## .next()\n", - "\n", - "Some HuggingFace models define methods to generate multiple outputs at a time. `LanguageModel` wraps that functionality to provide the same tracing features by using `.generate(...)` instead of `.trace(...)`. This calls the underlying model's `.generate` method. It passes the output through a `model.generator` module that we've added onto the model, allowing you to get the generate output at `model.generator.output`.\n", - "\n", - "In a case like this, the underlying model is called more than once; the modules of said model produce more than one output. Which iteration should a given `module.output` refer to? That's where `Module.next()` comes in.\n", - "\n", - "Each module has a call idx associated with it and `.next()` simply increments that attribute. At the time of execution, data is injected into the intervention graph only at the iteration that matches the call idx." - ], - "metadata": { - "id": "vvWA-CWqQtah" - } - }, - { - "cell_type": "code", - "source": [ - "with model.generate(\"The Eiffel Tower is in the city of\", max_new_tokens=3):\n", - "\n", - " token_ids_1 = model.lm_head.output.argmax(dim=-1).save()\n", - "\n", - " token_ids_2 = model.lm_head.next().output.argmax(dim=-1).save()\n", - "\n", - " token_ids_3 = model.lm_head.next().output.argmax(dim=-1).save()\n", - "\n", - " output = model.generator.output.save()\n", - "\n", - "print(\"Prediction 1: \", model.tokenizer.decode(token_ids_1[0][-1]))\n", - "print(\"Prediction 2: \", model.tokenizer.decode(token_ids_2[0][-1]))\n", - "print(\"Prediction 3: \", model.tokenizer.decode(token_ids_3[0][-1]))\n", - "\n", - "print(\"All token ids: \", output)\n", - "\n", - "print(\"All prediction: \", model.tokenizer.batch_decode(output))" - ], - "metadata": { - "colab": { - "base_uri": "https://localhost:8080/" + "e7a73cfc3fa04425afc08368532ef0be": { + "model_module": "@jupyter-widgets/controls", + "model_module_version": "1.5.0", + "model_name": "HTMLModel", + "state": { + "_dom_classes": [], + "_model_module": "@jupyter-widgets/controls", + "_model_module_version": "1.5.0", + "_model_name": "HTMLModel", + "_view_count": null, + "_view_module": "@jupyter-widgets/controls", + "_view_module_version": "1.5.0", + "_view_name": "HTMLView", + "description": "", + "description_tooltip": null, + "layout": "IPY_MODEL_b2748df9aff540df954f002310606a56", + "placeholder": "", + "style": "IPY_MODEL_f4caf2bc58704464b6ac7bf6da31b71d", + "value": "vocab.json: 100%" + } + }, + "eb0c489b913744a28304e5334bb66532": { + "model_module": "@jupyter-widgets/base", + "model_module_version": "1.2.0", + "model_name": "LayoutModel", + "state": { + "_model_module": "@jupyter-widgets/base", + "_model_module_version": "1.2.0", + "_model_name": "LayoutModel", + "_view_count": null, + "_view_module": "@jupyter-widgets/base", + "_view_module_version": "1.2.0", + "_view_name": "LayoutView", + "align_content": null, + "align_items": null, + "align_self": null, + "border": null, + "bottom": null, + "display": null, + "flex": null, + "flex_flow": null, + "grid_area": null, + "grid_auto_columns": null, + "grid_auto_flow": null, + "grid_auto_rows": null, + "grid_column": null, + "grid_gap": null, + "grid_row": null, + "grid_template_areas": null, + "grid_template_columns": null, + "grid_template_rows": null, + "height": null, + "justify_content": null, + "justify_items": null, + "left": null, + "margin": null, + "max_height": null, + "max_width": null, + "min_height": null, + "min_width": null, + "object_fit": null, + "object_position": null, + "order": null, + "overflow": null, + "overflow_x": null, + "overflow_y": null, + "padding": null, + "right": null, + "top": null, + "visibility": null, + "width": null + } + }, + "f28f557210b7402b8b2d0fd83f038dc1": { + "model_module": "@jupyter-widgets/base", + "model_module_version": "1.2.0", + "model_name": "LayoutModel", + "state": { + "_model_module": "@jupyter-widgets/base", + "_model_module_version": "1.2.0", + "_model_name": "LayoutModel", + "_view_count": null, + "_view_module": "@jupyter-widgets/base", + "_view_module_version": "1.2.0", + "_view_name": "LayoutView", + "align_content": null, + "align_items": null, + "align_self": null, + "border": null, + "bottom": null, + "display": null, + "flex": null, + "flex_flow": null, + "grid_area": null, + "grid_auto_columns": null, + "grid_auto_flow": null, + "grid_auto_rows": null, + "grid_column": null, + "grid_gap": null, + "grid_row": null, + "grid_template_areas": null, + "grid_template_columns": null, + "grid_template_rows": null, + "height": null, + "justify_content": null, + "justify_items": null, + "left": null, + "margin": null, + "max_height": null, + "max_width": null, + "min_height": null, + "min_width": null, + "object_fit": null, + "object_position": null, + "order": null, + "overflow": null, + "overflow_x": null, + "overflow_y": null, + "padding": null, + "right": null, + "top": null, + "visibility": null, + "width": null + } }, - "id": "Yy5Z9NE1GkaN", - "outputId": "0b4c1bd3-a6f6-4ba1-9cb8-6019c222311f" - }, - "execution_count": null, - "outputs": [ - { - "output_type": "stream", - "name": "stderr", - "text": [ - "Setting `pad_token_id` to `eos_token_id`:50256 for open-end generation.\n", - "A decoder-only architecture is being used, but right-padding was detected! For correct generation results, please set `padding_side='left'` when initializing the tokenizer.\n", - "Setting `pad_token_id` to `eos_token_id`:50256 for open-end generation.\n" - ] + "f3e24aa6619e4185a3e2f53f2e473e45": { + "model_module": "@jupyter-widgets/base", + "model_module_version": "1.2.0", + "model_name": "LayoutModel", + "state": { + "_model_module": "@jupyter-widgets/base", + "_model_module_version": "1.2.0", + "_model_name": "LayoutModel", + "_view_count": null, + "_view_module": "@jupyter-widgets/base", + "_view_module_version": "1.2.0", + "_view_name": "LayoutView", + "align_content": null, + "align_items": null, + "align_self": null, + "border": null, + "bottom": null, + "display": null, + "flex": null, + "flex_flow": null, + "grid_area": null, + "grid_auto_columns": null, + "grid_auto_flow": null, + "grid_auto_rows": null, + "grid_column": null, + "grid_gap": null, + "grid_row": null, + "grid_template_areas": null, + "grid_template_columns": null, + "grid_template_rows": null, + "height": null, + "justify_content": null, + "justify_items": null, + "left": null, + "margin": null, + "max_height": null, + "max_width": null, + "min_height": null, + "min_width": null, + "object_fit": null, + "object_position": null, + "order": null, + "overflow": null, + "overflow_x": null, + "overflow_y": null, + "padding": null, + "right": null, + "top": null, + "visibility": null, + "width": null + } }, - { - "output_type": "stream", - "name": "stdout", - "text": [ - "Prediction 1: Paris\n", - "Prediction 2: ,\n", - "Prediction 3: and\n", - "All token ids: tensor([[ 464, 412, 733, 417, 8765, 318, 287, 262, 1748, 286, 6342, 11,\n", - " 290]])\n", - "All prediction: ['The Eiffel Tower is in the city of Paris, and']\n" - ] + "f4caf2bc58704464b6ac7bf6da31b71d": { + "model_module": "@jupyter-widgets/controls", + "model_module_version": "1.5.0", + "model_name": "DescriptionStyleModel", + "state": { + "_model_module": "@jupyter-widgets/controls", + "_model_module_version": "1.5.0", + "_model_name": "DescriptionStyleModel", + "_view_count": null, + "_view_module": "@jupyter-widgets/base", + "_view_module_version": "1.2.0", + "_view_name": "StyleView", + "description_width": "" + } + }, + "f693dfed91df492c986afbd4ff925f89": { + "model_module": "@jupyter-widgets/controls", + "model_module_version": "1.5.0", + "model_name": "DescriptionStyleModel", + "state": { + "_model_module": "@jupyter-widgets/controls", + "_model_module_version": "1.5.0", + "_model_name": "DescriptionStyleModel", + "_view_count": null, + "_view_module": "@jupyter-widgets/base", + "_view_module_version": "1.2.0", + "_view_name": "StyleView", + "description_width": "" + } } - ] - }, - { - "cell_type": "markdown", - "source": [ - "# 3 I thought you said huge models?" - ], - "metadata": { - "id": "rpkX-LwBQZHo" - } - }, - { - "cell_type": "markdown", - "source": [ - "`NNsight` is only one part of our project to democratize access to AI internals. The other half is `NDIF` (National Deep Inference Facility).\n", - "\n", - "The interaction between the two is fairly straightforward. The `intervention graph` we create via the tracing context can be encoded into a custom json format and sent via an http request to the `NDIF` servers. `NDIF` then decodes the `intervention graph` and `interleaves` it alongside the specified model.\n", - "\n", - "To see which models are currently being hosted, check out the following status page: https://nnsight.net/status/" - ], - "metadata": { - "id": "uCZR65VmILEb" - } - }, - { - "cell_type": "markdown", - "source": [ - "## Remote execution" - ], - "metadata": { - "id": "65Ks1LUvQaER" - } - }, - { - "cell_type": "markdown", - "source": [ - "In its current state, `NDIF` requires you to recieve an API key. Therefore, to run the rest of this colab, you would need one of your own. To get one, simply join the [NDIF discord](https://discord.gg/6uFJmCSwW7) and introduce yourself on the `#introductions` channel. Then DM either @JadenFK or @caden and we'll create one for you.\n", - "\n", - "Once you have one, to register your api key with `nnsight`, do the following:" - ], - "metadata": { - "id": "Aa-ZuVOFKS5k" - } - }, - { - "cell_type": "code", - "source": [ - "from nnsight import CONFIG\n", - "\n", - "CONFIG.set_default_api_key(\"<your api key here>\")" - ], - "metadata": { - "id": "ax1NWoS9MJeZ" - }, - "execution_count": null, - "outputs": [] - }, - { - "cell_type": "markdown", - "source": [ - "This only needs to be run once as it will save this api key as the default in a config file along with the `nnsight` installation.\n", - "\n", - "To amp things up a few levels, let's demonstrate using `nnsight`'s tracing context with one of the larger open source language models, `Llama-2-70b`!" - ], - "metadata": { - "id": "mvXLOh65MXmF" - } - }, - { - "cell_type": "code", - "source": [ - "import os\n", - "\n", - "# llama2 70b is a gated model and you need access via your huggingface token\n", - "os.environ['HF_TOKEN'] = \"<your huggingface token>\"\n", - "\n", - "# llama response object requires the version of transformers from github\n", - "!pip uninstall -y transformers\n", - "!pip install git+https://github.com/huggingface/transformers\n", - "\n", - "clear_output()" - ], - "metadata": { - "id": "Q1N04sJPZnJt" - }, - "execution_count": null, - "outputs": [] - }, - { - "cell_type": "code", - "source": [ - "# We'll never actually load the parameters so no need to specify a device_map.\n", - "model = LanguageModel(\"meta-llama/Llama-2-70b-hf\")\n", - "\n", - "# All we need to specify using NDIF vs executing locally is remote=True.\n", - "with model.trace('The Eiffel Tower is in the city of', remote=True) as runner:\n", - "\n", - " hidden_states = model.model.layers[-1].output.save()\n", - "\n", - " output = model.output.save()\n", - "\n", - "print(hidden_states)\n", - "\n", - "print(output['logits'])" - ], - "metadata": { - "id": "oqVzjNoyNGc7" - }, - "execution_count": null, - "outputs": [] - }, - { - "cell_type": "markdown", - "source": [ - "It really is as simple as `remote=True`. All of the techniques we went through in earlier sections work just the same when running locally and remotely.\n", - "\n", - "Note that both `nnsight`, but especially `NDIF`, is in active development and therefore there may be caveats, changes, and errors to work through." - ], - "metadata": { - "id": "IceVpnZcZ9F0" - } - }, - { - "cell_type": "markdown", - "source": [ - "# Getting Involved!" - ], - "metadata": { - "id": "P3zRm-7VRRov" - } - }, - { - "cell_type": "markdown", - "source": [ - "If you're interested in following updates to `nnsight`, contributing, giving feedback, or finding collaborators, please join the [NDIF discord](https://discord.gg/6uFJmCSwW7)!\n", - "\n", - "The [Mech Interp discord](https://discord.gg/km2RQBzaUn) is also a fantastic place to discuss all things mech interp with a really cool community.\n", - "\n", - "Our website [nnsight.net](https://nnsight.net/), has a bunch more tutorials detailing more complex interpretability techniques using `nnsight`. If you want to share any of the work you do using `nnsight`, let others know on either of the discords above and we might turn it into a tutorial on our website.\n", - "\n", - "💟\n", - "\n", - "\n", - "\n" - ], - "metadata": { - "id": "BnCc9xgjvxEP" } } - ] -} \ No newline at end of file + }, + "nbformat": 4, + "nbformat_minor": 0 +} diff --git a/README.md b/README.md index 195f14e4..fc8d9b8c 100644 --- a/README.md +++ b/README.md @@ -11,7 +11,7 @@ The `nnsight` package enables interpreting and manipulating the internals of de Install this package through pip by running: -`pip install nnsight==0.2.3.dev0` +`pip install nnsight` #### Examples diff --git a/docs/.DS_Store b/docs/.DS_Store index a7839fe9..78f1ef6b 100644 Binary files a/docs/.DS_Store and b/docs/.DS_Store differ diff --git a/docs/contributing.md b/docs/contributing.md new file mode 100644 index 00000000..9af41725 --- /dev/null +++ b/docs/contributing.md @@ -0,0 +1,27 @@ +# Installation + +Run `pip install -r requirements.txt` while in the `nnsight/docs` directory. + +# Adding Tutorials + +Tutorials are written as Jupyter Notebooks in `.ipynb` format, and automatically converted to html by the `nbsphinx` extension. + +The only requirement of `nbsphinx` is that you add a header to the notebook to act as the rendered title on Sphinx docs. To do this, create a markdown cell at the top of your notebook with a header `#`. + +Then, just add your notebook to the `nnsight/docs/source/notebooks/tutorials` directory. + +If you're adding a new notebook, navigate to `nnsight/docs/source/tutorials.rst`. At the bottom of the page, add the path to your notebook under the `toctree`. + +``` +.. toctree:: + :maxdepth: 1 + + notebooks/tutorials/walkthrough.ipynb + ... + <YOUR NOTEBOOK PATH> + +``` + +# Compiling Sphinx to HTML + +Run `make dirhtml` from the `nnsight/docs` directory. The build is located in `nnsight/docs/build/dirhtml`. You can run `python3 -m http.server <PORT>` from that directory, and see the site at `http://localhost:<PORT>`. diff --git a/docs/requirements.txt b/docs/requirements.txt index a49f80f4..cf864581 100644 --- a/docs/requirements.txt +++ b/docs/requirements.txt @@ -1,6 +1,8 @@ nnsight pydata_sphinx_theme sphinx -sphinx_charts sphinx-copybutton +sphinx-design +nbsphinx plotly +ipython \ No newline at end of file diff --git a/docs/source/.DS_Store b/docs/source/.DS_Store index d39d1092..836f5368 100644 Binary files a/docs/source/.DS_Store and b/docs/source/.DS_Store differ diff --git a/docs/source/_static/.DS_Store b/docs/source/_static/.DS_Store index 3548ed0e..9ea5ea8f 100644 Binary files a/docs/source/_static/.DS_Store and b/docs/source/_static/.DS_Store differ diff --git a/docs/source/_static/css/custom.css b/docs/source/_static/css/custom.css index a5ece848..db595237 100644 --- a/docs/source/_static/css/custom.css +++ b/docs/source/_static/css/custom.css @@ -1,3 +1,5 @@ +/* Website light and dark theme colors / + html[data-theme="dark"] { --pst-color-secondary: #ce5d97; --pst-color-primary: #4385be; @@ -13,10 +15,29 @@ html[data-theme="dark"] { --pst-color-panel-background: #1C1B1A; } +html[data-theme="light"] { + --pst-color-secondary: #ce5d97; + --pst-color-primary: #4385be; + --pst-color-primary-highlight: #ce5d97; + --pst-color-background: #FFFCF0; + --pst-color-on-background: #F2F0E5; + --pst-color-surface: #E6E4D9; + --pst-color-on-surface: #100F0F; + --pst-color-text-base: #100F0F; + --pst-color-text-muted: #100F0F; + --pst-color-heading-color: #100F0F; + --pst-color-border: #F2F0E5; + --pst-color-panel-background: #F2F0E5; +} + +/ Editing the default width for pydata content / + .bd-main .bd-content .bd-article-container { max-width: 80%; / default is 60em / } +/ Front page buttons and hero / + .hero { height: 60vh !important; } @@ -39,34 +60,21 @@ html[data-theme="dark"] { color: var(--pst-color-on-surface) !important; } -.custom-card { - background-color : var(--pst-color-on-background) !important; - border-color : #878580 !important; - color: #878580 !important; -} +/ Hide the theme version note at the bottom of the page / -.custom-card:hover, -.custom-card:hover h5, -.custom-card:hover p { - border-color: var(--pst-color-on-surface) !important; - color: var(--pst-color-on-surface) !important; +.theme-version { + visibility: hidden; } - - .text { color : var(--pst-color-on-surface) !important; } +/ Default surface style for cards / + .surface { background-color : var(--pst-color-surface) !important; color : var(--pst-color-on-surface) !important; border-color: var(--pst-color-surface) !important; } - -pre { - border-color: var(--pst-color-surface) !important; -} - - diff --git a/docs/source/_static/css/front.css b/docs/source/_static/css/front.css index eb29888f..db44e323 100644 --- a/docs/source/_static/css/front.css +++ b/docs/source/_static/css/front.css @@ -1,26 +1,19 @@ / Custom Styles for List Group Items / .list-group-item { background-color: transparent; - / Make the button background transparent / border: none; - / Remove border / color: var(--pst-color-on-surface); - / Set text color / } .list-group-item:hover { background-color: var(--pst-color-on-background); color: var(--pst-color-on-surface); - / Light grey hover effect, adjust as needed / } .list-group-item.active { background-color: var(--pst-color-surface); - / White background for active state / color: var(--pst-color-on-surface); - / Text color for active state / border: none; - / Ensure no border for active state / } .tab-background { @@ -34,36 +27,23 @@ height: 100% !important; } -@media (max-width: 767px) { / Targets screens smaller than the md breakpoint / - .mid-section { - height: 50vh !important; - } - .static-on-small { - height: 35vh !important; - } -} - - +/ Middle section blurb / .mid-section { + margin-top: 5vh; height: 20vh; -} - -.middle-stuff { margin-bottom: 15vh; } + +/ Custom Styles for Code Widget / .code-surface { background-color : var(--pst-color-surface) !important; color : var(--pst-color-on-surface) !important; border-color: var(--pst-color-surface) !important; } - - -/ Custom Styles for Code Styles / - .fill-height { height: 100% !important; } @@ -71,20 +51,3 @@ pre { margin: 0 !important; } - - -.hover-border { - transition: border 0.3s ease-in-out; -} - -.hover-border:hover { - border: 1px solid #fff; / Adjust color to match the design / -} - - -.center-text { - display: flex; - justify-content: center; - align-items: center; - text-align: center; -} diff --git a/docs/source/_static/css/hljs.css b/docs/source/_static/css/hljs-dark.css similarity index 94% rename from docs/source/_static/css/hljs.css rename to docs/source/_static/css/hljs-dark.css index a5b3fdf9..82abee3b 100644 --- a/docs/source/_static/css/hljs.css +++ b/docs/source/_static/css/hljs-dark.css @@ -1,3 +1,5 @@ +/ Custom Highlight.js styles for the front page snippet / + .hljs { display: block !important; overflow-x: auto !important; diff --git a/docs/source/_static/css/hljs-light.css b/docs/source/_static/css/hljs-light.css new file mode 100644 index 00000000..c04c8e42 --- /dev/null +++ b/docs/source/_static/css/hljs-light.css @@ -0,0 +1,84 @@ +/ + +Orginal Style from ethanschoonover.com/solarized (c) Jeremy Hull <[email protected]> + +/ + +.hljs { + display: block; + overflow-x: auto; + padding: 0.5em; + background: var(--pst-color-surface); + color: var(--pst-color-on-surface); + } + + .hljs-comment, + .hljs-quote { + color: #93a1a1; + } + + / Solarized Green / + .hljs-keyword, + .hljs-selector-tag, + .hljs-addition { + color: #859900; + } + + / Solarized Cyan / + .hljs-number, + .hljs-string, + .hljs-meta .hljs-meta-string, + .hljs-literal, + .hljs-doctag, + .hljs-regexp { + color: #2aa198; + } + + / Solarized Blue / + .hljs-title, + .hljs-section, + .hljs-name, + .hljs-selector-id, + .hljs-selector-class { + color: #268bd2; + } + + / Solarized Yellow / + .hljs-attribute, + .hljs-attr, + .hljs-variable, + .hljs-template-variable, + .hljs-class .hljs-title, + .hljs-type { + color: #b58900; + } + + / Solarized Orange / + .hljs-symbol, + .hljs-bullet, + .hljs-subst, + .hljs-meta, + .hljs-meta .hljs-keyword, + .hljs-selector-attr, + .hljs-selector-pseudo, + .hljs-link { + color: #cb4b16; + } + + / Solarized Red / + .hljs-built_in, + .hljs-deletion { + color: #dc322f; + } + + .hljs-formula { + background: #eee8d5; + } + + .hljs-emphasis { + font-style: italic; + } + + .hljs-strong { + font-weight: bold; + } \ No newline at end of file diff --git a/docs/source/_static/css/status.css b/docs/source/_static/css/status.css index 3979c12c..7f5f243f 100644 --- a/docs/source/_static/css/status.css +++ b/docs/source/_static/css/status.css @@ -1,7 +1,10 @@ -.status-body { - color: "white"; -} - -.accordion-header { - margin: 0 !important; -} \ No newline at end of file +.custom-accordion-header { + background-color: #007bff; / Bootstrap's primary color for demonstration / + color: white; / Text color / + } + + .custom-accordion-body { + background-color: #f8f9fa; / A light gray background / + color: #212529; / Bootstrap default text color / + } + \ No newline at end of file diff --git a/docs/source/_static/css/tutorials.css b/docs/source/_static/css/tutorials.css deleted file mode 100644 index 2349f603..00000000 --- a/docs/source/_static/css/tutorials.css +++ /dev/null @@ -1,16 +0,0 @@ -.tutorials-cards { - font-size: .9rem; -} - -.tutorial-card-section { - overflow: scroll; - height: 50vh; -} - -.tutorial-card-section::-webkit-scrollbar { - display: none; -} - -.page-info { - z-index: 100; -} \ No newline at end of file diff --git a/docs/source/_static/images/.DS_Store b/docs/source/_static/images/.DS_Store index c671bc88..1847e05f 100644 Binary files a/docs/source/_static/images/.DS_Store and b/docs/source/_static/images/.DS_Store differ diff --git a/docs/source/_static/images/execution.png b/docs/source/_static/images/execution.png new file mode 100644 index 00000000..f9094444 Binary files /dev/null and b/docs/source/_static/images/execution.png differ diff --git a/docs/source/_static/images/interleaved.png b/docs/source/_static/images/interleaved.png new file mode 100644 index 00000000..6d17f285 Binary files /dev/null and b/docs/source/_static/images/interleaved.png differ diff --git a/docs/source/_static/images/intrgraph.png b/docs/source/_static/images/intrgraph.png deleted file mode 100644 index 3bb50229..00000000 Binary files a/docs/source/_static/images/intrgraph.png and /dev/null differ diff --git a/docs/source/_static/images/remote_execution.png b/docs/source/_static/images/remote_execution.png deleted file mode 100644 index 7020d5b9..00000000 Binary files a/docs/source/_static/images/remote_execution.png and /dev/null differ diff --git a/docs/source/_static/images/tutorials/.DS_Store b/docs/source/_static/images/tutorials/.DS_Store index ba36697a..cc81a870 100644 Binary files a/docs/source/_static/images/tutorials/.DS_Store and b/docs/source/_static/images/tutorials/.DS_Store differ diff --git a/docs/source/_static/images/tutorials/activation_patching.png b/docs/source/_static/images/tutorials/activation_patching.png deleted file mode 100644 index 4f82b933..00000000 Binary files a/docs/source/_static/images/tutorials/activation_patching.png and /dev/null differ diff --git a/docs/source/_static/images/tutorials/activation_patching.webp b/docs/source/_static/images/tutorials/activation_patching.webp new file mode 100644 index 00000000..05604c7f Binary files /dev/null and b/docs/source/_static/images/tutorials/activation_patching.webp differ diff --git a/docs/source/_static/images/tutorials/attribution_patching.png b/docs/source/_static/images/tutorials/attribution_patching.png deleted file mode 100644 index 4f012886..00000000 Binary files a/docs/source/_static/images/tutorials/attribution_patching.png and /dev/null differ diff --git a/docs/source/_static/images/tutorials/attribution_patching.webp b/docs/source/_static/images/tutorials/attribution_patching.webp new file mode 100644 index 00000000..abd71d1d Binary files /dev/null and b/docs/source/_static/images/tutorials/attribution_patching.webp differ diff --git a/docs/source/_static/images/tutorials/dictionary_learning.png b/docs/source/_static/images/tutorials/dictionary_learning.png deleted file mode 100644 index 40440864..00000000 Binary files a/docs/source/_static/images/tutorials/dictionary_learning.png and /dev/null differ diff --git a/docs/source/_static/images/tutorials/dictionary_learning.webp b/docs/source/_static/images/tutorials/dictionary_learning.webp new file mode 100644 index 00000000..00a4cb45 Binary files /dev/null and b/docs/source/_static/images/tutorials/dictionary_learning.webp differ diff --git a/docs/source/_static/images/tutorials/function_vectors.png b/docs/source/_static/images/tutorials/function_vectors.png deleted file mode 100644 index bee6d3fa..00000000 Binary files a/docs/source/_static/images/tutorials/function_vectors.png and /dev/null differ diff --git a/docs/source/_static/images/tutorials/function_vectors.webp b/docs/source/_static/images/tutorials/function_vectors.webp new file mode 100644 index 00000000..041b1db0 Binary files /dev/null and b/docs/source/_static/images/tutorials/function_vectors.webp differ diff --git a/docs/source/_static/images/tutorials/future_lens.png b/docs/source/_static/images/tutorials/future_lens.png deleted file mode 100644 index 4d7e0cf3..00000000 Binary files a/docs/source/_static/images/tutorials/future_lens.png and /dev/null differ diff --git a/docs/source/_static/images/tutorials/future_lens.webp b/docs/source/_static/images/tutorials/future_lens.webp new file mode 100644 index 00000000..19e29c79 Binary files /dev/null and b/docs/source/_static/images/tutorials/future_lens.webp differ diff --git a/docs/source/_static/images/tutorials/logit_lens.png b/docs/source/_static/images/tutorials/logit_lens.png deleted file mode 100644 index da8af979..00000000 Binary files a/docs/source/_static/images/tutorials/logit_lens.png and /dev/null differ diff --git a/docs/source/_static/images/tutorials/logit_lens.webp b/docs/source/_static/images/tutorials/logit_lens.webp new file mode 100644 index 00000000..24d1beca Binary files /dev/null and b/docs/source/_static/images/tutorials/logit_lens.webp differ diff --git a/docs/source/_static/images/tutorials/walkthrough.png b/docs/source/_static/images/tutorials/walkthrough.png deleted file mode 100644 index c07ac8f4..00000000 Binary files a/docs/source/_static/images/tutorials/walkthrough.png and /dev/null differ diff --git a/docs/source/_static/images/tutorials/walkthrough.webp b/docs/source/_static/images/tutorials/walkthrough.webp new file mode 100644 index 00000000..15e5b375 Binary files /dev/null and b/docs/source/_static/images/tutorials/walkthrough.webp differ diff --git a/docs/source/_static/js/code.js b/docs/source/_static/js/code.js index 0d18c802..c4b8da0e 100644 --- a/docs/source/_static/js/code.js +++ b/docs/source/_static/js/code.js @@ -46,4 +46,37 @@ document.addEventListener('DOMContentLoaded', (event) => { document.querySelectorAll('code.language-python.multi').forEach(el => { el.textContent = multi; }); + document.querySelectorAll('div.output_area.stderr').forEach(el => { + el.style.visibility = "hidden"; + el.style.display = "none"; + }); + document.querySelectorAll('div.output_area.stderr').forEach(el => { + el.style.visibility = "hidden"; + el.style.display = "none"; + }); + document.querySelectorAll('div.output_area').forEach(el => { + el.style.marginBottom = "1em"; + }); + document.querySelectorAll('div.input_area').forEach(el => { + el.style.border = "0px"; + el.style.marginTop = "0.5em"; + el.style.marginBottom = "0.5em"; + }); + document.querySelectorAll('summary').forEach(el => { + el.style.marginTop = "1em"; + el.style.marginBottom = "1em"; + document.querySelectorAll('p').forEach(p => { + const span = document.createElement('span'); + + // Copy the innerHTML from the <p> to the <span> + span.innerHTML = p.innerHTML; + + // Optional: Copy styles from the <p> to the <span> + span.style.cssText = p.style.cssText; + + // Replace the <p> with the <span> in the DOM + p.parentNode.replaceChild(span, p); + }); + }); }); + diff --git a/docs/source/_static/js/custom.js b/docs/source/_static/js/custom.js index a72b393d..8d712853 100644 --- a/docs/source/_static/js/custom.js +++ b/docs/source/_static/js/custom.js @@ -1,85 +1,115 @@ require.config({ - paths: { - 'three': 'https://cdnjs.cloudflare.com/ajax/libs/three.js/r134/three.min.js?MEH=BLAH', - 'VANTA': 'https://cdn.jsdelivr.net/npm/[email protected]/dist/vanta.dots.min.js?MEH=BLAH' - }, - shim: { - 'VANTA': { - deps: ['three'] // Removed the exports line - } + paths: { + 'three': 'https://cdnjs.cloudflare.com/ajax/libs/three.js/r134/three.min.js?MEH=BLAH', + 'VANTA': 'https://cdn.jsdelivr.net/npm/[email protected]/dist/vanta.dots.min.js?MEH=BLAH' + }, + shim: { + 'VANTA': { + deps: ['three'] } + } }); -// require(['three', 'VANTA'], function(THREE) { -// // Assuming VANTA attaches itself to a global variable, directly use it here -// // Check the correct global variable name (VANTA, _vantaEffect, etc.) -// if (window.VANTA && window.VANTA.WAVES) { -// window.VANTA.WAVES({ -// el: "#hero" -// }); -// } else { -// console.error('VANTA.WAVES is not available.'); -// } -// }); - -require(['three'], function(THREE) { +var vantaEffect; + +function initVanta() { + require(['three'], function (THREE) { window.THREE = THREE; // Make THREE globally accessible - require(['VANTA'], function() { - if (VANTA.DOTS) { - VANTA.DOTS({ - el: "#canvas", - mouseControls: false, - touchControls: false, - gyroControls: false, - minHeight: 200.00, - minWidth: 200.00, - scale: 1.00, - scaleMobile: 1.00, - color: 0xb3a094, - color2: 0xffffff, - backgroundColor: 0x100F0F, - size: 0.50, - // spacing: 68.00, - spacing: 10.00, - showLines: false - }) - } + require(['VANTA'], function () { + if (vantaEffect) { + vantaEffect.destroy(); + } + + const darkMode = document.documentElement.dataset.theme === 'dark'; + const color = darkMode ? 0xCECDC3 : 0x100F0F; // Example dark mode color + const color2 = darkMode ? 0xCECDC3 : 0x100F0F; // Adjust color2 for dark mode as needed + const backgroundColor = darkMode ? 0x100F0F : 0xFFFCF0; // Adjust background color for dark mode + + vantaEffect = VANTA.DOTS({ + el: "#canvas", + mouseControls: false, + touchControls: false, + gyroControls: false, + minHeight: 200.00, + minWidth: 200.00, + scale: 1.00, + scaleMobile: 1.00, + color: color, + color2: color2, + backgroundColor: backgroundColor, + size: 0.50, + spacing: 10.00, + showLines: false + }); }); + }); +} + +// Initial call +initVanta(); + +// Setup a mutation observer to detect theme changes +var observer = new MutationObserver(function(mutations) { + mutations.forEach(function(mutation) { + if (mutation.attributeName === "data-theme") { + initVanta(); // Reinitialize Vanta with the new theme settings + } + }); }); +// Start observing the document element for attribute changes +observer.observe(document.documentElement, {attributes: true, attributeFilter: ['data-theme']}); + + document.addEventListener('DOMContentLoaded', (event) => { - var listItems = document.querySelectorAll('#list-tab .list-group-item'); - listItems.forEach(function(listItem) { - listItem.addEventListener('click', function(e) { - e.preventDefault(); - var tabPaneId = this.getAttribute('href'); - - // Remove active class from all list items - listItems.forEach(function(item) { - item.classList.remove('active'); - }); - - // Add active class to the clicked list item - this.classList.add('active'); - - // Hide all tab panes - var tabPanes = document.querySelectorAll('.tab-content .tab-pane'); - tabPanes.forEach(function(tabPane) { - tabPane.classList.remove('show'); - tabPane.classList.remove('active'); - }); - - // Show the clicked tab pane - var activeTabPane = document.querySelector(tabPaneId); - activeTabPane.classList.add('active'); - - // We need to manually add 'show' class after a slight delay to trigger fade effect - setTimeout(function() { - activeTabPane.classList.add('show'); - }, 150); // Adjust delay as needed + var listItems = document.querySelectorAll('#list-tab .list-group-item'); + listItems.forEach(function (listItem) { + listItem.addEventListener('click', function (e) { + e.preventDefault(); + var tabPaneId = this.getAttribute('href'); + + // Remove active class from all list items + listItems.forEach(function (item) { + item.classList.remove('active'); }); + + // Add active class to the clicked list item + this.classList.add('active'); + + // Hide all tab panes + var tabPanes = document.querySelectorAll('.tab-content .tab-pane'); + tabPanes.forEach(function (tabPane) { + tabPane.classList.remove('show'); + tabPane.classList.remove('active'); + }); + + // Show the clicked tab pane + var activeTabPane = document.querySelector(tabPaneId); + activeTabPane.classList.add('active'); + + // We need to manually add 'show' class after a slight delay to trigger fade effect + setTimeout(function () { + activeTabPane.classList.add('show'); + }, 150); // Adjust delay as needed }); }); - \ No newline at end of file +}); + +document.addEventListener("DOMContentLoaded", function() { + fetch("https://ndif.dev/ping") + .then(response => response.json()) + .then(data => { + const statusElement = document.querySelector(".ndif"); + const status = (data === "pong") ? "Available" : "Unavailable"; + if (statusElement) { + statusElement.setAttribute('data-bs-original-title', `Status: ${status}`); + } + }) + .catch(error => console.error('Error fetching NDIF status:', error)); + + const data = document.querySelector("p.copyright"); + data.innerHTML = "(c) 2024 <a href='https://ndif.us/'>NDIF</a>" + +}); diff --git a/docs/source/_static/pygments/example.html b/docs/source/_static/pygments/example.html deleted file mode 100644 index d1546fd9..00000000 --- a/docs/source/_static/pygments/example.html +++ /dev/null @@ -1,17 +0,0 @@ -<div class="highlight"><pre><span></span><span class="kn">import</span> <span class="nn">json</span> -<span class="kn">from</span> <span class="nn">nnsight</span> <span class="kn">import</span> <span class="n">LanguageModel</span> -<span class="kn">import</span> <span class="nn">torch</span> <span class="k">as</span> <span class="nn">t</span> -<span class="kn">from</span> <span class="nn">tqdm</span> <span class="kn">import</span> <span class="n">tqdm</span> - -<span class="n">aligned</span> <span class="o">=</span> <span class="p">[</span> - <span class="s2">"truth_teller"</span><span class="p">,</span> - <span class="s2">"genie"</span><span class="p">,</span> - <span class="s2">"saint"</span><span class="p">,</span> -<span class="p">]</span> - -<span class="n">misaligned</span> <span class="o">=</span> <span class="p">[</span> - <span class="s2">"reward_maximizer"</span><span class="p">,</span> - <span class="s2">"fitness_maximizer"</span><span class="p">,</span> - <span class="s2">"money_maximizer"</span><span class="p">,</span> -<span class="p">]</span> -</pre></div> diff --git a/docs/source/_static/pygments/example.py b/docs/source/_static/pygments/example.py deleted file mode 100644 index d6f3c5b5..00000000 --- a/docs/source/_static/pygments/example.py +++ /dev/null @@ -1,16 +0,0 @@ -import json -from nnsight import LanguageModel -import torch as t -from tqdm import tqdm - -aligned = [ - "truth_teller", - "genie", - "saint", -] - -misaligned = [ - "reward_maximizer", - "fitness_maximizer", - "money_maximizer", -] diff --git a/docs/source/_templates/layout.html b/docs/source/_templates/layout.html new file mode 100644 index 00000000..de1d1c49 --- /dev/null +++ b/docs/source/_templates/layout.html @@ -0,0 +1,7 @@ +{% extends "!layout.html" %} + +{% block extrahead %} +{{ super() }} +<link href="{{ pathto('_static/css/custom.css', 1) }}?v={{ version_identifier }}" rel="stylesheet" type="text/css" /> +<link href="{{ pathto('_static/css/front.css', 1) }}?v={{ version_identifier }}" rel="stylesheet" type="text/css" /> +{% endblock %} diff --git a/docs/source/_templates/ndif_status.html b/docs/source/_templates/ndif_status.html index 348c8d61..4ac071d8 100644 --- a/docs/source/_templates/ndif_status.html +++ b/docs/source/_templates/ndif_status.html @@ -1,5 +1,4 @@ <script> - fetch("{{ndif_url}}") .then((response) => { if (response.status == 200) { diff --git a/docs/source/about.rst b/docs/source/about.rst index c58f101f..08fef7e9 100644 --- a/docs/source/about.rst +++ b/docs/source/about.rst @@ -1,3 +1,18 @@ +.. raw:: html + + <style> + .accordion-header { + margin: 0 !important; + } + </style> + + <script> + document.addEventListener('DOMContentLoaded', (event) => { + document.querySelectorAll('figure.align-default').forEach(el => { + el.style.marginBottom = "0px"; + }); + }); + </script> About NNsight ============= @@ -16,10 +31,11 @@ An API for transparent science on black-box AI. it lets you run experiments on huge open models easily, with full transparent access. The nnsight library is also terrific for studying smaller local models. -.. figure:: _static/images/remote_execution.png +.. figure:: _static/images/interleaved.png .. card:: + An overview of the nnsight/NDIF pipeline. Researchers write simple Python code to run along with the neural network locally or remotely. Unlike commercial inference, the experiment code can read or write any of the internal states of the neural networks being studied. This code creates a computation graph that can be sent to the remote service and interleaved with the execution of the neural network. How do I use NNsight? @@ -61,7 +77,7 @@ When using nnsight, it is helpful to understand that the operations are not exec An example of one such intervention graph can be seen below: -.. figure:: _static/images/intrgraph.png +.. figure:: _static/images/execution.png .. card:: diff --git a/docs/source/charts/chart_schema.json b/docs/source/charts/chart_schema.json deleted file mode 100644 index 6733fb47..00000000 --- a/docs/source/charts/chart_schema.json +++ /dev/null @@ -1,1124 +0,0 @@ -{ - "data": [ - { - "coloraxis": "coloraxis", - "name": "0", - "z": [ - [ - 0.0, - 0.0, - 0.0, - 0.0, - 0.0, - 0.0, - 0.0, - 0.0, - 0.0, - 0.9999988079071045, - 0.0, - 0.0, - 0.0, - 0.0, - 0.0, - 0.0 - ], - [ - 0.0, - 0.0, - 0.0, - 0.0, - 0.0, - 0.0, - 0.0, - 0.0, - 0.0, - 0.9974687695503235, - 0.0017737559974193573, - 0.0, - 0.0, - 0.0, - -0.001089064753614366, - 0.0 - ], - [ - 0.0, - 0.0, - 0.0, - 0.0, - 0.0, - 0.0, - 0.0, - 0.0, - 0.0, - 0.9971288442611694, - 0.0026111311744898558, - 0.0, - 0.0, - 0.0, - 0.0, - 0.0 - ], - [ - 0.0, - 0.0, - 0.0, - 0.0, - 0.0, - 0.0, - 0.0, - 0.0, - 0.0, - 0.9950985908508301, - 0.004562620539218187, - 0.0, - 0.0, - 0.0, - 0.0, - -0.0021960220765322447 - ], - [ - 0.0, - 0.0, - 0.0, - 0.0, - 0.0, - 0.0, - 0.0, - 0.0, - 0.0, - 0.9868704080581665, - 0.020100580528378487, - 0.0, - 0.0011236571008339524, - 0.0, - 0.0, - -0.0019157042261213064 - ], - [ - 0.0, - 0.0, - 0.0, - 0.0, - 0.0, - 0.0, - 0.0, - 0.0, - 0.0, - 0.9520763754844666, - 0.07812518626451492, - 0.002582503017038107, - 0.0021328015718609095, - 0.0, - 0.0, - -0.001445724512450397 - ], - [ - 0.0, - 0.0, - 0.0, - 0.0, - 0.0, - 0.0, - 0.0, - 0.0, - 0.0, - 0.9528278708457947, - 0.0788850262761116, - 0.0019860819447785616, - 0.0, - 0.0, - 0.0, - -0.004969379864633083 - ], - [ - 0.0, - 0.0, - 0.0, - 0.0, - 0.0, - 0.0, - 0.0, - 0.0, - 0.0, - 0.9087952971458435, - 0.10825876146554947, - 0.003998406231403351, - 0.0, - 0.0, - 0.0, - 0.009814703837037086 - ], - [ - 0.0, - 0.0, - 0.0, - 0.0, - 0.0, - 0.0, - 0.0, - 0.0, - 0.0, - 0.6858435869216919, - 0.035843707621097565, - 0.0, - 0.0, - 0.0, - 0.0013097404735162854, - 0.4153654873371124 - ], - [ - 0.0, - 0.0, - 0.0, - 0.0, - 0.0, - 0.0, - 0.0, - 0.0, - 0.0, - 0.10977605730295181, - 0.024682287126779556, - 0.0, - 0.0, - 0.0, - 0.001402782159857452, - 0.8544625639915466 - ], - [ - 0.0, - 0.0, - 0.0, - 0.0, - 0.0, - 0.0, - 0.0, - 0.0, - 0.0, - 0.01987036131322384, - 0.015400783158838749, - 0.0, - 0.0, - 0.0, - 0.0, - 0.9079328775405884 - ], - [ - 0.0, - 0.0, - 0.0, - 0.0, - 0.0, - 0.0, - 0.0, - 0.0, - 0.0, - 0.019981296733021736, - 0.0062970127910375595, - 0.0, - 0.0, - 0.0, - 0.0, - 0.9069046378135681 - ] - ], - "type": "heatmap", - "xaxis": "x", - "yaxis": "y", - "hovertemplate": "Position: %{x}\u003cbr\u003eLayer: %{y}\u003cbr\u003ecolor: %{z}\u003cextra\u003e\u003c\u002fextra\u003e" - } - ], - "layout": { - "template": { - "data": { - "histogram2dcontour": [ - { - "type": "histogram2dcontour", - "colorbar": { - "outlinewidth": 0, - "ticks": "" - }, - "colorscale": [ - [ - 0.0, - "#0d0887" - ], - [ - 0.1111111111111111, - "#46039f" - ], - [ - 0.2222222222222222, - "#7201a8" - ], - [ - 0.3333333333333333, - "#9c179e" - ], - [ - 0.4444444444444444, - "#bd3786" - ], - [ - 0.5555555555555556, - "#d8576b" - ], - [ - 0.6666666666666666, - "#ed7953" - ], - [ - 0.7777777777777778, - "#fb9f3a" - ], - [ - 0.8888888888888888, - "#fdca26" - ], - [ - 1.0, - "#f0f921" - ] - ] - } - ], - "choropleth": [ - { - "type": "choropleth", - "colorbar": { - "outlinewidth": 0, - "ticks": "" - } - } - ], - "histogram2d": [ - { - "type": "histogram2d", - "colorbar": { - "outlinewidth": 0, - "ticks": "" - }, - "colorscale": [ - [ - 0.0, - "#0d0887" - ], - [ - 0.1111111111111111, - "#46039f" - ], - [ - 0.2222222222222222, - "#7201a8" - ], - [ - 0.3333333333333333, - "#9c179e" - ], - [ - 0.4444444444444444, - "#bd3786" - ], - [ - 0.5555555555555556, - "#d8576b" - ], - [ - 0.6666666666666666, - "#ed7953" - ], - [ - 0.7777777777777778, - "#fb9f3a" - ], - [ - 0.8888888888888888, - "#fdca26" - ], - [ - 1.0, - "#f0f921" - ] - ] - } - ], - "heatmap": [ - { - "type": "heatmap", - "colorbar": { - "outlinewidth": 0, - "ticks": "" - }, - "colorscale": [ - [ - 0.0, - "#0d0887" - ], - [ - 0.1111111111111111, - "#46039f" - ], - [ - 0.2222222222222222, - "#7201a8" - ], - [ - 0.3333333333333333, - "#9c179e" - ], - [ - 0.4444444444444444, - "#bd3786" - ], - [ - 0.5555555555555556, - "#d8576b" - ], - [ - 0.6666666666666666, - "#ed7953" - ], - [ - 0.7777777777777778, - "#fb9f3a" - ], - [ - 0.8888888888888888, - "#fdca26" - ], - [ - 1.0, - "#f0f921" - ] - ] - } - ], - "heatmapgl": [ - { - "type": "heatmapgl", - "colorbar": { - "outlinewidth": 0, - "ticks": "" - }, - "colorscale": [ - [ - 0.0, - "#0d0887" - ], - [ - 0.1111111111111111, - "#46039f" - ], - [ - 0.2222222222222222, - "#7201a8" - ], - [ - 0.3333333333333333, - "#9c179e" - ], - [ - 0.4444444444444444, - "#bd3786" - ], - [ - 0.5555555555555556, - "#d8576b" - ], - [ - 0.6666666666666666, - "#ed7953" - ], - [ - 0.7777777777777778, - "#fb9f3a" - ], - [ - 0.8888888888888888, - "#fdca26" - ], - [ - 1.0, - "#f0f921" - ] - ] - } - ], - "contourcarpet": [ - { - "type": "contourcarpet", - "colorbar": { - "outlinewidth": 0, - "ticks": "" - } - } - ], - "contour": [ - { - "type": "contour", - "colorbar": { - "outlinewidth": 0, - "ticks": "" - }, - "colorscale": [ - [ - 0.0, - "#0d0887" - ], - [ - 0.1111111111111111, - "#46039f" - ], - [ - 0.2222222222222222, - "#7201a8" - ], - [ - 0.3333333333333333, - "#9c179e" - ], - [ - 0.4444444444444444, - "#bd3786" - ], - [ - 0.5555555555555556, - "#d8576b" - ], - [ - 0.6666666666666666, - "#ed7953" - ], - [ - 0.7777777777777778, - "#fb9f3a" - ], - [ - 0.8888888888888888, - "#fdca26" - ], - [ - 1.0, - "#f0f921" - ] - ] - } - ], - "surface": [ - { - "type": "surface", - "colorbar": { - "outlinewidth": 0, - "ticks": "" - }, - "colorscale": [ - [ - 0.0, - "#0d0887" - ], - [ - 0.1111111111111111, - "#46039f" - ], - [ - 0.2222222222222222, - "#7201a8" - ], - [ - 0.3333333333333333, - "#9c179e" - ], - [ - 0.4444444444444444, - "#bd3786" - ], - [ - 0.5555555555555556, - "#d8576b" - ], - [ - 0.6666666666666666, - "#ed7953" - ], - [ - 0.7777777777777778, - "#fb9f3a" - ], - [ - 0.8888888888888888, - "#fdca26" - ], - [ - 1.0, - "#f0f921" - ] - ] - } - ], - "mesh3d": [ - { - "type": "mesh3d", - "colorbar": { - "outlinewidth": 0, - "ticks": "" - } - } - ], - "scatter": [ - { - "fillpattern": { - "fillmode": "overlay", - "size": 10, - "solidity": 0.2 - }, - "type": "scatter" - } - ], - "parcoords": [ - { - "type": "parcoords", - "line": { - "colorbar": { - "outlinewidth": 0, - "ticks": "" - } - } - } - ], - "scatterpolargl": [ - { - "type": "scatterpolargl", - "marker": { - "colorbar": { - "outlinewidth": 0, - "ticks": "" - } - } - } - ], - "bar": [ - { - "error_x": { - "color": "#2a3f5f" - }, - "error_y": { - "color": "#2a3f5f" - }, - "marker": { - "line": { - "color": "#E5ECF6", - "width": 0.5 - }, - "pattern": { - "fillmode": "overlay", - "size": 10, - "solidity": 0.2 - } - }, - "type": "bar" - } - ], - "scattergeo": [ - { - "type": "scattergeo", - "marker": { - "colorbar": { - "outlinewidth": 0, - "ticks": "" - } - } - } - ], - "scatterpolar": [ - { - "type": "scatterpolar", - "marker": { - "colorbar": { - "outlinewidth": 0, - "ticks": "" - } - } - } - ], - "histogram": [ - { - "marker": { - "pattern": { - "fillmode": "overlay", - "size": 10, - "solidity": 0.2 - } - }, - "type": "histogram" - } - ], - "scattergl": [ - { - "type": "scattergl", - "marker": { - "colorbar": { - "outlinewidth": 0, - "ticks": "" - } - } - } - ], - "scatter3d": [ - { - "type": "scatter3d", - "line": { - "colorbar": { - "outlinewidth": 0, - "ticks": "" - } - }, - "marker": { - "colorbar": { - "outlinewidth": 0, - "ticks": "" - } - } - } - ], - "scattermapbox": [ - { - "type": "scattermapbox", - "marker": { - "colorbar": { - "outlinewidth": 0, - "ticks": "" - } - } - } - ], - "scatterternary": [ - { - "type": "scatterternary", - "marker": { - "colorbar": { - "outlinewidth": 0, - "ticks": "" - } - } - } - ], - "scattercarpet": [ - { - "type": "scattercarpet", - "marker": { - "colorbar": { - "outlinewidth": 0, - "ticks": "" - } - } - } - ], - "carpet": [ - { - "aaxis": { - "endlinecolor": "#2a3f5f", - "gridcolor": "white", - "linecolor": "white", - "minorgridcolor": "white", - "startlinecolor": "#2a3f5f" - }, - "baxis": { - "endlinecolor": "#2a3f5f", - "gridcolor": "white", - "linecolor": "white", - "minorgridcolor": "white", - "startlinecolor": "#2a3f5f" - }, - "type": "carpet" - } - ], - "table": [ - { - "cells": { - "fill": { - "color": "#EBF0F8" - }, - "line": { - "color": "white" - } - }, - "header": { - "fill": { - "color": "#C8D4E3" - }, - "line": { - "color": "white" - } - }, - "type": "table" - } - ], - "barpolar": [ - { - "marker": { - "line": { - "color": "#E5ECF6", - "width": 0.5 - }, - "pattern": { - "fillmode": "overlay", - "size": 10, - "solidity": 0.2 - } - }, - "type": "barpolar" - } - ], - "pie": [ - { - "automargin": true, - "type": "pie" - } - ] - }, - "layout": { - "autotypenumbers": "strict", - "colorway": [ - "#636efa", - "#EF553B", - "#00cc96", - "#ab63fa", - "#FFA15A", - "#19d3f3", - "#FF6692", - "#B6E880", - "#FF97FF", - "#FECB52" - ], - "font": { - "color": "#2a3f5f" - }, - "hovermode": "closest", - "hoverlabel": { - "align": "left" - }, - "paper_bgcolor": "white", - "plot_bgcolor": "#E5ECF6", - "polar": { - "bgcolor": "#E5ECF6", - "angularaxis": { - "gridcolor": "white", - "linecolor": "white", - "ticks": "" - }, - "radialaxis": { - "gridcolor": "white", - "linecolor": "white", - "ticks": "" - } - }, - "ternary": { - "bgcolor": "#E5ECF6", - "aaxis": { - "gridcolor": "white", - "linecolor": "white", - "ticks": "" - }, - "baxis": { - "gridcolor": "white", - "linecolor": "white", - "ticks": "" - }, - "caxis": { - "gridcolor": "white", - "linecolor": "white", - "ticks": "" - } - }, - "coloraxis": { - "colorbar": { - "outlinewidth": 0, - "ticks": "" - } - }, - "colorscale": { - "sequential": [ - [ - 0.0, - "#0d0887" - ], - [ - 0.1111111111111111, - "#46039f" - ], - [ - 0.2222222222222222, - "#7201a8" - ], - [ - 0.3333333333333333, - "#9c179e" - ], - [ - 0.4444444444444444, - "#bd3786" - ], - [ - 0.5555555555555556, - "#d8576b" - ], - [ - 0.6666666666666666, - "#ed7953" - ], - [ - 0.7777777777777778, - "#fb9f3a" - ], - [ - 0.8888888888888888, - "#fdca26" - ], - [ - 1.0, - "#f0f921" - ] - ], - "sequentialminus": [ - [ - 0.0, - "#0d0887" - ], - [ - 0.1111111111111111, - "#46039f" - ], - [ - 0.2222222222222222, - "#7201a8" - ], - [ - 0.3333333333333333, - "#9c179e" - ], - [ - 0.4444444444444444, - "#bd3786" - ], - [ - 0.5555555555555556, - "#d8576b" - ], - [ - 0.6666666666666666, - "#ed7953" - ], - [ - 0.7777777777777778, - "#fb9f3a" - ], - [ - 0.8888888888888888, - "#fdca26" - ], - [ - 1.0, - "#f0f921" - ] - ], - "diverging": [ - [ - 0, - "#8e0152" - ], - [ - 0.1, - "#c51b7d" - ], - [ - 0.2, - "#de77ae" - ], - [ - 0.3, - "#f1b6da" - ], - [ - 0.4, - "#fde0ef" - ], - [ - 0.5, - "#f7f7f7" - ], - [ - 0.6, - "#e6f5d0" - ], - [ - 0.7, - "#b8e186" - ], - [ - 0.8, - "#7fbc41" - ], - [ - 0.9, - "#4d9221" - ], - [ - 1, - "#276419" - ] - ] - }, - "xaxis": { - "gridcolor": "white", - "linecolor": "white", - "ticks": "", - "title": { - "standoff": 15 - }, - "zerolinecolor": "white", - "automargin": true, - "zerolinewidth": 2 - }, - "yaxis": { - "gridcolor": "white", - "linecolor": "white", - "ticks": "", - "title": { - "standoff": 15 - }, - "zerolinecolor": "white", - "automargin": true, - "zerolinewidth": 2 - }, - "scene": { - "xaxis": { - "backgroundcolor": "#E5ECF6", - "gridcolor": "white", - "linecolor": "white", - "showbackground": true, - "ticks": "", - "zerolinecolor": "white", - "gridwidth": 2 - }, - "yaxis": { - "backgroundcolor": "#E5ECF6", - "gridcolor": "white", - "linecolor": "white", - "showbackground": true, - "ticks": "", - "zerolinecolor": "white", - "gridwidth": 2 - }, - "zaxis": { - "backgroundcolor": "#E5ECF6", - "gridcolor": "white", - "linecolor": "white", - "showbackground": true, - "ticks": "", - "zerolinecolor": "white", - "gridwidth": 2 - } - }, - "shapedefaults": { - "line": { - "color": "#2a3f5f" - } - }, - "annotationdefaults": { - "arrowcolor": "#2a3f5f", - "arrowhead": 0, - "arrowwidth": 1 - }, - "geo": { - "bgcolor": "white", - "landcolor": "#E5ECF6", - "subunitcolor": "white", - "showland": true, - "showlakes": true, - "lakecolor": "white" - }, - "title": { - "x": 0.05 - }, - "mapbox": { - "style": "light" - } - } - }, - "xaxis": { - "anchor": "y", - "domain": [ - 0.0, - 1.0 - ], - "scaleanchor": "y", - "constrain": "domain", - "title": { - "text": "Position" - } - }, - "yaxis": { - "anchor": "x", - "domain": [ - 0.0, - 1.0 - ], - "autorange": "reversed", - "constrain": "domain", - "title": { - "text": "Layer" - } - }, - "coloraxis": { - "colorscale": [ - [ - 0.0, - "rgb(103,0,31)" - ], - [ - 0.1, - "rgb(178,24,43)" - ], - [ - 0.2, - "rgb(214,96,77)" - ], - [ - 0.3, - "rgb(244,165,130)" - ], - [ - 0.4, - "rgb(253,219,199)" - ], - [ - 0.5, - "rgb(247,247,247)" - ], - [ - 0.6, - "rgb(209,229,240)" - ], - [ - 0.7, - "rgb(146,197,222)" - ], - [ - 0.8, - "rgb(67,147,195)" - ], - [ - 0.9, - "rgb(33,102,172)" - ], - [ - 1.0, - "rgb(5,48,97)" - ] - ], - "cmid": 0.0 - }, - "margin": { - "t": 60 - } - } -} \ No newline at end of file diff --git a/docs/source/charts/induction.json b/docs/source/charts/induction.json deleted file mode 100644 index 4946b16a..00000000 --- a/docs/source/charts/induction.json +++ /dev/null @@ -1,1 +0,0 @@ -{"data":[{"coloraxis":"coloraxis","name":"0","z":[[0.010070409625768661,7.986390846781433e-05,0.01258260291069746,2.1361253743634734e-07,0.00020078140369150788,5.013069312553853e-05,0.008831579238176346,0.000892787182237953,0.009880675002932549,0.010911175981163979,0.007516765501350164,0.015246091410517693],[0.0004512245941441506,0.0006454214453697205,0.0025818662252277136,0.01306276023387909,0.005299094598740339,0.013112545944750309,0.015744850039482117,0.013699539005756378,0.012578608468174934,0.015512274578213692,0.006809852551668882,0.001121349399909377],[0.004063299857079983,0.01543406117707491,0.0010489624692127109,0.0013296647230163217,0.011215250939130783,0.0018478287383913994,0.005366469733417034,0.008304793387651443,0.0030799838714301586,0.0010017683962360024,0.0008394559845328331,0.009666861966252327],[0.013943175785243511,0.005458588246256113,0.0022658391389995813,0.009771357290446758,0.019387023523449898,0.010070898570120335,0.001896885340102017,0.0006945269415155053,0.004318578634411097,0.009726484306156635,0.007587844040244818,0.005201588850468397],[0.012102176435291767,0.01121105533093214,0.012845116667449474,0.007275603711605072,0.01644388772547245,0.010665860958397388,0.007502487860620022,0.0023389223497360945,0.01597411371767521,0.011477481573820114,0.017345573753118515,3.745994181514334e-09],[0.16879354417324066,0.3332799971103668,0.01330574695020914,0.006573637947440147,0.009730679914355278,0.3288743495941162,0.008749036118388176,0.015621461905539036,0.020145757123827934,0.023961644619703293,0.01803380250930786,0.015584372915327549],[0.00760197639465332,0.01550981029868126,0.017369290813803673,0.014239192940294743,0.01672932133078575,0.011254709213972092,0.024911386892199516,0.010701530613005161,0.010776473209261894,0.32112863659858704,0.024556512013077736,0.010988256894052029],[0.009945792146027088,0.11099910736083984,0.2978491187095642,0.01551186852157116,0.016720473766326904,0.013705338351428509,0.02495485544204712,0.09114369004964828,0.01463283784687519,0.016146264970302582,0.3241707682609558,0.03573969751596451],[0.014393663965165615,0.1397504210472107,0.012329380959272385,0.023921499028801918,0.01609531044960022,0.011505285277962685,0.06444355845451355,0.012670181691646576,0.017619885504245758,0.02077418565750122,0.030146704986691475,0.014868587255477905],[0.09159916639328003,0.08219298720359802,0.046227119863033295,0.0109515730291605,0.04310443252325058,0.019227860495448112,0.16419260203838348,0.020021013915538788,0.023179948329925537,0.17912650108337402,0.01535245031118393,0.019395975396037102],[0.12493489682674408,0.18007728457450867,0.02339624986052513,0.062402497977018356,0.030824515968561172,0.013725594617426395,0.11222781985998154,0.1664256453514099,0.02788865752518177,0.013621106743812561,0.06758259236812592,0.10711898654699326],[0.014790158718824387,0.030200988054275513,0.02301931381225586,0.00981247890740633,0.019670825451612473,0.04539807513356209,0.029655294492840767,0.03308337181806564,0.008349732495844364,0.1181706115603447,0.1325961947441101,0.013514983467757702]],"type":"heatmap","xaxis":"x","yaxis":"y","hovertemplate":"Position: %{x}\u003cbr\u003eLayer: %{y}\u003cbr\u003ecolor: %{z}\u003cextra\u003e\u003c\u002fextra\u003e"}],"layout":{"template":{"data":{"histogram2dcontour":[{"type":"histogram2dcontour","colorbar":{"outlinewidth":0,"ticks":""},"colorscale":[[0.0,"#0d0887"],[0.1111111111111111,"#46039f"],[0.2222222222222222,"#7201a8"],[0.3333333333333333,"#9c179e"],[0.4444444444444444,"#bd3786"],[0.5555555555555556,"#d8576b"],[0.6666666666666666,"#ed7953"],[0.7777777777777778,"#fb9f3a"],[0.8888888888888888,"#fdca26"],[1.0,"#f0f921"]]}],"choropleth":[{"type":"choropleth","colorbar":{"outlinewidth":0,"ticks":""}}],"histogram2d":[{"type":"histogram2d","colorbar":{"outlinewidth":0,"ticks":""},"colorscale":[[0.0,"#0d0887"],[0.1111111111111111,"#46039f"],[0.2222222222222222,"#7201a8"],[0.3333333333333333,"#9c179e"],[0.4444444444444444,"#bd3786"],[0.5555555555555556,"#d8576b"],[0.6666666666666666,"#ed7953"],[0.7777777777777778,"#fb9f3a"],[0.8888888888888888,"#fdca26"],[1.0,"#f0f921"]]}],"heatmap":[{"type":"heatmap","colorbar":{"outlinewidth":0,"ticks":""},"colorscale":[[0.0,"#0d0887"],[0.1111111111111111,"#46039f"],[0.2222222222222222,"#7201a8"],[0.3333333333333333,"#9c179e"],[0.4444444444444444,"#bd3786"],[0.5555555555555556,"#d8576b"],[0.6666666666666666,"#ed7953"],[0.7777777777777778,"#fb9f3a"],[0.8888888888888888,"#fdca26"],[1.0,"#f0f921"]]}],"heatmapgl":[{"type":"heatmapgl","colorbar":{"outlinewidth":0,"ticks":""},"colorscale":[[0.0,"#0d0887"],[0.1111111111111111,"#46039f"],[0.2222222222222222,"#7201a8"],[0.3333333333333333,"#9c179e"],[0.4444444444444444,"#bd3786"],[0.5555555555555556,"#d8576b"],[0.6666666666666666,"#ed7953"],[0.7777777777777778,"#fb9f3a"],[0.8888888888888888,"#fdca26"],[1.0,"#f0f921"]]}],"contourcarpet":[{"type":"contourcarpet","colorbar":{"outlinewidth":0,"ticks":""}}],"contour":[{"type":"contour","colorbar":{"outlinewidth":0,"ticks":""},"colorscale":[[0.0,"#0d0887"],[0.1111111111111111,"#46039f"],[0.2222222222222222,"#7201a8"],[0.3333333333333333,"#9c179e"],[0.4444444444444444,"#bd3786"],[0.5555555555555556,"#d8576b"],[0.6666666666666666,"#ed7953"],[0.7777777777777778,"#fb9f3a"],[0.8888888888888888,"#fdca26"],[1.0,"#f0f921"]]}],"surface":[{"type":"surface","colorbar":{"outlinewidth":0,"ticks":""},"colorscale":[[0.0,"#0d0887"],[0.1111111111111111,"#46039f"],[0.2222222222222222,"#7201a8"],[0.3333333333333333,"#9c179e"],[0.4444444444444444,"#bd3786"],[0.5555555555555556,"#d8576b"],[0.6666666666666666,"#ed7953"],[0.7777777777777778,"#fb9f3a"],[0.8888888888888888,"#fdca26"],[1.0,"#f0f921"]]}],"mesh3d":[{"type":"mesh3d","colorbar":{"outlinewidth":0,"ticks":""}}],"scatter":[{"fillpattern":{"fillmode":"overlay","size":10,"solidity":0.2},"type":"scatter"}],"parcoords":[{"type":"parcoords","line":{"colorbar":{"outlinewidth":0,"ticks":""}}}],"scatterpolargl":[{"type":"scatterpolargl","marker":{"colorbar":{"outlinewidth":0,"ticks":""}}}],"bar":[{"error_x":{"color":"#2a3f5f"},"error_y":{"color":"#2a3f5f"},"marker":{"line":{"color":"#E5ECF6","width":0.5},"pattern":{"fillmode":"overlay","size":10,"solidity":0.2}},"type":"bar"}],"scattergeo":[{"type":"scattergeo","marker":{"colorbar":{"outlinewidth":0,"ticks":""}}}],"scatterpolar":[{"type":"scatterpolar","marker":{"colorbar":{"outlinewidth":0,"ticks":""}}}],"histogram":[{"marker":{"pattern":{"fillmode":"overlay","size":10,"solidity":0.2}},"type":"histogram"}],"scattergl":[{"type":"scattergl","marker":{"colorbar":{"outlinewidth":0,"ticks":""}}}],"scatter3d":[{"type":"scatter3d","line":{"colorbar":{"outlinewidth":0,"ticks":""}},"marker":{"colorbar":{"outlinewidth":0,"ticks":""}}}],"scattermapbox":[{"type":"scattermapbox","marker":{"colorbar":{"outlinewidth":0,"ticks":""}}}],"scatterternary":[{"type":"scatterternary","marker":{"colorbar":{"outlinewidth":0,"ticks":""}}}],"scattercarpet":[{"type":"scattercarpet","marker":{"colorbar":{"outlinewidth":0,"ticks":""}}}],"carpet":[{"aaxis":{"endlinecolor":"#2a3f5f","gridcolor":"white","linecolor":"white","minorgridcolor":"white","startlinecolor":"#2a3f5f"},"baxis":{"endlinecolor":"#2a3f5f","gridcolor":"white","linecolor":"white","minorgridcolor":"white","startlinecolor":"#2a3f5f"},"type":"carpet"}],"table":[{"cells":{"fill":{"color":"#EBF0F8"},"line":{"color":"white"}},"header":{"fill":{"color":"#C8D4E3"},"line":{"color":"white"}},"type":"table"}],"barpolar":[{"marker":{"line":{"color":"#E5ECF6","width":0.5},"pattern":{"fillmode":"overlay","size":10,"solidity":0.2}},"type":"barpolar"}],"pie":[{"automargin":true,"type":"pie"}]},"layout":{"autotypenumbers":"strict","colorway":["#636efa","#EF553B","#00cc96","#ab63fa","#FFA15A","#19d3f3","#FF6692","#B6E880","#FF97FF","#FECB52"],"font":{"color":"#2a3f5f"},"hovermode":"closest","hoverlabel":{"align":"left"},"paper_bgcolor":"white","plot_bgcolor":"#E5ECF6","polar":{"bgcolor":"#E5ECF6","angularaxis":{"gridcolor":"white","linecolor":"white","ticks":""},"radialaxis":{"gridcolor":"white","linecolor":"white","ticks":""}},"ternary":{"bgcolor":"#E5ECF6","aaxis":{"gridcolor":"white","linecolor":"white","ticks":""},"baxis":{"gridcolor":"white","linecolor":"white","ticks":""},"caxis":{"gridcolor":"white","linecolor":"white","ticks":""}},"coloraxis":{"colorbar":{"outlinewidth":0,"ticks":""}},"colorscale":{"sequential":[[0.0,"#0d0887"],[0.1111111111111111,"#46039f"],[0.2222222222222222,"#7201a8"],[0.3333333333333333,"#9c179e"],[0.4444444444444444,"#bd3786"],[0.5555555555555556,"#d8576b"],[0.6666666666666666,"#ed7953"],[0.7777777777777778,"#fb9f3a"],[0.8888888888888888,"#fdca26"],[1.0,"#f0f921"]],"sequentialminus":[[0.0,"#0d0887"],[0.1111111111111111,"#46039f"],[0.2222222222222222,"#7201a8"],[0.3333333333333333,"#9c179e"],[0.4444444444444444,"#bd3786"],[0.5555555555555556,"#d8576b"],[0.6666666666666666,"#ed7953"],[0.7777777777777778,"#fb9f3a"],[0.8888888888888888,"#fdca26"],[1.0,"#f0f921"]],"diverging":[[0,"#8e0152"],[0.1,"#c51b7d"],[0.2,"#de77ae"],[0.3,"#f1b6da"],[0.4,"#fde0ef"],[0.5,"#f7f7f7"],[0.6,"#e6f5d0"],[0.7,"#b8e186"],[0.8,"#7fbc41"],[0.9,"#4d9221"],[1,"#276419"]]},"xaxis":{"gridcolor":"white","linecolor":"white","ticks":"","title":{"standoff":15},"zerolinecolor":"white","automargin":true,"zerolinewidth":2},"yaxis":{"gridcolor":"white","linecolor":"white","ticks":"","title":{"standoff":15},"zerolinecolor":"white","automargin":true,"zerolinewidth":2},"scene":{"xaxis":{"backgroundcolor":"#E5ECF6","gridcolor":"white","linecolor":"white","showbackground":true,"ticks":"","zerolinecolor":"white","gridwidth":2},"yaxis":{"backgroundcolor":"#E5ECF6","gridcolor":"white","linecolor":"white","showbackground":true,"ticks":"","zerolinecolor":"white","gridwidth":2},"zaxis":{"backgroundcolor":"#E5ECF6","gridcolor":"white","linecolor":"white","showbackground":true,"ticks":"","zerolinecolor":"white","gridwidth":2}},"shapedefaults":{"line":{"color":"#2a3f5f"}},"annotationdefaults":{"arrowcolor":"#2a3f5f","arrowhead":0,"arrowwidth":1},"geo":{"bgcolor":"white","landcolor":"#E5ECF6","subunitcolor":"white","showland":true,"showlakes":true,"lakecolor":"white"},"title":{"x":0.05},"mapbox":{"style":"light"}}},"xaxis":{"anchor":"y","domain":[0.0,1.0],"scaleanchor":"y","constrain":"domain","title":{"text":"Position"}},"yaxis":{"anchor":"x","domain":[0.0,1.0],"autorange":"reversed","constrain":"domain","title":{"text":"Layer"}},"coloraxis":{"colorscale":[[0.0,"rgb(103,0,31)"],[0.1,"rgb(178,24,43)"],[0.2,"rgb(214,96,77)"],[0.3,"rgb(244,165,130)"],[0.4,"rgb(253,219,199)"],[0.5,"rgb(247,247,247)"],[0.6,"rgb(209,229,240)"],[0.7,"rgb(146,197,222)"],[0.8,"rgb(67,147,195)"],[0.9,"rgb(33,102,172)"],[1.0,"rgb(5,48,97)"]],"cmid":0.0},"margin":{"t":60}}} \ No newline at end of file diff --git a/docs/source/charts/line.json b/docs/source/charts/line.json deleted file mode 100644 index b3dae70c..00000000 --- a/docs/source/charts/line.json +++ /dev/null @@ -1,1083 +0,0 @@ -{ - "data": [ - { - "hovertemplate": "variable=0\u003cbr\u003eindex=%{x}\u003cbr\u003evalue=%{y}\u003cextra\u003e\u003c\u002fextra\u003e", - "legendgroup": "0", - "line": { - "color": "#636efa", - "dash": "solid" - }, - "marker": { - "symbol": "circle" - }, - "mode": "lines", - "name": "0", - "orientation": "v", - "showlegend": true, - "x": [ - 0, - 1, - 2, - 3, - 4, - 5, - 6, - 7, - 8, - 9, - 10, - 11, - 12, - 13, - 14, - 15, - 16, - 17, - 18, - 19, - 20, - 21, - 22, - 23, - 24, - 25, - 26, - 27, - 28, - 29, - 30, - 31, - 32, - 33, - 34, - 35, - 36, - 37, - 38, - 39, - 40, - 41, - 42, - 43, - 44, - 45, - 46, - 47, - 48, - 49, - 50, - 51, - 52, - 53, - 54, - 55, - 56, - 57, - 58, - 59, - 60, - 61, - 62, - 63, - 64, - 65, - 66, - 67, - 68, - 69, - 70, - 71, - 72, - 73, - 74, - 75, - 76, - 77, - 78, - 79, - 80, - 81, - 82, - 83, - 84, - 85, - 86, - 87, - 88, - 89, - 90, - 91, - 92, - 93, - 94, - 95, - 96, - 97, - 98, - 99, - 100, - 101, - 102, - 103, - 104 - ], - "xaxis": "x", - "y": [ - 6.721836090087891, - 15.863195419311523, - 14.229825019836426, - 14.620223999023438, - 12.628325462341309, - 14.076154708862305, - 13.74431037902832, - 14.042016983032227, - 12.240949630737305, - 12.101900100708008, - 12.948463439941406, - 13.237577438354492, - 11.980368614196777, - 12.275359153747559, - 12.520895004272461, - 12.39179801940918, - 11.573649406433105, - 11.729515075683594, - 12.298530578613281, - 11.90172004699707, - 12.044195175170898, - 12.171052932739258, - 12.23204231262207, - 10.854452133178711, - 10.37388801574707, - 12.753911972045898, - 12.342389106750488, - 11.449339866638184, - 12.227272033691406, - 12.660356521606445, - 12.355462074279785, - 11.243412017822266, - 11.748628616333008, - 12.428620338439941, - 11.892528533935547, - 12.418279647827148, - 12.380900382995605, - 12.087124824523926, - 11.42609977722168, - 12.237627983093262, - 12.026906967163086, - 10.750848770141602, - 11.18249225616455, - 10.964658737182617, - 12.21753978729248, - 11.411855697631836, - 12.105101585388184, - 11.151140213012695, - 11.415960311889648, - 10.983814239501953, - 10.562061309814453, - 12.14051628112793, - 10.84417724609375, - 10.993143081665039, - 7.57098913192749, - 5.991965293884277, - 2.293006420135498, - 1.1911180019378662, - 1.4175719022750854, - 0.5377484560012817, - 0.6352464556694031, - 0.3934612572193146, - 0.3168984055519104, - 1.2923715114593506, - 0.25427520275115967, - 0.12562693655490875, - 0.1655888557434082, - 0.05889231711626053, - 0.13178208470344543, - 0.11886248737573624, - 0.10017742961645126, - 0.28241199254989624, - 0.2901555001735687, - 0.07609564065933228, - 0.13399086892604828, - 0.20270712673664093, - 0.22564712166786194, - 0.1278029978275299, - 0.065616674721241, - 0.11973398923873901, - 0.03500190004706383, - 0.5889385938644409, - 0.05313155800104141, - 0.0918748527765274, - 0.1541168987751007, - 0.07680816948413849, - 0.18667154014110565, - 0.1084548830986023, - 0.3974376916885376, - 0.14485009014606476, - 0.13042660057544708, - 0.21075209975242615, - 0.02071751281619072, - 0.016112055629491806, - 0.04902868717908859, - 0.11065000295639038, - 0.11377827823162079, - 0.10953978449106216, - 0.10317499935626984, - 0.07640652358531952, - 0.4654570519924164, - 0.05069968104362488, - 0.09305594116449356, - 0.3460657298564911, - 0.14819812774658203 - ], - "yaxis": "y", - "type": "scatter" - } - ], - "layout": { - "template": { - "data": { - "histogram2dcontour": [ - { - "type": "histogram2dcontour", - "colorbar": { - "outlinewidth": 0, - "ticks": "" - }, - "colorscale": [ - [ - 0.0, - "#0d0887" - ], - [ - 0.1111111111111111, - "#46039f" - ], - [ - 0.2222222222222222, - "#7201a8" - ], - [ - 0.3333333333333333, - "#9c179e" - ], - [ - 0.4444444444444444, - "#bd3786" - ], - [ - 0.5555555555555556, - "#d8576b" - ], - [ - 0.6666666666666666, - "#ed7953" - ], - [ - 0.7777777777777778, - "#fb9f3a" - ], - [ - 0.8888888888888888, - "#fdca26" - ], - [ - 1.0, - "#f0f921" - ] - ] - } - ], - "choropleth": [ - { - "type": "choropleth", - "colorbar": { - "outlinewidth": 0, - "ticks": "" - } - } - ], - "histogram2d": [ - { - "type": "histogram2d", - "colorbar": { - "outlinewidth": 0, - "ticks": "" - }, - "colorscale": [ - [ - 0.0, - "#0d0887" - ], - [ - 0.1111111111111111, - "#46039f" - ], - [ - 0.2222222222222222, - "#7201a8" - ], - [ - 0.3333333333333333, - "#9c179e" - ], - [ - 0.4444444444444444, - "#bd3786" - ], - [ - 0.5555555555555556, - "#d8576b" - ], - [ - 0.6666666666666666, - "#ed7953" - ], - [ - 0.7777777777777778, - "#fb9f3a" - ], - [ - 0.8888888888888888, - "#fdca26" - ], - [ - 1.0, - "#f0f921" - ] - ] - } - ], - "heatmap": [ - { - "type": "heatmap", - "colorbar": { - "outlinewidth": 0, - "ticks": "" - }, - "colorscale": [ - [ - 0.0, - "#0d0887" - ], - [ - 0.1111111111111111, - "#46039f" - ], - [ - 0.2222222222222222, - "#7201a8" - ], - [ - 0.3333333333333333, - "#9c179e" - ], - [ - 0.4444444444444444, - "#bd3786" - ], - [ - 0.5555555555555556, - "#d8576b" - ], - [ - 0.6666666666666666, - "#ed7953" - ], - [ - 0.7777777777777778, - "#fb9f3a" - ], - [ - 0.8888888888888888, - "#fdca26" - ], - [ - 1.0, - "#f0f921" - ] - ] - } - ], - "heatmapgl": [ - { - "type": "heatmapgl", - "colorbar": { - "outlinewidth": 0, - "ticks": "" - }, - "colorscale": [ - [ - 0.0, - "#0d0887" - ], - [ - 0.1111111111111111, - "#46039f" - ], - [ - 0.2222222222222222, - "#7201a8" - ], - [ - 0.3333333333333333, - "#9c179e" - ], - [ - 0.4444444444444444, - "#bd3786" - ], - [ - 0.5555555555555556, - "#d8576b" - ], - [ - 0.6666666666666666, - "#ed7953" - ], - [ - 0.7777777777777778, - "#fb9f3a" - ], - [ - 0.8888888888888888, - "#fdca26" - ], - [ - 1.0, - "#f0f921" - ] - ] - } - ], - "contourcarpet": [ - { - "type": "contourcarpet", - "colorbar": { - "outlinewidth": 0, - "ticks": "" - } - } - ], - "contour": [ - { - "type": "contour", - "colorbar": { - "outlinewidth": 0, - "ticks": "" - }, - "colorscale": [ - [ - 0.0, - "#0d0887" - ], - [ - 0.1111111111111111, - "#46039f" - ], - [ - 0.2222222222222222, - "#7201a8" - ], - [ - 0.3333333333333333, - "#9c179e" - ], - [ - 0.4444444444444444, - "#bd3786" - ], - [ - 0.5555555555555556, - "#d8576b" - ], - [ - 0.6666666666666666, - "#ed7953" - ], - [ - 0.7777777777777778, - "#fb9f3a" - ], - [ - 0.8888888888888888, - "#fdca26" - ], - [ - 1.0, - "#f0f921" - ] - ] - } - ], - "surface": [ - { - "type": "surface", - "colorbar": { - "outlinewidth": 0, - "ticks": "" - }, - "colorscale": [ - [ - 0.0, - "#0d0887" - ], - [ - 0.1111111111111111, - "#46039f" - ], - [ - 0.2222222222222222, - "#7201a8" - ], - [ - 0.3333333333333333, - "#9c179e" - ], - [ - 0.4444444444444444, - "#bd3786" - ], - [ - 0.5555555555555556, - "#d8576b" - ], - [ - 0.6666666666666666, - "#ed7953" - ], - [ - 0.7777777777777778, - "#fb9f3a" - ], - [ - 0.8888888888888888, - "#fdca26" - ], - [ - 1.0, - "#f0f921" - ] - ] - } - ], - "mesh3d": [ - { - "type": "mesh3d", - "colorbar": { - "outlinewidth": 0, - "ticks": "" - } - } - ], - "scatter": [ - { - "fillpattern": { - "fillmode": "overlay", - "size": 10, - "solidity": 0.2 - }, - "type": "scatter" - } - ], - "parcoords": [ - { - "type": "parcoords", - "line": { - "colorbar": { - "outlinewidth": 0, - "ticks": "" - } - } - } - ], - "scatterpolargl": [ - { - "type": "scatterpolargl", - "marker": { - "colorbar": { - "outlinewidth": 0, - "ticks": "" - } - } - } - ], - "bar": [ - { - "error_x": { - "color": "#2a3f5f" - }, - "error_y": { - "color": "#2a3f5f" - }, - "marker": { - "line": { - "color": "#E5ECF6", - "width": 0.5 - }, - "pattern": { - "fillmode": "overlay", - "size": 10, - "solidity": 0.2 - } - }, - "type": "bar" - } - ], - "scattergeo": [ - { - "type": "scattergeo", - "marker": { - "colorbar": { - "outlinewidth": 0, - "ticks": "" - } - } - } - ], - "scatterpolar": [ - { - "type": "scatterpolar", - "marker": { - "colorbar": { - "outlinewidth": 0, - "ticks": "" - } - } - } - ], - "histogram": [ - { - "marker": { - "pattern": { - "fillmode": "overlay", - "size": 10, - "solidity": 0.2 - } - }, - "type": "histogram" - } - ], - "scattergl": [ - { - "type": "scattergl", - "marker": { - "colorbar": { - "outlinewidth": 0, - "ticks": "" - } - } - } - ], - "scatter3d": [ - { - "type": "scatter3d", - "line": { - "colorbar": { - "outlinewidth": 0, - "ticks": "" - } - }, - "marker": { - "colorbar": { - "outlinewidth": 0, - "ticks": "" - } - } - } - ], - "scattermapbox": [ - { - "type": "scattermapbox", - "marker": { - "colorbar": { - "outlinewidth": 0, - "ticks": "" - } - } - } - ], - "scatterternary": [ - { - "type": "scatterternary", - "marker": { - "colorbar": { - "outlinewidth": 0, - "ticks": "" - } - } - } - ], - "scattercarpet": [ - { - "type": "scattercarpet", - "marker": { - "colorbar": { - "outlinewidth": 0, - "ticks": "" - } - } - } - ], - "carpet": [ - { - "aaxis": { - "endlinecolor": "#2a3f5f", - "gridcolor": "white", - "linecolor": "white", - "minorgridcolor": "white", - "startlinecolor": "#2a3f5f" - }, - "baxis": { - "endlinecolor": "#2a3f5f", - "gridcolor": "white", - "linecolor": "white", - "minorgridcolor": "white", - "startlinecolor": "#2a3f5f" - }, - "type": "carpet" - } - ], - "table": [ - { - "cells": { - "fill": { - "color": "#EBF0F8" - }, - "line": { - "color": "white" - } - }, - "header": { - "fill": { - "color": "#C8D4E3" - }, - "line": { - "color": "white" - } - }, - "type": "table" - } - ], - "barpolar": [ - { - "marker": { - "line": { - "color": "#E5ECF6", - "width": 0.5 - }, - "pattern": { - "fillmode": "overlay", - "size": 10, - "solidity": 0.2 - } - }, - "type": "barpolar" - } - ], - "pie": [ - { - "automargin": true, - "type": "pie" - } - ] - }, - "layout": { - "autotypenumbers": "strict", - "colorway": [ - "#636efa", - "#EF553B", - "#00cc96", - "#ab63fa", - "#FFA15A", - "#19d3f3", - "#FF6692", - "#B6E880", - "#FF97FF", - "#FECB52" - ], - "font": { - "color": "#2a3f5f" - }, - "hovermode": "closest", - "hoverlabel": { - "align": "left" - }, - "paper_bgcolor": "white", - "plot_bgcolor": "#E5ECF6", - "polar": { - "bgcolor": "#E5ECF6", - "angularaxis": { - "gridcolor": "white", - "linecolor": "white", - "ticks": "" - }, - "radialaxis": { - "gridcolor": "white", - "linecolor": "white", - "ticks": "" - } - }, - "ternary": { - "bgcolor": "#E5ECF6", - "aaxis": { - "gridcolor": "white", - "linecolor": "white", - "ticks": "" - }, - "baxis": { - "gridcolor": "white", - "linecolor": "white", - "ticks": "" - }, - "caxis": { - "gridcolor": "white", - "linecolor": "white", - "ticks": "" - } - }, - "coloraxis": { - "colorbar": { - "outlinewidth": 0, - "ticks": "" - } - }, - "colorscale": { - "sequential": [ - [ - 0.0, - "#0d0887" - ], - [ - 0.1111111111111111, - "#46039f" - ], - [ - 0.2222222222222222, - "#7201a8" - ], - [ - 0.3333333333333333, - "#9c179e" - ], - [ - 0.4444444444444444, - "#bd3786" - ], - [ - 0.5555555555555556, - "#d8576b" - ], - [ - 0.6666666666666666, - "#ed7953" - ], - [ - 0.7777777777777778, - "#fb9f3a" - ], - [ - 0.8888888888888888, - "#fdca26" - ], - [ - 1.0, - "#f0f921" - ] - ], - "sequentialminus": [ - [ - 0.0, - "#0d0887" - ], - [ - 0.1111111111111111, - "#46039f" - ], - [ - 0.2222222222222222, - "#7201a8" - ], - [ - 0.3333333333333333, - "#9c179e" - ], - [ - 0.4444444444444444, - "#bd3786" - ], - [ - 0.5555555555555556, - "#d8576b" - ], - [ - 0.6666666666666666, - "#ed7953" - ], - [ - 0.7777777777777778, - "#fb9f3a" - ], - [ - 0.8888888888888888, - "#fdca26" - ], - [ - 1.0, - "#f0f921" - ] - ], - "diverging": [ - [ - 0, - "#8e0152" - ], - [ - 0.1, - "#c51b7d" - ], - [ - 0.2, - "#de77ae" - ], - [ - 0.3, - "#f1b6da" - ], - [ - 0.4, - "#fde0ef" - ], - [ - 0.5, - "#f7f7f7" - ], - [ - 0.6, - "#e6f5d0" - ], - [ - 0.7, - "#b8e186" - ], - [ - 0.8, - "#7fbc41" - ], - [ - 0.9, - "#4d9221" - ], - [ - 1, - "#276419" - ] - ] - }, - "xaxis": { - "gridcolor": "white", - "linecolor": "white", - "ticks": "", - "title": { - "standoff": 15 - }, - "zerolinecolor": "white", - "automargin": true, - "zerolinewidth": 2 - }, - "yaxis": { - "gridcolor": "white", - "linecolor": "white", - "ticks": "", - "title": { - "standoff": 15 - }, - "zerolinecolor": "white", - "automargin": true, - "zerolinewidth": 2 - }, - "scene": { - "xaxis": { - "backgroundcolor": "#E5ECF6", - "gridcolor": "white", - "linecolor": "white", - "showbackground": true, - "ticks": "", - "zerolinecolor": "white", - "gridwidth": 2 - }, - "yaxis": { - "backgroundcolor": "#E5ECF6", - "gridcolor": "white", - "linecolor": "white", - "showbackground": true, - "ticks": "", - "zerolinecolor": "white", - "gridwidth": 2 - }, - "zaxis": { - "backgroundcolor": "#E5ECF6", - "gridcolor": "white", - "linecolor": "white", - "showbackground": true, - "ticks": "", - "zerolinecolor": "white", - "gridwidth": 2 - } - }, - "shapedefaults": { - "line": { - "color": "#2a3f5f" - } - }, - "annotationdefaults": { - "arrowcolor": "#2a3f5f", - "arrowhead": 0, - "arrowwidth": 1 - }, - "geo": { - "bgcolor": "white", - "landcolor": "#E5ECF6", - "subunitcolor": "white", - "showland": true, - "showlakes": true, - "lakecolor": "white" - }, - "title": { - "x": 0.05 - }, - "mapbox": { - "style": "light" - } - } - }, - "xaxis": { - "anchor": "y", - "domain": [ - 0.0, - 1.0 - ], - "title": { - "text": "index" - } - }, - "yaxis": { - "anchor": "x", - "domain": [ - 0.0, - 1.0 - ], - "title": { - "text": "value" - } - }, - "legend": { - "title": { - "text": "variable" - }, - "tracegroupgap": 0 - }, - "margin": { - "t": 60 - } - } -} \ No newline at end of file diff --git a/docs/source/conf.py b/docs/source/conf.py index b09d7c01..12b179d4 100644 --- a/docs/source/conf.py +++ b/docs/source/conf.py @@ -1,8 +1,10 @@ +import time + # Configuration file for the Sphinx documentation builder. # Project Information project = "nnsight" -copyright = "2023, NDIF" +copyright = "2024 NDIF" author = "Jaden Fiotto-Kaufman" @@ -26,19 +28,21 @@ html_title = "nnsight" html_logo = "_static/images/nnsight_logo.svg" html_static_path = ["_static"] +html_show_sphinx = False html_favicon = "_static/images/icon.ico" html_show_sourcelink = False html_context = { "default_mode": "dark", - "ndif_url": "https://ndif.dev/ping" + "ndif_url": "https://ndif.dev/ping", + "version_identifier": str(int(time.time())), } html_theme_options = { "show_nav_level": 2, - "navbar_end": ["navbar-icon-links", "ndif_status"], + "navbar_end": ["theme-switcher","navbar-icon-links", "ndif_status"], "navbar_align": "left", "icon_links": [ { @@ -52,25 +56,17 @@ "icon": "fa-brands fa-discord", }, { - "name": "Status", + "name": "Status: Unknown", "url": "/status", "icon": "fa-solid fa-circle-check", "attributes": {"class": "ndif"}, }, ], "show_prev_next": False, - "default_mode": "dark", - "pygment_dark_style": "monokai" + "pygment_dark_style": "monokai", + "pygment_light_style": "solarized-light" } -pygments_style = "rrt" - -html_css_files = [ - "css/custom.css", - "css/front.css", - "css/hljs.css" -] - html_js_files = [ 'js/custom.js', 'js/code.js' diff --git a/docs/source/features.rst b/docs/source/features.rst index 947da101..65d81ef4 100644 --- a/docs/source/features.rst +++ b/docs/source/features.rst @@ -1,5 +1,5 @@ Features -========== +========= .. raw:: html @@ -22,8 +22,8 @@ Features .. grid-item-card:: :link: notebooks/features/getting.ipynb - :class-card: code-surface - :class-body: code-surface + :class-card: surface + :class-body: surface .. raw:: html @@ -31,7 +31,7 @@ Features <div class="d-flex align-items-center"> <div class="d-flex justify-content-center" style="min-width: 50px; margin-right: 20px; height: 100%;"> <i class="fa-solid fa-wrench fa-2x"></i> - </div> <!-- Replace fa-icon-name with the correct icon class --> + </div> <div> <h5 class="card-title">Getting</h5> <p class="card-text">Accessing values</p> @@ -40,8 +40,8 @@ Features .. grid-item-card:: :link: notebooks/features/setting.ipynb - :class-card: code-surface - :class-body: code-surface + :class-card: surface + :class-body: surface .. raw:: html @@ -49,7 +49,7 @@ Features <div class="d-flex align-items-center"> <div class="d-flex justify-content-center" style="min-width: 50px; margin-right: 20px; height: 100%;"> <i class="fa-solid fa-code-pull-request fa-2x"></i> - </div> <!-- Replace fa-icon-name with the correct icon class --> + </div> <div> <h5 class="card-title">Setting</h5> <p class="card-text">Intervening on values</p> @@ -58,8 +58,8 @@ Features .. grid-item-card:: :link: notebooks/features/operations.ipynb - :class-card: code-surface - :class-body: code-surface + :class-card: surface + :class-body: surface .. raw:: html @@ -67,7 +67,7 @@ Features <div class="d-flex align-items-center"> <div class="d-flex justify-content-center" style="min-width: 50px; margin-right: 20px; height: 100%;"> <i class="fa-solid fa-glasses fa-2x"></i> - </div> <!-- Replace fa-icon-name with the correct icon class --> + </div> <div> <h5 class="card-title">Operations</h5> <p class="card-text">Editing values</p> @@ -76,8 +76,8 @@ Features .. grid-item-card:: :link: notebooks/features/modules.ipynb - :class-card: code-surface - :class-body: code-surface + :class-card: surface + :class-body: surface .. raw:: html @@ -85,7 +85,7 @@ Features <div class="d-flex align-items-center"> <div class="d-flex justify-content-center" style="min-width: 50px; margin-right: 20px; height: 100%;"> <i class="fa-solid fa-cubes fa-2x"></i> - </div> <!-- Replace fa-icon-name with the correct icon class --> + </div> <div> <h5 class="card-title">Modules</h5> <p class="card-text">Applying modules</p> @@ -94,8 +94,8 @@ Features .. grid-item-card:: :link: notebooks/features/cross_prompt.ipynb - :class-card: code-surface - :class-body: code-surface + :class-card: surface + :class-body: surface .. raw:: html @@ -112,8 +112,8 @@ Features .. grid-item-card:: :link: notebooks/features/multiple_token.ipynb - :class-card: code-surface - :class-body: code-surface + :class-card: surface + :class-body: surface .. raw:: html @@ -121,7 +121,7 @@ Features <div class="d-flex align-items-center"> <div class="d-flex justify-content-center" style="min-width: 50px; margin-right: 20px; height: 100%;"> <i class="fa-solid fa-gears fa-2x"></i> - </div> <!-- Replace fa-icon-name with the correct icon class --> + </div> <div> <h5 class="card-title">Generation</h5> <p class="card-text">Multiple tokens</p> @@ -130,8 +130,8 @@ Features .. grid-item-card:: :link: notebooks/features/gradients.ipynb - :class-card: code-surface - :class-body: code-surface + :class-card: surface + :class-body: surface .. raw:: html @@ -139,7 +139,7 @@ Features <div class="d-flex align-items-center"> <div class="d-flex justify-content-center" style="min-width: 50px; margin-right: 20px; height: 100%;"> <i class="fa-solid fa-backward fa-2x"></i> - </div> <!-- Replace fa-icon-name with the correct icon class --> + </div> <div> <h5 class="card-title">Working With Gradients</h5> <p class="card-text">Intervene on gradients</p> @@ -148,8 +148,8 @@ Features .. grid-item-card:: :link: notebooks/features/remote_execution.ipynb - :class-card: code-surface - :class-body: code-surface + :class-card: surface + :class-body: surface .. raw:: html @@ -157,7 +157,7 @@ Features <div class="d-flex align-items-center"> <div class="d-flex justify-content-center" style="min-width: 50px; margin-right: 20px; height: 100%;"> <i class="fa-solid fa-satellite-dish fa-2x"></i> - </div> <!-- Replace fa-icon-name with the correct icon class --> + </div> <div> <h5 class="card-title">Remote Execution</h5> <p class="card-text">Transparent deep learning</p> @@ -166,15 +166,15 @@ Features .. grid-item-card:: :link: notebooks/features/token_indexing.ipynb - :class-card: code-surface - :class-body: code-surface + :class-card: surface + :class-body: surface .. raw:: html <div class="d-flex align-items-center"> <div class="d-flex justify-content-center" style="min-width: 50px; margin-right: 20px; height: 100%;"> <i class="fa-regular fa-object-ungroup fa-2x"></i> - </div> <!-- Replace fa-icon-name with the correct icon class --> + </div> <div> <h5 class="card-title">Multi-Token Indexing</h5> <p class="card-text">Easy token indexing</p> diff --git a/docs/source/index.rst b/docs/source/index.rst index 8f4da88f..c57c20a2 100644 --- a/docs/source/index.rst +++ b/docs/source/index.rst @@ -33,31 +33,27 @@ nnsight </script> .. raw:: html - - <link rel="stylesheet" href="https://cdn.jsdelivr.net/npm/[email protected]/font/bootstrap-icons.min.css"> - <script src="dist/clipboard.min.js"></script> - - <script> - new ClipboardJS('.btn'); - </script> - - <script> - // Select the element with the class 'some-class' - var element = document.querySelector('.bd-container'); - - var newElement = document.createElement('div'); - newElement.id = 'canvas'; - - newElement.style.position = 'fixed'; - newElement.style.zIndex = '-1'; - newElement.style.height = '100vh'; - newElement.style.width = '100vw'; - newElement.style.top = '0'; - - element.appendChild(newElement); - - </script> + <script type="text/javascript"> + document.addEventListener('DOMContentLoaded', function() { + var observer = new MutationObserver(function(mutations) { + updateCSSLink(); + }); + observer.observe(document.documentElement, {attributes: true, attributeFilter: ['data-theme']}); + + function updateCSSLink() { + const dark = document.documentElement.dataset.theme == 'dark'; + var link = document.querySelector('link[rel="stylesheet"][href="hljs"]'); + if(link) { + link.href = dark ? '../_static/css/hljs-dark.css?v=0.2' : '../_static/css/hljs-light.css?v=0.2'; + } + } + + // Initial check in case the theme is already set before the observer is added + updateCSSLink(); + }); + </script> + <link rel="stylesheet" href="../_static/css/hljs-dark.css?v=0.2"> .. raw:: html @@ -82,7 +78,7 @@ nnsight </div> </section> - <section class="d-flex align-items-center middle-stuff mid-section"> + <section class="d-flex align-items-center mid-section d-none d-md-block"> <div class="px-4 text-center"> <div class="col-lg-10 mx-auto"> <p class="lead mb-4">Direct access to model internals, from one to one hundred billion parameters. Intervene on activations or gradients in transformers, train optimizations methods, perform out of order module applications, cross prompt interventions, and so much more.</p> @@ -90,7 +86,7 @@ nnsight </div> </section> - <section class="d-flex align-items-center" style="height: 60vh;"> + <section class="d-flex align-items-center d-none d-md-block" style="height: 60vh;"> <div class="container"> <div class="row align-items-end pb-3 border-bottom mb-5"> <div class="col"> @@ -135,4 +131,28 @@ nnsight </div> </section> - <section style="height: 20vh;"/> + <section class="d-none d-md-block" style="height: 20vh;"/> + + +.. raw:: html + + <script> + + var element = document.querySelector('.bd-container'); + + if (element) { + // If .bd-container exists, proceed to create the new element + var newElement = document.createElement('div'); + newElement.id = 'canvas'; + + newElement.style.position = 'fixed'; + newElement.style.zIndex = '-1'; + newElement.style.height = '100vh'; + newElement.style.width = '100vw'; + newElement.style.top = '0'; + + element.appendChild(newElement); + } else { + console.log(".bd-container element does not exist."); + } + </script> \ No newline at end of file diff --git a/docs/source/notebooks/tutorials/walkthrough.ipynb b/docs/source/notebooks/tutorials/walkthrough.ipynb index b148dafc..4ad70d96 100644 --- a/docs/source/notebooks/tutorials/walkthrough.ipynb +++ b/docs/source/notebooks/tutorials/walkthrough.ipynb @@ -1,16 +1,18 @@ { "cells": [ + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "# Walkthrough" + ] + }, { "cell_type": "markdown", "metadata": { "id": "-Scfr942GwO4" }, "source": [ - "<img src=\"https://nnsight.net/_images/nnsight_logo.svg\" alt=\"drawing\" width=\"200\"/>\n", - "\n", - "# NNsight\n", - "## The API for a transparent science on black-box AI\n", - "\n", "An interactive version of this walkthrough can be found [here](https://colab.research.google.com/github/JadenFiotto-Kaufman/nnsight/blob/dev/NNsight_v0_2.ipynb)\n", "\n", "In this era of large-scale deep learning, the most interesting AI models are massive black boxes that are hard to run. Ordinary commercial inference service APIs let you interact with huge models, but they do not let you access model internals.\n", @@ -54,7 +56,7 @@ "outputs": [], "source": [ "# Install nnsight\n", - "!pip install nnsight==0.2.3.dev0\n", + "!pip install nnsight\n", "\n", "from IPython.display import clear_output\n", "\n", @@ -1530,7 +1532,8 @@ "name": "python3" }, "language_info": { - "name": "python" + "name": "python", + "version": "3.8.0" }, "widgets": { "application/vnd.jupyter.widget-state+json": { diff --git a/docs/source/start.rst b/docs/source/start.rst index 7bec657a..732ca5f1 100644 --- a/docs/source/start.rst +++ b/docs/source/start.rst @@ -12,7 +12,7 @@ To get started with NNsight, install it with ``pip``. .. code-block:: console - pip install nnsight==0.2.3.dev0 + pip install nnsight First Steps diff --git a/docs/source/status.rst b/docs/source/status.rst index 4f308d6a..00d0650e 100644 --- a/docs/source/status.rst +++ b/docs/source/status.rst @@ -3,7 +3,50 @@ .. raw:: html - <link rel="stylesheet" href="../_static/css/status.css"> + <style> + .accordion-header { + margin: 0 !important; + } + /* Custom accordion styles / + .custom-accordion-header { + background-color: var(--pst-color-surface); / Default state background color / + color: var(--pst-color-text-base); / Text color / + border-bottom: var(--pst-color-border); / Border color / + } + + .custom-accordion-header: { + background-color: var(--pst-color-surface); / Default state background color / + color: var(--pst-color-text-base); / Text color / + border-bottom: var(--pst-color-border); / Border color / + } + + .accordion { + --bs-accordion-btn-icon: none; + --bs-accordion-btn-active-icon: none; + } + + .custom-accordion-header.collapsed { + background-color: var(--pst-color-on-background); / Collapsed state background color / + color: var(--pst-color-text-base); / Text color / + } + + .custom-accordion-header:not(.collapsed) { + background-color: var(--pst-color-surface); / Active/Expanded state background color / + color: var(--pst-color-text-base); / Text color / + } + + .custom-accordion-body { + background-color: var(--pst-color-on-background); / Body background color / + border-color: var(--pst-color-border); / Border color / + color: var(--pst-color-text-base); / Text color / + } + + .sd-card { + border-radius: 0 !important; + } + </style> + + <script> function autoFormatJsonString(jsonString) { @@ -24,13 +67,14 @@ fetch("https://ndif.dev/ping") .then((response) => { if (response.status == 200) { - Array.from(document.getElementsByClassName("status-container")).forEach((elm) => { - elm.style.backgroundColor = "#66800b"; - Array.from(elm.getElementsByClassName('sd-card-text')).forEach((text) => { - text.textContent = "All Systems Are Operational"; - text.style.color = "#CECDC3"; + document.querySelectorAll('div.sd-card-body.status-container').forEach(el => { + el.style.backgroundColor = "#66800b"; + el.querySelectorAll('p.sd-card-text').forEach(el => { + el.textContent = "All Systems Are Operational"; }); }); + + console.log('Ping success'); // Nested fetch to ndif.dev/stats @@ -82,7 +126,17 @@ prettyPrintedJson = prettyPrintedJson.replace(/"([^"]+)":/g, '"<b>$1</b>":'); let huggingFaceLink = `<a href="http://huggingface.co/${key}" target="_blank">HuggingFace Model Repository ↗</a>`; - infoString += `<div class="accordion-item"> <h2 class="accordion-header" id="${headingId}"> <button class="accordion-button collapsed" type="button" data-bs-toggle="collapse" data-bs-target="#${collapseId}" aria-expanded="false" aria-controls="${collapseId}"> (${summaryItem.number_of_copies}x) ${key} </button> </h2> <div id="${collapseId}" class="accordion-collapse collapse" aria-labelledby="${headingId}" data-bs-parent="#accordionExample"> <div class="accordion-body">${huggingFaceLink}</i> <pre>${prettyPrintedJson}</pre></div> </div> </div>`; + infoString += `<div class="accordion-item"> + <h2 class="accordion-header" id="${headingId}"> + <button class="accordion-button custom-accordion-header collapsed" type="button" data-bs-toggle="collapse" data-bs-target="#${collapseId}" aria-expanded="false" aria-controls="${collapseId}"> + (${summaryItem.number_of_copies}x) ${key} + </button> + </h2> + <div id="${collapseId}" class="accordion-collapse collapse" aria-labelledby="${headingId}" data-bs-parent="#accordionExample"> + <div class="accordion-body custom-accordion-body">${huggingFaceLink}<pre>${prettyPrintedJson}</pre></div> + </div> + </div>`; + index++; }); @@ -92,8 +146,6 @@ elm.innerHTML = infoString; - - console.log('Stats success'); }).catch((jsonError) => { console.log('JSON parsing error:', jsonError); @@ -106,22 +158,20 @@ console.log('Stats error'); }); } else { - Array.from(document.getElementsByClassName("status-container")).forEach((elm) => { - elm.style.backgroundColor = "#F84F31"; - Array.from(elm.getElementsByClassName('sd-card-text')).forEach((text) => { - text.textContent = "NDIF Is Unavailable"; - text.style.color = "#CECDC3"; + document.querySelectorAll('div.sd-card-body.status-container').forEach(el => { + el.style.backgroundColor = "#F84F31"; + el.querySelectorAll('p.sd-card-text').forEach(el => { + el.textContent = "NDIF Is Unavailable"; }); }); console.log('Ping error'); } }) .catch((pingError) => { - Array.from(document.getElementsByClassName("status-container")).forEach((elm) => { - elm.style.backgroundColor = "#F84F31"; - Array.from(elm.getElementsByClassName('sd-card-text')).forEach((text) => { - text.textContent = "NDIF Is Unavailable"; - text.style.color = "#CECDC3"; + document.querySelectorAll('div.sd-card-body.status-container').forEach(el => { + el.style.backgroundColor = "#F84F31"; + el.querySelectorAll('p.sd-card-text').forEach(el => { + el.textContent = "NDIF Is Unavailable"; }); }); console.error('Ping fetch failed:', pingError); @@ -139,10 +189,16 @@ Status :hidden: .. card:: - :class-card: status-container + :class-body: status-container + :shadow: none All Systems Are Operational +.. card:: + :shadow: none + + The library can be used to run local models without a key. Running experiments on remote models requires a free server API key. To obtain a key, join the `NDIF Discord <https://discord.gg/6uFJmCSwW7>`_ and say hello on the #introductions channel. + .. raw:: html <div class="accordion accordion-flush" id="accordionHook"> diff --git a/docs/source/tutorials.rst b/docs/source/tutorials.rst index 1728168a..3ad61261 100644 --- a/docs/source/tutorials.rst +++ b/docs/source/tutorials.rst @@ -28,7 +28,7 @@ Tutorials .. raw:: html <div class="d-flex align-items-center"> - <img src="../_static/images/tutorials/walkthrough.png" class="img-fluid" style="max-width: 50px; margin-right: 20px;"> + <img src="../_static/images/tutorials/walkthrough.webp" class="img-fluid" style="max-width: 50px; margin-right: 20px;"> <div> <h5 class="code-surface card-title">Walkthrough</h5> <p class="code-surface card-text">Learn the basics</p> @@ -43,7 +43,7 @@ Tutorials .. raw:: html <div class="d-flex align-items-center"> - <img src="../_static/images/tutorials/activation_patching.png" class="img-fluid" style="max-width: 50px; margin-right: 20px;"> + <img src="../_static/images/tutorials/activation_patching.webp" class="img-fluid" style="max-width: 50px; margin-right: 20px;"> <div> <h5 class="code-surface card-title">Activation Patching</h5> <p class="code-surface card-text">Causal intervention</p> @@ -58,7 +58,7 @@ Tutorials .. raw:: html <div class="d-flex align-items-center"> - <img src="../_static/images/tutorials/attribution_patching.png" class="img-fluid" style="max-width: 50px; margin-right: 20px;"> + <img src="../_static/images/tutorials/attribution_patching.webp" class="img-fluid" style="max-width: 50px; margin-right: 20px;"> <div> <h5 class="code-surface card-title">Attribution Patching</h5> <p class="code-surface card-text">Approximate patching</p> @@ -73,7 +73,7 @@ Tutorials .. raw:: html <div class="d-flex align-items-center"> - <img src="../_static/images/tutorials/logit_lens.png" class="img-fluid" style="max-width: 50px; margin-right: 20px;"> + <img src="../_static/images/tutorials/logit_lens.webp" class="img-fluid" style="max-width: 50px; margin-right: 20px;"> <div> <h5 class="code-surface card-title">Logit Lens</h5> <p class="code-surface card-text">Decode activations</p> @@ -88,7 +88,7 @@ Tutorials .. raw:: html <div class="d-flex align-items-center"> - <img src="../_static/images/tutorials/future_lens.png" class="img-fluid" style="max-width: 50px; margin-right: 20px;"> + <img src="../_static/images/tutorials/future_lens.webp" class="img-fluid" style="max-width: 50px; margin-right: 20px;"> <div> <h5 class="code-surface card-title">Future Lens</h5> <p class="code-surface card-text">Probe future tokens</p> @@ -103,7 +103,7 @@ Tutorials .. raw:: html <div class="d-flex align-items-center"> - <img src="../_static/images/tutorials/function_vectors.png" class="img-fluid" style="max-width: 50px; margin-right: 20px;"> + <img src="../_static/images/tutorials/function_vectors.webp" class="img-fluid" style="max-width: 50px; margin-right: 20px;"> <div> <h5 class="code-surface card-title">Function Vectors</h5> <p class="code-surface card-text">Lambdas</p> @@ -118,7 +118,7 @@ Tutorials .. raw:: html <div class="d-flex align-items-center"> - <img src="../_static/images/tutorials/dictionary_learning.png" class="img-fluid" style="max-width: 50px; margin-right: 20px;"> + <img src="../_static/images/tutorials/dictionary_learning.webp" class="img-fluid" style="max-width: 50px; margin-right: 20px;"> <div> <h5 class="code-surface card-title">Dictionary Learning</h5> <p class="code-surface card-text">Sparse autoencoders</p> diff --git a/src/.DS_Store b/src/.DS_Store new file mode 100644 index 00000000..1b454a83 Binary files /dev/null and b/src/.DS_Store differ diff --git a/src/nnsight/__init__.py b/src/nnsight/__init__.py index 5c9a94a2..218e7bb6 100644 --- a/src/nnsight/__init__.py +++ b/src/nnsight/__init__.py @@ -2,7 +2,7 @@ import os import yaml - +import torch from .patching import from .pydantics.Config import ConfigModel @@ -41,20 +41,49 @@ def _bool(self): def fake_tensor_new_wrapper(fn): - + @wraps(fn) def inner(cls, fake_mode, elem, device, constant=None): - + if isinstance(elem, FakeTensor): return elem - + else: return fn(cls, fake_mode, elem, device, constant=constant) - + return inner -DEFAULT_PATCHER.add(Patch(FakeTensor, fake_tensor_new_wrapper(FakeTensor.__new__), "__new__")) + +DEFAULT_PATCHER.add( + Patch(FakeTensor, fake_tensor_new_wrapper(FakeTensor.__new__), "__new__") +) + + +def onehot_wrapper(fn): + @wraps(fn) + def onehot(input: torch.Tensor, num_classes=-1): + if input.device.type == "meta": + return torch.zeros((input.shape, num_classes), device="meta") + + else: + return fn(input, num_classes=num_classes) + + return onehot + + +DEFAULT_PATCHER.add( + Patch(torch.nn.functional, onehot_wrapper(torch.nn.functional.one_hot), "one_hot") +) + +def noop_wrapper(fn): + @wraps(fn) + def noop(input: torch.Tensor, args, kwargs): + return input + + return noop + +DEFAULT_PATCHER.add(Patch(FakeTensor, noop_wrapper(FakeTensor.tolist), "tolist")) DEFAULT_PATCHER.__enter__() diff --git a/src/nnsight/config.yaml b/src/nnsight/config.yaml index ef2dc060..98cdc221 100644 --- a/src/nnsight/config.yaml +++ b/src/nnsight/config.yaml @@ -1,5 +1,5 @@ API: - APIKEY: vZ8HcTCC436SEX3qTfsq + APIKEY: null HOST: ndif.dev APP: LOGGING: false diff --git a/src/nnsight/contexts/Runner.py b/src/nnsight/contexts/Runner.py index 8dff9799..31366502 100644 --- a/src/nnsight/contexts/Runner.py +++ b/src/nnsight/contexts/Runner.py @@ -39,10 +39,10 @@ def __exit__(self, exc_type, exc_val, exc_tb) -> None: if isinstance(exc_val, BaseException): raise exc_val - self._graph.tracing = False - if self.remote: self.run_server() + + self._graph = None else: super().__exit__(exc_type, exc_val, exc_tb) @@ -110,7 +110,7 @@ def handle_response(self, event: str, data) -> bool: return True # Or if there was some error. elif response.status == pydantics.ResponseModel.JobStatus.ERROR: - return True + raise Exception(str(response)) return False diff --git a/src/nnsight/contexts/Tracer.py b/src/nnsight/contexts/Tracer.py index 7dfd626d..2110774b 100644 --- a/src/nnsight/contexts/Tracer.py +++ b/src/nnsight/contexts/Tracer.py @@ -64,8 +64,6 @@ def __exit__(self, exc_type, exc_val, exc_tb) -> None: if isinstance(exc_val, BaseException): raise exc_val - self._graph.tracing = False - output = self._model.interleave( self._model._execute, self._graph, @@ -73,6 +71,8 @@ def __exit__(self, exc_type, exc_val, exc_tb) -> None: self._kwargs, ) + self._graph = None + def invoke(self, inputs: Tuple[Any], kwargs) -> Invoker: """Create an Invoker context dor a given input. diff --git a/src/nnsight/envoy.py b/src/nnsight/envoy.py index 352e11b8..55dc6322 100644 --- a/src/nnsight/envoy.py +++ b/src/nnsight/envoy.py @@ -1,5 +1,6 @@ from __future__ import annotations +import warnings from typing import Any, Dict, Iterator, List, Optional, Union import torch @@ -59,7 +60,45 @@ def __init__(self, module: torch.nn.Module, module_path: str = ""): self._sub_envoys.append(envoy) - setattr(self, name, envoy) + # If the module already has a sub-module named 'input' or 'output', + # mount the proxy access to 'nns_input' or 'nns_output instead. + if hasattr(Envoy, name): + + self._handle_overloaded_mount(envoy, name) + + else: + + setattr(self, name, envoy) + + def _handle_overloaded_mount(self, envoy: Envoy, mount_point: str): + + warnings.warn( + f"Module of type `{type(self._module)}` has pre-defined a `{mount_point}` attribute. nnsight access for `{mount_point}` will be mounted at `.nns_{mount_point}` instead of `.{mount_point}` for this module only." + ) + + # If we already shifted a mount point dont create another new class. + if "Preserved" in self.__class__.__name__: + + new_cls = self.__class__ + + else: + + new_cls = type( + f"{Envoy.__name__}.Preserved", + (Envoy,), + {}, + ) + + # Get the normal proxy mount point + mount = getattr(new_cls, mount_point) + + # Move it to nns_<mount point> + setattr(new_cls, f"nns_{mount_point}", mount) + # Set the sub-module/envoy to the normal mount point on the CLASS itself not the instance. + setattr(new_cls, mount_point, envoy) + + # Update the class on the instance + self.__class__ = new_cls def _set_tracer(self, tracer: Tracer, propagate=True): """Set tracer object on Envoy. @@ -145,14 +184,76 @@ def next(self, increment: int = 1, propagate: bool = False) -> Envoy: return self - def __str__(self) -> str: + def _repr_module_list(self): + + list_of_reprs = [repr(item) for item in self._sub_envoys] + if len(list_of_reprs) == 0: + return self._module._get_name() + "()" + + start_end_indices = [[0, 0]] + repeated_blocks = [list_of_reprs[0]] + for i, r in enumerate(list_of_reprs[1:], 1): + if r == repeated_blocks[-1]: + start_end_indices[-1][1] += 1 + continue + + start_end_indices.append([i, i]) + repeated_blocks.append(r) + + lines = [] + main_str = self._module._get_name() + "(" + for (start_id, end_id), b in zip(start_end_indices, repeated_blocks): + local_repr = f"({start_id}): {b}" # default repr + + if start_id != end_id: + n = end_id - start_id + 1 + local_repr = f"({start_id}-{end_id}): {n} x {b}" + + local_repr = torch.nn.modules.module._addindent(local_repr, 2) + lines.append(local_repr) + + main_str += "\n " + "\n ".join(lines) + "\n" + main_str += ")" + return main_str + + def __repr__(self) -> str: """Wrapper method for underlying module's string representation. Returns: str: String. """ - return str(self._module) + if isinstance(self._module, torch.nn.ModuleList): + + return self._repr_module_list() + + extra_lines = [] + extra_repr = self._module.extra_repr() + # empty string will be split into list [''] + if extra_repr: + extra_lines = extra_repr.split("\n") + child_lines = [] + for attribute_name, attribute in self.__dict__.items(): + + if isinstance(attribute, Envoy): + + mod_str = repr(attribute) + mod_str = torch.nn.modules.module._addindent(mod_str, 2) + child_lines.append("(" + attribute_name + "): " + mod_str) + + lines = extra_lines + child_lines + + main_str = self._module._get_name() + "(" + if lines: + # simple one-liner info, which most builtin Modules will use + if len(extra_lines) == 1 and not child_lines: + main_str += extra_lines[0] + else: + main_str += "\n " + "\n ".join(lines) + "\n" + + main_str += ")" + + return main_str def __iter__(self) -> Iterator[Envoy]: """Wrapper method for underlying ModuleList iterator. @@ -214,8 +315,9 @@ def __call__(self, args: List[Any], *kwargs: Dict[str, Any]) -> InterventionPr torch._GLOBAL_DEVICE_CONTEXT = None return proxy - + torch.set_default_device + @property def output(self) -> InterventionProxy: """ diff --git a/src/nnsight/intervention.py b/src/nnsight/intervention.py index 0ec31e1b..c2b2af8f 100644 --- a/src/nnsight/intervention.py +++ b/src/nnsight/intervention.py @@ -64,7 +64,7 @@ def save(self) -> InterventionProxy: # Add a 'null' node with the save proxy as an argument to ensure the values are never deleted. # This is because 'null' nodes never actually get set and therefore there will always be a # dependency for the save proxy. - self.node.graph.add( + self.node.add( value=None, target="null", args=[self.node], @@ -83,8 +83,13 @@ def grad(self) -> InterventionProxy: """ if self._grad is None: - self.__dict__["_grad"] = self.node.graph.add( - value=self.node.proxy_value, target="grad", args=[self.node] + # We track how many times backward is called via an attribute on the Graph + if not hasattr(self.node.graph, 'n_backward_calls'): + + setattr(self.node.graph, 'n_backward_calls', 0) + + self.__dict__["_grad"] = self.node.add( + value=self.node.proxy_value, target="grad", args=[self.node,self.node.graph.n_backward_calls] ) return self._grad @@ -97,9 +102,45 @@ def grad(self, value: Union[InterventionProxy, Any]) -> None: Args: value (Union[InterventionProxy, Any]): Value to set output to. """ - self.node.graph.add(target="swap", args=[self.grad.node, value], value=True) + self.node.add(target="swap", args=[self.grad.node, value], value=True) - self.__dict__["_grad"] = None + def __call__(self, args, *kwargs) -> Self: + + + # We don't want to call backward on fake tensors + if ( + self.node.target is util.fetch_attr + and isinstance(self.node.args[1], str) + and self.node.args[1] == "backward" + ): + # We track how many times backward is called via an attribute on the Graph + if not hasattr(self.node.graph, 'n_backward_calls'): + + setattr(self.node.graph, 'n_backward_calls', 0) + + # Clear all .grad proxies + for node in self.node.graph.nodes.values(): + + try: + + if node.proxy._grad is not None: + + node.proxy.__dict__['_grad'] = None + + except ReferenceError: + pass + + self.node.graph.n_backward_calls += 1 + + return self.node.add( + value=None, + target=Proxy.proxy_call, + args=[self.node] + list(args), + kwargs=kwargs, + ) + + + return super().__call__(args, *kwargs) def __setattr__( self, key: Union[InterventionProxy, Any], value: Union[Self, Any] @@ -118,12 +159,26 @@ def shape(self) -> Collection[torch.Size]: Union[torch.Size,Collection[torch.Size]]: Proxy value shape or collection of shapes. """ - if not self.node.graph.tracing: + if self.node.is_graph_dereferenced(): return util.apply(self.value, lambda x: x.shape, torch.Tensor) return util.apply(self.node.proxy_value, lambda x: x.shape, torch.Tensor) + @property + def device(self) -> Collection[torch.device]: + """Property to retrieve the device of the traced proxy value or real value. + + Returns: + Union[torch.Size,Collection[torch.device]]: Proxy value shape or collection of shapes. + """ + + if self.node.is_graph_dereferenced(): + + return util.apply(self.value, lambda x: x.device, torch.Tensor) + + return util.apply(self.node.proxy_value, lambda x: x.device, torch.Tensor) + def concat( activations: Any, diff --git a/src/nnsight/tracing/Graph.py b/src/nnsight/tracing/Graph.py index c16e268c..3c63e00a 100644 --- a/src/nnsight/tracing/Graph.py +++ b/src/nnsight/tracing/Graph.py @@ -1,6 +1,7 @@ from __future__ import annotations import inspect +import weakref from typing import Any, Callable, Dict, List, Type, Union import torch @@ -40,8 +41,6 @@ def __init__( proxy_class: Type[Proxy] = Proxy, validate: bool = True, ) -> None: - - self.tracing = True self.proxy_class = proxy_class self.validate = validate @@ -56,7 +55,6 @@ def __init__( self.module_proxy = self.add( value=module, target="argument", args=["nnsight_root_module"] ) - def get_swap(self, value): if self.swap is not None: @@ -165,7 +163,7 @@ def add( node = Node( name=name, - graph=self, + graph=weakref.proxy(self), value=value, target=target, args=args, @@ -173,10 +171,6 @@ def add( meta={"line": proxy_frame.lineno, "file": proxy_frame.filename}, ) - if not self.tracing: - - return node.value - self.name_idx[target_name] += 1 self.nodes[name] = node diff --git a/src/nnsight/tracing/Node.py b/src/nnsight/tracing/Node.py index d43ac175..3a789e14 100644 --- a/src/nnsight/tracing/Node.py +++ b/src/nnsight/tracing/Node.py @@ -1,8 +1,7 @@ from __future__ import annotations import inspect -import weakref -from typing import TYPE_CHECKING, Any, Callable, Dict, List, Tuple, Union +from typing import TYPE_CHECKING, Any, Callable, Dict, List, Optional, Tuple, Union import torch @@ -28,7 +27,6 @@ class Node: listeners (List[Node]): Nodes that depend on this node. dependencies (List[Node]): Nodes that this node depends on. value (Any): Actual value to be populated during execution. - _proxy_device (torch.device): desc """ @staticmethod @@ -59,10 +57,12 @@ def slice_to_value(arg: slice): device = ( torch._GLOBAL_DEVICE_CONTEXT.device if torch._GLOBAL_DEVICE_CONTEXT is not None - else "cpu" + else None ) - values = util.apply(values, lambda x: x.to(device), torch.Tensor) + if device is not None: + + values = util.apply(values, lambda x: x.to(device), torch.Tensor) return values @@ -86,13 +86,15 @@ def __init__( meta = dict() self.name = name - self.graph: "Graph" = weakref.proxy(graph) + self.graph: "Graph" = graph self.proxy_value = value self.target = target self.args: List = util.apply(args, lambda x: x.node, Proxy) self.kwargs: Dict = util.apply(kwargs, lambda x: x.node, Proxy) self.meta = meta + self.proxy: Optional[Proxy] = None + self.value: Any = inspect._empty self.listeners: List[Node] = list() @@ -126,17 +128,62 @@ def __init__( self.remaining_listeners = 0 self.remaining_dependencies = 0 - self._proxy_device: torch.device = None - self.compile() - # (for when you want to apply things to proxies after model execution?) + # (for when you want to apply things to proxies after model execution) if self.fulfilled() and not isinstance(self.target, str): # So it doesn't get destroyed. self.remaining_listeners = 1 self.execute() + def is_graph_dereferenced(self) -> bool: + """Checks to see if the weakref to the Graph is deleted. If it is, we're no longer tracing. + + Returns: + bool: Is Graph dereferenced. + """ + + try: + + self.graph.add + + except: + return True + + return False + + def add( + self, + target: Union[Callable, str], + value: Any = inspect._empty, + args: List[Any] = None, + kwargs: Dict[str, Any] = None, + name: str = None, + ) -> Proxy: + """We use Node.add vs Graph.add in case the weakref to Graph is gone. + + Returns: + Proxy: Proxy + """ + + if self.is_graph_dereferenced(): + + return Proxy( + Node( + name=None, + graph=None, + value=None, + target=target, + args=args, + kwargs=kwargs, + ) + ).value + + return self.graph.add( + target=target, value=value, args=args, kwargs=kwargs, name=name + ) + def compile(self) -> None: """Resets this Nodes remaining_listeners and remaining_dependencies and sets its value to None.""" self.remaining_listeners = len(self.listeners) @@ -227,14 +274,30 @@ def execute(self) -> None: elif self.target == "grad": + tensor: torch.Tensor = args[0] + backward_idx: int = args[1] + def grad(value): - self.set_value(value) - value = self.graph.get_swap(value) + nonlocal backward_idx - return value + if backward_idx == 0: - tensor: torch.Tensor = args[0] + self.set_value(value) + + if not self.is_graph_dereferenced(): + + value = self.graph.get_swap(value) + + backward_idx = -1 + + return value + + else: + + backward_idx -= 1 + + return None tensor.register_hook(lambda value: grad(value)) diff --git a/src/nnsight/tracing/Proxy.py b/src/nnsight/tracing/Proxy.py index 3a981840..002a79e4 100644 --- a/src/nnsight/tracing/Proxy.py +++ b/src/nnsight/tracing/Proxy.py @@ -1,7 +1,7 @@ from __future__ import annotations -import inspect import operator +import weakref from typing import TYPE_CHECKING, Any, Callable, Union from typing_extensions import Self @@ -36,6 +36,7 @@ def __init__(self, node: "Node") -> None: self.__dict__["node"] = node self.node: "Node" + self.node.proxy = weakref.proxy(self) @property def value(self) -> Any: @@ -52,7 +53,7 @@ def value(self) -> Any: def __str__(self) -> str: - if not self.node.graph.tracing: + if self.node.is_graph_dereferenced(): return str(self.value) @@ -62,7 +63,7 @@ def __str__(self) -> str: def __repr__(self) -> str: - if not self.node.graph.tracing: + if self.node.is_graph_dereferenced(): return repr(self.value) @@ -76,38 +77,26 @@ def __call__(self, args, *kwargs) -> Self: Proxy: New call proxy. """ - value = inspect._empty - - # We don't want to call backward on fake tensors - if ( - self.node.target is util.fetch_attr - and isinstance(self.node.args[1], str) - and self.node.args[1] == "backward" - ): - - value = None - - return self.node.graph.add( - value=value, + return self.node.add( target=Proxy.proxy_call, args=[self.node] + list(args), kwargs=kwargs, ) def __getitem__(self, key: Union[Proxy, Any]) -> Self: - return self.node.graph.add( + return self.node.add( target=operator.getitem, args=[self.node, key], ) def __setitem__(self, key: Union[Proxy, Any], value: Union[Self, Any]) -> None: - self.node.graph.add( + self.node.add( target=operator.setitem, args=[self.node, key, value], ) def __getattr__(self, key: Union[Proxy, Any]) -> Self: - return self.node.graph.add( + return self.node.add( target=util.fetch_attr, args=[self.node, key], ) @@ -120,109 +109,109 @@ def __setattr__(self, key: Union[Proxy, Any], value: Union[Self, Any]) -> None: return - return self.node.graph.add( + return self.node.add( target=setattr, args=[self.node, key, value], ) def __len__(self) -> Self: - return self.node.graph.add( + return self.node.add( target=len, args=[self.node], ) def __abs__(self) -> Self: - return self.node.graph.add( + return self.node.add( target=operator.abs, args=[self.node], ) def __invert__(self) -> Self: - return self.node.graph.add( + return self.node.add( target=operator.invert, args=[self.node], ) def __add__(self, other: Union[Proxy, Any]) -> Self: - return self.node.graph.add( + return self.node.add( target=operator.add, args=[self.node, other], ) def __radd__(self, other: Union[Proxy, Any]) -> Self: - return self.node.graph.add( + return self.node.add( target=operator.add, args=[other, self.node], ) def __sub__(self, other: Union[Proxy, Any]) -> Self: - return self.node.graph.add( + return self.node.add( target=operator.sub, args=[self.node, other], ) def __rsub__(self, other: Union[Proxy, Any]) -> Self: - return self.node.graph.add( + return self.node.add( target=operator.sub, args=[other, self.node], ) def __pow__(self, other: Union[Proxy, Any]) -> Self: - return self.node.graph.add( + return self.node.add( target=operator.pow, args=[self.node, other], ) def __rpow__(self, other: Union[Proxy, Any]) -> Self: - return self.node.graph.add( + return self.node.add( target=operator.pow, args=[other, self.node], ) def __mul__(self, other: Union[Proxy, Any]) -> Self: - return self.node.graph.add( + return self.node.add( target=operator.mul, args=[self.node, other], ) def __rmul__(self, other: Union[Proxy, Any]) -> Self: - return self.node.graph.add( + return self.node.add( target=operator.mul, args=[other, self.node], ) def __mod__(self, other: Union[Proxy, Any]) -> Self: - return self.node.graph.add( + return self.node.add( target=operator.mod, args=[self.node, other], ) def __rmod__(self, other: Union[Proxy, Any]) -> Self: - return self.node.graph.add( + return self.node.add( target=operator.mod, args=[other, self.node], ) def __matmul__(self, other: Union[Proxy, Any]) -> Self: - return self.node.graph.add( + return self.node.add( target=operator.matmul, args=[self.node, other], ) def __rmatmul__(self, other: Union[Proxy, Any]) -> Self: - return self.node.graph.add( + return self.node.add( target=operator.matmul, args=[other, self.node], ) def __truediv__(self, other: Union[Proxy, Any]) -> Self: - return self.node.graph.add( + return self.node.add( target=operator.truediv, args=[self.node, other], ) def __rtruediv__(self, other: Union[Proxy, Any]) -> Self: - return self.node.graph.add( + return self.node.add( target=operator.truediv, args=[other, self.node], ) @@ -252,7 +241,7 @@ def get_proxy(arg): util.apply(args, get_proxy, Proxy) - return proxy.node.graph.add( + return proxy.node.add( target=orig_method, args=args, kwargs=kwargs, @@ -288,7 +277,7 @@ def get_node(proxy: Proxy): util.apply(list(args) + list(kwargs.values()), get_node, Proxy) if node is not None: - return node.graph.add(target=fn, args=args, kwargs=kwargs) + return node.add(target=fn, args=args, kwargs=kwargs) else: return fn(args, **kwargs)	Raise Exception on ERROR status of remote close #73		2024-02-23T20:01:02	[]	[]
ndif-team/nnsight	ndif-team__nnsight-36	cab401e14545e1a24c2671132404ee3a7ac60456	diff --git a/.DS_Store b/.DS_Store new file mode 100644 index 00000000..87cd3f95 Binary files /dev/null and b/.DS_Store differ diff --git a/README.md b/README.md index aae77ba0..d4e8aa23 100644 --- a/README.md +++ b/README.md @@ -198,7 +198,7 @@ model = LanguageModel('gpt2', device_map='cuda') with model.generate(max_new_tokens=1) as generator: with generator.invoke('The Eiffel Tower is in the city of') as invoker: - hidden_states_pre = model.transformer.h[-1].mlp.output.save() + hidden_states_pre = model.transformer.h[-1].mlp.output.clone().save() noise = (0.001*0.5)torch.randn(hidden_states_pre.shape) diff --git a/docs/.DS_Store b/docs/.DS_Store new file mode 100644 index 00000000..8bf5b4bc Binary files /dev/null and b/docs/.DS_Store differ diff --git a/docs/source/.DS_Store b/docs/source/.DS_Store new file mode 100644 index 00000000..bbf044e0 Binary files /dev/null and b/docs/source/.DS_Store differ diff --git a/docs/source/_static/.DS_Store b/docs/source/_static/.DS_Store new file mode 100644 index 00000000..f26591fc Binary files /dev/null and b/docs/source/_static/.DS_Store differ diff --git a/docs/source/_static/css/custom.css b/docs/source/_static/css/custom.css index 510f679a..d6257901 100644 --- a/docs/source/_static/css/custom.css +++ b/docs/source/_static/css/custom.css @@ -28,6 +28,10 @@ html[data-theme="light"] { } +.ndif { + color: "green"; +} + .bd-sidebar-primary { width: fit-content !important; padding-right: 40px !important @@ -48,6 +52,15 @@ img { pointer-events: none; } +.img-bottom-sizing { + max-height: 10vh; + width: auto +} + +.body-sizing{ + height: 10vh; +} + .title-bot { margin-bottom: -10px !important; line-height: normal !important; diff --git a/docs/source/_static/images/.DS_Store b/docs/source/_static/images/.DS_Store new file mode 100644 index 00000000..cc051dec Binary files /dev/null and b/docs/source/_static/images/.DS_Store differ diff --git a/docs/source/_static/images/attribute_lens.png b/docs/source/_static/images/attribute_lens.png new file mode 100644 index 00000000..ea63e99c Binary files /dev/null and b/docs/source/_static/images/attribute_lens.png differ diff --git a/docs/source/_static/images/ioi.png b/docs/source/_static/images/ioi.png new file mode 100644 index 00000000..6c323e55 Binary files /dev/null and b/docs/source/_static/images/ioi.png differ diff --git a/docs/source/_templates/ndif_status.html b/docs/source/_templates/ndif_status.html index 065ba225..b1e43700 100644 --- a/docs/source/_templates/ndif_status.html +++ b/docs/source/_templates/ndif_status.html @@ -1,27 +1,33 @@ -<style> - .ndif_status { - height: 18px; - width: 18px; - background-color: red; - border-radius: 50%; - display: inline-block; - margin-left: 5px; - vertical-align: middle; - } -</style> - -<span><span>NDIF Status: </span><span class="ndif_status"></span></span> - <script> + let statusIcon = document.querySelector('.ndif .fa-circle-check'); fetch("{{ndif_url}}") .then((response) => { if (response.status == 200) { - Array.from(document.getElementsByClassName("ndif_status")).forEach((elm) => elm.style.backgroundColor = "green"); + Array.from(document.getElementsByClassName("ndif")).forEach((ndifElement) => { + Array.from(ndifElement.getElementsByTagName('span')).forEach((spanElement) => { + spanElement.style.setProperty('color', '#23C552', 'important'); + }); + }); } else { - Array.from(document.getElementsByClassName("ndif_status")).forEach((elm) => elm.style.backgroundColor = "red"); + Array.from(document.getElementsByClassName("ndif")).forEach((ndifElement) => { + Array.from(ndifElement.getElementsByTagName('span')).forEach((spanElement) => { + spanElement.style.setProperty('color', '#F84F31', 'important'); + }); + }); + if (statusIcon) { + statusIcon.classList.replace('fa-circle-check', 'fa-circle-xmark'); + } + } + }) + .catch((response) => { + Array.from(document.getElementsByClassName("ndif")).forEach((ndifElement) => { + Array.from(ndifElement.getElementsByTagName('span')).forEach((spanElement) => { + spanElement.style.setProperty('color', '#F84F31', 'important'); + }); + }); + if (statusIcon) { + statusIcon.classList.replace('fa-circle-check', 'fa-circle-xmark'); } }) - .catch((response) => Array.from(document.getElementsByClassName("ndif_status")).forEach((elm) => elm.style.backgroundColor = "red")) - </script> \ No newline at end of file diff --git a/docs/source/about.rst b/docs/source/about.rst index 058f335a..bdbe3f06 100644 --- a/docs/source/about.rst +++ b/docs/source/about.rst @@ -38,7 +38,7 @@ Here is how it looks: model = LanguageModel('meta-llama/Llama-2-70b-hf') with model.forward(remote=True) as runner: with runner.invoke('The Eiffel Tower is in the city of ') as invoker: - hidden_state = model.layers[10].input.save() # save one hidden state + hidden_state = model.layers[10].input[0].save() # save one hidden state model.layers[11].mlp.output = 0 # change one MLP module output print('The model predicts', runner.output) print('The internal state was', hidden_state.value) diff --git a/docs/source/conf.py b/docs/source/conf.py index 5420d720..ba7bfc8f 100644 --- a/docs/source/conf.py +++ b/docs/source/conf.py @@ -31,9 +31,10 @@ html_favicon = "_static/images/icon.ico" html_show_sourcelink = False html_theme_options = { - "logo": {"text": "nnsight"}, + # "logo": {"text": "nnsight"}, "show_nav_level": 2, "navbar_end": ["navbar-icon-links", "ndif_status"], + # "navbar_end": ["navbar-icon-links"], "navbar_align": "left", "icon_links": [ { @@ -46,6 +47,12 @@ "url": "https://discord.gg/6uFJmCSwW7", "icon": "fa-brands fa-discord", }, + { + "name": "Status", + "url": "https://discord.gg/6uFJmCSwW7", + "icon": "fa-solid fa-circle-check", + "attributes": {"class": "ndif"}, + }, ], } diff --git a/docs/source/index.rst b/docs/source/index.rst index 6c6e2ce2..885a1d83 100644 --- a/docs/source/index.rst +++ b/docs/source/index.rst @@ -14,6 +14,10 @@ nnsight tutorials About <about> +.. toctree:: + :hidden: + status + .. grid:: 1 1 2 2 :class-container: hero :reverse: diff --git a/docs/source/notebooks/features/setting.ipynb b/docs/source/notebooks/features/setting.ipynb index d489854d..d2f1b571 100644 --- a/docs/source/notebooks/features/setting.ipynb +++ b/docs/source/notebooks/features/setting.ipynb @@ -30,13 +30,13 @@ "with model.generate(max_new_tokens=1) as generator:\n", " with generator.invoke('The Eiffel Tower is in the city of') as invoker:\n", "\n", - " hidden_states_pre = model.transformer.h[-1].output[0][:].save()\n", + " hidden_states_pre = model.transformer.h[-1].output[0].clone().save()\n", "\n", " noise = (0.001*0.5)torch.randn(hidden_states_pre.shape)\n", "\n", " model.transformer.h[-1].output[0][:] = hidden_states_pre + noise\n", "\n", - " hidden_states_post = model.transformer.h[-1].output[0][:].save()" + " hidden_states_post = model.transformer.h[-1].output[0].save()" ] }, { diff --git a/docs/source/notebooks/features/token_indexing.ipynb b/docs/source/notebooks/features/token_indexing.ipynb index 79080e8c..948979b8 100644 --- a/docs/source/notebooks/features/token_indexing.ipynb +++ b/docs/source/notebooks/features/token_indexing.ipynb @@ -102,10 +102,10 @@ "\n", "with model.forward() as runner:\n", " with runner.invoke('The') as invoker:\n", - " incorrect_a = model.transformer.input[0][:,0].save()\n", + " incorrect_a = model.transformer.input[0][0][:,0].save()\n", " \n", " with runner.invoke('The Eiffel Tower is in the city of''The Eiffel Tower is in the city of') as invoker:\n", - " incorrect_b = model.transformer.input[0][:,0].save()\n", + " incorrect_b = model.transformer.input[0][0][:,0].save()\n", "\n", "print(f\"Shorter input: {incorrect_a.value}\")\n", "print(f\"Longer input: {incorrect_b.value}\")" @@ -157,10 +157,10 @@ "source": [ "with model.forward() as runner:\n", " with runner.invoke('The') as invoker:\n", - " correct_a = model.transformer.input[0].t[0].save()\n", + " correct_a = model.transformer.input[0][0].t[0].save()\n", " \n", " with runner.invoke('The Eiffel Tower is in the city of') as invoker:\n", - " correct_b = model.transformer.input[0].t[0].save()\n", + " correct_b = model.transformer.input[0][0].t[0].save()\n", "\n", "print(f\"Shorter input: {correct_a.value}\")\n", "print(f\"Longer input: {correct_b.value}\")" diff --git a/docs/source/notebooks/tutorials/attribution_patching.ipynb b/docs/source/notebooks/tutorials/attribution_patching.ipynb index 46998c72..cb2a98c1 100644 --- a/docs/source/notebooks/tutorials/attribution_patching.ipynb +++ b/docs/source/notebooks/tutorials/attribution_patching.ipynb @@ -154,7 +154,7 @@ " clean_logit_diff = get_logit_diff(clean_logits, answer_token_indices).item()\n", "\n", " clean_cache = [\n", - " model.transformer.h[i].attn.attn_dropout.input.save()\n", + " model.transformer.h[i].attn.attn_dropout.input[0].save()\n", " for i in range(len(model.transformer.h))\n", " ]\n", "\n", @@ -171,7 +171,7 @@ " ).item()\n", "\n", " corrupted_cache = [\n", - " model.transformer.h[i].attn.attn_dropout.input.save()\n", + " model.transformer.h[i].attn.attn_dropout.input[0].save()\n", " for i in range(len(model.transformer.h))\n", " ]\n", "\n", diff --git a/docs/source/notebooks/tutorials/dictionary_learning.ipynb b/docs/source/notebooks/tutorials/dictionary_learning.ipynb deleted file mode 100644 index fb9e75a5..00000000 --- a/docs/source/notebooks/tutorials/dictionary_learning.ipynb +++ /dev/null @@ -1,25 +0,0 @@ -{ - "cells": [ - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "# Dictionary Learning" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "Coming Soon!" - ] - } - ], - "metadata": { - "language_info": { - "name": "python" - } - }, - "nbformat": 4, - "nbformat_minor": 2 -} diff --git a/docs/source/notebooks/walkthrough.ipynb b/docs/source/notebooks/walkthrough.ipynb index bae480d8..94e85469 100644 --- a/docs/source/notebooks/walkthrough.ipynb +++ b/docs/source/notebooks/walkthrough.ipynb @@ -75,7 +75,7 @@ "\n", "Use `dispatch=True` to load the model onto the specified device locally on initialization, not on first use.\n", "\n", - "`LanguageModel` is one wrapper for `nnsight` functionality. Another one `DiffusionModel` for text-to-image disffusion models. We encourage you to create you own by inheriting from `AbstractModel`\n", + "`LanguageModel` is one wrapper for `nnsight` functionality. Another one `DiffusionModel` for text-to-image disffusion models. We encourage you to create your own by inheriting from `AbstractModel`\n", "\n", "</details>\n", "\n", @@ -1303,7 +1303,7 @@ "px.line(\n", " [layer_probs.value[mary_token].item() for layer_probs in all_probs],\n", " title=\"Probability of Mary after each layer, according to logit lens\",\n", - " labels={\"value\":\"Layer\", \"index\":\"Probability\"}\n", + " labels={\"value\":\"Probability\", \"index\":\"Layer\"}\n", ")" ] }, diff --git a/docs/source/status.rst b/docs/source/status.rst new file mode 100644 index 00000000..8165df30 --- /dev/null +++ b/docs/source/status.rst @@ -0,0 +1,77 @@ +:html_theme.sidebar_secondary.remove: +:sd_hide_title: + +status +====== + +.. toctree:: + :maxdepth: 1 + :hidden: + + + +.. card:: + + All Systems Are Operational + + + +.. div:: sd-fs-1 sd-font-weight-bold sd-text-center sd-text-primary sd-mb-5 + + Key Features + +.. grid:: 1 1 2 2 + :class-container: features + + .. grid-item:: + + .. div:: features-container + + .. image:: _static/images/one.png + :width: 250 + + .. div:: + + Integration + + Pass in a ID from any HuggingFace Transformer repo and access its weights with nnsight. + + .. grid-item:: + + .. div:: features-container + + .. image:: _static/images/two.png + :width: 250 + + .. div:: + + Interpretability + + Access the internal gradients and activations at any point or module in a model. + + .. grid-item:: + + .. div:: features-container + + .. image:: _static/images/two.png + :width: 250 + + .. div:: + + Intuitive + + Set up a context block and manipulate model internals with only a couple lines of code. + + .. grid-item:: + + .. div:: features-container + + .. image:: _static/images/one.png + :width: 250 + + .. div:: + + Interoperable + + Enable grad and train interventions like LORA or probes on any point in a model. + diff --git a/docs/source/tutorials.rst b/docs/source/tutorials.rst index bb5b8e17..bc01aca5 100644 --- a/docs/source/tutorials.rst +++ b/docs/source/tutorials.rst @@ -1,17 +1,54 @@ Tutorials ============ +.. grid:: 2 + :gutter: 2 + + .. grid-item-card:: Walkthrough + :link: notebooks/walkthrough.ipynb + + Main features of the nnsight library. + + .. grid-item-card:: IOI Patching + :link: notebooks/tutorials/ioi_patching.ipynb + + :bdg-primary:`arXiv:2211.00593` + + .. grid-item-card:: Attribution Patching + :link: notebooks/tutorials/attribution_patching.ipynb + + :bdg-primary:`arXiv:2308.09124` + + .. grid-item-card:: Logit Lens + :link: notebooks/tutorials/logit_lens.ipynb + + :bdg-primary:`nostalgebraist` + + .. grid-item-card:: Future Lens + :link: notebooks/tutorials/future_lens.ipynb + + :bdg-primary:`arXiv:2311.04897` + + .. grid-item-card:: Function Vectors + :link: notebooks/tutorials/function_vectors.ipynb + + :bdg-primary:`arXiv:2310.15213` + + .. grid-item-card:: Dictionary Learning + :link: notebooks/tutorials/sae.ipynb + + :bdg-primary:`arXiv:2309.08600` + + .. toctree:: :maxdepth: 1 + :hidden: notebooks/tutorials/ioi_patching.ipynb notebooks/tutorials/attribution_patching.ipynb notebooks/tutorials/logit_lens.ipynb notebooks/tutorials/future_lens.ipynb notebooks/tutorials/function_vectors.ipynb - notebooks/tutorials/dictionary_learning.ipynb + notebooks/tutorials/sae.ipynb -.. toctree:: - :hidden: - notebooks/tutorials/sae.ipynb diff --git a/src/nnsight/contexts/Runner.py b/src/nnsight/contexts/Runner.py index 15890f2f..cd53714e 100644 --- a/src/nnsight/contexts/Runner.py +++ b/src/nnsight/contexts/Runner.py @@ -1,5 +1,8 @@ from __future__ import annotations +import pickle + +import requests import socketio from .. import CONFIG, pydantics @@ -101,7 +104,7 @@ def blocking_request(self, request: pydantics.RequestModel): f"wss://{CONFIG.API.HOST}", socketio_path="/ws/socket.io", transports=["websocket"], - wait_timeout=10 + wait_timeout=10, ) # Called when receiving a response from the server. @@ -116,10 +119,15 @@ def blocking_response(data): # If the status of the response is completed, update the local nodes that the user specified to save. # Then disconnect and continue. if response.status == pydantics.ResponseModel.JobStatus.COMPLETED: - for name, value in response.saves.items(): + with requests.get( + url=f"https://{CONFIG.API.HOST}/retrieve/{response.id}", stream=True + ) as stream: + result_response = pydantics.ResponseModel(**pickle.load(stream.raw)) + + for name, value in result_response.result.saves.items(): self.graph.nodes[name].value = value - self.output = response.output + self.output = result_response.result.output sio.disconnect() # Or if there was some error. diff --git a/src/nnsight/intervention.py b/src/nnsight/intervention.py index e533fc51..8dcadb1c 100644 --- a/src/nnsight/intervention.py +++ b/src/nnsight/intervention.py @@ -151,6 +151,7 @@ def _concat(values): key: _concat([value[key] for value in values]) for key in values[0].keys() } + return values[0] # As interventions are scoped only to their relevant batch, if we want to swap in values for this batch # we need to concatenate the batches before and after the relevant batch with the new values. @@ -194,7 +195,7 @@ def intervene(activations: Any, module_path: str, graph: Graph, key: str): # Key to module activation argument nodes has format: <module path>.<output/input>.<generation index> module_path = f"{module_path}.{key}.{graph.generation_idx}" - + if module_path in graph.argument_node_names: argument_node_names = graph.argument_node_names[module_path] @@ -206,6 +207,7 @@ def intervene(activations: Any, module_path: str, graph: Graph, key: str): _, batch_size, batch_start = node.args # We set its result to the activations, indexed by only the relevant batch idxs. + value = util.apply( activations, lambda x: x.narrow(0, batch_start, batch_size), @@ -279,10 +281,10 @@ def __enter__(self) -> HookModel: if hook_type == "input": - def input_hook(module, input, module_path=module_path): - return self.input_hook(input, module_path) + def input_hook(module, input, kwargs, module_path=module_path): + return self.input_hook((input, kwargs), module_path) - self.handles.append(module.register_forward_pre_hook(input_hook)) + self.handles.append(module.register_forward_pre_hook(input_hook, with_kwargs=True)) elif hook_type == "output": diff --git a/src/nnsight/models/AbstractModel.py b/src/nnsight/models/AbstractModel.py index 1874e380..320eca99 100644 --- a/src/nnsight/models/AbstractModel.py +++ b/src/nnsight/models/AbstractModel.py @@ -92,7 +92,7 @@ def meta_deepcopy(self, memo): with Patcher() as patcher: patcher.add( Patch( - torch.nn.parameter.Parameter.__deepcopy__, meta_deepcopy + torch.nn.parameter.Parameter, meta_deepcopy, "__deepcopy__" ) ) diff --git a/src/nnsight/pydantics/Response.py b/src/nnsight/pydantics/Response.py index ea4a6ec3..effbb44c 100644 --- a/src/nnsight/pydantics/Response.py +++ b/src/nnsight/pydantics/Response.py @@ -9,6 +9,16 @@ from pydantic import BaseModel, field_validator +class ResultModel(BaseModel): + id: str + output: Any = None + saves: Any = None + + @field_validator("output", "saves") + def unpickle(cls, value: bytes): + return pickle.loads(value) + + class ResponseModel(BaseModel): class JobStatus(Enum): RECEIVED = "RECEIVED" @@ -22,11 +32,11 @@ class JobStatus(Enum): description: str received: datetime = None - saves: Union[bytes, Any] = None - output: Union[bytes, Any] = None session_id: str = None blocking: bool = False + result: Union[bytes, ResultModel] = None + def __str__(self) -> str: return f"{self.id} - {self.status.name}: {self.description}" @@ -37,8 +47,3 @@ def log(self, logger: logging.Logger) -> ResponseModel: logger.info(str(self)) return self - - @field_validator("output", "saves") - @classmethod - def unpickle(cls, value): - return pickle.loads(value)	Tutorial Page Reformat and Status Tweaks Changed status logo and reformatted tutorial page		2024-01-01T21:04:32	[]	[]
Arize-ai/openinference	Arize-ai__openinference-1123	534908982d871ac7e481738109cb06d23f9294f9	diff --git a/js/.changeset/mean-walls-talk.md b/js/.changeset/mean-walls-talk.md new file mode 100644 index 000000000..579de1113 --- /dev/null +++ b/js/.changeset/mean-walls-talk.md @@ -0,0 +1,5 @@ +--- +"@arizeai/openinference-instrumentation-langchain": patch +--- + +fix: add support for capturing ChatBedrock token counts diff --git a/js/packages/openinference-instrumentation-langchain/src/utils.ts b/js/packages/openinference-instrumentation-langchain/src/utils.ts index d3f7abdb9..01423be37 100644 --- a/js/packages/openinference-instrumentation-langchain/src/utils.ts +++ b/js/packages/openinference-instrumentation-langchain/src/utils.ts @@ -478,6 +478,8 @@ function getTokenCount(maybeCount: unknown) { * Formats the token counts of a langchain run into OpenInference attributes. * @param outputs - The outputs of a langchain run * @returns The OpenInference attributes for the token counts + * + * @see https://github.com/langchain-ai/langchainjs/blob/main/langchain-core/src/language_models/chat_models.ts#L403 for how token counts get added to outputs / function formatTokenCounts( outputs: Run["outputs"], @@ -485,6 +487,34 @@ function formatTokenCounts( if (!isObject(outputs)) { return null; } + const firstGeneration = getFirstOutputGeneration(outputs); + + /* + * Some community models have non standard output structures and show token counts in different places notable ChatBedrock + * @see https://github.com/langchain-ai/langchainjs/blob/a173e300ef9ada416220876a2739e024b3a7f268/libs/langchain-community/src/chat_models/bedrock/web.ts + / + // Generations is an array of arrays containing messages + const maybeGenerationComponent = + firstGeneration != null ? firstGeneration[0] : null; + const maybeMessage = isObject(maybeGenerationComponent) + ? maybeGenerationComponent.message + : null; + const usageMetadata = isObject(maybeMessage) + ? maybeMessage.usage_metadata + : null; + if (isObject(usageMetadata)) { + return { + [SemanticConventions.LLM_TOKEN_COUNT_COMPLETION]: getTokenCount( + usageMetadata.output_tokens, + ), + [SemanticConventions.LLM_TOKEN_COUNT_PROMPT]: getTokenCount( + usageMetadata.input_tokens, + ), + [SemanticConventions.LLM_TOKEN_COUNT_TOTAL]: getTokenCount( + usageMetadata.total_tokens, + ), + }; + } const llmOutput = outputs.llmOutput; if (!isObject(llmOutput)) { return null; @@ -519,6 +549,33 @@ function formatTokenCounts( ), }; } + + /* + * In some cases community models have a different output structure do to the way they extend the base model + * Notably ChatBedrock may have tokens stored in this format instead of normalized + * @see https://github.com/langchain-ai/langchainjs/blob/a173e300ef9ada416220876a2739e024b3a7f268/libs/langchain-community/src/chat_models/bedrock/web.ts for ChatBedrock + * and + * @see https://github.com/langchain-ai/langchainjs/blob/main/langchain-core/src/language_models/chat_models.ts#L403 for nomalization + */ + if (isObject(llmOutput.usage)) { + const maybePromptTokens = getTokenCount(llmOutput.usage.input_tokens); + const maybeCompletionTokens = getTokenCount(llmOutput.usage.output_tokens); + let maybeTotalTokens = getTokenCount(llmOutput.usage.total_tokens); + if (maybeTotalTokens == null) { + maybeTotalTokens = + isNumber(maybePromptTokens) && isNumber(maybeCompletionTokens) + ? maybePromptTokens + maybeCompletionTokens + : undefined; + } + return { + [SemanticConventions.LLM_TOKEN_COUNT_COMPLETION]: getTokenCount( + maybeCompletionTokens, + ), + [SemanticConventions.LLM_TOKEN_COUNT_PROMPT]: + getTokenCount(maybePromptTokens), + [SemanticConventions.LLM_TOKEN_COUNT_TOTAL]: maybeTotalTokens, + }; + } return null; } @@ -589,7 +646,7 @@ function formatMetadata(run: Run) { return null; } return { - metadata: safelyJSONStringify(run.extra.metadata), + [SemanticConventions.METADATA]: safelyJSONStringify(run.extra.metadata), }; }	[bug][langchain][js] langchain js instrumentor missing token counts for claude 3.5 sonnet LangChain JS instrumentation reported to be missing token counts for Anthropic https://arize-ai.slack.com/archives/C04R3GXC8HK/p1731387916527799		2024-11-15T00:34:13	[]	[]
Arize-ai/openinference	Arize-ai__openinference-1113	a0e6f30c84011c5c743625bb69b66ba055ac17bd	diff --git a/python/instrumentation/openinference-instrumentation-dspy/examples/requirements.txt b/python/instrumentation/openinference-instrumentation-dspy/examples/requirements.txt index 3c0ec027a..76e21f6e3 100644 --- a/python/instrumentation/openinference-instrumentation-dspy/examples/requirements.txt +++ b/python/instrumentation/openinference-instrumentation-dspy/examples/requirements.txt @@ -1,4 +1,4 @@ -dspy-ai >= 2.1.0 +dspy >= 2.5.0 opentelemetry-sdk opentelemetry-exporter-otlp openinference-instrumentation-dspy \ No newline at end of file diff --git a/python/instrumentation/openinference-instrumentation-dspy/pyproject.toml b/python/instrumentation/openinference-instrumentation-dspy/pyproject.toml index 5c0c64f4e..b407373df 100644 --- a/python/instrumentation/openinference-instrumentation-dspy/pyproject.toml +++ b/python/instrumentation/openinference-instrumentation-dspy/pyproject.toml @@ -35,10 +35,10 @@ dependencies = [ [project.optional-dependencies] instruments = [ - "dspy-ai >= 2.5.0", + "dspy >= 2.5.0", ] test = [ - "dspy-ai==2.5.0", + "dspy>=2.5.27", "google-generativeai", "opentelemetry-sdk", "pytest-recording", diff --git a/python/instrumentation/openinference-instrumentation-dspy/src/openinference/instrumentation/dspy/package.py b/python/instrumentation/openinference-instrumentation-dspy/src/openinference/instrumentation/dspy/package.py index 5825ce077..02cd1b833 100644 --- a/python/instrumentation/openinference-instrumentation-dspy/src/openinference/instrumentation/dspy/package.py +++ b/python/instrumentation/openinference-instrumentation-dspy/src/openinference/instrumentation/dspy/package.py @@ -1,2 +1,2 @@ -_instruments = ("dspy-ai >= 2.1.0",) +_instruments = ("dspy >= 2.5.0",) _supports_metrics = False	[bug] dspy instrumentor failing CI unit tests for the latest dspy version		2024-11-12T04:40:05	[]	[]
Arize-ai/openinference	Arize-ai__openinference-1087	473701d3dd2f8bcda7afe12bb65a1145d0a4d893	diff --git a/python/instrumentation/openinference-instrumentation-anthropic/src/openinference/instrumentation/anthropic/_wrappers.py b/python/instrumentation/openinference-instrumentation-anthropic/src/openinference/instrumentation/anthropic/_wrappers.py index c5d0e8204..050200479 100644 --- a/python/instrumentation/openinference-instrumentation-anthropic/src/openinference/instrumentation/anthropic/_wrappers.py +++ b/python/instrumentation/openinference-instrumentation-anthropic/src/openinference/instrumentation/anthropic/_wrappers.py @@ -22,6 +22,7 @@ OpenInferenceMimeTypeValues, OpenInferenceSpanKindValues, SpanAttributes, + ToolAttributes, ToolCallAttributes, ) @@ -98,6 +99,7 @@ def __call__( _get_llm_prompts(llm_prompt), _get_inputs(arguments), _get_llm_invocation_parameters(llm_invocation_parameters), + _get_llm_tools(llm_invocation_parameters), ) ), ) as span: @@ -150,6 +152,7 @@ async def __call__( _get_llm_prompts(llm_prompt), _get_inputs(arguments), _get_llm_invocation_parameters(invocation_parameters), + _get_llm_tools(invocation_parameters), ) ), ) as span: @@ -201,6 +204,7 @@ def __call__( _get_llm_span_kind(), _get_llm_input_messages(llm_input_messages), _get_llm_invocation_parameters(invocation_parameters), + _get_llm_tools(invocation_parameters), _get_inputs(arguments), ) ), @@ -260,6 +264,7 @@ async def __call__( _get_llm_span_kind(), _get_llm_input_messages(llm_input_messages), _get_llm_invocation_parameters(invocation_parameters), + _get_llm_tools(invocation_parameters), _get_inputs(arguments), ) ), @@ -298,6 +303,12 @@ def _get_outputs(response: "BaseModel") -> Iterator[Tuple[str, Any]]: yield OUTPUT_MIME_TYPE, JSON +def _get_llm_tools(invocation_parameters: Mapping[str, Any]) -> Iterator[Tuple[str, Any]]: + if isinstance(tools := invocation_parameters.get("tools"), list): + for tool_index, tool_schema in enumerate(tools): + yield f"{LLM_TOOLS}.{tool_index}.{TOOL_JSON_SCHEMA}", safe_json_dumps(tool_schema) + + def _get_llm_span_kind() -> Iterator[Tuple[str, Any]]: yield OPENINFERENCE_SPAN_KIND, LLM @@ -464,6 +475,7 @@ def _validate_invocation_parameter(parameter: Any) -> bool: LLM_TOKEN_COUNT_COMPLETION = SpanAttributes.LLM_TOKEN_COUNT_COMPLETION LLM_TOKEN_COUNT_PROMPT = SpanAttributes.LLM_TOKEN_COUNT_PROMPT LLM_TOKEN_COUNT_TOTAL = SpanAttributes.LLM_TOKEN_COUNT_TOTAL +LLM_TOOLS = SpanAttributes.LLM_TOOLS MESSAGE_CONTENT = MessageAttributes.MESSAGE_CONTENT MESSAGE_CONTENTS = MessageAttributes.MESSAGE_CONTENTS MESSAGE_FUNCTION_CALL_ARGUMENTS_JSON = MessageAttributes.MESSAGE_FUNCTION_CALL_ARGUMENTS_JSON @@ -479,6 +491,7 @@ def _validate_invocation_parameter(parameter: Any) -> bool: TAG_TAGS = SpanAttributes.TAG_TAGS TOOL_CALL_FUNCTION_ARGUMENTS_JSON = ToolCallAttributes.TOOL_CALL_FUNCTION_ARGUMENTS_JSON TOOL_CALL_FUNCTION_NAME = ToolCallAttributes.TOOL_CALL_FUNCTION_NAME +TOOL_JSON_SCHEMA = ToolAttributes.TOOL_JSON_SCHEMA USER_ID = SpanAttributes.USER_ID LLM_PROVIDER = SpanAttributes.LLM_PROVIDER LLM_SYSTEM = SpanAttributes.LLM_SYSTEM	[feature request] capture tool_schema in anthropic requests see example: ```python from typing import Any, Dict from your_client_library import Client # Replace with actual import tool_schema = { "name": "record_travel_request_attributes", "description": "Records the attributes of a travel request", "input_schema": { "type": "object", "properties": { "location": { "type": "string", "description": 'The desired destination location. Use city, state, and country format when possible. If no destination is provided, return "not_stated".', }, "budget_level": { "type": "string", "enum": ["low", "medium", "high", "not_stated"], "description": 'The desired budget level. If no budget level is provided, return "not_stated".', }, "purpose": { "type": "string", "enum": ["business", "pleasure", "other", "not_stated"], "description": 'The purpose of the trip. If no purpose is provided, return "not_stated".', }, }, "required": ["location", "budget_level", "purpose"], }, } system_message = ( "You are an assistant that parses and records the attributes of a user's travel request." ) def extract_raw_travel_request_attributes_string( travel_request: str, client: Client, model: str = "claude-3-5-sonnet-20240620", ) -> str: response = client.messages.create( model=model, max_tokens=1024, messages=[ {"role": "user", "content": travel_request}, ], system=system_message, tools=[tool_schema], tool_choice={"type": "tool", "name": "record_travel_request_attributes"}, ) # Process the response to extract the relevant information # This is a placeholder - you'll need to implement the actual extraction logic extracted_attributes = process_response(response) return extracted_attributes def process_response(response: Any) -> str: # Implement the logic to extract the attributes from the response # This is a placeholder function pass ```	Related to playground	2024-10-30T06:15:55	[]	[]
Arize-ai/openinference	Arize-ai__openinference-1048	ff43e9ddc0a5f683f80d09139247ad194d6c29af	diff --git a/js/.changeset/tame-poems-argue.md b/js/.changeset/tame-poems-argue.md new file mode 100644 index 000000000..47b95c3fb --- /dev/null +++ b/js/.changeset/tame-poems-argue.md @@ -0,0 +1,5 @@ +--- +"@arizeai/openinference-core": minor +--- + +add OITracer and trace config to allow for masking of sensitive information on spans diff --git a/js/packages/openinference-core/jest.config.js b/js/packages/openinference-core/jest.config.js index 2fd4f78cf..de51c20f7 100644 --- a/js/packages/openinference-core/jest.config.js +++ b/js/packages/openinference-core/jest.config.js @@ -3,4 +3,7 @@ module.exports = { preset: "ts-jest", testEnvironment: "node", prettierPath: null, + moduleNameMapper: { + "^@core/(.)$": "<rootDir>/src/$1", + }, }; diff --git a/js/packages/openinference-core/package.json b/js/packages/openinference-core/package.json index b9caf8c8d..b4a1eac98 100644 --- a/js/packages/openinference-core/package.json +++ b/js/packages/openinference-core/package.json @@ -34,10 +34,8 @@ }, "dependencies": { "@arizeai/openinference-semantic-conventions": "workspace:", - "@opentelemetry/core": "^1.25.1" - }, - "peerDependencies": { - "@opentelemetry/api": ">=1.0.0 <1.9.0" + "@opentelemetry/core": "^1.25.1", + "@opentelemetry/api": "^1.9.0" }, "devDependencies": { "@opentelemetry/context-async-hooks": "^1.25.1", diff --git a/js/packages/openinference-core/src/trace/index.ts b/js/packages/openinference-core/src/trace/index.ts index 671871650..3c8e70b9e 100644 --- a/js/packages/openinference-core/src/trace/index.ts +++ b/js/packages/openinference-core/src/trace/index.ts @@ -1,2 +1,3 @@ export * from "./contextAttributes"; export * from "./types"; +export * from "./trace-config"; diff --git a/js/packages/openinference-core/src/trace/trace-config/OISpan.ts b/js/packages/openinference-core/src/trace/trace-config/OISpan.ts new file mode 100644 index 000000000..b1acf6d8a --- /dev/null +++ b/js/packages/openinference-core/src/trace/trace-config/OISpan.ts @@ -0,0 +1,90 @@ +import { + Span, + SpanContext, + SpanStatus, + TimeInput, + Attributes, + Exception, + AttributeValue, + Link, +} from "@opentelemetry/api"; +import { TraceConfig } from "./types"; +import { mask } from "./maskingRules"; + +/** + * A wrapper around the OpenTelemetry {@link Span} interface that masks sensitive information based on the passed in {@link TraceConfig}. + / +export class OISpan implements Span { + private readonly span: Span; + private readonly config: TraceConfig; + + constructor({ span, config }: { span: Span; config: TraceConfig }) { + this.span = span; + this.config = config; + } + + setAttribute(key: string, value: AttributeValue): this { + const maskedValue = mask({ config: this.config, key, value }); + if (maskedValue != null) { + this.span.setAttribute(key, maskedValue); + } + return this; + } + + setAttributes(attributes: Attributes): this { + const maskedAttributes = Object.entries(attributes).reduce( + (maskedAttributes, [key, value]) => { + maskedAttributes[key] = mask({ config: this.config, key, value }); + return maskedAttributes; + }, + {} as Attributes, + ); + this.span.setAttributes(maskedAttributes); + return this; + } + + addEvent( + name: string, + attributesOrStartTime?: Attributes \| TimeInput, + startTime?: TimeInput, + ): this { + this.span.addEvent(name, attributesOrStartTime, startTime); + return this; + } + + setStatus(status: SpanStatus): this { + this.span.setStatus(status); + return this; + } + + updateName(name: string): this { + this.span.updateName(name); + return this; + } + + end(endTime?: TimeInput): void { + this.span.end(endTime); + } + + isRecording(): boolean { + return this.span.isRecording(); + } + + recordException(exception: Exception, time?: TimeInput): void { + this.span.recordException(exception, time); + } + + spanContext(): SpanContext { + return this.span.spanContext(); + } + + addLink(link: Link): this { + this.span.addLink(link); + return this; + } + + addLinks(links: Link[]): this { + this.span.addLinks(links); + return this; + } +} diff --git a/js/packages/openinference-core/src/trace/trace-config/OITracer.ts b/js/packages/openinference-core/src/trace/trace-config/OITracer.ts new file mode 100644 index 000000000..d026df606 --- /dev/null +++ b/js/packages/openinference-core/src/trace/trace-config/OITracer.ts @@ -0,0 +1,130 @@ +import { + context as apiContext, + Context, + Span, + SpanOptions, + Tracer, +} from "@opentelemetry/api"; +import { + OpenInferenceActiveSpanCallback, + TraceConfig, + TraceConfigOptions, +} from "./types"; +import { OISpan } from "./OISpan"; +import { generateTraceConfig } from "./traceConfig"; +import { getAttributesFromContext } from "../contextAttributes"; + +/* + * Formats the params for the startActiveSpan method + * The method has multiple overloads, so we need to format the arguments + * Taken from @see https://github.com/open-telemetry/opentelemetry-js/blob/main/packages/opentelemetry-sdk-trace-base/src/Tracer.ts#L220C3-L235C6 + * + / +function formatStartActiveSpanParams<F extends OpenInferenceActiveSpanCallback>( + arg2?: SpanOptions \| F, + arg3?: Context \| F, + arg4?: F, +) { + let opts: SpanOptions \| undefined; + let ctx: Context \| undefined; + let fn: F; + + if (typeof arg2 === "function") { + fn = arg2; + } else if (typeof arg3 === "function") { + opts = arg2; + fn = arg3; + } else { + opts = arg2; + ctx = arg3; + fn = arg4 as F; + } + + opts = opts ?? {}; + ctx = ctx ?? apiContext.active(); + + return { opts, ctx, fn }; +} + +/* + * A wrapper around the OpenTelemetry {@link Tracer} interface that masks sensitive information based on the passed in {@link TraceConfig}. + / +export class OITracer implements Tracer { + private readonly tracer: Tracer; + private readonly config: TraceConfig; + /* + * + * @param tracer The OpenTelemetry {@link Tracer} to wrap + * @param traceConfig The {@link TraceConfigOptions} to set to control the behavior of the tracer + / + constructor({ + tracer, + traceConfig, + }: { + tracer: Tracer; + traceConfig?: TraceConfigOptions; + }) { + this.tracer = tracer; + this.config = generateTraceConfig(traceConfig); + } + startActiveSpan<F extends (span: OISpan) => unknown>( + name: string, + fn: F, + ): ReturnType<F>; + startActiveSpan<F extends (span: OISpan) => unknown>( + name: string, + options: SpanOptions, + fn: F, + ): ReturnType<F>; + startActiveSpan<F extends (span: OISpan) => unknown>( + name: string, + options: SpanOptions, + context: Context, + fn: F, + ): ReturnType<F>; + startActiveSpan<F extends (span: OISpan) => ReturnType<F>>( + name: string, + arg2?: F \| SpanOptions, + arg3?: F \| Context, + arg4?: F, + ): ReturnType<F> \| undefined { + const formattedArgs = formatStartActiveSpanParams(arg2, arg3, arg4); + if (formattedArgs == null) { + return; + } + const { opts, ctx, fn } = formattedArgs; + const { attributes } = opts ?? {}; + const contextAttributes = getAttributesFromContext(ctx); + const mergedAttributes = { ...contextAttributes, ...attributes }; + return this.tracer.startActiveSpan( + name, + { ...opts, attributes: undefined }, + ctx, + (span: Span) => { + const openInferenceSpan = new OISpan({ + span, + config: this.config, + }); + openInferenceSpan.setAttributes(mergedAttributes); + return fn(openInferenceSpan); + }, + ); + } + + startSpan(name: string, options?: SpanOptions, context?: Context): OISpan { + const attributes = options?.attributes; + const ctx = context ?? apiContext.active(); + const contextAttributes = getAttributesFromContext(ctx); + const mergedAttributes = { ...contextAttributes, ...attributes }; + const span = new OISpan({ + span: this.tracer.startSpan( + name, + { ...options, attributes: undefined }, + ctx, + ), + config: this.config, + }); + span.setAttributes(mergedAttributes); + return span; + } +} diff --git a/js/packages/openinference-core/src/trace/trace-config/constants.ts b/js/packages/openinference-core/src/trace/trace-config/constants.ts new file mode 100644 index 000000000..a9b878e7c --- /dev/null +++ b/js/packages/openinference-core/src/trace/trace-config/constants.ts @@ -0,0 +1,107 @@ +import { TraceConfigKey, TraceConfig, TraceConfigFlag } from "./types"; + +/* Hides input value & messages / +export const OPENINFERENCE_HIDE_INPUTS = "OPENINFERENCE_HIDE_INPUTS"; +/* Hides output value & messages / +export const OPENINFERENCE_HIDE_OUTPUTS = "OPENINFERENCE_HIDE_OUTPUTS"; +/* Hides all input messages / +export const OPENINFERENCE_HIDE_INPUT_MESSAGES = + "OPENINFERENCE_HIDE_INPUT_MESSAGES"; +/* Hides all output messages / +export const OPENINFERENCE_HIDE_OUTPUT_MESSAGES = + "OPENINFERENCE_HIDE_OUTPUT_MESSAGES"; +/* Hides images from input messages / +export const OPENINFERENCE_HIDE_INPUT_IMAGES = + "OPENINFERENCE_HIDE_INPUT_IMAGES"; +/* Hides text from input messages / +export const OPENINFERENCE_HIDE_INPUT_TEXT = "OPENINFERENCE_HIDE_INPUT_TEXT"; +/* Hides text from output messages / +export const OPENINFERENCE_HIDE_OUTPUT_TEXT = "OPENINFERENCE_HIDE_OUTPUT_TEXT"; +/* Hides embedding vectors / +export const OPENINFERENCE_HIDE_EMBEDDING_VECTORS = + "OPENINFERENCE_HIDE_EMBEDDING_VECTORS"; +/* Limits characters of a base64 encoding of an image / +export const OPENINFERENCE_BASE64_IMAGE_MAX_LENGTH = + "OPENINFERENCE_BASE64_IMAGE_MAX_LENGTH"; + +export const DEFAULT_HIDE_INPUTS = false; +export const DEFAULT_HIDE_OUTPUTS = false; + +export const DEFAULT_HIDE_INPUT_MESSAGES = false; +export const DEFAULT_HIDE_OUTPUT_MESSAGES = false; + +export const DEFAULT_HIDE_INPUT_IMAGES = false; +export const DEFAULT_HIDE_INPUT_TEXT = false; +export const DEFAULT_HIDE_OUTPUT_TEXT = false; + +export const DEFAULT_HIDE_EMBEDDING_VECTORS = false; +export const DEFAULT_BASE64_IMAGE_MAX_LENGTH = 32000; + +/* When a value is hidden, it will be replaced by this redacted value / +export const REDACTED_VALUE = "__REDACTED__"; + +/* + * The default, environment, and type information for each value on the TraceConfig + * Used to generate a full TraceConfig object with the correct types and default values + / +export const traceConfigMetadata: Readonly< + Record<TraceConfigKey, TraceConfigFlag> +> = { + hideInputs: { + default: DEFAULT_HIDE_INPUTS, + envKey: OPENINFERENCE_HIDE_INPUTS, + type: "boolean", + }, + hideOutputs: { + default: DEFAULT_HIDE_OUTPUTS, + envKey: OPENINFERENCE_HIDE_OUTPUTS, + type: "boolean", + }, + hideInputMessages: { + default: DEFAULT_HIDE_INPUT_MESSAGES, + envKey: OPENINFERENCE_HIDE_INPUT_MESSAGES, + type: "boolean", + }, + hideOutputMessages: { + default: DEFAULT_HIDE_OUTPUT_MESSAGES, + envKey: OPENINFERENCE_HIDE_OUTPUT_MESSAGES, + type: "boolean", + }, + hideInputImages: { + default: DEFAULT_HIDE_INPUT_IMAGES, + envKey: OPENINFERENCE_HIDE_INPUT_IMAGES, + type: "boolean", + }, + hideInputText: { + default: DEFAULT_HIDE_INPUT_TEXT, + envKey: OPENINFERENCE_HIDE_INPUT_TEXT, + type: "boolean", + }, + hideOutputText: { + default: DEFAULT_HIDE_OUTPUT_TEXT, + envKey: OPENINFERENCE_HIDE_OUTPUT_TEXT, + type: "boolean", + }, + hideEmbeddingVectors: { + default: DEFAULT_HIDE_EMBEDDING_VECTORS, + envKey: OPENINFERENCE_HIDE_EMBEDDING_VECTORS, + type: "boolean", + }, + base64ImageMaxLength: { + default: DEFAULT_BASE64_IMAGE_MAX_LENGTH, + envKey: OPENINFERENCE_BASE64_IMAGE_MAX_LENGTH, + type: "number", + }, +}; + +export const DefaultTraceConfig: TraceConfig = { + hideInputs: DEFAULT_HIDE_INPUTS, + hideOutputs: DEFAULT_HIDE_OUTPUTS, + hideInputMessages: DEFAULT_HIDE_INPUT_MESSAGES, + hideOutputMessages: DEFAULT_HIDE_OUTPUT_MESSAGES, + hideInputImages: DEFAULT_HIDE_INPUT_IMAGES, + hideInputText: DEFAULT_HIDE_INPUT_TEXT, + hideOutputText: DEFAULT_HIDE_OUTPUT_TEXT, + hideEmbeddingVectors: DEFAULT_HIDE_EMBEDDING_VECTORS, + base64ImageMaxLength: DEFAULT_BASE64_IMAGE_MAX_LENGTH, +}; diff --git a/js/packages/openinference-core/src/trace/trace-config/index.ts b/js/packages/openinference-core/src/trace/trace-config/index.ts new file mode 100644 index 000000000..f325bc598 --- /dev/null +++ b/js/packages/openinference-core/src/trace/trace-config/index.ts @@ -0,0 +1,24 @@ +export { TraceConfigOptions } from "./types"; +export { + OPENINFERENCE_BASE64_IMAGE_MAX_LENGTH, + OPENINFERENCE_HIDE_EMBEDDING_VECTORS, + DEFAULT_BASE64_IMAGE_MAX_LENGTH, + DEFAULT_HIDE_EMBEDDING_VECTORS, + DEFAULT_HIDE_INPUTS, + DEFAULT_HIDE_INPUT_IMAGES, + DEFAULT_HIDE_INPUT_MESSAGES, + DEFAULT_HIDE_INPUT_TEXT, + DEFAULT_HIDE_OUTPUTS, + DEFAULT_HIDE_OUTPUT_MESSAGES, + DEFAULT_HIDE_OUTPUT_TEXT, + OPENINFERENCE_HIDE_INPUTS, + OPENINFERENCE_HIDE_INPUT_IMAGES, + OPENINFERENCE_HIDE_INPUT_MESSAGES, + OPENINFERENCE_HIDE_INPUT_TEXT, + OPENINFERENCE_HIDE_OUTPUTS, + OPENINFERENCE_HIDE_OUTPUT_MESSAGES, + OPENINFERENCE_HIDE_OUTPUT_TEXT, + REDACTED_VALUE, +} from "./constants"; +export { OITracer } from "./OITracer"; +export { type OISpan } from "./OISpan"; diff --git a/js/packages/openinference-core/src/trace/trace-config/maskingRules.ts b/js/packages/openinference-core/src/trace/trace-config/maskingRules.ts new file mode 100644 index 000000000..c4cbed0f9 --- /dev/null +++ b/js/packages/openinference-core/src/trace/trace-config/maskingRules.ts @@ -0,0 +1,208 @@ +import { AttributeValue } from "@opentelemetry/api"; +import { REDACTED_VALUE } from "./constants"; +import { SemanticConventions } from "@arizeai/openinference-semantic-conventions"; +import { MaskingRule, MaskingRuleArgs } from "./types"; + +/* + * Masks (redacts) input text in LLM input messages. + * Will mask information stored under the key `llm.input_messages.[i].message.content`. + * @example + * ```typescript + * maskInputTextRule.condition({ + * config: {hideInputText: true}, + * key: "llm.input_messages.[i].message.content" + * }) // returns true so the rule applies and the value will be redacted + / +const maskInputTextRule: MaskingRule = { + condition: ({ config, key }) => + config.hideInputText && + key.includes(SemanticConventions.LLM_INPUT_MESSAGES) && + key.includes(SemanticConventions.MESSAGE_CONTENT) && + !key.includes(SemanticConventions.MESSAGE_CONTENTS), + action: () => REDACTED_VALUE, +}; + +/* + * Masks (redacts) output text in LLM output messages. + * Will mask information stored under the key `llm.output_messages.[i].message.content`. + * @example + * ```typescript + * maskOutputTextRule.condition({ + * config: {hideOutputText: true}, + * key: "llm.output_messages.[i].message.content" + * }) // returns true so the rule applies and the value will be redacted + * ``` + / +const maskOutputTextRule: MaskingRule = { + condition: ({ config, key }) => + config.hideOutputText && + key.includes(SemanticConventions.LLM_OUTPUT_MESSAGES) && + key.includes(SemanticConventions.MESSAGE_CONTENT) && + !key.includes(SemanticConventions.MESSAGE_CONTENTS), + action: () => REDACTED_VALUE, +}; + +/* + * Masks (redacts) input text content in LLM input messages. + * Will mask information stored under the key `llm.input_messages.[i].message.contents.[j].message_content.text`. + * @example + * ```typescript + * maskOutputTextRule.condition({ + * config: {hideInputText: true}, + * key: "llm.input_messages.[i].message.contents.[j].message_content.text" + * }) // returns true so the rule applies and the value will be redacted + / +const maskInputTextContentRule: MaskingRule = { + condition: ({ config, key }) => + config.hideInputText && + key.includes(SemanticConventions.LLM_INPUT_MESSAGES) && + key.includes(SemanticConventions.MESSAGE_CONTENT_TEXT), + action: () => REDACTED_VALUE, +}; + +/* + * Masks (redacts) output text content in LLM output messages. + * @example + * ```typescript + * maskOutputTextRule.condition({ + * config: {hideOutputText: true}, + * key: "llm.output_messages.[i].message.contents.[j].message_content.text" + * }) // returns true so the rule applies and the value will be redacted + / +const maskOutputTextContentRule: MaskingRule = { + condition: ({ config, key }) => + config.hideOutputText && + key.includes(SemanticConventions.LLM_OUTPUT_MESSAGES) && + key.includes(SemanticConventions.MESSAGE_CONTENT_TEXT), + action: () => REDACTED_VALUE, +}; + +/* + * Masks (removes) input images in LLM input messages. + * @example + * ```typescript + * maskOutputTextRule.condition({ + * config: {hideInputImages: true}, + * key: "llm.input_messages.[i].message.contents.[j].message_content.image" + * }) // returns true so the rule applies and the value will be removed + / +const maskInputImagesRule: MaskingRule = { + condition: ({ config, key }) => + config.hideInputImages && + key.includes(SemanticConventions.LLM_INPUT_MESSAGES) && + key.includes(SemanticConventions.MESSAGE_CONTENT_IMAGE), + action: () => undefined, +}; + +function isBase64Url(url?: AttributeValue): boolean { + return ( + typeof url === "string" && + url.startsWith("data:image/") && + url.includes("base64") + ); +} + +/* + * Masks (redacts) base64 images that are too long. + * * @example + * ```typescript + * maskOutputTextRule.condition({ + * config: {base64ImageMaxLength: 10}, + * key: "llm.input_messages.[i].message.contents.[j].message_content.image.url", + * value: "data:image/base64,verylongbase64string" + * }) // returns true so the rule applies and the value will be redacted + / +const maskLongBase64ImageRule: MaskingRule = { + condition: ({ config, key, value }) => + typeof value === "string" && + isBase64Url(value) && + value.length > config.base64ImageMaxLength && + key.includes(SemanticConventions.LLM_INPUT_MESSAGES) && + key.includes(SemanticConventions.MESSAGE_CONTENT_IMAGE) && + key.endsWith(SemanticConventions.IMAGE_URL), + action: () => REDACTED_VALUE, +}; + +/* + * Masks (removes) embedding vectors. + * * @example + * ```typescript + * maskOutputTextRule.condition({ + * config: {hideEmbeddingVectors: true}, + * key: "embedding.embeddings.[i].embedding.vector" + * }) // returns true so the rule applies and the value will be redacted + / +const maskEmbeddingVectorsRule: MaskingRule = { + condition: ({ config, key }) => + config.hideEmbeddingVectors && + key.includes(SemanticConventions.EMBEDDING_EMBEDDINGS) && + key.includes(SemanticConventions.EMBEDDING_VECTOR), + action: () => undefined, +}; + +/* + * A list of {@link MaskingRule}s that are applied to span attributes to either redact or remove sensitive information. + * The order of these rules is important as it can ensure appropriate masking of information + * Rules should go from more specific to more general so that things like `llm.input_messages.[i].message.content` are masked with {@link REDACTED_VALUE} before the more generic masking of `llm.input_messages` might happen with `undefined` might happen. + / +const maskingRules: MaskingRule[] = [ + { + condition: ({ config, key }) => + config.hideInputs && key === SemanticConventions.INPUT_VALUE, + action: () => REDACTED_VALUE, + }, + { + condition: ({ config, key }) => + config.hideInputs && key === SemanticConventions.INPUT_MIME_TYPE, + action: () => undefined, + }, + { + condition: ({ config, key }) => + config.hideOutputs && key === SemanticConventions.OUTPUT_VALUE, + action: () => REDACTED_VALUE, + }, + { + condition: ({ config, key }) => + config.hideOutputs && key === SemanticConventions.OUTPUT_MIME_TYPE, + action: () => undefined, + }, + { + condition: ({ config, key }) => + (config.hideInputs \|\| config.hideInputMessages) && + key.includes(SemanticConventions.LLM_INPUT_MESSAGES), + action: () => undefined, + }, + { + condition: ({ config, key }) => + (config.hideOutputs \|\| config.hideOutputMessages) && + key.includes(SemanticConventions.LLM_OUTPUT_MESSAGES), + action: () => undefined, + }, + maskInputTextRule, + maskOutputTextRule, + maskInputTextContentRule, + maskOutputTextContentRule, + maskInputImagesRule, + maskLongBase64ImageRule, + maskEmbeddingVectorsRule, +]; + +/* + * A function that masks (redacts or removes) sensitive information from span attributes based on the trace config. + * @param config The {@link TraceConfig} to use to determine if the value should be masked + * @param key The key of the attribute to mask + * @param value The value of the attribute to mask + * @returns The redacted value or undefined if the value should be masked, otherwise the original value + / +export function mask({ + config, + key, + value, +}: MaskingRuleArgs): AttributeValue \| undefined { + for (const rule of maskingRules) { + if (rule.condition({ config, key, value })) { + return rule.action(); + } + } + return value; +} diff --git a/js/packages/openinference-core/src/trace/trace-config/traceConfig.ts b/js/packages/openinference-core/src/trace/trace-config/traceConfig.ts new file mode 100644 index 000000000..2eef793c2 --- /dev/null +++ b/js/packages/openinference-core/src/trace/trace-config/traceConfig.ts @@ -0,0 +1,68 @@ +import { DefaultTraceConfig, traceConfigMetadata } from "./constants"; +import { TraceConfigKey, TraceConfig, TraceConfigOptions } from "./types"; +import { assertUnreachable, withSafety } from "../../utils"; + +const safelyParseInt = withSafety({ fn: parseInt }); + +type TraceConfigOptionMetadata = (typeof traceConfigMetadata)[TraceConfigKey]; + +/* + * Parses an option based on its type + * The order of precedence is: optionValue > envValue > defaultValue + * @param key - The key of the option. + * @param optionMetadata - The {@link TraceConfigOptionMetadata} for the option which includes its type, default value, and environment variable key. + * + / +function parseOption({ + optionValue, + optionMetadata, +}: { + optionValue?: number \| boolean; + optionMetadata: TraceConfigOptionMetadata; +}) { + if (optionValue !== undefined) { + return optionValue; + } + const envValue = process.env[optionMetadata.envKey]; + if (envValue !== undefined) { + switch (optionMetadata.type) { + case "number": { + const maybeEnvNumber = safelyParseInt(envValue); + return maybeEnvNumber != null && !isNaN(maybeEnvNumber) + ? maybeEnvNumber + : optionMetadata.default; + } + case "boolean": + return envValue.toLowerCase() === "true"; + default: + assertUnreachable(optionMetadata); + } + } + + return optionMetadata.default; +} + +/* + * Generates a full trace config object based on passed in options, environment variables, and default values. + * The order of precedence is: optionValue > envValue > defaultValue + * @param options - The user provided TraceConfigOptions. + * @returns A full TraceConfig object with all options set to their final values. + / +export function generateTraceConfig(options?: TraceConfigOptions): TraceConfig { + if (options == null) { + return DefaultTraceConfig; + } + return Object.entries(traceConfigMetadata).reduce( + (config, [key, optionMetadata]) => { + const TraceConfigKey = key as TraceConfigKey; + return { + ...config, + [TraceConfigKey]: parseOption({ + optionValue: options[TraceConfigKey], + optionMetadata, + }), + }; + }, + {} as TraceConfig, + ); +} diff --git a/js/packages/openinference-core/src/trace/trace-config/types.ts b/js/packages/openinference-core/src/trace/trace-config/types.ts new file mode 100644 index 000000000..cb18e8439 --- /dev/null +++ b/js/packages/openinference-core/src/trace/trace-config/types.ts @@ -0,0 +1,120 @@ +import { AttributeValue } from "@opentelemetry/api"; +import { OISpan } from "./OISpan"; + +/* + * Tracing configuration options that can be set to hide or redact sensitive information from traces + / +export type TraceConfigOptions = { + hideInputs?: boolean; + hideOutputs?: boolean; + hideInputMessages?: boolean; + hideOutputMessages?: boolean; + hideInputImages?: boolean; + hideInputText?: boolean; + hideOutputText?: boolean; + hideEmbeddingVectors?: boolean; + base64ImageMaxLength?: number; +}; + +/* + * A full tracing configuration object that includes all possible options with their final values + / +export type TraceConfig = Readonly<Required<TraceConfigOptions>>; + +export type TraceConfigKey = keyof TraceConfig; + +type BooleanTraceConfigFlag = { + /* + * The default value for the flag + / + default: boolean; + /* + * The environment variable key to use to override the default value + / + envKey: string; + /* + * The type of the flag + / + type: "boolean"; +}; + +type NumericTraceConfigFlag = { + /* + * The default value for the flag + / + default: number; + /* + * The environment variable key to use to override the default value + / + envKey: string; + /* + * The type of the flag + / + type: "number"; +}; + +/* + * The default and environment information for a value on the TraceConfig + * Used to generate a full TraceConfig object + * + * @example + * ```typescript + * { + * hideInputs: { + * default: DEFAULT_HIDE_INPUTS, // false + * envKey: OPENINFERENCE_HIDE_INPUTS, // "OPENINFERENCE_HIDE_INPUTS" + * type: "boolean", + * } + * } + * ``` + / +export type TraceConfigFlag = BooleanTraceConfigFlag \| NumericTraceConfigFlag; + +/* + * The arguments for a masking rule, used to determine if a value should be masked (redacted or omitted) + / +export type MaskingRuleArgs = { + /* + * The trace config to use to determine if the value should be masked + / + config: TraceConfig; + /* + * The key of the attribute to mask + / + key: string; + /* + * The value of the attribute to mask + / + value?: AttributeValue; +}; + +/* + * A condition to determine if a value should be masked + * and an action that masks (removes or redacts) the value if the condition is met + * @example + * ```typescript + * const config = {hideInputText: true} + * const key = "llm.input_messages.0.message.content" + * if (maskInputTextRule.condition({ config, key })) { + * return maskInputTextRule.action() + * } + * ``` + / +export type MaskingRule = { + /* + * + * @param args The {@linkcode MaskingRuleArgs} to determine if the value should be masked + * @returns true if the value should be masked, false otherwise + / + condition: (args: MaskingRuleArgs) => boolean; + /* + * An action to be applied if the condition is met + * @returns A redacted value or undefined + / + action: () => AttributeValue \| undefined; +}; + +/* + * A callback that is called when a new active {@link OISpan} is created + / +export type OpenInferenceActiveSpanCallback = (span: OISpan) => void; diff --git a/js/packages/openinference-core/src/utils/typeUtils.ts b/js/packages/openinference-core/src/utils/typeUtils.ts index 6d65f68dc..fae0fa660 100644 --- a/js/packages/openinference-core/src/utils/typeUtils.ts +++ b/js/packages/openinference-core/src/utils/typeUtils.ts @@ -48,3 +48,25 @@ export function isAttributes(value: unknown): value is Attributes { ) ); } + +/* + * A type check function to ensure that a switch or set of conditionals is exhaustive. + * Typscript will throw an error if the switch or conditionals are not exhaustive. + * @example + * ```typescript + * type MyType = "a" \| "b"; + * function myFunction(value: MyType) { + * switch (value) { + * case "a": + * return "A"; + * case "b": + * return "B"; + * default: + * assertUnreachable(value); + * } + * } + * ``` + / +export function assertUnreachable(value: never): never { + throw new Error(`Unreachable code reached with value: ${value}`); +} diff --git a/js/packages/openinference-core/tsconfig.esm.json b/js/packages/openinference-core/tsconfig.esm.json index 0b9947f92..737b185af 100644 --- a/js/packages/openinference-core/tsconfig.esm.json +++ b/js/packages/openinference-core/tsconfig.esm.json @@ -3,7 +3,8 @@ "compilerOptions": { "outDir": "dist/esm", "rootDir": "src", - "tsBuildInfoFile": "dist/esm/tsconfig.esm.tsbuildinfo" + "tsBuildInfoFile": "dist/esm/tsconfig.esm.tsbuildinfo", + "baseUrl": "./src" }, "include": ["src//.ts"], "references": [] diff --git a/js/packages/openinference-core/tsconfig.esnext.json b/js/packages/openinference-core/tsconfig.esnext.json index 84a00950b..7f1cd3f1f 100644 --- a/js/packages/openinference-core/tsconfig.esnext.json +++ b/js/packages/openinference-core/tsconfig.esnext.json @@ -3,7 +3,8 @@ "compilerOptions": { "outDir": "dist/esnext", "rootDir": "src", - "tsBuildInfoFile": "dist/esnext/tsconfig.esnext.tsbuildinfo" + "tsBuildInfoFile": "dist/esnext/tsconfig.esnext.tsbuildinfo", + "baseUrl": "./src" }, "include": ["src/*/.ts"], "references": [] diff --git a/js/packages/openinference-core/tsconfig.json b/js/packages/openinference-core/tsconfig.json index 64960dc0f..9099ee154 100644 --- a/js/packages/openinference-core/tsconfig.json +++ b/js/packages/openinference-core/tsconfig.json @@ -3,7 +3,8 @@ "compilerOptions": { "outDir": "dist", "rootDir": ".", - "types": ["node", "jest"] + "types": ["node", "jest"], + "baseUrl": "./src" }, "files": [], "include": ["src/*/.ts", "test/*/.ts"], diff --git a/js/packages/openinference-instrumentation-openai/package.json b/js/packages/openinference-instrumentation-openai/package.json index 9d5fe3d10..e1b16ec34 100644 --- a/js/packages/openinference-instrumentation-openai/package.json +++ b/js/packages/openinference-instrumentation-openai/package.json @@ -38,6 +38,7 @@ "@opentelemetry/sdk-trace-base": "^1.25.1", "@opentelemetry/sdk-trace-node": "^1.25.1", "@opentelemetry/semantic-conventions": "^1.25.1", + "jest": "^29.7.0", "openai": "^4.52.7" } } diff --git a/js/pnpm-lock.yaml b/js/pnpm-lock.yaml index 4c78f7a96..435a8855c 100644 --- a/js/pnpm-lock.yaml +++ b/js/pnpm-lock.yaml @@ -48,24 +48,24 @@ importers: specifier: workspace:* version: link:../openinference-semantic-conventions '@opentelemetry/api': - specifier: '>=1.0.0 <1.9.0' - version: 1.8.0 + specifier: ^1.9.0 + version: 1.9.0 '@opentelemetry/core': specifier: ^1.25.1 - version: 1.25.1(@opentelemetry/[email protected]) + version: 1.25.1(@opentelemetry/[email protected]) devDependencies: '@opentelemetry/context-async-hooks': specifier: ^1.25.1 - version: 1.25.1(@opentelemetry/[email protected]) + version: 1.25.1(@opentelemetry/[email protected]) '@opentelemetry/resources': specifier: ^1.19.0 - version: 1.25.1(@opentelemetry/[email protected]) + version: 1.25.1(@opentelemetry/[email protected]) '@opentelemetry/sdk-trace-base': specifier: ^1.19.0 - version: 1.25.1(@opentelemetry/[email protected]) + version: 1.25.1(@opentelemetry/[email protected]) '@opentelemetry/sdk-trace-node': specifier: ^1.19.0 - version: 1.25.1(@opentelemetry/[email protected]) + version: 1.25.1(@opentelemetry/[email protected]) '@opentelemetry/semantic-conventions': specifier: ^1.19.0 version: 1.26.0 @@ -191,6 +191,9 @@ importers: '@opentelemetry/semantic-conventions': specifier: ^1.25.1 version: 1.26.0 + jest: + specifier: ^29.7.0 + version: 29.7.0(@types/[email protected]) openai: specifier: ^4.52.7 version: 4.56.0([email protected]) @@ -3880,10 +3883,6 @@ snapshots: '@opentelemetry/[email protected]': {} - '@opentelemetry/[email protected](@opentelemetry/[email protected])': - dependencies: - '@opentelemetry/api': 1.8.0 - '@opentelemetry/[email protected](@opentelemetry/[email protected])': dependencies: '@opentelemetry/api': 1.9.0 @@ -3946,21 +3945,11 @@ snapshots: '@opentelemetry/sdk-metrics': 1.23.0(@opentelemetry/[email protected]) '@opentelemetry/sdk-trace-base': 1.23.0(@opentelemetry/[email protected]) - '@opentelemetry/[email protected](@opentelemetry/[email protected])': - dependencies: - '@opentelemetry/api': 1.8.0 - '@opentelemetry/core': 1.25.1(@opentelemetry/[email protected]) - '@opentelemetry/[email protected](@opentelemetry/[email protected])': dependencies: '@opentelemetry/api': 1.9.0 '@opentelemetry/core': 1.25.1(@opentelemetry/[email protected]) - '@opentelemetry/[email protected](@opentelemetry/[email protected])': - dependencies: - '@opentelemetry/api': 1.8.0 - '@opentelemetry/core': 1.25.1(@opentelemetry/[email protected]) - '@opentelemetry/[email protected](@opentelemetry/[email protected])': dependencies: '@opentelemetry/api': 1.9.0 @@ -4019,16 +4008,6 @@ snapshots: '@opentelemetry/resources': 1.25.1(@opentelemetry/[email protected]) '@opentelemetry/semantic-conventions': 1.25.1 - '@opentelemetry/[email protected](@opentelemetry/[email protected])': - dependencies: - '@opentelemetry/api': 1.8.0 - '@opentelemetry/context-async-hooks': 1.25.1(@opentelemetry/[email protected]) - '@opentelemetry/core': 1.25.1(@opentelemetry/[email protected]) - '@opentelemetry/propagator-b3': 1.25.1(@opentelemetry/[email protected]) - '@opentelemetry/propagator-jaeger': 1.25.1(@opentelemetry/[email protected]) - '@opentelemetry/sdk-trace-base': 1.25.1(@opentelemetry/[email protected]) - semver: 7.6.3 - '@opentelemetry/[email protected](@opentelemetry/[email protected])': dependencies: '@opentelemetry/api': 1.9.0	[instrumentation-js] TraceConfig for openinference-core Mimic the behavior of the trace config for python - e.g. in terms of scope and coherence with the environment variables		2024-09-19T00:01:49	[]	[]
Arize-ai/openinference	Arize-ai__openinference-1045	50cd7d5f58ab0a7d424f8c9950a22eb912330663	diff --git a/python/instrumentation/openinference-instrumentation-langchain/pyproject.toml b/python/instrumentation/openinference-instrumentation-langchain/pyproject.toml index 3ebed9e39..77c603020 100644 --- a/python/instrumentation/openinference-instrumentation-langchain/pyproject.toml +++ b/python/instrumentation/openinference-instrumentation-langchain/pyproject.toml @@ -45,6 +45,10 @@ test = [ "opentelemetry-sdk", "respx", "pytest-recording", + "langsmith", + "portpicker", + "starlette", + "uvicorn", ] type-check = [ "langchain_core == 0.1.5", @@ -73,7 +77,6 @@ testpaths = [ [tool.mypy] strict = true -explicit_package_bases = true exclude = [ "examples", "dist", diff --git a/python/tox.ini b/python/tox.ini index 0aef11523..181ee95e3 100644 --- a/python/tox.ini +++ b/python/tox.ini @@ -10,7 +10,7 @@ envlist = py3{8,12}-ci-{vertexai,vertexai-latest} py3{8,12}-ci-{llama_index,llama_index-latest} py3{9,12}-ci-{dspy,dspy-latest} - py3{8,12}-ci-{langchain,langchain-latest} + py3{9,12}-ci-{langchain,langchain-latest} ; py3{9,12}-ci-{guardrails,guardrails-latest} ; py3{10,12}-ci-{crewai,crewai-latest} py3{8,12}-ci-{haystack,haystack-latest} @@ -79,8 +79,8 @@ commands = ruff: ruff format {posargs:.} ruff: ruff check --fix {posargs:.} mypy: mypy {posargs:.} - test: pytest -n auto --nf {posargs:.} - ci: ruff format --diff {posargs:.} - ci: ruff check --no-fix {posargs:.} - ci: mypy {posargs:.} + test: pytest -rfEs -l --nf {posargs:.} + ci: ruff format --diff . + ci: ruff check --no-fix . + ci: mypy . ci: pytest -n auto -x -ra {posargs:.}	[bug] langchain failing on latest 0.3.0 LangChain CI is failing on latest. This could be due to 0.3.0 moving to pydantic 2. Needs investigation.		2024-09-18T19:31:07	[]	[]
Arize-ai/openinference	Arize-ai__openinference-1010	5bc4100c1a9683cff18e90538c857e5844159d31	diff --git a/python/openinference-instrumentation/src/openinference/instrumentation/config.py b/python/openinference-instrumentation/src/openinference/instrumentation/config.py index dd20abefc..1d441f6f8 100644 --- a/python/openinference-instrumentation/src/openinference/instrumentation/config.py +++ b/python/openinference-instrumentation/src/openinference/instrumentation/config.py @@ -1,6 +1,7 @@ import os from contextlib import contextmanager from dataclasses import dataclass, field, fields +from secrets import randbits from typing import ( Any, Callable, @@ -21,7 +22,16 @@ detach, set_value, ) -from opentelemetry.trace import Link, Span, SpanKind, Tracer, use_span +from opentelemetry.sdk.trace import IdGenerator +from opentelemetry.trace import ( + INVALID_SPAN_ID, + INVALID_TRACE_ID, + Link, + Span, + SpanKind, + Tracer, + use_span, +) from opentelemetry.util.types import Attributes, AttributeValue from openinference.semconv.trace import ( @@ -343,10 +353,32 @@ def end(self, end_time: Optional[int] = None) -> None: span.end(end_time) +class _IdGenerator(IdGenerator): + """ + An IdGenerator that uses a different source of randomness to + avoid being affected by seeds set by user application. + """ + + def generate_span_id(self) -> int: + while (span_id := randbits(64)) == INVALID_SPAN_ID: + continue + return span_id + + def generate_trace_id(self) -> int: + while (trace_id := randbits(128)) == INVALID_TRACE_ID: + continue + return trace_id + + class OITracer(wrapt.ObjectProxy): # type: ignore[misc] def __init__(self, wrapped: Tracer, config: TraceConfig) -> None: super().__init__(wrapped) self._self_config = config + self._self_id_generator = _IdGenerator() + + @property + def id_generator(self) -> IdGenerator: + return self._self_id_generator @contextmanager def start_as_current_span( @@ -391,7 +423,8 @@ def start_span( set_status_on_exception: bool = True, ) -> Span: tracer = cast(Tracer, self.__wrapped__) - span = tracer.start_span( + span = tracer.__class__.start_span( + self, name=name, context=context, kind=kind,	[bug] should use a different source of randomness, e.g. secrets, so span identifiers are not affected by seed from application [id_generator](https://github.com/open-telemetry/opentelemetry-python/blob/9697f805d1e5ba53cc55e2502ad1de0cc7c5900a/opentelemetry-sdk/src/opentelemetry/sdk/trace/id_generator.py) is affected by application seeds <img width="668" alt="Screenshot 2024-09-10 at 7 51 52 AM" src="https://github.com/user-attachments/assets/bc8c99ba-1d9c-411d-9e94-d0db2714b121">		2024-09-10T15:50:43	[]	[]
cvxgrp/pymde	cvxgrp__pymde-55	2a271c70ec36fa86efd539ed432c87997b75078b	diff --git a/pymde/quadratic.py b/pymde/quadratic.py index 73a6eac..82e3a60 100644 --- a/pymde/quadratic.py +++ b/pymde/quadratic.py @@ -44,10 +44,27 @@ def pca(Y, embedding_dim): return np.sqrt(float(n)) * U[:, :embedding_dim] -def _laplacian(n, m, edges, weights): - A = util.adjacency_matrix(n, m, edges, weights) +def _laplacian(n, m, edges, weights, use_scipy=True): + A = util.adjacency_matrix(n, m, edges, weights, use_scipy=use_scipy) L = -A - L.setdiag((np.array(A.sum(axis=1)).squeeze())) + if use_scipy: + L.setdiag(np.asarray(A.sum(axis=1)).squeeze()) + else: + # diag is currently 0 + diag_vals = torch.sparse.sum(A, dim=1).to_dense() + diag_inds = torch.tensor( + [[i, i] for i in range(len(A))], device=L.device, dtype=L.dtype + ) + diag_inds = diag_inds.transpose(0, 1) + diag_matrix = torch.sparse_coo_tensor( + diag_inds, + diag_vals, + size=(n, n), + dtype=torch.float32, + device=L.device, + ) + L = L + diag_matrix + L = L.coalesce() return L @@ -62,8 +79,8 @@ def _spectral( device=None, ): n = L.shape[0] + k = m + 1 if not cg: - k = m + 1 num_lanczos_vectors = max(2 * k + 1, int(np.sqrt(L.shape[0]))) eigenvalues, eigenvectors = scipy.sparse.linalg.eigsh( L, @@ -74,33 +91,34 @@ def _spectral( v0=np.ones(L.shape[0]), maxiter=L.shape[0] * 5, ) - order = np.argsort(eigenvalues)[1:k] else: - k = m if warm_start: mde = problem.MDE( - n, m, edges, f=penalties.Quadratic(weights), device=device + n, + k, + edges, + distortion_function=penalties.Quadratic(weights), + device=device, ) - X_init = mde.fit(max_iter=40, use_line_search=False) + X_init = mde.embed(max_iter=40) else: X_init = util.proj_standardized( - torch.tensor(util.np_rng().randn(n, m), device=device), + torch.randn((n, k), device=device), demean=True, ) - eigenvalues, eigenvectors = scipy.sparse.linalg.lobpcg( + eigenvalues, eigenvectors = torch.lobpcg( A=L, - X=X_init.cpu().numpy(), - # Y: search in the orthogonal complement of the ones vector - Y=np.ones((L.shape[0], 1)), + X=X_init, tol=None, # largest: find the smallest eigenvalues largest=False, - maxiter=max_iter, + niter=max_iter, ) - order = np.argsort(eigenvalues)[0:k] + order = np.argsort(eigenvalues)[1:k] return eigenvectors[:, order] +# TODO(akshayka): Deprecate device argument, infer from edges/weights def spectral( n_items, embedding_dim, edges, weights, cg=False, max_iter=40, device="cpu" ): @@ -116,8 +134,6 @@ def spectral( \\mbox{subject to} & (1/n) X^T X = I, \quad d_{ij} = \|x_i - x_j\|_2. \\end{array} - The weights may be negative. - By default, the problem is solved using a Lanczos method. If cg=True, LOBPCG is used; LOBPCG is warm-started by running a projected quasi-newton method for a small number of iterations. Use cg=True when the number @@ -134,7 +150,7 @@ def spectral( edges: torch.Tensor(shape=(n_edges, 2)) A list of edges (i, j), 0 <= i < j < n_items weights: torch.Tensor(shape=(n_edges,)) - A list of weights associated with each edge + A list of nonnegative weights associated with each edge cg: bool If True, uses a preconditioned CG method to find the embedding, which requires that the Laplacian matrix plus the identity is @@ -151,9 +167,13 @@ def spectral( torch.Tensor(shape=(n_items, embedding_dim)) A spectral embedding, projected onto the standardization constraint """ - L = _laplacian(n_items, embedding_dim, edges, weights) + # use torch sparse and linalg for lobpcg + use_scipy = not cg + L = _laplacian(n_items, embedding_dim, edges, weights, use_scipy=use_scipy) emb = _spectral(L, embedding_dim, cg=cg, device=device, max_iter=max_iter) - emb -= emb.mean(axis=0) - return util.proj_standardized( - torch.tensor(emb, dtype=weights.dtype, device=device) - ) + if use_scipy: + emb -= emb.mean(axis=0) + emb = torch.tensor(emb, dtype=weights.dtype, device=device) + else: + emb -= emb.mean(dim=0) + return util.proj_standardized(emb) diff --git a/pymde/recipes.py b/pymde/recipes.py index 925e9be..5bd7397 100644 --- a/pymde/recipes.py +++ b/pymde/recipes.py @@ -292,7 +292,11 @@ def preserve_neighbors( If not None, neighborhoods are restricted to have a radius no greater than ``max_distance``. init: str - Initialization strategy; 'quadratic' or 'random'. + Initialization strategy; 'quadratic' or 'random'. If the quadratic + initialization takes too much time, try a random initialization. + When `device=='cuda'` and `init=='quadratic'` a torch-implemented + version will be used, which is fully on device, but uses a CG-based + technique (versus Lanczos). device: str (optional) Device for the embedding (eg, 'cpu', 'cuda'). verbose: bool @@ -357,9 +361,11 @@ def preserve_neighbors( if init == "quadratic": if verbose: - problem.LOGGER.info("Computing quadratic initialization.") + problem.LOGGER.info(f"Computing {init} initialization.") + # use cg + torch when using GPU + cg = device == "cuda" X_init = quadratic.spectral( - n, embedding_dim, edges, weights, device=device + n, embedding_dim, edges, weights, device=device, cg=cg ) if not isinstance( constraint, (constraints._Centered, constraints._Standardized) diff --git a/pymde/util.py b/pymde/util.py index 770c1e4..0359e76 100644 --- a/pymde/util.py +++ b/pymde/util.py @@ -168,16 +168,30 @@ def proj_standardized(X, demean=False, inplace=False): return proj -def adjacency_matrix(n, m, edges, weights): - if isinstance(weights, torch.Tensor): - weights = weights.detach().cpu().numpy() - if isinstance(edges, torch.Tensor): - edges = edges.detach().cpu().numpy() - A = scipy.sparse.coo_matrix( - (weights, (edges[:, 0], edges[:, 1])), shape=(n, n), dtype=np.float32 - ) - A = A + A.T - return A.tocoo() +def adjacency_matrix(n, m, edges, weights, use_scipy=True): + + if use_scipy: + if isinstance(weights, torch.Tensor): + weights = weights.detach().cpu().numpy() + if isinstance(edges, torch.Tensor): + edges = edges.detach().cpu().numpy() + A = scipy.sparse.coo_matrix( + (weights, (edges[:, 0], edges[:, 1])), + shape=(n, n), + dtype=np.float32, + ) + A = A + A.T + return A.tocoo() + else: + A = torch.sparse_coo_tensor( + edges.transpose(0, 1), + weights, + size=(n, n), + dtype=torch.float32, + device=edges.device, + ) + A = A + A.transpose(0, 1) + return A @tensor_arguments @@ -317,7 +331,7 @@ def align(source, target): def scale_delta(delta, d_nat): # scale delta so RMS(delta) == d_nat N = delta.nelement() - rms = torch.sqrt(1 / N * torch.sum(delta ** 2)) + rms = torch.sqrt(1 / N * torch.sum(delta*2)) return delta d_nat / rms diff --git a/pyproject.toml b/pyproject.toml index 7f09190..0a52494 100644 --- a/pyproject.toml +++ b/pyproject.toml @@ -6,5 +6,8 @@ requires = [ "scipy >= 1.6", "setuptools", "wheel", - "cython" + "cython", ] + +[tool.black] +line-length = 80	Quadratic initialization on GPU The quadratic initialization takes the most time for a dataset with 150k observations and 15 nearest neighbors. Is it possible to use the GPU for this? PyTorch has support for both CUDA-backed COO matrices and [lobpcg](https://pytorch.org/docs/stable/generated/torch.lobpcg.html). See sparse supported linear algebra [operations](https://pytorch.org/docs/stable/sparse.html#supported-linear-algebra-operations).	> PyTorch has support for both CUDA-backed COO matrices and [lobpcg](https://pytorch.org/docs/stable/generated/torch.lobpcg.html). See sparse supported linear algebra [operations](https://pytorch.org/docs/stable/sparse.html#supported-linear-algebra-operations). Oh, that's awesome! I wasn't aware that sparse support had become this mature. I'm definitely open to having this in PyMDE. I could maybe try to implement it, but even if it was in there, it seems like it's not accessible through `preserve_neighbors` (`cg=True`) option for quadratic init	2022-02-25T19:42:47	[]	[]
nickmaccarthy/python-datemath	nickmaccarthy__python-datemath-26	be87139fe3f3a2c8ca681186ab0d17668491ef21	diff --git a/.github/workflows/release.yaml b/.github/workflows/release.yaml new file mode 100644 index 0000000..e912349 --- /dev/null +++ b/.github/workflows/release.yaml @@ -0,0 +1,44 @@ +name: Release Packages +on: + release: + types: + - released +jobs: + release-python2: + runs-on: ubuntu-20.04 + needs: + - build-python2 + steps: + - uses: actions/checkout@v2 + - name: install python2 for ubuntu + run: sudo apt install python2 + - name: get pip script for python2 + run: curl https://bootstrap.pypa.io/pip/2.7/get-pip.py --output get-pip.py + - name: install pip for python2 + run: sudo python2 get-pip.py + - name: install requirements for python2 + run: pip2 install -r requirements-2.txt + - name: package and upload python2 + env: + TWINE_USERNAME: __token__ + TWINE_PASSWORD: ${{ secrets.PYPI_API_TOKEN }} + PYTHON_DATEMATH_VERSION: ${{ github.event.release.tag_name }} + run: \| + python2.7 setup.py sdist bdist_wheel + twine upload dist/* + release-python3: + runs-on: ubuntu-20.04 + needs: + - build-python3 + steps: + - uses: actions/checkout@v2 + - name: install requirements for python3 + run: pip3 install -r requirements-3.txt + - name: package and upload python3 + env: + TWINE_USERNAME: __token__ + TWINE_PASSWORD: ${{ secrets.PYPI_API_TOKEN }} + PYTHON_DATEMATH_VERSION: ${{ github.event.release.tag_name }} + run: \| + python3 setup.py sdist bdist_wheel + twine upload dist/* \ No newline at end of file diff --git a/.gitignore b/.gitignore index 79658d6..2849d3d 100644 --- a/.gitignore +++ b/.gitignore @@ -1,10 +1,6 @@ .pyc -env -venv -arrow_test.py -dm_pretty.py -ranges.py -tztest.py +env +venv* # general things to ignore build/ diff --git a/CHANGELOG.md b/CHANGELOG.md index 4baaa19..7b766d6 100644 --- a/CHANGELOG.md +++ b/CHANGELOG.md @@ -1,5 +1,8 @@ # Changelog +## 1.5.3 (2021-04-16) +* [FIX] [Issue #25](https://github.com/nickmaccarthy/python-datemath/issues/25) - Fixed an issue where if you provided an invalid timestamp, i.e. `datemath('2')` you would not get an DateMathException back. Also bumped dependencies. + ## 1.5.2 (2020-10-01) * [FIX] [Issue #21](https://github.com/nickmaccarthy/python-datemath/issues/21) - Fixed an issue where if timezone offset was in a datetime string (ISO8601), the timezone of the returned datemath object would be UTC and not the timezone as specified in the datetime string. diff --git a/VERSION.txt b/VERSION.txt index 4cda8f1..8af85be 100644 --- a/VERSION.txt +++ b/VERSION.txt @@ -1,1 +1,1 @@ -1.5.2 +1.5.3 diff --git a/datemath/__init__.py b/datemath/__init__.py index e0246bf..e0e771e 100644 --- a/datemath/__init__.py +++ b/datemath/__init__.py @@ -1,4 +1,4 @@ -from .helpers import parse +from .helpers import parse, DateMathException def dm(expr, kwargs): ''' does our datemath and returns an arrow object ''' @@ -6,4 +6,4 @@ def dm(expr, kwargs): def datemath(expr, kwargs): ''' does our datemath and returns a datetime object ''' - return parse(expr, kwargs).datetime + return parse(expr, **kwargs) diff --git a/datemath/helpers.py b/datemath/helpers.py index ef3d995..f953611 100644 --- a/datemath/helpers.py +++ b/datemath/helpers.py @@ -136,6 +136,7 @@ def parse(expression, now=None, tz='UTC', type=None, roundDown=True): math = expression time = now else: + if debug: print('parse() - Found and expression that will hit the catchall') math = '' time = parseTime(expression, tz) @@ -174,6 +175,9 @@ def parseTime(timestamp, timezone='UTC'): ts = ts.replace(tzinfo=timezone) return ts + else: + if debug: print('parseTime() - Doesnt look like we have a valid timestamp, raise an exception. timestamp={}'.format(timestamp)) + raise DateMathException('Valid length timestamp not provide, you gave me a timestamp of "{}", but I need something that has a len() >= 4'.format(timestamp)) def roundDate(now, unit, tz='UTC', roundDown=True): ''' diff --git a/requirements-2.txt b/requirements-2.txt index 60ec35e..6e5b16d 100644 --- a/requirements-2.txt +++ b/requirements-2.txt @@ -7,8 +7,6 @@ packaging==16.8 pkginfo==1.4.2 pyparsing==2.2.0 python-dateutil==2.6.0 -requests==2.13.0 -requests-toolbelt>=0.8.0 six==1.10.0 traceback2==1.4.0 twine>=1.8.1 diff --git a/requirements-3.txt b/requirements-3.txt index 5eba2ad..268d6ea 100644 --- a/requirements-3.txt +++ b/requirements-3.txt @@ -1,7 +1,7 @@ appdirs==1.4.3 args==0.1.0 arrow==0.15.2 -bleach==3.1.2 +bleach==3.3.0 certifi==2019.9.11 chardet==3.0.4 clint==0.5.1 @@ -9,8 +9,8 @@ docutils==0.15.2 idna==2.7 linecache2==1.0.0 packaging==16.8 -pkginfo==1.4.1 -Pygments==2.4.2 +pkginfo==1.4.2 +Pygments==2.7.4 pyparsing==2.2.0 python-dateutil==2.6.0 readme-renderer==24.0	Invalid datemath expression can lead to NoneType Hi, when datemath gets an invalid argument, it can happen that `None `is returned instead of an exception. This is can lead to a weird NoneType if `datemath.datemath` is used instead of `datemath.parse`. ```python import datemath type(datemath.parse("2")) # --> None datemath.datemath("2") # AttributeError: 'NoneType' object has no attribute 'datetime' ``` I use version 1.5.2 (latest). It would be great if both cases could raise a parse exception. Thanks, Kai		2021-04-16T14:16:57	[]	[]
nickmaccarthy/python-datemath	nickmaccarthy__python-datemath-22	b71536403908faf74f513fd458ff577f6c41bffc	diff --git a/.gitignore b/.gitignore index 546d997..79658d6 100644 --- a/.gitignore +++ b/.gitignore @@ -1,5 +1,6 @@ .pyc env +venv arrow_test.py dm_pretty.py ranges.py diff --git a/CHANGELOG.md b/CHANGELOG.md index 46d3cfe..4baaa19 100644 --- a/CHANGELOG.md +++ b/CHANGELOG.md @@ -1,5 +1,8 @@ # Changelog +## 1.5.2 (2020-10-01) + [FIX] [Issue #21](https://github.com/nickmaccarthy/python-datemath/issues/21) - Fixed an issue where if timezone offset was in a datetime string (ISO8601), the timezone of the returned datemath object would be UTC and not the timezone as specified in the datetime string. + ## 1.5.1 (2020-03-25) * [FIX] [Issue #15](https://github.com/nickmaccarthy/python-datemath/issues/15) - Fixed issue with parser finding invalid timeunits and throwing correct errors diff --git a/README.md b/README.md index 55539ce..7d0f9cd 100644 --- a/README.md +++ b/README.md @@ -1,32 +1,37 @@ [![Build Status](https://travis-ci.org/nickmaccarthy/python-datemath.svg?branch=master)](https://travis-ci.org/nickmaccarthy/python-datemath.svg?branch=master) +# Python Datemath + +## What? -# What? A date math (aka datemath) parser compatiable with the elasticsearch 'date math' format -# Why? -Working with date objects in python has always been interesting. Having a background in php, I have been looking for quite some time ( no pun intended ) for a way to do date time interpolation similar to php's ```strtotime()``` function. While the arrow module comes close, I needed something that could turn date math type strings into datetime objects for use in tattle.io and other projects I use in elasticsearch. I have found even more uses for it, including AWS cloudwatch and various other projects and hopefully you will too. +## Why? + +Working with date objects in python has always been interesting. Having a background in php, I have been looking for quite some time ( no pun intended ) for a way to do date time interpolation similar to php's ```strtotime()``` function. While the arrow module comes close, I needed something that could turn date math type strings into datetime objects for use in [tattle.io](http://tattle.io) and other projects I use in elasticsearch. I have found even more uses for it, including AWS cloudwatch and various other projects and hopefully you will too. + +## What is date math? -# What is date math ? Date Math is the short hand arithmetic to find relative time to fixed moments in date and time. Similar to the SOLR date math format, Elasticsearch has its own built in format for short hand date math and this module aims to support that same coverage in python. Documentation from elasticsearch: -http://www.elasticsearch.org/guide/en/elasticsearch/reference/current/mapping-date-format.html#date-math +[http://www.elasticsearch.org/guide/en/elasticsearch/reference/current/mapping-date-format.html#date-math](http://www.elasticsearch.org/guide/en/elasticsearch/reference/current/mapping-date-format.html#date-math) > The date type supports using date math expression when using it in a query/filter (mainly makes sense in range query/filter). - +> > The expression starts with an "anchor" date, which can be either now or a date string (in the applicable format) ending with `\|\|`. - +> > It can then follow by a math expression, supporting `+`, `-` and `/` (rounding). - +> > The units supported are `y` (year), `M` (month), `w` (week), `d` (day), `h` (hour), `m` (minute), and `s` (second). - +> > Here are some samples: `now+1h`, `now+1h+1m`, `now+1h/d`, `2012-01-01\|\|+1M/d`. - +> > Note, when doing range type searches, and the upper value is inclusive, the rounding will properly be rounded to the ceiling instead of flooring it. -# Unit Maps -``` +## Unit Maps + +```yaml y or Y = 'year' M = 'month' m = 'minute' @@ -36,13 +41,17 @@ h or H = 'hour' s or S = 'second' ``` -# Install +## Install + ```python pip install python-datemath ``` -# Examples -Assuming our datetime is currently: '2016-01-01T00:00:00-00:00' -``` + +## Examples + +Assuming our datetime is currently: `2016-01-01T00:00:00-00:00` + +```yaml Expression: Result: now-1h 2015-12-31T23:00:00+00:00 now-1y 2015-01-01T00:00:00+00:00 @@ -62,10 +71,11 @@ now/d 2016-01-01T23:59:59+00:00 now/Y 2016-12-31T23:59:59+00:00 ``` -# Usage +## Usage + By default datemath return an arrow date object representing your timestamp. -``` +```python >>> from datemath import dm >>> >>> dm('now+1h') @@ -92,7 +102,8 @@ By default datemath return an arrow date object representing your timestamp. If you would rather have a string, you can use arrow's ```.format()``` method. > For for info on string formatting, check out arrows tokens section: http://crsmithdev.com/arrow/#tokens -``` + +```python >>> from datemath import dm >>> >>> src_timestamp = dm('2016-01-01') @@ -105,11 +116,11 @@ If you would rather have a string, you can use arrow's ```.format()``` method. >>> >>> new_timestamp.format('YYYY.MM.DD') u'2015.12.18' ->>> ``` Rather have a python datetime object instead? Just pass along the 'datetime' type -``` + +```python from datemath import dm >>> dm('now', type='datetime') datetime.datetime(2016, 1, 22, 22, 58, 28, 338060, tzinfo=tzutc()) @@ -119,7 +130,8 @@ datetime.datetime(2016, 1, 24, 22, 57, 45, 394470, tzinfo=tzutc()) ``` Or you can just import the `datemath` module, this will always give us a native `datetime` object -``` + +```python >>> from datemath import datemath >>> >>> datemath('2016-01-01T16:20:00\|\|/d', roundDown=False) @@ -129,20 +141,28 @@ datetime.datetime(2016, 1, 1, 23, 59, 59, 999999, tzinfo=tzutc()) >>> # roundDown=True is default and implied >>> datemath('2016-01-01T16:20:00\|\|/d') datetime.datetime(2016, 1, 1, 0, 0, tzinfo=tzutc()) ->>> ``` If you want a Epoch timestamp back instead, we can do that. + ```python >>> dm('now+2d-1m', type='timestamp') 1453676321 ``` -# What timezone are my objects in? -By default all objects returned by datemath are in UTC. If you want them them in a different timezone, just pass along the ```tz``` argument. -Timezone list can be found here: https://gist.github.com/pamelafox/986163 -``` -from datemath import dm +## What timezone are my objects in? + +By default all object returned by datemath are in UTC. + +If you want them them back in a different timezone, just pass along the ```tz``` argument. Timezone list can be found here: [https://gist.github.com/pamelafox/986163](https://gist.github.com/pamelafox/986163) + +If you provide a timezone offset in your timestring, datemath will return your time object as that timezone offset in the string. + +Note - currently timestrings with a timezone offset and the usage of the ```tz``` argument will result in the time object being returned with the timezone of what was in the timezone offset in the original string + +```python +>>> from datemath import dm +>>> >>> dm('now') <Arrow [2016-01-26T01:00:53.601088+00:00]> >>> @@ -154,9 +174,26 @@ from datemath import dm >>> >>> dm('2017-10-20 09:15:20', tz='US/Pacific') <Arrow [2017-10-20T09:15:20.000000-08:00]> +>>> +>>> # Timestring with TZ offset in the string (ISO8601 format only) +>>> dm('2016-01-01T00:00:00-05:00') +<Arrow [2016-01-01T00:00:00-05:00]> +>>> +>>> # Timestring with TZ offset with datemath added (again, TS must be in ISO8601) +>>> dm('2016-01-01T00:00:00-05:00\|\|+2d+3h+5m') +<Arrow [2016-01-03T03:05:00-05:00]> +>>> +>>> # Note, timestrings with TZ offsets will be returned as the timezone of the offset in the string even if the "tz" option is used. +>>> dm('2016-01-01T00:00:00-05:00', tz='US/Central') +<Arrow [2016-01-01T00:00:00-05:00]> ``` +## Debugging + +If you would like more verbose output to debug the process of what datemath is doing, simply set `export DATEMATH_DEBUG=true` in your shell then run some datemath tests. To stop debugging, run `unset DATEMATH_DEBUG`. + ## Changes + See CHANGELOG.md # Happy date math'ing! diff --git a/VERSION.txt b/VERSION.txt index 26ca594..4cda8f1 100644 --- a/VERSION.txt +++ b/VERSION.txt @@ -1,1 +1,1 @@ -1.5.1 +1.5.2 diff --git a/datemath/helpers.py b/datemath/helpers.py index df9aba7..ef3d995 100644 --- a/datemath/helpers.py +++ b/datemath/helpers.py @@ -42,9 +42,11 @@ import re import os from dateutil import tz +import dateutil import sys +from pprint import pprint -debug = True if os.environ.get('DEBUG') == 'true' else False +debug = True if os.environ.get('DATEMATH_DEBUG') else False class DateMathException(Exception): pass @@ -75,7 +77,7 @@ def unitMap(c): def as_datetime(expression, now, tz='UTC'): ''' - returs our datemath expression as a python datetime object + returns our datemath expression as a python datetime object note: this has been deprecated and the 'type' argument in parse is the current way ''' return parse(expression, now, tz) @@ -89,26 +91,28 @@ def parse(expression, now=None, tz='UTC', type=None, roundDown=True): :param type - if we are dealing with a arrow or datetime object :param roundDown - wether or not we should round up or round down on this. default is roundDown=True, which means if it was 12:00:00, `/d` would be '00:00:00', and with roundDown=False, `/d` would be '29:59:59' ''' + if debug: print("parse() - starting for expression: {}".format(expression)) if now is None: + if debug: print("parse() - Now is None, setting now to utcnow()") now = arrow.utcnow() - if debug: print("Orig Expression: {0}".format(expression)) + if debug: print("parse() - Orig Expression: {0}".format(expression)) math = '' time = '' if 'UTC' not in tz: - if debug: print("will now convert tz to {0}".format(tz)) + if debug: print("parse() - will now convert tz to {0}".format(tz)) now = now.to(tz) if expression == 'now': - if debug: print("Now, no dm: {0}".format(now)) + if debug: print("parse() - Now, no dm: {0}".format(now)) if type: return getattr(now, type) else: return now elif re.match('\d{10,}', str(expression)): - if debug: print('found an epoch timestamp') + if debug: print('parse() - found an epoch timestamp') if len(str(expression)) == 13: raise DateMathException('Unable to parse epoch timestamps in millis, please convert to the nearest second to continue - i.e. 1451610061 / 1000') ts = arrow.get(int(expression)) @@ -118,7 +122,7 @@ def parse(expression, now=None, tz='UTC', type=None, roundDown=True): ''' parse our standard "now+1d" kind of queries ''' math = expression[3:] time = now - if debug: print('now expression: {0}'.format(now)) + if debug: print('parse() - now expression: {0}'.format(now)) else: ''' parse out datemath with date, ex "2015-10-20\|\|+1d" ''' if '\|\|' in expression: @@ -137,6 +141,7 @@ def parse(expression, now=None, tz='UTC', type=None, roundDown=True): if not math or math == '': rettime = time + rettime = evaluate(math, time, tz, roundDown) if type: @@ -144,17 +149,32 @@ def parse(expression, now=None, tz='UTC', type=None, roundDown=True): else: return rettime - def parseTime(timestamp, timezone='UTC'): ''' - parses a date/time stamp and returns and arrow object + parses a datetime string and returns and arrow object ''' if timestamp and len(timestamp) >= 4: ts = arrow.get(timestamp) - ts = ts.replace(tzinfo=timezone) - return ts + if debug: print("parseTime() - ts = {} :: vars :: {}".format(ts, vars(ts))) + if debug: print("parseTime() - ts timezone = {}".format(ts.tzinfo)) + if debug: print("parseTime() - tzinfo type = {}".format(type(ts.tzinfo))) + if debug: print("parseTime() - timezone that came in = {}".format(timezone)) + + if ts.tzinfo: + import dateutil + if isinstance(ts.tzinfo, dateutil.tz.tz.tzoffset): + # this means our TZ probably came in via our datetime string + # then lets set our tz to whatever tzoffset is + ts = ts.replace(tzinfo=ts.tzinfo) + elif isinstance(ts.tzinfo, dateutil.tz.tz.tzutc): + # otherwise if we are utc, then lets just set it to be as such + ts = ts.replace(tzinfo=timezone) + else: + # otherwise lets just ensure its set to whatever timezone came in + ts = ts.replace(tzinfo=timezone) - + return ts + def roundDate(now, unit, tz='UTC', roundDown=True): ''' rounds our date object @@ -163,7 +183,7 @@ def roundDate(now, unit, tz='UTC', roundDown=True): now = now.floor(unit) else: now = now.ceil(unit) - if debug: print("roundDate Now: {0}".format(now)) + if debug: print("roundDate() Now: {0}".format(now)) return now def calculate(now, offsetval, unit): @@ -175,7 +195,7 @@ def calculate(now, offsetval, unit): offsetval = int(offsetval) try: now = now.shift(**{unit: offsetval}) - if debug: print("Calculate called: now: {}, offsetval: {}, offsetval-type: {}, unit: {}".format(now, offsetval, type(offsetval), unit)) + if debug: print("calculate() called: now: {}, offsetval: {}, offsetval-type: {}, unit: {}".format(now, offsetval, type(offsetval), unit)) return now except Exception as e: raise DateMathException('Unable to calculate date: now: {0}, offsetvalue: {1}, unit: {2} - reason: {3}'.format(now,offsetval,unit,e)) @@ -184,8 +204,8 @@ def evaluate(expression, now, timeZone='UTC', roundDown=True): ''' evaluates our datemath style expression ''' - if debug: print('Expression: {0}'.format(expression)) - if debug: print('Now: {0}'.format(now)) + if debug: print('evaluate() - Expression: {0}'.format(expression)) + if debug: print('evaluate() - Now: {0}'.format(now)) val = 0 i = 0 while i < len(expression): @@ -221,7 +241,7 @@ def evaluate(expression, now, timeZone='UTC', roundDown=True): raise DateMathException(''''{}' is not a valid timeunit for expression: '{}' '''.format(char, expression)) i += 1 - if debug: print("Fin: {0}".format(now)) + if debug: print("evaluate() - Finished: {0}".format(now)) if debug: print('\n\n') return now	Timezone in string doesn't work/isn't blocked Hi, I tried some examples with this library and it accepts timezones in the string but seems to ignore them: ```python datemath("2016-01-01T16:20:00.5+12:00") datetime.datetime(2016, 1, 1, 16, 20, 0, 500000, tzinfo=tzutc()) ``` If you then try to do math on it it fails: ```python datemath("2016-01-01T16:20:00.5+12:00\|\|+1d") datemath.helpers.DateMathException: Unable to calculate date: now: 2016-01-01T16:20:00.500000+00:00, offsetvalue: 1.0, unit: days - reason: 'Arrow' object has no attribute 'shift' ``` Can you please check if timezones in the string are supported and if not, block them? datemath version: 1.5.1 python version: 3.7.1 PS: I'm aware of the tz parameter but was curious if it's supported inside the string		2020-10-01T20:17:31	[]	[]
pybamm-team/PyBaMM	pybamm-team__PyBaMM-4641	c471f0db7d841033d9edbccd54d19b99d8c70a15	diff --git a/src/pybamm/experiment/step/base_step.py b/src/pybamm/experiment/step/base_step.py index a7dfa9c9ba..8dd829935e 100644 --- a/src/pybamm/experiment/step/base_step.py +++ b/src/pybamm/experiment/step/base_step.py @@ -1,6 +1,4 @@ -# -# Private classes and functions for experiment steps -# +from __future__ import annotations import pybamm import numpy as np from datetime import datetime @@ -492,6 +490,10 @@ def set_up(self, new_model, new_parameter_values): } +def get_unit_from(a_string: str) -> str: + return a_string.lstrip("0123456789.-eE ") + + def _convert_time_to_seconds(time_and_units): """Convert a time in seconds, minutes or hours to a time in seconds""" if time_and_units is None: @@ -501,11 +503,10 @@ def _convert_time_to_seconds(time_and_units): if isinstance(time_and_units, numbers.Number): if time_and_units <= 0: raise ValueError("time must be positive") - else: - return time_and_units + return time_and_units # Split number and units - units = time_and_units.lstrip("0123456789.- ") + units = get_unit_from(time_and_units) time = time_and_units[: -len(units)] if units in ["second", "seconds", "s", "sec"]: time_in_seconds = float(time) @@ -528,7 +529,7 @@ def _convert_temperature_to_kelvin(temperature_and_units): return temperature_and_units # Split number and units - units = temperature_and_units.lstrip("0123456789.- ") + units = get_unit_from(temperature_and_units) temperature = temperature_and_units[: -len(units)] if units in ["K"]: temperature_in_kelvin = float(temperature) @@ -547,7 +548,7 @@ def _convert_electric(value_string): value = 1 / float(value_string[2:]) else: # All other cases e.g. 4 A, 2.5 V, 1.5 Ohm - unit = value_string.lstrip("0123456789.- ") + unit = get_unit_from(value_string) value = float(value_string[: -len(unit)]) # Catch milli- prefix if unit.startswith("m"):	[Bug]: pybam.Experiment parse units error ### PyBaMM Version 24.9.0 ### Python Version python 3.10 ### Describe the bug in https://github.com/pybamm-team/PyBaMM/blob/develop/src/pybamm/experiment/step/base_step.py function def _convert_electric(value_string), if the value_string float value contains 'e' or 'E', the units incorrect parsed: Python 3.12.3 (main, Nov 6 2024, 18:32:19) [GCC 13.2.0] on linux Type "help", "copyright", "credits" or "license" for more information. >>> value_string = "-1.1729285959021311e-05 W" >>> unit = value_string.lstrip("0123456789.- ") >>> unit 'e-05 W' >>> unit = value_string.lstrip("0123456789eE.- ") >>> unit 'W' >>> for python3, according to https://docs.python.org/3/library/functions.html#float, should accept 'e' or 'E' as well ### Steps to Reproduce 1. value_string = "-1.1729285959021311e-05 W" 2. unit = value_string.lstrip("0123456789.- ") get incorrect unit 'e-05 W', it should be 'W' ### Relevant log output ```shell ```	Thanks @chuckliu1979 I will take a look at this	2024-12-03T02:00:53	[]	[]
pybamm-team/PyBaMM	pybamm-team__PyBaMM-4622	0a46a3b7b833c86e3d50f3a9358022551dec4b1f	diff --git a/CHANGELOG.md b/CHANGELOG.md index ff376c2edc..5748b9f227 100644 --- a/CHANGELOG.md +++ b/CHANGELOG.md @@ -3,6 +3,10 @@ ## Features - Added two more submodels (options) for the SEI: Lars von Kolzenberg (2020) model and Tunneling Limit model ([#4394](https://github.com/pybamm-team/PyBaMM/pull/4394)) +## Bug fixes + +- Reverted modifications to quickplot from [#4529](https://github.com/pybamm-team/PyBaMM/pull/4529) which caused issues with the plots displaying correct variable names. ([#4622](https://github.com/pybamm-team/PyBaMM/pull/4622)) + ## Breaking changes - Double-layer SEI models have been removed (with the corresponding parameters). All models assume now a single SEI layer. ([#4470](https://github.com/pybamm-team/PyBaMM/pull/4470)) diff --git a/src/pybamm/plotting/quick_plot.py b/src/pybamm/plotting/quick_plot.py index e3ad40855e..babfd2e761 100644 --- a/src/pybamm/plotting/quick_plot.py +++ b/src/pybamm/plotting/quick_plot.py @@ -1,7 +1,6 @@ # # Class for quick plotting of variables from models # -from __future__ import annotations import os import numpy as np import pybamm @@ -480,13 +479,13 @@ def reset_axis(self): ): # pragma: no cover raise ValueError(f"Axis limits cannot be NaN for variables '{key}'") - def plot(self, t: float \| list[float], dynamic: bool = False): + def plot(self, t, dynamic=False): """Produces a quick plot with the internal states at time t. Parameters ---------- - t : float or list of float - Dimensional time (in 'time_units') at which to plot. Can be a single time or a list of times. + t : float + Dimensional time (in 'time_units') at which to plot. dynamic : bool, optional Determine whether to allocate space for a slider at the bottom of the plot when generating a dynamic plot. If True, creates a dynamic plot with a slider. @@ -494,10 +493,10 @@ def plot(self, t: float \| list[float], dynamic: bool = False): plt = import_optional_dependency("matplotlib.pyplot") gridspec = import_optional_dependency("matplotlib.gridspec") + cm = import_optional_dependency("matplotlib", "cm") + colors = import_optional_dependency("matplotlib", "colors") - if not isinstance(t, list): - t = [t] - + t_in_seconds = t * self.time_scaling_factor self.fig = plt.figure(figsize=self.figsize) self.gridspec = gridspec.GridSpec(self.n_rows, self.n_cols) @@ -509,11 +508,6 @@ def plot(self, t: float \| list[float], dynamic: bool = False): # initialize empty handles, to be created only if the appropriate plots are made solution_handles = [] - # Generate distinct colors for each time point - time_colors = plt.cm.coolwarm( - np.linspace(0, 1, len(t)) - ) # Use a colormap for distinct colors - for k, (key, variable_lists) in enumerate(self.variables.items()): ax = self.fig.add_subplot(self.gridspec[k]) self.axes.add(key, ax) @@ -524,17 +518,19 @@ def plot(self, t: float \| list[float], dynamic: bool = False): ax.xaxis.set_major_locator(plt.MaxNLocator(3)) self.plots[key] = defaultdict(dict) variable_handles = [] - + # Set labels for the first subplot only (avoid repetition) if variable_lists[0][0].dimensions == 0: - # 0D plot: plot as a function of time, indicating multiple times with lines + # 0D plot: plot as a function of time, indicating time t with a line ax.set_xlabel(f"Time [{self.time_unit}]") for i, variable_list in enumerate(variable_lists): for j, variable in enumerate(variable_list): - linestyle = ( - self.linestyles[i] - if len(variable_list) == 1 - else self.linestyles[j] - ) + if len(variable_list) == 1: + # single variable -> use linestyle to differentiate model + linestyle = self.linestyles[i] + else: + # multiple variables -> use linestyle to differentiate + # variables (color differentiates models) + linestyle = self.linestyles[j] full_t = self.ts_seconds[i] (self.plots[key][i][j],) = ax.plot( full_t / self.time_scaling_factor, @@ -546,104 +542,128 @@ def plot(self, t: float \| list[float], dynamic: bool = False): solution_handles.append(self.plots[key][i][0]) y_min, y_max = ax.get_ylim() ax.set_ylim(y_min, y_max) - - # Add vertical lines for each time in the list, using different colors for each time - for idx, t_single in enumerate(t): - t_in_seconds = t_single * self.time_scaling_factor - (self.time_lines[key],) = ax.plot( - [ - t_in_seconds / self.time_scaling_factor, - t_in_seconds / self.time_scaling_factor, - ], - [y_min, y_max], - "--", # Dashed lines - lw=1.5, - color=time_colors[idx], # Different color for each time - label=f"t = {t_single:.2f} {self.time_unit}", - ) - ax.legend() - + (self.time_lines[key],) = ax.plot( + [ + t_in_seconds / self.time_scaling_factor, + t_in_seconds / self.time_scaling_factor, + ], + [y_min, y_max], + "k--", + lw=1.5, + ) elif variable_lists[0][0].dimensions == 1: - # 1D plot: plot as a function of x at different times + # 1D plot: plot as a function of x at time t + # Read dictionary of spatial variables spatial_vars = self.spatial_variable_dict[key] spatial_var_name = next(iter(spatial_vars.keys())) - ax.set_xlabel(f"{spatial_var_name} [{self.spatial_unit}]") - - for idx, t_single in enumerate(t): - t_in_seconds = t_single * self.time_scaling_factor - - for i, variable_list in enumerate(variable_lists): - for j, variable in enumerate(variable_list): - linestyle = ( - self.linestyles[i] - if len(variable_list) == 1 - else self.linestyles[j] - ) - (self.plots[key][i][j],) = ax.plot( - self.first_spatial_variable[key], - variable(t_in_seconds, spatial_vars), - color=time_colors[idx], # Different color for each time - linestyle=linestyle, - label=f"t = {t_single:.2f} {self.time_unit}", # Add time label - zorder=10, - ) - variable_handles.append(self.plots[key][0][j]) - solution_handles.append(self.plots[key][i][0]) - - # Add a legend to indicate which plot corresponds to which time - ax.legend() - + ax.set_xlabel( + f"{spatial_var_name} [{self.spatial_unit}]", + ) + for i, variable_list in enumerate(variable_lists): + for j, variable in enumerate(variable_list): + if len(variable_list) == 1: + # single variable -> use linestyle to differentiate model + linestyle = self.linestyles[i] + else: + # multiple variables -> use linestyle to differentiate + # variables (color differentiates models) + linestyle = self.linestyles[j] + (self.plots[key][i][j],) = ax.plot( + self.first_spatial_variable[key], + variable(t_in_seconds, spatial_vars), + color=self.colors[i], + linestyle=linestyle, + zorder=10, + ) + variable_handles.append(self.plots[key][0][j]) + solution_handles.append(self.plots[key][i][0]) + # add lines for boundaries between subdomains + for boundary in variable_lists[0][0].internal_boundaries: + boundary_scaled = boundary * self.spatial_factor + ax.axvline(boundary_scaled, color="0.5", lw=1, zorder=0) elif variable_lists[0][0].dimensions == 2: - # 2D plot: superimpose plots at different times + # Read dictionary of spatial variables spatial_vars = self.spatial_variable_dict[key] + # there can only be one entry in the variable list variable = variable_lists[0][0] - - for t_single in t: - t_in_seconds = t_single * self.time_scaling_factor + # different order based on whether the domains are x-r, x-z or y-z, etc + if self.x_first_and_y_second[key] is False: + x_name = list(spatial_vars.keys())[1][0] + y_name = next(iter(spatial_vars.keys()))[0] + x = self.second_spatial_variable[key] + y = self.first_spatial_variable[key] + var = variable(t_in_seconds, spatial_vars) + else: + x_name = next(iter(spatial_vars.keys()))[0] + y_name = list(spatial_vars.keys())[1][0] x = self.first_spatial_variable[key] y = self.second_spatial_variable[key] var = variable(t_in_seconds, spatial_vars).T - - ax.set_xlabel( - f"{next(iter(spatial_vars.keys()))[0]} [{self.spatial_unit}]" - ) - ax.set_ylabel( - f"{list(spatial_vars.keys())[1][0]} [{self.spatial_unit}]" + ax.set_xlabel(f"{x_name} [{self.spatial_unit}]") + ax.set_ylabel(f"{y_name} [{self.spatial_unit}]") + vmin, vmax = self.variable_limits[key] + # store the plot and the var data (for testing) as cant access + # z data from QuadMesh or QuadContourSet object + if self.is_y_z[key] is True: + self.plots[key][0][0] = ax.pcolormesh( + x, + y, + var, + vmin=vmin, + vmax=vmax, + shading=self.shading, ) - vmin, vmax = self.variable_limits[key] - - # Use contourf and colorbars to represent the values - contour_plot = ax.contourf( - x, y, var, levels=100, vmin=vmin, vmax=vmax, cmap="coolwarm" + else: + self.plots[key][0][0] = ax.contourf( + x, y, var, levels=100, vmin=vmin, vmax=vmax ) - self.plots[key][0][0] = contour_plot - self.colorbars[key] = self.fig.colorbar(contour_plot, ax=ax) - - self.plots[key][0][1] = var - - ax.set_title(f"t = {t_single:.2f} {self.time_unit}") + self.plots[key][0][1] = var + if vmin is None and vmax is None: + vmin = ax_min(var) + vmax = ax_max(var) + self.colorbars[key] = self.fig.colorbar( + cm.ScalarMappable(colors.Normalize(vmin=vmin, vmax=vmax)), + ax=ax, + ) + # Set either y label or legend entries + if len(key) == 1: + title = split_long_string(key[0]) + ax.set_title(title, fontsize="medium") + else: + ax.legend( + variable_handles, + [split_long_string(s, 6) for s in key], + bbox_to_anchor=(0.5, 1), + loc="lower center", + ) - # Set global legend if there are multiple models + # Set global legend if len(self.labels) > 1: fig_legend = self.fig.legend( solution_handles, self.labels, loc="lower right" ) + # Get the position of the top of the legend in relative figure units + # There may be a better way ... + try: + legend_top_inches = fig_legend.get_window_extent( + renderer=self.fig.canvas.get_renderer() + ).get_points()[1, 1] + fig_height_inches = (self.fig.get_size_inches() * self.fig.dpi)[1] + legend_top = legend_top_inches / fig_height_inches + except AttributeError: # pragma: no cover + # When testing the examples we set the matplotlib backend to "Template" + # which means that the above code doesn't work. Since this is just for + # that particular test we can just skip it + legend_top = 0 else: - fig_legend = None + legend_top = 0 - # Fix layout for sliders if dynamic + # Fix layout if dynamic: slider_top = 0.05 else: slider_top = 0 - bottom = max( - fig_legend.get_window_extent( - renderer=self.fig.canvas.get_renderer() - ).get_points()[1, 1] - if fig_legend - else 0, - slider_top, - ) + bottom = max(legend_top, slider_top) self.gridspec.tight_layout(self.fig, rect=[0, bottom, 1, 1]) def dynamic_plot(self, show_plot=True, step=None):	[Bug]: Variables names missing from plot ### PyBaMM Version 24.11.1 ### Python Version 3.11 ### Describe the bug Dynamic plot has no variable labels and shows the legend entries as `t=0.00h` for all plots ![Image](https://github.com/user-attachments/assets/4ab94412-5ffd-4298-8c7a-fa93a5c76e35) ### Steps to Reproduce ```python3 import pybamm model = pybamm.lithium_ion.SPM() sim = pybamm.Simulation(model) sim.solve([0, 3600]) sim.plot() ``` ### Relevant log output ```shell ```	From https://github.com/pybamm-team/PyBaMM/pull/4529 @rtimms Sorry missed in the review cc: @medha-14, this might be worth a look! Ah, I must have missed this , will get to fixing this in a few days. Another issue that I believe is related: when plotting 2D variables the colourbar range is wrong. Edit: digging a bit, I suspect the issue for the colourmaps is also to do with the plotting multiple times in the same plot. For 2D this should not be possible so the fix there only requires reverting that code to the previous version and throwing an error if multiple times are passed for a 2D variable. ## 24.9 ![Image](https://github.com/user-attachments/assets/31be8b94-3128-4fdd-8886-a936f8dd036b) ## 24.11 ![Image](https://github.com/user-attachments/assets/d807d9f8-2ba9-4751-8396-13e027d1c62f) Digging a bit deeper, there also some inconsistencies when plotting multiple solutions in one plot. Given that there are multiple issues to fix (basically the issue is a lot harder than I anticipated in #4271 😅) and plotting is quite a major feature, I suggest reverting #4529 and making a minor release. We can then address this for the next major release. ![Image](https://github.com/user-attachments/assets/9aa0319c-f71b-4489-add9-3ad444a46f5d)	2024-11-27T15:45:19	[]	[]
pybamm-team/PyBaMM	pybamm-team__PyBaMM-4591	e17b549fb20f036cf6da474329f7c58ca11365fe	diff --git a/.github/workflows/benchmark_on_push.yml b/.github/workflows/benchmark_on_push.yml index 2883eb5f26..3aa9fce9c0 100644 --- a/.github/workflows/benchmark_on_push.yml +++ b/.github/workflows/benchmark_on_push.yml @@ -8,6 +8,9 @@ concurrency: group: ${{ github.workflow }}-${{ github.ref }} cancel-in-progress: true +env: + PYBAMM_DISABLE_TELEMETRY: "true" + jobs: benchmarks: runs-on: ubuntu-latest diff --git a/.github/workflows/periodic_benchmarks.yml b/.github/workflows/periodic_benchmarks.yml index 14f97c55d8..30603a0ea9 100644 --- a/.github/workflows/periodic_benchmarks.yml +++ b/.github/workflows/periodic_benchmarks.yml @@ -15,6 +15,9 @@ on: # workflow manually workflow_dispatch: +env: + PYBAMM_DISABLE_TELEMETRY: "true" + jobs: benchmarks: runs-on: ubuntu-latest diff --git a/.github/workflows/publish_pypi.yml b/.github/workflows/publish_pypi.yml index 89c9b057c1..944c900e9a 100644 --- a/.github/workflows/publish_pypi.yml +++ b/.github/workflows/publish_pypi.yml @@ -18,6 +18,7 @@ on: # Set options available for all jobs that use cibuildwheel env: + PYBAMM_DISABLE_TELEMETRY: "true" # Increase pip debugging output, equivalent to `pip -vv` CIBW_BUILD_VERBOSITY: 2 # Disable build isolation to allow pre-installing build-time dependencies. @@ -75,6 +76,7 @@ jobs: run: pipx run cibuildwheel --output-dir wheelhouse env: CIBW_ENVIRONMENT: > + PYBAMM_DISABLE_TELEMETRY="true" PYBAMM_USE_VCPKG=ON VCPKG_ROOT_DIR=C:\vcpkg VCPKG_DEFAULT_TRIPLET=x64-windows-static-md @@ -116,6 +118,8 @@ jobs: - name: Build wheels on Linux run: pipx run cibuildwheel --output-dir wheelhouse env: + CIBW_ENVIRONMENT: > + PYBAMM_DISABLE_TELEMETRY="true" CIBW_ARCHS_LINUX: x86_64 CIBW_BEFORE_ALL_LINUX: > yum -y install openblas-devel lapack-devel && @@ -242,6 +246,8 @@ jobs: python scripts/install_KLU_Sundials.py python -m cibuildwheel --output-dir wheelhouse env: + CIBW_ENVIRONMENT: > + PYBAMM_DISABLE_TELEMETRY="true" # 10.13 for Intel (macos-13), 11.0 for Apple Silicon (macos-14 and macos-latest) MACOSX_DEPLOYMENT_TARGET: ${{ matrix.os == 'macos-14' && '11.0' \|\| '10.13' }} CIBW_ARCHS_MACOS: auto diff --git a/.github/workflows/run_benchmarks_over_history.yml b/.github/workflows/run_benchmarks_over_history.yml index d66704a635..28960fb4da 100644 --- a/.github/workflows/run_benchmarks_over_history.yml +++ b/.github/workflows/run_benchmarks_over_history.yml @@ -18,6 +18,10 @@ on: ncommits: description: "Number of commits to benchmark between commit_start and commit_end" default: "100" + +env: + PYBAMM_DISABLE_TELEMETRY: "true" + jobs: benchmarks: runs-on: ubuntu-latest diff --git a/.github/workflows/work_precision_sets.yml b/.github/workflows/work_precision_sets.yml index c167bf5747..5810956786 100644 --- a/.github/workflows/work_precision_sets.yml +++ b/.github/workflows/work_precision_sets.yml @@ -5,6 +5,9 @@ on: types: [published] workflow_dispatch: +env: + PYBAMM_DISABLE_TELEMETRY: "true" + jobs: benchmarks_on_release: if: github.repository_owner == 'pybamm-team' diff --git a/noxfile.py b/noxfile.py index d65812b8ed..5ab32f463f 100644 --- a/noxfile.py +++ b/noxfile.py @@ -49,7 +49,6 @@ def set_iree_state(): "IREE_INDEX_URL": os.getenv( "IREE_INDEX_URL", "https://iree.dev/pip-release-links.html" ), - "PYBAMM_DISABLE_TELEMETRY": "true", } VENV_DIR = Path("./venv").resolve() diff --git a/src/pybamm/config.py b/src/pybamm/config.py index ba7171e5d2..0bd5c96eb6 100644 --- a/src/pybamm/config.py +++ b/src/pybamm/config.py @@ -8,6 +8,18 @@ import time +def check_env_opt_out(): + return os.getenv("PYBAMM_DISABLE_TELEMETRY", "false").lower() != "false" + + +def check_opt_out(): + opt_out = check_env_opt_out() + config = pybamm.config.read() + if config: + opt_out = opt_out or not config["enable_telemetry"] + return opt_out + + def is_running_tests(): # pragma: no cover """ Detect if the code is being run as part of a test suite or building docs with Sphinx. @@ -21,20 +33,18 @@ def is_running_tests(): # pragma: no cover ): return True - # Check for GitHub Actions environment variable - if "GITHUB_ACTIONS" in os.environ: - return True - # Check for other common CI environment variables - ci_env_vars = ["CI", "TRAVIS", "CIRCLECI", "JENKINS_URL", "GITLAB_CI"] + ci_env_vars = [ + "GITHUB_ACTIONS", + "CI", + "TRAVIS", + "CIRCLECI", + "JENKINS_URL", + "GITLAB_CI", + ] if any(var in os.environ for var in ci_env_vars): return True - # Check for common test runner names in command-line arguments - test_runners = ["pytest", "unittest", "nose", "trial", "nox", "tox"] - if any(runner in arg.lower() for arg in sys.argv for runner in test_runners): - return True - # Check if building docs with Sphinx if any(mod == "sphinx" or mod.startswith("sphinx.") for mod in sys.modules): print( @@ -110,7 +120,7 @@ def get_input(): # pragma: no cover def generate(): - if is_running_tests(): + if is_running_tests() or check_opt_out(): return # Check if the config file already exists diff --git a/src/pybamm/telemetry.py b/src/pybamm/telemetry.py index b3319b9237..2dad19f814 100644 --- a/src/pybamm/telemetry.py +++ b/src/pybamm/telemetry.py @@ -1,38 +1,45 @@ from posthog import Posthog -import os import pybamm import sys -_posthog = Posthog( - # this is the public, write only API key, so it's ok to include it here - project_api_key="phc_bLZKBW03XjgiRhbWnPsnKPr0iw0z03fA6ZZYjxgW7ej", - host="https://us.i.posthog.com", -) -_posthog.log.setLevel("CRITICAL") +class MockTelemetry: + def __init__(self): + self.disabled = True + @staticmethod + def capture(**kwargs): # pragma: no cover + pass -def disable(): - _posthog.disabled = True + +if pybamm.config.check_opt_out(): + _posthog = MockTelemetry() +else: # pragma: no cover + _posthog = Posthog( + # this is the public, write only API key, so it's ok to include it here + project_api_key="phc_bLZKBW03XjgiRhbWnPsnKPr0iw0z03fA6ZZYjxgW7ej", + host="https://us.i.posthog.com", + ) + _posthog.log.setLevel("CRITICAL") -_opt_out = os.getenv("PYBAMM_DISABLE_TELEMETRY", "false").lower() -if _opt_out != "false": # pragma: no cover - disable() +def disable(): + _posthog.disabled = True def capture(event): # pragma: no cover - # don't capture events in automated testing if pybamm.config.is_running_tests() or _posthog.disabled: return - properties = { - "python_version": sys.version, - "pybamm_version": pybamm.__version__, - } + if pybamm.config.check_opt_out(): + disable() + return config = pybamm.config.read() if config: - if config["enable_telemetry"]: - user_id = config["uuid"] - _posthog.capture(user_id, event, properties=properties) + properties = { + "python_version": sys.version, + "pybamm_version": pybamm.__version__, + } + user_id = config["uuid"] + _posthog.capture(distinct_id=user_id, event=event, properties=properties)	Fortnightly build for wheels failed The build is failing with the following logs - https://github.com/pybamm-team/PyBaMM/actions/runs/11854441290		2024-11-15T22:02:44	[]	[]
pybamm-team/PyBaMM	pybamm-team__PyBaMM-4562	bf8f6238cfbd6fc6dadec8a527307d14533dc9c1	diff --git a/CHANGELOG.md b/CHANGELOG.md index bf376c1a13..eee7de05d7 100644 --- a/CHANGELOG.md +++ b/CHANGELOG.md @@ -2,6 +2,7 @@ ## Features +- Added `CoupledVariable` which provides a placeholder variable whose equation can be elsewhere in the model. ([#4556](https://github.com/pybamm-team/PyBaMM/pull/4556)) - Adds support to `pybamm.Experiment` for the `output_variables` option in the `IDAKLUSolver`. ([#4534](https://github.com/pybamm-team/PyBaMM/pull/4534)) - Adds an option "voltage as a state" that can be "false" (default) or "true". If "true" adds an explicit algebraic equation for the voltage. ([#4507](https://github.com/pybamm-team/PyBaMM/pull/4507)) - Improved `QuickPlot` accuracy for simulations with Hermite interpolation. ([#4483](https://github.com/pybamm-team/PyBaMM/pull/4483)) @@ -21,6 +22,8 @@ - Removed the `start_step_offset` setting and disabled minimum `dt` warnings for drive cycles with the (`IDAKLUSolver`). ([#4416](https://github.com/pybamm-team/PyBaMM/pull/4416)) ## Bug Fixes +- Added error for binary operators on two concatenations with different numbers of children. Previously, the extra children were dropped. Also fixed bug where Q_rxn was dropped from the total heating term in half-cell models. ([#4562](https://github.com/pybamm-team/PyBaMM/pull/4562)) +- Fixed bug where Q_rxn was set to 0 for the negative electrode in half-cell models. ([#4557](https://github.com/pybamm-team/PyBaMM/pull/4557)) - Fixed bug in post-processing solutions with infeasible experiments using the (`IDAKLUSolver`). ([#4541](https://github.com/pybamm-team/PyBaMM/pull/4541)) - Disabled IREE on MacOS due to compatibility issues and added the CasADI path to the environment to resolve issues on MacOS and Linux. Windows diff --git a/src/pybamm/expression_tree/binary_operators.py b/src/pybamm/expression_tree/binary_operators.py index 42f2f74e24..be3df653ad 100644 --- a/src/pybamm/expression_tree/binary_operators.py +++ b/src/pybamm/expression_tree/binary_operators.py @@ -856,12 +856,17 @@ def _simplified_binary_broadcast_concatenation( elif isinstance(right, pybamm.Concatenation) and not isinstance( right, pybamm.ConcatenationVariable ): - return left.create_copy( - [ - operator(left_child, right_child) - for left_child, right_child in zip(left.orphans, right.orphans) - ] - ) + if len(left.orphans) == len(right.orphans): + return left.create_copy( + [ + operator(left_child, right_child) + for left_child, right_child in zip(left.orphans, right.orphans) + ] + ) + else: + raise AssertionError( + "Concatenations must have the same number of children" + ) if isinstance(right, pybamm.Concatenation) and not isinstance( right, pybamm.ConcatenationVariable ): diff --git a/src/pybamm/models/full_battery_models/base_battery_model.py b/src/pybamm/models/full_battery_models/base_battery_model.py index a445f47d19..153c447ed3 100644 --- a/src/pybamm/models/full_battery_models/base_battery_model.py +++ b/src/pybamm/models/full_battery_models/base_battery_model.py @@ -989,7 +989,6 @@ def options(self, extra_options): raise pybamm.OptionError( f"must use surface formulation to solve {self!s} with hydrolysis" ) - self._options = options def set_standard_output_variables(self): diff --git a/src/pybamm/models/submodels/thermal/base_thermal.py b/src/pybamm/models/submodels/thermal/base_thermal.py index 384cb1723a..d5a39435fd 100644 --- a/src/pybamm/models/submodels/thermal/base_thermal.py +++ b/src/pybamm/models/submodels/thermal/base_thermal.py @@ -88,6 +88,12 @@ def _get_standard_coupled_variables(self, variables): # TODO: change full stefan-maxwell conductivity so that i_e is always # a Concatenation i_e = variables["Electrolyte current density [A.m-2]"] + # Special case for half cell -- i_e has to be a concatenation for this to work due to a mismatch with Q_ohm, so we make a new i_e which is a concatenation. + if (not isinstance(i_e, pybamm.Concatenation)) and ( + self.options.electrode_types["negative"] == "planar" + ): + i_e = self._get_i_e_for_half_cell_thermal(variables) + phi_e = variables["Electrolyte potential [V]"] if isinstance(i_e, pybamm.Concatenation): # compute by domain if possible @@ -411,3 +417,23 @@ def _yz_average(self, var): return pybamm.z_average(var) elif self.options["dimensionality"] == 2: return pybamm.yz_average(var) + + def _get_i_e_for_half_cell_thermal(self, variables): + c_e_s = variables["Separator electrolyte concentration [mol.m-3]"] + c_e_p = variables["Positive electrolyte concentration [mol.m-3]"] + grad_phi_e_s = variables["Gradient of separator electrolyte potential [V.m-1]"] + grad_phi_e_p = variables["Gradient of positive electrolyte potential [V.m-1]"] + grad_c_e_s = pybamm.grad(c_e_s) + grad_c_e_p = pybamm.grad(c_e_p) + T_s = variables["Separator temperature [K]"] + T_p = variables["Positive electrode temperature [K]"] + tor_s = variables["Separator electrolyte transport efficiency"] + tor_p = variables["Positive electrolyte transport efficiency"] + i_e_s = (self.param.kappa_e(c_e_s, T_s) * tor_s) * ( + self.param.chiRT_over_Fc(c_e_s, T_s) * grad_c_e_s - grad_phi_e_s + ) + i_e_p = (self.param.kappa_e(c_e_p, T_p) * tor_p) * ( + self.param.chiRT_over_Fc(c_e_p, T_p) * grad_c_e_p - grad_phi_e_p + ) + i_e = pybamm.concatenation(i_e_s, i_e_p) + return i_e	[Bug]: Binary operators on concatenations assume that each concatenation has the same number of orphans ### PyBaMM Version develop ### Python Version 3.12 ### Describe the bug If a binary operator, such as `+`, is invoked with 2 concatenations (i.e. `isinstance(left, pybamm.Concatenation) and isinstance(right, pybamm.Concatenation) == True`), and the concatenations have a different number of orphans, the "extra orphans" are dropped. This can be easily seen [here](https://github.com/aabills/PyBaMM/blob/8ce4a20090fee6bae2251d54fd28b4b0d49abdfb/src/pybamm/expression_tree/binary_operators.py#L859). This is causing reaction heating to not be present in DFN half-cell models (see comments in #4557) ### Steps to Reproduce no MWE yet, but should be self evident in the code. ### Relevant log output _No response_	I think there are two different options for "solving" this: 1) Raise an error when this situation occurs 2) Pass the identity of the "unmatched orphans". I implemented (1) to check how impactful it will be, and the only test that fails is the specific case I found this bug in: Half-cell DFN models. Therefore, I am of the opinion that we should implement (1) and fix the underlying issue that causes it in the half-cell case. On board with option 1.	2024-11-04T20:02:11	[]	[]
pybamm-team/PyBaMM	pybamm-team__PyBaMM-4456	1390ea306e4200e04e6d63a9eac42333a75961e8	diff --git a/CHANGELOG.md b/CHANGELOG.md index 7070ebaf52..f70aae0272 100644 --- a/CHANGELOG.md +++ b/CHANGELOG.md @@ -10,6 +10,7 @@ ## Optimizations +- Performance refactor of JAX BDF Solver with default Jax method set to `"BDF"`. ([#4456](https://github.com/pybamm-team/PyBaMM/pull/4456)) - Improved performance of initialization and reinitialization of ODEs in the (`IDAKLUSolver`). ([#4453](https://github.com/pybamm-team/PyBaMM/pull/4453)) - Removed the `start_step_offset` setting and disabled minimum `dt` warnings for drive cycles with the (`IDAKLUSolver`). ([#4416](https://github.com/pybamm-team/PyBaMM/pull/4416)) diff --git a/examples/scripts/multiprocess_jax_solver.py b/examples/scripts/multiprocess_jax_solver.py new file mode 100644 index 0000000000..8192256ed1 --- /dev/null +++ b/examples/scripts/multiprocess_jax_solver.py @@ -0,0 +1,57 @@ +import pybamm +import time +import numpy as np + + +# This script provides an example for massively vectorised +# model solves using the JAX BDF solver. First, +# we set up the model and process parameters +model = pybamm.lithium_ion.SPM() +model.convert_to_format = "jax" +model.events = [] # remove events (not supported in jax) +geometry = model.default_geometry +param = pybamm.ParameterValues("Chen2020") +param.update({"Current function [A]": "[input]"}) +param.process_geometry(geometry) +param.process_model(model) + +# Discretise and setup solver +mesh = pybamm.Mesh(geometry, model.default_submesh_types, model.default_var_pts) +disc = pybamm.Discretisation(mesh, model.default_spatial_methods) +disc.process_model(model) +t_eval = np.linspace(0, 3600, 100) +solver = pybamm.JaxSolver(atol=1e-6, rtol=1e-6, method="BDF") + +# Set number of vectorised solves +values = np.linspace(0.01, 1.0, 1000) +inputs = [{"Current function [A]": value} for value in values] + +# Run solve for all inputs, with a just-in-time compilation +# occurring on the first solve. All sequential solves will +# use the compiled code, with a large performance improvement. +start_time = time.time() +sol = solver.solve(model, t_eval, inputs=inputs) +print(f"Time taken: {time.time() - start_time}") # 1.3s + +# Rerun the vectorised solve, showing performance improvement +start_time = time.time() +compiled_sol = solver.solve(model, t_eval, inputs=inputs) +print(f"Compiled time taken: {time.time() - start_time}") # 0.42s + +# Plot one of the solves +plot = pybamm.QuickPlot( + sol[5], + [ + "Negative particle concentration [mol.m-3]", + "Electrolyte concentration [mol.m-3]", + "Positive particle concentration [mol.m-3]", + "Current [A]", + "Negative electrode potential [V]", + "Electrolyte potential [V]", + "Positive electrode potential [V]", + "Voltage [V]", + ], + time_unit="seconds", + spatial_unit="um", +) +plot.dynamic_plot() diff --git a/noxfile.py b/noxfile.py index 6567ed167c..14bafcca47 100644 --- a/noxfile.py +++ b/noxfile.py @@ -222,7 +222,7 @@ def run_scripts(session): # https://bitbucket.org/pybtex-devs/pybtex/issues/169/replace-pkg_resources-with # is fixed session.install("setuptools", silent=False) - session.install("-e", ".[all,dev]", silent=False) + session.install("-e", ".[all,dev,jax]", silent=False) session.run("python", "-m", "pytest", "-m", "scripts") diff --git a/src/pybamm/solvers/base_solver.py b/src/pybamm/solvers/base_solver.py index 6334169cf0..a7fc79fe3a 100644 --- a/src/pybamm/solvers/base_solver.py +++ b/src/pybamm/solvers/base_solver.py @@ -90,6 +90,10 @@ def root_method(self): def supports_parallel_solve(self): return False + @property + def requires_explicit_sensitivities(self): + return True + @root_method.setter def root_method(self, method): if method == "casadi": @@ -141,7 +145,7 @@ def set_up(self, model, inputs=None, t_eval=None, ics_only=False): # see if we need to form the explicit sensitivity equations calculate_sensitivities_explicit = ( - model.calculate_sensitivities and not isinstance(self, pybamm.IDAKLUSolver) + model.calculate_sensitivities and self.requires_explicit_sensitivities ) self._set_up_model_sensitivities_inplace( @@ -494,11 +498,7 @@ def _set_up_model_sensitivities_inplace( # if we have a mass matrix, we need to extend it def extend_mass_matrix(M): M_extend = [M.entries] * (num_parameters + 1) - M_extend_pybamm = pybamm.Matrix(block_diag(M_extend, format="csr")) - return M_extend_pybamm - - model.mass_matrix = extend_mass_matrix(model.mass_matrix) - model.mass_matrix = extend_mass_matrix(model.mass_matrix) + return pybamm.Matrix(block_diag(M_extend, format="csr")) model.mass_matrix = extend_mass_matrix(model.mass_matrix) diff --git a/src/pybamm/solvers/idaklu_solver.py b/src/pybamm/solvers/idaklu_solver.py index 32048d89c0..2b2d852697 100644 --- a/src/pybamm/solvers/idaklu_solver.py +++ b/src/pybamm/solvers/idaklu_solver.py @@ -736,6 +736,10 @@ def _demote_64_to_32(self, x: pybamm.EvaluatorJax): def supports_parallel_solve(self): return True + @property + def requires_explicit_sensitivities(self): + return False + def _integrate(self, model, t_eval, inputs_list=None, t_interp=None): """ Solve a DAE model defined by residuals with initial conditions y0. diff --git a/src/pybamm/solvers/jax_bdf_solver.py b/src/pybamm/solvers/jax_bdf_solver.py index 6f0c62b9a8..3db82ca0da 100644 --- a/src/pybamm/solvers/jax_bdf_solver.py +++ b/src/pybamm/solvers/jax_bdf_solver.py @@ -111,68 +111,63 @@ def fun_bind_inputs(y, t): jac_bind_inputs = jax.jacfwd(fun_bind_inputs, argnums=0) - t0 = t_eval[0] - h0 = t_eval[1] - t0 - + t0, h0 = t_eval[0], t_eval[1] - t_eval[0] stepper = _bdf_init( fun_bind_inputs, jac_bind_inputs, mass, t0, y0, h0, rtol, atol ) - i = 0 y_out = jnp.empty((len(t_eval), len(y0)), dtype=y0.dtype) - init_state = [stepper, t_eval, i, y_out] - def cond_fun(state): - _, t_eval, i, _ = state + _, _, i, _ = state return i < len(t_eval) def body_fun(state): stepper, t_eval, i, y_out = state stepper = _bdf_step(stepper, fun_bind_inputs, jac_bind_inputs) - index = jnp.searchsorted(t_eval, stepper.t) - index = index.astype( - "int" + t_eval.dtype.name[-2:] - ) # Coerce index to correct type + index = jnp.searchsorted(t_eval, stepper.t).astype(jnp.int32) - def for_body(j, y_out): - t = t_eval[j] - y_out = y_out.at[jnp.index_exp[j, :]].set(_bdf_interpolate(stepper, t)) - return y_out + def interpolate_and_update(j, y_out): + y = _bdf_interpolate(stepper, t_eval[j]) + return y_out.at[j].set(y) - y_out = jax.lax.fori_loop(i, index, for_body, y_out) - return [stepper, t_eval, index, y_out] + y_out = jax.lax.fori_loop(i, index, interpolate_and_update, y_out) + return stepper, t_eval, index, y_out + + init_state = (stepper, t_eval, 0, y_out) + _, _, _, y_out = jax.lax.while_loop(cond_fun, body_fun, init_state) - stepper, t_eval, i, y_out = jax.lax.while_loop(cond_fun, body_fun, init_state) return y_out - BDFInternalStates = [ - "t", - "atol", - "rtol", - "M", - "newton_tol", - "order", - "h", - "n_equal_steps", - "D", - "y0", - "scale_y0", - "kappa", - "gamma", - "alpha", - "c", - "error_const", - "J", - "LU", - "U", - "psi", - "n_function_evals", - "n_jacobian_evals", - "n_lu_decompositions", - "n_steps", - "consistent_y0_failed", - ] - BDFState = collections.namedtuple("BDFState", BDFInternalStates) + BDFState = collections.namedtuple( + "BDFState", + [ + "t", + "atol", + "rtol", + "M", + "newton_tol", + "order", + "h", + "n_equal_steps", + "D", + "y0", + "scale_y0", + "kappa", + "gamma", + "alpha", + "c", + "error_const", + "J", + "LU", + "U", + "psi", + "n_function_evals", + "n_jacobian_evals", + "n_lu_decompositions", + "n_steps", + "consistent_y0_failed", + ], + ) jax.tree_util.register_pytree_node( BDFState, lambda xs: (tuple(xs), None), lambda _, xs: BDFState(xs) @@ -211,62 +206,70 @@ def _bdf_init(fun, jac, mass, t0, y0, h0, rtol, atol): absolute tolerance for the solver """ - state = {} - state["t"] = t0 - state["atol"] = atol - state["rtol"] = rtol - state["M"] = mass EPS = jnp.finfo(y0.dtype).eps - state["newton_tol"] = jnp.maximum(10 EPS / rtol, jnp.minimum(0.03, rtol*0.5)) + # Scaling and tolerance initialisation scale_y0 = atol + rtol jnp.abs(y0) + newton_tol = jnp.maximum(10 * EPS / rtol, jnp.minimum(0.03, rtol*0.5)) + y0, not_converged = _select_initial_conditions( - fun, mass, t0, y0, state["newton_tol"], scale_y0 + fun, mass, t0, y0, newton_tol, scale_y0 ) - state["consistent_y0_failed"] = not_converged + # Compute initial function and step size f0 = fun(y0, t0) - order = 1 - state["order"] = order - state["h"] = _select_initial_step(atol, rtol, fun, t0, y0, f0, h0) - state["n_equal_steps"] = 0 + h = _select_initial_step(atol, rtol, fun, t0, y0, f0, h0) + + # Initialise the difference matrix, D D = jnp.empty((MAX_ORDER + 1, len(y0)), dtype=y0.dtype) D = D.at[jnp.index_exp[0, :]].set(y0) - D = D.at[jnp.index_exp[1, :]].set(f0 state["h"]) - state["D"] = D - state["y0"] = y0 - state["scale_y0"] = scale_y0 + D = D.at[jnp.index_exp[1, :]].set(f0 * h) # kappa values for difference orders, taken from Table 1 of [1] kappa = jnp.array([0, -0.1850, -1 / 9, -0.0823, -0.0415, 0]) gamma = jnp.hstack((0, jnp.cumsum(1 / jnp.arange(1, MAX_ORDER + 1)))) alpha = 1.0 / ((1 - kappa) * gamma) - c = state["h"] * alpha[order] + c = h * alpha[1] error_const = kappa * gamma + 1 / jnp.arange(1, MAX_ORDER + 2) - state["kappa"] = kappa - state["gamma"] = gamma - state["alpha"] = alpha - state["c"] = c - state["error_const"] = error_const - + # Jacobian and LU decomp J = jac(y0, t0) - state["J"] = J - - state["LU"] = jax.scipy.linalg.lu_factor(state["M"] - c * J) - - state["U"] = _compute_R(order, 1) - state["psi"] = None - - state["n_function_evals"] = 2 - state["n_jacobian_evals"] = 1 - state["n_lu_decompositions"] = 1 - state["n_steps"] = 0 + LU = jax.scipy.linalg.lu_factor(mass - c * J) + U = _compute_R(1, 1) # Order 1 + + # Create initial BDFState + state = BDFState( + t=t0, + atol=atol, + rtol=rtol, + M=mass, + newton_tol=newton_tol, + consistent_y0_failed=not_converged, + order=1, + h=h, + n_equal_steps=0, + D=D, + y0=y0, + scale_y0=scale_y0, + kappa=kappa, + gamma=gamma, + alpha=alpha, + c=c, + error_const=error_const, + J=J, + LU=LU, + U=U, + psi=None, + n_function_evals=2, + n_jacobian_evals=1, + n_lu_decompositions=1, + n_steps=0, + ) - tuple_state = BDFState([state[k] for k in BDFInternalStates]) - y0, scale_y0 = _predict(tuple_state, D) - psi = _update_psi(tuple_state, D) - return tuple_state._replace(y0=y0, scale_y0=scale_y0, psi=psi) + # Predict initial y0, scale_yo, update state + y0, scale_y0 = _predict(state, D) + psi = _update_psi(state, D) + return state._replace(y0=y0, scale_y0=scale_y0, psi=psi) def _compute_R(order, factor): """ @@ -374,10 +377,8 @@ def _predict(state, D): """ predict forward to new step (eq 2 in [1]) """ - n = len(state.y0) - order = state.order - orders = jnp.repeat(jnp.arange(MAX_ORDER + 1).reshape(-1, 1), n, axis=1) - subD = jnp.where(orders <= order, D, 0) + orders = jnp.arange(MAX_ORDER + 1)[:, None] + subD = jnp.where(orders <= state.order, D, 0) y0 = jnp.sum(subD, axis=0) scale_y0 = state.atol + state.rtol jnp.abs(state.y0) return y0, scale_y0 @@ -397,7 +398,7 @@ def _update_psi(state, D): def _update_difference_for_next_step(state, d): """ - update of difference equations can be done efficiently + Update of difference equations can be done efficiently by reusing d and D. From first equation on page 4 of [1]: @@ -409,34 +410,21 @@ def _update_difference_for_next_step(state, d): Combining these gives the following algorithm """ order = state.order - D = state.D - D = D.at[jnp.index_exp[order + 2]].set(d - D[order + 1]) - D = D.at[jnp.index_exp[order + 1]].set(d) - i = order - while_state = [i, D] - - def while_cond(while_state): - i, _ = while_state - return i >= 0 + D = state.D.at[order + 2].set(d - state.D[order + 1]) + D = D.at[order + 1].set(d) - def while_body(while_state): - i, D = while_state - D = D.at[jnp.index_exp[i]].add(D[i + 1]) - i -= 1 - return [i, D] + def update_D(i, D): + return D.at[order - i].add(D[order - i + 1]) - i, D = jax.lax.while_loop(while_cond, while_body, while_state) - - return D + return jax.lax.fori_loop(0, order + 1, update_D, D) def _update_step_size_and_lu(state, factor): + """ + Update step size and recompute LU decomposition. + """ state = _update_step_size(state, factor) - - # redo lu (c has changed) LU = jax.scipy.linalg.lu_factor(state.M - state.c * state.J) - n_lu_decompositions = state.n_lu_decompositions + 1 - - return state._replace(LU=LU, n_lu_decompositions=n_lu_decompositions) + return state._replace(LU=LU, n_lu_decompositions=state.n_lu_decompositions + 1) def _update_step_size(state, factor): """ @@ -449,7 +437,6 @@ def _update_step_size(state, factor): """ order = state.order h = state.h * factor - n_equal_steps = 0 c = h * state.alpha[order] # update D using equations in section 3.2 of [1] @@ -461,19 +448,14 @@ def _update_step_size(state, factor): RU = jnp.where( jnp.logical_and(I <= order, J <= order), RU, jnp.identity(MAX_ORDER + 1) ) - D = state.D - D = jnp.dot(RU.T, D) - # D = jax.ops.index_update(D, jax.ops.index[:order + 1], - # jnp.dot(RU.T, D[:order + 1])) + D = jnp.dot(RU.T, state.D) - # update psi (D has changed) + # update psi, y0 (D has changed) psi = _update_psi(state, D) - - # update y0 (D has changed) y0, scale_y0 = _predict(state, D) return state._replace( - n_equal_steps=n_equal_steps, + n_equal_steps=0, h=h, c=c, D=D, @@ -484,27 +466,23 @@ def _update_step_size(state, factor): def _update_jacobian(state, jac): """ - we update the jacobian using J(t_{n+1}, y^0_{n+1}) + Update the jacobian using J(t_{n+1}, y^0_{n+1}) following the scipy bdf implementation rather than J(t_n, y_n) as per [1] """ J = jac(state.y0, state.t + state.h) - n_jacobian_evals = state.n_jacobian_evals + 1 LU = jax.scipy.linalg.lu_factor(state.M - state.c * J) - n_lu_decompositions = state.n_lu_decompositions + 1 return state._replace( J=J, - n_jacobian_evals=n_jacobian_evals, + n_jacobian_evals=state.n_jacobian_evals + 1, LU=LU, - n_lu_decompositions=n_lu_decompositions, + n_lu_decompositions=state.n_lu_decompositions + 1, ) def _newton_iteration(state, fun): - tol = state.newton_tol - c = state.c - psi = state.psi + """ + Perform Newton iteration to solve the system. + """ y0 = state.y0 - LU = state.LU - M = state.M scale_y0 = state.scale_y0 t = state.t + state.h d = jnp.zeros(y0.shape, dtype=y0.dtype) @@ -522,17 +500,20 @@ def while_cond(while_state): def while_body(while_state): k, converged, dy_norm_old, d, y, n_function_evals = while_state + f_eval = fun(y, t) n_function_evals += 1 - b = c * f_eval - M @ (psi + d) - dy = jax.scipy.linalg.lu_solve(LU, b) + b = state.c * f_eval - state.M @ (state.psi + d) + dy = jax.scipy.linalg.lu_solve(state.LU, b) dy_norm = jnp.sqrt(jnp.mean((dy / scale_y0) 2)) rate = dy_norm / dy_norm_old # if iteration is not going to converge in NEWTON_MAXITER # (assuming the current rate), then abort pred = rate >= 1 - pred += rate (NEWTON_MAXITER - k) / (1 - rate) * dy_norm > tol + pred += ( + rate ** (NEWTON_MAXITER - k) / (1 - rate) * dy_norm > state.newton_tol + ) pred = dy_norm_old >= 0 k += pred (NEWTON_MAXITER - k - 1) @@ -541,7 +522,7 @@ def while_body(while_state): # if converged then break out of iteration early pred = dy_norm_old >= 0.0 - pred = rate / (1 - rate) dy_norm < tol + pred = rate / (1 - rate) dy_norm < state.newton_tol converged = (dy_norm == 0.0) + pred dy_norm_old = dy_norm @@ -564,7 +545,6 @@ def _prepare_next_step(state, d): def _prepare_next_step_order_change(state, d, y, n_iter): order = state.order - D = _update_difference_for_next_step(state, d) # Note: we are recalculating these from the while loop above, could re-use? @@ -586,7 +566,6 @@ def _prepare_next_step_order_change(state, d, y, n_iter): rms_norm(state.error_const[order + 1] * D[order + 2] / scale_y), jnp.inf, ) - error_norms = jnp.array([error_m_norm, error_norm, error_p_norm]) factors = error_norms ** (-1 / (jnp.arange(3) + order)) @@ -595,111 +574,89 @@ def _prepare_next_step_order_change(state, d, y, n_iter): max_index = jnp.argmax(factors) order += max_index - 1 + # New step size factor factor = jnp.minimum(MAX_FACTOR, safety * factors[max_index]) - - new_state = _update_step_size_and_lu(state._replace(D=D, order=order), factor) - return new_state + new_state = state._replace(D=D, order=order) + return _update_step_size_and_lu(new_state, factor) def _bdf_step(state, fun, jac): - # print('bdf_step', state.t, state.h) - # we will try and use the old jacobian unless convergence of newton iteration - # fails - updated_jacobian = False - # initialise step size and try to make the step, - # iterate, reducing step size until error is in bounds - step_accepted = False - y = jnp.empty_like(state.y0) - d = jnp.empty_like(state.y0) - n_iter = -1 - - # loop until step is accepted - while_state = [state, step_accepted, updated_jacobian, y, d, n_iter] + """ + Perform a BDF step. - def while_cond(while_state): - _, step_accepted, _, _, _, _ = while_state - return step_accepted == False # noqa: E712 + We will try and use the old jacobian unless + convergence of newton iteration fails. + """ - def while_body(while_state): - state, step_accepted, updated_jacobian, y, d, n_iter = while_state + def step_iteration(while_state): + state, updated_jacobian = while_state - # solve BDF equation using y0 as starting point + # Solve BDF equation using Newton iteration converged, n_iter, y, d, state = _newton_iteration(state, fun) - not_converged = converged == False # noqa: E712 - - # newton iteration did not converge, but jacobian has already been - # evaluated so reduce step size by 0.3 (as per [1]) and try again - state = tree_map( - partial(jnp.where, not_converged * updated_jacobian), - _update_step_size_and_lu(state, 0.3), - state, - ) - # if not_converged * updated_jacobian: - # print('not converged, update step size by 0.3') - # if not_converged * (updated_jacobian == False): - # print('not converged, update jacobian') - - # if not converged and jacobian not updated, then update the jacobian and - # try again - (state, updated_jacobian) = tree_map( - partial( - jnp.where, - not_converged * (updated_jacobian == False), # noqa: E712 + # Update Jacobian or reduce step size if not converged + # Evaluated so reduce step size by 0.3 (as per [1]) and try again + state, updated_jacobian = jax.lax.cond( + ~converged, + lambda s, uj: jax.lax.cond( + uj, + lambda s: (_update_step_size_and_lu(s, 0.3), True), + lambda s: (_update_jacobian(s, jac), True), + s, ), - (_update_jacobian(state, jac), True), - (state, False + updated_jacobian), + lambda s, uj: (s, uj), + state, + updated_jacobian, ) safety = 0.9 * (2 * NEWTON_MAXITER + 1) / (2 * NEWTON_MAXITER + n_iter) scale_y = state.atol + state.rtol * jnp.abs(y) + # Calculate error and updated step size factor # combine eq 3, 4 and 6 from [1] to obtain error # Note that error = C_k * h^{k+1} y^{k+1} # and d = D^{k+1} y_{n+1} \approx h^{k+1} y^{k+1} error = state.error_const[state.order] * d - error_norm = rms_norm(error / scale_y) - # calculate optimal step size factor as per eq 2.46 of [2] - factor = jnp.maximum( - MIN_FACTOR, safety * error_norm ** (-1 / (state.order + 1)) + # Calculate optimal step size factor as per eq 2.46 of [2] + factor = jnp.clip( + safety * error_norm ** (-1 / (state.order + 1)), MIN_FACTOR, None ) - # if converged * (error_norm > 1): - # print( - # "converged, but error is too large", - # error_norm, - # factor, - # d, - # scale_y, - # ) - - (state, step_accepted) = tree_map( - partial(jnp.where, converged * (error_norm > 1)), - (_update_step_size_and_lu(state, factor), False), - (state, converged), + # Update step size if error is too large + state = jax.lax.cond( + converged & (error_norm > 1), + lambda s: _update_step_size_and_lu(s, factor), + lambda s: s, + state, ) - return [state, step_accepted, updated_jacobian, y, d, n_iter] - - state, step_accepted, updated_jacobian, y, d, n_iter = jax.lax.while_loop( - while_cond, while_body, while_state + step_accepted = converged & (error_norm <= 1) + return (state, updated_jacobian), (step_accepted, y, d, n_iter) + + # Iterate until step is accepted + (state, _), (_, y, d, n_iter) = jax.lax.while_loop( + lambda carry_and_aux: ~carry_and_aux[1][0], + lambda carry_and_aux: step_iteration(carry_and_aux[0]), + ( + (state, False), + (False, jnp.empty_like(state.y0), jnp.empty_like(state.y0), -1), + ), ) - # take the accepted step + # Update state for the accepted step n_steps = state.n_steps + 1 t = state.t + state.h - - # a change in order is only done after running at order k for k + 1 steps - # (see page 83 of [2]) n_equal_steps = state.n_equal_steps + 1 - state = state._replace(n_equal_steps=n_equal_steps, t=t, n_steps=n_steps) - state = tree_map( - partial(jnp.where, n_equal_steps < state.order + 1), - _prepare_next_step(state, d), - _prepare_next_step_order_change(state, d, y, n_iter), + # Prepare for the next step, potentially changing order + # (see page 83 of [2]) + state = jax.lax.cond( + n_equal_steps < state.order + 1, + lambda s: _prepare_next_step(s, d), + lambda s: _prepare_next_step_order_change(s, d, y, n_iter), + state, ) return state @@ -710,8 +667,6 @@ def _bdf_interpolate(state, t_eval): definition of the interpolating polynomial can be found on page 7 of [1] """ - order = state.order - t = state.t h = state.h D = state.D j = 0 @@ -721,11 +676,11 @@ def _bdf_interpolate(state, t_eval): def while_cond(while_state): j, _, _ = while_state - return j < order + return j < state.order def while_body(while_state): j, time_factor, order_summation = while_state - time_factor = (t_eval - (t - h j)) / (h * (1 + j)) + time_factor = (t_eval - (state.t - h j)) / (h * (1 + j)) order_summation += D[j + 1] * time_factor j += 1 return [j, time_factor, order_summation] @@ -972,13 +927,13 @@ def jax_bdf_integrate(func, y0, t_eval, args, rtol=1e-6, atol=1e-6, mass=None): """ Backward Difference formula (BDF) implicit multistep integrator. The basic algorithm is derived in :footcite:t:`byrne1975polyalgorithm`. This particular implementation - follows that implemented in the Matlab routine ode15s described in - :footcite:t:`shampine1997matlab` and the SciPy implementation - :footcite:t:`Virtanen2020` which features the NDF formulas for improved stability, - with associated differences in the error constants, and calculates the jacobian at - J(t_{n+1}, y^0_{n+1}). This implementation was based on that implemented in the - SciPy library :footcite:t:`Virtanen2020`, which also mainly follows - :footcite:t:`shampine1997matlab` but uses the more standard jacobian update. + follows the Matlab routine ode15s described in :footcite:t:`shampine1997matlab` + and the SciPy implementation :footcite:t:`Virtanen2020` which features + the NDF formulas for improved stability, with associated differences in the + error constants, and calculates the jacobian at J(t_{n+1}, y^0_{n+1}). This + implementation was based on that implemented in the SciPy library + :footcite:t:`Virtanen2020`, which also mainly follows :footcite:t:`shampine1997matlab` + but uses the more standard jacobian update. Parameters ---------- diff --git a/src/pybamm/solvers/jax_solver.py b/src/pybamm/solvers/jax_solver.py index bfcdef1882..a1f1733ed6 100644 --- a/src/pybamm/solvers/jax_solver.py +++ b/src/pybamm/solvers/jax_solver.py @@ -31,10 +31,10 @@ class JaxSolver(pybamm.BaseSolver): Parameters ---------- method: str, optional (see `jax.experimental.ode.odeint` for details) - 'RK45' (default) uses jax.experimental.ode.odeint - * 'BDF' uses custom jax_bdf_integrate (see `jax_bdf_integrate.py` for details) + * 'BDF' (default) uses custom jax_bdf_integrate (see `jax_bdf_integrate.py` for details) + * 'RK45' uses jax.experimental.ode.odeint root_method: str, optional - Method to use to calculate consistent initial conditions. By default this uses + Method to use to calculate consistent initial conditions. By default, this uses the newton chord method internal to the jax bdf solver, otherwise choose from the set of default options defined in docs for pybamm.BaseSolver rtol : float, optional @@ -52,7 +52,7 @@ class JaxSolver(pybamm.BaseSolver): def __init__( self, - method="RK45", + method="BDF", root_method=None, rtol=1e-6, atol=1e-6, @@ -189,6 +189,10 @@ def solve_model_bdf(inputs): def supports_parallel_solve(self): return True + @property + def requires_explicit_sensitivities(self): + return False + def _integrate(self, model, t_eval, inputs=None, t_interp=None): """ Solve a model defined by dydt with initial conditions y0.	[Bug]: JAX BDF solver sensitivities bug ### PyBaMM Version develop ### Python Version 3.12 ### Describe the bug While sensitivities are not currently supported with the JAX solvers, passing the `calculate_sensitivities` argument currently results in an internal error due to extending the mass matrix, but not the corresponding initial conditions. ### Steps to Reproduce Pass `calculate_sensitivities` to Jax BDF `solver.solve()`. I.e, ```python ... solver = pybamm.JaxSolver(atol=1e-6, rtol=1e-6, method="BDF") sol = solver.solve(model, t_eval, calculate_sensitivities=True) ``` ### Relevant log output ```shell Traceback (most recent call last): File "/Users/Documents/Git/forks/PyBaMM/examples/scripts/compare_lithium_ion.py", line 31, in <module> solution = solver.solve(model, t_eval, inputs=inputs, calculate_sensitivities=True) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "/Users/Documents/Git/forks/PyBaMM/src/pybamm/solvers/base_solver.py", line 905, in solve new_solutions = self._integrate( ^^^^^^^^^^^^^^^^ File "/Users/Documents/Git/forks/PyBaMM/src/pybamm/solvers/jax_solver.py", line 231, in _integrate y = asyncio.run(solve_model_for_inputs()) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "/Users/.pyenv/versions/3.12.2/lib/python3.12/asyncio/runners.py", line 194, in run return runner.run(main) ^^^^^^^^^^^^^^^^ File "/Users/.pyenv/versions/3.12.2/lib/python3.12/asyncio/runners.py", line 118, in run return self._loop.run_until_complete(task) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "/Users/.pyenv/versions/3.12.2/lib/python3.12/asyncio/base_events.py", line 685, in run_until_complete return future.result() ^^^^^^^^^^^^^^^ File "/Users/Documents/Git/forks/PyBaMM/src/pybamm/solvers/jax_solver.py", line 229, in solve_model_for_inputs return await asyncio.gather(coro) ^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "/Users/Documents/Git/forks/PyBaMM/src/pybamm/solvers/jax_solver.py", line 224, in solve_model_async return self._cached_solves[model](inputs_v) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "/Users/Documents/Git/forks/PyBaMM/src/pybamm/solvers/jax_solver.py", line 171, in solve_model_bdf y = pybamm.jax_bdf_integrate( ^^^^^^^^^^^^^^^^^^^^^^^^^ File "/Users/Documents/Git/forks/PyBaMM/src/pybamm/solvers/jax_bdf_solver.py", line 1030, in jax_bdf_integrate return _bdf_odeint_wrapper(converted, mass, rtol, atol, y0, t_eval, consts, args) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "/Users/Documents/Git/forks/PyBaMM/src/pybamm/solvers/jax_bdf_solver.py", line 63, in caller return callee(args) ^^^^^^^^^^^^^ File "/Users/Documents/Git/forks/PyBaMM/src/pybamm/solvers/jax_bdf_solver.py", line 57, in callee return fun(args) ^^^^^^^^^^ File "/Users/Documents/Git/forks/PyBaMM/src/pybamm/solvers/jax_bdf_solver.py", line 812, in _bdf_odeint_wrapper out = _bdf_odeint(func, mass, rtol, atol, y0, ts, args) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "/Users/Documents/Git/forks/PyBaMM/src/pybamm/solvers/jax_bdf_solver.py", line 117, in _bdf_odeint stepper = _bdf_init( ^^^^^^^^^^ File "/Users/Documents/Git/forks/PyBaMM/src/pybamm/solvers/jax_bdf_solver.py", line 256, in _bdf_init state["LU"] = jax.scipy.linalg.lu_factor(state["M"] - c * J) ~~~~~~~~~~~^~~~~~~ File "/Users/.pyenv/versions/pybamm-v24.5/lib/python3.12/site-packages/jax/_src/numpy/array_methods.py", line 737, in op return getattr(self.aval, f"_{name}")(self, args) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "/Users/.pyenv/versions/pybamm-v24.5/lib/python3.12/site-packages/jax/_src/numpy/array_methods.py", line 265, in deferring_binary_op return binary_op(args) ^^^^^^^^^^^^^^^^ File "/Users/.pyenv/versions/pybamm-v24.5/lib/python3.12/site-packages/jax/_src/numpy/ufuncs.py", line 86, in <lambda> fn = lambda x1, x2, /: lax_fn(promote_args(numpy_fn.__name__, x1, x2)) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ TypeError: sub got incompatible shapes for broadcasting: (84, 84), (42, 42). ``` [Bug]: JAX BDF solver sensitivities bug ### PyBaMM Version develop ### Python Version 3.12 ### Describe the bug While sensitivities are not currently supported with the JAX solvers, passing the `calculate_sensitivities` argument currently results in an internal error due to extending the mass matrix, but not the corresponding initial conditions. ### Steps to Reproduce Pass `calculate_sensitivities` to Jax BDF `solver.solve()`. I.e, ```python ... solver = pybamm.JaxSolver(atol=1e-6, rtol=1e-6, method="BDF") sol = solver.solve(model, t_eval, calculate_sensitivities=True) ``` ### Relevant log output ```shell Traceback (most recent call last): File "/Users/Documents/Git/forks/PyBaMM/examples/scripts/compare_lithium_ion.py", line 31, in <module> solution = solver.solve(model, t_eval, inputs=inputs, calculate_sensitivities=True) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "/Users/Documents/Git/forks/PyBaMM/src/pybamm/solvers/base_solver.py", line 905, in solve new_solutions = self._integrate( ^^^^^^^^^^^^^^^^ File "/Users/Documents/Git/forks/PyBaMM/src/pybamm/solvers/jax_solver.py", line 231, in _integrate y = asyncio.run(solve_model_for_inputs()) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "/Users/.pyenv/versions/3.12.2/lib/python3.12/asyncio/runners.py", line 194, in run return runner.run(main) ^^^^^^^^^^^^^^^^ File "/Users/.pyenv/versions/3.12.2/lib/python3.12/asyncio/runners.py", line 118, in run return self._loop.run_until_complete(task) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "/Users/.pyenv/versions/3.12.2/lib/python3.12/asyncio/base_events.py", line 685, in run_until_complete return future.result() ^^^^^^^^^^^^^^^ File "/Users/Documents/Git/forks/PyBaMM/src/pybamm/solvers/jax_solver.py", line 229, in solve_model_for_inputs return await asyncio.gather(coro) ^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "/Users/Documents/Git/forks/PyBaMM/src/pybamm/solvers/jax_solver.py", line 224, in solve_model_async return self._cached_solves[model](inputs_v) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "/Users/Documents/Git/forks/PyBaMM/src/pybamm/solvers/jax_solver.py", line 171, in solve_model_bdf y = pybamm.jax_bdf_integrate( ^^^^^^^^^^^^^^^^^^^^^^^^^ File "/Users/Documents/Git/forks/PyBaMM/src/pybamm/solvers/jax_bdf_solver.py", line 1030, in jax_bdf_integrate return _bdf_odeint_wrapper(converted, mass, rtol, atol, y0, t_eval, consts, args) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "/Users/Documents/Git/forks/PyBaMM/src/pybamm/solvers/jax_bdf_solver.py", line 63, in caller return callee(args) ^^^^^^^^^^^^^ File "/Users/Documents/Git/forks/PyBaMM/src/pybamm/solvers/jax_bdf_solver.py", line 57, in callee return fun(args) ^^^^^^^^^^ File "/Users/Documents/Git/forks/PyBaMM/src/pybamm/solvers/jax_bdf_solver.py", line 812, in _bdf_odeint_wrapper out = _bdf_odeint(func, mass, rtol, atol, y0, ts, args) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "/Users/Documents/Git/forks/PyBaMM/src/pybamm/solvers/jax_bdf_solver.py", line 117, in _bdf_odeint stepper = _bdf_init( ^^^^^^^^^^ File "/Users/Documents/Git/forks/PyBaMM/src/pybamm/solvers/jax_bdf_solver.py", line 256, in _bdf_init state["LU"] = jax.scipy.linalg.lu_factor(state["M"] - c J) ~~~~~~~~~~~^~~~~~~ File "/Users/.pyenv/versions/pybamm-v24.5/lib/python3.12/site-packages/jax/_src/numpy/array_methods.py", line 737, in op return getattr(self.aval, f"_{name}")(self, args) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "/Users/.pyenv/versions/pybamm-v24.5/lib/python3.12/site-packages/jax/_src/numpy/array_methods.py", line 265, in deferring_binary_op return binary_op(args) ^^^^^^^^^^^^^^^^ File "/Users/.pyenv/versions/pybamm-v24.5/lib/python3.12/site-packages/jax/_src/numpy/ufuncs.py", line 86, in <lambda> fn = lambda x1, x2, /: lax_fn(*promote_args(numpy_fn.__name__, x1, x2)) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ TypeError: sub got incompatible shapes for broadcasting: (84, 84), (42, 42). ```		2024-09-23T12:19:12	[]	[]
pybamm-team/PyBaMM	pybamm-team__PyBaMM-4300	5408ce3063a9d0f0e2a21c9deafc003703e531f6	diff --git a/CHANGELOG.md b/CHANGELOG.md index 044c85609b..1de6947420 100644 --- a/CHANGELOG.md +++ b/CHANGELOG.md @@ -4,6 +4,10 @@ - Added additional user-configurable options to the (`IDAKLUSolver`) and adjusted the default values to improve performance. ([#4282](https://github.com/pybamm-team/PyBaMM/pull/4282)) +## Bug Fixes + +- Fixed bug where IDAKLU solver failed when `output variables` were specified and an event triggered. ([#4300](https://github.com/pybamm-team/PyBaMM/pull/4300)) + # [v24.5](https://github.com/pybamm-team/PyBaMM/tree/v24.5) - 2024-07-26 ## Features diff --git a/pybamm/solvers/c_solvers/idaklu.cpp b/pybamm/solvers/c_solvers/idaklu.cpp index 3afed5faa8..3427c01853 100644 --- a/pybamm/solvers/c_solvers/idaklu.cpp +++ b/pybamm/solvers/c_solvers/idaklu.cpp @@ -187,5 +187,6 @@ PYBIND11_MODULE(idaklu, m) .def_readwrite("t", &Solution::t) .def_readwrite("y", &Solution::y) .def_readwrite("yS", &Solution::yS) + .def_readwrite("y_term", &Solution::y_term) .def_readwrite("flag", &Solution::flag); } diff --git a/pybamm/solvers/c_solvers/idaklu/IDAKLUSolverOpenMP.inl b/pybamm/solvers/c_solvers/idaklu/IDAKLUSolverOpenMP.inl index 340baa2d30..96ebb40d10 100644 --- a/pybamm/solvers/c_solvers/idaklu/IDAKLUSolverOpenMP.inl +++ b/pybamm/solvers/c_solvers/idaklu/IDAKLUSolverOpenMP.inl @@ -401,6 +401,7 @@ Solution IDAKLUSolverOpenMP<ExprSet>::solve( // set return vectors int length_of_return_vector = 0; + int length_of_final_sv_slice = 0; size_t max_res_size = 0; // maximum result size (for common result buffer) size_t max_res_dvar_dy = 0, max_res_dvar_dp = 0; if (functions->var_fcns.size() > 0) { @@ -414,6 +415,7 @@ Solution IDAKLUSolverOpenMP<ExprSet>::solve( for (auto& dvar_fcn : functions->dvar_dp_fcns) { max_res_dvar_dp = std::max(max_res_dvar_dp, size_t(dvar_fcn->out_shape(0))); } + length_of_final_sv_slice = number_of_states; } } else { // Return full y state-vector @@ -425,6 +427,7 @@ Solution IDAKLUSolverOpenMP<ExprSet>::solve( realtype yS_return = new realtype[number_of_parameters number_of_timesteps * length_of_return_vector]; + realtype yterm_return = new realtype[length_of_final_sv_slice]; res.resize(max_res_size); res_dvar_dy.resize(max_res_dvar_dy); @@ -451,6 +454,13 @@ Solution IDAKLUSolverOpenMP<ExprSet>::solve( delete[] vect; } ); + py::capsule free_yterm_when_done( + yterm_return, + [](void f) { + realtype vect = reinterpret_cast<realtype >(f); + delete[] vect; + } + ); // Initial state (t_i=0) int t_i = 0; @@ -518,6 +528,10 @@ Solution IDAKLUSolverOpenMP<ExprSet>::solve( t_i += 1; if (retval == IDA_SUCCESS \|\| retval == IDA_ROOT_RETURN) { + if (functions->var_fcns.size() > 0) { + // store final state slice if outout variables are specified + yterm_return = yval; + } break; } } @@ -532,7 +546,7 @@ Solution IDAKLUSolverOpenMP<ExprSet>::solve( &y_return[0], free_y_when_done ); - // Note: Ordering of vector is differnet if computing variables vs returning + // Note: Ordering of vector is different if computing variables vs returning // the complete state vector np_array yS_ret; if (functions->var_fcns.size() > 0) { @@ -556,8 +570,13 @@ Solution IDAKLUSolverOpenMP<ExprSet>::solve( free_yS_when_done ); } + np_array y_term = np_array( + length_of_final_sv_slice, + &yterm_return[0], + free_yterm_when_done + ); - Solution sol(retval, t_ret, y_ret, yS_ret); + Solution sol(retval, t_ret, y_ret, yS_ret, y_term); if (solver_opts.print_stats) { long nsteps, nrevals, nlinsetups, netfails; diff --git a/pybamm/solvers/c_solvers/idaklu/Solution.hpp b/pybamm/solvers/c_solvers/idaklu/Solution.hpp index 92e22d02b6..ad0ea06762 100644 --- a/pybamm/solvers/c_solvers/idaklu/Solution.hpp +++ b/pybamm/solvers/c_solvers/idaklu/Solution.hpp @@ -12,8 +12,8 @@ class Solution /** * @brief Constructor / - Solution(int retval, np_array t_np, np_array y_np, np_array yS_np) - : flag(retval), t(t_np), y(y_np), yS(yS_np) + Solution(int retval, np_array t_np, np_array y_np, np_array yS_np, np_array y_term_np) + : flag(retval), t(t_np), y(y_np), yS(yS_np), y_term(y_term_np) { } @@ -21,6 +21,7 @@ class Solution np_array t; np_array y; np_array yS; + np_array y_term; }; #endif // PYBAMM_IDAKLU_COMMON_HPP diff --git a/pybamm/solvers/c_solvers/idaklu/python.cpp b/pybamm/solvers/c_solvers/idaklu/python.cpp index 03090c9850..015f504086 100644 --- a/pybamm/solvers/c_solvers/idaklu/python.cpp +++ b/pybamm/solvers/c_solvers/idaklu/python.cpp @@ -478,8 +478,9 @@ Solution solve_python(np_array t_np, np_array y0_np, np_array yp0_np, std::vector<ptrdiff_t> {number_of_parameters, number_of_timesteps, number_of_states}, &yS_return[0] ); + np_array yterm_ret = np_array(0); - Solution sol(retval, t_ret, y_ret, yS_ret); + Solution sol(retval, t_ret, y_ret, yS_ret, yterm_ret); return sol; } diff --git a/pybamm/solvers/idaklu_solver.py b/pybamm/solvers/idaklu_solver.py index 8741a595f8..fa92563e3b 100644 --- a/pybamm/solvers/idaklu_solver.py +++ b/pybamm/solvers/idaklu_solver.py @@ -816,6 +816,7 @@ def _make_iree_function(self, fcn, args, sparse_index=False): def fcn(args): return fcn_inner(args)[coo.row, coo.col] + elif coo.nnz != iree_fcn.numel: iree_fcn.nnz = iree_fcn.numel iree_fcn.col = list(range(iree_fcn.numel)) @@ -969,8 +970,10 @@ def _integrate(self, model, t_eval, inputs_dict=None): if self.output_variables: # Substitute empty vectors for state vector 'y' y_out = np.zeros((number_of_timesteps * number_of_states, 0)) + y_event = sol.y_term else: y_out = sol.y.reshape((number_of_timesteps, number_of_states)) + y_event = y_out[-1] # return sensitivity solution, we need to flatten yS to # (#timesteps * #states (where t is changing the quickest),) @@ -1000,7 +1003,7 @@ def _integrate(self, model, t_eval, inputs_dict=None): model, inputs_dict, np.array([t[-1]]), - np.transpose(y_out[-1])[:, np.newaxis], + np.transpose(y_event)[:, np.newaxis], termination, sensitivities=yS_out, ) diff --git a/pybamm/solvers/solution.py b/pybamm/solvers/solution.py index 39281ef4b0..c3c8451634 100644 --- a/pybamm/solvers/solution.py +++ b/pybamm/solvers/solution.py @@ -100,10 +100,6 @@ def __init__( self.sensitivities = sensitivities - self._t_event = t_event - self._y_event = y_event - self._termination = termination - # Check no ys are too large if check_solution: self.check_ys_are_not_too_large()	[Bug]: output variables option in idaklu solver fails when event triggered ### PyBaMM Version 24.1 ### Python Version 3.10 ### Describe the bug When the output_variables option of the IDAKLU solver is used in a simulation with an event, the solve fails with the error message below when trying to get the event reason ### Steps to Reproduce ```python def test(solver): model = pybamm.lithium_ion.DFN() parameter_values = pybamm.ParameterValues("Chen2020") sim = pybamm.Simulation( model, parameter_values=parameter_values, solver=solver, ) sol = sim.solve(np.linspace(0, 3600, 1000)) sol = sol["Terminal voltage [V]"].entries test(pybamm.IDAKLUSolver(output_variables=["Terminal voltage [V]"])) ``` ### Relevant log output ```shell Traceback (most recent call last): File "/home/mrobins/git/tmp/test.py", line 17, in <module> test(pybamm.IDAKLUSolver(output_variables=["Terminal voltage [V]"])) File "/home/mrobins/git/tmp/test.py", line 14, in test sol = sim.solve(np.linspace(0, 3600, 1000)) File "/home/mrobins/git/tmp/env/lib/python3.10/site-packages/pybamm/simulation.py", line 584, in solve self._solution = solver.solve(self.built_model, t_eval, **kwargs) File "/home/mrobins/git/tmp/env/lib/python3.10/site-packages/pybamm/solvers/base_solver.py", line 959, in solve solutions[i], termination = self.get_termination_reason( File "/home/mrobins/git/tmp/env/lib/python3.10/site-packages/pybamm/solvers/base_solver.py", line 1303, in get_termination_reason final_event_values[event.name] = event.expression.evaluate( File "/home/mrobins/git/tmp/env/lib/python3.10/site-packages/pybamm/expression_tree/binary_operators.py", line 132, in evaluate right = self.right.evaluate(t, y, y_dot, inputs) File "/home/mrobins/git/tmp/env/lib/python3.10/site-packages/pybamm/expression_tree/binary_operators.py", line 132, in evaluate right = self.right.evaluate(t, y, y_dot, inputs) File "/home/mrobins/git/tmp/env/lib/python3.10/site-packages/pybamm/expression_tree/binary_operators.py", line 132, in evaluate right = self.right.evaluate(t, y, y_dot, inputs) File "/home/mrobins/git/tmp/env/lib/python3.10/site-packages/pybamm/expression_tree/symbol.py", line 778, in evaluate return self._base_evaluate(t, y, y_dot, inputs) File "/home/mrobins/git/tmp/env/lib/python3.10/site-packages/pybamm/expression_tree/state_vector.py", line 284, in _base_evaluate raise ValueError( ValueError: y is too short, so value with slice is smaller than expected (env) ➜ tmp pip list \|grep pybamm ```		2024-07-29T11:29:43	[]	[]
pybamm-team/PyBaMM	pybamm-team__PyBaMM-4239	5e04abdd13fdebb1d8d01f7ba53c4bd02f8a7dfd	diff --git a/docs/source/examples/notebooks/getting_started/tutorial-5-run-experiments.ipynb b/docs/source/examples/notebooks/getting_started/tutorial-5-run-experiments.ipynb index eb82f59719..85be34e421 100644 --- a/docs/source/examples/notebooks/getting_started/tutorial-5-run-experiments.ipynb +++ b/docs/source/examples/notebooks/getting_started/tutorial-5-run-experiments.ipynb @@ -25,18 +25,8 @@ "name": "stdout", "output_type": "stream", "text": [ - "\n", - "\u001b[1m[\u001b[0m\u001b[34;49mnotice\u001b[0m\u001b[1;39;49m]\u001b[0m\u001b[39;49m A new release of pip is available: \u001b[0m\u001b[31;49m23.3.1\u001b[0m\u001b[39;49m -> \u001b[0m\u001b[32;49m24.0\u001b[0m\n", - "\u001b[1m[\u001b[0m\u001b[34;49mnotice\u001b[0m\u001b[1;39;49m]\u001b[0m\u001b[39;49m To update, run: \u001b[0m\u001b[32;49mpip install --upgrade pip\u001b[0m\n", "Note: you may need to restart the kernel to use updated packages.\n" ] - }, - { - "name": "stderr", - "output_type": "stream", - "text": [ - "An NVIDIA GPU may be present on this machine, but a CUDA-enabled jaxlib is not installed. Falling back to cpu.\n" - ] } ], "source": [ @@ -163,18 +153,19 @@ "name": "stderr", "output_type": "stream", "text": [ - "At t = 522.66 and h = 1.1556e-13, the corrector convergence failed repeatedly or with \|h\| = hmin.\n" + "At t = 339.952 and h = 1.4337e-18, the corrector convergence failed repeatedly or with \|h\| = hmin.\n", + "At t = 522.687 and h = 4.04917e-14, the corrector convergence failed repeatedly or with \|h\| = hmin.\n" ] }, { "data": { "application/vnd.jupyter.widget-view+json": { - "model_id": "5ab1f22de6af4878b6ca43d27ffc01c5", + "model_id": "93feca98298f4111909ae487e2a1e273", "version_major": 2, "version_minor": 0 }, "text/plain": [ - "interactive(children=(FloatSlider(value=0.0, description='t', max=40.132949019384355, step=0.40132949019384356…" + "interactive(children=(FloatSlider(value=0.0, description='t', max=40.13268704803602, step=0.4013268704803602),…" ] }, "metadata": {}, @@ -183,7 +174,7 @@ { "data": { "text/plain": [ - "<pybamm.plotting.quick_plot.QuickPlot at 0x7f1a11f76690>" + "<pybamm.plotting.quick_plot.QuickPlot at 0x16520e690>" ] }, "execution_count": 5, @@ -211,12 +202,12 @@ { "data": { "application/vnd.jupyter.widget-view+json": { - "model_id": "7cdac234d74241a2814918053454f6a6", + "model_id": "4d6e43032f4e4aa6be5843c4916b4b50", "version_major": 2, "version_minor": 0 }, "text/plain": [ - "interactive(children=(FloatSlider(value=0.0, description='t', max=13.076977041121545, step=0.13076977041121546…" + "interactive(children=(FloatSlider(value=0.0, description='t', max=13.076887099589111, step=0.1307688709958911)…" ] }, "metadata": {}, @@ -225,7 +216,7 @@ { "data": { "text/plain": [ - "<pybamm.plotting.quick_plot.QuickPlot at 0x7f1a105ecb50>" + "<pybamm.plotting.quick_plot.QuickPlot at 0x1696cf290>" ] }, "execution_count": 6, @@ -255,7 +246,7 @@ { "data": { "text/plain": [ - "_Step(C-rate, 1.0, duration=1 hour, period=1 minute, temperature=25oC, tags=['tag1'], description=Discharge at 1C for 1 hour)" + "Step(1.0, duration=1 hour, period=1 minute, temperature=25oC, tags=['tag1'], description=Discharge at 1C for 1 hour, direction=Discharge)" ] }, "execution_count": 7, @@ -293,7 +284,7 @@ { "data": { "text/plain": [ - "_Step(current, 1, duration=1 hour, termination=2.5 V)" + "Step(1, duration=1 hour, termination=2.5 V, direction=Discharge)" ] }, "execution_count": 8, @@ -321,7 +312,7 @@ { "data": { "text/plain": [ - "_Step(current, 1.0, duration=1 hour, termination=2.5V, description=Discharge at 1A for 1 hour or until 2.5V)" + "Step(1.0, duration=1 hour, termination=2.5V, description=Discharge at 1A for 1 hour or until 2.5V, direction=Discharge)" ] }, "execution_count": 9, @@ -348,10 +339,78 @@ "execution_count": 10, "metadata": {}, "outputs": [ + { + "name": "stderr", + "output_type": "stream", + "text": [ + "2024-07-10 14:41:02.625 - [WARNING] callbacks.on_experiment_infeasible_time(240): \n", + "\n", + "\tExperiment is infeasible: default duration (1.0 seconds) was reached during 'Step([[ 0.00000000e+00 0.00000000e+00]\n", + " [ 1.69491525e-02 5.31467428e-02]\n", + " [ 3.38983051e-02 1.05691312e-01]\n", + " [ 5.08474576e-02 1.57038356e-01]\n", + " [ 6.77966102e-02 2.06606093e-01]\n", + " [ 8.47457627e-02 2.53832900e-01]\n", + " [ 1.01694915e-01 2.98183679e-01]\n", + " [ 1.18644068e-01 3.39155918e-01]\n", + " [ 1.35593220e-01 3.76285385e-01]\n", + " [ 1.52542373e-01 4.09151388e-01]\n", + " [ 1.69491525e-01 4.37381542e-01]\n", + " [ 1.86440678e-01 4.60655989e-01]\n", + " [ 2.03389831e-01 4.78711019e-01]\n", + " [ 2.20338983e-01 4.91342062e-01]\n", + " [ 2.37288136e-01 4.98406004e-01]\n", + " [ 2.54237288e-01 4.99822806e-01]\n", + " [ 2.71186441e-01 4.95576416e-01]\n", + " [ 2.88135593e-01 4.85714947e-01]\n", + " [ 3.05084746e-01 4.70350133e-01]\n", + " [ 3.22033898e-01 4.49656065e-01]\n", + " [ 3.38983051e-01 4.23867214e-01]\n", + " [ 3.55932203e-01 3.93275778e-01]\n", + " [ 3.72881356e-01 3.58228370e-01]\n", + " [ 3.89830508e-01 3.19122092e-01]\n", + " [ 4.06779661e-01 2.76400033e-01]\n", + " [ 4.23728814e-01 2.30546251e-01]\n", + " [ 4.40677966e-01 1.82080288e-01]\n", + " [ 4.57627119e-01 1.31551282e-01]\n", + " [ 4.74576271e-01 7.95317480e-02]\n", + " [ 4.91525424e-01 2.66110874e-02]\n", + " [ 5.08474576e-01 -2.66110874e-02]\n", + " [ 5.25423729e-01 -7.95317480e-02]\n", + " [ 5.42372881e-01 -1.31551282e-01]\n", + " [ 5.59322034e-01 -1.82080288e-01]\n", + " [ 5.76271186e-01 -2.30546251e-01]\n", + " [ 5.93220339e-01 -2.76400033e-01]\n", + " [ 6.10169492e-01 -3.19122092e-01]\n", + " [ 6.27118644e-01 -3.58228370e-01]\n", + " [ 6.44067797e-01 -3.93275778e-01]\n", + " [ 6.61016949e-01 -4.23867214e-01]\n", + " [ 6.77966102e-01 -4.49656065e-01]\n", + " [ 6.94915254e-01 -4.70350133e-01]\n", + " [ 7.11864407e-01 -4.85714947e-01]\n", + " [ 7.28813559e-01 -4.95576416e-01]\n", + " [ 7.45762712e-01 -4.99822806e-01]\n", + " [ 7.62711864e-01 -4.98406004e-01]\n", + " [ 7.79661017e-01 -4.91342062e-01]\n", + " [ 7.96610169e-01 -4.78711019e-01]\n", + " [ 8.13559322e-01 -4.60655989e-01]\n", + " [ 8.30508475e-01 -4.37381542e-01]\n", + " [ 8.47457627e-01 -4.09151388e-01]\n", + " [ 8.64406780e-01 -3.76285385e-01]\n", + " [ 8.81355932e-01 -3.39155918e-01]\n", + " [ 8.98305085e-01 -2.98183679e-01]\n", + " [ 9.15254237e-01 -2.53832900e-01]\n", + " [ 9.32203390e-01 -2.06606093e-01]\n", + " [ 9.49152542e-01 -1.57038356e-01]\n", + " [ 9.66101695e-01 -1.05691312e-01]\n", + " [ 9.83050847e-01 -5.31467428e-02]\n", + " [ 1.00000000e+00 -1.22464680e-16]], duration=1.0, period=0.016949152542372836, direction=Rest)'. The returned solution only contains up to step 1 of cycle 1. Please specify a duration in the step instructions.\n" + ] + }, { "data": { "application/vnd.jupyter.widget-view+json": { - "model_id": "730d5e19b17e447ebde5679de68c46ef", + "model_id": "6364b4579fc447e2a607f2f8414172ba", "version_major": 2, "version_minor": 0 }, @@ -365,7 +424,7 @@ { "data": { "text/plain": [ - "<pybamm.plotting.quick_plot.QuickPlot at 0x7f1a1a4c2ed0>" + "<pybamm.plotting.quick_plot.QuickPlot at 0x16a1d8d90>" ] }, "execution_count": 10, @@ -419,13 +478,14 @@ "output_type": "stream", "text": [ "[1] Joel A. E. Andersson, Joris Gillis, Greg Horn, James B. Rawlings, and Moritz Diehl. CasADi – A software framework for nonlinear optimization and optimal control. Mathematical Programming Computation, 11(1):1–36, 2019. doi:10.1007/s12532-018-0139-4.\n", - "[2] Marc Doyle, Thomas F. Fuller, and John Newman. Modeling of galvanostatic charge and discharge of the lithium/polymer/insertion cell. Journal of the Electrochemical society, 140(6):1526–1533, 1993. doi:10.1149/1.2221597.\n", - "[3] Charles R. Harris, K. Jarrod Millman, Stéfan J. van der Walt, Ralf Gommers, Pauli Virtanen, David Cournapeau, Eric Wieser, Julian Taylor, Sebastian Berg, Nathaniel J. Smith, and others. Array programming with NumPy. Nature, 585(7825):357–362, 2020. doi:10.1038/s41586-020-2649-2.\n", - "[4] Scott G. Marquis, Valentin Sulzer, Robert Timms, Colin P. Please, and S. Jon Chapman. An asymptotic derivation of a single particle model with electrolyte. Journal of The Electrochemical Society, 166(15):A3693–A3706, 2019. doi:10.1149/2.0341915jes.\n", - "[5] Peyman Mohtat, Suhak Lee, Jason B Siegel, and Anna G Stefanopoulou. Towards better estimability of electrode-specific state of health: decoding the cell expansion. Journal of Power Sources, 427:101–111, 2019.\n", - "[6] Valentin Sulzer, Scott G. Marquis, Robert Timms, Martin Robinson, and S. Jon Chapman. Python Battery Mathematical Modelling (PyBaMM). Journal of Open Research Software, 9(1):14, 2021. doi:10.5334/jors.309.\n", - "[7] Pauli Virtanen, Ralf Gommers, Travis E. Oliphant, Matt Haberland, Tyler Reddy, David Cournapeau, Evgeni Burovski, Pearu Peterson, Warren Weckesser, Jonathan Bright, and others. SciPy 1.0: fundamental algorithms for scientific computing in Python. Nature Methods, 17(3):261–272, 2020. doi:10.1038/s41592-019-0686-2.\n", - "[8] Andrew Weng, Jason B Siegel, and Anna Stefanopoulou. Differential voltage analysis for battery manufacturing process control. arXiv preprint arXiv:2303.07088, 2023.\n", + "[2] Von DAG Bruggeman. Berechnung verschiedener physikalischer konstanten von heterogenen substanzen. i. dielektrizitätskonstanten und leitfähigkeiten der mischkörper aus isotropen substanzen. Annalen der physik, 416(7):636–664, 1935.\n", + "[3] Marc Doyle, Thomas F. Fuller, and John Newman. Modeling of galvanostatic charge and discharge of the lithium/polymer/insertion cell. Journal of the Electrochemical society, 140(6):1526–1533, 1993. doi:10.1149/1.2221597.\n", + "[4] Charles R. Harris, K. Jarrod Millman, Stéfan J. van der Walt, Ralf Gommers, Pauli Virtanen, David Cournapeau, Eric Wieser, Julian Taylor, Sebastian Berg, Nathaniel J. Smith, and others. Array programming with NumPy. Nature, 585(7825):357–362, 2020. doi:10.1038/s41586-020-2649-2.\n", + "[5] Scott G. Marquis, Valentin Sulzer, Robert Timms, Colin P. Please, and S. Jon Chapman. An asymptotic derivation of a single particle model with electrolyte. Journal of The Electrochemical Society, 166(15):A3693–A3706, 2019. doi:10.1149/2.0341915jes.\n", + "[6] Peyman Mohtat, Suhak Lee, Jason B Siegel, and Anna G Stefanopoulou. Towards better estimability of electrode-specific state of health: decoding the cell expansion. Journal of Power Sources, 427:101–111, 2019.\n", + "[7] Valentin Sulzer, Scott G. Marquis, Robert Timms, Martin Robinson, and S. Jon Chapman. Python Battery Mathematical Modelling (PyBaMM). Journal of Open Research Software, 9(1):14, 2021. doi:10.5334/jors.309.\n", + "[8] Pauli Virtanen, Ralf Gommers, Travis E. Oliphant, Matt Haberland, Tyler Reddy, David Cournapeau, Evgeni Burovski, Pearu Peterson, Warren Weckesser, Jonathan Bright, and others. SciPy 1.0: fundamental algorithms for scientific computing in Python. Nature Methods, 17(3):261–272, 2020. doi:10.1038/s41592-019-0686-2.\n", + "[9] Andrew Weng, Jason B Siegel, and Anna Stefanopoulou. Differential voltage analysis for battery manufacturing process control. arXiv preprint arXiv:2303.07088, 2023.\n", "\n" ] } diff --git a/pybamm/callbacks.py b/pybamm/callbacks.py index 57f2426d0f..4e8c67c8be 100644 --- a/pybamm/callbacks.py +++ b/pybamm/callbacks.py @@ -36,37 +36,37 @@ def on_experiment_start(self, logs): """ Called at the start of an experiment simulation. """ - pass + pass # pragma: no cover def on_cycle_start(self, logs): """ Called at the start of each cycle in an experiment simulation. """ - pass + pass # pragma: no cover def on_step_start(self, logs): """ Called at the start of each step in an experiment simulation. """ - pass + pass # pragma: no cover def on_step_end(self, logs): """ Called at the end of each step in an experiment simulation. """ - pass + pass # pragma: no cover def on_cycle_end(self, logs): """ Called at the end of each cycle in an experiment simulation. """ - pass + pass # pragma: no cover def on_experiment_end(self, logs): """ Called at the end of an experiment simulation. """ - pass + pass # pragma: no cover def on_experiment_error(self, logs): """ @@ -75,13 +75,19 @@ def on_experiment_error(self, logs): For example, this could be used to send an error alert with a bug report when running batch simulations in the cloud. """ - pass + pass # pragma: no cover - def on_experiment_infeasible(self, logs): + def on_experiment_infeasible_time(self, logs): """ - Called when an experiment simulation is infeasible. + Called when an experiment simulation is infeasible due to reaching maximum time. """ - pass + pass # pragma: no cover + + def on_experiment_infeasible_event(self, logs): + """ + Called when an experiment simulation is infeasible due to an event. + """ + pass # pragma: no cover ######################################################################################## @@ -226,7 +232,19 @@ def on_experiment_error(self, logs): error = logs["error"] pybamm.logger.error(f"Simulation error: {error}") - def on_experiment_infeasible(self, logs): + def on_experiment_infeasible_time(self, logs): + duration = logs["step duration"] + cycle_num = logs["cycle number"][0] + step_num = logs["step number"][0] + operating_conditions = logs["step operating conditions"] + self.logger.warning( + f"\n\n\tExperiment is infeasible: default duration ({duration} seconds) " + f"was reached during '{operating_conditions}'. The returned solution only " + f"contains up to step {step_num} of cycle {cycle_num}. " + "Please specify a duration in the step instructions." + ) + + def on_experiment_infeasible_event(self, logs): termination = logs["termination"] cycle_num = logs["cycle number"][0] step_num = logs["step number"][0] diff --git a/pybamm/experiment/step/base_step.py b/pybamm/experiment/step/base_step.py index 16224a6afa..6b77bed2cf 100644 --- a/pybamm/experiment/step/base_step.py +++ b/pybamm/experiment/step/base_step.py @@ -71,6 +71,8 @@ def __init__( description=None, direction=None, ): + self.input_duration = duration + self.input_value = value # Check if drive cycle is_drive_cycle = isinstance(value, np.ndarray) is_python_function = callable(value) @@ -100,8 +102,11 @@ def __init__( f"Input function must return a real number output at t = {t0}" ) + # Record whether the step uses the default duration + # This will be used by the experiment to check whether the step is feasible + self.uses_default_duration = duration is None # Set duration - if duration is None: + if self.uses_default_duration: duration = self.default_duration(value) self.duration = _convert_time_to_seconds(duration) @@ -195,8 +200,8 @@ def copy(self): A copy of the step. """ return self.__class__( - self.value, - duration=self.duration, + self.input_value, + duration=self.input_duration, termination=self.termination, period=self.period, temperature=self.temperature, @@ -259,7 +264,7 @@ def default_duration(self, value): t = value[:, 0] return t[-1] else: - return 24 * 3600 # 24 hours in seconds + return 24 * 3600 # one day in seconds def process_model(self, model, parameter_values): new_model = model.new_copy() @@ -411,10 +416,16 @@ def set_up(self, new_model, new_parameter_values): def _convert_time_to_seconds(time_and_units): """Convert a time in seconds, minutes or hours to a time in seconds""" - # If the time is a number, assume it is in seconds - if isinstance(time_and_units, numbers.Number) or time_and_units is None: + if time_and_units is None: return time_and_units + # If the time is a number, assume it is in seconds + if isinstance(time_and_units, numbers.Number): + if time_and_units <= 0: + raise ValueError("time must be positive") + else: + return time_and_units + # Split number and units units = time_and_units.lstrip("0123456789.- ") time = time_and_units[: -len(units)] diff --git a/pybamm/experiment/step/steps.py b/pybamm/experiment/step/steps.py index e3322104bf..e66178dc81 100644 --- a/pybamm/experiment/step/steps.py +++ b/pybamm/experiment/step/steps.py @@ -156,6 +156,11 @@ def __init__(self, value, *kwargs): def current_value(self, variables): return self.value pybamm.Parameter("Nominal cell capacity [A.h]") + def default_duration(self, value): + # "value" is C-rate, so duration is "1 / value" hours in seconds + # with a 2x safety factor + return 1 / abs(value) * 3600 * 2 + def c_rate(value, **kwargs): """ diff --git a/pybamm/simulation.py b/pybamm/simulation.py index 8ec85d67d4..a55310870e 100644 --- a/pybamm/simulation.py +++ b/pybamm/simulation.py @@ -679,6 +679,7 @@ def solve( logs["step number"] = (step_num, cycle_length) logs["step operating conditions"] = step_str + logs["step duration"] = step.duration callbacks.on_step_start(logs) inputs = { @@ -767,23 +768,33 @@ def solve( callbacks.on_step_end(logs) logs["termination"] = step_solution.termination - # Only allow events specified by experiment - if not ( + + # Check for some cases that would make the experiment end early + if step_termination == "final time" and step.uses_default_duration: + # reached the default duration of a step (typically we should + # reach an event before the default duration) + callbacks.on_experiment_infeasible_time(logs) + feasible = False + break + + elif not ( isinstance(step_solution, pybamm.EmptySolution) or step_termination == "final time" or "[experiment]" in step_termination ): - callbacks.on_experiment_infeasible(logs) + # Step has reached an event that is not specified in the + # experiment + callbacks.on_experiment_infeasible_event(logs) feasible = False break - if time_stop is not None: - max_time = cycle_solution.t[-1] - if max_time >= time_stop: - break + elif time_stop is not None and logs["experiment time"] >= time_stop: + # reached the time limit of the experiment + break - # Increment index for next iteration - idx += 1 + else: + # Increment index for next iteration, then continue + idx += 1 if save_this_cycle or feasible is False: self._solution = self._solution + cycle_solution	[Bug]: [24.5rc0] Simulations with `pybamm.Experiment` terminate after 24 h ### PyBaMM Version 24.5rc0 ### Python Version 3.9.13 ### Describe the bug Simulations using `pybamm.Experiment` with a voltage cut-off only appear to terminate with a time cut-off after 24 h. I'm afraid I haven't had time to get into logs to figure out why this is happening! ### Steps to Reproduce The following simulation should proceed for approximately 100 h = 360000 s, until the specified voltage cut-off. However, it terminates after 24 h. ```python import pybamm parameter_values = pybamm.ParameterValues("Chen2020") model = pybamm.lithium_ion.DFN() experiment = pybamm.Experiment( [ "Discharge at 0.01C until 2.5 V", ] ) sim = pybamm.Simulation( model, parameter_values=parameter_values, experiment=experiment, ) sol = sim.solve() sim.plot() print(f'Last time step at {sim.solution["Time [s]"].entries[-1] / 3600} h') ``` ### Relevant log output _No response_	Hi, @ejfdickinson – looks like a duplicate of https://github.com/pybamm-team/PyBaMM/issues/4155? @agriyakhetarpal Yes, it is. v24.1 does not do this, so it should be in your release notes for v24.5rc0. Any recommendations on how to work around? > Any recommendations on how to work around? @ejfdickinson you can include a duration in the experiment step, e.g., ```python experiment = pybamm.Experiment( [ "Discharge at 0.01C for 100 hours or until 2.5 V", ] ) ``` and it won't cut off at 24 hours. <img width="1484" alt="image" src="https://github.com/pybamm-team/PyBaMM/assets/34451391/cb30fa0d-f333-464d-951a-4dc644437e0a"> Thanks @MarcBerliner ! > v24.1 does not do this, so it should be in your release notes for v24.5rc0. I think we should mark this as a breaking change, @valentinsulzer? IMO we should fix this before 24.5 proper. I don't know what logic changed or if there was a longer default before. Update: the logic used to be ```python3 if op_conds.type == "current": # Current control: max simulation time: 3h / C-rate Crate = op_conds.value / capacity dt = 3 / abs(Crate) * 3600 # seconds else: # max simulation time: 1 day dt = 24 * 3600 # seconds ``` Alternatively, should we require a duration to avoid unexpected behaviour? Why did we move away from that logic? As in, if we bring it back will it break anything? It got moved to the `Step` class instead of the `Simulation` class. It's easy to put the same logic back in for C-rate, but not current, as the default duration needs to return a float, not a symbol (see the commit above). It's fixable but needs a bit of work, and I'm leaning towards requiring a duration as it's less ambiguous. It would be a breaking change but easy for users to fix. @rtimms For our use cases, a maximum duration would be fine, so long as it's declared on simulation exit like a `"Time"` stop rather than occurring silently. As noted in #4155 , 24 h is well within typical simulation times for normal experiments, so if the purpose is just to avoid hanging the solver due to an absurd duration, adding a few orders of magnitude to the default might be more sensible!	2024-07-03T17:32:40	[]	[]
pybamm-team/PyBaMM	pybamm-team__PyBaMM-4221	b50dfa595dd93a2943aa2a5a8aa14ac9468aea8e	diff --git a/pybamm/models/full_battery_models/base_battery_model.py b/pybamm/models/full_battery_models/base_battery_model.py index fb8d001f93..8a2e443338 100644 --- a/pybamm/models/full_battery_models/base_battery_model.py +++ b/pybamm/models/full_battery_models/base_battery_model.py @@ -633,6 +633,20 @@ def __init__(self, extra_options): "be 'none': 'particle mechanics', 'loss of active material'" ) + if "true" in options["SEI on cracks"]: + sei_on_cr = options["SEI on cracks"] + p_mechanics = options["particle mechanics"] + if isinstance(p_mechanics, str) and isinstance(sei_on_cr, tuple): + p_mechanics = (p_mechanics, p_mechanics) + if any( + sei == "true" and mech != "swelling and cracking" + for mech, sei in zip(p_mechanics, sei_on_cr) + ): + raise pybamm.OptionError( + "If 'SEI on cracks' is 'true' then 'particle mechanics' must be " + "'swelling and cracking'." + ) + # Check options are valid for option, value in options.items(): if isinstance(value, str) or option in [	Raise OptionError if "SEI on cracks" is "true" but cracking isn't enabled. If you do e.g. ```python3 model = pybamm.lithium_ion.SPMe( { "SEI": "solvent-diffusion limited", "SEI porosity change": "true", "lithium plating": "partially reversible", "lithium plating porosity change": "true", # alias for "SEI porosity change" "particle mechanics": ("swelling only", "swelling only"), "SEI on cracks": "true", "loss of active material": "stress-driven", } ) ``` you get a `ZeroDivisionError`. It doesn't make sense to have SEI n cracks if there are no cracks, so we should catch this and raise an `OptionError`	As an alternative the cracking could be turned on if the "SEI on cracks" is enabled. Going by the name of the parameter, I think I would expect that it also enables cracking	2024-06-26T16:48:05	[]	[]
pybamm-team/PyBaMM	pybamm-team__PyBaMM-4208	9b5766921af1f6d192e8b1974343eb7ccbd72efe	diff --git a/pybamm/parameters/parameter_values.py b/pybamm/parameters/parameter_values.py index c8011c735c..57ebf65058 100644 --- a/pybamm/parameters/parameter_values.py +++ b/pybamm/parameters/parameter_values.py @@ -65,7 +65,10 @@ def __init__(self, values, chemistry=None): values.pop("chemistry", None) self.update(values, check_already_exists=False) else: - raise ValueError("Invalid Parameter Value") + valid_sets = "\n".join(pybamm.parameter_sets.keys()) + raise ValueError( + f"'{values}' is not a valid parameter set. Parameter set must be one of:\n{valid_sets}" + ) # Initialise empty _processed_symbols dict (for caching) self._processed_symbols = {}	Better error on unrecognised parameter set Doing `pybamm.ParameterValues("bad")` gives `ValueError: Invalid Parameter Value`. It would be better to say something like `ValueError: 'bad' is not a recognised parameter set` and possibly list the valid options (including custom parameter sets that have been registered as entry points).		2024-06-24T15:06:53	[]	[]
pybamm-team/PyBaMM	pybamm-team__PyBaMM-4092	33d9ea7f28c26fcbe16461857adbada1349fe0ff	diff --git a/.github/workflows/publish_pypi.yml b/.github/workflows/publish_pypi.yml index fb4a203e5d..01bf7ad95f 100644 --- a/.github/workflows/publish_pypi.yml +++ b/.github/workflows/publish_pypi.yml @@ -99,7 +99,7 @@ jobs: strategy: fail-fast: false matrix: - os: [ubuntu-latest, macos-12] + os: [ubuntu-latest, macos-13] steps: - uses: actions/checkout@v4 name: Check out PyBaMM repository @@ -112,15 +112,6 @@ jobs: - name: Clone pybind11 repo (no history) run: git clone --depth 1 --branch v2.11.1 https://github.com/pybind/pybind11.git - # sometimes gfortran cannot be found, so reinstall gcc just to be sure - - name: Install SuiteSparse and SUNDIALS on macOS - if: matrix.os == 'macos-12' - run: \| - brew install graphviz libomp - brew reinstall gcc - python -m pip install cmake wget - python scripts/install_KLU_Sundials.py - - name: Build wheels on Linux run: pipx run cibuildwheel --output-dir wheelhouse if: matrix.os == 'ubuntu-latest' @@ -134,11 +125,50 @@ jobs: CIBW_TEST_COMMAND: python -c "import pybamm; pybamm.IDAKLUSolver()" - name: Build wheels on macOS amd64 - if: matrix.os == 'macos-12' + if: matrix.os == 'macos-13' run: pipx run cibuildwheel --output-dir wheelhouse env: - CIBW_BEFORE_BUILD_MACOS: > - python -m pip install --upgrade cmake casadi setuptools wheel + MACOSX_DEPLOYMENT_TARGET: 11.0 + # Sourced from + # License: BSD-3-Clause + # https://github.com/scipy/scipy/blob/f2d4775e7762fad984f8f0acd8227c725ff21630/tools/wheels/cibw_before_build_macos.sh#L23-L49 + CIBW_BEFORE_ALL_MACOS: \| + set -e -x + + # download gfortran with proper macos minimum version (11.0) + curl -L https://github.com/isuruf/gcc/releases/download/gcc-11.3.0-2/gfortran-darwin-x86_64-native.tar.gz -o gfortran.tar.gz + + GFORTRAN_SHA256=$(shasum --algorithm 256 gfortran.tar.gz) + KNOWN_SHA256="981367dd0ad4335613e91bbee453d60b6669f5d7e976d18c7bdb7f1966f26ae4 gfortran.tar.gz" + if [ "$GFORTRAN_SHA256" != "$KNOWN_SHA256" ]; then + echo "SHA256 mismatch for gfortran.tar.gz" + echo "expected: $KNOWN_SHA256" + echo "got: $GFORTRAN_SHA256" + exit 1 + fi + + mkdir -p gfortran_installed/ + tar -xv -C gfortran_installed/ -f gfortran.tar.gz + + export FC=$(pwd)/gfortran_installed/gfortran-darwin-x86_64-native/bin/gfortran + export PATH=$(pwd)/gfortran_installed/gfortran-darwin-x86_64-native/bin:$PATH + + # link libgfortran.5.dylib, libgfortran.dylib, libquadmath.0.dylib, libquadmath.dylib, libgcc_s.1.dylib, libgcc_s.1.1.dylib + # and place them in $HOME/.local/lib, and then change rpath + # to $HOME/.local/lib + + mkdir -p $HOME/.local/lib + lib_dir=$(pwd)/gfortran_installed/gfortran-darwin-x86_64-native/lib + for lib in libgfortran.5.dylib libgfortran.dylib libquadmath.0.dylib libquadmath.dylib libgcc_s.1.dylib libgcc_s.1.1.dylib; do + cp $lib_dir/$lib $HOME/.local/lib/ + install_name_tool -id $HOME/.local/lib/$lib $HOME/.local/lib/$lib + done + + export SDKROOT=${SDKROOT:-$(xcrun --show-sdk-path)} + + python -m pip install cmake wget + python scripts/install_KLU_Sundials.py + CIBW_BEFORE_BUILD_MACOS: python -m pip install --upgrade pip cmake casadi setuptools wheel delocate CIBW_REPAIR_WHEEL_COMMAND_MACOS: delocate-listdeps {wheel} && delocate-wheel -v -w {dest_dir} {wheel} CIBW_TEST_COMMAND: python -c "import pybamm; pybamm.IDAKLUSolver()" @@ -162,18 +192,57 @@ jobs: - name: Clone pybind11 repo (no history) run: git clone --depth 1 --branch v2.11.1 https://github.com/pybind/pybind11.git - - name: Install SuiteSparse and SUNDIALS on macOS - run: \| - brew install graphviz libomp - brew reinstall gcc - python -m pip install cmake pipx - python scripts/install_KLU_Sundials.py - - name: Build wheels on macOS arm64 - run: python -m pipx run cibuildwheel --output-dir wheelhouse + run: \| + python -m pip install cibuildwheel + python -m cibuildwheel --output-dir wheelhouse env: - CIBW_BEFORE_BUILD: python -m pip install cmake casadi setuptools wheel delocate - CIBW_REPAIR_WHEEL_COMMAND: delocate-listdeps {wheel} && delocate-wheel -v -w {dest_dir} {wheel} + MACOSX_DEPLOYMENT_TARGET: 11.0 + # Sourced from + # License: BSD-3-Clause + # https://github.com/scipy/scipy/blob/f2d4775e7762fad984f8f0acd8227c725ff21630/tools/wheels/cibw_before_build_macos.sh#L23-L49 + CIBW_BEFORE_ALL_MACOS: \| + set -e -x + + # download gfortran with proper macos minimum version (11.0) + curl -L https://github.com/isuruf/gcc/releases/download/gcc-11.3.0-2/gfortran-darwin-arm64-native.tar.gz -o gfortran.tar.gz + + GFORTRAN_SHA256=$(shasum --algorithm 256 gfortran.tar.gz) + KNOWN_SHA256="84364eee32ba843d883fb8124867e2bf61a0cd73b6416d9897ceff7b85a24604 gfortran.tar.gz" + if [ "$GFORTRAN_SHA256" != "$KNOWN_SHA256" ]; then + echo "SHA256 mismatch for gfortran.tar.gz" + echo "expected: $KNOWN_SHA256" + echo "got: $GFORTRAN_SHA256" + exit 1 + fi + + mkdir -p gfortran_installed/ + tar -xv -C gfortran_installed/ -f gfortran.tar.gz + + export FC=$(pwd)/gfortran_installed/gfortran-darwin-arm64-native/bin/gfortran + export PATH=$(pwd)/gfortran_installed/gfortran-darwin-arm64-native/bin:$PATH + + # link libgfortran.5.dylib, libgfortran.dylib, libquadmath.0.dylib, libquadmath.dylib, libgcc_s.1.1.dylib + # and place them in $HOME/.local/lib, and then change rpath + # note: libgcc_s.1.dylib not present for arm64, skip for now + # to $HOME/.local/lib + + mkdir -p $HOME/.local/lib + lib_dir=$(pwd)/gfortran_installed/gfortran-darwin-arm64-native/lib + for lib in libgfortran.5.dylib libgfortran.dylib libquadmath.0.dylib libquadmath.dylib libgcc_s.1.1.dylib; do + cp $lib_dir/$lib $HOME/.local/lib/ + install_name_tool -id $HOME/.local/lib/$lib $HOME/.local/lib/$lib + done + + export SDKROOT=${SDKROOT:-$(xcrun --show-sdk-path)} + + python -m pip install cmake wget + python scripts/install_KLU_Sundials.py + CIBW_BEFORE_BUILD_MACOS: python -m pip install --upgrade pip cmake casadi setuptools wheel delocate + # Use higher macOS target for now: https://github.com/casadi/casadi/issues/3698 + CIBW_REPAIR_WHEEL_COMMAND_MACOS: \| + delocate-listdeps {wheel} && delocate-wheel -v -w {dest_dir} {wheel} --require-target-macos-version 11.1 + for file in {dest_dir}/.whl; do mv "$file" "${file//macosx_11_1/macosx_11_0}"; done CIBW_TEST_COMMAND: python -c "import pybamm; pybamm.IDAKLUSolver()" - name: Upload wheels for macOS arm64 diff --git a/.gitignore b/.gitignore index 03750e18b2..6f1f201617 100644 --- a/.gitignore +++ b/.gitignore @@ -137,3 +137,7 @@ results/ # tests test_callback.log + +# openmp downloads +install_KLU_Sundials/openmp- +install_KLU_Sundials/usr/ diff --git a/.lycheeignore b/.lycheeignore index 338303b4ba..1543b69f03 100644 --- a/.lycheeignore +++ b/.lycheeignore @@ -1,6 +1,7 @@ # a list of links/files to be ignored by lychee link checker (see workflow file) https://github.com/DrTimothyAldenDavis/SuiteSparse/archive/v%7BSUITESPARSE_VERSION%7D.tar.gz https://github.com/LLNL/sundials/releases/download/v%7BSUNDIALS_VERSION%7D/sundials-%7BSUNDIALS_VERSION%7D.tar.gz +https://mac.r-project.org/openmp/openmp-%7BOPENMP_VERSION%7D-darwin20-Release.tar.gz # Errors in docs/source/user_guide/getting_started.md file:///home/runner/work/PyBaMM/PyBaMM/docs/source/user_guide/api_docs diff --git a/docs/source/user_guide/installation/install-from-source.rst b/docs/source/user_guide/installation/install-from-source.rst index ea664b4a5b..a2a7637c9e 100644 --- a/docs/source/user_guide/installation/install-from-source.rst +++ b/docs/source/user_guide/installation/install-from-source.rst @@ -43,11 +43,11 @@ You can install the above with Where ``X`` is the version sub-number. -.. tab:: MacOS +.. tab:: macOS .. code:: bash - brew install python openblas gcc gfortran graphviz libomp cmake pandoc + brew install python openblas gcc gfortran graphviz cmake pandoc .. note:: @@ -82,8 +82,9 @@ If you are running windows, you can simply skip this section and jump to :ref:`p # in the PyBaMM/ directory nox -s pybamm-requires -This will download, compile and install the SuiteSparse and SUNDIALS libraries. -Both libraries are installed in ``~/.local``. +This will download, compile and install the SuiteSparse and SUNDIALS (with OpenMP) libraries +and the ``pybind11`` headers. +SuiteSparse and SUNDIALS are installed in ``~/.local`` by default. For users requiring more control over the installation process, the ``pybamm-requires`` session supports additional command-line arguments: @@ -110,7 +111,7 @@ If you'd rather do things yourself, 1. Make sure you have CMake installed 2. Compile and install SuiteSparse (PyBaMM only requires the ``KLU`` component). -3. Compile and install SUNDIALS. +3. Compile and install SUNDIALS with `OpenMP support <https://mac.r-project.org/openmp/>`_. 4. Clone the pybind11 repository in the ``PyBaMM/`` directory (make sure the directory is named ``pybind11``). @@ -315,12 +316,12 @@ source files to your current directory. ``ValueError: Integrator name ida does not exist``, or ``ValueError: Integrator name cvode does not exist``. -Solution: This could mean that you have not installed -``scikits.odes`` correctly, check the instructions given above and make +Solution: This could mean that you have not linked to +SUNDIALS correctly, check the instructions given above and make sure each command was successful. One possibility is that you have not set your ``LD_LIBRARY_PATH`` to -point to the sundials library, type ``echo $LD_LIBRARY_PATH`` and make +point to the SUNDIALS library, type ``echo $LD_LIBRARY_PATH`` and make sure one of the directories printed out corresponds to where the SUNDIALS libraries are located. diff --git a/scripts/install_KLU_Sundials.py b/scripts/install_KLU_Sundials.py index 6224e40776..c0e046c48d 100755 --- a/scripts/install_KLU_Sundials.py +++ b/scripts/install_KLU_Sundials.py @@ -14,18 +14,35 @@ SUITESPARSE_VERSION = "6.0.3" SUNDIALS_VERSION = "6.5.0" + SUITESPARSE_URL = f"https://github.com/DrTimothyAldenDavis/SuiteSparse/archive/v{SUITESPARSE_VERSION}.tar.gz" SUNDIALS_URL = f"https://github.com/LLNL/sundials/releases/download/v{SUNDIALS_VERSION}/sundials-{SUNDIALS_VERSION}.tar.gz" + SUITESPARSE_CHECKSUM = ( "7111b505c1207f6f4bd0be9740d0b2897e1146b845d73787df07901b4f5c1fb7" ) SUNDIALS_CHECKSUM = "4e0b998dff292a2617e179609b539b511eb80836f5faacf800e688a886288502" + +# universal binaries for macOS 11.0 and later; sourced from https://mac.r-project.org/openmp/ +OPENMP_VERSION = "16.0.4" +OPENMP_URL = ( + f"https://mac.r-project.org/openmp/openmp-{OPENMP_VERSION}-darwin20-Release.tar.gz" +) +OPENMP_CHECKSUM = "a763f0bdc9115c4f4933accc81f514f3087d56d6528778f38419c2a0d2231972" + + DEFAULT_INSTALL_DIR = os.path.join(os.getenv("HOME"), ".local") def safe_remove_dir(path): - if os.path.exists(path): - shutil.rmtree(path) + """Remove a directory or file if it exists.""" + try: + if os.path.isfile(path): + os.remove(path) + elif os.path.isdir(path): + shutil.rmtree(path) + except Exception as e: + print(f"Error while removing {path}: {e}") def install_suitesparse(download_dir): @@ -51,11 +68,11 @@ def install_suitesparse(download_dir): env = os.environ.copy() for libdir in ["SuiteSparse_config", "AMD", "COLAMD", "BTF", "KLU"]: build_dir = os.path.join(suitesparse_src, libdir) - # We want to ensure that libsuitesparseconfig.dylib is not repeated in - # multiple paths at the time of wheel repair. Therefore, it should not be - # built with an RPATH since it is copied to the install prefix. + # Set an RPATH in order for libsuitesparseconfig.dylib to find libomp.dylib if libdir == "SuiteSparse_config": - env["CMAKE_OPTIONS"] = f"-DCMAKE_INSTALL_PREFIX={install_dir}" + env["CMAKE_OPTIONS"] = ( + f"-DCMAKE_INSTALL_PREFIX={install_dir} -DCMAKE_INSTALL_RPATH={install_dir}/lib" + ) else: # For AMD, COLAMD, BTF and KLU; do not set a BUILD RPATH but use an # INSTALL RPATH in order to ensure that the dynamic libraries are found @@ -93,20 +110,9 @@ def install_sundials(download_dir, install_dir): # try to find OpenMP on mac if platform.system() == "Darwin": # flags to find OpenMP on mac - if platform.processor() == "arm": - OpenMP_C_FLAGS = ( - "-Xpreprocessor -fopenmp -I/opt/homebrew/opt/libomp/include" - ) - OpenMP_C_LIB_NAMES = "omp" - OpenMP_omp_LIBRARY = "/opt/homebrew/opt/libomp/lib/libomp.dylib" - elif platform.processor() == "i386": - OpenMP_C_FLAGS = "-Xpreprocessor -fopenmp -I/usr/local/opt/libomp/include" - OpenMP_C_LIB_NAMES = "omp" - OpenMP_omp_LIBRARY = "/usr/local/opt/libomp/lib/libomp.dylib" - else: - raise NotImplementedError( - f"Unsupported processor architecture: {platform.processor()}. Only 'arm' and 'i386' architectures are supported." - ) + OpenMP_C_FLAGS = f"-Xpreprocessor -fopenmp -lomp -L{os.path.join(KLU_LIBRARY_DIR)} -I{os.path.join(KLU_INCLUDE_DIR)}" + OpenMP_C_LIB_NAMES = "omp" + OpenMP_omp_LIBRARY = os.path.join(KLU_LIBRARY_DIR, "libomp.dylib") cmake_args += [ "-DOpenMP_C_FLAGS=" + OpenMP_C_FLAGS, @@ -130,6 +136,47 @@ def install_sundials(download_dir, install_dir): subprocess.run(make_cmd, cwd=build_dir, check=True) +# Relevant for macOS only because recent Xcode Clang versions do not include OpenMP headers. +# Other compilers (e.g. GCC) include the OpenMP specification by default. +def set_up_openmp(download_dir, install_dir): + print("-" * 10, "Extracting OpenMP archive", "-" * 40) + + openmp_dir = f"openmp-{OPENMP_VERSION}" + openmp_src = os.path.join(download_dir, openmp_dir) + + # extract OpenMP archive + with tarfile.open( + os.path.join(download_dir, f"{openmp_dir}-darwin20-Release.tar.gz") + ) as tar: + tar.extractall(openmp_src) + + # create directories + os.makedirs(os.path.join(install_dir, "lib"), exist_ok=True) + os.makedirs(os.path.join(install_dir, "include"), exist_ok=True) + + # copy files + shutil.copy( + os.path.join(openmp_src, "usr", "local", "lib", "libomp.dylib"), + os.path.join(install_dir, "lib"), + ) + for file in os.listdir(os.path.join(openmp_src, "usr", "local", "include")): + shutil.copy( + os.path.join(openmp_src, "usr", "local", "include", file), + os.path.join(install_dir, "include"), + ) + + # fix rpath; for some reason the downloaded dylib has an absolute path + # to /usr/local/lib/, so use self-referential rpath + subprocess.check_call( + [ + "install_name_tool", + "-id", + "@rpath/libomp.dylib", + f"{os.path.join(install_dir, 'lib', 'libomp.dylib')}", + ] + ) + + def check_libraries_installed(install_dir): # Define the directories to check for SUNDIALS and SuiteSparse libraries lib_dirs = [install_dir] @@ -177,7 +224,7 @@ def check_libraries_installed(install_dir): suitesparse_files = [file + ".dylib" for file in suitesparse_files] else: raise NotImplementedError( - f"Unsupported operating system: {platform.system()}. This script currently supports only Linux and macOS." + f"Unsupported operating system: {platform.system()}. This script supports only Linux and macOS." ) suitesparse_lib_found = True @@ -199,6 +246,16 @@ def check_libraries_installed(install_dir): return sundials_lib_found, suitesparse_lib_found +def check_openmp_installed_on_macos(install_dir): + openmp_lib_found = isfile(join(install_dir, "lib", "libomp.dylib")) + openmp_headers_found = isfile(join(install_dir, "include", "omp.h")) + if not openmp_lib_found or not openmp_headers_found: + print("libomp.dylib or omp.h not found. Proceeding with OpenMP installation.") + else: + print(f"libomp.dylib and omp.h found in {install_dir}.") + return openmp_lib_found + + def calculate_sha256(file_path): sha256_hash = hashlib.sha256() with open(file_path, "rb") as f: @@ -274,7 +331,7 @@ def parallel_download(urls, download_dir): # Get installation location parser = argparse.ArgumentParser( - description="Download, compile and install Sundials and SuiteSparse." + description="Download, compile and install SUNDIALS and SuiteSparse." ) parser.add_argument( "--force", @@ -296,10 +353,20 @@ def parallel_download(urls, download_dir): safe_remove_dir(os.path.join(download_dir, "build_sundials")) safe_remove_dir(os.path.join(download_dir, f"SuiteSparse-{SUITESPARSE_VERSION}")) safe_remove_dir(os.path.join(download_dir, f"sundials-{SUNDIALS_VERSION}")) - sundials_found, suitesparse_found = False, False + if platform.system() == "Darwin": + safe_remove_dir(os.path.join(install_dir, "lib", "libomp.dylib")) + safe_remove_dir(os.path.join(install_dir, "include", "omp.h")) + sundials_found, suitesparse_found, openmp_found = False, False, False + else: + sundials_found, suitesparse_found = False, False else: # Check whether the libraries are installed - sundials_found, suitesparse_found = check_libraries_installed(install_dir) + if platform.system() == "Darwin": + sundials_found, suitesparse_found = check_libraries_installed(install_dir) + openmp_found = check_openmp_installed_on_macos(install_dir) + else: # Linux + sundials_found, suitesparse_found = check_libraries_installed(install_dir) + if __name__ == "__main__": # Determine which libraries to download based on whether they were found @@ -312,12 +379,22 @@ def parallel_download(urls, download_dir): ], download_dir, ) + + if platform.system() == "Darwin" and not openmp_found: + download_extract_library(OPENMP_URL, OPENMP_CHECKSUM, download_dir) + set_up_openmp(download_dir, install_dir) + install_suitesparse(download_dir) install_sundials(download_dir, install_dir) + else: if not sundials_found: # Only SUNDIALS is missing, download and install it parallel_download([(SUNDIALS_URL, SUNDIALS_CHECKSUM)], download_dir) + if platform.system() == "Darwin" and not openmp_found: + download_extract_library(OPENMP_URL, OPENMP_CHECKSUM, download_dir) + set_up_openmp(download_dir, install_dir) + # openmp needed for SUNDIALS on macOS install_sundials(download_dir, install_dir) if not suitesparse_found: # Only SuiteSparse is missing, download and install it	Fortnightly build for wheels failed / PyBaMM import kills process on macOS M-series The build is failing with the following logs - https://github.com/pybamm-team/PyBaMM/actions/runs/9093914939 Note: after #4092, this is failing for the self-hosted runners as well after the wheels were fixed: https://github.com/pybamm-team/PyBaMM/actions/runs/9184757819/job/25257629262 most likely due to a SIGKILL or some sort of memory leak.	Something is up either on the `cibuildwheel` side or when we build SuiteSparse and SUNDIALS. Looks like `MACOSX_DEPLOYMENT_TARGET` wasn't set for the latter. I shall take a look at this.	2024-05-15T13:18:18	[]	[]
pybamm-team/PyBaMM	pybamm-team__PyBaMM-4083	b972782ab9a8e5b308c29c04a7212fddcb71a273	diff --git a/pybamm/models/submodels/thermal/base_thermal.py b/pybamm/models/submodels/thermal/base_thermal.py index 757181655b..65a3f72a59 100644 --- a/pybamm/models/submodels/thermal/base_thermal.py +++ b/pybamm/models/submodels/thermal/base_thermal.py @@ -138,7 +138,7 @@ def _get_standard_coupled_variables(self, variables): num_phases = int(self.options.negative["particle phases"]) phase_names = [""] if num_phases > 1: - phase_names = ["primary", "secondary"] + phase_names = ["primary ", "secondary "] if self.options.electrode_types["negative"] == "planar": Q_rxn_n = pybamm.FullBroadcast(	[Bug]: thermal option does not work in 2 phase dfn model ### PyBaMM Version 24.1 ### Python Version 3.9 ### Describe the bug thermal option does not work in 2 phase dfn model ### Steps to Reproduce model = pybamm.lithium_ion.DFN( { "particle phases": ("2", "1"), "open-circuit potential": (("single", "current sigmoid"), "single"), "thermal": "lumped", "cell geometry": "pouch", } ) ### Relevant log output ```shell ModelError: Missing variable for submodel 'thermal': 'Negative electrode primaryvolumetric interfacial current density [A.m-3]'. Check the selected submodels provide all of the required variables. ```	Adding a space after {phase} in base_thermal.py _get_standard_coupled_variables() seems to solve the problem Thanks, the way to fix this is to replace [this line](https://github.com/pybamm-team/PyBaMM/blob/1dae0bddd25c2b998ef6b635313c7bd0e6d8a33c/pybamm/models/submodels/thermal/base_thermal.py#L141) with ` phase_names = ["primary ", "secondary "]` so that you don't end up with an extra space in the single phase model. Could you open a PR with a fix? I will do a quick patch for this so we can get this into the release	2024-05-13T15:44:20	[]	[]
pybamm-team/PyBaMM	pybamm-team__PyBaMM-4020	5d22165e7d97ae48ad2e567d8a7742cfbde1eb08	diff --git a/CHANGELOG.md b/CHANGELOG.md index ea458b4b60..a8d7d155de 100644 --- a/CHANGELOG.md +++ b/CHANGELOG.md @@ -2,6 +2,7 @@ ## Features +- Added `plot_thermal_components` to plot the contributions to the total heat generation in a battery ([#4021](https://github.com/pybamm-team/PyBaMM/pull/4021)) - Added functions for normal probability density function (`pybamm.normal_pdf`) and cumulative distribution function (`pybamm.normal_cdf`) ([#3999](https://github.com/pybamm-team/PyBaMM/pull/3999)) - Updates multiprocess `Pool` in `BaseSolver.solve()` to be constructed with context `fork`. Adds small example for multiprocess inputs. ([#3974](https://github.com/pybamm-team/PyBaMM/pull/3974)) - Added custom experiment steps ([#3835](https://github.com/pybamm-team/PyBaMM/pull/3835)) @@ -14,6 +15,7 @@ ## Bug Fixes +- Set the `remove_independent_variables_from_rhs` to `False` by default, and moved the option from `Discretisation.process_model` to `Discretisation.__init__`. This fixes a bug related to the discharge capacity, but may make the simulation slower in some cases. To set the option to `True`, use `Simulation(..., discretisation_kwargs={"remove_independent_variables_from_rhs": True})`. ([#4020](https://github.com/pybamm-team/PyBaMM/pull/4020)) - Fixed a bug where independent variables were removed from models even if they appeared in events ([#4019](https://github.com/pybamm-team/PyBaMM/pull/4019)) - Fix bug with upwind and downwind schemes producing the wrong discretised system ([#3979](https://github.com/pybamm-team/PyBaMM/pull/3979)) - Allow evaluation of an `Interpolant` object with a number ([#3932](https://github.com/pybamm-team/PyBaMM/pull/3932)) @@ -33,6 +35,7 @@ ## Breaking changes +- Removed `check_model` argument from `Simulation.solve`. To change the `check_model` option, use `Simulation(..., discretisation_kwargs={"check_model": False})`. ([#4020](https://github.com/pybamm-team/PyBaMM/pull/4020)) - Removed multiple Docker images. Here on, a single Docker image tagged `pybamm/pybamm:latest` will be provided with both solvers (`IDAKLU` and `JAX`) pre-installed. ([#3992](https://github.com/pybamm-team/PyBaMM/pull/3992)) - Removed support for Python 3.8 ([#3961](https://github.com/pybamm-team/PyBaMM/pull/3961)) - Renamed "ocp_soc_0_dimensional" to "ocp_soc_0" and "ocp_soc_100_dimensional" to "ocp_soc_100" ([#3942](https://github.com/pybamm-team/PyBaMM/pull/3942)) diff --git a/pybamm/batch_study.py b/pybamm/batch_study.py index 1503c3048c..e854c94e00 100644 --- a/pybamm/batch_study.py +++ b/pybamm/batch_study.py @@ -102,7 +102,6 @@ def solve( self, t_eval=None, solver=None, - check_model=True, save_at_cycles=None, calc_esoh=True, starting_solution=None, @@ -158,7 +157,6 @@ def solve( sol = sim.solve( t_eval, solver, - check_model, save_at_cycles, calc_esoh, starting_solution, diff --git a/pybamm/discretisations/discretisation.py b/pybamm/discretisations/discretisation.py index 92e9e4e24f..d61d8a4d60 100644 --- a/pybamm/discretisations/discretisation.py +++ b/pybamm/discretisations/discretisation.py @@ -26,14 +26,32 @@ class Discretisation: Parameters ---------- mesh : pybamm.Mesh - contains all submeshes to be used on each domain + contains all submeshes to be used on each domain spatial_methods : dict - a dictionary of the spatial methods to be used on each - domain. The keys correspond to the model domains and the - values to the spatial method. + a dictionary of the spatial methods to be used on each + domain. The keys correspond to the model domains and the + values to the spatial method. + check_model : bool, optional + If True, model checks are performed after discretisation. For large + systems these checks can be slow, so can be skipped by setting this + option to False. When developing, testing or debugging it is recommended + to leave this option as True as it may help to identify any errors. + Default is True. + remove_independent_variables_from_rhs : bool, optional + If True, model checks to see whether any variables from the RHS are used + in any other equation. If a variable meets all of the following criteria + (not used anywhere in the model, len(rhs)>1), then the variable + is moved to be explicitly integrated when called by the solution object. + Default is False. """ - def __init__(self, mesh=None, spatial_methods=None): + def __init__( + self, + mesh=None, + spatial_methods=None, + check_model=True, + remove_independent_variables_from_rhs=False, + ): self._mesh = mesh if mesh is None: self._spatial_methods = {} @@ -60,6 +78,10 @@ def __init__(self, mesh=None, spatial_methods=None): self._bcs = {} self.y_slices = {} self._discretised_symbols = {} + self._check_model_flag = check_model + self._remove_independent_variables_from_rhs_flag = ( + remove_independent_variables_from_rhs + ) @property def mesh(self): @@ -90,13 +112,7 @@ def bcs(self, value): # reset discretised_symbols self._discretised_symbols = {} - def process_model( - self, - model, - inplace=True, - check_model=True, - remove_independent_variables_from_rhs=True, - ): + def process_model(self, model, inplace=True): """ Discretise a model. Currently inplace, could be changed to return a new model. @@ -108,18 +124,6 @@ def process_model( inplace : bool, optional If True, discretise the model in place. Otherwise, return a new discretised model. Default is True. - check_model : bool, optional - If True, model checks are performed after discretisation. For large - systems these checks can be slow, so can be skipped by setting this - option to False. When developing, testing or debugging it is recommended - to leave this option as True as it may help to identify any errors. - Default is True. - remove_independent_variables_from_rhs : bool, optional - If True, model checks to see whether any variables from the RHS are used - in any other equation. If a variable meets all of the following criteria - (not used anywhere in the model, len(rhs)>1), then the variable - is moved to be explicitly integrated when called by the solution object. - Default is True. Returns ------- @@ -158,7 +162,7 @@ def process_model( # set variables (we require the full variable not just id) # Search Equations for Independence - if remove_independent_variables_from_rhs: + if self._remove_independent_variables_from_rhs_flag: model = self.remove_independent_variables_from_rhs(model) variables = list(model.rhs.keys()) + list(model.algebraic.keys()) # Find those RHS's that are constant @@ -240,7 +244,7 @@ def process_model( model_disc._geometry = getattr(self.mesh, "_geometry", None) # Check that resulting model makes sense - if check_model: + if self._check_model_flag: pybamm.logger.verbose(f"Performing model checks for {model.name}") self.check_model(model_disc) @@ -1123,6 +1127,7 @@ def remove_independent_variables_from_rhs(self, model): # in variables twice under different names for key in model.variables: if model.variables[key] == var: + print("here") model.variables[key] = model.variables[var.name] del model.rhs[var] del model.initial_conditions[var] diff --git a/pybamm/simulation.py b/pybamm/simulation.py index 060d1cdd8c..a81f7b62b2 100644 --- a/pybamm/simulation.py +++ b/pybamm/simulation.py @@ -63,6 +63,9 @@ class Simulation: A list of variables to plot automatically C_rate: float (optional) The C-rate at which you would like to run a constant current (dis)charge. + discretisation_kwargs: dict (optional) + Any keyword arguments to pass to the Discretisation class. + See :class:`pybamm.Discretisation` for details. """ def __init__( @@ -77,6 +80,7 @@ def __init__( solver=None, output_variables=None, C_rate=None, + discretisation_kwargs=None, ): self._parameter_values = parameter_values or model.default_parameter_values self._unprocessed_parameter_values = self._parameter_values @@ -126,6 +130,7 @@ def __init__( self._spatial_methods = spatial_methods or self._model.default_spatial_methods self._solver = solver or self._model.default_solver self._output_variables = output_variables + self._discretisation_kwargs = discretisation_kwargs or {} # Initialize empty built states self._model_with_set_params = None @@ -268,7 +273,7 @@ def set_initial_soc(self, initial_soc, inputs=None): # Save solved initial SOC in case we need to re-build the model self._built_initial_soc = initial_soc - def build(self, check_model=True, initial_soc=None, inputs=None): + def build(self, initial_soc=None, inputs=None): """ A method to build the model into a system of matrices and vectors suitable for performing numerical computations. If the model has already been built or @@ -278,13 +283,12 @@ def build(self, check_model=True, initial_soc=None, inputs=None): Parameters ---------- - check_model : bool, optional - If True, model checks are performed after discretisation (see - :meth:`pybamm.Discretisation.process_model`). Default is True. initial_soc : float, optional Initial State of Charge (SOC) for the simulation. Must be between 0 and 1. If given, overwrites the initial concentrations provided in the parameter set. + inputs : dict, optional + A dictionary of input parameters to pass to the model when solving. """ if initial_soc is not None: self.set_initial_soc(initial_soc, inputs=inputs) @@ -297,14 +301,16 @@ def build(self, check_model=True, initial_soc=None, inputs=None): else: self.set_parameters() self._mesh = pybamm.Mesh(self._geometry, self._submesh_types, self._var_pts) - self._disc = pybamm.Discretisation(self._mesh, self._spatial_methods) + self._disc = pybamm.Discretisation( + self._mesh, self._spatial_methods, self._discretisation_kwargs + ) self._built_model = self._disc.process_model( - self._model_with_set_params, inplace=False, check_model=check_model + self._model_with_set_params, inplace=False ) # rebuilt model so clear solver setup self._solver._model_set_up = {} - def build_for_experiment(self, check_model=True, initial_soc=None, inputs=None): + def build_for_experiment(self, initial_soc=None, inputs=None): """ Similar to :meth:`Simulation.build`, but for the case of simulating an experiment, where there may be several models and solvers to build. @@ -322,7 +328,9 @@ def build_for_experiment(self, check_model=True, initial_soc=None, inputs=None): self._parameter_values.process_geometry(self._geometry) # Only needs to set up mesh and discretisation once self._mesh = pybamm.Mesh(self._geometry, self._submesh_types, self._var_pts) - self._disc = pybamm.Discretisation(self._mesh, self._spatial_methods) + self._disc = pybamm.Discretisation( + self._mesh, self._spatial_methods, self._discretisation_kwargs + ) # Process all the different models self.steps_to_built_models = {} self.steps_to_built_solvers = {} @@ -333,7 +341,7 @@ def build_for_experiment(self, check_model=True, initial_soc=None, inputs=None): # It's ok to modify the model with set parameters in place as it's # not returned anywhere built_model = self._disc.process_model( - model_with_set_params, inplace=True, check_model=check_model + model_with_set_params, inplace=True ) solver = self._solver.copy() self.steps_to_built_solvers[step] = solver @@ -343,7 +351,6 @@ def solve( self, t_eval=None, solver=None, - check_model=True, save_at_cycles=None, calc_esoh=True, starting_solution=None, @@ -377,9 +384,6 @@ def solve( provided in the data. solver : :class:`pybamm.BaseSolver`, optional The solver to use to solve the model. If None, Simulation.solver is used - check_model : bool, optional - If True, model checks are performed after discretisation (see - :meth:`pybamm.Discretisation.process_model`). Default is True. save_at_cycles : int or list of ints, optional Which cycles to save the full sub-solutions for. If None, all cycles are saved. If int, every multiple of save_at_cycles is saved. If list, every @@ -416,7 +420,7 @@ def solve( inputs = inputs or {} if self.operating_mode in ["without experiment", "drive cycle"]: - self.build(check_model=check_model, initial_soc=initial_soc, inputs=inputs) + self.build(initial_soc=initial_soc, inputs=inputs) if save_at_cycles is not None: raise ValueError( "'save_at_cycles' option can only be used if simulating an " @@ -493,9 +497,7 @@ def solve( elif self.operating_mode == "with experiment": callbacks.on_experiment_start(logs) - self.build_for_experiment( - check_model=check_model, initial_soc=initial_soc, inputs=inputs - ) + self.build_for_experiment(initial_soc=initial_soc, inputs=inputs) if t_eval is not None: pybamm.logger.warning( "Ignoring t_eval as solution times are specified by the experiment"	"Throughput capacity [A.h]" is different for sol.cycles[0] and sol.cycles[0].steps[0] ### Discussed in https://github.com/pybamm-team/PyBaMM/discussions/2523 <div type='discussions-op-text'> <sup>Originally posted by RuiheLi December 5, 2022</sup> Dear PyBaMM, I am using PyBaMM 22.9 and Python 3.9.6. I find when using that set_initial_conditions_from, "Throughput capacity [A.h]" is different for sol.cycles[0] and sol.cycles[0].steps[0] for a later solution. Figure below: ![image](https://user-images.githubusercontent.com/84007676/205644495-558eabbe-1f36-4dd5-9935-44f131704896.png) Script below: ``` import pybamm import matplotlib as mpl fs=17; font = {'family' : 'DejaVu Sans','size' : fs} mpl.rc('font', *font) V_max = 4.0; V_min = 2.7; model_0 = pybamm.lithium_ion.DFN(options={ "calculate discharge energy":"true", },) exp_AGE_text = [(f"Discharge at 1 C until {V_min} V", f"Charge at 1 C until {V_max} V", ), ] Experiment = pybamm.Experiment( exp_AGE_text 1) Para_0 = pybamm.ParameterValues("OKane2022" ) Sim_0= pybamm.Simulation( model_0, experiment = Experiment, parameter_values=Para_0, solver = pybamm.CasadiSolver(),) Sol_0 = Sim_0.solve(calc_esoh=False,) model_1 = model_0.set_initial_conditions_from(Sol_0, inplace=False) Sim_1= pybamm.Simulation( model_1, experiment = Experiment, parameter_values=Para_0, solver = pybamm.CasadiSolver(),) Sol_1 = Sim_1.solve(calc_esoh=False,) model_2 = model_0.set_initial_conditions_from(Sol_1, inplace=False) Sim_2= pybamm.Simulation( model_2, experiment = Experiment, parameter_values=Para_0, solver = pybamm.CasadiSolver(),) Sol_2 = Sim_2.solve(calc_esoh=False,) label = ["Sol_0","Sol_1","Sol_2",] output_variables3 = [ "Throughput energy [W.h]", "Throughput capacity [A.h]", "Terminal voltage [V]", ] quick_plot = pybamm.QuickPlot( [Sol_0,Sol_1,Sol_2,], output_variables3,label,variable_limits='fixed', time_unit='hours',n_rows=1) #figsize = (18,12), quick_plot.dynamic_plot() label = ["Sol_2 whole","Sol_2 cycles","Sol_2 steps",] quick_plot = pybamm.QuickPlot( [Sol_2,Sol_2.cycles[0],Sol_2.cycles[0].steps[0]], output_variables3,label,variable_limits='fixed', time_unit='hours',n_rows=1) #figsize = (18,12), quick_plot.dynamic_plot() ``` Do you have any ideas on this? Big thanks! </div>	Basically the problem is that `initial_value` for `ExplicitTimeIntegral` objects, which is used here https://github.com/pybamm-team/PyBaMM/blob/b072d22f1344f062fc84ca44e0c540873669b6f7/pybamm/solvers/solution.py#L472 is not updated properly when doing `solution.last_state`, which is used here https://github.com/pybamm-team/PyBaMM/blob/b072d22f1344f062fc84ca44e0c540873669b6f7/pybamm/models/base_model.py#L596 To solve this properly will require reformatting how we do `ExplicitTimeIntegral`, which in turn will depend on other improvements to the backend that we are working on. In the meantime, what you could do is set this to `remove_independent_variables_from_rhs=False` in your branch https://github.com/pybamm-team/PyBaMM/blob/b072d22f1344f062fc84ca44e0c540873669b6f7/pybamm/discretisations/discretisation.py#L102 (but note that it might take longer to solve the model as a result) Possibly related to this, I encountered a similar issue in a much simpler problem (free fall problem) ```python3 # Create model model = pybamm.BaseModel() u = pybamm.Variable("u") v = pybamm.Variable("v") model.rhs = {u: v, v: -1} model.initial_conditions = {u: 2, v: 0} model.variables = {"u": u, "v": v} # Discretise using default discretisation disc = pybamm.Discretisation() disc.process_model(model); # Solve t_eval = np.linspace(0, 5, 30) ode_solver = pybamm.ScipySolver() solution = ode_solver.solve(model, t_eval) # Plot solution.plot(["u", "v"]) ``` It seems that after the first step, the variable `u` is set to zero and then evolves according to the equation. ![image](https://github.com/pybamm-team/PyBaMM/assets/28443643/3401215a-570b-42ff-9a70-9e0fd4b10c0a)	2024-04-17T14:42:40	[]	[]
pybamm-team/PyBaMM	pybamm-team__PyBaMM-4001	8cd1cec79134e60aed9dfa04b2f14883b9b5a8b5	diff --git a/pybamm/solvers/base_solver.py b/pybamm/solvers/base_solver.py index 28dd4bdd9d..0e60aca189 100644 --- a/pybamm/solvers/base_solver.py +++ b/pybamm/solvers/base_solver.py @@ -13,10 +13,6 @@ import pybamm from pybamm.expression_tree.binary_operators import _Heaviside -# Set context for parallel processing depending on the platform -if platform.system() == "Darwin" or platform.system() == "Linux": - mp.set_start_method("fork") - class BaseSolver: """Solve a discretised model. @@ -70,6 +66,7 @@ def __init__( self.algebraic_solver = False self._on_extrapolation = "warn" self.computed_var_fcns = {} + self._mp_context = self.get_platform_context(platform.system()) @property def root_method(self): @@ -917,7 +914,7 @@ def solve( model_inputs_list, ) else: - with mp.Pool(processes=nproc) as p: + with mp.get_context(self._mp_context).Pool(processes=nproc) as p: new_solutions = p.starmap( self._integrate, zip( @@ -1395,6 +1392,12 @@ def check_extrapolation(self, solution, events): "outside these bounds." ) + def get_platform_context(self, system_type: str): + # Set context for parallel processing depending on the platform + if system_type.lower() in ["linux", "darwin"]: + return "fork" + return "spawn" + @staticmethod def _set_up_model_inputs(model, inputs): """Set up input parameters"""	[Bug]: PyBaMM fails to import when user has already imported a library that uses `mp.set_start_method` ### PyBaMM Version develop ### Python Version 3.11 ### Describe the bug PyBaMM fails to import when importing a library that uses multiprocessing and has already set the start method to something other than fork, as a result of #3975. This should be apparent when reading the documentation for `mp.set_start_method` Pending a long-term solution, I think that #3975 should at least have `mp.set_start_method` wrapped in a try-except -- most people, most of the time, do not really need the multiprocessing functionality, and I don't think pybamm should force those users to modify their multiprocessing configuration. It can throw a warning saying multiprocessing is not available or whatever. Other solutions are welcome, of course. ### Steps to Reproduce module `thing`: ``` import multiprocessing as mp mp.set_start_method("spawn") def hello(): print("hello, world") ``` script: ``` import thing import pybamm ``` ### Relevant log output ```shell ❯ python main.py Traceback (most recent call last): File "/Users/abills/.../main.py", line 2, in <module> import pybamm File "/Users/abills/.../python3.11/site-packages/pybamm/__init__.py", line 204, in <module> from .solvers.base_solver import BaseSolver File "/Users/abills/.../python3.11/site-packages/pybamm/solvers/base_solver.py", line 18, in <module> mp.set_start_method("fork") File "/Users/abills/.../python3.11/multiprocessing/context.py", line 247, in set_start_method raise RuntimeError('context has already been set') RuntimeError: context has already been set ```	@BradyPlanden Tagging you since the bug most likely comes from #3975	2024-04-12T08:03:36	[]	[]
pybamm-team/PyBaMM	pybamm-team__PyBaMM-3988	51c1f76efabbd9a1444f472bab42c70dfba3b4f9	diff --git a/docs/source/api/util.rst b/docs/source/api/util.rst index 03e7388e3e..f187cfbabb 100644 --- a/docs/source/api/util.rst +++ b/docs/source/api/util.rst @@ -3,8 +3,6 @@ Utility functions .. autofunction:: pybamm.get_git_commit_info -.. autofunction:: pybamm.rmse - .. autofunction:: pybamm.root_dir .. autoclass:: pybamm.Timer diff --git a/pybamm/__init__.py b/pybamm/__init__.py index 68195ca279..6d6e1b855c 100644 --- a/pybamm/__init__.py +++ b/pybamm/__init__.py @@ -1,38 +1,12 @@ import sys -import os - -# -# Version info -# from pybamm.version import __version__ -# -# Constants -# -# Float format: a float can be converted to a 17 digit decimal and back without -# loss of information -FLOAT_FORMAT = "{: .17e}" -# Absolute path to the PyBaMM repo -script_path = os.path.abspath(__file__) - -from .util import root_dir - -ABSOLUTE_PATH = root_dir() -PARAMETER_PATH = [ - root_dir(), - os.getcwd(), - os.path.join(root_dir(), "pybamm", "input", "parameters"), -] - - -# # Utility classes and methods -# +from .util import root_dir from .util import Timer, TimerTime, FuzzyDict from .util import ( root_dir, - rmse, load, is_constant_and_can_evaluate, ) @@ -48,9 +22,7 @@ from .settings import settings from .citations import Citations, citations, print_citations -# # Classes for the Expression Tree -# from .expression_tree.symbol import * from .expression_tree.binary_operators import * from .expression_tree.concatenations import * @@ -88,9 +60,7 @@ from .expression_tree.operations.convert_to_casadi import CasadiConverter from .expression_tree.operations.unpack_symbols import SymbolUnpacker -# # Model classes -# from .models.base_model import BaseModel from .models.event import Event from .models.event import EventType @@ -104,9 +74,7 @@ from .models.full_battery_models import lithium_ion from .models.full_battery_models import equivalent_circuit -# # Submodel classes -# from .models.submodels.base_submodel import BaseSubModel from .models.submodels import ( @@ -132,18 +100,14 @@ from .models.submodels.interface import interface_utilisation from .models.submodels.interface import open_circuit_potential -# # Geometry -# from .geometry.geometry import Geometry from .geometry.battery_geometry import battery_geometry from .expression_tree.independent_variable import KNOWN_COORD_SYS from .geometry import standard_spatial_vars -# # Parameter classes and methods -# from .parameters.parameter_values import ParameterValues from .parameters import constants from .parameters.geometric_parameters import geometric_parameters, GeometricParameters @@ -158,9 +122,7 @@ from .parameters.size_distribution_parameters import * from .parameters.parameter_sets import parameter_sets -# # Mesh and Discretisation classes -# from .discretisations.discretisation import Discretisation from .discretisations.discretisation import has_bc_of_form from .meshes.meshes import Mesh, SubMesh, MeshGenerator @@ -181,23 +143,17 @@ UserSupplied2DSubMesh, ) -# # Serialisation -# from .models.base_model import load_model -# # Spatial Methods -# from .spatial_methods.spatial_method import SpatialMethod from .spatial_methods.zero_dimensional_method import ZeroDimensionalSpatialMethod from .spatial_methods.finite_volume import FiniteVolume from .spatial_methods.spectral_volume import SpectralVolume from .spatial_methods.scikit_finite_element import ScikitFiniteElement -# # Solver classes -# from .solvers.solution import Solution, EmptySolution, make_cycle_solution from .solvers.processed_variable import ProcessedVariable from .solvers.processed_variable_computed import ProcessedVariableComputed @@ -214,17 +170,12 @@ from .solvers.idaklu_jax import IDAKLUJax from .solvers.idaklu_solver import IDAKLUSolver, have_idaklu -# # Experiments -# from .experiment.experiment import Experiment from . import experiment from .experiment import step - -# # Plotting -# from .plotting.quick_plot import QuickPlot, close_plots, QuickPlotAxes from .plotting.plot import plot from .plotting.plot2D import plot2D @@ -232,22 +183,14 @@ from .plotting.plot_summary_variables import plot_summary_variables from .plotting.dynamic_plot import dynamic_plot -# # Simulation -# from .simulation import Simulation, load_sim, is_notebook -# # Batch Study -# from .batch_study import BatchStudy -# # Callbacks -# from . import callbacks -# # Remove any imported modules, so we don't expose them as part of pybamm -# del sys diff --git a/pybamm/util.py b/pybamm/util.py index c3d7b4e263..f37248dabd 100644 --- a/pybamm/util.py +++ b/pybamm/util.py @@ -18,7 +18,6 @@ import difflib from warnings import warn -import numpy as np import pybamm # Versions of jax and jaxlib compatible with PyBaMM. Note: these are also defined in @@ -244,16 +243,6 @@ def __eq__(self, other): return self.value == other.value -def rmse(x, y): - """ - Calculate the root-mean-square-error between two vectors x and y, ignoring NaNs - """ - # Check lengths - if len(x) != len(y): - raise ValueError("Vectors must have the same length") - return np.sqrt(np.nanmean((x - y) ** 2)) - - def load(filename): """Load a saved object""" with open(filename, "rb") as f:	pybamm.rmse. document return type and args types ### PyBaMM Version 23.9 ### Python Version 3.11.7 ### Describe the bug Minor issue for the PyBaMM training workshop ### Steps to Reproduce Read the documentation of pybamm.rmse ### Relevant log output _No response_	Let's get rid of pybamm.rmse, it doesn't fit within the scope of the package	2024-04-10T20:46:33	[]	[]
pybamm-team/PyBaMM	pybamm-team__PyBaMM-3968	09ee982ba84055ec2c56f73b4cc63742c2ca0721	diff --git a/CONTRIBUTING.md b/CONTRIBUTING.md index fc8e848bb5..aeb62b9b00 100644 --- a/CONTRIBUTING.md +++ b/CONTRIBUTING.md @@ -126,26 +126,18 @@ This allows people to (1) use PyBaMM without importing optional dependencies by Writing Tests for Optional Dependencies -Whenever a new optional dependency is added for optional functionality, it is recommended to write a corresponding unit test in `test_util.py`. This ensures that an error is raised upon the absence of said dependency. Here's an example: +Below, we list the currently available test functions to provide an overview. If you find it useful to add new test cases please do so within `tests/unit/test_util.py`. -```python -from tests import TestCase -import pybamm - - -class TestUtil(TestCase): - def test_optional_dependency(self): - # Test that an error is raised when pybtex is not available - with self.assertRaisesRegex( - ModuleNotFoundError, "Optional dependency pybtex is not available" - ): - sys.modules["pybtex"] = None - pybamm.function_using_pybtex(x, y, z) - - # Test that the function works when pybtex is available - sys.modules["pybtex"] = pybamm.util.import_optional_dependency("pybtex") - pybamm.function_using_pybtex(x, y, z) -``` +Currently, there are three functions to test what concerns optional dependencies: +- `test_import_optional_dependency` +- `test_pybamm_import` +- `test_optional_dependencies` + +The `test_import_optional_dependency` function extracts the optional dependencies installed in the setup environment, makes them unimportable (by setting them to `None` among the `sys.modules`), and tests that the `pybamm.util.import_optional_dependency` function throws a `ModuleNotFoundError` exception when their import is attempted. + +The `test_pybamm_import` function extracts the optional dependencies installed in the setup environment and makes them unimportable (by setting them to `None` among the `sys.modules`), unloads `pybamm` and its sub-modules, and finally tests that `pybamm` can be imported successfully. In fact, it is essential that the `pybamm` package is importable with only the mandatory dependencies. + +The `test_optional_dependencies` function extracts `pybamm` mandatory distribution packages and verifies that they are not present in the optional distribution packages list in `pyproject.toml`. This test is crucial for ensuring the consistency of the released package information and potential updates to dependencies during development. ## Testing diff --git a/pybamm/citations.py b/pybamm/citations.py index ff1851bfaa..ae18f9adc7 100644 --- a/pybamm/citations.py +++ b/pybamm/citations.py @@ -67,29 +67,41 @@ def read_citations(self): """Reads the citations in `pybamm.CITATIONS.bib`. Other works can be cited by passing a BibTeX citation to :meth:`register`. """ - parse_file = import_optional_dependency("pybtex.database", "parse_file") - citations_file = os.path.join(pybamm.root_dir(), "pybamm", "CITATIONS.bib") - bib_data = parse_file(citations_file, bib_format="bibtex") - for key, entry in bib_data.entries.items(): - self._add_citation(key, entry) + try: + parse_file = import_optional_dependency("pybtex.database", "parse_file") + citations_file = os.path.join(pybamm.root_dir(), "pybamm", "CITATIONS.bib") + bib_data = parse_file(citations_file, bib_format="bibtex") + for key, entry in bib_data.entries.items(): + self._add_citation(key, entry) + except ModuleNotFoundError: # pragma: no cover + pybamm.logger.warning( + "Citations could not be read because the 'pybtex' library is not installed. " + "Install 'pybamm[cite]' to enable citation reading." + ) def _add_citation(self, key, entry): """Adds `entry` to `self._all_citations` under `key`, warning the user if a previous entry is overwritten """ - Entry = import_optional_dependency("pybtex.database", "Entry") - # Check input types are correct - if not isinstance(key, str) or not isinstance(entry, Entry): - raise TypeError() - - # Warn if overwriting a previous citation - new_citation = entry.to_string("bibtex") - if key in self._all_citations and new_citation != self._all_citations[key]: - warnings.warn(f"Replacing citation for {key}", stacklevel=2) - - # Add to database - self._all_citations[key] = new_citation + try: + Entry = import_optional_dependency("pybtex.database", "Entry") + # Check input types are correct + if not isinstance(key, str) or not isinstance(entry, Entry): + raise TypeError() + + # Warn if overwriting a previous citation + new_citation = entry.to_string("bibtex") + if key in self._all_citations and new_citation != self._all_citations[key]: + warnings.warn(f"Replacing citation for {key}", stacklevel=2) + + # Add to database + self._all_citations[key] = new_citation + except ModuleNotFoundError: # pragma: no cover + pybamm.logger.warning( + f"Could not add citation for '{key}' because the 'pybtex' library is not installed. " + "Install 'pybamm[cite]' to enable adding citations." + ) def _add_citation_tag(self, key, entry): """Adds a tag for a citation key in the dict, which represents the name of the @@ -143,24 +155,32 @@ def _parse_citation(self, key): key: str A BibTeX formatted citation """ - PybtexError = import_optional_dependency("pybtex.scanner", "PybtexError") - parse_string = import_optional_dependency("pybtex.database", "parse_string") try: - # Parse string as a bibtex citation, and check that a citation was found - bib_data = parse_string(key, bib_format="bibtex") - if not bib_data.entries: - raise PybtexError("no entries found") - - # Add and register all citations - for key, entry in bib_data.entries.items(): - # Add to _all_citations dictionary - self._add_citation(key, entry) - # Add to _papers_to_cite set - self._papers_to_cite.add(key) - return - except PybtexError as error: - # Unable to parse / unknown key - raise KeyError(f"Not a bibtex citation or known citation: {key}") from error + PybtexError = import_optional_dependency("pybtex.scanner", "PybtexError") + parse_string = import_optional_dependency("pybtex.database", "parse_string") + try: + # Parse string as a bibtex citation, and check that a citation was found + bib_data = parse_string(key, bib_format="bibtex") + if not bib_data.entries: + raise PybtexError("no entries found") + + # Add and register all citations + for key, entry in bib_data.entries.items(): + # Add to _all_citations dictionary + self._add_citation(key, entry) + # Add to _papers_to_cite set + self._papers_to_cite.add(key) + return + except PybtexError as error: + # Unable to parse / unknown key + raise KeyError( + f"Not a bibtex citation or known citation: {key}" + ) from error + except ModuleNotFoundError: # pragma: no cover + pybamm.logger.warning( + f"Could not parse citation for '{key}' because the 'pybtex' library is not installed. " + "Install 'pybamm[cite]' to enable citation parsing." + ) def _tag_citations(self): """Prints the citation tags for the citations that have been registered @@ -211,38 +231,44 @@ def print(self, filename=None, output_format="text", verbose=False): """ # Parse citations that were not known keys at registration, but do not # fail if they cannot be parsed - pybtex = import_optional_dependency("pybtex") try: - for key in self._unknown_citations: - self._parse_citation(key) - except KeyError: # pragma: no cover - warnings.warn( - message=f'\nCitation with key "{key}" is invalid. Please try again\n', - category=UserWarning, - stacklevel=2, - ) - # delete the invalid citation from the set - self._unknown_citations.remove(key) - - if output_format == "text": - citations = pybtex.format_from_strings( - self._cited, style="plain", output_backend="plaintext" + pybtex = import_optional_dependency("pybtex") + try: + for key in self._unknown_citations: + self._parse_citation(key) + except KeyError: # pragma: no cover + warnings.warn( + message=f'\nCitation with key "{key}" is invalid. Please try again\n', + category=UserWarning, + stacklevel=2, + ) + # delete the invalid citation from the set + self._unknown_citations.remove(key) + + if output_format == "text": + citations = pybtex.format_from_strings( + self._cited, style="plain", output_backend="plaintext" + ) + elif output_format == "bibtex": + citations = "\n".join(self._cited) + else: + raise pybamm.OptionError( + f"Output format {output_format} not recognised." + "It should be 'text' or 'bibtex'." + ) + + if filename is None: + print(citations) + if verbose: + self._tag_citations() # pragma: no cover + else: + with open(filename, "w") as f: + f.write(citations) + except ModuleNotFoundError: # pragma: no cover + pybamm.logger.warning( + "Could not print citations because the 'pybtex' library is not installed. " + "Please, install 'pybamm[cite]' to print citations." ) - elif output_format == "bibtex": - citations = "\n".join(self._cited) - else: - raise pybamm.OptionError( - f"Output format {output_format} not recognised." - "It should be 'text' or 'bibtex'." - ) - - if filename is None: - print(citations) - if verbose: - self._tag_citations() # pragma: no cover - else: - with open(filename, "w") as f: - f.write(citations) def print_citations(filename=None, output_format="text", verbose=False):	PyBaMM import error - Cannot run anything (import broken due to `pybtex`) ### Discussed in https://github.com/pybamm-team/PyBaMM/discussions/3928 <div type='discussions-op-text'> <sup>Originally posted by abhishek-appana March 25, 2024</sup> Hello, I am running Python 3.11.8 and installed PyBaMM 24.1 using pip on a Windows system. Earlier I was using Python 3.10 and everything was fine. But I am getting the below error after upgrading Python. When I run `import pybamm`, it gives me the error: ModuleNotFoundError: Optional dependency pybtex.database is not available. See https://docs.pybamm.org/en/latest/source/user_guide/installation/index.html#optional-dependencies for more details. I tried installing pybamm using `pip install pybamm[all]`. I was still getting this error. How do I resolve this?</div> <hr> I think that this is a valid bug and it is still failing, and `import pybamm` should work without installing `pybamm[all]`. It can be noticed in the logs here: https://github.com/pybamm-team/PyBaMM/actions/runs/8461852686/job/23182296194, so I've converted this discussion to an issue instead (cc: @Saransh-cpp @arjxn-py). Perhaps this isn't being properly checked through #3892, @lorenzofavaro?	Perhaps #3886 broke this where the try-catch blocks everywhere were removed? I haven't looked at this in enough detail, but the breakage is coming from this instantiation of the `Citations` class at the end of the file. ```python citations = Citations() ``` where `__init__` tries to call the function `_reset()`, and that in-turn is trying to register the PyBaMM and NumPy citation, which will fail? Hmm, I somehow can't reproduce this locally. I'm not sure where is this originating from. Me neither I haven't looked at this again, but I think having it fail in a fresh CI environment means there is indeed a bug, even if we can't reproduce it on our own machines. I added the "high priority" label for now. Maybe a fix can be to register the PyBaMM and NumPy papers only inside the `print_citations()` method and only if they don't exist in the `_papers_to_cite` variable, so that they get cited when someone actually runs `pybamm.print_citations()`. I'm not sure if this is a complete fix, though. Let me look into this. ~I have looked into this and looks like this is failing as we merged #3932 & haven't removed the jobs to test `pybamm_install_odes`~ Oh wait, I see that they've been removed already I had a chance to look at the issue and understand the problem. Actually at the moment pybamm has a hidden mandatory dependency, namely `pybtex`. It's imported during `pybamm/citations.py:41`, that in turn trigger `pybamm/citations.py:70`: ```python parse_file = import_optional_dependency("pybtex.database", "parse_file") ``` During #3892 the error was still masked in `pybamm/citations` in the `except`: ```python try: self.read_citations() self._reset() except Exception as e: # pragma: no cover self._citation_err_msg = e ``` Beyond that, the error was not reported after for a subtle issue. The issue is that `test_pybamm_import` is done inside `tests/unit/test_util.py` and if we look at the module `tests/unit/_init__.py` we find that it imports `StandardModelTest` which in turn imports `pybamm`. When we don't have the optional libraries installed locally (e.g., from local with `pybamm.[dev]`), this import avoid the tests execution during the `nox` session (it stops before), even though it reports them as `PASSED`. In fact, if you try to run the specific tests individually (i.e. `python -m unittest tests.unit.test_util.TestUtil.test_pybamm_import`), you would get an exception before the test run. Instead, when we have the optional libraries installed (e.g. from CI with `pybamm.[all]`), we don't notice the error because digging deep into the `importlib` documentation, `importlib.reload(...)` does not reload all sub-modules. So when it initially in `StandardModelTest` imports `pybamm`, it also imports `pybtex.database` which will no longer be imported even with the reload (not triggering the error we expect). To solve this, we should edit the `test_pybamm_import` function. Possible solution steps: 1. Remove the hidden mandatory dependency from `pybtex` - I know that @prady0t is investigating to solve this. 2. Fix error detection - Edit `tests/unit/test_util.py`, in particular edit `test_pybamm_import`: - Do not use `importlib.reload(...)` to reload `pybamm`, instead remove `pybamm` from the `sys.modules` and all its sub-modules (so also `pybamm.citations` etc..). - In the `sys.modules` set to `None` all the optional modules (e.g. `pybtex`) and all their sub-modules (so also `pybtex.database`, etc..). I could work on this in a shared PR with @prady0t. Thanks for posting this here, @lorenzofavaro, I think this is a reasonable solution based on our conversation earlier. You mentioned that you're working on some aspects of this too, so I'll assign you to this alongside @prady0t.	2024-04-07T13:55:34	[]	[]
pybamm-team/PyBaMM	pybamm-team__PyBaMM-3963	426e441fb6d26d390f6add9d32631ccf219cd940	diff --git a/CHANGELOG.md b/CHANGELOG.md index d13c17ad25..351b644ac8 100644 --- a/CHANGELOG.md +++ b/CHANGELOG.md @@ -28,6 +28,7 @@ ## Breaking changes +- Removed support for Python 3.8 ([#3961](https://github.com/pybamm-team/PyBaMM/pull/3961)) - Renamed "ocp_soc_0_dimensional" to "ocp_soc_0" and "ocp_soc_100_dimensional" to "ocp_soc_100" ([#3942](https://github.com/pybamm-team/PyBaMM/pull/3942)) - The ODES solver was removed due to compatibility issues. Users should use IDAKLU, Casadi, or JAX instead. ([#3932](https://github.com/pybamm-team/PyBaMM/pull/3932)) - Integrated the `[pandas]` extra into the core PyBaMM package, deprecating the `pybamm[pandas]` optional dependency. Pandas is now a required dependency and will be installed upon installing PyBaMM ([#3892](https://github.com/pybamm-team/PyBaMM/pull/3892))	Flaky JAX test cases While running unit-tests, some tests appaers to be flaky. See https://github.com/pybamm-team/PyBaMM/pull/3857#issuecomment-2015300918 #3857 Is a prerequisite to this. Once it gets merged, we can discuss more such tests here and come up with a solution for them.	For additional context – we have seen this happen just with the JAX evaluator tests, where we do not receive consistent outputs. Right now, a stopgap solution is to run these tests again by adding `pytest-flaky` as a dependency and using a `@pytest.mark.flaky` decorator on them – it will then run those tests thrice until they (hopefully) pass Upon a closer look, these failures seem to be coming from comparing the time taken to get to a second solution after the first solution, which speaks more about how the test has been written rather than the computation, I feel. It can be inherently flaky, especially with how we have been noticing that the these tests in parallel mode are quite slow, cf. #3948.	2024-04-03T22:19:39	[]	[]
pybamm-team/PyBaMM	pybamm-team__PyBaMM-3961	6bb959c933808a91f01feb7047e2943da41fb761	diff --git a/.github/workflows/benchmark_on_push.yml b/.github/workflows/benchmark_on_push.yml index 41eeb7d637..b0da71461e 100644 --- a/.github/workflows/benchmark_on_push.yml +++ b/.github/workflows/benchmark_on_push.yml @@ -13,10 +13,10 @@ jobs: runs-on: ubuntu-latest steps: - uses: actions/checkout@v4 - - name: Set up Python 3.8 + - name: Set up Python 3.12 uses: actions/setup-python@v5 with: - python-version: 3.8 + python-version: 3.12 - name: Install Linux system dependencies run: \| diff --git a/.github/workflows/periodic_benchmarks.yml b/.github/workflows/periodic_benchmarks.yml index 276724d274..dd5cc39173 100644 --- a/.github/workflows/periodic_benchmarks.yml +++ b/.github/workflows/periodic_benchmarks.yml @@ -21,10 +21,10 @@ jobs: steps: - uses: actions/checkout@v4 - - name: Set up Python 3.8 + - name: Set up Python 3.12 uses: actions/setup-python@v5 with: - python-version: 3.8 + python-version: 3.12 - name: Install Linux system dependencies run: \| @@ -58,10 +58,10 @@ jobs: needs: benchmarks runs-on: ubuntu-latest steps: - - name: Set up Python 3.8 + - name: Set up Python 3.12 uses: actions/setup-python@v5 with: - python-version: 3.8 + python-version: 3.12 - name: Install asv run: pip install asv diff --git a/.github/workflows/publish_pypi.yml b/.github/workflows/publish_pypi.yml index bbff62201d..d1c960ec43 100644 --- a/.github/workflows/publish_pypi.yml +++ b/.github/workflows/publish_pypi.yml @@ -35,7 +35,7 @@ jobs: - uses: actions/checkout@v4 - uses: actions/setup-python@v5 with: - python-version: 3.8 + python-version: 3.11 - name: Get number of cores on Windows id: get_num_cores @@ -107,7 +107,7 @@ jobs: - uses: actions/setup-python@v5 name: Set up Python with: - python-version: 3.8 + python-version: 3.11 - name: Clone pybind11 repo (no history) run: git clone --depth 1 --branch v2.11.1 https://github.com/pybind/pybind11.git diff --git a/.github/workflows/run_benchmarks_over_history.yml b/.github/workflows/run_benchmarks_over_history.yml index 4f7302a4a5..c8d6892e66 100644 --- a/.github/workflows/run_benchmarks_over_history.yml +++ b/.github/workflows/run_benchmarks_over_history.yml @@ -23,10 +23,10 @@ jobs: runs-on: ubuntu-latest steps: - uses: actions/checkout@v4 - - name: Set up Python 3.8 + - name: Set up Python 3.12 uses: actions/setup-python@v5 with: - python-version: 3.8 + python-version: 3.12 - name: Install nox and asv run: pip install -U pip nox asv - name: Fetch develop branch @@ -53,10 +53,10 @@ jobs: needs: benchmarks runs-on: ubuntu-latest steps: - - name: Set up Python 3.8 + - name: Set up Python 3.12 uses: actions/setup-python@v5 with: - python-version: 3.8 + python-version: 3.12 - name: Install asv run: pip install asv - name: Checkout pybamm-bench repo diff --git a/.github/workflows/update_version.yml b/.github/workflows/update_version.yml index 85f85cebe8..eb62469778 100644 --- a/.github/workflows/update_version.yml +++ b/.github/workflows/update_version.yml @@ -44,7 +44,7 @@ jobs: - name: Set up Python uses: actions/setup-python@v5 with: - python-version: 3.8 + python-version: 3.12 - name: Install dependencies run: \| diff --git a/docs/source/user_guide/installation/gnu-linux-mac.rst b/docs/source/user_guide/installation/gnu-linux-mac.rst index 46f3d1281f..0be4b98e4c 100644 --- a/docs/source/user_guide/installation/gnu-linux-mac.rst +++ b/docs/source/user_guide/installation/gnu-linux-mac.rst @@ -95,10 +95,6 @@ Optional - JaxSolver Users can install ``jax`` and ``jaxlib`` to use the Jax solver. -.. note:: - - The Jax solver is only supported for Python versions 3.9 through 3.12. - .. code:: bash pip install "pybamm[jax]" diff --git a/docs/source/user_guide/installation/install-from-source.rst b/docs/source/user_guide/installation/install-from-source.rst index edcbdf79aa..827f6c2750 100644 --- a/docs/source/user_guide/installation/install-from-source.rst +++ b/docs/source/user_guide/installation/install-from-source.rst @@ -25,7 +25,7 @@ or download the source archive on the repository's homepage. To install PyBaMM, you will need: -- Python 3 (PyBaMM supports versions 3.8, 3.9, 3.10, 3.11, and 3.12) +- Python 3 (PyBaMM supports versions 3.9, 3.10, 3.11, and 3.12) - The Python headers file for your current Python version. - A BLAS library (for instance `openblas <https://www.openblas.net/>`_). - A C compiler (ex: ``gcc``). diff --git a/docs/source/user_guide/installation/windows.rst b/docs/source/user_guide/installation/windows.rst index d99d1f2eb2..02d9f8dd29 100644 --- a/docs/source/user_guide/installation/windows.rst +++ b/docs/source/user_guide/installation/windows.rst @@ -6,7 +6,7 @@ Windows Prerequisites ------------- -To use PyBaMM, you must have Python 3.8, 3.9, 3.10, 3.11, or 3.12 installed. +To use PyBaMM, you must have Python 3.9, 3.10, 3.11, or 3.12 installed. To install Python 3 download the installation files from `Python’s website <https://www.python.org/downloads/windows/>`__. Make sure to @@ -71,10 +71,6 @@ Optional - JaxSolver Users can install ``jax`` and ``jaxlib`` to use the Jax solver. -.. note:: - - The Jax solver is only supported for Python versions 3.9 through 3.12. - .. code:: bash pip install "pybamm[jax]" diff --git a/noxfile.py b/noxfile.py index 17d466e690..c93f3d808d 100644 --- a/noxfile.py +++ b/noxfile.py @@ -61,10 +61,7 @@ def run_coverage(session): """Run the coverage tests and generate an XML report.""" set_environment_variables(PYBAMM_ENV, session=session) session.install("coverage", silent=False) - if sys.version_info < (3, 9): - session.install("-e", ".[all,dev]", silent=False) - else: - session.install("-e", ".[all,dev,jax]", silent=False) + session.install("-e", ".[all,dev,jax]", silent=False) session.run("pytest", "--cov=pybamm", "--cov-report=xml", "tests/unit") @@ -72,10 +69,7 @@ def run_coverage(session): def run_integration(session): """Run the integration tests.""" set_environment_variables(PYBAMM_ENV, session=session) - if sys.version_info < (3, 9): - session.install("-e", ".[all,dev]", silent=False) - else: - session.install("-e", ".[all,dev,jax]", silent=False) + session.install("-e", ".[all,dev,jax]", silent=False) session.run("python", "run-tests.py", "--integration") @@ -90,10 +84,7 @@ def run_doctests(session): def run_unit(session): """Run the unit tests.""" set_environment_variables(PYBAMM_ENV, session=session) - if sys.version_info < (3, 9): - session.install("-e", ".[all,dev]", silent=False) - else: - session.install("-e", ".[all,dev,jax]", silent=False) + session.install("-e", ".[all,dev,jax]", silent=False) session.run("python", "run-tests.py", "--unit") @@ -129,36 +120,22 @@ def set_dev(session): # https://bitbucket.org/pybtex-devs/pybtex/issues/169/replace-pkg_resources-with # is fixed session.run(python, "-m", "pip", "install", "setuptools", external=True) - if sys.version_info < (3, 9): - session.run( - python, - "-m", - "pip", - "install", - "-e", - ".[all,dev]", - external=True, - ) - else: - session.run( - python, - "-m", - "pip", - "install", - "-e", - ".[all,dev,jax]", - external=True, - ) + session.run( + python, + "-m", + "pip", + "install", + "-e", + ".[all,dev,jax]", + external=True, + ) @nox.session(name="tests") def run_tests(session): """Run the unit tests and integration tests sequentially.""" set_environment_variables(PYBAMM_ENV, session=session) - if sys.version_info < (3, 9): - session.install("-e", ".[all,dev]", silent=False) - else: - session.install("-e", ".[all,dev,jax]", silent=False) + session.install("-e", ".[all,dev,jax]", silent=False) session.run("python", "run-tests.py", "--all") diff --git a/pybamm/expression_tree/concatenations.py b/pybamm/expression_tree/concatenations.py index a25ac8e0b3..11e695425f 100644 --- a/pybamm/expression_tree/concatenations.py +++ b/pybamm/expression_tree/concatenations.py @@ -8,7 +8,7 @@ import numpy as np import sympy from scipy.sparse import issparse, vstack -from typing import Sequence +from collections.abc import Sequence import pybamm diff --git a/pybamm/expression_tree/functions.py b/pybamm/expression_tree/functions.py index 7f43f0b7c2..6eb301a24c 100644 --- a/pybamm/expression_tree/functions.py +++ b/pybamm/expression_tree/functions.py @@ -6,7 +6,8 @@ import numpy as np from scipy import special import sympy -from typing import Sequence, Callable +from typing import Callable +from collections.abc import Sequence from typing_extensions import TypeVar import pybamm diff --git a/pybamm/expression_tree/interpolant.py b/pybamm/expression_tree/interpolant.py index 16714f9db3..10881d3084 100644 --- a/pybamm/expression_tree/interpolant.py +++ b/pybamm/expression_tree/interpolant.py @@ -4,7 +4,7 @@ from __future__ import annotations import numpy as np from scipy import interpolate -from typing import Sequence +from collections.abc import Sequence import numbers import pybamm diff --git a/pybamm/expression_tree/operations/unpack_symbols.py b/pybamm/expression_tree/operations/unpack_symbols.py index 1933eada76..97e1d08d98 100644 --- a/pybamm/expression_tree/operations/unpack_symbols.py +++ b/pybamm/expression_tree/operations/unpack_symbols.py @@ -2,7 +2,8 @@ # Helper function to unpack a symbol # from __future__ import annotations -from typing import TYPE_CHECKING, Sequence +from typing import TYPE_CHECKING +from collections.abc import Sequence if TYPE_CHECKING: # pragma: no cover import pybamm diff --git a/pybamm/expression_tree/symbol.py b/pybamm/expression_tree/symbol.py index 466e5b7be6..d50b8a6c89 100644 --- a/pybamm/expression_tree/symbol.py +++ b/pybamm/expression_tree/symbol.py @@ -8,7 +8,8 @@ import sympy from scipy.sparse import csr_matrix, issparse from functools import cached_property -from typing import TYPE_CHECKING, Sequence, cast +from typing import TYPE_CHECKING, cast +from collections.abc import Sequence import pybamm from pybamm.util import import_optional_dependency diff --git a/pybamm/solvers/idaklu_jax.py b/pybamm/solvers/idaklu_jax.py index 9a5315e120..ba1c80c1c4 100644 --- a/pybamm/solvers/idaklu_jax.py +++ b/pybamm/solvers/idaklu_jax.py @@ -5,7 +5,6 @@ import numbers from typing import Union -from typing import List from functools import lru_cache @@ -275,7 +274,7 @@ def jax_value( self, t: np.ndarray = None, inputs: Union[dict, None] = None, - output_variables: Union[List[str], None] = None, + output_variables: Union[list[str], None] = None, ): """Helper function to compute the gradient of a jaxified expression @@ -308,7 +307,7 @@ def jax_grad( self, t: np.ndarray = None, inputs: Union[dict, None] = None, - output_variables: Union[List[str], None] = None, + output_variables: Union[list[str], None] = None, ): """Helper function to compute the gradient of a jaxified expression diff --git a/pybamm/type_definitions.py b/pybamm/type_definitions.py index c3d0a56faa..376189b41a 100644 --- a/pybamm/type_definitions.py +++ b/pybamm/type_definitions.py @@ -1,9 +1,6 @@ -# -# Common type definitions for PyBaMM -# from __future__ import annotations -from typing import Union, List, Dict +from typing import Union from typing_extensions import TypeAlias import numpy as np import pybamm @@ -15,6 +12,6 @@ ChildValue: TypeAlias = Union[float, np.ndarray] ChildSymbol: TypeAlias = Union[float, np.ndarray, pybamm.Symbol] -DomainType: TypeAlias = Union[List[str], str, None] -AuxiliaryDomainType: TypeAlias = Union[Dict[str, str], None] -DomainsType: TypeAlias = Union[Dict[str, Union[List[str], str]], None] +DomainType: TypeAlias = Union[list[str], str, None] +AuxiliaryDomainType: TypeAlias = Union[dict[str, str], None] +DomainsType: TypeAlias = Union[dict[str, Union[list[str], str]], None] diff --git a/pyproject.toml b/pyproject.toml index 07b2909983..b8b1469ece 100644 --- a/pyproject.toml +++ b/pyproject.toml @@ -18,7 +18,7 @@ license = { file = "LICENSE.txt" } description = "Python Battery Mathematical Modelling" authors = [{name = "The PyBaMM Team", email = "[email protected]"}] maintainers = [{name = "The PyBaMM Team", email = "[email protected]"}] -requires-python = ">=3.8, <3.13" +requires-python = ">=3.9, <3.13" readme = {file = "README.md", content-type = "text/markdown"} classifiers = [ "Development Status :: 5 - Production/Stable", @@ -28,7 +28,6 @@ classifiers = [ "Programming Language :: Python", "Programming Language :: Python :: 3", "Programming Language :: Python :: 3 :: Only", - "Programming Language :: Python :: 3.8", "Programming Language :: Python :: 3.9", "Programming Language :: Python :: 3.10", "Programming Language :: Python :: 3.11", @@ -54,7 +53,6 @@ Releases = "https://github.com/pybamm-team/PyBaMM/releases" Changelog = "https://github.com/pybamm-team/PyBaMM/blob/develop/CHANGELOG.md" [project.optional-dependencies] -# For the generation of documentation docs = [ "sphinx>=6", "sphinx_rtd_theme>=0.5", @@ -77,7 +75,6 @@ docs = [ examples = [ "jupyter", ] -# Plotting functionality plot = [ "imageio>=2.3.0", # Note: matplotlib is loaded for debug plots, but to ensure PyBaMM runs @@ -85,7 +82,6 @@ plot = [ # outside of plot() methods. "matplotlib>=3.6.0", ] -# For the Citations class cite = [ "pybtex>=0.24.0", ] @@ -97,7 +93,6 @@ bpx = [ tqdm = [ "tqdm", ] -# Dependencies intended for use by developers dev = [ # For working with pre-commit hooks "pre-commit", diff --git a/scripts/install_KLU_Sundials.py b/scripts/install_KLU_Sundials.py index c30a3cf3e5..6224e40776 100755 --- a/scripts/install_KLU_Sundials.py +++ b/scripts/install_KLU_Sundials.py @@ -1,14 +1,3 @@ -# /// pyproject -# [run] -# requires-python = "">=3.8, <3.13"" -# dependencies = [ -# "cmake", -# ] -# -# [additional-info] -# repository = "https://github.com/pybamm-team/PyBaMM" -# documentation = "https://docs.pybamm.org" -# /// import os import subprocess import tarfile	Drop support for Python `3.8`, support Python `3.12`? NumPy>=1.26 and SciPy>=1.11 support Python 3.9+ now and already have wheels for Python 3.12. The blocker should be CasADi, which might have a new release soon because the [pre-release](https://github.com/casadi/casadi/releases/tag/nightly-py312) is available. In any case, we should be able to drop Python `3.8` support at this time and `3.12` can wait	I don't think we should drop support for Python 3.8 unless PyBaMM breaks on 3.8 or 3.8 reaches `end-of-life`. We can actually start testing PyBaMM on 3.12 in the CI. IIRC, `pybind11` has already migrated to Python 3.12, and we can fetch CasADi's nightly wheels in the 3.12 job. Yes, it would be good to try supporting Python 3.12, which comes tomorrow. `xarray` doesn't run CI against 3.12 yet, but they list their Python dependency as >=3.9. I think we should be fine with using the nightly wheels for CasADi, should we let the optional dependencies catch up later? > I don't think we should drop support for Python 3.8 unless PyBaMM breaks on 3.8 or 3.8 reaches end-of-life. That makes sense, I just checked and Python 3.8 EOL is in October 2024, so we have just about a year to remove it We should only build and ship pybamm wheels once every dependency works, but testing pybamm (just with the required dependencies) in the CI should be good. (release candidates of 3.12 have been available on GH Actions for quite some time now) Good idea, we could also do these checks in the nightly releases when ready (#3251) since it would help catch errors and deprecations early Copying from slack - > https://scientific-python.org/specs/ - like PEPs but for the Scientific Python ecosystem. I think we should follow them to stay up-to-date with other libraries. > > The first SPEC suggests removing Python 3.9 support by the end of this year, but I’m not sure if we should do that :grimacing: There are a lot of interesting things available in the Scientific Python ecosystem, perhaps we should also try using [`sp-repo-review`](https://learn.scientific-python.org/development/guides/repo-review/)? I guess this would be better off as a separate issue, please let me know if I should open one Some of the failing checks in repo review are intentional from our end (for example - we removed codespell, prettier, and isort as pre-commit hooks because they were doing more harm than good), and some of them are potential GSoC projects (for example - pybamm wide typing hints), but yes an issue to track repo review's suggestions would be nice. We could either implement the suggestions, discard them, or mark them as implemented. PS: we weren't using repo review before because it was Scikit-HEP specific, but now they've moved it to Scientific Python, which is nice! Makes sense. I will run the tool locally and see what we are missing, I cannot get it to to work inside the browser at all for some reason. There has been some discussion about a PR for adding type hints to PyBaMM models after they have been serialised in #3397 (or they were probably for the expression tree: not sure) in the developer meeting yesterday > Makes sense. I will run the tool locally and see what we are missing, I cannot get it to to work inside the browser at all for some reason. There has been some discussion about a PR for adding type hints to PyBaMM models after they have been serialised in #3397 (or they were probably for the expression tree: not sure) in the developer meeting yesterday I've been going through and adding type hints for the expression trees yes (they're more of a by-product from trying various methods for serialisation) - I'll make a PR asap for those. `casadi` now supports Python 3.12 (https://github.com/casadi/casadi/releases/tag/3.6.4) which checks all four of our core dependencies. I installed all dependencies with Python 3.12 locally without issues, except for `[jax]` (which will now require bumping the versions) and `[odes]` (I see that `numpy` removed `distutils` with Python 3.12+ which broke the wheel building process – I can file an issue upstream requesting for a fix). @Saransh-cpp: should we try for Python 3.12? We can still keep Python 3.8 as discussed above. I was supposed to work on #3443, which I can then close in the same PR to avoid merge conflicts. In any case, it will be merged only after the stable release, so we have a lot of time to look into any failures that occur. > should we try for Python 3.12? We can still keep Python 3.8 as discussed above. Yes! We can add a note in the docs about optional dependencies not yet supported for Python 3.12. > and [odes] Ah, I remember the maintainers of scikits.odes were waiting for NumPy's migration to meson to stop using `numpy.distutils`, but the repository is inactive. See my old PR https://github.com/bmcage/odes/pull/137 for some discussion on this. Creating an issue there should be good.	2024-04-03T17:37:07	[]	[]
pybamm-team/PyBaMM	pybamm-team__PyBaMM-3866	7a8268710a4c6ca72f650c08f6f323a23f73b105	diff --git a/CHANGELOG.md b/CHANGELOG.md index 069093015a..41856b3fd7 100644 --- a/CHANGELOG.md +++ b/CHANGELOG.md @@ -18,6 +18,7 @@ ## Breaking changes +- Renamed "have_optional_dependency" to "import_optional_dependency" ([#3866](https://github.com/pybamm-team/PyBaMM/pull/3866)) - Integrated the `[latexify]` extra into the core PyBaMM package, deprecating the `pybamm[latexify]` set of optional dependencies. SymPy is now a required dependency and will be installed upon installing PyBaMM ([#3848](https://github.com/pybamm-team/PyBaMM/pull/3848)) - Renamed "testing" argument for plots to "show_plot" and flipped its meaning (show_plot=True is now the default and shows the plot) ([#3842](https://github.com/pybamm-team/PyBaMM/pull/3842)) - Dropped support for BPX version 0.3.0 and below ([#3414](https://github.com/pybamm-team/PyBaMM/pull/3414)) diff --git a/CONTRIBUTING.md b/CONTRIBUTING.md index b9800dcd61..fc8e848bb5 100644 --- a/CONTRIBUTING.md +++ b/CONTRIBUTING.md @@ -104,13 +104,13 @@ Only 'core pybamm' is installed by default. The others have to be specified expl PyBaMM utilizes optional dependencies to allow users to choose which additional libraries they want to use. Managing these optional dependencies and their imports is essential to provide flexibility to PyBaMM users. -PyBaMM provides a utility function `have_optional_dependency`, to check for the availability of optional dependencies within methods. This function can be used to conditionally import optional dependencies only if they are available. Here's how to use it: +PyBaMM provides a utility function `import_optional_dependency`, to check for the availability of optional dependencies within methods. This function can be used to conditionally import optional dependencies only if they are available. Here's how to use it: Optional dependencies should never be imported at the module level, but always inside methods. For example: ``` def use_pybtex(x,y,z): - pybtex = have_optional_dependency("pybtex") + pybtex = import_optional_dependency("pybtex") ... ``` @@ -118,7 +118,7 @@ While importing a specific module instead of an entire package/library: ```python def use_parse_file(x, y, z): - parse_file = have_optional_dependency("pybtex.database", "parse_file") + parse_file = import_optional_dependency("pybtex.database", "parse_file") ... ``` @@ -143,7 +143,7 @@ class TestUtil(TestCase): pybamm.function_using_pybtex(x, y, z) # Test that the function works when pybtex is available - sys.modules["pybtex"] = pybamm.util.have_optional_dependency("pybtex") + sys.modules["pybtex"] = pybamm.util.import_optional_dependency("pybtex") pybamm.function_using_pybtex(x, y, z) ``` diff --git a/pybamm/__init__.py b/pybamm/__init__.py index ab2e72ed28..c2654ea9cf 100644 --- a/pybamm/__init__.py +++ b/pybamm/__init__.py @@ -47,7 +47,7 @@ get_parameters_filepath, have_jax, install_jax, - have_optional_dependency, + import_optional_dependency, is_jax_compatible, get_git_commit_info, ) diff --git a/pybamm/citations.py b/pybamm/citations.py index da371bbd84..74f477c117 100644 --- a/pybamm/citations.py +++ b/pybamm/citations.py @@ -7,7 +7,7 @@ import os import warnings from sys import _getframe -from pybamm.util import have_optional_dependency +from pybamm.util import import_optional_dependency class Citations: @@ -74,7 +74,7 @@ def read_citations(self): """Reads the citations in `pybamm.CITATIONS.bib`. Other works can be cited by passing a BibTeX citation to :meth:`register`. """ - parse_file = have_optional_dependency("pybtex.database", "parse_file") + parse_file = import_optional_dependency("pybtex.database", "parse_file") citations_file = os.path.join(pybamm.root_dir(), "pybamm", "CITATIONS.bib") bib_data = parse_file(citations_file, bib_format="bibtex") for key, entry in bib_data.entries.items(): @@ -85,7 +85,7 @@ def _add_citation(self, key, entry): previous entry is overwritten """ - Entry = have_optional_dependency("pybtex.database", "Entry") + Entry = import_optional_dependency("pybtex.database", "Entry") # Check input types are correct if not isinstance(key, str) or not isinstance(entry, Entry): raise TypeError() @@ -151,8 +151,8 @@ def _parse_citation(self, key): key: str A BibTeX formatted citation """ - PybtexError = have_optional_dependency("pybtex.scanner", "PybtexError") - parse_string = have_optional_dependency("pybtex.database", "parse_string") + PybtexError = import_optional_dependency("pybtex.scanner", "PybtexError") + parse_string = import_optional_dependency("pybtex.database", "parse_string") try: # Parse string as a bibtex citation, and check that a citation was found bib_data = parse_string(key, bib_format="bibtex") @@ -219,7 +219,7 @@ def print(self, filename=None, output_format="text", verbose=False): """ # Parse citations that were not known keys at registration, but do not # fail if they cannot be parsed - pybtex = have_optional_dependency("pybtex") + pybtex = import_optional_dependency("pybtex") try: for key in self._unknown_citations: self._parse_citation(key) diff --git a/pybamm/expression_tree/functions.py b/pybamm/expression_tree/functions.py index 7f0986441f..7f43f0b7c2 100644 --- a/pybamm/expression_tree/functions.py +++ b/pybamm/expression_tree/functions.py @@ -10,7 +10,7 @@ from typing_extensions import TypeVar import pybamm -from pybamm.util import have_optional_dependency +from pybamm.util import import_optional_dependency class Function(pybamm.Symbol): @@ -98,7 +98,7 @@ def _function_diff(self, children: Sequence[pybamm.Symbol], idx: float): Derivative with respect to child number 'idx'. See :meth:`pybamm.Symbol._diff()`. """ - autograd = have_optional_dependency("autograd") + autograd = import_optional_dependency("autograd") # Store differentiated function, needed in case we want to convert to CasADi if self.derivative == "autograd": return Function( diff --git a/pybamm/expression_tree/symbol.py b/pybamm/expression_tree/symbol.py index 5c4bd5f097..e68e436749 100644 --- a/pybamm/expression_tree/symbol.py +++ b/pybamm/expression_tree/symbol.py @@ -11,7 +11,7 @@ from typing import TYPE_CHECKING, Sequence, cast import pybamm -from pybamm.util import have_optional_dependency +from pybamm.util import import_optional_dependency from pybamm.expression_tree.printing.print_name import prettify_print_name if TYPE_CHECKING: # pragma: no cover @@ -479,7 +479,7 @@ def render(self): # pragma: no cover """ Print out a visual representation of the tree (this node and its children) """ - anytree = have_optional_dependency("anytree") + anytree = import_optional_dependency("anytree") for pre, _, node in anytree.RenderTree(self): if isinstance(node, pybamm.Scalar) and node.name != str(node.value): print(f"{pre}{node.name} = {node.value}") @@ -498,7 +498,7 @@ def visualise(self, filename: str): filename to output, must end in ".png" """ - DotExporter = have_optional_dependency("anytree.exporter", "DotExporter") + DotExporter = import_optional_dependency("anytree.exporter", "DotExporter") # check that filename ends in .png. if filename[-4:] != ".png": raise ValueError("filename should end in .png") @@ -518,7 +518,7 @@ def relabel_tree(self, symbol: Symbol, counter: int): Finds all children of a symbol and assigns them a new id so that they can be visualised properly using the graphviz output """ - anytree = have_optional_dependency("anytree") + anytree = import_optional_dependency("anytree") name = symbol.name if name == "div": name = "∇⋅" @@ -561,7 +561,7 @@ def pre_order(self): a b """ - anytree = have_optional_dependency("anytree") + anytree = import_optional_dependency("anytree") return anytree.PreOrderIter(self) def __str__(self): diff --git a/pybamm/expression_tree/unary_operators.py b/pybamm/expression_tree/unary_operators.py index cb5bc905fd..9669c2596a 100644 --- a/pybamm/expression_tree/unary_operators.py +++ b/pybamm/expression_tree/unary_operators.py @@ -7,7 +7,7 @@ from scipy.sparse import csr_matrix, issparse import sympy import pybamm -from pybamm.util import have_optional_dependency +from pybamm.util import import_optional_dependency from pybamm.type_definitions import DomainsType @@ -450,7 +450,9 @@ def _unary_new_copy(self, child): def _sympy_operator(self, child): """Override :meth:`pybamm.UnaryOperator._sympy_operator`""" - sympy_Gradient = have_optional_dependency("sympy.vector.operators", "Gradient") + sympy_Gradient = import_optional_dependency( + "sympy.vector.operators", "Gradient" + ) return sympy_Gradient(child) @@ -484,7 +486,7 @@ def _unary_new_copy(self, child): def _sympy_operator(self, child): """Override :meth:`pybamm.UnaryOperator._sympy_operator`""" - sympy_Divergence = have_optional_dependency( + sympy_Divergence = import_optional_dependency( "sympy.vector.operators", "Divergence" ) return sympy_Divergence(child) diff --git a/pybamm/meshes/scikit_fem_submeshes.py b/pybamm/meshes/scikit_fem_submeshes.py index 82a7bd72f1..e52f58f069 100644 --- a/pybamm/meshes/scikit_fem_submeshes.py +++ b/pybamm/meshes/scikit_fem_submeshes.py @@ -5,7 +5,7 @@ from .meshes import SubMesh import numpy as np -from pybamm.util import have_optional_dependency +from pybamm.util import import_optional_dependency class ScikitSubMesh2D(SubMesh): @@ -27,7 +27,7 @@ class ScikitSubMesh2D(SubMesh): """ def __init__(self, edges, coord_sys, tabs): - skfem = have_optional_dependency("skfem") + skfem = import_optional_dependency("skfem") self.edges = edges self.nodes = dict.fromkeys(["y", "z"]) for var in self.nodes.keys(): diff --git a/pybamm/models/base_model.py b/pybamm/models/base_model.py index 194b8938af..007897370f 100644 --- a/pybamm/models/base_model.py +++ b/pybamm/models/base_model.py @@ -13,7 +13,6 @@ import pybamm from pybamm.expression_tree.operations.serialise import Serialise -import sympy class BaseModel: @@ -1185,8 +1184,7 @@ def latexify(self, filename=None, newline=True, output_variables=None): This will return first five model equations >>> model.latexify(newline=False)[1:5] """ - if sympy: - from pybamm.expression_tree.operations.latexify import Latexify + from pybamm.expression_tree.operations.latexify import Latexify return Latexify(self, filename, newline).latexify( output_variables=output_variables diff --git a/pybamm/plotting/plot.py b/pybamm/plotting/plot.py index cf5c972a87..4037ab8fbf 100644 --- a/pybamm/plotting/plot.py +++ b/pybamm/plotting/plot.py @@ -3,7 +3,7 @@ # import pybamm from .quick_plot import ax_min, ax_max -from pybamm.util import have_optional_dependency +from pybamm.util import import_optional_dependency def plot(x, y, ax=None, show_plot=True, kwargs): @@ -27,7 +27,7 @@ def plot(x, y, ax=None, show_plot=True, kwargs): Keyword arguments, passed to plt.plot """ - plt = have_optional_dependency("matplotlib.pyplot") + plt = import_optional_dependency("matplotlib.pyplot") if not isinstance(x, pybamm.Array): raise TypeError("x must be 'pybamm.Array'") diff --git a/pybamm/plotting/plot2D.py b/pybamm/plotting/plot2D.py index a37cd1e2ed..7d1f3c6bae 100644 --- a/pybamm/plotting/plot2D.py +++ b/pybamm/plotting/plot2D.py @@ -3,7 +3,7 @@ # import pybamm from .quick_plot import ax_min, ax_max -from pybamm.util import have_optional_dependency +from pybamm.util import import_optional_dependency def plot2D(x, y, z, ax=None, show_plot=True, kwargs): @@ -27,7 +27,7 @@ def plot2D(x, y, z, ax=None, show_plot=True, kwargs): only display the plot after plt.show() has been called. """ - plt = have_optional_dependency("matplotlib.pyplot") + plt = import_optional_dependency("matplotlib.pyplot") if not isinstance(x, pybamm.Array): raise TypeError("x must be 'pybamm.Array'") diff --git a/pybamm/plotting/plot_summary_variables.py b/pybamm/plotting/plot_summary_variables.py index 33642c4d5a..bd4db0ee6c 100644 --- a/pybamm/plotting/plot_summary_variables.py +++ b/pybamm/plotting/plot_summary_variables.py @@ -3,7 +3,7 @@ # import numpy as np import pybamm -from pybamm.util import have_optional_dependency +from pybamm.util import import_optional_dependency def plot_summary_variables( @@ -27,7 +27,7 @@ def plot_summary_variables( Keyword arguments, passed to plt.subplots. """ - plt = have_optional_dependency("matplotlib.pyplot") + plt = import_optional_dependency("matplotlib.pyplot") if isinstance(solutions, pybamm.Solution): solutions = [solutions] diff --git a/pybamm/plotting/plot_voltage_components.py b/pybamm/plotting/plot_voltage_components.py index 3b155b71de..0d1bb7b573 100644 --- a/pybamm/plotting/plot_voltage_components.py +++ b/pybamm/plotting/plot_voltage_components.py @@ -3,7 +3,7 @@ # import numpy as np -from pybamm.util import have_optional_dependency +from pybamm.util import import_optional_dependency from pybamm.simulation import Simulation from pybamm.solvers.solution import Solution @@ -42,7 +42,7 @@ def plot_voltage_components( solution = input_data.solution elif isinstance(input_data, Solution): solution = input_data - plt = have_optional_dependency("matplotlib.pyplot") + plt = import_optional_dependency("matplotlib.pyplot") # Set a default value for alpha, the opacity kwargs_fill = {"alpha": 0.6, *kwargs_fill} diff --git a/pybamm/plotting/quick_plot.py b/pybamm/plotting/quick_plot.py index c6d3054abb..dbb8f37555 100644 --- a/pybamm/plotting/quick_plot.py +++ b/pybamm/plotting/quick_plot.py @@ -5,7 +5,7 @@ import numpy as np import pybamm from collections import defaultdict -from pybamm.util import have_optional_dependency +from pybamm.util import import_optional_dependency class LoopList(list): @@ -46,7 +46,7 @@ def split_long_string(title, max_words=None): def close_plots(): """Close all open figures""" - plt = have_optional_dependency("matplotlib.pyplot") + plt = import_optional_dependency("matplotlib.pyplot") plt.close("all") @@ -473,10 +473,10 @@ def plot(self, t, dynamic=False): Dimensional time (in 'time_units') at which to plot. """ - plt = have_optional_dependency("matplotlib.pyplot") - gridspec = have_optional_dependency("matplotlib.gridspec") - cm = have_optional_dependency("matplotlib", "cm") - colors = have_optional_dependency("matplotlib", "colors") + plt = import_optional_dependency("matplotlib.pyplot") + gridspec = import_optional_dependency("matplotlib.gridspec") + cm = import_optional_dependency("matplotlib", "cm") + colors = import_optional_dependency("matplotlib", "colors") t_in_seconds = t self.time_scaling_factor self.fig = plt.figure(figsize=self.figsize) @@ -674,8 +674,8 @@ def dynamic_plot(self, show_plot=True, step=None): continuous_update=False, ) else: - plt = have_optional_dependency("matplotlib.pyplot") - Slider = have_optional_dependency("matplotlib.widgets", "Slider") + plt = import_optional_dependency("matplotlib.pyplot") + Slider = import_optional_dependency("matplotlib.widgets", "Slider") # create an initial plot at time self.min_t self.plot(self.min_t, dynamic=True) @@ -779,8 +779,8 @@ def create_gif(self, number_of_images=80, duration=0.1, output_filename="plot.gi Name of the generated GIF file. """ - imageio = have_optional_dependency("imageio.v2") - plt = have_optional_dependency("matplotlib.pyplot") + imageio = import_optional_dependency("imageio.v2") + plt = import_optional_dependency("matplotlib.pyplot") # time stamps at which the images/plots will be created time_array = np.linspace(self.min_t, self.max_t, num=number_of_images) diff --git a/pybamm/simulation.py b/pybamm/simulation.py index c1d62a47e9..aa58022b6d 100644 --- a/pybamm/simulation.py +++ b/pybamm/simulation.py @@ -11,7 +11,7 @@ import sys from functools import lru_cache from datetime import timedelta -from pybamm.util import have_optional_dependency +from pybamm.util import import_optional_dependency from pybamm.expression_tree.operations.serialise import Serialise @@ -699,7 +699,7 @@ def solve( # check if a user has tqdm installed if showprogress: - tqdm = have_optional_dependency("tqdm") + tqdm = import_optional_dependency("tqdm") cycle_lengths = tqdm.tqdm( self.experiment.cycle_lengths, desc="Cycling", diff --git a/pybamm/spatial_methods/scikit_finite_element.py b/pybamm/spatial_methods/scikit_finite_element.py index 07a3c0e1be..e65e29f7f8 100644 --- a/pybamm/spatial_methods/scikit_finite_element.py +++ b/pybamm/spatial_methods/scikit_finite_element.py @@ -7,7 +7,7 @@ from scipy.sparse.linalg import inv import numpy as np -from pybamm.util import have_optional_dependency +from pybamm.util import import_optional_dependency class ScikitFiniteElement(pybamm.SpatialMethod): @@ -88,7 +88,7 @@ def gradient(self, symbol, discretised_symbol, boundary_conditions): to the y-component of the gradient and the second column corresponds to the z component of the gradient. """ - skfem = have_optional_dependency("skfem") + skfem = import_optional_dependency("skfem") domain = symbol.domain[0] mesh = self.mesh[domain] @@ -144,7 +144,7 @@ def gradient_matrix(self, symbol, boundary_conditions): :class:`pybamm.Matrix` The (sparse) finite element gradient matrix for the domain """ - skfem = have_optional_dependency("skfem") + skfem = import_optional_dependency("skfem") # get primary domain mesh domain = symbol.domain[0] mesh = self.mesh[domain] @@ -190,7 +190,7 @@ def laplacian(self, symbol, discretised_symbol, boundary_conditions): Contains the result of acting the discretised gradient on the child discretised_symbol """ - skfem = have_optional_dependency("skfem") + skfem = import_optional_dependency("skfem") domain = symbol.domain[0] mesh = self.mesh[domain] @@ -258,7 +258,7 @@ def stiffness_matrix(self, symbol, boundary_conditions): :class:`pybamm.Matrix` The (sparse) finite element stiffness matrix for the domain """ - skfem = have_optional_dependency("skfem") + skfem = import_optional_dependency("skfem") # get primary domain mesh domain = symbol.domain[0] mesh = self.mesh[domain] @@ -321,7 +321,7 @@ def definite_integral_matrix(self, child, vector_type="row"): :class:`pybamm.Matrix` The finite element integral vector for the domain """ - skfem = have_optional_dependency("skfem") + skfem = import_optional_dependency("skfem") # get primary domain mesh domain = child.domain[0] mesh = self.mesh[domain] @@ -383,7 +383,7 @@ def boundary_integral_vector(self, domain, region): :class:`pybamm.Matrix` The finite element integral vector for the domain """ - skfem = have_optional_dependency("skfem") + skfem = import_optional_dependency("skfem") # get primary domain mesh mesh = self.mesh[domain[0]] @@ -501,7 +501,7 @@ def assemble_mass_form(self, symbol, boundary_conditions, region="interior"): :class:`pybamm.Matrix` The (sparse) mass matrix for the spatial method. """ - skfem = have_optional_dependency("skfem") + skfem = import_optional_dependency("skfem") # get primary domain mesh domain = symbol.domain[0] mesh = self.mesh[domain] diff --git a/pybamm/util.py b/pybamm/util.py index 1149327cf7..0722b8174b 100644 --- a/pybamm/util.py +++ b/pybamm/util.py @@ -353,24 +353,19 @@ def install_jax(arguments=None): # pragma: no cover # https://docs.pybamm.org/en/latest/source/user_guide/contributing.html#managing-optional-dependencies-and-their-imports -def have_optional_dependency(module_name, attribute=None): +def import_optional_dependency(module_name, attribute=None): err_msg = f"Optional dependency {module_name} is not available. See https://docs.pybamm.org/en/latest/source/user_guide/installation/index.html#optional-dependencies for more details." try: - # Attempt to import the specified module module = importlib.import_module(module_name) - if attribute: - # If an attribute is specified, check if it's available if hasattr(module, attribute): imported_attribute = getattr(module, attribute) - return imported_attribute # Return the imported attribute + # Return the imported attribute + return imported_attribute else: - # Raise an ModuleNotFoundError if the attribute is not available raise ModuleNotFoundError(err_msg) # pragma: no cover else: # Return the entire module if no attribute is specified return module - except ModuleNotFoundError: - # Raise an ModuleNotFoundError if the module or attribute is not available raise ModuleNotFoundError(err_msg)	Rename `have_optional_dependency` to `import_optional_dependency` We use `have_optional_dependency()` to import optional dependencies across codebase which is a wrapper over `importlib.import_module()` to throw a helpful error message.		2024-03-05T14:44:19	[]	[]
pybamm-team/PyBaMM	pybamm-team__PyBaMM-3848	ed6df7e8b444f7aa1fd18dea48254592ef2eb3f4	diff --git a/CHANGELOG.md b/CHANGELOG.md index 6cade77684..069093015a 100644 --- a/CHANGELOG.md +++ b/CHANGELOG.md @@ -18,6 +18,7 @@ ## Breaking changes +- Integrated the `[latexify]` extra into the core PyBaMM package, deprecating the `pybamm[latexify]` set of optional dependencies. SymPy is now a required dependency and will be installed upon installing PyBaMM ([#3848](https://github.com/pybamm-team/PyBaMM/pull/3848)) - Renamed "testing" argument for plots to "show_plot" and flipped its meaning (show_plot=True is now the default and shows the plot) ([#3842](https://github.com/pybamm-team/PyBaMM/pull/3842)) - Dropped support for BPX version 0.3.0 and below ([#3414](https://github.com/pybamm-team/PyBaMM/pull/3414)) diff --git a/docs/source/user_guide/installation/index.rst b/docs/source/user_guide/installation/index.rst index bf83e38835..1a773c65a5 100644 --- a/docs/source/user_guide/installation/index.rst +++ b/docs/source/user_guide/installation/index.rst @@ -59,15 +59,17 @@ Required dependencies PyBaMM requires the following dependencies. -================================================================ ========================== -Package Minimum supported version -================================================================ ========================== -`NumPy <https://numpy.org>`__ 1.23.5 -`SciPy <https://docs.scipy.org/doc/scipy/>`__ 1.9.3 -`CasADi <https://web.casadi.org/docs/>`__ 3.6.3 -`Xarray <https://docs.xarray.dev/en/stable/>`__ 2022.6.0 -`Anytree <https://anytree.readthedocs.io/en/stable/>`__ 2.8.0 -================================================================ ========================== +=================================================================== ========================== +Package Minimum supported version +=================================================================== ========================== +`NumPy <https://numpy.org>`__ 1.23.5 +`SciPy <https://docs.scipy.org/doc/scipy/>`__ 1.9.3 +`CasADi <https://web.casadi.org/docs/>`__ 3.6.3 +`Xarray <https://docs.xarray.dev/en/stable/>`__ 2022.6.0 +`Anytree <https://anytree.readthedocs.io/en/stable/>`__ 2.8.0 +`SymPy <https://docs.sympy.org/latest/index.html>`__ 1.9.3 +`typing-extensions <https://pypi.org/project/typing-extensions/>`__ 4.10.0 +=================================================================== ========================== .. _install.optional_dependencies: @@ -174,19 +176,6 @@ Dependency Minimum Version p `pybtex <https://docs.pybtex.org/>`__ 0.24.0 cite BibTeX-compatible bibliography processor. =========================================================== ================== ================== ========================================= -.. _install.latexify_dependencies: - -Latexify dependencies -^^^^^^^^^^^^^^^^^^^^^ - -Installable with ``pip install "pybamm[latexify]"`` - -=========================================================== ================== ================== ========================= -Dependency Minimum Version pip extra Notes -=========================================================== ================== ================== ========================= -`sympy <https://docs.sympy.org/latest/index.html>`__ 1.9.3 latexify For symbolic mathematics. -=========================================================== ================== ================== ========================= - .. _install.bpx_dependencies: bpx dependencies diff --git a/pybamm/expression_tree/array.py b/pybamm/expression_tree/array.py index 0bb6168a7c..dbaa5cd144 100644 --- a/pybamm/expression_tree/array.py +++ b/pybamm/expression_tree/array.py @@ -4,14 +4,10 @@ from __future__ import annotations import numpy as np from scipy.sparse import csr_matrix, issparse -from typing import TYPE_CHECKING import pybamm -from pybamm.util import have_optional_dependency from pybamm.type_definitions import DomainType, AuxiliaryDomainType, DomainsType - -if TYPE_CHECKING: # pragma: no cover - import sympy +import sympy class Array(pybamm.Symbol): @@ -157,7 +153,6 @@ def is_constant(self): def to_equation(self) -> sympy.Array: """Returns the value returned by the node when evaluated.""" - sympy = have_optional_dependency("sympy") entries_list = self.entries.tolist() return sympy.Array(entries_list) diff --git a/pybamm/expression_tree/binary_operators.py b/pybamm/expression_tree/binary_operators.py index d10844798b..a8086ebdd1 100644 --- a/pybamm/expression_tree/binary_operators.py +++ b/pybamm/expression_tree/binary_operators.py @@ -5,11 +5,11 @@ import numbers import numpy as np +import sympy from scipy.sparse import csr_matrix, issparse import functools import pybamm -from pybamm.util import have_optional_dependency from typing import Callable, cast @@ -180,7 +180,6 @@ def _sympy_operator(self, left, right): def to_equation(self): """Convert the node and its subtree into a SymPy equation.""" - sympy = have_optional_dependency("sympy") if self.print_name is not None: return sympy.Symbol(self.print_name) else: @@ -388,7 +387,6 @@ def _binary_evaluate(self, left, right): def _sympy_operator(self, left, right): """Override :meth:`pybamm.BinaryOperator._sympy_operator`""" - sympy = have_optional_dependency("sympy") left = sympy.Matrix(left) right = sympy.Matrix(right) return left * right @@ -737,7 +735,6 @@ def _binary_new_copy( def _sympy_operator(self, left, right): """Override :meth:`pybamm.BinaryOperator._sympy_operator`""" - sympy = have_optional_dependency("sympy") return sympy.Min(left, right) @@ -782,7 +779,6 @@ def _binary_new_copy( def _sympy_operator(self, left, right): """Override :meth:`pybamm.BinaryOperator._sympy_operator`""" - sympy = have_optional_dependency("sympy") return sympy.Max(left, right) diff --git a/pybamm/expression_tree/concatenations.py b/pybamm/expression_tree/concatenations.py index 06718c311d..52d22b062b 100644 --- a/pybamm/expression_tree/concatenations.py +++ b/pybamm/expression_tree/concatenations.py @@ -6,11 +6,11 @@ from collections import defaultdict import numpy as np +import sympy from scipy.sparse import issparse, vstack from typing import Sequence import pybamm -from pybamm.util import have_optional_dependency class Concatenation(pybamm.Symbol): @@ -159,7 +159,6 @@ def is_constant(self): def _sympy_operator(self, *children): """Apply appropriate SymPy operators.""" - sympy = have_optional_dependency("sympy") self.concat_latex = tuple(map(sympy.latex, children)) if self.print_name is not None: diff --git a/pybamm/expression_tree/functions.py b/pybamm/expression_tree/functions.py index 72c9d4074a..7f0986441f 100644 --- a/pybamm/expression_tree/functions.py +++ b/pybamm/expression_tree/functions.py @@ -5,6 +5,7 @@ import numpy as np from scipy import special +import sympy from typing import Sequence, Callable from typing_extensions import TypeVar @@ -210,7 +211,6 @@ def _sympy_operator(self, child): def to_equation(self): """Convert the node and its subtree into a SymPy equation.""" - sympy = have_optional_dependency("sympy") if self.print_name is not None: return sympy.Symbol(self.print_name) else: @@ -275,7 +275,6 @@ def _function_new_copy(self, children): def _sympy_operator(self, child): """Apply appropriate SymPy operators.""" - sympy = have_optional_dependency("sympy") class_name = self.__class__.__name__.lower() sympy_function = getattr(sympy, class_name) return sympy_function(child) @@ -332,7 +331,6 @@ def _function_diff(self, children, idx): def _sympy_operator(self, child): """Override :meth:`pybamm.Function._sympy_operator`""" - sympy = have_optional_dependency("sympy") return sympy.asinh(child) @@ -360,7 +358,6 @@ def _function_diff(self, children, idx): def _sympy_operator(self, child): """Override :meth:`pybamm.Function._sympy_operator`""" - sympy = have_optional_dependency("sympy") return sympy.atan(child) diff --git a/pybamm/expression_tree/independent_variable.py b/pybamm/expression_tree/independent_variable.py index 0dca6dba46..01e9dd661a 100644 --- a/pybamm/expression_tree/independent_variable.py +++ b/pybamm/expression_tree/independent_variable.py @@ -6,7 +6,6 @@ import numpy as np import pybamm -from pybamm.util import have_optional_dependency from pybamm.type_definitions import DomainType, AuxiliaryDomainType, DomainsType KNOWN_COORD_SYS = ["cartesian", "cylindrical polar", "spherical polar"] @@ -58,7 +57,6 @@ def _jac(self, variable) -> pybamm.Scalar: def to_equation(self) -> sympy.Symbol: """Convert the node and its subtree into a SymPy equation.""" - sympy = have_optional_dependency("sympy") if self.print_name is not None: return sympy.Symbol(self.print_name) else: @@ -102,7 +100,6 @@ def _evaluate_for_shape(self): def to_equation(self): """Convert the node and its subtree into a SymPy equation.""" - sympy = have_optional_dependency("sympy") return sympy.Symbol("t") diff --git a/pybamm/expression_tree/operations/latexify.py b/pybamm/expression_tree/operations/latexify.py index c16ab4b83d..f1c3734deb 100644 --- a/pybamm/expression_tree/operations/latexify.py +++ b/pybamm/expression_tree/operations/latexify.py @@ -9,7 +9,7 @@ import pybamm from pybamm.expression_tree.printing.sympy_overrides import custom_print_func -from pybamm.util import have_optional_dependency +import sympy def get_rng_min_max_name(rng, min_or_max): @@ -89,7 +89,6 @@ def _get_bcs_displays(self, var): Returns a list of boundary condition equations with ranges in front of the equations. """ - sympy = have_optional_dependency("sympy") bcs_eqn_list = [] bcs = self.model.boundary_conditions.get(var, None) @@ -120,7 +119,6 @@ def _get_bcs_displays(self, var): def _get_param_var(self, node): """Returns a list of parameters and a list of variables.""" - sympy = have_optional_dependency("sympy") param_list = [] var_list = [] dfs_nodes = [node] @@ -163,7 +161,6 @@ def _get_param_var(self, node): return param_list, var_list def latexify(self, output_variables=None): - sympy = have_optional_dependency("sympy") # Voltage is the default output variable if it exists if output_variables is None: if "Voltage [V]" in self.model.variables: diff --git a/pybamm/expression_tree/parameter.py b/pybamm/expression_tree/parameter.py index e646ff234d..4294d7ebc1 100644 --- a/pybamm/expression_tree/parameter.py +++ b/pybamm/expression_tree/parameter.py @@ -5,13 +5,11 @@ import sys import numpy as np -from typing import TYPE_CHECKING, Literal +from typing import Literal -if TYPE_CHECKING: # pragma: no cover - import sympy +import sympy import pybamm -from pybamm.util import have_optional_dependency class Parameter(pybamm.Symbol): @@ -48,7 +46,6 @@ def is_constant(self) -> Literal[False]: def to_equation(self) -> sympy.Symbol: """Convert the node and its subtree into a SymPy equation.""" - sympy = have_optional_dependency("sympy") if self.print_name is not None: return sympy.Symbol(self.print_name) else: @@ -244,7 +241,6 @@ def _evaluate_for_shape(self): def to_equation(self) -> sympy.Symbol: """Convert the node and its subtree into a SymPy equation.""" - sympy = have_optional_dependency("sympy") if self.print_name is not None: return sympy.Symbol(self.print_name) else: diff --git a/pybamm/expression_tree/scalar.py b/pybamm/expression_tree/scalar.py index 26bbabfcf0..a860ae45ec 100644 --- a/pybamm/expression_tree/scalar.py +++ b/pybamm/expression_tree/scalar.py @@ -3,10 +3,10 @@ # from __future__ import annotations import numpy as np +import sympy from typing import Literal import pybamm -from pybamm.util import have_optional_dependency from pybamm.type_definitions import Numeric @@ -87,7 +87,6 @@ def is_constant(self) -> Literal[True]: def to_equation(self): """Returns the value returned by the node when evaluated.""" - sympy = have_optional_dependency("sympy") if self.print_name is not None: return sympy.Symbol(self.print_name) else: diff --git a/pybamm/expression_tree/symbol.py b/pybamm/expression_tree/symbol.py index 70f4e82db6..5c4bd5f097 100644 --- a/pybamm/expression_tree/symbol.py +++ b/pybamm/expression_tree/symbol.py @@ -5,6 +5,7 @@ import numbers import numpy as np +import sympy from scipy.sparse import csr_matrix, issparse from functools import lru_cache, cached_property from typing import TYPE_CHECKING, Sequence, cast @@ -1056,7 +1057,6 @@ def print_name(self, name): self._print_name = prettify_print_name(name) def to_equation(self): - sympy = have_optional_dependency("sympy") return sympy.Symbol(str(self.name)) def to_json(self): diff --git a/pybamm/expression_tree/unary_operators.py b/pybamm/expression_tree/unary_operators.py index 319499a9fc..cb5bc905fd 100644 --- a/pybamm/expression_tree/unary_operators.py +++ b/pybamm/expression_tree/unary_operators.py @@ -5,6 +5,7 @@ import numpy as np from scipy.sparse import csr_matrix, issparse +import sympy import pybamm from pybamm.util import have_optional_dependency from pybamm.type_definitions import DomainsType @@ -108,7 +109,6 @@ def _sympy_operator(self, child): def to_equation(self): """Convert the node and its subtree into a SymPy equation.""" - sympy = have_optional_dependency("sympy") if self.print_name is not None: return sympy.Symbol(self.print_name) else: @@ -672,7 +672,6 @@ def _evaluates_on_edges(self, dimension: str) -> bool: def _sympy_operator(self, child): """Override :meth:`pybamm.UnaryOperator._sympy_operator`""" - sympy = have_optional_dependency("sympy") return sympy.Integral(child, sympy.Symbol("xn")) @@ -996,7 +995,6 @@ def _unary_new_copy(self, child): def _sympy_operator(self, child): """Override :meth:`pybamm.UnaryOperator._sympy_operator`""" - sympy = have_optional_dependency("sympy") if ( self.child.domain[0] in ["negative particle", "positive particle"] and self.side == "right" diff --git a/pybamm/expression_tree/variable.py b/pybamm/expression_tree/variable.py index eb0d90cdb6..238dab05fc 100644 --- a/pybamm/expression_tree/variable.py +++ b/pybamm/expression_tree/variable.py @@ -5,7 +5,7 @@ import numpy as np import numbers import pybamm -from pybamm.util import have_optional_dependency +import sympy from pybamm.type_definitions import ( DomainType, AuxiliaryDomainType, @@ -135,7 +135,6 @@ def _evaluate_for_shape(self): def to_equation(self): """Convert the node and its subtree into a SymPy equation.""" - sympy = have_optional_dependency("sympy") if self.print_name is not None: return sympy.Symbol(self.print_name) else: diff --git a/pybamm/models/base_model.py b/pybamm/models/base_model.py index b6b5a9b2da..194b8938af 100644 --- a/pybamm/models/base_model.py +++ b/pybamm/models/base_model.py @@ -13,7 +13,7 @@ import pybamm from pybamm.expression_tree.operations.serialise import Serialise -from pybamm.util import have_optional_dependency +import sympy class BaseModel: @@ -1185,7 +1185,6 @@ def latexify(self, filename=None, newline=True, output_variables=None): This will return first five model equations >>> model.latexify(newline=False)[1:5] """ - sympy = have_optional_dependency("sympy") if sympy: from pybamm.expression_tree.operations.latexify import Latexify diff --git a/pyproject.toml b/pyproject.toml index 13ea6df786..65e8f8fb48 100644 --- a/pyproject.toml +++ b/pyproject.toml @@ -41,6 +41,8 @@ dependencies = [ "casadi>=3.6.3", "xarray>=2022.6.0", "anytree>=2.8.0", + "sympy>=1.12", + "typing-extensions>=4.10.0", ] [project.urls] @@ -86,10 +88,6 @@ plot = [ cite = [ "pybtex>=0.24.0", ] -# To generate LaTeX strings -latexify = [ - "sympy>=1.12", -] # Battery Parameter eXchange format bpx = [ "bpx>=0.4.0",	[Bug]: non-required dependency `typing_extensions` is imported by default ### PyBaMM Version 24.1 ### Python Version 3.8-3.12 ### Describe the bug There might be some instance in #3578 where non required dependency has been imported by default, which is causing failure, [related build logs](https://github.com/arjxn-py/PyBaMM/actions/runs/8032837736/job/21942676799). ### Steps to Reproduce * try installing latest built wheels on develop * do `import pybamm` ### Relevant log output _No response_	Note: this happens with `sympy` as well Also, it's weird that the tests for `have_optional_dependency` in `test_util.py` didn't catch `typing-extensions` and `sympy` – is this because we are not mocking all of the optional dependencies inside that test case? IMO, we should mock all of the optional dependencies to ensure that this problem doesn't get repeated in the future.	2024-02-28T17:50:20	[]	[]
pybamm-team/PyBaMM	pybamm-team__PyBaMM-3829	84eb098e3140d774e39ab3d3c842c22462faf4aa	diff --git a/CHANGELOG.md b/CHANGELOG.md index bdacb00ae3..067e6ee025 100644 --- a/CHANGELOG.md +++ b/CHANGELOG.md @@ -9,6 +9,7 @@ ## Bug Fixes +- Fixed a bug where 1+1D and 2+1D models would not work with voltage or power controlled experiments([#3829](https://github.com/pybamm-team/PyBaMM/pull/3829)) - Updated `_steps_util.py` to throw a specific exception when drive cycle starts at t>0 ([#3756](https://github.com/pybamm-team/PyBaMM/pull/3756)) - Updated `plot_voltage_components.py` to support both `Simulation` and `Solution` objects. Added new methods in both `Simulation` and `Solution` classes for allow the syntax `simulation.plot_voltage_components` and `solution.plot_voltage_components`. Updated `test_plot_voltage_components.py` to reflect these changes ([#3723](https://github.com/pybamm-team/PyBaMM/pull/3723)). - The SEI thickness decreased at some intervals when the 'electron-migration limited' model was used. It has been corrected ([#3622](https://github.com/pybamm-team/PyBaMM/pull/3622)) diff --git a/pybamm/models/submodels/current_collector/potential_pair.py b/pybamm/models/submodels/current_collector/potential_pair.py index f2fd3aee83..68a9066da3 100644 --- a/pybamm/models/submodels/current_collector/potential_pair.py +++ b/pybamm/models/submodels/current_collector/potential_pair.py @@ -58,13 +58,12 @@ def set_algebraic(self, variables): } def set_initial_conditions(self, variables): - applied_current = self.param.current_with_time phi_s_cn = variables["Negative current collector potential [V]"] i_boundary_cc = variables["Current collector current density [A.m-2]"] self.initial_conditions = { phi_s_cn: pybamm.Scalar(0), - i_boundary_cc: applied_current, + i_boundary_cc: pybamm.Scalar(0), }	[Bug]: pouch models fail with voltage or power control ### PyBaMM Version 24.1 ### Python Version 3.11 ### Describe the bug Pouch cell simulations with voltage or power control fail with the error ``` TypeError: Initial conditions cannot contain 'Variable' objects, but 'Variable(-0x4b072f39e284004e, Current variable [A], children=[], domains={})' found in initial conditions for 'Current collector current density [A.m-2]' ``` ### Steps to Reproduce ```python3 import pybamm model = pybamm.lithium_ion.DFN( { "cell geometry": "pouch", "current collector": "potential pair", "dimensionality": 1, "thermal": "x-lumped" }, ) experiment = pybamm.Experiment(["Hold at 4.2 V until C/20 (5 seconds period)"]) sim = pybamm.Simulation(model, experiment = experiment) sol = sim.solve() ``` ### Relevant log output _No response_		2024-02-19T22:57:01	[]	[]
pybamm-team/PyBaMM	pybamm-team__PyBaMM-3811	84bfd6159baace67bad87d231c945f241ac20d50	diff --git a/noxfile.py b/noxfile.py index 380fae6af6..b74cfbab64 100644 --- a/noxfile.py +++ b/noxfile.py @@ -16,6 +16,7 @@ PYBAMM_ENV = { "SUNDIALS_INST": f"{homedir}/.local", "LD_LIBRARY_PATH": f"{homedir}/.local/lib", + "PIP_NO_BINARY": "scikits.odes", } VENV_DIR = Path("./venv").resolve() @@ -68,6 +69,15 @@ def run_coverage(session): if sys.version_info > (3, 12): session.install("-e", ".[all,jax]", silent=False) else: + session.run_always( + sys.executable, + "-m", + "pip", + "cache", + "remove", + "scikits.odes", + external=True, + ) session.install("-e", ".[all,jax,odes]", silent=False) else: if sys.version_info < (3, 9): @@ -91,6 +101,15 @@ def run_integration(session): if sys.version_info > (3, 12): session.install("-e", ".[all,jax]", silent=False) else: + session.run_always( + sys.executable, + "-m", + "pip", + "cache", + "remove", + "scikits.odes", + external=True, + ) session.install("-e", ".[all,jax,odes]", silent=False) else: if sys.version_info < (3, 9): @@ -123,6 +142,15 @@ def run_unit(session): if sys.version_info > (3, 12): session.install("-e", ".[all,jax]", silent=False) else: + session.run_always( + sys.executable, + "-m", + "pip", + "cache", + "remove", + "scikits.odes", + external=True, + ) session.install("-e", ".[all,jax,odes]", silent=False) else: if sys.version_info < (3, 9): @@ -180,6 +208,15 @@ def set_dev(session): external=True, ) else: + session.run_always( + sys.executable, + "-m", + "pip", + "cache", + "remove", + "scikits.odes", + external=True, + ) session.run( python, "-m", @@ -224,6 +261,15 @@ def run_tests(session): if sys.version_info > (3, 12): session.install("-e", ".[all,jax]", silent=False) else: + session.run_always( + sys.executable, + "-m", + "pip", + "cache", + "remove", + "scikits.odes", + external=True, + ) session.install("-e", ".[all,jax,odes]", silent=False) else: if sys.version_info < (3, 9):	Fortran compiler (`FC` variable) missing in macOS M-series runners, breaks SUNDIALS compilation See https://github.com/agriyakhetarpal/PyBaMM/actions/runs/7794200267/job/21255141901	I will look into the issue of the failing Python 3.11 unit tests here since that is a small fix	2024-02-07T18:48:07	[]	[]
pybamm-team/PyBaMM	pybamm-team__PyBaMM-3795	705baff8689106b546ab06746b91a4b8e672123f	diff --git a/pybamm/batch_study.py b/pybamm/batch_study.py index 4f1c37e71c..1503c3048c 100644 --- a/pybamm/batch_study.py +++ b/pybamm/batch_study.py @@ -196,7 +196,8 @@ def create_gif(self, number_of_images=80, duration=0.1, output_filename="plot.gi Name of the generated GIF file. """ - + if not hasattr(self, "sims"): + raise ValueError("The simulations have not been solved yet.") if self.quick_plot is None: self.quick_plot = pybamm.QuickPlot(self.sims) diff --git a/pybamm/simulation.py b/pybamm/simulation.py index 73c8188be7..a4c0d4bb32 100644 --- a/pybamm/simulation.py +++ b/pybamm/simulation.py @@ -1072,6 +1072,8 @@ def create_gif(self, number_of_images=80, duration=0.1, output_filename="plot.gi Name of the generated GIF file. """ + if self.solution is None: + raise ValueError("The simulation has not been solved yet.") if self.quick_plot is None: self.quick_plot = pybamm.QuickPlot(self._solution)	Check if the solution exists before creating a GIF (`Simulation.create_gif` and `BatchStudy.create_gif`) I liked how #3723 checks if the simulations have been solved before proceeding ahead to the plotting functionality. It would be nice to have similar checks in - https://github.com/pybamm-team/PyBaMM/blob/c3def31a04df163769851733657d9b49b3700bca/pybamm/batch_study.py#L184-L187 and https://github.com/pybamm-team/PyBaMM/blob/c3def31a04df163769851733657d9b49b3700bca/pybamm/simulation.py#L1047-L1050	Would like to take this on. Hi @prady0t are you still working on this issue? yes @Saransh-cpp BatchStudy class doesn't have an attribute `solution` unlike Simulation. How to check for empty solutions there? You can check if the `self.sims` list is empty and throw an error if it is - https://github.com/pybamm-team/PyBaMM/blob/c3def31a04df163769851733657d9b49b3700bca/pybamm/batch_study.py#L116 The simulations are being initialised and solved in the `solve()` method, so this would be a good indication of if the user has called the `solve()` method before calling the `create_gif()` method. why is https://github.com/pybamm-team/PyBaMM/blob/063a3830afd5547f7a78e283ae81058465cbdd3e/pybamm/batch_study.py#L116 declared outside `__init__` should'nt it be inside it?	2024-02-01T09:30:59	[]	[]
pybamm-team/PyBaMM	pybamm-team__PyBaMM-3708	64681dd7ee9ca6e9c6c18a9e704e5d984a90466c	diff --git a/CHANGELOG.md b/CHANGELOG.md index 17adc3a31f..89b104298c 100644 --- a/CHANGELOG.md +++ b/CHANGELOG.md @@ -1,5 +1,8 @@ # [Unreleased](https://github.com/pybamm-team/PyBaMM/) +## Bug Fixes + +- Fixed a bug where if the first step(s) in a cycle are skipped then the cycle solution started from the model's initial conditions instead of from the last state of the previous cycle ([#3708](https://github.com/pybamm-team/PyBaMM/pull/3708)) # [v24.1rc0](https://github.com/pybamm-team/PyBaMM/tree/v24.1rc0) - 2024-01-31 ## Features diff --git a/pybamm/simulation.py b/pybamm/simulation.py index c95ab3039c..8a6150cc4e 100644 --- a/pybamm/simulation.py +++ b/pybamm/simulation.py @@ -839,7 +839,20 @@ def solve( steps.append(step_solution) - cycle_solution = cycle_solution + step_solution + # If there haven't been any successful steps yet in this cycle, then + # carry the solution over from the previous cycle (but + # `step_solution` should still be an EmptySolution so that in the + # list of returned step solutions we can see which steps were + # skipped) + if ( + cycle_solution is None + and isinstance(step_solution, pybamm.EmptySolution) + and not isinstance(current_solution, pybamm.EmptySolution) + ): + cycle_solution = current_solution.last_state + else: + cycle_solution = cycle_solution + step_solution + current_solution = cycle_solution callbacks.on_step_end(logs)	[Bug]: Some variables not copied correctly if experiment steps are skipped ### PyBaMM Version 23,9 ### Python Version 3.11 ### Describe the bug Sometimes experiement steps are skipped (e.g. skip a charge if you're already at that voltage). When this happens the solver returns an `EmptySolution`. The next step of the experiment then starts from the wrong point. ### Steps to Reproduce ```python3 import pybamm import matplotlib.pyplot as plt model = pybamm.lithium_ion.SPMe( { "SEI": "solvent-diffusion limited", } ) exp = pybamm.Experiment( [(f"Rest for 24 hours (1 hour period)",)] + [ ( "Charge at C/3 until 4.1 V", "Hold at 4.1V until C/20", "Discharge at C/3 until 2.5 V", ) ] ) sim1 = pybamm.Simulation(model, experiment=exp) sim1.solve(initial_soc=1) sim2 = pybamm.Simulation(model, experiment=exp) sim2.solve(initial_soc=0.9) sols = [sim1.solution, sim2.solution] labels = ["SOC=1", "SOC=0.9"] fig, ax = plt.subplots(2, 2, figsize=(10, 4)) for sol, label in zip(sols, labels): ax[0,0].plot(sol["Time [h]"].data, sol["Terminal voltage [V]"].data, label=label) ax[0,1].plot(sol["Time [h]"].data, sol["X-averaged negative particle surface concentration [mol.m-3]"].data, label=label) ax[1,0].plot(sol["Time [h]"].data, sol["Total capacity lost to side reactions [A.h]"].data, label=label) ax[1,1].plot(sol["Time [h]"].data, sol["X-averaged negative total SEI thickness [m]"].data, label=label) ax[0, 0].legend() sim1.solution.cycles[1].steps sim2.solution.cycles[1].steps ``` ![image](https://github.com/pybamm-team/PyBaMM/assets/43040151/d3d0d3cf-9a58-4ee0-a69a-8df6e657ea00) ### Relevant log output _No response_		2024-01-10T16:13:59	[]	[]
pybamm-team/PyBaMM	pybamm-team__PyBaMM-3627	2e354155e447fc3b3947c75375012942b271dd1d	diff --git a/CHANGELOG.md b/CHANGELOG.md index fdaf360d6f..b1e093f452 100644 --- a/CHANGELOG.md +++ b/CHANGELOG.md @@ -3,9 +3,10 @@ ## Features - Added support for macOS arm64 (M-series) platforms. ([#3789](https://github.com/pybamm-team/PyBaMM/pull/3789)) +- Added the ability to specify a custom solver tolerance in `get_initial_stoichiometries` and related functions ([#3714](https://github.com/pybamm-team/PyBaMM/pull/3714)) +- Modified `step` function to take an array of time `t_eval` as an argument and deprecated use of `npts`. ([#3627](https://github.com/pybamm-team/PyBaMM/pull/3627)) - Renamed "electrode diffusivity" to "particle diffusivity" as a non-breaking change with a deprecation warning ([#3624](https://github.com/pybamm-team/PyBaMM/pull/3624)) - Add support for BPX version 0.4.0 which allows for blended electrodes and user-defined parameters in BPX([#3414](https://github.com/pybamm-team/PyBaMM/pull/3414)) -- Added the ability to specify a custom solver tolerance in `get_initial_stoichiometries` and related functions ([#3714](https://github.com/pybamm-team/PyBaMM/pull/3714)) ## Bug Fixes diff --git a/examples/scripts/SPMe_step.py b/examples/scripts/SPMe_step.py index 90d6f3d017..f277c0e790 100644 --- a/examples/scripts/SPMe_step.py +++ b/examples/scripts/SPMe_step.py @@ -32,12 +32,14 @@ # step model dt = 500 +# t_eval is an array of time in the interval 0 to dt, dt being size of the step. +t_eval = np.array([0, 50, 100, 200, 500]) time = 0 end_time = solution.t[-1] step_solver = pybamm.CasadiSolver() step_solution = None while time < end_time: - step_solution = step_solver.step(step_solution, model, dt=dt, npts=10) + step_solution = step_solver.step(step_solution, model, dt=dt, t_eval=t_eval) time += dt # plot diff --git a/pybamm/simulation.py b/pybamm/simulation.py index a2b260ab43..3dfdac94b5 100644 --- a/pybamm/simulation.py +++ b/pybamm/simulation.py @@ -789,7 +789,7 @@ def solve( current_solution, model, dt, - npts=npts, + t_eval=np.linspace(0, dt, npts), save=False, kwargs, ) @@ -970,7 +970,7 @@ def run_padding_rest(self, kwargs, rest_time, step_solution): step_solution, model, rest_time, - npts=npts, + t_eval=np.linspace(0, rest_time, npts), save=False, kwargs, ) @@ -978,7 +978,13 @@ def run_padding_rest(self, kwargs, rest_time, step_solution): return step_solution_with_rest def step( - self, dt, solver=None, npts=2, save=True, starting_solution=None, kwargs + self, + dt, + solver=None, + t_eval=None, + save=True, + starting_solution=None, + kwargs, ): """ A method to step the model forward one timestep. This method will @@ -990,9 +996,10 @@ def step( The timestep over which to step the solution solver : :class:`pybamm.BaseSolver` The solver to use to solve the model. - npts : int, optional - The number of points at which the solution will be returned during - the step dt. Default is 2 (returns the solution at t0 and t0 + dt). + t_eval : list or numpy.ndarray, optional + An array of times at which to return the solution during the step + (Note: t_eval is the time measured from the start of the step, so should start at 0 and end at dt). + By default, the solution is returned at t0 and t0 + dt. save : bool Turn on to store the solution of all previous timesteps starting_solution : :class:`pybamm.Solution` @@ -1012,7 +1019,12 @@ def step( starting_solution = self._solution self._solution = solver.step( - starting_solution, self.built_model, dt, npts=npts, save=save, kwargs + starting_solution, + self.built_model, + dt, + t_eval=t_eval, + save=save, + kwargs, ) return self.solution diff --git a/pybamm/solvers/base_solver.py b/pybamm/solvers/base_solver.py index a2b4c305c2..be26306bbd 100644 --- a/pybamm/solvers/base_solver.py +++ b/pybamm/solvers/base_solver.py @@ -1090,7 +1090,8 @@ def step( old_solution, model, dt, - npts=2, + t_eval=None, + npts=None, inputs=None, save=True, ): @@ -1108,9 +1109,11 @@ def step( initial_conditions dt : numeric type The timestep (in seconds) over which to step the solution - npts : int, optional - The number of points at which the solution will be returned during - the step dt. default is 2 (returns the solution at t0 and t0 + dt). + t_eval : list or numpy.ndarray, optional + An array of times at which to return the solution during the step + (Note: t_eval is the time measured from the start of the step, so should start at 0 and end at dt). + By default, the solution is returned at t0 and t0 + dt. + npts : deprecated inputs : dict, optional Any input parameters to pass to the model when solving save : bool @@ -1149,10 +1152,28 @@ def step( f"Step time must be at least {pybamm.TimerTime(step_start_offset)}" ) + # Raise deprecation warning for npts and convert it to t_eval + if npts is not None: + warnings.warn( + "The 'npts' parameter is deprecated, use 't_eval' instead.", + DeprecationWarning, + stacklevel=2, + ) + t_eval = np.linspace(0, dt, npts) + + if t_eval is not None: + # Checking if t_eval lies within range + if t_eval[0] != 0 or t_eval[-1] != dt: + raise pybamm.SolverError( + "Elements inside array t_eval must lie in the closed interval 0 to dt" + ) + + else: + t_eval = np.array([0, dt]) + t_start = old_solution.t[-1] + t_eval = t_start + t_eval t_end = t_start + dt - # Calculate t_eval - t_eval = np.linspace(t_start, t_end, npts) if t_start == 0: t_start_shifted = t_start	Pass array of times to `step` When you call `solver.step` you pass in `dt`, the time time increment you want to step forward in. You can also pass `npts` to get the solution back at a linearly spaced array of times between the initial time, `t0`, and `t0+dt`. It would be good to instead be able to pass an array of times, `t_eval`, like you pass to solve, so you can get step solutions returned at non-linearly spaced times. The motivation for this is that right now you can't pass a `starting_solution` to `Simualtion.solve` if you aren't using an experiment, so if you want to chain solutions together you have to call `Simulation.step` instead. An alternative is that we allow starting solutions without experiments in `Simulation.solve`. The whole logic for `step` and `solve` is pretty clunky both in `Simulation` and `BaseSolver`, but that's a separate issue...	Hey I would like to work on this. Can you please give examples of how to use step() function to get solutions which I can use for debugging? I'm quite new here and I cant find it in project doc. Sorry for the slow reply. Take a look [here](https://github.com/pybamm-team/PyBaMM/blob/develop/examples/scripts/SPMe_step.py). Thanks 😄 Maybe if we change step function to something like: ```py def step( self, old_solution, model, dt=0, non_linear_time=None, npts=2, inputs=None, save=True, ) ``` Here `non_linear_time` is an array of time (example: [0,10,15,17,20]). We can assign `dt = non_linear_time[i] - non_linear_time[i-1]` inside a loop and make changes to the function accordingly. Is this a correct solution? > Here `non_linear_time` is an array of time (example: [0,10,15,17,20]). Hi @prady0t, theoretically it looks promising, though it would be nicer to open a PR to keep things moving & also would be easier for us to suggest changes there 🙂 Sure will do!	2023-12-17T07:27:09	[]	[]
pybamm-team/PyBaMM	pybamm-team__PyBaMM-3557	a4632463c7a9bed4f644bc7c1549726c9d976769	diff --git a/.coveragerc b/.coveragerc deleted file mode 100644 index a174f5aced..0000000000 --- a/.coveragerc +++ /dev/null @@ -1,3 +0,0 @@ -[run] -source = pybamm -concurrency = multiprocessing diff --git a/.git-blame-ignore-revs b/.git-blame-ignore-revs index 228a76373e..9e59bd7f07 100644 --- a/.git-blame-ignore-revs +++ b/.git-blame-ignore-revs @@ -4,3 +4,5 @@ a63e49ece0f9336d1f5c2562f7459e555c6e6693 # activated standard pre-commits - https://github.com/pybamm-team/PyBaMM/pull/3192 5273214b585c5a4286609aed40e0b092d0e05f42 +# migrate config to pyproject.toml - https://github.com/pybamm-team/PyBaMM/pull/3557 +12c5d77203bd93542785d237bac00bad5ed5469a diff --git a/benchmarks/work_precision_sets/time_vs_abstols.py b/benchmarks/work_precision_sets/time_vs_abstols.py index 6447884083..9a96f07514 100644 --- a/benchmarks/work_precision_sets/time_vs_abstols.py +++ b/benchmarks/work_precision_sets/time_vs_abstols.py @@ -96,7 +96,7 @@ plt.savefig(f"benchmarks/benchmark_images/time_vs_abstols_{pybamm.__version__}.png") -content = f"# PyBaMM {pybamm.__version__}\n## Solve Time vs Abstols\n<img src='./benchmark_images/time_vs_abstols_{pybamm.__version__}.png'>\n" # noqa +content = f"# PyBaMM {pybamm.__version__}\n## Solve Time vs Abstols\n<img src='./benchmark_images/time_vs_abstols_{pybamm.__version__}.png'>\n" with open("./benchmarks/release_work_precision_sets.md", "r") as original: data = original.read() diff --git a/benchmarks/work_precision_sets/time_vs_dt_max.py b/benchmarks/work_precision_sets/time_vs_dt_max.py index 3926a4bcd6..3e428b702c 100644 --- a/benchmarks/work_precision_sets/time_vs_dt_max.py +++ b/benchmarks/work_precision_sets/time_vs_dt_max.py @@ -98,7 +98,7 @@ plt.savefig(f"benchmarks/benchmark_images/time_vs_dt_max_{pybamm.__version__}.png") -content = f"## Solve Time vs dt_max\n<img src='./benchmark_images/time_vs_dt_max_{pybamm.__version__}.png'>\n" # noqa +content = f"## Solve Time vs dt_max\n<img src='./benchmark_images/time_vs_dt_max_{pybamm.__version__}.png'>\n" with open("./benchmarks/release_work_precision_sets.md", "r") as original: data = original.read() diff --git a/benchmarks/work_precision_sets/time_vs_mesh_size.py b/benchmarks/work_precision_sets/time_vs_mesh_size.py index 7b4d4145d4..f0f13f706b 100644 --- a/benchmarks/work_precision_sets/time_vs_mesh_size.py +++ b/benchmarks/work_precision_sets/time_vs_mesh_size.py @@ -78,7 +78,7 @@ plt.savefig(f"benchmarks/benchmark_images/time_vs_mesh_size_{pybamm.__version__}.png") -content = f"## Solve Time vs Mesh size\n<img src='./benchmark_images/time_vs_mesh_size_{pybamm.__version__}.png'>\n" # noqa +content = f"## Solve Time vs Mesh size\n<img src='./benchmark_images/time_vs_mesh_size_{pybamm.__version__}.png'>\n" with open("./benchmarks/release_work_precision_sets.md", "r") as original: data = original.read() diff --git a/benchmarks/work_precision_sets/time_vs_no_of_states.py b/benchmarks/work_precision_sets/time_vs_no_of_states.py index 0a88ac8b52..eb27aba322 100644 --- a/benchmarks/work_precision_sets/time_vs_no_of_states.py +++ b/benchmarks/work_precision_sets/time_vs_no_of_states.py @@ -82,7 +82,7 @@ ) -content = f"## Solve Time vs Number of states\n<img src='./benchmark_images/time_vs_no_of_states_{pybamm.__version__}.png'>\n" # noqa +content = f"## Solve Time vs Number of states\n<img src='./benchmark_images/time_vs_no_of_states_{pybamm.__version__}.png'>\n" with open("./benchmarks/release_work_precision_sets.md", "r") as original: data = original.read() diff --git a/benchmarks/work_precision_sets/time_vs_reltols.py b/benchmarks/work_precision_sets/time_vs_reltols.py index 12e41b526f..93964910a8 100644 --- a/benchmarks/work_precision_sets/time_vs_reltols.py +++ b/benchmarks/work_precision_sets/time_vs_reltols.py @@ -102,7 +102,7 @@ plt.savefig(f"benchmarks/benchmark_images/time_vs_reltols_{pybamm.__version__}.png") -content = f"## Solve Time vs Reltols\n<img src='./benchmark_images/time_vs_reltols_{pybamm.__version__}.png'>\n" # noqa +content = f"## Solve Time vs Reltols\n<img src='./benchmark_images/time_vs_reltols_{pybamm.__version__}.png'>\n" with open("./benchmarks/release_work_precision_sets.md", "r") as original: data = original.read() diff --git a/docs/source/examples/notebooks/models/jelly-roll-model.ipynb b/docs/source/examples/notebooks/models/jelly-roll-model.ipynb index fe6173f1ce..557366099a 100644 --- a/docs/source/examples/notebooks/models/jelly-roll-model.ipynb +++ b/docs/source/examples/notebooks/models/jelly-roll-model.ipynb @@ -84,7 +84,7 @@ "delta = pybamm.Parameter(\"Current collector thickness\")\n", "delta_p = delta # assume same thickness\n", "delta_n = delta # assume same thickness\n", - "l = 1/2 - delta_p - delta_n # active material thickness # noqa: E741\n", + "l = 1/2 - delta_p - delta_n # active material thickness\n", "sigma_p = pybamm.Parameter(\"Positive current collector conductivity\")\n", "sigma_n = pybamm.Parameter(\"Negative current collector conductivity\")\n", "sigma_a = pybamm.Parameter(\"Active material conductivity\")" diff --git a/docs/source/examples/notebooks/models/latexify.ipynb b/docs/source/examples/notebooks/models/latexify.ipynb index 63e7c0d519..c2a45ff2c8 100644 --- a/docs/source/examples/notebooks/models/latexify.ipynb +++ b/docs/source/examples/notebooks/models/latexify.ipynb @@ -1252,7 +1252,7 @@ "source": [ "spme_latex = model_spme.latexify(newline=False)\n", "for line in spme_latex:\n", - " display(line) # noqa: F821" + " display(line)" ] }, { diff --git a/docs/source/user_guide/installation/index.rst b/docs/source/user_guide/installation/index.rst index 983f66842e..e771611a37 100644 --- a/docs/source/user_guide/installation/index.rst +++ b/docs/source/user_guide/installation/index.rst @@ -154,6 +154,7 @@ Dependency `pre-commit <https://pre-commit.com/index.html>`__ \- dev For managing and maintaining multi-language pre-commit hooks. `ruff <https://beta.ruff.rs/docs/>`__ \- dev For code formatting. `nox <https://nox.thea.codes/en/stable/>`__ \- dev For running testing sessions in multiple environments. +`coverage <https://coverage.readthedocs.io/en/>`__ \- dev For calculating coverage of tests. `pytest <https://docs.pytest.org/en/stable/>`__ 6.0.0 dev For running Jupyter notebooks tests. `pytest-xdist <https://pytest-xdist.readthedocs.io/en/latest/>`__ \- dev For running tests in parallel across distributed workers. `nbmake <https://github.com/treebeardtech/nbmake/>`__ \- dev A ``pytest`` plugin for executing Jupyter notebooks. diff --git a/examples/scripts/print_model_parameter_combinations.py b/examples/scripts/print_model_parameter_combinations.py index 8d24f919c3..f7f00714fb 100644 --- a/examples/scripts/print_model_parameter_combinations.py +++ b/examples/scripts/print_model_parameter_combinations.py @@ -19,13 +19,13 @@ try: model = pybamm.lithium_ion.SPM(options.copy()) except pybamm.OptionError as e: - print(f"Cannot create model with {options}. (OptionError: {str(e)})") + print(f"Cannot create model with {options}. (OptionError: {e!s})") except pybamm.ModelError as e: # todo: properly resolve the cases that raise these errors - print(f"Cannot create model with {options}. (ModelError: {str(e)})") + print(f"Cannot create model with {options}. (ModelError: {e!s})") except AttributeError as e: # todo: properly resolve the cases that raise these errors - print(f"Cannot create model with {options}. (AttributeError: {str(e)})") + print(f"Cannot create model with {options}. (AttributeError: {e!s})") else: output = f"{options} with {parameter_set} parameters: " try: diff --git a/noxfile.py b/noxfile.py index 7a57ad5820..5db08ee92f 100644 --- a/noxfile.py +++ b/noxfile.py @@ -38,7 +38,7 @@ def set_environment_variables(env_dict, session): @nox.session(name="pybamm-requires") def run_pybamm_requires(session): - """Download, compile, and install the build-time requirements for Linux and macOS: the SuiteSparse and SUNDIALS libraries.""" # noqa: E501 + """Download, compile, and install the build-time requirements for Linux and macOS: the SuiteSparse and SUNDIALS libraries.""" set_environment_variables(PYBAMM_ENV, session=session) if sys.platform != "win32": session.install("wget", "cmake", silent=False) @@ -64,7 +64,7 @@ def run_coverage(session): session.install("-e", ".[all,odes,jax]", silent=False) else: session.install("-e", ".[all]", silent=False) - session.run("coverage", "run", "--rcfile=.coveragerc", "run-tests.py", "--nosub") + session.run("coverage", "run", "run-tests.py", "--nosub") session.run("coverage", "combine") session.run("coverage", "xml") diff --git a/pybamm/citations.py b/pybamm/citations.py index b72262989b..e73351a4c6 100644 --- a/pybamm/citations.py +++ b/pybamm/citations.py @@ -267,7 +267,7 @@ def print_citations(filename=None, output_format="text", verbose=False): if verbose: # pragma: no cover if filename is not None: # pragma: no cover raise Exception( - "Verbose output is available only for the terminal and not for printing to files", # noqa: E501 + "Verbose output is available only for the terminal and not for printing to files", ) else: citations.print(filename, output_format, verbose=True) diff --git a/pybamm/discretisations/discretisation.py b/pybamm/discretisations/discretisation.py index a120adecc0..bb6e678f4c 100644 --- a/pybamm/discretisations/discretisation.py +++ b/pybamm/discretisations/discretisation.py @@ -461,7 +461,7 @@ def process_boundary_conditions(self, model): if ( self.mesh[subdomain].coord_sys in ["spherical polar", "cylindrical polar"] - and list(self.mesh.geometry[subdomain].values())[0]["min"] == 0 + and next(iter(self.mesh.geometry[subdomain].values()))["min"] == 0 ): if bcs["left"][0].value != 0 or bcs["left"][1] != "Neumann": raise pybamm.ModelError( @@ -753,7 +753,7 @@ def _process_symbol(self, symbol): spatial_method = self.spatial_methods[symbol.domain[0]] # If boundary conditions are provided, need to check for BCs on tabs if self.bcs: - key_id = list(self.bcs.keys())[0] + key_id = next(iter(self.bcs.keys())) if any("tab" in side for side in list(self.bcs[key_id].keys())): self.bcs[key_id] = self.check_tab_conditions( symbol, self.bcs[key_id] diff --git a/pybamm/expression_tree/array.py b/pybamm/expression_tree/array.py index 2736886d95..adbdc88dc2 100644 --- a/pybamm/expression_tree/array.py +++ b/pybamm/expression_tree/array.py @@ -97,7 +97,7 @@ def entries_string(self, value): def set_id(self): """See :meth:`pybamm.Symbol.set_id()`.""" self._id = hash( - (self.__class__, self.name) + self.entries_string + tuple(self.domain) + (self.__class__, self.name, self.entries_string, tuple(self.domain)) ) def _jac(self, variable): diff --git a/pybamm/expression_tree/interpolant.py b/pybamm/expression_tree/interpolant.py index cd0df4d077..28188ce68f 100644 --- a/pybamm/expression_tree/interpolant.py +++ b/pybamm/expression_tree/interpolant.py @@ -222,9 +222,7 @@ def entries_string(self, value): def set_id(self): """See :meth:`pybamm.Symbol.set_id()`.""" self._id = hash( - (self.__class__, self.name, self.entries_string) - + tuple([child.id for child in self.children]) - + tuple(self.domain) + (self.__class__, self.name, self.entries_string, tuple([child.id for child in self.children]), tuple(self.domain)) ) def _function_new_copy(self, children): diff --git a/pybamm/expression_tree/operations/evaluate_python.py b/pybamm/expression_tree/operations/evaluate_python.py index 1f44a69784..d0cd4c776d 100644 --- a/pybamm/expression_tree/operations/evaluate_python.py +++ b/pybamm/expression_tree/operations/evaluate_python.py @@ -42,7 +42,7 @@ class JaxCooMatrix: def __init__(self, row, col, data, shape): if not pybamm.have_jax(): # pragma: no cover raise ModuleNotFoundError( - "Jax or jaxlib is not installed, please see https://docs.pybamm.org/en/latest/source/user_guide/installation/GNU-linux.html#optional-jaxsolver" # noqa: E501 + "Jax or jaxlib is not installed, please see https://docs.pybamm.org/en/latest/source/user_guide/installation/GNU-linux.html#optional-jaxsolver" ) self.row = jax.numpy.array(row) @@ -537,7 +537,7 @@ class EvaluatorJax: def __init__(self, symbol): if not pybamm.have_jax(): # pragma: no cover raise ModuleNotFoundError( - "Jax or jaxlib is not installed, please see https://docs.pybamm.org/en/latest/source/user_guide/installation/GNU-linux.html#optional-jaxsolver" # noqa: E501 + "Jax or jaxlib is not installed, please see https://docs.pybamm.org/en/latest/source/user_guide/installation/GNU-linux.html#optional-jaxsolver" ) constants, python_str = pybamm.to_python(symbol, debug=False, output_jax=True) diff --git a/pybamm/expression_tree/operations/latexify.py b/pybamm/expression_tree/operations/latexify.py index 9f2949069e..572f01a560 100644 --- a/pybamm/expression_tree/operations/latexify.py +++ b/pybamm/expression_tree/operations/latexify.py @@ -93,9 +93,9 @@ def _get_bcs_displays(self, var): if bcs: # Take range minimum from the first domain - var_name = list(self.model.default_geometry[var.domain[0]].keys())[0] - rng_left = list(self.model.default_geometry[var.domain[0]].values())[0] - rng_right = list(self.model.default_geometry[var.domain[-1]].values())[0] + var_name = next(iter(self.model.default_geometry[var.domain[0]].keys())) + rng_left = next(iter(self.model.default_geometry[var.domain[0]].values())) + rng_right = next(iter(self.model.default_geometry[var.domain[-1]].values())) # Trim name (r_n --> r) var_name = re.findall(r"(.)_.", str(var_name))[0] diff --git a/pybamm/expression_tree/parameter.py b/pybamm/expression_tree/parameter.py index eebe77ad2f..00a28017b8 100644 --- a/pybamm/expression_tree/parameter.py +++ b/pybamm/expression_tree/parameter.py @@ -152,9 +152,7 @@ def input_names(self, inp=None): def set_id(self): """See :meth:`pybamm.Symbol.set_id`""" self._id = hash( - (self.__class__, self.name, self.diff_variable) - + tuple([child.id for child in self.children]) - + tuple(self.domain) + (self.__class__, self.name, self.diff_variable, tuple([child.id for child in self.children]), tuple(self.domain)) ) def diff(self, variable): diff --git a/pybamm/expression_tree/state_vector.py b/pybamm/expression_tree/state_vector.py index 6ef8bee904..437ba752ed 100644 --- a/pybamm/expression_tree/state_vector.py +++ b/pybamm/expression_tree/state_vector.py @@ -107,8 +107,7 @@ def set_evaluation_array(self, y_slices, evaluation_array): def set_id(self): """See :meth:`pybamm.Symbol.set_id()`""" self._id = hash( - (self.__class__, self.name, tuple(self.evaluation_array)) - + tuple(self.domain) + (self.__class__, self.name, tuple(self.evaluation_array), tuple(self.domain)) ) def _jac_diff_vector(self, variable): diff --git a/pybamm/expression_tree/symbol.py b/pybamm/expression_tree/symbol.py index 8f1608e7ba..5fe765af33 100644 --- a/pybamm/expression_tree/symbol.py +++ b/pybamm/expression_tree/symbol.py @@ -403,9 +403,7 @@ def set_id(self): need to hash once. """ self._id = hash( - (self.__class__, self.name) - + tuple([child.id for child in self.children]) - + tuple([(k, tuple(v)) for k, v in self.domains.items() if v != []]) + (self.__class__, self.name, tuple([child.id for child in self.children]), tuple([(k, tuple(v)) for k, v in self.domains.items() if v != []])) ) @property diff --git a/pybamm/expression_tree/unary_operators.py b/pybamm/expression_tree/unary_operators.py index 81c3dc28c2..95306ebad5 100644 --- a/pybamm/expression_tree/unary_operators.py +++ b/pybamm/expression_tree/unary_operators.py @@ -287,14 +287,7 @@ def _unary_jac(self, child_jac): def set_id(self): """See :meth:`pybamm.Symbol.set_id()`""" self._id = hash( - ( - self.__class__, - self.name, - self.slice.start, - self.slice.stop, - self.children[0].id, - ) - + tuple(self.domain) + (self.__class__, self.name, self.slice.start, self.slice.stop, self.children[0].id, tuple(self.domain)) ) def _unary_evaluate(self, child): @@ -554,15 +547,7 @@ def integration_variable(self): def set_id(self): """See :meth:`pybamm.Symbol.set_id()`""" self._id = hash( - (self.__class__, self.name) - + tuple( - [ - integration_variable.id - for integration_variable in self.integration_variable - ] - ) - + (self.children[0].id,) - + tuple(self.domain) + (self.__class__, self.name, tuple([integration_variable.id for integration_variable in self.integration_variable]), self.children[0].id, tuple(self.domain)) ) def _unary_new_copy(self, child): @@ -702,9 +687,7 @@ def __init__(self, child, vector_type="row"): def set_id(self): """See :meth:`pybamm.Symbol.set_id()`""" self._id = hash( - (self.__class__, self.name, self.vector_type) - + (self.children[0].id,) - + tuple(self.domain) + (self.__class__, self.name, self.vector_type, self.children[0].id, tuple(self.domain)) ) def _unary_new_copy(self, child): @@ -757,7 +740,7 @@ def __init__(self, child, region="entire"): def set_id(self): """See :meth:`pybamm.Symbol.set_id()`""" self._id = hash( - (self.__class__, self.name) + (self.children[0].id,) + tuple(self.domain) + (self.__class__, self.name, self.children[0].id, tuple(self.domain)) ) def _unary_new_copy(self, child): @@ -798,8 +781,7 @@ def __init__(self, child, side, domain): def set_id(self): """See :meth:`pybamm.Symbol.set_id()`""" self._id = hash( - (self.__class__, self.name, self.side, self.children[0].id) - + tuple([(k, tuple(v)) for k, v in self.domains.items()]) + (self.__class__, self.name, self.side, self.children[0].id, tuple([(k, tuple(v)) for k, v in self.domains.items()])) ) def _evaluates_on_edges(self, dimension): @@ -857,8 +839,7 @@ def __init__(self, name, child, side): def set_id(self): """See :meth:`pybamm.Symbol.set_id()`""" self._id = hash( - (self.__class__, self.name, self.side, self.children[0].id) - + tuple([(k, tuple(v)) for k, v in self.domains.items()]) + (self.__class__, self.name, self.side, self.children[0].id, tuple([(k, tuple(v)) for k, v in self.domains.items()])) ) def _unary_new_copy(self, child): diff --git a/pybamm/expression_tree/variable.py b/pybamm/expression_tree/variable.py index 0d1e1fd424..22b176b6fc 100644 --- a/pybamm/expression_tree/variable.py +++ b/pybamm/expression_tree/variable.py @@ -103,8 +103,7 @@ def bounds(self, values): def set_id(self): self._id = hash( - (self.__class__, self.name, self.scale, self.reference) - + tuple([(k, tuple(v)) for k, v in self.domains.items() if v != []]) + (self.__class__, self.name, self.scale, self.reference, tuple([(k, tuple(v)) for k, v in self.domains.items() if v != []])) ) def create_copy(self): diff --git a/pybamm/install_odes.py b/pybamm/install_odes.py index 4bf310a0f2..0fbbcdc637 100644 --- a/pybamm/install_odes.py +++ b/pybamm/install_odes.py @@ -66,7 +66,7 @@ def install_sundials(download_dir, install_dir): print("-" 10, "Running CMake prepare", "-" * 40) subprocess.run( - ["cmake", "../sundials-{}".format(sundials_version)] + cmake_args, + ["cmake", "../sundials-{}".format(sundials_version), cmake_args], cwd=build_directory, check=True, ) diff --git a/pybamm/meshes/zero_dimensional_submesh.py b/pybamm/meshes/zero_dimensional_submesh.py index 5b2f38e29f..82e8cb6524 100644 --- a/pybamm/meshes/zero_dimensional_submesh.py +++ b/pybamm/meshes/zero_dimensional_submesh.py @@ -31,7 +31,7 @@ def __init__(self, position, npts=None): raise pybamm.GeometryError("position should only contain a single variable") # extract the position - position = list(position.values())[0] + position = next(iter(position.values())) spatial_position = position["position"] self.nodes = np.array([spatial_position]) self.edges = np.array([spatial_position]) diff --git a/pybamm/models/base_model.py b/pybamm/models/base_model.py index 08890757b7..a88f6f4255 100644 --- a/pybamm/models/base_model.py +++ b/pybamm/models/base_model.py @@ -1205,7 +1205,7 @@ def check_and_convert_equations(self, equations): for var, eqn in equations.items(): if eqn.has_symbol_of_classes(pybamm.Variable): unpacker = pybamm.SymbolUnpacker(pybamm.Variable) - variable_in_equation = list(unpacker.unpack_symbol(eqn))[0] + variable_in_equation = next(iter(unpacker.unpack_symbol(eqn))) raise TypeError( "Initial conditions cannot contain 'Variable' objects, " "but '{!r}' found in initial conditions for '{}'".format( diff --git a/pybamm/models/submodels/electrolyte_conductivity/surface_potential_form/full_surface_form_conductivity.py b/pybamm/models/submodels/electrolyte_conductivity/surface_potential_form/full_surface_form_conductivity.py index 7afeeac47e..83bcfb8027 100644 --- a/pybamm/models/submodels/electrolyte_conductivity/surface_potential_form/full_surface_form_conductivity.py +++ b/pybamm/models/submodels/electrolyte_conductivity/surface_potential_form/full_surface_form_conductivity.py @@ -224,7 +224,7 @@ class FullAlgebraic(BaseModel): The parameters to use for this submodel options : dict, optional A dictionary of options to be passed to the model. - """ # noqa: E501 + """ def __init__(self, param, domain, options=None): super().__init__(param, domain, options) @@ -258,7 +258,7 @@ class FullDifferential(BaseModel): The parameters to use for this submodel options : dict, optional A dictionary of options to be passed to the model. - """ # noqa: E501 + """ def __init__(self, param, domain, options=None): super().__init__(param, domain, options) diff --git a/pybamm/parameters_cli.py b/pybamm/parameters_cli.py index 90950b23ee..e3d4a273b8 100644 --- a/pybamm/parameters_cli.py +++ b/pybamm/parameters_cli.py @@ -2,7 +2,7 @@ def raise_error(): raise NotImplementedError( "parameters cli has been deprecated. " "Parameters should now be defined via python files (see " - "https://github.com/pybamm-team/PyBaMM/tree/develop/pybamm/input/parameters/lithium_ion/Ai2020.py" # noqa: E501 + "https://github.com/pybamm-team/PyBaMM/tree/develop/pybamm/input/parameters/lithium_ion/Ai2020.py" " for example)" ) diff --git a/pybamm/plotting/quick_plot.py b/pybamm/plotting/quick_plot.py index ff657ee375..0e56c17c75 100644 --- a/pybamm/plotting/quick_plot.py +++ b/pybamm/plotting/quick_plot.py @@ -534,7 +534,7 @@ def plot(self, t, dynamic=False): # 1D plot: plot as a function of x at time t # Read dictionary of spatial variables spatial_vars = self.spatial_variable_dict[key] - spatial_var_name = list(spatial_vars.keys())[0] + spatial_var_name = next(iter(spatial_vars.keys())) ax.set_xlabel( "{} [{}]".format(spatial_var_name, self.spatial_unit), ) @@ -568,12 +568,12 @@ def plot(self, t, dynamic=False): # different order based on whether the domains are x-r, x-z or y-z, etc if self.x_first_and_y_second[key] is False: x_name = list(spatial_vars.keys())[1][0] - y_name = list(spatial_vars.keys())[0][0] + y_name = next(iter(spatial_vars.keys()))[0] x = self.second_spatial_variable[key] y = self.first_spatial_variable[key] var = variable(t_in_seconds, spatial_vars, warn=False) else: - x_name = list(spatial_vars.keys())[0][0] + x_name = next(iter(spatial_vars.keys()))[0] y_name = list(spatial_vars.keys())[1][0] x = self.first_spatial_variable[key] y = self.second_spatial_variable[key] diff --git a/pybamm/simulation.py b/pybamm/simulation.py index f9aebb1c54..bdd97f6894 100644 --- a/pybamm/simulation.py +++ b/pybamm/simulation.py @@ -375,7 +375,7 @@ def set_initial_soc(self, initial_soc): param = self._model.param if options["open-circuit potential"] == "MSMR": self._parameter_values = ( - self._unprocessed_parameter_values.set_initial_ocps( # noqa: E501 + self._unprocessed_parameter_values.set_initial_ocps( initial_soc, param=param, inplace=False, options=options ) ) diff --git a/pybamm/solvers/jax_bdf_solver.py b/pybamm/solvers/jax_bdf_solver.py index 2f334ed8ec..8f5b8ed817 100644 --- a/pybamm/solvers/jax_bdf_solver.py +++ b/pybamm/solvers/jax_bdf_solver.py @@ -676,7 +676,7 @@ def while_body(while_state): # ) (state, step_accepted) = tree_map( - partial(jnp.where, converged (error_norm > 1)), # noqa: E712 + partial(jnp.where, converged * (error_norm > 1)), (_update_step_size_and_lu(state, factor), False), (state, converged), ) @@ -883,9 +883,7 @@ def arg_dicts_to_values(args): """ return sum((tuple(b.values()) for b in args if isinstance(b, dict)), ()) - aug_mass = (mass, mass, onp.array(1.0)) + arg_dicts_to_values( - tree_map(arg_to_identity, args) - ) + aug_mass = (mass, mass, onp.array(1.0), arg_dicts_to_values(tree_map(arg_to_identity, args))) def scan_fun(carry, i): y_bar, t0_bar, args_bar = carry @@ -961,7 +959,7 @@ def ravel_first_arg(f, unravel): @lu.transformation def ravel_first_arg_(unravel, y_flat, args): y = unravel(y_flat) - ans = yield (y,) + args, {} + ans = yield (y, args), {} ans_flat, _ = ravel_pytree(ans) yield ans_flat @@ -1007,7 +1005,7 @@ def jax_bdf_integrate(func, y0, t_eval, args, rtol=1e-6, atol=1e-6, mass=None): """ if not pybamm.have_jax(): raise ModuleNotFoundError( - "Jax or jaxlib is not installed, please see https://docs.pybamm.org/en/latest/source/user_guide/installation/GNU-linux.html#optional-jaxsolver" # noqa: E501 + "Jax or jaxlib is not installed, please see https://docs.pybamm.org/en/latest/source/user_guide/installation/GNU-linux.html#optional-jaxsolver" ) def _check_arg(arg): diff --git a/pybamm/solvers/jax_solver.py b/pybamm/solvers/jax_solver.py index 4c9759008a..313fddc208 100644 --- a/pybamm/solvers/jax_solver.py +++ b/pybamm/solvers/jax_solver.py @@ -61,7 +61,7 @@ def __init__( ): if not pybamm.have_jax(): raise ModuleNotFoundError( - "Jax or jaxlib is not installed, please see https://docs.pybamm.org/en/latest/source/user_guide/installation/GNU-linux.html#optional-jaxsolver" # noqa: E501 + "Jax or jaxlib is not installed, please see https://docs.pybamm.org/en/latest/source/user_guide/installation/GNU-linux.html#optional-jaxsolver" ) # note: bdf solver itself calculates consistent initial conditions so can set diff --git a/pybamm/spatial_methods/finite_volume.py b/pybamm/spatial_methods/finite_volume.py index ecbae69796..5c32e5a2c0 100644 --- a/pybamm/spatial_methods/finite_volume.py +++ b/pybamm/spatial_methods/finite_volume.py @@ -620,10 +620,10 @@ def add_ghost_nodes(self, symbol, discretised_symbol, bcs): # Dirichlet boundary conditions n_bcs = 0 if lbc_type == "Dirichlet": - domain = [domain[0] + "_left ghost cell"] + domain + domain = [domain[0] + "_left ghost cell", domain] n_bcs += 1 if rbc_type == "Dirichlet": - domain = domain + [domain[-1] + "_right ghost cell"] + domain = [domain, domain[-1] + "_right ghost cell"] n_bcs += 1 # Calculate values for ghost nodes for any Dirichlet boundary conditions diff --git a/pyproject.toml b/pyproject.toml index 4569c7c6c3..eae0575117 100644 --- a/pyproject.toml +++ b/pyproject.toml @@ -102,6 +102,8 @@ dev = [ "ruff", # For running testing sessions "nox", + # For coverage + "coverage[toml]", # For testing Jupyter notebooks "pytest>=6", "pytest-xdist", @@ -172,3 +174,76 @@ pybamm = [ [tool.setuptools.packages.find] include = ["pybamm", "pybamm."] + +[tool.ruff] +extend-include = [".ipynb"] +extend-exclude = ["__init__.py"] + +[tool.ruff.lint] +extend-select = [ + # "B", # flake8-bugbear + # "I", # isort + # "ARG", # flake8-unused-arguments + # "C4", # flake8-comprehensions + # "ICN", # flake8-import-conventions + # "ISC", # flake8-implicit-str-concat + # "PGH", # pygrep-hooks + # "PIE", # flake8-pie + # "PL", # pylint + # "PT", # flake8-pytest-style + # "PTH", # flake8-use-pathlib + # "RET", # flake8-return + "RUF", # Ruff-specific + # "SIM", # flake8-simplify + # "T20", # flake8-print + # "UP", # pyupgrade + "YTT", # flake8-2020 +] +ignore = [ + "E741", # Ambiguous variable name + "RUF012", # Mutable class attributes should be annotated with `typing.ClassVar` + "SIM108", # Use ternary operator + "ARG001", # Unused function argument: + "ARG002", # Unused method arguments + "PLR2004", # Magic value used in comparison + "PLR0915", # Too many statements + "PLR0913", # Too many arguments + "PLR0912", # Too many branches + "RET504", # Unnecessary assignment + "RET505", # Unnecessary `else` + "RET506", # Unnecessary `elif` + "B018", # Found useless expression + "RUF002", # Docstring contains ambiguous +] + +[tool.ruff.lint.per-file-ignores] +"tests/" = ["T20"] +"docs/" = ["T20"] +"examples/" = ["T20"] +".ipynb" = ["E402", "E703"] + +# NOTE: currently used only for notebook tests with the nbmake plugin. +[tool.pytest.ini_options] +# Use pytest-xdist to run tests in parallel by default, exit with +# error if not installed +required_plugins = [ + "pytest-xdist", +] +addopts = [ + "-nauto", + "-v", +] +testpaths = [ + "docs/source/examples/", +] +console_output_style = "progress" + +# Logging configuration +log_cli = "true" +log_cli_level = "INFO" +log_format = "%(asctime)s - %(name)s - %(levelname)s - %(message)s" +log_date_format = "%Y-%m-%d %H:%M:%S" + +[tool.coverage.run] +source = ["pybamm"] +concurrency = ["multiprocessing"] diff --git a/ruff.toml b/ruff.toml deleted file mode 100644 index 7304d64570..0000000000 --- a/ruff.toml +++ /dev/null @@ -1,8 +0,0 @@ -extend-include = [".ipynb"] -extend-exclude = ["__init__.py"] - -[lint] -ignore = ["E741"] - -[lint.per-file-ignores] -"*.ipynb" = ["E402", "E703"] diff --git a/scripts/fix_casadi_rpath_mac.py b/scripts/fix_casadi_rpath_mac.py index 23c8a32d59..3f7f71e834 100644 --- a/scripts/fix_casadi_rpath_mac.py +++ b/scripts/fix_casadi_rpath_mac.py @@ -30,15 +30,15 @@ os.path.join(casadi_dir, libcasadi_37_name), ] -subprocess.run(["otool"] + ["-L", os.path.join(casadi_dir, libcasadi_name)]) +subprocess.run(["otool", "-L", os.path.join(casadi_dir, libcasadi_name)]) -print(" ".join(["install_name_tool"] + install_name_tool_args_for_libcasadi_name)) -subprocess.run(["install_name_tool"] + install_name_tool_args_for_libcasadi_name) +print(" ".join(["install_name_tool", install_name_tool_args_for_libcasadi_name])) +subprocess.run(["install_name_tool", install_name_tool_args_for_libcasadi_name]) -print(" ".join(["install_name_tool"] + install_name_tool_args_for_libcasadi_37_name)) -subprocess.run(["install_name_tool"] + install_name_tool_args_for_libcasadi_37_name) +print(" ".join(["install_name_tool", install_name_tool_args_for_libcasadi_37_name])) +subprocess.run(["install_name_tool", install_name_tool_args_for_libcasadi_37_name]) -subprocess.run(["otool"] + ["-L", os.path.join(casadi_dir, libcasadi_name)]) +subprocess.run(["otool", "-L", os.path.join(casadi_dir, libcasadi_name)]) install_name_tool_args = [ "-change", @@ -46,12 +46,12 @@ os.path.join(casadi_dir, libcppabi_name), os.path.join(casadi_dir, libcpp_name), ] -subprocess.run(["otool"] + ["-L", os.path.join(casadi_dir, libcpp_name)]) +subprocess.run(["otool", "-L", os.path.join(casadi_dir, libcpp_name)]) -print(" ".join(["install_name_tool"] + install_name_tool_args)) -subprocess.run(["install_name_tool"] + install_name_tool_args) +print(" ".join(["install_name_tool", install_name_tool_args])) +subprocess.run(["install_name_tool", install_name_tool_args]) -subprocess.run(["otool"] + ["-L", os.path.join(casadi_dir, libcpp_name)]) +subprocess.run(["otool", "-L", os.path.join(casadi_dir, libcpp_name)]) # Copy libcasadi.3.7.dylib and libc++.1.0.dylib to LD_LIBRARY_PATH # This is needed for the casadi python bindings to work while repairing the wheel diff --git a/scripts/install_KLU_Sundials.py b/scripts/install_KLU_Sundials.py index 96e60aeb0e..8f41f5969a 100755 --- a/scripts/install_KLU_Sundials.py +++ b/scripts/install_KLU_Sundials.py @@ -156,7 +156,7 @@ def download_extract_library(url, download_dir): sundials_src = "../sundials-{}".format(sundials_version) print("-" 10, "Running CMake prepare", "-" * 40) -subprocess.run(["cmake", sundials_src] + cmake_args, cwd=build_dir, check=True) +subprocess.run(["cmake", sundials_src, cmake_args], cwd=build_dir, check=True) print("-" 10, "Building the sundials", "-" * 40) make_cmd = ["make", "install"] diff --git a/scripts/update_version.py b/scripts/update_version.py index 8a2d832e59..ccdec661e2 100644 --- a/scripts/update_version.py +++ b/scripts/update_version.py @@ -78,7 +78,7 @@ def update_version(): file.write(replace_version) # Get latest commit id from pybamm-team/sundials-vcpkg-registry - cmd = "git ls-remote https://github.com/pybamm-team/sundials-vcpkg-registry \| grep refs/heads/main \| cut -f 1 \| tr -d '\n'" # noqa: E501 + cmd = "git ls-remote https://github.com/pybamm-team/sundials-vcpkg-registry \| grep refs/heads/main \| cut -f 1 \| tr -d '\n'" latest_commit_id = os.popen(cmd).read() # vcpkg-configuration.json @@ -93,7 +93,7 @@ def update_version(): file.write(replace_commit_id) changelog_line1 = "# [Unreleased](https://github.com/pybamm-team/PyBaMM/)\n" - changelog_line2 = f"# [v{release_version}](https://github.com/pybamm-team/PyBaMM/tree/v{release_version}) - {last_day_of_month}\n\n" # noqa: E501 + changelog_line2 = f"# [v{release_version}](https://github.com/pybamm-team/PyBaMM/tree/v{release_version}) - {last_day_of_month}\n\n" # CHANGELOG.md with open(os.path.join(pybamm.root_dir(), "CHANGELOG.md"), "r+") as file: diff --git a/setup.py b/setup.py index 9cfc4df4ff..ef82e65e70 100644 --- a/setup.py +++ b/setup.py @@ -42,10 +42,7 @@ def set_vcpkg_environment_variables(): # ---------- CMakeBuild class (custom build_ext for IDAKLU target) --------------------- class CMakeBuild(build_ext): - user_options = build_ext.user_options + [ - ("suitesparse-root=", None, "suitesparse source location"), - ("sundials-root=", None, "sundials source location"), - ] + user_options = [build_ext.user_options, ("suitesparse-root=", None, "suitesparse source location"), ("sundials-root=", None, "sundials source location")] def initialize_options(self): build_ext.initialize_options(self) @@ -136,7 +133,7 @@ def run(self): cmake_list_dir = os.path.abspath(os.path.dirname(__file__)) print("-" 10, "Running CMake for IDAKLU solver", "-" * 40) subprocess.run( - ["cmake", cmake_list_dir] + cmake_args, cwd=build_dir, env=build_env + ["cmake", cmake_list_dir, cmake_args], cwd=build_dir, env=build_env , check=True) if os.path.isfile(os.path.join(build_dir, "CMakeError.log")): @@ -144,7 +141,7 @@ def run(self): "cmake configuration steps encountered errors, and the IDAKLU module" " could not be built. Make sure dependencies are correctly " "installed. See " - "https://docs.pybamm.org/en/latest/source/user_guide/installation/install-from-source.html" # noqa: E501 + "https://docs.pybamm.org/en/latest/source/user_guide/installation/install-from-source.html" ) raise RuntimeError(msg) else: @@ -201,10 +198,7 @@ def move_output(self, ext): class CustomInstall(install): """A custom install command to add 2 build options""" - user_options = install.user_options + [ - ("suitesparse-root=", None, "suitesparse source location"), - ("sundials-root=", None, "sundials source location"), - ] + user_options = [install.user_options, ("suitesparse-root=", None, "suitesparse source location"), ("sundials-root=", None, "sundials source location")] def initialize_options(self): install.initialize_options(self) @@ -228,10 +222,7 @@ def run(self): class bdist_wheel(orig.bdist_wheel): """A custom install command to add 2 build options""" - user_options = orig.bdist_wheel.user_options + [ - ("suitesparse-root=", None, "suitesparse source location"), - ("sundials-root=", None, "sundials source location"), - ] + user_options = [*orig.bdist_wheel.user_options, ("suitesparse-root=", None, "suitesparse source location"), ("sundials-root=", None, "sundials source location")] def initialize_options(self): orig.bdist_wheel.initialize_options(self)	Move all the extra configurations to `pyproject.toml` Once #3301 is merged, we can move the configurations of pytest, ruff, and others to `pyproject.toml`	I was going to open this issue, you beat me to it! 😄 Some more tools whose configuration can be moved: `cibuildwheel`, `coverage.py`, and subsequently `towncrier` – after #3469 is merged)	2023-11-25T15:20:47	[]	[]
eclipse-volttron/volttron-openadr-ven	eclipse-volttron__volttron-openadr-ven-17	da1dd24743d0b35c59fa7f8d90cf909654168979	diff --git a/.gitignore b/.gitignore index 94cae34..6063ad4 100644 --- a/.gitignore +++ b/.gitignore @@ -132,3 +132,4 @@ dmypy.json secret config_local.json +.DS_Store diff --git a/.gitmodules b/.gitmodules deleted file mode 100644 index 8dd3851..0000000 --- a/.gitmodules +++ /dev/null @@ -1,3 +0,0 @@ -[submodule "openleadr"] - path = openleadr - url = [email protected]:bonicim/openleadr-python.git diff --git a/README.md b/README.md index 933493b..0a6db97 100644 --- a/README.md +++ b/README.md @@ -14,15 +14,14 @@ correct configuration with optional parameters added. { "ven_name": "PNNLVEN", "vtn_url": "https://eiss2demo.ipkeys.com/oadr2/OpenADR2/Simple/2.0b", + "openadr_client_type": "IPKeysClient" # the list of valid client types are the class names of the OpenADRClient subclasses in ~volttron_openadr_ven/volttron_openadr_client.py # below are optional parameters - # if you want/need to sign outgoing messages using a public-private key pair, you need to add a directory named "secret" in the top level - # of this repo; then provide the relative path to the cert and key - # below is an example - # test certificates for IP Keys VTN's can be obtained at https://testcerts.kyrio.com/#/ - "cert": "secret/TEST_RSA_VEN_210923215148_certs/TEST_RSA_VEN_210923215148_cert.pem", - "key": "secret/TEST_RSA_VEN_210923215148_certs/TEST_RSA_VEN_210923215148_privkey.pem", + # if you want/need to sign outgoing messages using a public-private key pair, provide the relative path to the cert_path and key_path + # in this example, the keypair is stored in the directory named '~/.ssh/secret' + "cert_path": "~/.ssh/secret/TEST_RSA_VEN_210923215148_cert.pem", + "key_path": "~/.ssh/secret/TEST_RSA_VEN_210923215148_privkey.pem", # other optional parameters "debug": true, @@ -30,32 +29,58 @@ correct configuration with optional parameters added. } ``` -## Volttron integration +Save this configuration in a JSON file in your preferred location; we recommend saving it in the `volttron_openadr_ven` directory. +An example of such a configuration is saved in the root of this repository; the file is named `config_example1.json` -To ensure that this agent is created within the Volttron framework, ensure that you set the following environmental variable -to the absolute path of the configuration JSON file for this agent. For example, on Linux, use the following example to set -the environmental variable AGENT_CONFIG to a path to your JSON config file: -`export AGENT_CONFIG /home/usr/path-to-my-config-file` +# Quickstart -# Development +## Installing the agent on Monolith Volttron -## Prerequisites +1. Start the Volttron platform on your machine + +```shell +volttron -vv -l volttron.log & +``` + +2. Tail the logs so you can observe OpenADRClient logs + +```shell +tail -f volttron.log +``` + +3. Install the agent on Volttron + +```shell +vctl -v install <path to root directory of volttron-openadr-ven> \ +--tag openadr \ +--agent-config <path to agent config> +``` + +4. Verify status of agent +```shell +vctl status +``` + +5. Start the agent +```shell +vctl start --tag openadr +``` + +## Starting the agent on Modular Volttron + + +### Prerequisites Python >=3.7 * Pip >=20.1 * Poetry >=1.16 - -## Environment setup +### 1. Environment setup Note: This repo uses [Poetry](https://python-poetry.org/), a dependency management and packaging tool for Python. If you don't have Poetry installed on your machine, follow [these steps](https://python-poetry.org/docs/#installation) to install it on your machine. To check if Poetry is installed, run `poetry --version`. If you receive the error 'command not found: poetry', add the following line to your '~/.bashrc' script: ```export PATH=$PATH:$HOME/.poetry/bin```. -0. Install the required submodules - -```git submodule update --init --recursive``` - 1. Create virtual environment By default, poetry creates a virtual environment in {cache-dir}/virtualenvs @@ -71,11 +96,6 @@ Then to create the virtual environment itself, run the following command: poetry shell ``` -NOTE: Poetry needs the command 'python' in order to create a new shell. On Linux, if you only have python3 installed, run the following to link 'python' to 'python3': -```shell -sudo apt install python-is-python3 -``` - 2. Install requirements ```shell @@ -83,80 +103,141 @@ poetry install ``` -## Code Quality and Standardization +The environment is now setup to run the OpenADRVen agent. Next, we need to set up a local Volttron platform and configure the +remote VTN server that this OpenADRVen agent will connect to. -## Git Hooks and Pre-commit +### 2. Local Volttron platform setup -This repo provides pre-commit hooks to enforce code quality and standardization before pushing any commit to the repo. To use this tool locally, -run the following command. +We need to start a local Volttron platform on the same machine that is hosting this OpenADRVen agent. Follow the [installation steps](https://volttron.readthedocs.io/en/develop/introduction/platform-install.html) +for installing a single instance of the Volttron platform. After you have installed and started the Volttron platform, we need +to get the platform's server key and create a public and secret keys. We will use these keys to configure the credentials of the +OpenADRVen agent so that the local Volttron platform can successfully authenticate the OpenADRVen agent. + +Steps + +1. With the local Volttron platform running, open a new shell and run `vctl auth serverkey`. Copy and paste output in the agent's config file. See example below: ```shell -pre-commit install +$ vctl auth serverkey +FDIH6OBg0m3L2T6Jv_zUuQlSxYdgvTD3QOEye-vM-iI + +# Paste the output in config file for agent + +{ + { + "ven_name": "PNNLVEN", + ... + "server_key": "FDIH6OBg0m3L2T6Jv_zUuQlSxYdgvTD3QOEye-vM-iI" + } +} ``` -To check that pre-commit scripts are installed and properly run, run pre-commit on all the files: +2. On the same shell, run `vctl auth keypair`. Copy the values of `public` and `secret` and paste them in the agent's config file. See example below: +```shell +$ vctl auth keypair +{ + "public": "dyMf8g58ptiQTqI7EHbNr7PPFsAr2y-OIy1J-he_DCU", + "secret": "iJa4mlZBomb3bDP8oRAN4IK8uXEqpl1MDg_te3aJzMo" +} + +# Paste the output in config file for agent + +{ + { + "ven_name": "PNNLVEN", + ... + "server_key": "FDIH6OBg0m3L2T6Jv_zUuQlSxYdgvTD3QOEye-vM-iI", + "agent_public": "csJBQqQDZ-pP_8E9FIgM9hvkAak6HriLkIQhP46ZFl4", + "agent_secret": "4rv_l3oEzgmRxbPkGma2-tMhfPu47yyhi48ygF5hVQY" + } +} ``` -pre-commit run --all + +3. Add the public key, which was generated in the previous step, to the Volttron platform. Use `vctl auth add --credentials <public key>'. +An example of the command being run is shown below: + +```shell +$ vctl auth add --credentials 'csJBQqQDZ-pP_8E9FIgM9hvkAak6HriLkIQhP46ZFl4' ``` -With pre-commit installed and configured, every time you run `git commit`, git will check your commit against a set of code linters. For details, see .pre-commit-config.yaml +### 3. VTN Server setup -This project installs the following tools to aid in code standardization and best practices: +Depending on the type of VTN that you are using, you need to configure your VTN to send events so that the OpenADRVen agent +can receive such events from your VTN. -## Black -This repo uses [Black](https://pypi.org/project/black/), an opinionated Python code formatter. To use Black to check and automagically reformat your code, -run the following command on your chosen Python file: +#### IPKeys VTN configuration -```shell -black <path to file> -``` +The community is currently testing this OpenADRVen agent against a IPKeys VTN. To configure the agent with the right +certificates, follow the instructions below: + +Get VEN certificates at https://testcerts.kyrio.com/#/. Certificates can be stored anywhere on your machine, however, +it is recommended to place your keypair files in your ~/.ssh/ folder and make those files read-only. + +* IPKeys VTN web browser console: https://eiss2demo.ipkeys.com/ui/home/#/login + +To create an event, click "Events" tab. Then click the "+" icon to create an event. Use the template "PNNLTestEvent" or +"PNNLnontestevent" to create an event. + +![Alt text](screenshots/test_event_screenshot_ipkeys.png?raw=true "Screenshot of Events page of IPKeys GUI") + +### 4. Starting the OpenADRVen agent + +Now that we have both the environment for OpenADRVen agent setup and the Volttron platform and VTN's configured properly, +we can finally run our OpenADRVen agent. Note: we are NOT installing this agent on the local Volttron platform. We will +simply run this agent on the host machine. Navigate to the root level of this repo. Then run the ~/volttron_openadr_ven/agent.py +module with the path to the agent configuration file that you created as noted in the "Agent Configuration" section at the beginning of this README. -## MyPy -This repo uses [MyPy](https://mypy.readthedocs.io/en/stable/), a static type checker for Python. To use MyPy on a Python file, run the following command: ```shell -mypy <path to file> +poetry run python -m volttron_openadr_ven.agent <path to the config file> ``` -# CI/CD -TODO: +# Development, Code Quality, and Standardization -# Building Wheel and Publishing to PyPi +## OpenLeadr Notes (IMPORTANT) -TODO: +The OpenADRVen agent uses a third-party library, OpenLeader[https://openleadr.org/docs/index.html], as the actual client. +However, the OpenADRVen agent had to modify some parts of that client in order to connect to the IPKeys VTN for testing. Specifically, +OpenADRVen agent extends the OpenLeadr client class and overrides some class methods. -Build wheel -```python setup.py -q sdist bdist_wheel``` +Those parts that were modified are located in ~/volttron_openadr_ven/volttron_openleadr.py. Future releases of OpenLeadr could potentially and adversely +affect OpenADRVen agent if those releases directly or indirectly affect the modified code blocks. Thus, maintenance of this agent should closely monitor future changes to OpenLeadr. -Check wheel -```check-wheel-contents dist/.whl``` -```tar tzf <>``` -Test uploading -```twine upload --repository-url https://test.pypi.org/legacy/ dist/``` +## Git Hooks and Pre-commit -Upload to PyPi -```python -m twin upload dist/``` +This repo provides pre-commit hooks to enforce code quality and standardization before pushing any commit to the repo. To use this tool locally, +run the following command. +```shell +pre-commit install +``` -# Testing +To check that pre-commit scripts are installed and properly run, run pre-commit on all the files: -## Ven Certificates +``` +pre-commit run --all +``` -https://testcerts.kyrio.com/#/ +With pre-commit installed and configured, every time you run `git commit`, git will check your commit against a set of code linters. For details, see .pre-commit-config.yaml -## Test VTN's +This project installs the following tools to aid in code standardization and best practices: - IPKeys: https://eiss2demo.ipkeys.com/ui/home/#/login +## Black -# Other +This repo uses [Black](https://pypi.org/project/black/), an opinionated Python code formatter. To use Black to check and automagically reformat your code, +run the following command on your chosen Python file: -## Installing dependencies behind a proxy +```shell +black <path to file> +``` -If running this behind a PNNL VPN, run the following so Poetry can install all the dependencies: -See: https://leifengblog.net/blog/how-to-use-pip-behind-a-proxy/ +## MyPy -```export https_proxy=http://proxy01.pnl.gov:3128``` +This repo uses [MyPy](https://mypy.readthedocs.io/en/stable/), a static type checker for Python. To use MyPy on a Python file, run the following command: +```shell +mypy <path to file> +``` diff --git a/config_example1.json b/config_example1.json new file mode 100644 index 0000000..bfed53d --- /dev/null +++ b/config_example1.json @@ -0,0 +1,13 @@ +{ + "ven_name": "PNNLVEN", + "vtn_url": "https://eiss2demo.ipkeys.com/oadr2/OpenADR2/Simple/2.0b", + "cert_path": "~/.ssh/secret/path_to_cert.pem", + "key_path": "~/.ssh/secret/path_to_privkey.pem", + "debug": true, + "disable_signature": true, + "vip_address": "tcp://127.0.0.1:22916", + "server_key": "FDIH6OBg0m3L2T6Jv_zUuQlSxYdgvTD3QOEye-vM-iI", + "agent_public": "csJBQqQDZ-pP_8E9FIgM9hvkAak6HriLkIQhP46ZFl4", + "agent_secret": "4rv_l3oEzgmRxbPkGma2-tMhfPu47yyhi48ygF5hVQY", + "openadr_client_type": "IPKeysClient" +} diff --git a/openleadr b/openleadr deleted file mode 160000 index e306798..0000000 --- a/openleadr +++ /dev/null @@ -1,1 +0,0 @@ -Subproject commit e306798f9f4af36f0349c10fec4607aaa18d34c8 diff --git a/poetry.lock b/poetry.lock index b2f2e1c..0568674 100644 --- a/poetry.lock +++ b/poetry.lock @@ -1,25 +1,37 @@ [[package]] name = "aiohttp" -version = "3.7.4.post0" +version = "3.8.1" description = "Async http client/server framework (asyncio)" category = "main" optional = false python-versions = ">=3.6" [package.dependencies] -async-timeout = ">=3.0,<4.0" +aiosignal = ">=1.1.2" +async-timeout = ">=4.0.0a3,<5.0" attrs = ">=17.3.0" -chardet = ">=2.0,<5.0" +charset-normalizer = ">=2.0,<3.0" +frozenlist = ">=1.1.1" multidict = ">=4.5,<7.0" -typing-extensions = ">=3.6.5" yarl = ">=1.0,<2.0" [package.extras] -speedups = ["aiodns", "brotlipy", "cchardet"] +speedups = ["aiodns", "brotli", "cchardet"] + +[[package]] +name = "aiosignal" +version = "1.2.0" +description = "aiosignal: a list of registered asynchronous callbacks" +category = "main" +optional = false +python-versions = ">=3.6" + +[package.dependencies] +frozenlist = ">=1.1.0" [[package]] name = "apscheduler" -version = "3.8.0" +version = "3.8.1" description = "In-process task scheduler with Cron-like capabilities" category = "main" optional = false @@ -28,7 +40,7 @@ python-versions = ">=2.7, !=3.0., !=3.1., !=3.2., !=3.3., !=3.4., <4" [package.dependencies] pytz = "" six = ">=1.4.0" -tzlocal = ">=2.0,<3.0" +tzlocal = ">=2.0,<3.0.0 \|\| >=4.0.0" [package.extras] asyncio = ["trollius"] @@ -53,11 +65,14 @@ python-versions = "" [[package]] name = "async-timeout" -version = "3.0.1" +version = "4.0.1" description = "Timeout context manager for asyncio programs" category = "main" optional = false -python-versions = ">=3.5.3" +python-versions = ">=3.6" + +[package.dependencies] +typing-extensions = ">=3.6.5" [[package]] name = "atomicwrites" @@ -83,7 +98,7 @@ tests_no_zope = ["coverage[toml] (>=5.0.2)", "hypothesis", "pympler", "pytest (> [[package]] name = "backports.entry-points-selectable" -version = "1.1.0" +version = "1.1.1" description = "Compatibility shim providing selectable entry points for older implementations" category = "main" optional = false @@ -91,11 +106,11 @@ python-versions = ">=2.7" [package.extras] docs = ["sphinx", "jaraco.packaging (>=8.2)", "rst.linker (>=1.9)"] -testing = ["pytest (>=4.6)", "pytest-flake8", "pytest-cov", "pytest-black (>=0.3.7)", "pytest-mypy", "pytest-checkdocs (>=2.4)", "pytest-enabler (>=1.0.1)"] +testing = ["pytest", "pytest-flake8", "pytest-cov", "pytest-black (>=0.3.7)", "pytest-mypy", "pytest-checkdocs (>=2.4)", "pytest-enabler (>=1.0.1)"] [[package]] name = "black" -version = "21.9b0" +version = "21.11b1" description = "The uncompromising code formatter." category = "main" optional = false @@ -106,7 +121,7 @@ click = ">=7.1.2" mypy-extensions = ">=0.4.3" pathspec = ">=0.9.0,<1" platformdirs = ">=2" -regex = ">=2020.1.8" +regex = ">=2021.4.4" tomli = ">=0.2.6,<2.0.0" typing-extensions = [ {version = ">=3.10.0.0", markers = "python_version < \"3.10\""}, @@ -115,9 +130,9 @@ typing-extensions = [ [package.extras] colorama = ["colorama (>=0.4.3)"] -d = ["aiohttp (>=3.6.0)", "aiohttp-cors (>=0.4.0)"] +d = ["aiohttp (>=3.7.4)"] jupyter = ["ipython (>=7.8.0)", "tokenize-rt (>=3.2.0)"] -python2 = ["typed-ast (>=1.4.2)"] +python2 = ["typed-ast (>=1.4.3)"] uvloop = ["uvloop (>=0.15.2)"] [[package]] @@ -147,17 +162,9 @@ category = "main" optional = false python-versions = ">=3.6.1" -[[package]] -name = "chardet" -version = "4.0.0" -description = "Universal encoding detector for Python 2 and 3" -category = "main" -optional = false -python-versions = ">=2.7, !=3.0., !=3.1., !=3.2., !=3.3., !=3.4." - [[package]] name = "charset-normalizer" -version = "2.0.7" +version = "2.0.9" description = "The Real First Universal Charset Detector. Open, modern and actively maintained alternative to Chardet." category = "main" optional = false @@ -238,7 +245,7 @@ future = ">=0.16,<0.19" [[package]] name = "filelock" -version = "3.3.1" +version = "3.4.0" description = "A platform independent file lock." category = "main" optional = false @@ -248,6 +255,14 @@ python-versions = ">=3.6" docs = ["furo (>=2021.8.17b43)", "sphinx (>=4.1)", "sphinx-autodoc-typehints (>=1.12)"] testing = ["covdefaults (>=1.2.0)", "coverage (>=4)", "pytest (>=4)", "pytest-cov", "pytest-timeout (>=1.4.2)"] +[[package]] +name = "frozenlist" +version = "1.2.0" +description = "A list-like structure which implements collections.abc.MutableSequence" +category = "main" +optional = false +python-versions = ">=3.6" + [[package]] name = "future" version = "0.18.2" @@ -299,22 +314,22 @@ requests = "" [[package]] name = "identify" -version = "2.3.0" +version = "2.4.0" description = "File identification library for Python" category = "main" optional = false python-versions = ">=3.6.1" [package.extras] -license = ["editdistance-s"] +license = ["ukkonen"] [[package]] name = "idna" -version = "2.10" +version = "3.3" description = "Internationalized Domain Names in Applications (IDNA)" category = "main" optional = false -python-versions = ">=2.7, !=3.0., !=3.1., !=3.2., !=3.3." +python-versions = ">=3.5" [[package]] name = "iniconfig" @@ -326,7 +341,7 @@ python-versions = "" [[package]] name = "jinja2" -version = "3.0.2" +version = "3.0.3" description = "A very fast and expressive template engine." category = "main" optional = false @@ -340,7 +355,7 @@ i18n = ["Babel (>=2.7)"] [[package]] name = "lxml" -version = "4.6.3" +version = "4.6.4" description = "Powerful and Pythonic XML processing library combining libxml2/libxslt with the ElementTree API." category = "main" optional = false @@ -403,11 +418,10 @@ python-versions = "" [[package]] name = "openleadr" version = "0.5.24" -description = "" +description = "Python3 library for building OpenADR Clients (VENs) and Servers (VTNs)" category = "main" optional = false python-versions = ">=3.7.0" -develop = true [package.dependencies] aiohttp = "" @@ -416,20 +430,16 @@ jinja2 = "" signxml-openadr = "2.9.1" xmltodict = "" -[package.source] -type = "directory" -url = "openleadr" - [[package]] name = "packaging" -version = "21.0" +version = "21.3" description = "Core utilities for Python packages" category = "dev" optional = false python-versions = ">=3.6" [package.dependencies] -pyparsing = ">=2.0.2" +pyparsing = ">=2.0.2,<3.0.5 \|\| >3.0.5" [[package]] name = "pathspec" @@ -439,6 +449,14 @@ category = "main" optional = false python-versions = "!=3.0.,!=3.1.,!=3.2.,!=3.3.,!=3.4.,>=2.7" +[[package]] +name = "pbr" +version = "5.8.0" +description = "Python Build Reasonableness" +category = "main" +optional = false +python-versions = ">=2.6" + [[package]] name = "platformdirs" version = "2.4.0" @@ -465,7 +483,7 @@ testing = ["pytest", "pytest-benchmark"] [[package]] name = "pre-commit" -version = "2.15.0" +version = "2.16.0" description = "A framework for managing and maintaining multi-language pre-commit hooks." category = "main" optional = false @@ -492,15 +510,15 @@ test = ["ipaddress", "mock", "unittest2", "enum34", "pywin32", "wmi"] [[package]] name = "py" -version = "1.10.0" +version = "1.11.0" description = "library with cross-python path, ini-parsing, io, code, log facilities" category = "main" optional = false -python-versions = ">=2.7, !=3.0., !=3.1., !=3.2., !=3.3." +python-versions = ">=2.7, !=3.0., !=3.1., !=3.2., !=3.3., !=3.4." [[package]] name = "pycparser" -version = "2.20" +version = "2.21" description = "C parser in Python" category = "main" optional = false @@ -524,11 +542,14 @@ test = ["flaky", "pretend", "pytest (>=3.0.1)"] [[package]] name = "pyparsing" -version = "2.4.7" +version = "3.0.6" description = "Python parsing module" category = "dev" optional = false -python-versions = ">=2.6, !=3.0., !=3.1., !=3.2." +python-versions = ">=3.6" + +[package.extras] +diagrams = ["jinja2", "railroad-diagrams"] [[package]] name = "pytest" @@ -572,11 +593,11 @@ python-versions = "" [[package]] name = "pyyaml" -version = "5.4.1" +version = "6.0" description = "YAML parser and emitter for Python" category = "main" optional = false -python-versions = ">=2.7, !=3.0., !=3.1., !=3.2., !=3.3., !=3.4., !=3.5." +python-versions = ">=3.6" [[package]] name = "pyzmq" @@ -592,7 +613,7 @@ py = {version = "", markers = "implementation_name == \"pypy\""} [[package]] name = "regex" -version = "2021.10.21" +version = "2021.11.10" description = "Alternative regular expression module, to replace re." category = "main" optional = false @@ -649,7 +670,7 @@ python-versions = ">=2.6, !=3.0., !=3.1., !=3.2." [[package]] name = "tomli" -version = "1.2.1" +version = "1.2.2" description = "A lil' TOML parser" category = "main" optional = false @@ -665,11 +686,11 @@ python-versions = "" [[package]] name = "typing-extensions" -version = "3.10.0.2" -description = "Backported and Experimental Type Hints for Python 3.5+" +version = "4.0.1" +description = "Backported and Experimental Type Hints for Python 3.6+" category = "main" optional = false -python-versions = "" +python-versions = ">=3.6" [[package]] name = "tzlocal" @@ -697,7 +718,7 @@ socks = ["PySocks (>=1.5.6,!=1.5.7,<2.0)"] [[package]] name = "virtualenv" -version = "20.8.1" +version = "20.10.0" description = "Virtual Python Environment builder" category = "main" optional = false @@ -706,49 +727,83 @@ python-versions = "!=3.0.,!=3.1.,!=3.2.,!=3.3.,!=3.4.,>=2.7" [package.dependencies] "backports.entry-points-selectable" = ">=1.0.4" distlib = ">=0.3.1,<1" -filelock = ">=3.0.0,<4" +filelock = ">=3.2,<4" platformdirs = ">=2,<3" six = ">=1.9.0,<2" [package.extras] -docs = ["proselint (>=0.10.2)", "sphinx (>=3)", "sphinx-argparse (>=0.2.5)", "sphinx-rtd-theme (>=0.4.3)", "towncrier (>=19.9.0rc1)"] +docs = ["proselint (>=0.10.2)", "sphinx (>=3)", "sphinx-argparse (>=0.2.5)", "sphinx-rtd-theme (>=0.4.3)", "towncrier (>=21.3)"] testing = ["coverage (>=4)", "coverage-enable-subprocess (>=1)", "flaky (>=3)", "pytest (>=4)", "pytest-env (>=0.6.2)", "pytest-freezegun (>=0.4.1)", "pytest-mock (>=2)", "pytest-randomly (>=1)", "pytest-timeout (>=1)", "packaging (>=20.0)"] [[package]] name = "volttron-client" -version = "0.3.4" -description = "Client for connecting to a volttron server" +version = "0.3.8" +description = "\"VOLTTRON client library to connect to volttron servers\"" category = "main" optional = false -python-versions = ">=3.6.2,<4" +python-versions = ">=3.7, <4" [package.dependencies] -dateutils = ">=0.6,<0.7" -grequests = ">=0.6.0,<0.7.0" -idna = ">=2.5,<3" -requests = ">=2.26.0,<3.0.0" -toml = ">=0.10.2,<0.11.0" -volttron-utils = ">=0.3,<0.4" +asn1crypto = "1.4.0" +certifi = "2021.10.8" +cffi = "1.15.0" +charset-normalizer = {version = "2.0.9", markers = "python_version >= \"3\""} +cryptography = "2.3" +dateutils = "0.6.12" +gevent = "20.6.1" +greenlet = {version = "0.4.16", markers = "platform_python_implementation == \"CPython\""} +grequests = "0.6.0" +idna = {version = "3.3", markers = "python_version >= \"3\""} +pbr = "5.8.0" +psutil = "5.8.0" +pycparser = "2.21" +pyopenssl = "19.0.0" +python-dateutil = "2.8.2" +pytz = "2021.3" +pyyaml = "6.0" +pyzmq = "22.3.0" +requests = "2.26.0" +six = "1.16.0" +tzlocal = "2.1" +urllib3 = "1.26.7" +volttron-utils = "0.3.4" +watchdog = "2.1.6" +watchdog-gevent = "0.1.1" +zmq = "0.0.0" +"zope.event" = "4.5.0" +"zope.interface" = "5.4.0" [[package]] name = "volttron-utils" -version = "0.3.2" +version = "0.3.4" description = "Utility classes and methods for VOLTTRON" category = "main" optional = false -python-versions = ">=3.6.2,<4.0" +python-versions = ">=3.7, <4" [package.dependencies] +asn1crypto = "1.4.0" +cffi = "1.15.0" cryptography = "2.3" -dateutils = ">=0.6.7,<0.7.0" +dateutils = "0.6.12" gevent = "20.6.1" -greenlet = "0.4.16" -psutil = ">=5.7.0,<6.0.0" -pyOpenSSL = "19.0.0" -PyYAML = ">=5.4.1,<6.0.0" +greenlet = {version = "0.4.16", markers = "platform_python_implementation == \"CPython\""} +idna = "3.3" +pbr = "5.8.0" +psutil = "5.8.0" +pycparser = "2.21" +pyopenssl = "19.0.0" +python-dateutil = "2.8.2" +pytz = "2021.3" +pyyaml = "6.0" +pyzmq = "22.3.0" +six = "1.16.0" tzlocal = "2.1" -watchdog-gevent = ">=0.1.1,<0.2.0" -zmq = ">=0.0.0,<0.0.1" +watchdog = "2.1.6" +watchdog-gevent = "0.1.1" +zmq = "0.0.0" +"zope.event" = "4.5.0" +"zope.interface" = "5.4.0" [[package]] name = "watchdog" @@ -783,7 +838,7 @@ python-versions = ">=2.7, !=3.0., !=3.1., !=3.2., !=3.3." [[package]] name = "yarl" -version = "1.7.0" +version = "1.7.2" description = "Yet another URL library" category = "main" optional = false @@ -832,59 +887,98 @@ testing = ["coverage (>=5.0.3)", "zope.event", "zope.testing"] [metadata] lock-version = "1.1" python-versions = "^3.8" -content-hash = "05aefd0e959b8da3ef1df6dbd3f3ec194bfe8b7cbc8d49f393a5919cd78c075e" +content-hash = "a9c58b9265431cbbc61fc25e65851b944f320d1f21f4c74be7f043d40e4c7c6b" [metadata.files] aiohttp = [ - {file = "aiohttp-3.7.4.post0-cp36-cp36m-macosx_10_14_x86_64.whl", hash = "sha256:3cf75f7cdc2397ed4442594b935a11ed5569961333d49b7539ea741be2cc79d5"}, - {file = "aiohttp-3.7.4.post0-cp36-cp36m-manylinux1_i686.whl", hash = "sha256:4b302b45040890cea949ad092479e01ba25911a15e648429c7c5aae9650c67a8"}, - {file = "aiohttp-3.7.4.post0-cp36-cp36m-manylinux2014_aarch64.whl", hash = "sha256:fe60131d21b31fd1a14bd43e6bb88256f69dfc3188b3a89d736d6c71ed43ec95"}, - {file = "aiohttp-3.7.4.post0-cp36-cp36m-manylinux2014_i686.whl", hash = "sha256:393f389841e8f2dfc86f774ad22f00923fdee66d238af89b70ea314c4aefd290"}, - {file = "aiohttp-3.7.4.post0-cp36-cp36m-manylinux2014_ppc64le.whl", hash = "sha256:c6e9dcb4cb338d91a73f178d866d051efe7c62a7166653a91e7d9fb18274058f"}, - {file = "aiohttp-3.7.4.post0-cp36-cp36m-manylinux2014_s390x.whl", hash = "sha256:5df68496d19f849921f05f14f31bd6ef53ad4b00245da3195048c69934521809"}, - {file = "aiohttp-3.7.4.post0-cp36-cp36m-manylinux2014_x86_64.whl", hash = "sha256:0563c1b3826945eecd62186f3f5c7d31abb7391fedc893b7e2b26303b5a9f3fe"}, - {file = "aiohttp-3.7.4.post0-cp36-cp36m-win32.whl", hash = "sha256:3d78619672183be860b96ed96f533046ec97ca067fd46ac1f6a09cd9b7484287"}, - {file = "aiohttp-3.7.4.post0-cp36-cp36m-win_amd64.whl", hash = "sha256:f705e12750171c0ab4ef2a3c76b9a4024a62c4103e3a55dd6f99265b9bc6fcfc"}, - {file = "aiohttp-3.7.4.post0-cp37-cp37m-macosx_10_14_x86_64.whl", hash = "sha256:230a8f7e24298dea47659251abc0fd8b3c4e38a664c59d4b89cca7f6c09c9e87"}, - {file = "aiohttp-3.7.4.post0-cp37-cp37m-manylinux1_i686.whl", hash = "sha256:2e19413bf84934d651344783c9f5e22dee452e251cfd220ebadbed2d9931dbf0"}, - {file = "aiohttp-3.7.4.post0-cp37-cp37m-manylinux2014_aarch64.whl", hash = "sha256:e4b2b334e68b18ac9817d828ba44d8fcb391f6acb398bcc5062b14b2cbeac970"}, - {file = "aiohttp-3.7.4.post0-cp37-cp37m-manylinux2014_i686.whl", hash = "sha256:d012ad7911653a906425d8473a1465caa9f8dea7fcf07b6d870397b774ea7c0f"}, - {file = "aiohttp-3.7.4.post0-cp37-cp37m-manylinux2014_ppc64le.whl", hash = "sha256:40eced07f07a9e60e825554a31f923e8d3997cfc7fb31dbc1328c70826e04cde"}, - {file = "aiohttp-3.7.4.post0-cp37-cp37m-manylinux2014_s390x.whl", hash = "sha256:209b4a8ee987eccc91e2bd3ac36adee0e53a5970b8ac52c273f7f8fd4872c94c"}, - {file = "aiohttp-3.7.4.post0-cp37-cp37m-manylinux2014_x86_64.whl", hash = "sha256:14762875b22d0055f05d12abc7f7d61d5fd4fe4642ce1a249abdf8c700bf1fd8"}, - {file = "aiohttp-3.7.4.post0-cp37-cp37m-win32.whl", hash = "sha256:7615dab56bb07bff74bc865307aeb89a8bfd9941d2ef9d817b9436da3a0ea54f"}, - {file = "aiohttp-3.7.4.post0-cp37-cp37m-win_amd64.whl", hash = "sha256:d9e13b33afd39ddeb377eff2c1c4f00544e191e1d1dee5b6c51ddee8ea6f0cf5"}, - {file = "aiohttp-3.7.4.post0-cp38-cp38-macosx_10_14_x86_64.whl", hash = "sha256:547da6cacac20666422d4882cfcd51298d45f7ccb60a04ec27424d2f36ba3eaf"}, - {file = "aiohttp-3.7.4.post0-cp38-cp38-manylinux1_i686.whl", hash = "sha256:af9aa9ef5ba1fd5b8c948bb11f44891968ab30356d65fd0cc6707d989cd521df"}, - {file = "aiohttp-3.7.4.post0-cp38-cp38-manylinux2014_aarch64.whl", hash = "sha256:64322071e046020e8797117b3658b9c2f80e3267daec409b350b6a7a05041213"}, - {file = "aiohttp-3.7.4.post0-cp38-cp38-manylinux2014_i686.whl", hash = "sha256:bb437315738aa441251214dad17428cafda9cdc9729499f1d6001748e1d432f4"}, - {file = "aiohttp-3.7.4.post0-cp38-cp38-manylinux2014_ppc64le.whl", hash = "sha256:e54962802d4b8b18b6207d4a927032826af39395a3bd9196a5af43fc4e60b009"}, - {file = "aiohttp-3.7.4.post0-cp38-cp38-manylinux2014_s390x.whl", hash = "sha256:a00bb73540af068ca7390e636c01cbc4f644961896fa9363154ff43fd37af2f5"}, - {file = "aiohttp-3.7.4.post0-cp38-cp38-manylinux2014_x86_64.whl", hash = "sha256:79ebfc238612123a713a457d92afb4096e2148be17df6c50fb9bf7a81c2f8013"}, - {file = "aiohttp-3.7.4.post0-cp38-cp38-win32.whl", hash = "sha256:515dfef7f869a0feb2afee66b957cc7bbe9ad0cdee45aec7fdc623f4ecd4fb16"}, - {file = "aiohttp-3.7.4.post0-cp38-cp38-win_amd64.whl", hash = "sha256:114b281e4d68302a324dd33abb04778e8557d88947875cbf4e842c2c01a030c5"}, - {file = "aiohttp-3.7.4.post0-cp39-cp39-macosx_10_14_x86_64.whl", hash = "sha256:7b18b97cf8ee5452fa5f4e3af95d01d84d86d32c5e2bfa260cf041749d66360b"}, - {file = "aiohttp-3.7.4.post0-cp39-cp39-manylinux1_i686.whl", hash = "sha256:15492a6368d985b76a2a5fdd2166cddfea5d24e69eefed4630cbaae5c81d89bd"}, - {file = "aiohttp-3.7.4.post0-cp39-cp39-manylinux2014_aarch64.whl", hash = "sha256:bdb230b4943891321e06fc7def63c7aace16095be7d9cf3b1e01be2f10fba439"}, - {file = "aiohttp-3.7.4.post0-cp39-cp39-manylinux2014_i686.whl", hash = "sha256:cffe3ab27871bc3ea47df5d8f7013945712c46a3cc5a95b6bee15887f1675c22"}, - {file = "aiohttp-3.7.4.post0-cp39-cp39-manylinux2014_ppc64le.whl", hash = "sha256:f881853d2643a29e643609da57b96d5f9c9b93f62429dcc1cbb413c7d07f0e1a"}, - {file = "aiohttp-3.7.4.post0-cp39-cp39-manylinux2014_s390x.whl", hash = "sha256:a5ca29ee66f8343ed336816c553e82d6cade48a3ad702b9ffa6125d187e2dedb"}, - {file = "aiohttp-3.7.4.post0-cp39-cp39-manylinux2014_x86_64.whl", hash = "sha256:17c073de315745a1510393a96e680d20af8e67e324f70b42accbd4cb3315c9fb"}, - {file = "aiohttp-3.7.4.post0-cp39-cp39-win32.whl", hash = "sha256:932bb1ea39a54e9ea27fc9232163059a0b8855256f4052e776357ad9add6f1c9"}, - {file = "aiohttp-3.7.4.post0-cp39-cp39-win_amd64.whl", hash = "sha256:02f46fc0e3c5ac58b80d4d56eb0a7c7d97fcef69ace9326289fb9f1955e65cfe"}, - {file = "aiohttp-3.7.4.post0.tar.gz", hash = "sha256:493d3299ebe5f5a7c66b9819eacdcfbbaaf1a8e84911ddffcdc48888497afecf"}, + {file = "aiohttp-3.8.1-cp310-cp310-macosx_10_9_universal2.whl", hash = "sha256:1ed0b6477896559f17b9eaeb6d38e07f7f9ffe40b9f0f9627ae8b9926ae260a8"}, + {file = "aiohttp-3.8.1-cp310-cp310-macosx_10_9_x86_64.whl", hash = "sha256:7dadf3c307b31e0e61689cbf9e06be7a867c563d5a63ce9dca578f956609abf8"}, + {file = "aiohttp-3.8.1-cp310-cp310-macosx_11_0_arm64.whl", hash = "sha256:a79004bb58748f31ae1cbe9fa891054baaa46fb106c2dc7af9f8e3304dc30316"}, + {file = "aiohttp-3.8.1-cp310-cp310-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:12de6add4038df8f72fac606dff775791a60f113a725c960f2bab01d8b8e6b15"}, + {file = "aiohttp-3.8.1-cp310-cp310-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:6f0d5f33feb5f69ddd57a4a4bd3d56c719a141080b445cbf18f238973c5c9923"}, + {file = "aiohttp-3.8.1-cp310-cp310-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:eaba923151d9deea315be1f3e2b31cc39a6d1d2f682f942905951f4e40200922"}, + {file = "aiohttp-3.8.1-cp310-cp310-manylinux_2_5_i686.manylinux1_i686.manylinux_2_12_i686.manylinux2010_i686.whl", hash = "sha256:099ebd2c37ac74cce10a3527d2b49af80243e2a4fa39e7bce41617fbc35fa3c1"}, + {file = "aiohttp-3.8.1-cp310-cp310-manylinux_2_5_x86_64.manylinux1_x86_64.manylinux_2_12_x86_64.manylinux2010_x86_64.whl", hash = "sha256:2e5d962cf7e1d426aa0e528a7e198658cdc8aa4fe87f781d039ad75dcd52c516"}, + {file = "aiohttp-3.8.1-cp310-cp310-musllinux_1_1_aarch64.whl", hash = "sha256:fa0ffcace9b3aa34d205d8130f7873fcfefcb6a4dd3dd705b0dab69af6712642"}, + {file = "aiohttp-3.8.1-cp310-cp310-musllinux_1_1_i686.whl", hash = "sha256:61bfc23df345d8c9716d03717c2ed5e27374e0fe6f659ea64edcd27b4b044cf7"}, + {file = "aiohttp-3.8.1-cp310-cp310-musllinux_1_1_ppc64le.whl", hash = "sha256:31560d268ff62143e92423ef183680b9829b1b482c011713ae941997921eebc8"}, + {file = "aiohttp-3.8.1-cp310-cp310-musllinux_1_1_s390x.whl", hash = "sha256:01d7bdb774a9acc838e6b8f1d114f45303841b89b95984cbb7d80ea41172a9e3"}, + {file = "aiohttp-3.8.1-cp310-cp310-musllinux_1_1_x86_64.whl", hash = "sha256:97ef77eb6b044134c0b3a96e16abcb05ecce892965a2124c566af0fd60f717e2"}, + {file = "aiohttp-3.8.1-cp310-cp310-win32.whl", hash = "sha256:c2aef4703f1f2ddc6df17519885dbfa3514929149d3ff900b73f45998f2532fa"}, + {file = "aiohttp-3.8.1-cp310-cp310-win_amd64.whl", hash = "sha256:713ac174a629d39b7c6a3aa757b337599798da4c1157114a314e4e391cd28e32"}, + {file = "aiohttp-3.8.1-cp36-cp36m-macosx_10_9_x86_64.whl", hash = "sha256:473d93d4450880fe278696549f2e7aed8cd23708c3c1997981464475f32137db"}, + {file = "aiohttp-3.8.1-cp36-cp36m-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:99b5eeae8e019e7aad8af8bb314fb908dd2e028b3cdaad87ec05095394cce632"}, + {file = "aiohttp-3.8.1-cp36-cp36m-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:3af642b43ce56c24d063325dd2cf20ee012d2b9ba4c3c008755a301aaea720ad"}, + {file = "aiohttp-3.8.1-cp36-cp36m-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:c3630c3ef435c0a7c549ba170a0633a56e92629aeed0e707fec832dee313fb7a"}, + {file = "aiohttp-3.8.1-cp36-cp36m-manylinux_2_5_i686.manylinux1_i686.manylinux_2_12_i686.manylinux2010_i686.whl", hash = "sha256:4a4a4e30bf1edcad13fb0804300557aedd07a92cabc74382fdd0ba6ca2661091"}, + {file = "aiohttp-3.8.1-cp36-cp36m-manylinux_2_5_x86_64.manylinux1_x86_64.manylinux_2_12_x86_64.manylinux2010_x86_64.whl", hash = "sha256:6f8b01295e26c68b3a1b90efb7a89029110d3a4139270b24fda961893216c440"}, + {file = "aiohttp-3.8.1-cp36-cp36m-musllinux_1_1_aarch64.whl", hash = "sha256:a25fa703a527158aaf10dafd956f7d42ac6d30ec80e9a70846253dd13e2f067b"}, + {file = "aiohttp-3.8.1-cp36-cp36m-musllinux_1_1_i686.whl", hash = "sha256:5bfde62d1d2641a1f5173b8c8c2d96ceb4854f54a44c23102e2ccc7e02f003ec"}, + {file = "aiohttp-3.8.1-cp36-cp36m-musllinux_1_1_ppc64le.whl", hash = "sha256:51467000f3647d519272392f484126aa716f747859794ac9924a7aafa86cd411"}, + {file = "aiohttp-3.8.1-cp36-cp36m-musllinux_1_1_s390x.whl", hash = "sha256:03a6d5349c9ee8f79ab3ff3694d6ce1cfc3ced1c9d36200cb8f08ba06bd3b782"}, + {file = "aiohttp-3.8.1-cp36-cp36m-musllinux_1_1_x86_64.whl", hash = "sha256:102e487eeb82afac440581e5d7f8f44560b36cf0bdd11abc51a46c1cd88914d4"}, + {file = "aiohttp-3.8.1-cp36-cp36m-win32.whl", hash = "sha256:4aed991a28ea3ce320dc8ce655875e1e00a11bdd29fe9444dd4f88c30d558602"}, + {file = "aiohttp-3.8.1-cp36-cp36m-win_amd64.whl", hash = "sha256:b0e20cddbd676ab8a64c774fefa0ad787cc506afd844de95da56060348021e96"}, + {file = "aiohttp-3.8.1-cp37-cp37m-macosx_10_9_x86_64.whl", hash = "sha256:37951ad2f4a6df6506750a23f7cbabad24c73c65f23f72e95897bb2cecbae676"}, + {file = "aiohttp-3.8.1-cp37-cp37m-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:5c23b1ad869653bc818e972b7a3a79852d0e494e9ab7e1a701a3decc49c20d51"}, + {file = "aiohttp-3.8.1-cp37-cp37m-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:15b09b06dae900777833fe7fc4b4aa426556ce95847a3e8d7548e2d19e34edb8"}, + {file = "aiohttp-3.8.1-cp37-cp37m-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:477c3ea0ba410b2b56b7efb072c36fa91b1e6fc331761798fa3f28bb224830dd"}, + {file = "aiohttp-3.8.1-cp37-cp37m-manylinux_2_5_i686.manylinux1_i686.manylinux_2_12_i686.manylinux2010_i686.whl", hash = "sha256:2f2f69dca064926e79997f45b2f34e202b320fd3782f17a91941f7eb85502ee2"}, + {file = "aiohttp-3.8.1-cp37-cp37m-manylinux_2_5_x86_64.manylinux1_x86_64.manylinux_2_12_x86_64.manylinux2010_x86_64.whl", hash = "sha256:ef9612483cb35171d51d9173647eed5d0069eaa2ee812793a75373447d487aa4"}, + {file = "aiohttp-3.8.1-cp37-cp37m-musllinux_1_1_aarch64.whl", hash = "sha256:6d69f36d445c45cda7b3b26afef2fc34ef5ac0cdc75584a87ef307ee3c8c6d00"}, + {file = "aiohttp-3.8.1-cp37-cp37m-musllinux_1_1_i686.whl", hash = "sha256:55c3d1072704d27401c92339144d199d9de7b52627f724a949fc7d5fc56d8b93"}, + {file = "aiohttp-3.8.1-cp37-cp37m-musllinux_1_1_ppc64le.whl", hash = "sha256:b9d00268fcb9f66fbcc7cd9fe423741d90c75ee029a1d15c09b22d23253c0a44"}, + {file = "aiohttp-3.8.1-cp37-cp37m-musllinux_1_1_s390x.whl", hash = "sha256:07b05cd3305e8a73112103c834e91cd27ce5b4bd07850c4b4dbd1877d3f45be7"}, + {file = "aiohttp-3.8.1-cp37-cp37m-musllinux_1_1_x86_64.whl", hash = "sha256:c34dc4958b232ef6188c4318cb7b2c2d80521c9a56c52449f8f93ab7bc2a8a1c"}, + {file = "aiohttp-3.8.1-cp37-cp37m-win32.whl", hash = "sha256:d2f9b69293c33aaa53d923032fe227feac867f81682f002ce33ffae978f0a9a9"}, + {file = "aiohttp-3.8.1-cp37-cp37m-win_amd64.whl", hash = "sha256:6ae828d3a003f03ae31915c31fa684b9890ea44c9c989056fea96e3d12a9fa17"}, + {file = "aiohttp-3.8.1-cp38-cp38-macosx_10_9_universal2.whl", hash = "sha256:0c7ebbbde809ff4e970824b2b6cb7e4222be6b95a296e46c03cf050878fc1785"}, + {file = "aiohttp-3.8.1-cp38-cp38-macosx_10_9_x86_64.whl", hash = "sha256:8b7ef7cbd4fec9a1e811a5de813311ed4f7ac7d93e0fda233c9b3e1428f7dd7b"}, + {file = "aiohttp-3.8.1-cp38-cp38-macosx_11_0_arm64.whl", hash = "sha256:c3d6a4d0619e09dcd61021debf7059955c2004fa29f48788a3dfaf9c9901a7cd"}, + {file = "aiohttp-3.8.1-cp38-cp38-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:718626a174e7e467f0558954f94af117b7d4695d48eb980146016afa4b580b2e"}, + {file = "aiohttp-3.8.1-cp38-cp38-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:589c72667a5febd36f1315aa6e5f56dd4aa4862df295cb51c769d16142ddd7cd"}, + {file = "aiohttp-3.8.1-cp38-cp38-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:2ed076098b171573161eb146afcb9129b5ff63308960aeca4b676d9d3c35e700"}, + {file = "aiohttp-3.8.1-cp38-cp38-manylinux_2_5_i686.manylinux1_i686.manylinux_2_12_i686.manylinux2010_i686.whl", hash = "sha256:086f92daf51a032d062ec5f58af5ca6a44d082c35299c96376a41cbb33034675"}, + {file = "aiohttp-3.8.1-cp38-cp38-manylinux_2_5_x86_64.manylinux1_x86_64.manylinux_2_12_x86_64.manylinux2010_x86_64.whl", hash = "sha256:11691cf4dc5b94236ccc609b70fec991234e7ef8d4c02dd0c9668d1e486f5abf"}, + {file = "aiohttp-3.8.1-cp38-cp38-musllinux_1_1_aarch64.whl", hash = "sha256:31d1e1c0dbf19ebccbfd62eff461518dcb1e307b195e93bba60c965a4dcf1ba0"}, + {file = "aiohttp-3.8.1-cp38-cp38-musllinux_1_1_i686.whl", hash = "sha256:11a67c0d562e07067c4e86bffc1553f2cf5b664d6111c894671b2b8712f3aba5"}, + {file = "aiohttp-3.8.1-cp38-cp38-musllinux_1_1_ppc64le.whl", hash = "sha256:bb01ba6b0d3f6c68b89fce7305080145d4877ad3acaed424bae4d4ee75faa950"}, + {file = "aiohttp-3.8.1-cp38-cp38-musllinux_1_1_s390x.whl", hash = "sha256:44db35a9e15d6fe5c40d74952e803b1d96e964f683b5a78c3cc64eb177878155"}, + {file = "aiohttp-3.8.1-cp38-cp38-musllinux_1_1_x86_64.whl", hash = "sha256:844a9b460871ee0a0b0b68a64890dae9c415e513db0f4a7e3cab41a0f2fedf33"}, + {file = "aiohttp-3.8.1-cp38-cp38-win32.whl", hash = "sha256:7d08744e9bae2ca9c382581f7dce1273fe3c9bae94ff572c3626e8da5b193c6a"}, + {file = "aiohttp-3.8.1-cp38-cp38-win_amd64.whl", hash = "sha256:04d48b8ce6ab3cf2097b1855e1505181bdd05586ca275f2505514a6e274e8e75"}, + {file = "aiohttp-3.8.1-cp39-cp39-macosx_10_9_universal2.whl", hash = "sha256:f5315a2eb0239185af1bddb1abf472d877fede3cc8d143c6cddad37678293237"}, + {file = "aiohttp-3.8.1-cp39-cp39-macosx_10_9_x86_64.whl", hash = "sha256:a996d01ca39b8dfe77440f3cd600825d05841088fd6bc0144cc6c2ec14cc5f74"}, + {file = "aiohttp-3.8.1-cp39-cp39-macosx_11_0_arm64.whl", hash = "sha256:13487abd2f761d4be7c8ff9080de2671e53fff69711d46de703c310c4c9317ca"}, + {file = "aiohttp-3.8.1-cp39-cp39-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:ea302f34477fda3f85560a06d9ebdc7fa41e82420e892fc50b577e35fc6a50b2"}, + {file = "aiohttp-3.8.1-cp39-cp39-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:a2f635ce61a89c5732537a7896b6319a8fcfa23ba09bec36e1b1ac0ab31270d2"}, + {file = "aiohttp-3.8.1-cp39-cp39-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:e999f2d0e12eea01caeecb17b653f3713d758f6dcc770417cf29ef08d3931421"}, + {file = "aiohttp-3.8.1-cp39-cp39-manylinux_2_5_i686.manylinux1_i686.manylinux_2_12_i686.manylinux2010_i686.whl", hash = "sha256:0770e2806a30e744b4e21c9d73b7bee18a1cfa3c47991ee2e5a65b887c49d5cf"}, + {file = "aiohttp-3.8.1-cp39-cp39-manylinux_2_5_x86_64.manylinux1_x86_64.manylinux_2_12_x86_64.manylinux2010_x86_64.whl", hash = "sha256:d15367ce87c8e9e09b0f989bfd72dc641bcd04ba091c68cd305312d00962addd"}, + {file = "aiohttp-3.8.1-cp39-cp39-musllinux_1_1_aarch64.whl", hash = "sha256:6c7cefb4b0640703eb1069835c02486669312bf2f12b48a748e0a7756d0de33d"}, + {file = "aiohttp-3.8.1-cp39-cp39-musllinux_1_1_i686.whl", hash = "sha256:71927042ed6365a09a98a6377501af5c9f0a4d38083652bcd2281a06a5976724"}, + {file = "aiohttp-3.8.1-cp39-cp39-musllinux_1_1_ppc64le.whl", hash = "sha256:28d490af82bc6b7ce53ff31337a18a10498303fe66f701ab65ef27e143c3b0ef"}, + {file = "aiohttp-3.8.1-cp39-cp39-musllinux_1_1_s390x.whl", hash = "sha256:b6613280ccedf24354406caf785db748bebbddcf31408b20c0b48cb86af76866"}, + {file = "aiohttp-3.8.1-cp39-cp39-musllinux_1_1_x86_64.whl", hash = "sha256:81e3d8c34c623ca4e36c46524a3530e99c0bc95ed068fd6e9b55cb721d408fb2"}, + {file = "aiohttp-3.8.1-cp39-cp39-win32.whl", hash = "sha256:7187a76598bdb895af0adbd2fb7474d7f6025d170bc0a1130242da817ce9e7d1"}, + {file = "aiohttp-3.8.1-cp39-cp39-win_amd64.whl", hash = "sha256:1c182cb873bc91b411e184dab7a2b664d4fea2743df0e4d57402f7f3fa644bac"}, + {file = "aiohttp-3.8.1.tar.gz", hash = "sha256:fc5471e1a54de15ef71c1bc6ebe80d4dc681ea600e68bfd1cbce40427f0b7578"}, +] +aiosignal = [ + {file = "aiosignal-1.2.0-py3-none-any.whl", hash = "sha256:26e62109036cd181df6e6ad646f91f0dcfd05fe16d0cb924138ff2ab75d64e3a"}, + {file = "aiosignal-1.2.0.tar.gz", hash = "sha256:78ed67db6c7b7ced4f98e495e572106d5c432a93e1ddd1bf475e1dc05f5b7df2"}, ] apscheduler = [ - {file = "APScheduler-3.8.0-py2.py3-none-any.whl", hash = "sha256:82d6d21b2f0343510d07bb35618333a794653439a265635555b12935647b460a"}, - {file = "APScheduler-3.8.0.tar.gz", hash = "sha256:793b2d37c52ece53e34626619e6142e99b20b59a12155f39e1e6932e324f079d"}, + {file = "APScheduler-3.8.1-py2.py3-none-any.whl", hash = "sha256:c22cb14b411a31435eb2c530dfbbec948ac63015b517087c7978adb61b574865"}, + {file = "APScheduler-3.8.1.tar.gz", hash = "sha256:5cf344ebcfbdaa48ae178c029c055cec7bc7a4a47c21e315e4d1f08bd35f2355"}, ] asn1crypto = [ {file = "asn1crypto-1.4.0-py2.py3-none-any.whl", hash = "sha256:4bcdf33c861c7d40bdcd74d8e4dd7661aac320fcdf40b9a3f95b4ee12fde2fa8"}, {file = "asn1crypto-1.4.0.tar.gz", hash = "sha256:f4f6e119474e58e04a2b1af817eb585b4fd72bdd89b998624712b5c99be7641c"}, ] async-timeout = [ - {file = "async-timeout-3.0.1.tar.gz", hash = "sha256:0c3c816a028d47f659d6ff5c745cb2acf1f966da1fe5c19c77a70282b25f4c5f"}, - {file = "async_timeout-3.0.1-py3-none-any.whl", hash = "sha256:4291ca197d287d274d0b6cb5d6f8f8f82d434ed288f962539ff18cc9012f9ea3"}, + {file = "async-timeout-4.0.1.tar.gz", hash = "sha256:b930cb161a39042f9222f6efb7301399c87eeab394727ec5437924a36d6eef51"}, + {file = "async_timeout-4.0.1-py3-none-any.whl", hash = "sha256:a22c0b311af23337eb05fcf05a8b51c3ea53729d46fb5460af62bee033cec690"}, ] atomicwrites = [ {file = "atomicwrites-1.4.0-py2.py3-none-any.whl", hash = "sha256:6d1784dea7c0c8d4a5172b6c620f40b6e4cbfdf96d783691f2e1302a7b88e197"}, @@ -895,12 +989,12 @@ attrs = [ {file = "attrs-21.2.0.tar.gz", hash = "sha256:ef6aaac3ca6cd92904cdd0d83f629a15f18053ec84e6432106f7a4d04ae4f5fb"}, ] "backports.entry-points-selectable" = [ - {file = "backports.entry_points_selectable-1.1.0-py2.py3-none-any.whl", hash = "sha256:a6d9a871cde5e15b4c4a53e3d43ba890cc6861ec1332c9c2428c92f977192acc"}, - {file = "backports.entry_points_selectable-1.1.0.tar.gz", hash = "sha256:988468260ec1c196dab6ae1149260e2f5472c9110334e5d51adcb77867361f6a"}, + {file = "backports.entry_points_selectable-1.1.1-py2.py3-none-any.whl", hash = "sha256:7fceed9532a7aa2bd888654a7314f864a3c16a4e710b34a58cfc0f08114c663b"}, + {file = "backports.entry_points_selectable-1.1.1.tar.gz", hash = "sha256:914b21a479fde881635f7af5adc7f6e38d6b274be32269070c53b698c60d5386"}, ] black = [ - {file = "black-21.9b0-py3-none-any.whl", hash = "sha256:380f1b5da05e5a1429225676655dddb96f5ae8c75bdf91e53d798871b902a115"}, - {file = "black-21.9b0.tar.gz", hash = "sha256:7de4cfc7eb6b710de325712d40125689101d21d25283eed7e9998722cf10eb91"}, + {file = "black-21.11b1-py3-none-any.whl", hash = "sha256:802c6c30b637b28645b7fde282ed2569c0cd777dbe493a41b6a03c1d903f99ac"}, + {file = "black-21.11b1.tar.gz", hash = "sha256:a042adbb18b3262faad5aff4e834ff186bb893f95ba3a8013f09de1e5569def2"}, ] certifi = [ {file = "certifi-2021.10.8-py2.py3-none-any.whl", hash = "sha256:d62a0163eb4c2344ac042ab2bdf75399a71a2d8c7d47eac2e2ee91b9d6339569"}, @@ -962,13 +1056,9 @@ cfgv = [ {file = "cfgv-3.3.1-py2.py3-none-any.whl", hash = "sha256:c6a0883f3917a037485059700b9e75da2464e6c27051014ad85ba6aaa5884426"}, {file = "cfgv-3.3.1.tar.gz", hash = "sha256:f5a830efb9ce7a445376bb66ec94c638a9787422f96264c98edc6bdeed8ab736"}, ] -chardet = [ - {file = "chardet-4.0.0-py2.py3-none-any.whl", hash = "sha256:f864054d66fd9118f2e67044ac8981a54775ec5b67aed0441892edb553d21da5"}, - {file = "chardet-4.0.0.tar.gz", hash = "sha256:0d6f53a15db4120f2b08c94f11e7d93d2c911ee118b6b30a04ec3ee8310179fa"}, -] charset-normalizer = [ - {file = "charset-normalizer-2.0.7.tar.gz", hash = "sha256:e019de665e2bcf9c2b64e2e5aa025fa991da8720daa3c1138cadd2fd1856aed0"}, - {file = "charset_normalizer-2.0.7-py3-none-any.whl", hash = "sha256:f7af805c321bfa1ce6714c51f254e0d5bb5e5834039bc17db7ebe3a4cec9492b"}, + {file = "charset-normalizer-2.0.9.tar.gz", hash = "sha256:b0b883e8e874edfdece9c28f314e3dd5badf067342e42fb162203335ae61aa2c"}, + {file = "charset_normalizer-2.0.9-py3-none-any.whl", hash = "sha256:1eecaa09422db5be9e29d7fc65664e6c33bd06f9ced7838578ba40d58bdf3721"}, ] click = [ {file = "click-8.0.3-py3-none-any.whl", hash = "sha256:353f466495adaeb40b6b5f592f9f91cb22372351c84caeb068132442a4518ef3"}, @@ -1012,8 +1102,82 @@ eight = [ {file = "eight-1.0.1.tar.gz", hash = "sha256:6465970d3c12a1d759151690629b10b50ffb124c9b62e912999ee9233378a562"}, ] filelock = [ - {file = "filelock-3.3.1-py3-none-any.whl", hash = "sha256:2b5eb3589e7fdda14599e7eb1a50e09b4cc14f34ed98b8ba56d33bfaafcbef2f"}, - {file = "filelock-3.3.1.tar.gz", hash = "sha256:34a9f35f95c441e7b38209775d6e0337f9a3759f3565f6c5798f19618527c76f"}, + {file = "filelock-3.4.0-py3-none-any.whl", hash = "sha256:2e139a228bcf56dd8b2274a65174d005c4a6b68540ee0bdbb92c76f43f29f7e8"}, + {file = "filelock-3.4.0.tar.gz", hash = "sha256:93d512b32a23baf4cac44ffd72ccf70732aeff7b8050fcaf6d3ec406d954baf4"}, +] +frozenlist = [ + {file = "frozenlist-1.2.0-cp310-cp310-macosx_10_9_universal2.whl", hash = "sha256:977a1438d0e0d96573fd679d291a1542097ea9f4918a8b6494b06610dfeefbf9"}, + {file = "frozenlist-1.2.0-cp310-cp310-macosx_10_9_x86_64.whl", hash = "sha256:a8d86547a5e98d9edd47c432f7a14b0c5592624b496ae9880fb6332f34af1edc"}, + {file = "frozenlist-1.2.0-cp310-cp310-macosx_11_0_arm64.whl", hash = "sha256:181754275d5d32487431a0a29add4f897968b7157204bc1eaaf0a0ce80c5ba7d"}, + {file = "frozenlist-1.2.0-cp310-cp310-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:5df31bb2b974f379d230a25943d9bf0d3bc666b4b0807394b131a28fca2b0e5f"}, + {file = "frozenlist-1.2.0-cp310-cp310-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:4766632cd8a68e4f10f156a12c9acd7b1609941525569dd3636d859d79279ed3"}, + {file = "frozenlist-1.2.0-cp310-cp310-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:16eef427c51cb1203a7c0ab59d1b8abccaba9a4f58c4bfca6ed278fc896dc193"}, + {file = "frozenlist-1.2.0-cp310-cp310-manylinux_2_5_i686.manylinux1_i686.manylinux_2_12_i686.manylinux2010_i686.whl", hash = "sha256:01d79515ed5aa3d699b05f6bdcf1fe9087d61d6b53882aa599a10853f0479c6c"}, + {file = "frozenlist-1.2.0-cp310-cp310-manylinux_2_5_x86_64.manylinux1_x86_64.manylinux_2_12_x86_64.manylinux2010_x86_64.whl", hash = "sha256:28e164722ea0df0cf6d48c4d5bdf3d19e87aaa6dfb39b0ba91153f224b912020"}, + {file = "frozenlist-1.2.0-cp310-cp310-musllinux_1_1_aarch64.whl", hash = "sha256:e63ad0beef6ece06475d29f47d1f2f29727805376e09850ebf64f90777962792"}, + {file = "frozenlist-1.2.0-cp310-cp310-musllinux_1_1_i686.whl", hash = "sha256:41de4db9b9501679cf7cddc16d07ac0f10ef7eb58c525a1c8cbff43022bddca4"}, + {file = "frozenlist-1.2.0-cp310-cp310-musllinux_1_1_ppc64le.whl", hash = "sha256:c6a9d84ee6427b65a81fc24e6ef589cb794009f5ca4150151251c062773e7ed2"}, + {file = "frozenlist-1.2.0-cp310-cp310-musllinux_1_1_s390x.whl", hash = "sha256:f5f3b2942c3b8b9bfe76b408bbaba3d3bb305ee3693e8b1d631fe0a0d4f93673"}, + {file = "frozenlist-1.2.0-cp310-cp310-musllinux_1_1_x86_64.whl", hash = "sha256:c98d3c04701773ad60d9545cd96df94d955329efc7743fdb96422c4b669c633b"}, + {file = "frozenlist-1.2.0-cp310-cp310-win32.whl", hash = "sha256:72cfbeab7a920ea9e74b19aa0afe3b4ad9c89471e3badc985d08756efa9b813b"}, + {file = "frozenlist-1.2.0-cp310-cp310-win_amd64.whl", hash = "sha256:11ff401951b5ac8c0701a804f503d72c048173208490c54ebb8d7bb7c07a6d00"}, + {file = "frozenlist-1.2.0-cp36-cp36m-macosx_10_9_x86_64.whl", hash = "sha256:b46f997d5ed6d222a863b02cdc9c299101ee27974d9bbb2fd1b3c8441311c408"}, + {file = "frozenlist-1.2.0-cp36-cp36m-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:351686ca020d1bcd238596b1fa5c8efcbc21bffda9d0efe237aaa60348421e2a"}, + {file = "frozenlist-1.2.0-cp36-cp36m-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:bfbaa08cf1452acad9cb1c1d7b89394a41e712f88df522cea1a0f296b57782a0"}, + {file = "frozenlist-1.2.0-cp36-cp36m-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:b2ae2f5e9fa10805fb1c9adbfefaaecedd9e31849434be462c3960a0139ed729"}, + {file = "frozenlist-1.2.0-cp36-cp36m-manylinux_2_5_i686.manylinux1_i686.manylinux_2_12_i686.manylinux2010_i686.whl", hash = "sha256:6790b8d96bbb74b7a6f4594b6f131bd23056c25f2aa5d816bd177d95245a30e3"}, + {file = "frozenlist-1.2.0-cp36-cp36m-manylinux_2_5_x86_64.manylinux1_x86_64.manylinux_2_12_x86_64.manylinux2010_x86_64.whl", hash = "sha256:41f62468af1bd4e4b42b5508a3fe8cc46a693f0cdd0ca2f443f51f207893d837"}, + {file = "frozenlist-1.2.0-cp36-cp36m-musllinux_1_1_aarch64.whl", hash = "sha256:ec6cf345771cdb00791d271af9a0a6fbfc2b6dd44cb753f1eeaa256e21622adb"}, + {file = "frozenlist-1.2.0-cp36-cp36m-musllinux_1_1_i686.whl", hash = "sha256:14a5cef795ae3e28fb504b73e797c1800e9249f950e1c964bb6bdc8d77871161"}, + {file = "frozenlist-1.2.0-cp36-cp36m-musllinux_1_1_ppc64le.whl", hash = "sha256:8b54cdd2fda15467b9b0bfa78cee2ddf6dbb4585ef23a16e14926f4b076dfae4"}, + {file = "frozenlist-1.2.0-cp36-cp36m-musllinux_1_1_s390x.whl", hash = "sha256:f025f1d6825725b09c0038775acab9ae94264453a696cc797ce20c0769a7b367"}, + {file = "frozenlist-1.2.0-cp36-cp36m-musllinux_1_1_x86_64.whl", hash = "sha256:84e97f59211b5b9083a2e7a45abf91cfb441369e8bb6d1f5287382c1c526def3"}, + {file = "frozenlist-1.2.0-cp36-cp36m-win32.whl", hash = "sha256:c5328ed53fdb0a73c8a50105306a3bc013e5ca36cca714ec4f7bd31d38d8a97f"}, + {file = "frozenlist-1.2.0-cp36-cp36m-win_amd64.whl", hash = "sha256:9ade70aea559ca98f4b1b1e5650c45678052e76a8ab2f76d90f2ac64180215a2"}, + {file = "frozenlist-1.2.0-cp37-cp37m-macosx_10_9_x86_64.whl", hash = "sha256:a0d3ffa8772464441b52489b985d46001e2853a3b082c655ec5fad9fb6a3d618"}, + {file = "frozenlist-1.2.0-cp37-cp37m-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:3457f8cf86deb6ce1ba67e120f1b0128fcba1332a180722756597253c465fc1d"}, + {file = "frozenlist-1.2.0-cp37-cp37m-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:5a72eecf37eface331636951249d878750db84034927c997d47f7f78a573b72b"}, + {file = "frozenlist-1.2.0-cp37-cp37m-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:acc4614e8d1feb9f46dd829a8e771b8f5c4b1051365d02efb27a3229048ade8a"}, + {file = "frozenlist-1.2.0-cp37-cp37m-manylinux_2_5_i686.manylinux1_i686.manylinux_2_12_i686.manylinux2010_i686.whl", hash = "sha256:87521e32e18a2223311afc2492ef2d99946337da0779ddcda77b82ee7319df59"}, + {file = "frozenlist-1.2.0-cp37-cp37m-manylinux_2_5_x86_64.manylinux1_x86_64.manylinux_2_12_x86_64.manylinux2010_x86_64.whl", hash = "sha256:8b4c7665a17c3a5430edb663e4ad4e1ad457614d1b2f2b7f87052e2ef4fa45ca"}, + {file = "frozenlist-1.2.0-cp37-cp37m-musllinux_1_1_aarch64.whl", hash = "sha256:ed58803563a8c87cf4c0771366cf0ad1aa265b6b0ae54cbbb53013480c7ad74d"}, + {file = "frozenlist-1.2.0-cp37-cp37m-musllinux_1_1_i686.whl", hash = "sha256:aa44c4740b4e23fcfa259e9dd52315d2b1770064cde9507457e4c4a65a04c397"}, + {file = "frozenlist-1.2.0-cp37-cp37m-musllinux_1_1_ppc64le.whl", hash = "sha256:2de5b931701257d50771a032bba4e448ff958076380b049fd36ed8738fdb375b"}, + {file = "frozenlist-1.2.0-cp37-cp37m-musllinux_1_1_s390x.whl", hash = "sha256:6e105013fa84623c057a4381dc8ea0361f4d682c11f3816cc80f49a1f3bc17c6"}, + {file = "frozenlist-1.2.0-cp37-cp37m-musllinux_1_1_x86_64.whl", hash = "sha256:705c184b77565955a99dc360f359e8249580c6b7eaa4dc0227caa861ef46b27a"}, + {file = "frozenlist-1.2.0-cp37-cp37m-win32.whl", hash = "sha256:a37594ad6356e50073fe4f60aa4187b97d15329f2138124d252a5a19c8553ea4"}, + {file = "frozenlist-1.2.0-cp37-cp37m-win_amd64.whl", hash = "sha256:25b358aaa7dba5891b05968dd539f5856d69f522b6de0bf34e61f133e077c1a4"}, + {file = "frozenlist-1.2.0-cp38-cp38-macosx_10_9_universal2.whl", hash = "sha256:af2a51c8a381d76eabb76f228f565ed4c3701441ecec101dd18be70ebd483cfd"}, + {file = "frozenlist-1.2.0-cp38-cp38-macosx_10_9_x86_64.whl", hash = "sha256:82d22f6e6f2916e837c91c860140ef9947e31194c82aaeda843d6551cec92f19"}, + {file = "frozenlist-1.2.0-cp38-cp38-macosx_11_0_arm64.whl", hash = "sha256:1cfe6fef507f8bac40f009c85c7eddfed88c1c0d38c75e72fe10476cef94e10f"}, + {file = "frozenlist-1.2.0-cp38-cp38-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:26f602e380a5132880fa245c92030abb0fc6ff34e0c5500600366cedc6adb06a"}, + {file = "frozenlist-1.2.0-cp38-cp38-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:4ad065b2ebd09f32511ff2be35c5dfafee6192978b5a1e9d279a5c6e121e3b03"}, + {file = "frozenlist-1.2.0-cp38-cp38-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:bc93f5f62df3bdc1f677066327fc81f92b83644852a31c6aa9b32c2dde86ea7d"}, + {file = "frozenlist-1.2.0-cp38-cp38-manylinux_2_5_i686.manylinux1_i686.manylinux_2_12_i686.manylinux2010_i686.whl", hash = "sha256:89fdfc84c6bf0bff2ff3170bb34ecba8a6911b260d318d377171429c4be18c73"}, + {file = "frozenlist-1.2.0-cp38-cp38-manylinux_2_5_x86_64.manylinux1_x86_64.manylinux_2_12_x86_64.manylinux2010_x86_64.whl", hash = "sha256:47b2848e464883d0bbdcd9493c67443e5e695a84694efff0476f9059b4cb6257"}, + {file = "frozenlist-1.2.0-cp38-cp38-musllinux_1_1_aarch64.whl", hash = "sha256:4f52d0732e56906f8ddea4bd856192984650282424049c956857fed43697ea43"}, + {file = "frozenlist-1.2.0-cp38-cp38-musllinux_1_1_i686.whl", hash = "sha256:16ef7dd5b7d17495404a2e7a49bac1bc13d6d20c16d11f4133c757dd94c4144c"}, + {file = "frozenlist-1.2.0-cp38-cp38-musllinux_1_1_ppc64le.whl", hash = "sha256:1cf63243bc5f5c19762943b0aa9e0d3fb3723d0c514d820a18a9b9a5ef864315"}, + {file = "frozenlist-1.2.0-cp38-cp38-musllinux_1_1_s390x.whl", hash = "sha256:54a1e09ab7a69f843cd28fefd2bcaf23edb9e3a8d7680032c8968b8ac934587d"}, + {file = "frozenlist-1.2.0-cp38-cp38-musllinux_1_1_x86_64.whl", hash = "sha256:954b154a4533ef28bd3e83ffdf4eadf39deeda9e38fb8feaf066d6069885e034"}, + {file = "frozenlist-1.2.0-cp38-cp38-win32.whl", hash = "sha256:cb3957c39668d10e2b486acc85f94153520a23263b6401e8f59422ef65b9520d"}, + {file = "frozenlist-1.2.0-cp38-cp38-win_amd64.whl", hash = "sha256:0a7c7cce70e41bc13d7d50f0e5dd175f14a4f1837a8549b0936ed0cbe6170bf9"}, + {file = "frozenlist-1.2.0-cp39-cp39-macosx_10_9_universal2.whl", hash = "sha256:4c457220468d734e3077580a3642b7f682f5fd9507f17ddf1029452450912cdc"}, + {file = "frozenlist-1.2.0-cp39-cp39-macosx_10_9_x86_64.whl", hash = "sha256:e74f8b4d8677ebb4015ac01fcaf05f34e8a1f22775db1f304f497f2f88fdc697"}, + {file = "frozenlist-1.2.0-cp39-cp39-macosx_11_0_arm64.whl", hash = "sha256:fbd4844ff111449f3bbe20ba24fbb906b5b1c2384d0f3287c9f7da2354ce6d23"}, + {file = "frozenlist-1.2.0-cp39-cp39-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:f0081a623c886197ff8de9e635528fd7e6a387dccef432149e25c13946cb0cd0"}, + {file = "frozenlist-1.2.0-cp39-cp39-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:9b6e21e5770df2dea06cb7b6323fbc008b13c4a4e3b52cb54685276479ee7676"}, + {file = "frozenlist-1.2.0-cp39-cp39-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:406aeb340613b4b559db78d86864485f68919b7141dec82aba24d1477fd2976f"}, + {file = "frozenlist-1.2.0-cp39-cp39-manylinux_2_5_i686.manylinux1_i686.manylinux_2_12_i686.manylinux2010_i686.whl", hash = "sha256:878ebe074839d649a1cdb03a61077d05760624f36d196884a5cafb12290e187b"}, + {file = "frozenlist-1.2.0-cp39-cp39-manylinux_2_5_x86_64.manylinux1_x86_64.manylinux_2_12_x86_64.manylinux2010_x86_64.whl", hash = "sha256:1fef737fd1388f9b93bba8808c5f63058113c10f4e3c0763ced68431773f72f9"}, + {file = "frozenlist-1.2.0-cp39-cp39-musllinux_1_1_aarch64.whl", hash = "sha256:4a495c3d513573b0b3f935bfa887a85d9ae09f0627cf47cad17d0cc9b9ba5c38"}, + {file = "frozenlist-1.2.0-cp39-cp39-musllinux_1_1_i686.whl", hash = "sha256:e7d0dd3e727c70c2680f5f09a0775525229809f1a35d8552b92ff10b2b14f2c2"}, + {file = "frozenlist-1.2.0-cp39-cp39-musllinux_1_1_ppc64le.whl", hash = "sha256:66a518731a21a55b7d3e087b430f1956a36793acc15912e2878431c7aec54210"}, + {file = "frozenlist-1.2.0-cp39-cp39-musllinux_1_1_s390x.whl", hash = "sha256:94728f97ddf603d23c8c3dd5cae2644fa12d33116e69f49b1644a71bb77b89ae"}, + {file = "frozenlist-1.2.0-cp39-cp39-musllinux_1_1_x86_64.whl", hash = "sha256:c1e8e9033d34c2c9e186e58279879d78c94dd365068a3607af33f2bc99357a53"}, + {file = "frozenlist-1.2.0-cp39-cp39-win32.whl", hash = "sha256:83334e84a290a158c0c4cc4d22e8c7cfe0bba5b76d37f1c2509dabd22acafe15"}, + {file = "frozenlist-1.2.0-cp39-cp39-win_amd64.whl", hash = "sha256:735f386ec522e384f511614c01d2ef9cf799f051353876b4c6fb93ef67a6d1ee"}, + {file = "frozenlist-1.2.0.tar.gz", hash = "sha256:68201be60ac56aff972dc18085800b6ee07973c49103a8aba669dee3d71079de"}, ] future = [ {file = "future-0.18.2.tar.gz", hash = "sha256:b1bead90b70cf6ec3f0710ae53a525360fa360d306a86583adc6bf83a4db537d"}, @@ -1067,70 +1231,82 @@ grequests = [ {file = "grequests-0.6.0.tar.gz", hash = "sha256:7dec890c6668e6755a1ea968565535867956639301268394d24df67b478df666"}, ] identify = [ - {file = "identify-2.3.0-py2.py3-none-any.whl", hash = "sha256:d1e82c83d063571bb88087676f81261a4eae913c492dafde184067c584bc7c05"}, - {file = "identify-2.3.0.tar.gz", hash = "sha256:fd08c97f23ceee72784081f1ce5125c8f53a02d3f2716dde79a6ab8f1039fea5"}, + {file = "identify-2.4.0-py2.py3-none-any.whl", hash = "sha256:eba31ca80258de6bb51453084bff4a923187cd2193b9c13710f2516ab30732cc"}, + {file = "identify-2.4.0.tar.gz", hash = "sha256:a33ae873287e81651c7800ca309dc1f84679b763c9c8b30680e16fbfa82f0107"}, ] idna = [ - {file = "idna-2.10-py2.py3-none-any.whl", hash = "sha256:b97d804b1e9b523befed77c48dacec60e6dcb0b5391d57af6a65a312a90648c0"}, - {file = "idna-2.10.tar.gz", hash = "sha256:b307872f855b18632ce0c21c5e45be78c0ea7ae4c15c828c20788b26921eb3f6"}, + {file = "idna-3.3-py3-none-any.whl", hash = "sha256:84d9dd047ffa80596e0f246e2eab0b391788b0503584e8945f2368256d2735ff"}, + {file = "idna-3.3.tar.gz", hash = "sha256:9d643ff0a55b762d5cdb124b8eaa99c66322e2157b69160bc32796e824360e6d"}, ] iniconfig = [ {file = "iniconfig-1.1.1-py2.py3-none-any.whl", hash = "sha256:011e24c64b7f47f6ebd835bb12a743f2fbe9a26d4cecaa7f53bc4f35ee9da8b3"}, {file = "iniconfig-1.1.1.tar.gz", hash = "sha256:bc3af051d7d14b2ee5ef9969666def0cd1a000e121eaea580d4a313df4b37f32"}, ] jinja2 = [ - {file = "Jinja2-3.0.2-py3-none-any.whl", hash = "sha256:8569982d3f0889eed11dd620c706d39b60c36d6d25843961f33f77fb6bc6b20c"}, - {file = "Jinja2-3.0.2.tar.gz", hash = "sha256:827a0e32839ab1600d4eb1c4c33ec5a8edfbc5cb42dafa13b81f182f97784b45"}, + {file = "Jinja2-3.0.3-py3-none-any.whl", hash = "sha256:077ce6014f7b40d03b47d1f1ca4b0fc8328a692bd284016f806ed0eaca390ad8"}, + {file = "Jinja2-3.0.3.tar.gz", hash = "sha256:611bb273cd68f3b993fabdc4064fc858c5b47a973cb5aa7999ec1ba405c87cd7"}, ] lxml = [ - {file = "lxml-4.6.3-cp27-cp27m-macosx_10_9_x86_64.whl", hash = "sha256:df7c53783a46febb0e70f6b05df2ba104610f2fb0d27023409734a3ecbb78fb2"}, - {file = "lxml-4.6.3-cp27-cp27m-manylinux1_i686.whl", hash = "sha256:1b7584d421d254ab86d4f0b13ec662a9014397678a7c4265a02a6d7c2b18a75f"}, - {file = "lxml-4.6.3-cp27-cp27m-manylinux1_x86_64.whl", hash = "sha256:079f3ae844f38982d156efce585bc540c16a926d4436712cf4baee0cce487a3d"}, - {file = "lxml-4.6.3-cp27-cp27m-win32.whl", hash = "sha256:bc4313cbeb0e7a416a488d72f9680fffffc645f8a838bd2193809881c67dd106"}, - {file = "lxml-4.6.3-cp27-cp27m-win_amd64.whl", hash = "sha256:8157dadbb09a34a6bd95a50690595e1fa0af1a99445e2744110e3dca7831c4ee"}, - {file = "lxml-4.6.3-cp27-cp27mu-manylinux1_i686.whl", hash = "sha256:7728e05c35412ba36d3e9795ae8995e3c86958179c9770e65558ec3fdfd3724f"}, - {file = "lxml-4.6.3-cp27-cp27mu-manylinux1_x86_64.whl", hash = "sha256:4bff24dfeea62f2e56f5bab929b4428ae6caba2d1eea0c2d6eb618e30a71e6d4"}, - {file = "lxml-4.6.3-cp310-cp310-manylinux_2_12_i686.manylinux2010_i686.manylinux_2_24_i686.whl", hash = "sha256:64812391546a18896adaa86c77c59a4998f33c24788cadc35789e55b727a37f4"}, - {file = "lxml-4.6.3-cp310-cp310-manylinux_2_17_x86_64.manylinux2014_x86_64.manylinux_2_24_x86_64.whl", hash = "sha256:c1a40c06fd5ba37ad39caa0b3144eb3772e813b5fb5b084198a985431c2f1e8d"}, - {file = "lxml-4.6.3-cp35-cp35m-manylinux1_i686.whl", hash = "sha256:74f7d8d439b18fa4c385f3f5dfd11144bb87c1da034a466c5b5577d23a1d9b51"}, - {file = "lxml-4.6.3-cp35-cp35m-manylinux1_x86_64.whl", hash = "sha256:f90ba11136bfdd25cae3951af8da2e95121c9b9b93727b1b896e3fa105b2f586"}, - {file = "lxml-4.6.3-cp35-cp35m-manylinux2010_i686.whl", hash = "sha256:4c61b3a0db43a1607d6264166b230438f85bfed02e8cff20c22e564d0faff354"}, - {file = "lxml-4.6.3-cp35-cp35m-manylinux2014_x86_64.whl", hash = "sha256:5c8c163396cc0df3fd151b927e74f6e4acd67160d6c33304e805b84293351d16"}, - {file = "lxml-4.6.3-cp35-cp35m-win32.whl", hash = "sha256:f2380a6376dfa090227b663f9678150ef27543483055cc327555fb592c5967e2"}, - {file = "lxml-4.6.3-cp35-cp35m-win_amd64.whl", hash = "sha256:c4f05c5a7c49d2fb70223d0d5bcfbe474cf928310ac9fa6a7c6dddc831d0b1d4"}, - {file = "lxml-4.6.3-cp36-cp36m-macosx_10_9_x86_64.whl", hash = "sha256:d2e35d7bf1c1ac8c538f88d26b396e73dd81440d59c1ef8522e1ea77b345ede4"}, - {file = "lxml-4.6.3-cp36-cp36m-manylinux1_i686.whl", hash = "sha256:289e9ca1a9287f08daaf796d96e06cb2bc2958891d7911ac7cae1c5f9e1e0ee3"}, - {file = "lxml-4.6.3-cp36-cp36m-manylinux1_x86_64.whl", hash = "sha256:bccbfc27563652de7dc9bdc595cb25e90b59c5f8e23e806ed0fd623755b6565d"}, - {file = "lxml-4.6.3-cp36-cp36m-manylinux2010_i686.whl", hash = "sha256:d916d31fd85b2f78c76400d625076d9124de3e4bda8b016d25a050cc7d603f24"}, - {file = "lxml-4.6.3-cp36-cp36m-manylinux2014_aarch64.whl", hash = "sha256:820628b7b3135403540202e60551e741f9b6d3304371712521be939470b454ec"}, - {file = "lxml-4.6.3-cp36-cp36m-manylinux2014_x86_64.whl", hash = "sha256:c47ff7e0a36d4efac9fd692cfa33fbd0636674c102e9e8d9b26e1b93a94e7617"}, - {file = "lxml-4.6.3-cp36-cp36m-win32.whl", hash = "sha256:5a0a14e264069c03e46f926be0d8919f4105c1623d620e7ec0e612a2e9bf1c04"}, - {file = "lxml-4.6.3-cp36-cp36m-win_amd64.whl", hash = "sha256:92e821e43ad382332eade6812e298dc9701c75fe289f2a2d39c7960b43d1e92a"}, - {file = "lxml-4.6.3-cp37-cp37m-macosx_10_9_x86_64.whl", hash = "sha256:efd7a09678fd8b53117f6bae4fa3825e0a22b03ef0a932e070c0bdbb3a35e654"}, - {file = "lxml-4.6.3-cp37-cp37m-manylinux1_i686.whl", hash = "sha256:efac139c3f0bf4f0939f9375af4b02c5ad83a622de52d6dfa8e438e8e01d0eb0"}, - {file = "lxml-4.6.3-cp37-cp37m-manylinux1_x86_64.whl", hash = "sha256:0fbcf5565ac01dff87cbfc0ff323515c823081c5777a9fc7703ff58388c258c3"}, - {file = "lxml-4.6.3-cp37-cp37m-manylinux2010_i686.whl", hash = "sha256:36108c73739985979bf302006527cf8a20515ce444ba916281d1c43938b8bb96"}, - {file = "lxml-4.6.3-cp37-cp37m-manylinux2014_aarch64.whl", hash = "sha256:122fba10466c7bd4178b07dba427aa516286b846b2cbd6f6169141917283aae2"}, - {file = "lxml-4.6.3-cp37-cp37m-manylinux2014_x86_64.whl", hash = "sha256:cdaf11d2bd275bf391b5308f86731e5194a21af45fbaaaf1d9e8147b9160ea92"}, - {file = "lxml-4.6.3-cp37-cp37m-win32.whl", hash = "sha256:3439c71103ef0e904ea0a1901611863e51f50b5cd5e8654a151740fde5e1cade"}, - {file = "lxml-4.6.3-cp37-cp37m-win_amd64.whl", hash = "sha256:4289728b5e2000a4ad4ab8da6e1db2e093c63c08bdc0414799ee776a3f78da4b"}, - {file = "lxml-4.6.3-cp38-cp38-macosx_10_9_x86_64.whl", hash = "sha256:b007cbb845b28db4fb8b6a5cdcbf65bacb16a8bd328b53cbc0698688a68e1caa"}, - {file = "lxml-4.6.3-cp38-cp38-manylinux1_i686.whl", hash = "sha256:76fa7b1362d19f8fbd3e75fe2fb7c79359b0af8747e6f7141c338f0bee2f871a"}, - {file = "lxml-4.6.3-cp38-cp38-manylinux1_x86_64.whl", hash = "sha256:26e761ab5b07adf5f555ee82fb4bfc35bf93750499c6c7614bd64d12aaa67927"}, - {file = "lxml-4.6.3-cp38-cp38-manylinux2010_i686.whl", hash = "sha256:e1cbd3f19a61e27e011e02f9600837b921ac661f0c40560eefb366e4e4fb275e"}, - {file = "lxml-4.6.3-cp38-cp38-manylinux2014_aarch64.whl", hash = "sha256:66e575c62792c3f9ca47cb8b6fab9e35bab91360c783d1606f758761810c9791"}, - {file = "lxml-4.6.3-cp38-cp38-manylinux2014_x86_64.whl", hash = "sha256:1b38116b6e628118dea5b2186ee6820ab138dbb1e24a13e478490c7db2f326ae"}, - {file = "lxml-4.6.3-cp38-cp38-win32.whl", hash = "sha256:89b8b22a5ff72d89d48d0e62abb14340d9e99fd637d046c27b8b257a01ffbe28"}, - {file = "lxml-4.6.3-cp38-cp38-win_amd64.whl", hash = "sha256:2a9d50e69aac3ebee695424f7dbd7b8c6d6eb7de2a2eb6b0f6c7db6aa41e02b7"}, - {file = "lxml-4.6.3-cp39-cp39-macosx_10_9_x86_64.whl", hash = "sha256:ce256aaa50f6cc9a649c51be3cd4ff142d67295bfc4f490c9134d0f9f6d58ef0"}, - {file = "lxml-4.6.3-cp39-cp39-manylinux1_i686.whl", hash = "sha256:7610b8c31688f0b1be0ef882889817939490a36d0ee880ea562a4e1399c447a1"}, - {file = "lxml-4.6.3-cp39-cp39-manylinux1_x86_64.whl", hash = "sha256:f8380c03e45cf09f8557bdaa41e1fa7c81f3ae22828e1db470ab2a6c96d8bc23"}, - {file = "lxml-4.6.3-cp39-cp39-manylinux2010_i686.whl", hash = "sha256:3082c518be8e97324390614dacd041bb1358c882d77108ca1957ba47738d9d59"}, - {file = "lxml-4.6.3-cp39-cp39-manylinux2014_aarch64.whl", hash = "sha256:884ab9b29feaca361f7f88d811b1eea9bfca36cf3da27768d28ad45c3ee6f969"}, - {file = "lxml-4.6.3-cp39-cp39-manylinux2014_x86_64.whl", hash = "sha256:6f12e1427285008fd32a6025e38e977d44d6382cf28e7201ed10d6c1698d2a9a"}, - {file = "lxml-4.6.3-cp39-cp39-win32.whl", hash = "sha256:33bb934a044cf32157c12bfcfbb6649807da20aa92c062ef51903415c704704f"}, - {file = "lxml-4.6.3-cp39-cp39-win_amd64.whl", hash = "sha256:542d454665a3e277f76954418124d67516c5f88e51a900365ed54a9806122b83"}, - {file = "lxml-4.6.3.tar.gz", hash = "sha256:39b78571b3b30645ac77b95f7c69d1bffc4cf8c3b157c435a34da72e78c82468"}, + {file = "lxml-4.6.4-cp27-cp27m-macosx_10_14_x86_64.whl", hash = "sha256:bbf2dc330bd44bfc0254ab37677ec60f7c7ecea55ad8ba1b8b2ea7bf20c265f5"}, + {file = "lxml-4.6.4-cp27-cp27m-manylinux_2_5_i686.manylinux1_i686.whl", hash = "sha256:b667c51682fe9b9788c69465956baa8b6999531876ccedcafc895c74ad716cd8"}, + {file = "lxml-4.6.4-cp27-cp27m-manylinux_2_5_x86_64.manylinux1_x86_64.whl", hash = "sha256:72e730d33fe2e302fd07285f14624fca5e5e2fb2bb4fb2c3941e318c41c443d1"}, + {file = "lxml-4.6.4-cp27-cp27m-win32.whl", hash = "sha256:433df8c7dde0f9e41cbf4f36b0829d50a378116ef5e962ba3881f2f5f025c7be"}, + {file = "lxml-4.6.4-cp27-cp27m-win_amd64.whl", hash = "sha256:35752ee40f7bbf6adc9ff4e1f4b84794a3593736dcce80db32e3c2aa85e294ac"}, + {file = "lxml-4.6.4-cp27-cp27mu-manylinux_2_5_i686.manylinux1_i686.whl", hash = "sha256:5ff5bb2a198ea67403bb6818705e9a4f90e0313f2215428ec51001ce56d939fb"}, + {file = "lxml-4.6.4-cp27-cp27mu-manylinux_2_5_x86_64.manylinux1_x86_64.whl", hash = "sha256:9b87727561c1150c0cc91c5d9d389448b37a7d15f0ba939ed3d1acb2f11bf6c5"}, + {file = "lxml-4.6.4-cp310-cp310-macosx_10_14_x86_64.whl", hash = "sha256:45fdb2899c755138722797161547a40b3e2a06feda620cc41195ee7e97806d81"}, + {file = "lxml-4.6.4-cp310-cp310-manylinux_2_12_i686.manylinux2010_i686.manylinux_2_24_i686.whl", hash = "sha256:38b9de0de3aa689fe9fb9877ae1be1e83b8cf9621f7e62049d0436b9ecf4ad64"}, + {file = "lxml-4.6.4-cp310-cp310-manylinux_2_17_aarch64.manylinux2014_aarch64.manylinux_2_24_aarch64.whl", hash = "sha256:662523cd2a0246740225c7e32531f2e766544122e58bee70e700a024cfc0cf81"}, + {file = "lxml-4.6.4-cp310-cp310-manylinux_2_17_x86_64.manylinux2014_x86_64.manylinux_2_24_x86_64.whl", hash = "sha256:4aa349c5567651f34d4eaae7de6ed5b523f6d70a288f9c6fbac22d13a0784e04"}, + {file = "lxml-4.6.4-cp310-cp310-musllinux_1_1_x86_64.whl", hash = "sha256:08eb9200d88b376a8ed5e50f1dc1d1a45b49305169674002a3b5929943390591"}, + {file = "lxml-4.6.4-cp310-cp310-win32.whl", hash = "sha256:bdc224f216ead849e902151112efef6e96c41ee1322e15d4e5f7c8a826929aee"}, + {file = "lxml-4.6.4-cp310-cp310-win_amd64.whl", hash = "sha256:ab6db93a2b6b66cbf62b4e4a7135f476e708e8c5c990d186584142c77d7f975a"}, + {file = "lxml-4.6.4-cp35-cp35m-manylinux_2_5_i686.manylinux1_i686.whl", hash = "sha256:50790313df028aa05cf22be9a8da033b86c42fa32523e4fd944827b482b17bf0"}, + {file = "lxml-4.6.4-cp35-cp35m-manylinux_2_5_x86_64.manylinux1_x86_64.whl", hash = "sha256:6764998345552b1dfc9326a932d2bad6367c6b37a176bb73ada6b9486bf602f7"}, + {file = "lxml-4.6.4-cp35-cp35m-win32.whl", hash = "sha256:543b239b191bb3b6d9bef5f09f1fb2be5b7eb09ab4d386aa655e4d53fbe9ff47"}, + {file = "lxml-4.6.4-cp35-cp35m-win_amd64.whl", hash = "sha256:a75c1ad05eedb1a3ff2a34a52a4f0836cfaa892e12796ba39a7732c82701eff4"}, + {file = "lxml-4.6.4-cp36-cp36m-macosx_10_14_x86_64.whl", hash = "sha256:47e955112ce64241fdb357acf0216081f9f3255b3ac9c502ca4b3323ec1ca558"}, + {file = "lxml-4.6.4-cp36-cp36m-manylinux_2_12_i686.manylinux2010_i686.manylinux_2_24_i686.whl", hash = "sha256:20d7c8d90d449c6a353b15ee0459abae8395dbe59ad01e406ccbf30cd81c6f98"}, + {file = "lxml-4.6.4-cp36-cp36m-manylinux_2_17_x86_64.manylinux2014_x86_64.manylinux_2_24_x86_64.whl", hash = "sha256:240db6f3228d26e3c6f4fad914b9ddaaf8707254e8b3efd564dc680c8ec3c264"}, + {file = "lxml-4.6.4-cp36-cp36m-manylinux_2_5_i686.manylinux1_i686.whl", hash = "sha256:351482da8dd028834028537f08724b1de22d40dcf3bb723b469446564f409074"}, + {file = "lxml-4.6.4-cp36-cp36m-manylinux_2_5_x86_64.manylinux1_x86_64.whl", hash = "sha256:e678a643177c0e5ec947b645fa7bc84260dfb9b6bf8fb1fdd83008dfc2ca5928"}, + {file = "lxml-4.6.4-cp36-cp36m-musllinux_1_1_x86_64.whl", hash = "sha256:15d0381feb56f08f78c5cc4fc385ddfe0bde1456e37f54a9322833371aec4060"}, + {file = "lxml-4.6.4-cp36-cp36m-win32.whl", hash = "sha256:4ba74afe5ee5cb5e28d83b513a6e8f0875fda1dc1a9aea42cc0065f029160d2a"}, + {file = "lxml-4.6.4-cp36-cp36m-win_amd64.whl", hash = "sha256:9c91a73971a922c13070fd8fa5a114c858251791ba2122a941e6aa781c713e44"}, + {file = "lxml-4.6.4-cp37-cp37m-macosx_10_14_x86_64.whl", hash = "sha256:6020c70ff695106bf80651953a23e37718ef1fee9abd060dcad8e32ab2dc13f3"}, + {file = "lxml-4.6.4-cp37-cp37m-manylinux_2_12_i686.manylinux2010_i686.manylinux_2_24_i686.whl", hash = "sha256:f5dd358536b8a964bf6bd48de038754c1609e72e5f17f5d21efe2dda17594dbf"}, + {file = "lxml-4.6.4-cp37-cp37m-manylinux_2_17_aarch64.manylinux2014_aarch64.manylinux_2_24_aarch64.whl", hash = "sha256:7ae7089d81fc502df4b217ad77f03c54039fe90dac0acbe70448d7e53bfbc57e"}, + {file = "lxml-4.6.4-cp37-cp37m-manylinux_2_17_x86_64.manylinux2014_x86_64.manylinux_2_24_x86_64.whl", hash = "sha256:80d10d53d3184837445ff8562021bdd37f57c4cadacbf9d8726cc16220a00d54"}, + {file = "lxml-4.6.4-cp37-cp37m-manylinux_2_5_i686.manylinux1_i686.whl", hash = "sha256:e95da348d57eb448d226a44b868ff2ca5786fbcbe417ac99ff62d0a7d724b9c7"}, + {file = "lxml-4.6.4-cp37-cp37m-manylinux_2_5_x86_64.manylinux1_x86_64.whl", hash = "sha256:ffd65cfa33fed01735c82aca640fde4cc63f0414775cba11e06f84fae2085a6e"}, + {file = "lxml-4.6.4-cp37-cp37m-musllinux_1_1_x86_64.whl", hash = "sha256:877666418598f6cb289546c77ff87590cfd212f903b522b0afa0b9fb73b3ccfb"}, + {file = "lxml-4.6.4-cp37-cp37m-win32.whl", hash = "sha256:e91d24623e747eeb2d8121f4a94c6a7ad27dc48e747e2dc95bfe88632bd028a2"}, + {file = "lxml-4.6.4-cp37-cp37m-win_amd64.whl", hash = "sha256:4ec9a80dd5704ecfde54319b6964368daf02848c8954d3bacb9b64d1c7659159"}, + {file = "lxml-4.6.4-cp38-cp38-macosx_10_14_x86_64.whl", hash = "sha256:2901625f4a878a055d275beedc20ba9cb359cefc4386a967222fee29eb236038"}, + {file = "lxml-4.6.4-cp38-cp38-manylinux_2_12_i686.manylinux2010_i686.manylinux_2_24_i686.whl", hash = "sha256:b567178a74a2261345890eac66fbf394692a6e002709d329f28a673ca6042473"}, + {file = "lxml-4.6.4-cp38-cp38-manylinux_2_17_aarch64.manylinux2014_aarch64.manylinux_2_24_aarch64.whl", hash = "sha256:4717123f7c11c81e0da69989e5a64079c3f402b0efeb4c6241db6c369d657bd8"}, + {file = "lxml-4.6.4-cp38-cp38-manylinux_2_17_x86_64.manylinux2014_x86_64.manylinux_2_24_x86_64.whl", hash = "sha256:cf201bf5594d1aab139fe53e3fca457e4f8204a5bbd65d48ab3b82a16f517868"}, + {file = "lxml-4.6.4-cp38-cp38-manylinux_2_5_i686.manylinux1_i686.whl", hash = "sha256:a77a3470ba37e11872c75ca95baf9b3312133a3d5a5dc720803b23098c653976"}, + {file = "lxml-4.6.4-cp38-cp38-manylinux_2_5_x86_64.manylinux1_x86_64.whl", hash = "sha256:619c6d2b552bba00491e96c0518aad94002651c108a0f7364ff2d7798812c00e"}, + {file = "lxml-4.6.4-cp38-cp38-musllinux_1_1_x86_64.whl", hash = "sha256:601f0ab75538b280aaf1e720eb9d68d4fa104ac274e1e9e6971df488f4dcdb0f"}, + {file = "lxml-4.6.4-cp38-cp38-win32.whl", hash = "sha256:75d3c5bbc0ddbad03bb68b9be638599f67e4b98ed3dcd0fec9f6f39e41ee96cb"}, + {file = "lxml-4.6.4-cp38-cp38-win_amd64.whl", hash = "sha256:4341d135f5660db10184963d9c3418c3e28d7f868aaf8b11a323ebf85813f7f4"}, + {file = "lxml-4.6.4-cp39-cp39-macosx_10_14_x86_64.whl", hash = "sha256:9db24803fa71e3305fe4a7812782b708da21a0b774b130dd1860cf40a6d7a3ee"}, + {file = "lxml-4.6.4-cp39-cp39-manylinux_2_12_i686.manylinux2010_i686.manylinux_2_24_i686.whl", hash = "sha256:afd60230ad9d8bcba005945ec3a343722f09e0b7f8ae804246e5d2cfc6bd71a6"}, + {file = "lxml-4.6.4-cp39-cp39-manylinux_2_17_aarch64.manylinux2014_aarch64.manylinux_2_24_aarch64.whl", hash = "sha256:0c15e1cd55055956e77b0732270f1c6005850696bc3ef3e03d01e78af84eaa42"}, + {file = "lxml-4.6.4-cp39-cp39-manylinux_2_17_x86_64.manylinux2014_x86_64.manylinux_2_24_x86_64.whl", hash = "sha256:6d422b3c729737d8a39279a25fa156c983a56458f8b2f97661ee6fb22b80b1d6"}, + {file = "lxml-4.6.4-cp39-cp39-manylinux_2_5_i686.manylinux1_i686.whl", hash = "sha256:2eb90f6ec3c236ef2f1bb38aee7c0d23e77d423d395af6326e7cca637519a4cb"}, + {file = "lxml-4.6.4-cp39-cp39-manylinux_2_5_x86_64.manylinux1_x86_64.whl", hash = "sha256:51a0e5d243687596f46e24e464121d4b232ad772e2d1785b2a2c0eb413c285d4"}, + {file = "lxml-4.6.4-cp39-cp39-musllinux_1_1_x86_64.whl", hash = "sha256:d43bd68714049c84e297c005456a15ecdec818f7b5aa5868c8b0a865cfb78a44"}, + {file = "lxml-4.6.4-cp39-cp39-win32.whl", hash = "sha256:ee9e4b07b0eba4b6a521509e9e1877476729c1243246b6959de697ebea739643"}, + {file = "lxml-4.6.4-cp39-cp39-win_amd64.whl", hash = "sha256:48eaac2991b3036175b42ee8d3c23f4cca13f2be8426bf29401a690ab58c88f4"}, + {file = "lxml-4.6.4-pp37-pypy37_pp73-macosx_10_14_x86_64.whl", hash = "sha256:2b06a91cf7b8acea7793006e4ae50646cef0fe35ce5acd4f5cb1c77eb228e4a1"}, + {file = "lxml-4.6.4-pp37-pypy37_pp73-manylinux_2_12_i686.manylinux2010_i686.manylinux_2_24_i686.whl", hash = "sha256:523f195948a1ba4f9f5b7294d83c6cd876547dc741820750a7e5e893a24bbe38"}, + {file = "lxml-4.6.4-pp37-pypy37_pp73-manylinux_2_17_x86_64.manylinux2014_x86_64.manylinux_2_24_x86_64.whl", hash = "sha256:b0ca0ada9d3bc18bd6f611bd001a28abdd49ab9698bd6d717f7f5394c8e94628"}, + {file = "lxml-4.6.4-pp38-pypy38_pp73-macosx_10_14_x86_64.whl", hash = "sha256:197b7cb7a753cf553a45115739afd8458464a28913da00f5c525063f94cd3f48"}, + {file = "lxml-4.6.4-pp38-pypy38_pp73-manylinux_2_12_i686.manylinux2010_i686.manylinux_2_24_i686.whl", hash = "sha256:6298f5b42a26581206ef63fffa97c754245d329414108707c525512a5197f2ba"}, + {file = "lxml-4.6.4-pp38-pypy38_pp73-manylinux_2_17_x86_64.manylinux2014_x86_64.manylinux_2_24_x86_64.whl", hash = "sha256:0b12c95542f04d10cba46b3ff28ea52ea56995b78cf918f0b11b05e75812bb79"}, + {file = "lxml-4.6.4.tar.gz", hash = "sha256:daf9bd1fee31f1c7a5928b3e1059e09a8d683ea58fb3ffc773b6c88cb8d1399c"}, ] markupsafe = [ {file = "MarkupSafe-2.0.1-cp310-cp310-macosx_10_9_universal2.whl", hash = "sha256:d8446c54dc28c01e5a2dbac5a25f071f6653e6e40f3a8818e8b45d790fe6ef53"}, @@ -1294,15 +1470,22 @@ nodeenv = [ {file = "nodeenv-1.6.0-py2.py3-none-any.whl", hash = "sha256:621e6b7076565ddcacd2db0294c0381e01fd28945ab36bcf00f41c5daf63bef7"}, {file = "nodeenv-1.6.0.tar.gz", hash = "sha256:3ef13ff90291ba2a4a7a4ff9a979b63ffdd00a464dbe04acf0ea6471517a4c2b"}, ] -openleadr = [] +openleadr = [ + {file = "openleadr-0.5.24-py3-none-any.whl", hash = "sha256:8723ff25eb02ab8b31b0cdca7d2429f9bb21e337b9eb809978237c644bbf5390"}, + {file = "openleadr-0.5.24.tar.gz", hash = "sha256:c06078040ca4246c85ca6e8378c386737df8daf7df86dbebe82a74d666185de8"}, +] packaging = [ - {file = "packaging-21.0-py3-none-any.whl", hash = "sha256:c86254f9220d55e31cc94d69bade760f0847da8000def4dfe1c6b872fd14ff14"}, - {file = "packaging-21.0.tar.gz", hash = "sha256:7dc96269f53a4ccec5c0670940a4281106dd0bb343f47b7471f779df49c2fbe7"}, + {file = "packaging-21.3-py3-none-any.whl", hash = "sha256:ef103e05f519cdc783ae24ea4e2e0f508a9c99b2d4969652eed6a2e1ea5bd522"}, + {file = "packaging-21.3.tar.gz", hash = "sha256:dd47c42927d89ab911e606518907cc2d3a1f38bbd026385970643f9c5b8ecfeb"}, ] pathspec = [ {file = "pathspec-0.9.0-py2.py3-none-any.whl", hash = "sha256:7d15c4ddb0b5c802d161efc417ec1a2558ea2653c2e8ad9c19098201dc1c993a"}, {file = "pathspec-0.9.0.tar.gz", hash = "sha256:e564499435a2673d586f6b2130bb5b95f04a3ba06f81b8f895b651a3c76aabb1"}, ] +pbr = [ + {file = "pbr-5.8.0-py2.py3-none-any.whl", hash = "sha256:176e8560eaf61e127817ef93d8a844803abb27a4d4637f0ff3bb783129be2e0a"}, + {file = "pbr-5.8.0.tar.gz", hash = "sha256:672d8ebee84921862110f23fcec2acea191ef58543d34dfe9ef3d9f13c31cddf"}, +] platformdirs = [ {file = "platformdirs-2.4.0-py3-none-any.whl", hash = "sha256:8868bbe3c3c80d42f20156f22e7131d2fb321f5bc86a2a345375c6481a67021d"}, {file = "platformdirs-2.4.0.tar.gz", hash = "sha256:367a5e80b3d04d2428ffa76d33f124cf11e8fff2acdaa9b43d545f5c7d661ef2"}, @@ -1312,8 +1495,8 @@ pluggy = [ {file = "pluggy-1.0.0.tar.gz", hash = "sha256:4224373bacce55f955a878bf9cfa763c1e360858e330072059e10bad68531159"}, ] pre-commit = [ - {file = "pre_commit-2.15.0-py2.py3-none-any.whl", hash = "sha256:a4ed01000afcb484d9eb8d504272e642c4c4099bbad3a6b27e519bd6a3e928a6"}, - {file = "pre_commit-2.15.0.tar.gz", hash = "sha256:3c25add78dbdfb6a28a651780d5c311ac40dd17f160eb3954a0c59da40a505a7"}, + {file = "pre_commit-2.16.0-py2.py3-none-any.whl", hash = "sha256:758d1dc9b62c2ed8881585c254976d66eae0889919ab9b859064fc2fe3c7743e"}, + {file = "pre_commit-2.16.0.tar.gz", hash = "sha256:fe9897cac830aa7164dbd02a4e7b90cae49630451ce88464bca73db486ba9f65"}, ] psutil = [ {file = "psutil-5.8.0-cp27-cp27m-macosx_10_9_x86_64.whl", hash = "sha256:0066a82f7b1b37d334e68697faba68e5ad5e858279fd6351c8ca6024e8d6ba64"}, @@ -1346,20 +1529,20 @@ psutil = [ {file = "psutil-5.8.0.tar.gz", hash = "sha256:0c9ccb99ab76025f2f0bbecf341d4656e9c1351db8cc8a03ccd62e318ab4b5c6"}, ] py = [ - {file = "py-1.10.0-py2.py3-none-any.whl", hash = "sha256:3b80836aa6d1feeaa108e046da6423ab8f6ceda6468545ae8d02d9d58d18818a"}, - {file = "py-1.10.0.tar.gz", hash = "sha256:21b81bda15b66ef5e1a777a21c4dcd9c20ad3efd0b3f817e7a809035269e1bd3"}, + {file = "py-1.11.0-py2.py3-none-any.whl", hash = "sha256:607c53218732647dff4acdfcd50cb62615cedf612e72d1724fb1a0cc6405b378"}, + {file = "py-1.11.0.tar.gz", hash = "sha256:51c75c4126074b472f746a24399ad32f6053d1b34b68d2fa41e558e6f4a98719"}, ] pycparser = [ - {file = "pycparser-2.20-py2.py3-none-any.whl", hash = "sha256:7582ad22678f0fcd81102833f60ef8d0e57288b6b5fb00323d101be910e35705"}, - {file = "pycparser-2.20.tar.gz", hash = "sha256:2d475327684562c3a96cc71adf7dc8c4f0565175cf86b6d7a404ff4c771f15f0"}, + {file = "pycparser-2.21-py2.py3-none-any.whl", hash = "sha256:8ee45429555515e1f6b185e78100aea234072576aa43ab53aefcae078162fca9"}, + {file = "pycparser-2.21.tar.gz", hash = "sha256:e644fdec12f7872f86c58ff790da456218b10f863970249516d60a5eaca77206"}, ] pyopenssl = [ {file = "pyOpenSSL-19.0.0-py2.py3-none-any.whl", hash = "sha256:c727930ad54b10fc157015014b666f2d8b41f70c0d03e83ab67624fd3dd5d1e6"}, {file = "pyOpenSSL-19.0.0.tar.gz", hash = "sha256:aeca66338f6de19d1aa46ed634c3b9ae519a64b458f8468aec688e7e3c20f200"}, ] pyparsing = [ - {file = "pyparsing-2.4.7-py2.py3-none-any.whl", hash = "sha256:ef9d7589ef3c200abe66653d3f1ab1033c3c419ae9b9bdb1240a85b024efc88b"}, - {file = "pyparsing-2.4.7.tar.gz", hash = "sha256:c203ec8783bf771a155b207279b9bccb8dea02d8f0c9e5f8ead507bc3246ecc1"}, + {file = "pyparsing-3.0.6-py3-none-any.whl", hash = "sha256:04ff808a5b90911829c55c4e26f75fa5ca8a2f5f36aa3a51f68e27033341d3e4"}, + {file = "pyparsing-3.0.6.tar.gz", hash = "sha256:d9bdec0013ef1eb5a84ab39a3b3868911598afa494f5faa038647101504e2b81"}, ] pytest = [ {file = "pytest-6.2.5-py3-none-any.whl", hash = "sha256:7310f8d27bc79ced999e760ca304d69f6ba6c6649c0b60fb0e04a4a77cacc134"}, @@ -1374,35 +1557,39 @@ pytz = [ {file = "pytz-2021.3.tar.gz", hash = "sha256:acad2d8b20a1af07d4e4c9d2e9285c5ed9104354062f275f3fcd88dcef4f1326"}, ] pyyaml = [ - {file = "PyYAML-5.4.1-cp27-cp27m-macosx_10_9_x86_64.whl", hash = "sha256:3b2b1824fe7112845700f815ff6a489360226a5609b96ec2190a45e62a9fc922"}, - {file = "PyYAML-5.4.1-cp27-cp27m-win32.whl", hash = "sha256:129def1b7c1bf22faffd67b8f3724645203b79d8f4cc81f674654d9902cb4393"}, - {file = "PyYAML-5.4.1-cp27-cp27m-win_amd64.whl", hash = "sha256:4465124ef1b18d9ace298060f4eccc64b0850899ac4ac53294547536533800c8"}, - {file = "PyYAML-5.4.1-cp27-cp27mu-manylinux1_x86_64.whl", hash = "sha256:bb4191dfc9306777bc594117aee052446b3fa88737cd13b7188d0e7aa8162185"}, - {file = "PyYAML-5.4.1-cp36-cp36m-macosx_10_9_x86_64.whl", hash = "sha256:6c78645d400265a062508ae399b60b8c167bf003db364ecb26dcab2bda048253"}, - {file = "PyYAML-5.4.1-cp36-cp36m-manylinux1_x86_64.whl", hash = "sha256:4e0583d24c881e14342eaf4ec5fbc97f934b999a6828693a99157fde912540cc"}, - {file = "PyYAML-5.4.1-cp36-cp36m-manylinux2014_aarch64.whl", hash = "sha256:72a01f726a9c7851ca9bfad6fd09ca4e090a023c00945ea05ba1638c09dc3347"}, - {file = "PyYAML-5.4.1-cp36-cp36m-manylinux2014_s390x.whl", hash = "sha256:895f61ef02e8fed38159bb70f7e100e00f471eae2bc838cd0f4ebb21e28f8541"}, - {file = "PyYAML-5.4.1-cp36-cp36m-win32.whl", hash = "sha256:3bd0e463264cf257d1ffd2e40223b197271046d09dadf73a0fe82b9c1fc385a5"}, - {file = "PyYAML-5.4.1-cp36-cp36m-win_amd64.whl", hash = "sha256:e4fac90784481d221a8e4b1162afa7c47ed953be40d31ab4629ae917510051df"}, - {file = "PyYAML-5.4.1-cp37-cp37m-macosx_10_9_x86_64.whl", hash = "sha256:5accb17103e43963b80e6f837831f38d314a0495500067cb25afab2e8d7a4018"}, - {file = "PyYAML-5.4.1-cp37-cp37m-manylinux1_x86_64.whl", hash = "sha256:e1d4970ea66be07ae37a3c2e48b5ec63f7ba6804bdddfdbd3cfd954d25a82e63"}, - {file = "PyYAML-5.4.1-cp37-cp37m-manylinux2014_aarch64.whl", hash = "sha256:cb333c16912324fd5f769fff6bc5de372e9e7a202247b48870bc251ed40239aa"}, - {file = "PyYAML-5.4.1-cp37-cp37m-manylinux2014_s390x.whl", hash = "sha256:fe69978f3f768926cfa37b867e3843918e012cf83f680806599ddce33c2c68b0"}, - {file = "PyYAML-5.4.1-cp37-cp37m-win32.whl", hash = "sha256:dd5de0646207f053eb0d6c74ae45ba98c3395a571a2891858e87df7c9b9bd51b"}, - {file = "PyYAML-5.4.1-cp37-cp37m-win_amd64.whl", hash = "sha256:08682f6b72c722394747bddaf0aa62277e02557c0fd1c42cb853016a38f8dedf"}, - {file = "PyYAML-5.4.1-cp38-cp38-macosx_10_9_x86_64.whl", hash = "sha256:d2d9808ea7b4af864f35ea216be506ecec180628aced0704e34aca0b040ffe46"}, - {file = "PyYAML-5.4.1-cp38-cp38-manylinux1_x86_64.whl", hash = "sha256:8c1be557ee92a20f184922c7b6424e8ab6691788e6d86137c5d93c1a6ec1b8fb"}, - {file = "PyYAML-5.4.1-cp38-cp38-manylinux2014_aarch64.whl", hash = "sha256:fd7f6999a8070df521b6384004ef42833b9bd62cfee11a09bda1079b4b704247"}, - {file = "PyYAML-5.4.1-cp38-cp38-manylinux2014_s390x.whl", hash = "sha256:bfb51918d4ff3d77c1c856a9699f8492c612cde32fd3bcd344af9be34999bfdc"}, - {file = "PyYAML-5.4.1-cp38-cp38-win32.whl", hash = "sha256:fa5ae20527d8e831e8230cbffd9f8fe952815b2b7dae6ffec25318803a7528fc"}, - {file = "PyYAML-5.4.1-cp38-cp38-win_amd64.whl", hash = "sha256:0f5f5786c0e09baddcd8b4b45f20a7b5d61a7e7e99846e3c799b05c7c53fa696"}, - {file = "PyYAML-5.4.1-cp39-cp39-macosx_10_9_x86_64.whl", hash = "sha256:294db365efa064d00b8d1ef65d8ea2c3426ac366c0c4368d930bf1c5fb497f77"}, - {file = "PyYAML-5.4.1-cp39-cp39-manylinux1_x86_64.whl", hash = "sha256:74c1485f7707cf707a7aef42ef6322b8f97921bd89be2ab6317fd782c2d53183"}, - {file = "PyYAML-5.4.1-cp39-cp39-manylinux2014_aarch64.whl", hash = "sha256:d483ad4e639292c90170eb6f7783ad19490e7a8defb3e46f97dfe4bacae89122"}, - {file = "PyYAML-5.4.1-cp39-cp39-manylinux2014_s390x.whl", hash = "sha256:fdc842473cd33f45ff6bce46aea678a54e3d21f1b61a7750ce3c498eedfe25d6"}, - {file = "PyYAML-5.4.1-cp39-cp39-win32.whl", hash = "sha256:49d4cdd9065b9b6e206d0595fee27a96b5dd22618e7520c33204a4a3239d5b10"}, - {file = "PyYAML-5.4.1-cp39-cp39-win_amd64.whl", hash = "sha256:c20cfa2d49991c8b4147af39859b167664f2ad4561704ee74c1de03318e898db"}, - {file = "PyYAML-5.4.1.tar.gz", hash = "sha256:607774cbba28732bfa802b54baa7484215f530991055bb562efbed5b2f20a45e"}, + {file = "PyYAML-6.0-cp310-cp310-macosx_10_9_x86_64.whl", hash = "sha256:d4db7c7aef085872ef65a8fd7d6d09a14ae91f691dec3e87ee5ee0539d516f53"}, + {file = "PyYAML-6.0-cp310-cp310-macosx_11_0_arm64.whl", hash = "sha256:9df7ed3b3d2e0ecfe09e14741b857df43adb5a3ddadc919a2d94fbdf78fea53c"}, + {file = "PyYAML-6.0-cp310-cp310-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:77f396e6ef4c73fdc33a9157446466f1cff553d979bd00ecb64385760c6babdc"}, + {file = "PyYAML-6.0-cp310-cp310-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:a80a78046a72361de73f8f395f1f1e49f956c6be882eed58505a15f3e430962b"}, + {file = "PyYAML-6.0-cp310-cp310-manylinux_2_5_x86_64.manylinux1_x86_64.manylinux_2_12_x86_64.manylinux2010_x86_64.whl", hash = "sha256:f84fbc98b019fef2ee9a1cb3ce93e3187a6df0b2538a651bfb890254ba9f90b5"}, + {file = "PyYAML-6.0-cp310-cp310-win32.whl", hash = "sha256:2cd5df3de48857ed0544b34e2d40e9fac445930039f3cfe4bcc592a1f836d513"}, + {file = "PyYAML-6.0-cp310-cp310-win_amd64.whl", hash = "sha256:daf496c58a8c52083df09b80c860005194014c3698698d1a57cbcfa182142a3a"}, + {file = "PyYAML-6.0-cp36-cp36m-macosx_10_9_x86_64.whl", hash = "sha256:897b80890765f037df3403d22bab41627ca8811ae55e9a722fd0392850ec4d86"}, + {file = "PyYAML-6.0-cp36-cp36m-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:50602afada6d6cbfad699b0c7bb50d5ccffa7e46a3d738092afddc1f9758427f"}, + {file = "PyYAML-6.0-cp36-cp36m-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:48c346915c114f5fdb3ead70312bd042a953a8ce5c7106d5bfb1a5254e47da92"}, + {file = "PyYAML-6.0-cp36-cp36m-manylinux_2_5_x86_64.manylinux1_x86_64.manylinux_2_12_x86_64.manylinux2010_x86_64.whl", hash = "sha256:98c4d36e99714e55cfbaaee6dd5badbc9a1ec339ebfc3b1f52e293aee6bb71a4"}, + {file = "PyYAML-6.0-cp36-cp36m-win32.whl", hash = "sha256:0283c35a6a9fbf047493e3a0ce8d79ef5030852c51e9d911a27badfde0605293"}, + {file = "PyYAML-6.0-cp36-cp36m-win_amd64.whl", hash = "sha256:07751360502caac1c067a8132d150cf3d61339af5691fe9e87803040dbc5db57"}, + {file = "PyYAML-6.0-cp37-cp37m-macosx_10_9_x86_64.whl", hash = "sha256:819b3830a1543db06c4d4b865e70ded25be52a2e0631ccd2f6a47a2822f2fd7c"}, + {file = "PyYAML-6.0-cp37-cp37m-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:473f9edb243cb1935ab5a084eb238d842fb8f404ed2193a915d1784b5a6b5fc0"}, + {file = "PyYAML-6.0-cp37-cp37m-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:0ce82d761c532fe4ec3f87fc45688bdd3a4c1dc5e0b4a19814b9009a29baefd4"}, + {file = "PyYAML-6.0-cp37-cp37m-manylinux_2_5_x86_64.manylinux1_x86_64.manylinux_2_12_x86_64.manylinux2010_x86_64.whl", hash = "sha256:231710d57adfd809ef5d34183b8ed1eeae3f76459c18fb4a0b373ad56bedcdd9"}, + {file = "PyYAML-6.0-cp37-cp37m-win32.whl", hash = "sha256:c5687b8d43cf58545ade1fe3e055f70eac7a5a1a0bf42824308d868289a95737"}, + {file = "PyYAML-6.0-cp37-cp37m-win_amd64.whl", hash = "sha256:d15a181d1ecd0d4270dc32edb46f7cb7733c7c508857278d3d378d14d606db2d"}, + {file = "PyYAML-6.0-cp38-cp38-macosx_10_9_x86_64.whl", hash = "sha256:0b4624f379dab24d3725ffde76559cff63d9ec94e1736b556dacdfebe5ab6d4b"}, + {file = "PyYAML-6.0-cp38-cp38-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:213c60cd50106436cc818accf5baa1aba61c0189ff610f64f4a3e8c6726218ba"}, + {file = "PyYAML-6.0-cp38-cp38-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:9fa600030013c4de8165339db93d182b9431076eb98eb40ee068700c9c813e34"}, + {file = "PyYAML-6.0-cp38-cp38-manylinux_2_5_x86_64.manylinux1_x86_64.manylinux_2_12_x86_64.manylinux2010_x86_64.whl", hash = "sha256:277a0ef2981ca40581a47093e9e2d13b3f1fbbeffae064c1d21bfceba2030287"}, + {file = "PyYAML-6.0-cp38-cp38-win32.whl", hash = "sha256:d4eccecf9adf6fbcc6861a38015c2a64f38b9d94838ac1810a9023a0609e1b78"}, + {file = "PyYAML-6.0-cp38-cp38-win_amd64.whl", hash = "sha256:1e4747bc279b4f613a09eb64bba2ba602d8a6664c6ce6396a4d0cd413a50ce07"}, + {file = "PyYAML-6.0-cp39-cp39-macosx_10_9_x86_64.whl", hash = "sha256:055d937d65826939cb044fc8c9b08889e8c743fdc6a32b33e2390f66013e449b"}, + {file = "PyYAML-6.0-cp39-cp39-macosx_11_0_arm64.whl", hash = "sha256:e61ceaab6f49fb8bdfaa0f92c4b57bcfbea54c09277b1b4f7ac376bfb7a7c174"}, + {file = "PyYAML-6.0-cp39-cp39-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:d67d839ede4ed1b28a4e8909735fc992a923cdb84e618544973d7dfc71540803"}, + {file = "PyYAML-6.0-cp39-cp39-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:cba8c411ef271aa037d7357a2bc8f9ee8b58b9965831d9e51baf703280dc73d3"}, + {file = "PyYAML-6.0-cp39-cp39-manylinux_2_5_x86_64.manylinux1_x86_64.manylinux_2_12_x86_64.manylinux2010_x86_64.whl", hash = "sha256:40527857252b61eacd1d9af500c3337ba8deb8fc298940291486c465c8b46ec0"}, + {file = "PyYAML-6.0-cp39-cp39-win32.whl", hash = "sha256:b5b9eccad747aabaaffbc6064800670f0c297e52c12754eb1d976c57e4f74dcb"}, + {file = "PyYAML-6.0-cp39-cp39-win_amd64.whl", hash = "sha256:b3d267842bf12586ba6c734f89d1f5b871df0273157918b0ccefa29deb05c21c"}, + {file = "PyYAML-6.0.tar.gz", hash = "sha256:68fb519c14306fec9720a2a5b45bc9f0c8d1b9c72adf45c37baedfcd949c35a2"}, ] pyzmq = [ {file = "pyzmq-22.3.0-cp310-cp310-macosx_10_15_universal2.whl", hash = "sha256:6b217b8f9dfb6628f74b94bdaf9f7408708cb02167d644edca33f38746ca12dd"}, @@ -1444,42 +1631,80 @@ pyzmq = [ {file = "pyzmq-22.3.0.tar.gz", hash = "sha256:8eddc033e716f8c91c6a2112f0a8ebc5e00532b4a6ae1eb0ccc48e027f9c671c"}, ] regex = [ - {file = "regex-2021.10.21-cp310-cp310-macosx_10_9_x86_64.whl", hash = "sha256:edff4e31d159672a7b9d70164b21289e4b53b239ce1dc945bf9643d266537573"}, - {file = "regex-2021.10.21-cp310-cp310-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:d6432daf42f2c487b357e1aa0bdc43193f050ff53a3188bfab20b88202b53027"}, - {file = "regex-2021.10.21-cp310-cp310-manylinux_2_5_i686.manylinux1_i686.manylinux_2_12_i686.manylinux2010_i686.whl", hash = "sha256:468de52dd3f20187ab5ca4fd265c1bea61a5346baef01ad0333a5e89fa9fad29"}, - {file = "regex-2021.10.21-cp310-cp310-manylinux_2_5_i686.manylinux1_i686.manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:c5a2ac760f2fc13a1c58131ec217779911890899ce1a0a63c9409bd23fecde6f"}, - {file = "regex-2021.10.21-cp310-cp310-manylinux_2_5_x86_64.manylinux1_x86_64.manylinux_2_12_x86_64.manylinux2010_x86_64.whl", hash = "sha256:ad1fedca001fefc3030d1e9022b038af429e58dc06a7e9c55e40bd1f834582ec"}, - {file = "regex-2021.10.21-cp310-cp310-win32.whl", hash = "sha256:9c613d797a3790f6b12e78a61e1cd29df7fc88135218467cf8b0891353292b9c"}, - {file = "regex-2021.10.21-cp310-cp310-win_amd64.whl", hash = "sha256:678d9a4ce79e1eaa4ebe88bc9769df52919eb30c597576a0deba1f3cf2360e65"}, - {file = "regex-2021.10.21-cp36-cp36m-macosx_10_9_x86_64.whl", hash = "sha256:2044174af237bb9c56ecc07294cf38623ee379e8dca14b01e970f8b015c71917"}, - {file = "regex-2021.10.21-cp36-cp36m-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:98743a2d827a135bf3390452be18d95839b947a099734d53c17e09a64fc09480"}, - {file = "regex-2021.10.21-cp36-cp36m-manylinux_2_5_i686.manylinux1_i686.manylinux_2_12_i686.manylinux2010_i686.whl", hash = "sha256:f1b23304855303bd97b5954edab63b8ddd56c91c41c6d4eba408228c0bae95f3"}, - {file = "regex-2021.10.21-cp36-cp36m-manylinux_2_5_i686.manylinux1_i686.manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:19c4fd59747236423016ccd89b9a6485d958bf1aa7a8a902a6ba28029107a87f"}, - {file = "regex-2021.10.21-cp36-cp36m-manylinux_2_5_x86_64.manylinux1_x86_64.manylinux_2_12_x86_64.manylinux2010_x86_64.whl", hash = "sha256:130a002fa386c976615a2f6d6dff0fcc25da24858994a36b14d2e3129dce7de2"}, - {file = "regex-2021.10.21-cp36-cp36m-win32.whl", hash = "sha256:8bd83d9b8ee125350cd666b55294f4bc9993c4f0d9b1be9344a318d0762e94cc"}, - {file = "regex-2021.10.21-cp36-cp36m-win_amd64.whl", hash = "sha256:98fe0e1b07a314f0a86dc58af4e717c379d48a403eddd8d966ab9b8bf91ce164"}, - {file = "regex-2021.10.21-cp37-cp37m-macosx_10_9_x86_64.whl", hash = "sha256:ded4748c7be6f31fb207387ee83a3a0f625e700defe32f268cb1d350ed6e4a66"}, - {file = "regex-2021.10.21-cp37-cp37m-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:f3da121de36a9ead0f32b44ea720ee8c87edbb59dca6bb980d18377d84ad58a3"}, - {file = "regex-2021.10.21-cp37-cp37m-manylinux_2_5_i686.manylinux1_i686.manylinux_2_12_i686.manylinux2010_i686.whl", hash = "sha256:b9dfba513eae785e3d868803f5a7e21a032cb2b038fa4a1ea7ec691037426ad3"}, - {file = "regex-2021.10.21-cp37-cp37m-manylinux_2_5_i686.manylinux1_i686.manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:2ff91696888755e96230138355cbe8ce2965d930d967d6cff7c636082d038c78"}, - {file = "regex-2021.10.21-cp37-cp37m-manylinux_2_5_x86_64.manylinux1_x86_64.manylinux_2_12_x86_64.manylinux2010_x86_64.whl", hash = "sha256:0f82de529d7595011a40573cc0f27422e59cafa94943b64a4d17d966d75f2c01"}, - {file = "regex-2021.10.21-cp37-cp37m-win32.whl", hash = "sha256:164e51ace4d00f07c519f85ec2209e8faaeab18bc77be6b35685c18d4ac1c22a"}, - {file = "regex-2021.10.21-cp37-cp37m-win_amd64.whl", hash = "sha256:e39eafa854e469d7225066c806c76b9a0acba5ff5ce36c82c0224b75e24888f2"}, - {file = "regex-2021.10.21-cp38-cp38-macosx_10_9_x86_64.whl", hash = "sha256:740a28580520b099b804776db1e919360fcbf30a734a14c5985d5e39a39e7237"}, - {file = "regex-2021.10.21-cp38-cp38-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:de7dbf72ae80f06e79444ff9614fb5e3a7956645d513b0e12d1bbe6f3ccebd11"}, - {file = "regex-2021.10.21-cp38-cp38-manylinux_2_5_i686.manylinux1_i686.manylinux_2_12_i686.manylinux2010_i686.whl", hash = "sha256:dc1a9bedf389bf3d3627a4d2b21cbdc5fe5e0f029d1f465972f4437833dcc946"}, - {file = "regex-2021.10.21-cp38-cp38-manylinux_2_5_i686.manylinux1_i686.manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:9cd14f22425beecf727f6dbdf5c893e46ecbc5ff16197c16a6f38a9066f2d4d5"}, - {file = "regex-2021.10.21-cp38-cp38-manylinux_2_5_x86_64.manylinux1_x86_64.manylinux_2_12_x86_64.manylinux2010_x86_64.whl", hash = "sha256:5b75a3db3aab0bfa51b6af3f820760779d360eb79f59e32c88c7fba648990b4f"}, - {file = "regex-2021.10.21-cp38-cp38-win32.whl", hash = "sha256:f68c71aabb10b1352a06515e25a425a703ba85660ae04cf074da5eb91c0af5e5"}, - {file = "regex-2021.10.21-cp38-cp38-win_amd64.whl", hash = "sha256:c0f49f1f03be3e4a5faaadc35db7afa2b83a871943b889f9f7bba56e0e2e8bd5"}, - {file = "regex-2021.10.21-cp39-cp39-macosx_10_9_x86_64.whl", hash = "sha256:201890fdc8a65396cfb6aa4493201353b2a6378e27d2de65234446f8329233cb"}, - {file = "regex-2021.10.21-cp39-cp39-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:fd1bfc6b7347de9f0ae1fb6f9080426bed6a9ca55b5766fa4fdf7b3a29ccae9c"}, - {file = "regex-2021.10.21-cp39-cp39-manylinux_2_5_i686.manylinux1_i686.manylinux_2_12_i686.manylinux2010_i686.whl", hash = "sha256:72a0b98d41c4508ed23a96eef41090f78630b44ba746e28cd621ecbe961e0a16"}, - {file = "regex-2021.10.21-cp39-cp39-manylinux_2_5_i686.manylinux1_i686.manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:3b5a0660a63b0703380758a7141b96cc1c1a13dee2b8e9c280a2522962fd12af"}, - {file = "regex-2021.10.21-cp39-cp39-manylinux_2_5_x86_64.manylinux1_x86_64.manylinux_2_12_x86_64.manylinux2010_x86_64.whl", hash = "sha256:f82d3adde46ac9188db3aa7e6e1690865ebb6448d245df5a3ea22284f70d9e46"}, - {file = "regex-2021.10.21-cp39-cp39-win32.whl", hash = "sha256:bc4637390235f1e3e2fcdd3e904ca0b42aa655ae28a78072248b2992b4ad4c08"}, - {file = "regex-2021.10.21-cp39-cp39-win_amd64.whl", hash = "sha256:74d03c256cf0aed81997e87be8e24297b5792c9718f3a735f5055ddfad392f06"}, - {file = "regex-2021.10.21.tar.gz", hash = "sha256:4832736b3f24617e63dc919ce8c4215680ba94250a5d9e710fcc0c5f457b5028"}, + {file = "regex-2021.11.10-cp310-cp310-macosx_10_9_x86_64.whl", hash = "sha256:9345b6f7ee578bad8e475129ed40123d265464c4cfead6c261fd60fc9de00bcf"}, + {file = "regex-2021.11.10-cp310-cp310-macosx_11_0_arm64.whl", hash = "sha256:416c5f1a188c91e3eb41e9c8787288e707f7d2ebe66e0a6563af280d9b68478f"}, + {file = "regex-2021.11.10-cp310-cp310-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:e0538c43565ee6e703d3a7c3bdfe4037a5209250e8502c98f20fea6f5fdf2965"}, + {file = "regex-2021.11.10-cp310-cp310-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:7ee1227cf08b6716c85504aebc49ac827eb88fcc6e51564f010f11a406c0a667"}, + {file = "regex-2021.11.10-cp310-cp310-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:6650f16365f1924d6014d2ea770bde8555b4a39dc9576abb95e3cd1ff0263b36"}, + {file = "regex-2021.11.10-cp310-cp310-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:30ab804ea73972049b7a2a5c62d97687d69b5a60a67adca07eb73a0ddbc9e29f"}, + {file = "regex-2021.11.10-cp310-cp310-manylinux_2_5_i686.manylinux1_i686.manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:68a067c11463de2a37157930d8b153005085e42bcb7ad9ca562d77ba7d1404e0"}, + {file = "regex-2021.11.10-cp310-cp310-manylinux_2_5_x86_64.manylinux1_x86_64.manylinux_2_12_x86_64.manylinux2010_x86_64.whl", hash = "sha256:162abfd74e88001d20cb73ceaffbfe601469923e875caf9118333b1a4aaafdc4"}, + {file = "regex-2021.11.10-cp310-cp310-musllinux_1_1_aarch64.whl", hash = "sha256:b9ed0b1e5e0759d6b7f8e2f143894b2a7f3edd313f38cf44e1e15d360e11749b"}, + {file = "regex-2021.11.10-cp310-cp310-musllinux_1_1_i686.whl", hash = "sha256:473e67837f786404570eae33c3b64a4b9635ae9f00145250851a1292f484c063"}, + {file = "regex-2021.11.10-cp310-cp310-musllinux_1_1_ppc64le.whl", hash = "sha256:2fee3ed82a011184807d2127f1733b4f6b2ff6ec7151d83ef3477f3b96a13d03"}, + {file = "regex-2021.11.10-cp310-cp310-musllinux_1_1_s390x.whl", hash = "sha256:d5fd67df77bab0d3f4ea1d7afca9ef15c2ee35dfb348c7b57ffb9782a6e4db6e"}, + {file = "regex-2021.11.10-cp310-cp310-musllinux_1_1_x86_64.whl", hash = "sha256:5d408a642a5484b9b4d11dea15a489ea0928c7e410c7525cd892f4d04f2f617b"}, + {file = "regex-2021.11.10-cp310-cp310-win32.whl", hash = "sha256:98ba568e8ae26beb726aeea2273053c717641933836568c2a0278a84987b2a1a"}, + {file = "regex-2021.11.10-cp310-cp310-win_amd64.whl", hash = "sha256:780b48456a0f0ba4d390e8b5f7c661fdd218934388cde1a974010a965e200e12"}, + {file = "regex-2021.11.10-cp36-cp36m-macosx_10_9_x86_64.whl", hash = "sha256:dba70f30fd81f8ce6d32ddeef37d91c8948e5d5a4c63242d16a2b2df8143aafc"}, + {file = "regex-2021.11.10-cp36-cp36m-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:e1f54b9b4b6c53369f40028d2dd07a8c374583417ee6ec0ea304e710a20f80a0"}, + {file = "regex-2021.11.10-cp36-cp36m-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:fbb9dc00e39f3e6c0ef48edee202f9520dafb233e8b51b06b8428cfcb92abd30"}, + {file = "regex-2021.11.10-cp36-cp36m-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:666abff54e474d28ff42756d94544cdfd42e2ee97065857413b72e8a2d6a6345"}, + {file = "regex-2021.11.10-cp36-cp36m-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:5537f71b6d646f7f5f340562ec4c77b6e1c915f8baae822ea0b7e46c1f09b733"}, + {file = "regex-2021.11.10-cp36-cp36m-manylinux_2_5_i686.manylinux1_i686.manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:ed2e07c6a26ed4bea91b897ee2b0835c21716d9a469a96c3e878dc5f8c55bb23"}, + {file = "regex-2021.11.10-cp36-cp36m-manylinux_2_5_x86_64.manylinux1_x86_64.manylinux_2_12_x86_64.manylinux2010_x86_64.whl", hash = "sha256:ca5f18a75e1256ce07494e245cdb146f5a9267d3c702ebf9b65c7f8bd843431e"}, + {file = "regex-2021.11.10-cp36-cp36m-musllinux_1_1_aarch64.whl", hash = "sha256:74cbeac0451f27d4f50e6e8a8f3a52ca074b5e2da9f7b505c4201a57a8ed6286"}, + {file = "regex-2021.11.10-cp36-cp36m-musllinux_1_1_i686.whl", hash = "sha256:3598893bde43091ee5ca0a6ad20f08a0435e93a69255eeb5f81b85e81e329264"}, + {file = "regex-2021.11.10-cp36-cp36m-musllinux_1_1_ppc64le.whl", hash = "sha256:50a7ddf3d131dc5633dccdb51417e2d1910d25cbcf842115a3a5893509140a3a"}, + {file = "regex-2021.11.10-cp36-cp36m-musllinux_1_1_s390x.whl", hash = "sha256:61600a7ca4bcf78a96a68a27c2ae9389763b5b94b63943d5158f2a377e09d29a"}, + {file = "regex-2021.11.10-cp36-cp36m-musllinux_1_1_x86_64.whl", hash = "sha256:563d5f9354e15e048465061509403f68424fef37d5add3064038c2511c8f5e00"}, + {file = "regex-2021.11.10-cp36-cp36m-win32.whl", hash = "sha256:93a5051fcf5fad72de73b96f07d30bc29665697fb8ecdfbc474f3452c78adcf4"}, + {file = "regex-2021.11.10-cp36-cp36m-win_amd64.whl", hash = "sha256:b483c9d00a565633c87abd0aaf27eb5016de23fed952e054ecc19ce32f6a9e7e"}, + {file = "regex-2021.11.10-cp37-cp37m-macosx_10_9_x86_64.whl", hash = "sha256:fff55f3ce50a3ff63ec8e2a8d3dd924f1941b250b0aac3d3d42b687eeff07a8e"}, + {file = "regex-2021.11.10-cp37-cp37m-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:e32d2a2b02ccbef10145df9135751abea1f9f076e67a4e261b05f24b94219e36"}, + {file = "regex-2021.11.10-cp37-cp37m-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:53db2c6be8a2710b359bfd3d3aa17ba38f8aa72a82309a12ae99d3c0c3dcd74d"}, + {file = "regex-2021.11.10-cp37-cp37m-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:2207ae4f64ad3af399e2d30dde66f0b36ae5c3129b52885f1bffc2f05ec505c8"}, + {file = "regex-2021.11.10-cp37-cp37m-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:d5ca078bb666c4a9d1287a379fe617a6dccd18c3e8a7e6c7e1eb8974330c626a"}, + {file = "regex-2021.11.10-cp37-cp37m-manylinux_2_5_i686.manylinux1_i686.manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:dd33eb9bdcfbabab3459c9ee651d94c842bc8a05fabc95edf4ee0c15a072495e"}, + {file = "regex-2021.11.10-cp37-cp37m-manylinux_2_5_x86_64.manylinux1_x86_64.manylinux_2_12_x86_64.manylinux2010_x86_64.whl", hash = "sha256:05b7d6d7e64efe309972adab77fc2af8907bb93217ec60aa9fe12a0dad35874f"}, + {file = "regex-2021.11.10-cp37-cp37m-musllinux_1_1_aarch64.whl", hash = "sha256:42b50fa6666b0d50c30a990527127334d6b96dd969011e843e726a64011485da"}, + {file = "regex-2021.11.10-cp37-cp37m-musllinux_1_1_i686.whl", hash = "sha256:6e1d2cc79e8dae442b3fa4a26c5794428b98f81389af90623ffcc650ce9f6732"}, + {file = "regex-2021.11.10-cp37-cp37m-musllinux_1_1_ppc64le.whl", hash = "sha256:0416f7399e918c4b0e074a0f66e5191077ee2ca32a0f99d4c187a62beb47aa05"}, + {file = "regex-2021.11.10-cp37-cp37m-musllinux_1_1_s390x.whl", hash = "sha256:ce298e3d0c65bd03fa65ffcc6db0e2b578e8f626d468db64fdf8457731052942"}, + {file = "regex-2021.11.10-cp37-cp37m-musllinux_1_1_x86_64.whl", hash = "sha256:dc07f021ee80510f3cd3af2cad5b6a3b3a10b057521d9e6aaeb621730d320c5a"}, + {file = "regex-2021.11.10-cp37-cp37m-win32.whl", hash = "sha256:e71255ba42567d34a13c03968736c5d39bb4a97ce98188fafb27ce981115beec"}, + {file = "regex-2021.11.10-cp37-cp37m-win_amd64.whl", hash = "sha256:07856afef5ffcc052e7eccf3213317fbb94e4a5cd8177a2caa69c980657b3cb4"}, + {file = "regex-2021.11.10-cp38-cp38-macosx_10_9_x86_64.whl", hash = "sha256:ba05430e819e58544e840a68b03b28b6d328aff2e41579037e8bab7653b37d83"}, + {file = "regex-2021.11.10-cp38-cp38-macosx_11_0_arm64.whl", hash = "sha256:7f301b11b9d214f83ddaf689181051e7f48905568b0c7017c04c06dfd065e244"}, + {file = "regex-2021.11.10-cp38-cp38-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:4aaa4e0705ef2b73dd8e36eeb4c868f80f8393f5f4d855e94025ce7ad8525f50"}, + {file = "regex-2021.11.10-cp38-cp38-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:788aef3549f1924d5c38263104dae7395bf020a42776d5ec5ea2b0d3d85d6646"}, + {file = "regex-2021.11.10-cp38-cp38-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:f8af619e3be812a2059b212064ea7a640aff0568d972cd1b9e920837469eb3cb"}, + {file = "regex-2021.11.10-cp38-cp38-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:85bfa6a5413be0ee6c5c4a663668a2cad2cbecdee367630d097d7823041bdeec"}, + {file = "regex-2021.11.10-cp38-cp38-manylinux_2_5_i686.manylinux1_i686.manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:f23222527b307970e383433daec128d769ff778d9b29343fb3496472dc20dabe"}, + {file = "regex-2021.11.10-cp38-cp38-manylinux_2_5_x86_64.manylinux1_x86_64.manylinux_2_12_x86_64.manylinux2010_x86_64.whl", hash = "sha256:da1a90c1ddb7531b1d5ff1e171b4ee61f6345119be7351104b67ff413843fe94"}, + {file = "regex-2021.11.10-cp38-cp38-musllinux_1_1_aarch64.whl", hash = "sha256:f5be7805e53dafe94d295399cfbe5227f39995a997f4fd8539bf3cbdc8f47ca8"}, + {file = "regex-2021.11.10-cp38-cp38-musllinux_1_1_i686.whl", hash = "sha256:a955b747d620a50408b7fdf948e04359d6e762ff8a85f5775d907ceced715129"}, + {file = "regex-2021.11.10-cp38-cp38-musllinux_1_1_ppc64le.whl", hash = "sha256:139a23d1f5d30db2cc6c7fd9c6d6497872a672db22c4ae1910be22d4f4b2068a"}, + {file = "regex-2021.11.10-cp38-cp38-musllinux_1_1_s390x.whl", hash = "sha256:ca49e1ab99593438b204e00f3970e7a5f70d045267051dfa6b5f4304fcfa1dbf"}, + {file = "regex-2021.11.10-cp38-cp38-musllinux_1_1_x86_64.whl", hash = "sha256:96fc32c16ea6d60d3ca7f63397bff5c75c5a562f7db6dec7d412f7c4d2e78ec0"}, + {file = "regex-2021.11.10-cp38-cp38-win32.whl", hash = "sha256:0617383e2fe465732af4509e61648b77cbe3aee68b6ac8c0b6fe934db90be5cc"}, + {file = "regex-2021.11.10-cp38-cp38-win_amd64.whl", hash = "sha256:a3feefd5e95871872673b08636f96b61ebef62971eab044f5124fb4dea39919d"}, + {file = "regex-2021.11.10-cp39-cp39-macosx_10_9_x86_64.whl", hash = "sha256:f7f325be2804246a75a4f45c72d4ce80d2443ab815063cdf70ee8fb2ca59ee1b"}, + {file = "regex-2021.11.10-cp39-cp39-macosx_11_0_arm64.whl", hash = "sha256:537ca6a3586931b16a85ac38c08cc48f10fc870a5b25e51794c74df843e9966d"}, + {file = "regex-2021.11.10-cp39-cp39-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:eef2afb0fd1747f33f1ee3e209bce1ed582d1896b240ccc5e2697e3275f037c7"}, + {file = "regex-2021.11.10-cp39-cp39-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:432bd15d40ed835a51617521d60d0125867f7b88acf653e4ed994a1f8e4995dc"}, + {file = "regex-2021.11.10-cp39-cp39-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:b43c2b8a330a490daaef5a47ab114935002b13b3f9dc5da56d5322ff218eeadb"}, + {file = "regex-2021.11.10-cp39-cp39-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:962b9a917dd7ceacbe5cd424556914cb0d636001e393b43dc886ba31d2a1e449"}, + {file = "regex-2021.11.10-cp39-cp39-manylinux_2_5_i686.manylinux1_i686.manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:fa8c626d6441e2d04b6ee703ef2d1e17608ad44c7cb75258c09dd42bacdfc64b"}, + {file = "regex-2021.11.10-cp39-cp39-manylinux_2_5_x86_64.manylinux1_x86_64.manylinux_2_12_x86_64.manylinux2010_x86_64.whl", hash = "sha256:3c5fb32cc6077abad3bbf0323067636d93307c9fa93e072771cf9a64d1c0f3ef"}, + {file = "regex-2021.11.10-cp39-cp39-musllinux_1_1_aarch64.whl", hash = "sha256:cd410a1cbb2d297c67d8521759ab2ee3f1d66206d2e4328502a487589a2cb21b"}, + {file = "regex-2021.11.10-cp39-cp39-musllinux_1_1_i686.whl", hash = "sha256:e6096b0688e6e14af6a1b10eaad86b4ff17935c49aa774eac7c95a57a4e8c296"}, + {file = "regex-2021.11.10-cp39-cp39-musllinux_1_1_ppc64le.whl", hash = "sha256:529801a0d58809b60b3531ee804d3e3be4b412c94b5d267daa3de7fadef00f49"}, + {file = "regex-2021.11.10-cp39-cp39-musllinux_1_1_s390x.whl", hash = "sha256:0f594b96fe2e0821d026365f72ac7b4f0b487487fb3d4aaf10dd9d97d88a9737"}, + {file = "regex-2021.11.10-cp39-cp39-musllinux_1_1_x86_64.whl", hash = "sha256:2409b5c9cef7054dde93a9803156b411b677affc84fca69e908b1cb2c540025d"}, + {file = "regex-2021.11.10-cp39-cp39-win32.whl", hash = "sha256:3b5df18db1fccd66de15aa59c41e4f853b5df7550723d26aa6cb7f40e5d9da5a"}, + {file = "regex-2021.11.10-cp39-cp39-win_amd64.whl", hash = "sha256:83ee89483672b11f8952b158640d0c0ff02dc43d9cb1b70c1564b49abe92ce29"}, + {file = "regex-2021.11.10.tar.gz", hash = "sha256:f341ee2df0999bfdf7a95e448075effe0db212a59387de1a70690e4acb03d4c6"}, ] requests = [ {file = "requests-2.26.0-py2.py3-none-any.whl", hash = "sha256:6c1246513ecd5ecd4528a0906f910e8f0f9c6b8ec72030dc9fd154dc1a6efd24"}, @@ -1498,8 +1723,8 @@ toml = [ {file = "toml-0.10.2.tar.gz", hash = "sha256:b3bda1d108d5dd99f4a20d24d9c348e91c4db7ab1b749200bded2f839ccbe68f"}, ] tomli = [ - {file = "tomli-1.2.1-py3-none-any.whl", hash = "sha256:8dd0e9524d6f386271a36b41dbf6c57d8e32fd96fd22b6584679dc569d20899f"}, - {file = "tomli-1.2.1.tar.gz", hash = "sha256:a5b75cb6f3968abb47af1b40c1819dc519ea82bcc065776a866e8d74c5ca9442"}, + {file = "tomli-1.2.2-py3-none-any.whl", hash = "sha256:f04066f68f5554911363063a30b108d2b5a5b1a010aa8b6132af78489fe3aade"}, + {file = "tomli-1.2.2.tar.gz", hash = "sha256:c6ce0015eb38820eaf32b5db832dbc26deb3dd427bd5f6556cf0acac2c214fee"}, ] typed-ast = [ {file = "typed_ast-1.4.3-cp35-cp35m-manylinux1_i686.whl", hash = "sha256:2068531575a125b87a41802130fa7e29f26c09a2833fea68d9a40cf33902eba6"}, @@ -1534,9 +1759,8 @@ typed-ast = [ {file = "typed_ast-1.4.3.tar.gz", hash = "sha256:fb1bbeac803adea29cedd70781399c99138358c26d05fcbd23c13016b7f5ec65"}, ] typing-extensions = [ - {file = "typing_extensions-3.10.0.2-py2-none-any.whl", hash = "sha256:d8226d10bc02a29bcc81df19a26e56a9647f8b0a6d4a83924139f4a8b01f17b7"}, - {file = "typing_extensions-3.10.0.2-py3-none-any.whl", hash = "sha256:f1d25edafde516b146ecd0613dabcc61409817af4766fbbcfb8d1ad4ec441a34"}, - {file = "typing_extensions-3.10.0.2.tar.gz", hash = "sha256:49f75d16ff11f1cd258e1b988ccff82a3ca5570217d7ad8c5f48205dd99a677e"}, + {file = "typing_extensions-4.0.1-py3-none-any.whl", hash = "sha256:7f001e5ac290a0c0401508864c7ec868be4e701886d5b573a9528ed3973d9d3b"}, + {file = "typing_extensions-4.0.1.tar.gz", hash = "sha256:4ca091dea149f945ec56afb48dae714f21e8692ef22a395223bcd328961b6a0e"}, ] tzlocal = [ {file = "tzlocal-2.1-py2.py3-none-any.whl", hash = "sha256:e2cb6c6b5b604af38597403e9852872d7f534962ae2954c7f35efcb1ccacf4a4"}, @@ -1547,16 +1771,16 @@ urllib3 = [ {file = "urllib3-1.26.7.tar.gz", hash = "sha256:4987c65554f7a2dbf30c18fd48778ef124af6fab771a377103da0585e2336ece"}, ] virtualenv = [ - {file = "virtualenv-20.8.1-py2.py3-none-any.whl", hash = "sha256:10062e34c204b5e4ec5f62e6ef2473f8ba76513a9a617e873f1f8fb4a519d300"}, - {file = "virtualenv-20.8.1.tar.gz", hash = "sha256:bcc17f0b3a29670dd777d6f0755a4c04f28815395bca279cdcb213b97199a6b8"}, + {file = "virtualenv-20.10.0-py2.py3-none-any.whl", hash = "sha256:4b02e52a624336eece99c96e3ab7111f469c24ba226a53ec474e8e787b365814"}, + {file = "virtualenv-20.10.0.tar.gz", hash = "sha256:576d05b46eace16a9c348085f7d0dc8ef28713a2cabaa1cf0aea41e8f12c9218"}, ] volttron-client = [ - {file = "volttron-client-0.3.4.tar.gz", hash = "sha256:a426efd8c5d10c5f574b78501082c3ac76714903c29d4fe9c8c9d335abc03d47"}, - {file = "volttron_client-0.3.4-py3-none-any.whl", hash = "sha256:60ba299b75431e5d249ff0cf871e1978ab75e65814a28b6252a5f7612f1ec658"}, + {file = "volttron-client-0.3.8.tar.gz", hash = "sha256:e8ec4953f27f28de851a2ebdd55c89eb5df3163ad594c1c8eec3b4ae035632e9"}, + {file = "volttron_client-0.3.8-py3-none-any.whl", hash = "sha256:75c2fef11b5032b664e47105d1d9f73e67ddab28d8c7febf45cba5109e695a1b"}, ] volttron-utils = [ - {file = "volttron-utils-0.3.2.tar.gz", hash = "sha256:abb35d335ba60ecc06e9e1bbcbb5dd0b3dbc9445e58a366dfaa6071e4ce5ce80"}, - {file = "volttron_utils-0.3.2-py3-none-any.whl", hash = "sha256:240e6f84ef164d7740b0c035b01efbd00c194c1b5b12bb3b73fe28386fff6f5f"}, + {file = "volttron-utils-0.3.4.tar.gz", hash = "sha256:0d499f5ab228d669e5c99b39987af3e0fbba9ad46a4cbb9bc6c5d8e320c28ca2"}, + {file = "volttron_utils-0.3.4-py3-none-any.whl", hash = "sha256:e565dc2535a9cd1dd24006ad61c35627e06d701e923f3ac75f8ed2d8959a715c"}, ] watchdog = [ {file = "watchdog-2.1.6-cp310-cp310-macosx_10_9_universal2.whl", hash = "sha256:9693f35162dc6208d10b10ddf0458cc09ad70c30ba689d9206e02cd836ce28a3"}, @@ -1592,78 +1816,78 @@ xmltodict = [ {file = "xmltodict-0.12.0.tar.gz", hash = "sha256:50d8c638ed7ecb88d90561beedbf720c9b4e851a9fa6c47ebd64e99d166d8a21"}, ] yarl = [ - {file = "yarl-1.7.0-cp310-cp310-macosx_10_9_universal2.whl", hash = "sha256:e35d8230e4b08d86ea65c32450533b906a8267a87b873f2954adeaecede85169"}, - {file = "yarl-1.7.0-cp310-cp310-macosx_10_9_x86_64.whl", hash = "sha256:eb4b3f277880c314e47720b4b6bb2c85114ab3c04c5442c9bc7006b3787904d8"}, - {file = "yarl-1.7.0-cp310-cp310-macosx_11_0_arm64.whl", hash = "sha256:c7015dcedb91d90a138eebdc7e432aec8966e0147ab2a55f2df27b1904fa7291"}, - {file = "yarl-1.7.0-cp310-cp310-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:bb3e478175e15e00d659fb0354a6a8db71a7811a2a5052aed98048bc972e5d2b"}, - {file = "yarl-1.7.0-cp310-cp310-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:8b8c409aa3a7966647e7c1c524846b362a6bcbbe120bf8a176431f940d2b9a2e"}, - {file = "yarl-1.7.0-cp310-cp310-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:b22ea41c7e98170474a01e3eded1377d46b2dfaef45888a0005c683eaaa49285"}, - {file = "yarl-1.7.0-cp310-cp310-manylinux_2_5_i686.manylinux1_i686.manylinux_2_12_i686.manylinux2010_i686.whl", hash = "sha256:a7dfc46add4cfe5578013dbc4127893edc69fe19132d2836ff2f6e49edc5ecd6"}, - {file = "yarl-1.7.0-cp310-cp310-manylinux_2_5_x86_64.manylinux1_x86_64.manylinux_2_12_x86_64.manylinux2010_x86_64.whl", hash = "sha256:82ff6f85f67500a4f74885d81659cd270eb24dfe692fe44e622b8a2fd57e7279"}, - {file = "yarl-1.7.0-cp310-cp310-musllinux_1_1_aarch64.whl", hash = "sha256:f3cd2158b2ed0fb25c6811adfdcc47224efe075f2d68a750071dacc03a7a66e4"}, - {file = "yarl-1.7.0-cp310-cp310-musllinux_1_1_i686.whl", hash = "sha256:59c0f13f9592820c51280d1cf811294d753e4a18baf90f0139d1dc93d4b6fc5f"}, - {file = "yarl-1.7.0-cp310-cp310-musllinux_1_1_ppc64le.whl", hash = "sha256:7f7655ad83d1a8afa48435a449bf2f3009293da1604f5dd95b5ddcf5f673bd69"}, - {file = "yarl-1.7.0-cp310-cp310-musllinux_1_1_s390x.whl", hash = "sha256:aa9f0d9b62d15182341b3e9816582f46182cab91c1a57b2d308b9a3c4e2c4f78"}, - {file = "yarl-1.7.0-cp310-cp310-musllinux_1_1_x86_64.whl", hash = "sha256:fdd1b90c225a653b1bd1c0cae8edf1957892b9a09c8bf7ee6321eeb8208eac0f"}, - {file = "yarl-1.7.0-cp310-cp310-win32.whl", hash = "sha256:7c8d0bb76eabc5299db203e952ec55f8f4c53f08e0df4285aac8c92bd9e12675"}, - {file = "yarl-1.7.0-cp310-cp310-win_amd64.whl", hash = "sha256:622a36fa779efb4ff9eff5fe52730ff17521431379851a31e040958fc251670c"}, - {file = "yarl-1.7.0-cp36-cp36m-macosx_10_9_x86_64.whl", hash = "sha256:3d461b7a8e139b9e4b41f62eb417ffa0b98d1c46d4caf14c845e6a3b349c0bb1"}, - {file = "yarl-1.7.0-cp36-cp36m-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:81cfacdd1e40bc931b5519499342efa388d24d262c30a3d31187bfa04f4a7001"}, - {file = "yarl-1.7.0-cp36-cp36m-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:821b978f2152be7695d4331ef0621d207aedf9bbd591ba23a63412a3efc29a01"}, - {file = "yarl-1.7.0-cp36-cp36m-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:b64bd24c8c9a487f4a12260dc26732bf41028816dbf0c458f17864fbebdb3131"}, - {file = "yarl-1.7.0-cp36-cp36m-manylinux_2_5_i686.manylinux1_i686.manylinux_2_12_i686.manylinux2010_i686.whl", hash = "sha256:98c9ddb92b60a83c21be42c776d3d9d5ec632a762a094c41bda37b7dfbd2cd83"}, - {file = "yarl-1.7.0-cp36-cp36m-manylinux_2_5_x86_64.manylinux1_x86_64.manylinux_2_12_x86_64.manylinux2010_x86_64.whl", hash = "sha256:a532d75ca74431c053a88a802e161fb3d651b8bf5821a3440bc3616e38754583"}, - {file = "yarl-1.7.0-cp36-cp36m-musllinux_1_1_aarch64.whl", hash = "sha256:053e09817eafb892e94e172d05406c1b3a22a93bc68f6eff5198363a3d764459"}, - {file = "yarl-1.7.0-cp36-cp36m-musllinux_1_1_i686.whl", hash = "sha256:98c51f02d542945d306c8e934aa2c1e66ba5e9c1c86b5bf37f3a51c8a747067e"}, - {file = "yarl-1.7.0-cp36-cp36m-musllinux_1_1_ppc64le.whl", hash = "sha256:15ec41a5a5fdb7bace6d7b16701f9440007a82734f69127c0fbf6d87e10f4a1e"}, - {file = "yarl-1.7.0-cp36-cp36m-musllinux_1_1_s390x.whl", hash = "sha256:a7f08819dba1e1255d6991ed37448a1bf4b1352c004bcd899b9da0c47958513d"}, - {file = "yarl-1.7.0-cp36-cp36m-musllinux_1_1_x86_64.whl", hash = "sha256:8e3ffab21db0542ffd1887f3b9575ddd58961f2cf61429cb6458afc00c4581e0"}, - {file = "yarl-1.7.0-cp36-cp36m-win32.whl", hash = "sha256:50127634f519b2956005891507e3aa4ac345f66a7ea7bbc2d7dcba7401f41898"}, - {file = "yarl-1.7.0-cp36-cp36m-win_amd64.whl", hash = "sha256:36ec44f15193f6d5288d42ebb8e751b967ebdfb72d6830983838d45ab18edb4f"}, - {file = "yarl-1.7.0-cp37-cp37m-macosx_10_9_x86_64.whl", hash = "sha256:ec1b5a25a25c880c976d0bb3d107def085bb08dbb3db7f4442e0a2b980359d24"}, - {file = "yarl-1.7.0-cp37-cp37m-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:b36f5a63c891f813c6f04ef19675b382efc190fd5ce7e10ab19386d2548bca06"}, - {file = "yarl-1.7.0-cp37-cp37m-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:38173b8c3a29945e7ecade9a3f6ff39581eee8201338ee6a2c8882db5df3e806"}, - {file = "yarl-1.7.0-cp37-cp37m-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:8ba402f32184f0b405fb281b93bd0d8ab7e3257735b57b62a6ed2e94cdf4fe50"}, - {file = "yarl-1.7.0-cp37-cp37m-manylinux_2_5_i686.manylinux1_i686.manylinux_2_12_i686.manylinux2010_i686.whl", hash = "sha256:be52bc5208d767cdd8308a9e93059b3b36d1e048fecbea0e0346d0d24a76adc0"}, - {file = "yarl-1.7.0-cp37-cp37m-manylinux_2_5_x86_64.manylinux1_x86_64.manylinux_2_12_x86_64.manylinux2010_x86_64.whl", hash = "sha256:08c2044a956f4ef30405f2f433ce77f1f57c2c773bf81ae43201917831044d5a"}, - {file = "yarl-1.7.0-cp37-cp37m-musllinux_1_1_aarch64.whl", hash = "sha256:484d61c047c45670ef5967653a1d0783e232c54bf9dd786a7737036828fa8d54"}, - {file = "yarl-1.7.0-cp37-cp37m-musllinux_1_1_i686.whl", hash = "sha256:b7de92a4af85cfcaf4081f8aa6165b1d63ee5de150af3ee85f954145f93105a7"}, - {file = "yarl-1.7.0-cp37-cp37m-musllinux_1_1_ppc64le.whl", hash = "sha256:376e41775aab79c5575534924a386c8e0f1a5d91db69fc6133fd27a489bcaf10"}, - {file = "yarl-1.7.0-cp37-cp37m-musllinux_1_1_s390x.whl", hash = "sha256:8a8b10d0e7bac154f959b709fcea593cda527b234119311eb950096653816a86"}, - {file = "yarl-1.7.0-cp37-cp37m-musllinux_1_1_x86_64.whl", hash = "sha256:f46cd4c43e6175030e2a56def8f1d83b64e6706eeb2bb9ab0ef4756f65eab23f"}, - {file = "yarl-1.7.0-cp37-cp37m-win32.whl", hash = "sha256:b28cfb46140efe1a6092b8c5c4994a1fe70dc83c38fbcea4992401e0c6fb9cce"}, - {file = "yarl-1.7.0-cp37-cp37m-win_amd64.whl", hash = "sha256:9624154ec9c02a776802da1086eed7f5034bd1971977f5146233869c2ac80297"}, - {file = "yarl-1.7.0-cp38-cp38-macosx_10_9_universal2.whl", hash = "sha256:69945d13e1bbf81784a9bc48824feb9cd66491e6a503d4e83f6cd7c7cc861361"}, - {file = "yarl-1.7.0-cp38-cp38-macosx_10_9_x86_64.whl", hash = "sha256:46a742ed9e363bd01be64160ce7520e92e11989bd4cb224403cfd31c101cc83d"}, - {file = "yarl-1.7.0-cp38-cp38-macosx_11_0_arm64.whl", hash = "sha256:cb4ff1ac7cb4500f43581b3f4cbd627d702143aa6be1fdc1fa3ebffaf4dc1be5"}, - {file = "yarl-1.7.0-cp38-cp38-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:3ad51e17cd65ea3debb0e10f0120cf8dd987c741fe423ed2285087368090b33d"}, - {file = "yarl-1.7.0-cp38-cp38-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:7e37786ea89a5d3ffbbf318ea9790926f8dfda83858544f128553c347ad143c6"}, - {file = "yarl-1.7.0-cp38-cp38-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:c63c1e208f800daad71715786bfeb1cecdc595d87e2e9b1cd234fd6e597fd71d"}, - {file = "yarl-1.7.0-cp38-cp38-manylinux_2_5_i686.manylinux1_i686.manylinux_2_12_i686.manylinux2010_i686.whl", hash = "sha256:91cbe24300c11835ef186436363352b3257db7af165e0a767f4f17aa25761388"}, - {file = "yarl-1.7.0-cp38-cp38-manylinux_2_5_x86_64.manylinux1_x86_64.manylinux_2_12_x86_64.manylinux2010_x86_64.whl", hash = "sha256:e510dbec7c59d32eaa61ffa48173d5e3d7170a67f4a03e8f5e2e9e3971aca622"}, - {file = "yarl-1.7.0-cp38-cp38-musllinux_1_1_aarch64.whl", hash = "sha256:3def6e681cc02397e5d8141ee97b41d02932b2bcf0fb34532ad62855eab7c60e"}, - {file = "yarl-1.7.0-cp38-cp38-musllinux_1_1_i686.whl", hash = "sha256:263c81b94e6431942b27f6f671fa62f430a0a5c14bb255f2ab69eeb9b2b66ff7"}, - {file = "yarl-1.7.0-cp38-cp38-musllinux_1_1_ppc64le.whl", hash = "sha256:e78c91faefe88d601ddd16e3882918dbde20577a2438e2320f8239c8b7507b8f"}, - {file = "yarl-1.7.0-cp38-cp38-musllinux_1_1_s390x.whl", hash = "sha256:22b2430c49713bfb2f0a0dd4a8d7aab218b28476ba86fd1c78ad8899462cbcf2"}, - {file = "yarl-1.7.0-cp38-cp38-musllinux_1_1_x86_64.whl", hash = "sha256:2e7ad9db939082f5d0b9269cfd92c025cb8f2fbbb1f1b9dc5a393c639db5bd92"}, - {file = "yarl-1.7.0-cp38-cp38-win32.whl", hash = "sha256:3a31e4a8dcb1beaf167b7e7af61b88cb961b220db8d3ba1c839723630e57eef7"}, - {file = "yarl-1.7.0-cp38-cp38-win_amd64.whl", hash = "sha256:d579957439933d752358c6a300c93110f84aae67b63dd0c19dde6ecbf4056f6b"}, - {file = "yarl-1.7.0-cp39-cp39-macosx_10_9_universal2.whl", hash = "sha256:87721b549505a546eb003252185103b5ec8147de6d3ad3714d148a5a67b6fe53"}, - {file = "yarl-1.7.0-cp39-cp39-macosx_10_9_x86_64.whl", hash = "sha256:a1fa866fa24d9f4108f9e58ea8a2135655419885cdb443e36b39a346e1181532"}, - {file = "yarl-1.7.0-cp39-cp39-macosx_11_0_arm64.whl", hash = "sha256:1d3b8449dfedfe94eaff2b77954258b09b24949f6818dfa444b05dbb05ae1b7e"}, - {file = "yarl-1.7.0-cp39-cp39-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:db2372e350794ce8b9f810feb094c606b7e0e4aa6807141ac4fadfe5ddd75bb0"}, - {file = "yarl-1.7.0-cp39-cp39-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:a06d9d0b9a97fa99b84fee71d9dd11e69e21ac8a27229089f07b5e5e50e8d63c"}, - {file = "yarl-1.7.0-cp39-cp39-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:a3455c2456d6307bcfa80bc1157b8603f7d93573291f5bdc7144489ca0df4628"}, - {file = "yarl-1.7.0-cp39-cp39-manylinux_2_5_i686.manylinux1_i686.manylinux_2_12_i686.manylinux2010_i686.whl", hash = "sha256:d30d67e3486aea61bb2cbf7cf81385364c2e4f7ce7469a76ed72af76a5cdfe6b"}, - {file = "yarl-1.7.0-cp39-cp39-manylinux_2_5_x86_64.manylinux1_x86_64.manylinux_2_12_x86_64.manylinux2010_x86_64.whl", hash = "sha256:c18a4b286e8d780c3a40c31d7b79836aa93b720f71d5743f20c08b7e049ca073"}, - {file = "yarl-1.7.0-cp39-cp39-musllinux_1_1_aarch64.whl", hash = "sha256:d54c925396e7891666cabc0199366ca55b27d003393465acef63fd29b8b7aa92"}, - {file = "yarl-1.7.0-cp39-cp39-musllinux_1_1_i686.whl", hash = "sha256:64773840952de17851a1c7346ad7f71688c77e74248d1f0bc230e96680f84028"}, - {file = "yarl-1.7.0-cp39-cp39-musllinux_1_1_ppc64le.whl", hash = "sha256:acbf1756d9dc7cd0ae943d883be72e84e04396f6c2ff93a6ddeca929d562039f"}, - {file = "yarl-1.7.0-cp39-cp39-musllinux_1_1_s390x.whl", hash = "sha256:2e48f27936aa838939c798f466c851ba4ae79e347e8dfce43b009c64b930df12"}, - {file = "yarl-1.7.0-cp39-cp39-musllinux_1_1_x86_64.whl", hash = "sha256:1beef4734ca1ad40a9d8c6b20a76ab46e3a2ed09f38561f01e4aa2ea82cafcef"}, - {file = "yarl-1.7.0-cp39-cp39-win32.whl", hash = "sha256:8ee78c9a5f3c642219d4607680a4693b59239c27a3aa608b64ef79ddc9698039"}, - {file = "yarl-1.7.0-cp39-cp39-win_amd64.whl", hash = "sha256:d750503682605088a14d29a4701548c15c510da4f13c8b17409c4097d5b04c52"}, - {file = "yarl-1.7.0.tar.gz", hash = "sha256:8e7ebaf62e19c2feb097ffb7c94deb0f0c9fab52590784c8cd679d30ab009162"}, + {file = "yarl-1.7.2-cp310-cp310-macosx_10_9_universal2.whl", hash = "sha256:f2a8508f7350512434e41065684076f640ecce176d262a7d54f0da41d99c5a95"}, + {file = "yarl-1.7.2-cp310-cp310-macosx_10_9_x86_64.whl", hash = "sha256:da6df107b9ccfe52d3a48165e48d72db0eca3e3029b5b8cb4fe6ee3cb870ba8b"}, + {file = "yarl-1.7.2-cp310-cp310-macosx_11_0_arm64.whl", hash = "sha256:a1d0894f238763717bdcfea74558c94e3bc34aeacd3351d769460c1a586a8b05"}, + {file = "yarl-1.7.2-cp310-cp310-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:dfe4b95b7e00c6635a72e2d00b478e8a28bfb122dc76349a06e20792eb53a523"}, + {file = "yarl-1.7.2-cp310-cp310-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:c145ab54702334c42237a6c6c4cc08703b6aa9b94e2f227ceb3d477d20c36c63"}, + {file = "yarl-1.7.2-cp310-cp310-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:1ca56f002eaf7998b5fcf73b2421790da9d2586331805f38acd9997743114e98"}, + {file = "yarl-1.7.2-cp310-cp310-manylinux_2_5_i686.manylinux1_i686.manylinux_2_12_i686.manylinux2010_i686.whl", hash = "sha256:1d3d5ad8ea96bd6d643d80c7b8d5977b4e2fb1bab6c9da7322616fd26203d125"}, + {file = "yarl-1.7.2-cp310-cp310-manylinux_2_5_x86_64.manylinux1_x86_64.manylinux_2_12_x86_64.manylinux2010_x86_64.whl", hash = "sha256:167ab7f64e409e9bdd99333fe8c67b5574a1f0495dcfd905bc7454e766729b9e"}, + {file = "yarl-1.7.2-cp310-cp310-musllinux_1_1_aarch64.whl", hash = "sha256:95a1873b6c0dd1c437fb3bb4a4aaa699a48c218ac7ca1e74b0bee0ab16c7d60d"}, + {file = "yarl-1.7.2-cp310-cp310-musllinux_1_1_i686.whl", hash = "sha256:6152224d0a1eb254f97df3997d79dadd8bb2c1a02ef283dbb34b97d4f8492d23"}, + {file = "yarl-1.7.2-cp310-cp310-musllinux_1_1_ppc64le.whl", hash = "sha256:5bb7d54b8f61ba6eee541fba4b83d22b8a046b4ef4d8eb7f15a7e35db2e1e245"}, + {file = "yarl-1.7.2-cp310-cp310-musllinux_1_1_s390x.whl", hash = "sha256:9c1f083e7e71b2dd01f7cd7434a5f88c15213194df38bc29b388ccdf1492b739"}, + {file = "yarl-1.7.2-cp310-cp310-musllinux_1_1_x86_64.whl", hash = "sha256:f44477ae29025d8ea87ec308539f95963ffdc31a82f42ca9deecf2d505242e72"}, + {file = "yarl-1.7.2-cp310-cp310-win32.whl", hash = "sha256:cff3ba513db55cc6a35076f32c4cdc27032bd075c9faef31fec749e64b45d26c"}, + {file = "yarl-1.7.2-cp310-cp310-win_amd64.whl", hash = "sha256:c9c6d927e098c2d360695f2e9d38870b2e92e0919be07dbe339aefa32a090265"}, + {file = "yarl-1.7.2-cp36-cp36m-macosx_10_9_x86_64.whl", hash = "sha256:9b4c77d92d56a4c5027572752aa35082e40c561eec776048330d2907aead891d"}, + {file = "yarl-1.7.2-cp36-cp36m-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:c01a89a44bb672c38f42b49cdb0ad667b116d731b3f4c896f72302ff77d71656"}, + {file = "yarl-1.7.2-cp36-cp36m-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:c19324a1c5399b602f3b6e7db9478e5b1adf5cf58901996fc973fe4fccd73eed"}, + {file = "yarl-1.7.2-cp36-cp36m-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:3abddf0b8e41445426d29f955b24aeecc83fa1072be1be4e0d194134a7d9baee"}, + {file = "yarl-1.7.2-cp36-cp36m-manylinux_2_5_i686.manylinux1_i686.manylinux_2_12_i686.manylinux2010_i686.whl", hash = "sha256:6a1a9fe17621af43e9b9fcea8bd088ba682c8192d744b386ee3c47b56eaabb2c"}, + {file = "yarl-1.7.2-cp36-cp36m-manylinux_2_5_x86_64.manylinux1_x86_64.manylinux_2_12_x86_64.manylinux2010_x86_64.whl", hash = "sha256:8b0915ee85150963a9504c10de4e4729ae700af11df0dc5550e6587ed7891e92"}, + {file = "yarl-1.7.2-cp36-cp36m-musllinux_1_1_aarch64.whl", hash = "sha256:29e0656d5497733dcddc21797da5a2ab990c0cb9719f1f969e58a4abac66234d"}, + {file = "yarl-1.7.2-cp36-cp36m-musllinux_1_1_i686.whl", hash = "sha256:bf19725fec28452474d9887a128e98dd67eee7b7d52e932e6949c532d820dc3b"}, + {file = "yarl-1.7.2-cp36-cp36m-musllinux_1_1_ppc64le.whl", hash = "sha256:d6f3d62e16c10e88d2168ba2d065aa374e3c538998ed04996cd373ff2036d64c"}, + {file = "yarl-1.7.2-cp36-cp36m-musllinux_1_1_s390x.whl", hash = "sha256:ac10bbac36cd89eac19f4e51c032ba6b412b3892b685076f4acd2de18ca990aa"}, + {file = "yarl-1.7.2-cp36-cp36m-musllinux_1_1_x86_64.whl", hash = "sha256:aa32aaa97d8b2ed4e54dc65d241a0da1c627454950f7d7b1f95b13985afd6c5d"}, + {file = "yarl-1.7.2-cp36-cp36m-win32.whl", hash = "sha256:87f6e082bce21464857ba58b569370e7b547d239ca22248be68ea5d6b51464a1"}, + {file = "yarl-1.7.2-cp36-cp36m-win_amd64.whl", hash = "sha256:ac35ccde589ab6a1870a484ed136d49a26bcd06b6a1c6397b1967ca13ceb3913"}, + {file = "yarl-1.7.2-cp37-cp37m-macosx_10_9_x86_64.whl", hash = "sha256:a467a431a0817a292121c13cbe637348b546e6ef47ca14a790aa2fa8cc93df63"}, + {file = "yarl-1.7.2-cp37-cp37m-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:6ab0c3274d0a846840bf6c27d2c60ba771a12e4d7586bf550eefc2df0b56b3b4"}, + {file = "yarl-1.7.2-cp37-cp37m-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:d260d4dc495c05d6600264a197d9d6f7fc9347f21d2594926202fd08cf89a8ba"}, + {file = "yarl-1.7.2-cp37-cp37m-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:fc4dd8b01a8112809e6b636b00f487846956402834a7fd59d46d4f4267181c41"}, + {file = "yarl-1.7.2-cp37-cp37m-manylinux_2_5_i686.manylinux1_i686.manylinux_2_12_i686.manylinux2010_i686.whl", hash = "sha256:c1164a2eac148d85bbdd23e07dfcc930f2e633220f3eb3c3e2a25f6148c2819e"}, + {file = "yarl-1.7.2-cp37-cp37m-manylinux_2_5_x86_64.manylinux1_x86_64.manylinux_2_12_x86_64.manylinux2010_x86_64.whl", hash = "sha256:67e94028817defe5e705079b10a8438b8cb56e7115fa01640e9c0bb3edf67332"}, + {file = "yarl-1.7.2-cp37-cp37m-musllinux_1_1_aarch64.whl", hash = "sha256:89ccbf58e6a0ab89d487c92a490cb5660d06c3a47ca08872859672f9c511fc52"}, + {file = "yarl-1.7.2-cp37-cp37m-musllinux_1_1_i686.whl", hash = "sha256:8cce6f9fa3df25f55521fbb5c7e4a736683148bcc0c75b21863789e5185f9185"}, + {file = "yarl-1.7.2-cp37-cp37m-musllinux_1_1_ppc64le.whl", hash = "sha256:211fcd65c58bf250fb994b53bc45a442ddc9f441f6fec53e65de8cba48ded986"}, + {file = "yarl-1.7.2-cp37-cp37m-musllinux_1_1_s390x.whl", hash = "sha256:c10ea1e80a697cf7d80d1ed414b5cb8f1eec07d618f54637067ae3c0334133c4"}, + {file = "yarl-1.7.2-cp37-cp37m-musllinux_1_1_x86_64.whl", hash = "sha256:52690eb521d690ab041c3919666bea13ab9fbff80d615ec16fa81a297131276b"}, + {file = "yarl-1.7.2-cp37-cp37m-win32.whl", hash = "sha256:695ba021a9e04418507fa930d5f0704edbce47076bdcfeeaba1c83683e5649d1"}, + {file = "yarl-1.7.2-cp37-cp37m-win_amd64.whl", hash = "sha256:c17965ff3706beedafd458c452bf15bac693ecd146a60a06a214614dc097a271"}, + {file = "yarl-1.7.2-cp38-cp38-macosx_10_9_universal2.whl", hash = "sha256:fce78593346c014d0d986b7ebc80d782b7f5e19843ca798ed62f8e3ba8728576"}, + {file = "yarl-1.7.2-cp38-cp38-macosx_10_9_x86_64.whl", hash = "sha256:c2a1ac41a6aa980db03d098a5531f13985edcb451bcd9d00670b03129922cd0d"}, + {file = "yarl-1.7.2-cp38-cp38-macosx_11_0_arm64.whl", hash = "sha256:39d5493c5ecd75c8093fa7700a2fb5c94fe28c839c8e40144b7ab7ccba6938c8"}, + {file = "yarl-1.7.2-cp38-cp38-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:1eb6480ef366d75b54c68164094a6a560c247370a68c02dddb11f20c4c6d3c9d"}, + {file = "yarl-1.7.2-cp38-cp38-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:5ba63585a89c9885f18331a55d25fe81dc2d82b71311ff8bd378fc8004202ff6"}, + {file = "yarl-1.7.2-cp38-cp38-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:e39378894ee6ae9f555ae2de332d513a5763276a9265f8e7cbaeb1b1ee74623a"}, + {file = "yarl-1.7.2-cp38-cp38-manylinux_2_5_i686.manylinux1_i686.manylinux_2_12_i686.manylinux2010_i686.whl", hash = "sha256:c0910c6b6c31359d2f6184828888c983d54d09d581a4a23547a35f1d0b9484b1"}, + {file = "yarl-1.7.2-cp38-cp38-manylinux_2_5_x86_64.manylinux1_x86_64.manylinux_2_12_x86_64.manylinux2010_x86_64.whl", hash = "sha256:6feca8b6bfb9eef6ee057628e71e1734caf520a907b6ec0d62839e8293e945c0"}, + {file = "yarl-1.7.2-cp38-cp38-musllinux_1_1_aarch64.whl", hash = "sha256:8300401dc88cad23f5b4e4c1226f44a5aa696436a4026e456fe0e5d2f7f486e6"}, + {file = "yarl-1.7.2-cp38-cp38-musllinux_1_1_i686.whl", hash = "sha256:788713c2896f426a4e166b11f4ec538b5736294ebf7d5f654ae445fd44270832"}, + {file = "yarl-1.7.2-cp38-cp38-musllinux_1_1_ppc64le.whl", hash = "sha256:fd547ec596d90c8676e369dd8a581a21227fe9b4ad37d0dc7feb4ccf544c2d59"}, + {file = "yarl-1.7.2-cp38-cp38-musllinux_1_1_s390x.whl", hash = "sha256:737e401cd0c493f7e3dd4db72aca11cfe069531c9761b8ea474926936b3c57c8"}, + {file = "yarl-1.7.2-cp38-cp38-musllinux_1_1_x86_64.whl", hash = "sha256:baf81561f2972fb895e7844882898bda1eef4b07b5b385bcd308d2098f1a767b"}, + {file = "yarl-1.7.2-cp38-cp38-win32.whl", hash = "sha256:ede3b46cdb719c794427dcce9d8beb4abe8b9aa1e97526cc20de9bd6583ad1ef"}, + {file = "yarl-1.7.2-cp38-cp38-win_amd64.whl", hash = "sha256:cc8b7a7254c0fc3187d43d6cb54b5032d2365efd1df0cd1749c0c4df5f0ad45f"}, + {file = "yarl-1.7.2-cp39-cp39-macosx_10_9_universal2.whl", hash = "sha256:580c1f15500e137a8c37053e4cbf6058944d4c114701fa59944607505c2fe3a0"}, + {file = "yarl-1.7.2-cp39-cp39-macosx_10_9_x86_64.whl", hash = "sha256:3ec1d9a0d7780416e657f1e405ba35ec1ba453a4f1511eb8b9fbab81cb8b3ce1"}, + {file = "yarl-1.7.2-cp39-cp39-macosx_11_0_arm64.whl", hash = "sha256:3bf8cfe8856708ede6a73907bf0501f2dc4e104085e070a41f5d88e7faf237f3"}, + {file = "yarl-1.7.2-cp39-cp39-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:1be4bbb3d27a4e9aa5f3df2ab61e3701ce8fcbd3e9846dbce7c033a7e8136746"}, + {file = "yarl-1.7.2-cp39-cp39-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:534b047277a9a19d858cde163aba93f3e1677d5acd92f7d10ace419d478540de"}, + {file = "yarl-1.7.2-cp39-cp39-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:c6ddcd80d79c96eb19c354d9dca95291589c5954099836b7c8d29278a7ec0bda"}, + {file = "yarl-1.7.2-cp39-cp39-manylinux_2_5_i686.manylinux1_i686.manylinux_2_12_i686.manylinux2010_i686.whl", hash = "sha256:9bfcd43c65fbb339dc7086b5315750efa42a34eefad0256ba114cd8ad3896f4b"}, + {file = "yarl-1.7.2-cp39-cp39-manylinux_2_5_x86_64.manylinux1_x86_64.manylinux_2_12_x86_64.manylinux2010_x86_64.whl", hash = "sha256:f64394bd7ceef1237cc604b5a89bf748c95982a84bcd3c4bbeb40f685c810794"}, + {file = "yarl-1.7.2-cp39-cp39-musllinux_1_1_aarch64.whl", hash = "sha256:044daf3012e43d4b3538562da94a88fb12a6490652dbc29fb19adfa02cf72eac"}, + {file = "yarl-1.7.2-cp39-cp39-musllinux_1_1_i686.whl", hash = "sha256:368bcf400247318382cc150aaa632582d0780b28ee6053cd80268c7e72796dec"}, + {file = "yarl-1.7.2-cp39-cp39-musllinux_1_1_ppc64le.whl", hash = "sha256:bab827163113177aee910adb1f48ff7af31ee0289f434f7e22d10baf624a6dfe"}, + {file = "yarl-1.7.2-cp39-cp39-musllinux_1_1_s390x.whl", hash = "sha256:0cba38120db72123db7c58322fa69e3c0efa933040ffb586c3a87c063ec7cae8"}, + {file = "yarl-1.7.2-cp39-cp39-musllinux_1_1_x86_64.whl", hash = "sha256:59218fef177296451b23214c91ea3aba7858b4ae3306dde120224cfe0f7a6ee8"}, + {file = "yarl-1.7.2-cp39-cp39-win32.whl", hash = "sha256:1edc172dcca3f11b38a9d5c7505c83c1913c0addc99cd28e993efeaafdfaa18d"}, + {file = "yarl-1.7.2-cp39-cp39-win_amd64.whl", hash = "sha256:797c2c412b04403d2da075fb93c123df35239cd7b4cc4e0cd9e5839b73f52c58"}, + {file = "yarl-1.7.2.tar.gz", hash = "sha256:45399b46d60c253327a460e99856752009fcee5f5d3c80b2f7c0cae1c38d56dd"}, ] zmq = [ {file = "zmq-0.0.0.tar.gz", hash = "sha256:6b1a1de53338646e8c8405803cffb659e8eb7bb02fff4c9be62a7acfac8370c9"}, diff --git a/pyproject.toml b/pyproject.toml index 3f39fdf..64002f8 100644 --- a/pyproject.toml +++ b/pyproject.toml @@ -10,9 +10,8 @@ black = "^21.5b2" mypy = "^0.812" pre-commit = "^2.13.0" volttron-client = "^0.3.4" +openleadr = "^0.5.24" -# local dependencies -openleadr= {path="./openleadr", develop=true} [tool.poetry.dev-dependencies] pytest = "^6.2.4" diff --git a/requirements.txt b/requirements.txt new file mode 100644 index 0000000..27803e4 --- /dev/null +++ b/requirements.txt @@ -0,0 +1,1 @@ +openleadr==0.5.24 diff --git a/scripts/__init__.py b/scripts/__init__.py new file mode 100644 index 0000000..e69de29 diff --git a/scripts/run_ipkeys_openadr_client.py b/scripts/run_ipkeys_openadr_client.py new file mode 100644 index 0000000..a365235 --- /dev/null +++ b/scripts/run_ipkeys_openadr_client.py @@ -0,0 +1,51 @@ +import logging +import asyncio + +from pprint import pformat +from openleadr.enums import OPT +from volttron_openadr_ven.volttron_openadr_client import IPKeysClient +from volttron_openadr_ven.constants import ( + VEN_NAME, + VTN_URL, + DEBUG, + DISABLE_SIGNATURE, + CERT, + KEY, +) + +logging.basicConfig(level=logging.INFO) +_log = logging.getLogger(__name__) +__version__ = "1.0" + + +def handle_event(event): + _log.info(f"Received event:\n {pformat(event)}") + return OPT.OPT_IN + + +config = { + # Change the paths to the location of the cert and key to the paths of the cert and key in your local machine + CERT: "secret/TEST_RSA_VEN_210923215148_certs/TEST_RSA_VEN_210923215148_cert.pem", + KEY: "secret/TEST_RSA_VEN_210923215148_certs/TEST_RSA_VEN_210923215148_privkey.pem", + # Do not change these values + VEN_NAME: "PNNLVEN", + VTN_URL: "https://eiss2demo.ipkeys.com/oadr2/OpenADR2/Simple/2.0b", + DEBUG: True, + DISABLE_SIGNATURE: True, +} + + +ven = IPKeysClient( + ven_name=config[VEN_NAME], + vtn_url=config[VTN_URL], + debug=True, + disable_signature=True, + cert=config[CERT], + key=config[KEY], +) +ven.add_handler("on_event", handle_event) + + +loop = asyncio.get_event_loop() +loop.create_task(ven.run()) +loop.run_forever() diff --git a/setup.cfg b/setup.cfg new file mode 100644 index 0000000..4930738 --- /dev/null +++ b/setup.cfg @@ -0,0 +1,43 @@ +[metadata] +name = volttron-openadr-ven +description = OpenAdr VEN Implementation +description-file = README.md +description-content-type = text/markdown +author = The VOLTTRON Team +author-email = [email protected] +license = Apache 2 +keywords = volttron, iot platform, building automation, ADR, openadr, Automated Demand Response +project_urls = + Bug Reports = https://github.com/VOLTTRON/volttron-openadr-ven/issues + Source = https://github.com/VOLTTRON/volttron-openadr-ven/ + +# Classifiers help users find your project by categorizing it. +# +# For a list of valid classifiers, see https://pypi.org/classifiers/ +classifiers= + # How mature is this project? Common values are + # 3 - Alpha + # 4 - Beta + # 5 - Production/Stable + Development Status :: 3 - Alpha + # Indicate who your project is intended for + Intended Audience :: Developers + Intended Audience :: Science/Research + # Pick your license as you wish + License :: OSI Approved :: Apache Software License + # Specify the Python versions you support here. In particular, ensure + # that you indicate whether you support Python 2, Python 3 or both. + # These classifiers are not* checked by 'pip install'. See instead + # 'python_requires' below. + Programming Language :: Python :: 3.7 + Programming Language :: Python :: 3.8 + Programming Language :: Python :: 3.9 + +[files] +packages = + volttron_openadr_ven + +;[entry_points] +;console_scripts = +; volttron = volttron.server.__main__:main + diff --git a/setup.py b/setup.py new file mode 100644 index 0000000..8e2caba --- /dev/null +++ b/setup.py @@ -0,0 +1,43 @@ +# Always prefer setuptools over distutils +from setuptools import setup, find_packages + +setup( + setup_requires=["pbr"], + pbr=True, + # Specify which Python versions you support. In contrast to the + # 'Programming Language' classifiers above, 'pip install' will check this + # and refuse to install the project if the version does not match. If you + # do not support Python 2, you can simplify this to '>=3.5' or similar, see + # https://packaging.python.org/guides/distributing-packages-using-setuptools/#python-requires + python_requires=">=3.7, <4", + + # For an analysis of "install_requires" vs pip's requirements files see: + # https://packaging.python.org/en/latest/requirements.html + install_requires=[], # Optional + # List additional groups of dependencies here (e.g. development + # dependencies). Users will be able to install these using the "extras" + # syntax, for example: + # + # $ pip install sampleproject[dev] + # + # Similar to `install_requires` above, these must be valid existing + # projects. + extras_require={"dev": []}, # Optional + # If there are data files included in your packages that need to be + # installed, specify them here. + # + # Sometimes you’ll want to use packages that are properly arranged with + # setuptools, but are not published to PyPI. In those cases, you can specify + # a list of one or more dependency_links URLs where the package can + # be downloaded, along with some additional hints, and setuptools + # will find and install the package correctly. + # see https://python-packaging.readthedocs.io/en/latest/dependencies.html#packages-not-on-pypi + # + dependency_links=[], + entry_points={ + 'setuptools.installation': [ + 'eggsecutable = volttron_openadr_ven.agent:main', + ] + } + +) \ No newline at end of file diff --git a/volttron_openadr_ven/__init__.py b/volttron_openadr_ven/__init__.py new file mode 100644 index 0000000..f7557d5 --- /dev/null +++ b/volttron_openadr_ven/__init__.py @@ -0,0 +1,82 @@ +# -- coding: utf-8 -- {{{ +# vim: set fenc=utf-8 ft=python sw=4 ts=4 sts=4 et: +# +# Copyright 2020, Battelle Memorial Institute. +# +# Licensed under the Apache License, Version 2.0 (the "License"); +# you may not use this file except in compliance with the License. +# You may obtain a copy of the License at +# +# http://www.apache.org/licenses/LICENSE-2.0 +# +# Unless required by applicable law or agreed to in writing, software +# distributed under the License is distributed on an "AS IS" BASIS, +# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. +# See the License for the specific language governing permissions and +# limitations under the License. +# +# This material was prepared as an account of work sponsored by an agency of +# the United States Government. Neither the United States Government nor the +# United States Department of Energy, nor Battelle, nor any of their +# employees, nor any jurisdiction or organization that has cooperated in the +# development of these materials, makes any warranty, express or +# implied, or assumes any legal liability or responsibility for the accuracy, +# completeness, or usefulness or any information, apparatus, product, +# software, or process disclosed, or represents that its use would not infringe +# privately owned rights. Reference herein to any specific commercial product, +# process, or service by trade name, trademark, manufacturer, or otherwise +# does not necessarily constitute or imply its endorsement, recommendation, or +# favoring by the United States Government or any agency thereof, or +# Battelle Memorial Institute. The views and opinions of authors expressed +# herein do not necessarily state or reflect those of the +# United States Government or any agency thereof. +# +# PACIFIC NORTHWEST NATIONAL LABORATORY operated by +# BATTELLE for the UNITED STATES DEPARTMENT OF ENERGY +# under Contract DE-AC05-76RL01830 +# }}} +""" +This __init__.py file is a shim that allows the agent to run in either the 8.x version of VOLTTRON +or the new modular volttron. The imports are the primary difference and therefore if the +8.x version is loaded but the modular version is only available then an import exception will +be thrown. + +If neither are available then an import exception will be thrown. + +In order to use this, the agent itself should use this import by using + +from . import ( + get_aware_utc_now, + .... + .... +) + +This redirects the correct version of the code to either the modular or 8.x version to import. +""" +try: + # Attempt to import from 8.x version of VOLTTRON if not successful to import then + # attempt to import from modular version of VOLTTRON. + from volttron.platform import jsonapi + from volttron.platform.agent.utils import ( + get_aware_utc_now, + setup_logging, + format_timestamp, + load_config, + isapipe + ) + from volttron.platform.messaging import topics, headers + from volttron.platform.vip.agent import Core, Agent + +except ImportError: + + from volttron.utils import ( + get_aware_utc_now, + jsonapi, + setup_logging, + format_timestamp, + load_config, + isapipe, + ) + + from volttron.client.vip.agent import Agent, Core + from volttron.client.messaging import topics, headers diff --git a/volttron_openadr_ven/agent.py b/volttron_openadr_ven/agent.py index ef2c3e2..1334e10 100644 --- a/volttron_openadr_ven/agent.py +++ b/volttron_openadr_ven/agent.py @@ -35,45 +35,58 @@ # BATTELLE for the UNITED STATES DEPARTMENT OF ENERGY # under Contract DE-AC05-76RL01830 # }}} - import logging import asyncio import sys -import openleadr.enums +import os +import gevent +from pathlib import Path from pprint import pformat -from datetime import datetime +from datetime import timedelta from openleadr.enums import OPT, REPORT_NAME, MEASUREMENTS -from openleadr import OpenADRClient +from openleadr.client import OpenADRClient -from volttron.utils import ( +from . import ( get_aware_utc_now, - jsonapi, setup_logging, - vip_main, format_timestamp, load_config, + isapipe, + jsonapi, + topics, + headers, + Agent, +) + +from volttron_openadr_ven.constants import ( + REQUIRED_KEYS, + VEN_NAME, + VTN_URL, + VIP_ADDRESS, + SERVER_KEY, + AGENT_PUBLIC, + AGENT_SECRET, + DEBUG, + CERT, + KEY, + PASSPHRASE, + VTN_FINGERPRINT, + SHOW_FINGERPRINT, + CA_FILE, + VEN_ID, + DISABLE_SIGNATURE, + OPENADR_CLIENT_TYPE, + IDENTITY, ) -from volttron.client.vip.agent import Agent, Core -from volttron.client.messaging import topics, headers +from .volttron_openadr_client import openadr_clients -setup_logging(level=logging.DEBUG) + +setup_logging() _log = logging.getLogger(__name__) __version__ = "1.0" -VEN_NAME = "ven_name" -VTN_URL = "vtn_url" -DEBUG = "debug" -CERT = "cert" -KEY = "key" -PASSPHRASE = "passphrase" -VTN_FINGERPRINT = "vtn_fingerprint" -SHOW_FINGERPRINT = "show_fingerprint" -CA_FILE = "ca_file" -VEN_ID = "ven_id" -DISABLE_SIGNATURE = "disable_signature" - class OpenADRVenAgent(Agent): """ @@ -88,6 +101,7 @@ def __init__( self, ven_name: str, vtn_url: str, + openadr_client_type: str, debug: bool = False, cert: str = None, key: str = None, @@ -119,6 +133,7 @@ def __init__( certificate. str ven_id: The ID for this VEN. If you leave this blank, a VEN_ID will be assigned by the VTN. bool disable_signature: Whether to disable digital signatures + str openadr_client_type: The type of openadr client to use. Valid client types are the openadr client class names in ~/volttron_openadr_ven/volttron_openadr_client.py """ super(OpenADRVenAgent, self).__init__(enable_web=True, kwargs) @@ -136,6 +151,7 @@ def __init__( CA_FILE: ca_file, VEN_ID: ven_id, DISABLE_SIGNATURE: disable_signature, + OPENADR_CLIENT_TYPE: openadr_client_type, } # SubSystem/ConfigStore @@ -147,17 +163,24 @@ def __init__( # initialize all attributes of this class self.configure_agent(self.default_config) - def configure_agent(self, config: dict) -> None: + def configure_agent(self, config) -> None: """ Initialize the agent's configuration. Create an OpenADR Client using OpenLeadr. """ - _log.debug(f"Configuring agent with: \n {pformat(config)} ") + _log.info(f"Configuring agent with: \n {pformat(config)} ") # instantiate and add handlers to the OpenADR Client - _log.info("Creating OpenLeadrVen Client...") - self.ven_client = OpenADRClient( - ven_name=config.get(VEN_NAME), - vtn_url=config.get(VTN_URL), + client_type = config.get(OPENADR_CLIENT_TYPE) + openADRClient = openadr_clients().get(client_type) + if openADRClient is None: + msg = f"Invalid client type: {client_type}. Please use valid client types: {list(openadr_clients().keys())}" + _log.debug(msg) + raise KeyError(msg) + + _log.info(f"Creating OpenADRClient type: {client_type}...") + self.ven_client = openADRClient( + config.get(VEN_NAME), + config.get(VTN_URL), debug=config.get(DEBUG), cert=config.get(CERT), key=config.get(KEY), @@ -168,7 +191,11 @@ def configure_agent(self, config: dict) -> None: ven_id=config.get(VEN_ID), disable_signature=config.get(DISABLE_SIGNATURE), ) + _log.info(f"{client_type} successfully created.") + _log.info( + f"Adding capabilities (e.g. handlers, reports) to {client_type}..." + ) # Add event handling capability to the client # if you want to add more handlers on a specific event, you must create a coroutine in this class # and then add it as the second input for 'self.ven_client.add_handler(<some event>, <coroutine>)' @@ -183,7 +210,8 @@ def configure_agent(self, config: dict) -> None: measurement=MEASUREMENTS.VOLTAGE, ) - _log.info("Configuration complete.") + _log.info("Capabilities successfully added.") + gevent.spawn_later(3, self.start_asyncio_loop) def _configure(self, config_name, action, contents: dict) -> None: """The agent's config may have changed. Re-initialize it.""" @@ -191,39 +219,15 @@ def _configure(self, config_name, action, contents: dict) -> None: config.update(contents) self.configure_agent(config) - # ************* Methods for Managing the Agent on Volttron **************** - - @Core.receiver("onstart") - def onstart(self, sender) -> None: - """The agent has started.""" - _log.info(f"Sender {sender}") + def start_asyncio_loop(self): _log.info("Starting agent...") loop = asyncio.get_event_loop() loop.create_task(self.ven_client.run()) loop.run_forever() - # # TODO: Identify actions needed to be done before shutdown - # @Core.receiver("onstop") - # def onstop(self, sender, kwargs): - # """ - # This method is called when the Agent is about to shutdown, but before it disconnects from - # the message bus. - # """ - # pass - # - # # TODO: Identify what, if at all, RPC methods should be available to other Agents - # @RPC.export - # def rpc_method(self, arg1, arg2, kwarg1=None, kwarg2=None): - # """ - # RPC method - # - # May be called from another agent via self.vip.rpc.call - # """ - # pass - # *************** Methods for Servicing VTN Requests **************** - async def handle_event(self, event: dict) -> openleadr.enums.OPT: + async def handle_event(self, event) -> OPT: """ Publish event to the Volttron message bus. Return OPT response. This coroutine will be called when there is an event to be handled. @@ -252,7 +256,7 @@ async def handle_event(self, event: dict) -> openleadr.enums.OPT: """ try: _log.debug("Received event. Processing event now...") - signal = event.get("event_signals")[0] # type: ignore[index] + signal = event.get("event_signals")[0] _log.info(f"Event signal:\n {pformat(signal)}") except IndexError as e: _log.debug( @@ -299,14 +303,79 @@ def _json_object(self, obj: dict): """Ensure that an object is valid JSON by dumping it with json_converter and then reloading it.""" obj_string = jsonapi.dumps( obj, - default=lambda x: x.__str__() if isinstance(x, datetime) else None, + default=lambda x: int(x.total_seconds()) + if isinstance(x, timedelta) + else None, ) return jsonapi.loads(obj_string) def main(): """Main method called to start the agent.""" - vip_main(ven_agent, version=__version__) + # TODO: when volttron.utils gets fixed by https://github.com/VOLTTRON/volttron-utils/issues/6, uncomment the line below and remove vip_main_tmp + # vip_main(ven_agent, version=__version__) + vip_main_tmp() + + +def vip_main_tmp(): + config_path = os.environ.get("AGENT_CONFIG") + if not config_path: + # this function borrows code from volttron.utils.commands.vip_main + # it allows the user of this agent to set the certificates so that the remote volttron platform can authenticate this agent + import argparse + + # Instantiate the parser + parser = argparse.ArgumentParser() + parser.add_argument("config_path") + args = parser.parse_args() + config_path = args.config_path + + if isapipe(sys.stdout): + # Hold a reference to the previous file object so it doesn't + # get garbage collected and close the underlying descriptor. + stdout = sys.stdout + sys.stdout = os.fdopen(stdout.fileno(), "w", 1) + + agent = ven_agent(config_path) + + try: + run = agent.run + except AttributeError: + run = agent.core.run + task = gevent.spawn(run) + try: + task.join() + finally: + task.kill() + + +def check_required_key(required_key, key_actual): + if required_key == VEN_NAME and not key_actual: + raise KeyError(f"{VEN_NAME} is required.") + elif required_key == VTN_URL and not key_actual: + raise KeyError( + f"{VTN_URL} is required. Ensure {VTN_URL} is given a URL to the VTN." + ) + elif required_key == VIP_ADDRESS and not key_actual: + raise KeyError( + f"{VIP_ADDRESS} is required. For example, if running volttron instance locally, use tcp://127.0.0.1" + ) + elif required_key == SERVER_KEY and not key_actual: + raise KeyError( + f"{SERVER_KEY} is required. To get the server key from the volttron instance that this agent will be connected to, use the command: vctl auth serverkey" + ) + elif required_key == AGENT_PUBLIC and not key_actual: + raise KeyError( + f"{AGENT_PUBLIC} is required. To generate a public key and associated secret key from the volttron instance that this agent will be connected to, run the command: vctl auth keypair" + ) + elif required_key == AGENT_SECRET and not key_actual: + raise KeyError( + f"{AGENT_SECRET} is required. To generate a public key and associated secret key from the volttron instance that this agent will be connected to, run the command: vctl auth keypair" + ) + elif required_key == OPENADR_CLIENT_TYPE and not key_actual: + raise KeyError( + f"{OPENADR_CLIENT_TYPE} is required. Specify one of the following valid client types: {list(openadr_clients().keys())}" + ) def ven_agent(config_path: str, kwargs) -> OpenADRVenAgent: @@ -331,27 +400,42 @@ def ven_agent(config_path: str, kwargs) -> OpenADRVenAgent: if not config: raise Exception("Configuration cannot be empty.") - ven_name = config.get(VEN_NAME) - if not ven_name: - raise Exception(f"{VEN_NAME} is required.") - vtn_url = config.get(VTN_URL) - if not vtn_url: - raise Exception( - f"{VTN_URL} is required. Ensure {VTN_URL} is given a URL to the VTN." - ) + req_keys_actual = {k: "" for k in REQUIRED_KEYS} + for required_key in REQUIRED_KEYS: + key_actual = config.get(required_key) + check_required_key(required_key, key_actual) + req_keys_actual[required_key] = key_actual + + remote_url = ( + f"{req_keys_actual[VIP_ADDRESS]}" + f"?serverkey={req_keys_actual[SERVER_KEY]}" + f"&publickey={req_keys_actual[AGENT_PUBLIC]}" + f"&secretkey={req_keys_actual[AGENT_SECRET]}" + ) + + # optional configurations debug = config.get(DEBUG) + + # keypair paths cert = config.get(CERT) + if cert: + cert = str(Path(cert).expanduser().resolve(strict=True)) key = config.get(KEY) + if key: + key = str(Path(key).expanduser().resolve(strict=True)) + passphrase = config.get(PASSPHRASE) vtn_fingerprint = config.get(VTN_FINGERPRINT) show_fingerprint = config.get(SHOW_FINGERPRINT) ca_file = config.get(CA_FILE) ven_id = config.get(VEN_ID) disable_signature = bool(config.get(DISABLE_SIGNATURE)) + openadr_client_type = req_keys_actual[OPENADR_CLIENT_TYPE] return OpenADRVenAgent( - ven_name, - vtn_url, + req_keys_actual[VEN_NAME], + req_keys_actual[VTN_URL], + openadr_client_type, debug=debug, cert=cert, key=key, @@ -361,6 +445,9 @@ def ven_agent(config_path: str, kwargs) -> OpenADRVenAgent: ca_file=ca_file, ven_id=ven_id, disable_signature=disable_signature, + identity=IDENTITY, + # TODO: when volttron.utils gets fixed by https://github.com/VOLTTRON/volttron-utils/issues/6, remove the input 'address' + address=remote_url, kwargs, ) diff --git a/volttron_openadr_ven/config.json b/volttron_openadr_ven/config.json deleted file mode 100644 index acba83b..0000000 --- a/volttron_openadr_ven/config.json +++ /dev/null @@ -1,8 +0,0 @@ -{ - "ven_name": "PNNLVEN", - "vtn_url": "https://eiss2demo.ipkeys.com/oadr2/OpenADR2/Simple/2.0b", - "cert": "secret/", - "key": "secret/", - "debug": true, - "disable_signature": true -} diff --git a/volttron_openadr_ven/constants.py b/volttron_openadr_ven/constants.py new file mode 100644 index 0000000..7588f47 --- /dev/null +++ b/volttron_openadr_ven/constants.py @@ -0,0 +1,28 @@ +# key names for processing config file for OpenADR VEN agent +VEN_NAME = "ven_name" +VTN_URL = "vtn_url" +DEBUG = "debug" +CERT = "cert_path" +KEY = "key_path" +PASSPHRASE = "passphrase" +VTN_FINGERPRINT = "vtn_fingerprint" +SHOW_FINGERPRINT = "show_fingerprint" +CA_FILE = "ca_file" +VEN_ID = "ven_id" +DISABLE_SIGNATURE = "disable_signature" +VIP_ADDRESS = "vip_address" +SERVER_KEY = "server_key" +AGENT_PUBLIC = "agent_public" +AGENT_SECRET = "agent_secret" +OPENADR_CLIENT_TYPE = "openadr_client_type" +REQUIRED_KEYS = [ + VEN_NAME, + VTN_URL, + VIP_ADDRESS, + SERVER_KEY, + AGENT_PUBLIC, + AGENT_SECRET, + OPENADR_CLIENT_TYPE, +] + +IDENTITY = "openadr_ven" diff --git a/volttron_openadr_ven/volttron_openadr_client.py b/volttron_openadr_ven/volttron_openadr_client.py new file mode 100644 index 0000000..377bd91 --- /dev/null +++ b/volttron_openadr_ven/volttron_openadr_client.py @@ -0,0 +1,206 @@ +import logging +import asyncio + +from abc import ABC +from functools import partial +from lxml import etree + +from openleadr.client import OpenADRClient +from openleadr.preflight import preflight_message +from openleadr.messaging import TEMPLATES, SIGNER, _create_replay_protect +from openleadr import utils, enums + +from . import setup_logging + +setup_logging() +logger = logging.getLogger(__name__) + + +""" +The VolttronOpenADR VEN agent uses the python library OpenLEADR https://github.com/openleadr/openleadr-python to create +an OpenADR VEN client. This agent extends the OpenADRClient class from OpenLEADR so that it can have the flexibility +to connect to any implementation of and OpenADR VTN. For example, to connect to an IPKeys VTN that was implemented +on an old OpenADR protocol, the IPKeysDemoVTNOpenADRClient is created so that it can successfully connect to an IPKeys VTN. + +If you have a specific VTN that you want to connect to and require further customization of the VEN client, create your +own OpenADRClient by extending the base class VolttronOpenADRClientBase, updating your client with your business logic, and putting that subclass in this module +""" + + +class OpenADRClientBase(OpenADRClient, ABC): + def __init__(self, ven_name, vtn_url, disable_signature=False, kwargs): + super().__init__(ven_name, vtn_url, kwargs) + self.disable_signature = disable_signature + + +class IPKeysClient(OpenADRClientBase, ABC): + def __init__(self, ven_name, vtn_url, disable_signature, kwargs): + super().__init__(ven_name, vtn_url, disable_signature, *kwargs) + + self._create_message = partial( + self.create_message_ipkeys, + cert=self.cert_path, + key=self.key_path, + passphrase=self.passphrase, + disable_signature=self.disable_signature, + ) + + async def _on_event(self, message): + logger.debug("The VEN received an event") + events = message["events"] + try: + results = [] + + for event in message["events"]: + event_id = event["event_descriptor"]["event_id"] + event_status = event["event_descriptor"]["event_status"] + modification_number = event["event_descriptor"][ + "modification_number" + ] + received_event = utils.find_by( + self.received_events, "event_descriptor.event_id", event_id + ) + + if received_event: + if ( + received_event["event_descriptor"][ + "modification_number" + ] + == modification_number + ): + # Re-submit the same opt type as we already had previously + result = self.responded_events[event_id] + else: + # Replace the event with the fresh copy + utils.pop_by( + self.received_events, + "event_descriptor.event_id", + event_id, + ) + self.received_events.append(event) + # Wait for the result of the on_update_event handler + result = await utils.await_if_required( + self.on_update_event(event) + ) + else: + # Wait for the result of the on_event + self.received_events.append(event) + result = self.on_event(event) + if asyncio.iscoroutine(result): + result = await result + results.append(result) + if ( + event_status + in ( + enums.EVENT_STATUS.COMPLETED, + enums.EVENT_STATUS.CANCELLED, + ) + and event_id in self.responded_events + ): + self.responded_events.pop(event_id) + else: + self.responded_events[event_id] = result + for i, result in enumerate(results): + if ( + result not in ("optIn", "optOut") + and events[i]["response_required"] == "always" + ): + logger.error( + "Your on_event or on_update_event handler must return 'optIn' or 'optOut'; " + f"you supplied {result}. Please fix your on_event handler." + ) + results[i] = "optOut" + except Exception as err: + logger.error( + "Your on_event handler encountered an error. Will Opt Out of the event. " + f"The error was {err.__class__.__name__}: {str(err)}" + ) + results = ["optOut"] len(events) + + event_responses = [ + { + "response_code": 200, + "response_description": "OK", + "opt_type": results[i], + "request_id": message["request_id"], + "modification_number": events[i]["event_descriptor"][ + "modification_number" + ], + "event_id": events[i]["event_descriptor"]["event_id"], + } + for i, event in enumerate(events) + if event["response_required"] == "always" + and not utils.determine_event_status(event["active_period"]) + == "completed" + ] + + if len(event_responses) > 0: + response = { + "response_code": 200, + "response_description": "OK", + "request_id": message["request_id"], + } + message = self._create_message( + "oadrCreatedEvent", + response=response, + event_responses=event_responses, + ven_id=self.ven_id, + ) + service = "EiEvent" + response_type, response_payload = await self._perform_request( + service, message + ) + logger.info(response_type, response_payload) + else: + logger.info( + "Not sending any event responses, because a response was not required/allowed by the VTN." + ) + + @staticmethod + def create_message_ipkeys( + message_type, + cert=None, + key=None, + passphrase=None, + disable_signature=False, + message_payload, + ): + """ + Create and optionally sign an OpenADR message. Returns an XML string. + """ + message_payload = preflight_message(message_type, message_payload) + template = TEMPLATES.get_template(f"{message_type}.xml") + signed_object = utils.flatten_xml(template.render(message_payload)) + envelope = TEMPLATES.get_template("oadrPayload.xml") + if cert and key and not disable_signature: + tree = etree.fromstring(signed_object) + signature_tree = SIGNER.sign( + tree, + key=key, + cert=cert, + passphrase=utils.ensure_bytes(passphrase), + reference_uri="#oadrSignedObject", + signature_properties=_create_replay_protect(), + ) + signature = etree.tostring(signature_tree).decode("utf-8") + else: + signature = None + msg = envelope.render( + template=f"{message_type}", + signature=signature, + signed_object=signed_object, + ) + return msg + + +def openadr_clients(): + clients = {} + work = [OpenADRClient] + while work: + parent = work.pop() + for child in parent.__subclasses__(): + child_name = child.__name__ + if child_name not in clients: + clients[child_name] = child + work.append(child) + return clients	Connect to IPKeys vtn and see what gaps we have with openleadr (see ben's stuff in email) Create Skeleton Agent Structure This skeleton will have volttron agent properties as well as being able to start the async loop and be able to call the openleadr source code https://pypi.org/project/openleadr/	Was finally able to connect to IPKeys demo VTN using OpenLEADR, albeit with some strings attached. I had to modify the OpenLEADR library to create a VEN client that can disable signatures for clients that use TLS and certs. See https://github.com/bonicim/openleadr-python/tree/feature/disable-signature I opened up a PR to OpenLEADR with my changes: https://github.com/OpenLEADR/openleadr-python/pull/98 Since Openleadr has yet to review my PR, I have forked it, made the required changes to work with the Volttron Agent framework and to receive messages from the IPKeys VTN. The forked version used for the agent is at: https://github.com/bonicim/openleadr-python/tree/e306798f9f4af36f0349c10fec4607aaa18d34c8 Additional Notes: * This agent will take the following dependencies: * https://github.com/VOLTTRON/volttron-client/tree/main/volttron/platform * https://github.com/VOLTTRON/volttron-utils * This agent is a pared-down version of OpenADRAgent in volttron core; see https://github.com/VOLTTRON/volttron/blob/develop/services/core/OpenADRVenAgent/openadrven/agent.py#L414 * Agent should have the following methods * publish_event * process_event * process_report * Agent should ideally be a gevent-based agent that can make asyn calls; see example from openleader: https://openleadr.org/docs/#getting-started * Agent should be packaged and able to be installed via wheel on a volttron platform using `vctl install`	2021-12-04T00:40:06	[]	[]
eclipse-volttron/volttron-openadr-ven	eclipse-volttron__volttron-openadr-ven-13	7b0f5a919c37a8588736c5c85eb52a081685cd57	diff --git a/.gitignore b/.gitignore index b97131e..94cae34 100644 --- a/.gitignore +++ b/.gitignore @@ -129,3 +129,6 @@ dmypy.json .pyre/ .idea/ + +secret +config_local.json diff --git a/.gitmodules b/.gitmodules new file mode 100644 index 0000000..8dd3851 --- /dev/null +++ b/.gitmodules @@ -0,0 +1,3 @@ +[submodule "openleadr"] + path = openleadr + url = [email protected]:bonicim/openleadr-python.git diff --git a/README.md b/README.md index 0191cf6..933493b 100644 --- a/README.md +++ b/README.md @@ -1,43 +1,85 @@ -# Prerequisites +# OpenADRVen Agent + +OpenADR (Automated Demand Response) is a standard for alerting and responding to the need to adjust electric power +consumption in response to fluctuations in grid demand. + +For further information about OpenADR and this agent, please see the OpenADR documentation in VOLTTRON ReadTheDocs. + +## Agent Configuration + +The only required parameters for this agent are "ven_name" and "vtn_url". Below is an example of a +correct configuration with optional parameters added. + +```jsonpath + { + "ven_name": "PNNLVEN", + "vtn_url": "https://eiss2demo.ipkeys.com/oadr2/OpenADR2/Simple/2.0b", + + # below are optional parameters + + # if you want/need to sign outgoing messages using a public-private key pair, you need to add a directory named "secret" in the top level + # of this repo; then provide the relative path to the cert and key + # below is an example + # test certificates for IP Keys VTN's can be obtained at https://testcerts.kyrio.com/#/ + "cert": "secret/TEST_RSA_VEN_210923215148_certs/TEST_RSA_VEN_210923215148_cert.pem", + "key": "secret/TEST_RSA_VEN_210923215148_certs/TEST_RSA_VEN_210923215148_privkey.pem", + + # other optional parameters + "debug": true, + "disable_signature": true + } +``` + +## Volttron integration + +To ensure that this agent is created within the Volttron framework, ensure that you set the following environmental variable +to the absolute path of the configuration JSON file for this agent. For example, on Linux, use the following example to set +the environmental variable AGENT_CONFIG to a path to your JSON config file: + +`export AGENT_CONFIG /home/usr/path-to-my-config-file` + +# Development + +## Prerequisites Python >=3.7 * Pip >=20.1 * Poetry >=1.16 -# Development ## Environment setup Note: This repo uses [Poetry](https://python-poetry.org/), a dependency management and packaging tool for Python. If you don't have Poetry installed on your machine, follow [these steps](https://python-poetry.org/docs/#installation) to install it on your machine. -To check if Poetry is installed, run ```poetry --version```. If your receive 'command not found: poetry' error, add the following line to your '~/.bashrc' script: ```export PATH=$PATH:$HOME/.poetry/bin```. +To check if Poetry is installed, run `poetry --version`. If you receive the error 'command not found: poetry', add the following line to your '~/.bashrc' script: ```export PATH=$PATH:$HOME/.poetry/bin```. + +0. Install the required submodules -To setup your development environment, do the following: +```git submodule update --init --recursive``` 1. Create virtual environment -Run the following command on the CLI. By default, poetry creates a virtual environment in {cache-dir}/virtualenvs -({cache-dir}\virtualenvs on Windows). For more info on creating virtual environments through Poetry, see [this](https://python-poetry.org/docs/basic-usage/#using-your-virtual-environment). +By default, poetry creates a virtual environment in {cache-dir}/virtualenvs +({cache-dir}\virtualenvs on Windows). To configure 'poetry' to create the virtualenv inside this project's root +directory, run the following command: + +[```poetry config virtualenvs.in-project true```](https://python-poetry.org/docs/configuration +) + +Then to create the virtual environment itself, run the following command: ```shell -$ poetry shell +poetry shell ``` NOTE: Poetry needs the command 'python' in order to create a new shell. On Linux, if you only have python3 installed, run the following to link 'python' to 'python3': ```shell -$ sudo apt install python-is-python3 +sudo apt install python-is-python3 ``` 2. Install requirements ```shell -(env) $ poetry install -``` - - -3. Install Volttron-specific dependencies for this agent. - -```shell -(env) $ pip install volttron_utils-0.1.1-py3-none-any.whl volttron_client-0.1.2-py3-none-any.whl +poetry install ``` @@ -46,16 +88,21 @@ $ sudo apt install python-is-python3 ## Git Hooks and Pre-commit This repo provides pre-commit hooks to enforce code quality and standardization before pushing any commit to the repo. To use this tool locally, -run the following commands: +run the following command. ```shell -(env) $ pre-commit install +pre-commit install +``` + +To check that pre-commit scripts are installed and properly run, run pre-commit on all the files: -# To check that pre-commit scripts are installed and properly run, run pre-commit on all the files -(env) $ pre-commit run --all +``` +pre-commit run --all ``` -With pre-commit setup, every time you run "git commit", git will check your commit against a set of code linters. For details, see .pre-commit-config.yaml +With pre-commit installed and configured, every time you run `git commit`, git will check your commit against a set of code linters. For details, see .pre-commit-config.yaml + +This project installs the following tools to aid in code standardization and best practices: ## Black @@ -63,14 +110,14 @@ This repo uses [Black](https://pypi.org/project/black/), an opinionated Python c run the following command on your chosen Python file: ```shell -(env) $ black <path to file> +black <path to file> ``` ## MyPy This repo uses [MyPy](https://mypy.readthedocs.io/en/stable/), a static type checker for Python. To use MyPy on a Python file, run the following command: ```shell -(env) $ mypy <path to file> +mypy <path to file> ``` # CI/CD @@ -93,3 +140,23 @@ Test uploading Upload to PyPi ```python -m twin upload dist/``` + + +# Testing + +## Ven Certificates + +https://testcerts.kyrio.com/#/ + +## Test VTN's + + IPKeys: https://eiss2demo.ipkeys.com/ui/home/#/login + +# Other + +## Installing dependencies behind a proxy + +If running this behind a PNNL VPN, run the following so Poetry can install all the dependencies: +See: https://leifengblog.net/blog/how-to-use-pip-behind-a-proxy/ + +```export https_proxy=http://proxy01.pnl.gov:3128``` diff --git a/mypy.ini b/mypy.ini index 03aefad..8f59715 100644 --- a/mypy.ini +++ b/mypy.ini @@ -2,3 +2,4 @@ show_error_context = True pretty = True show_column_numbers = True +show_error_codes = True diff --git a/openleadr b/openleadr new file mode 160000 index 0000000..e306798 --- /dev/null +++ b/openleadr @@ -0,0 +1,1 @@ +Subproject commit e306798f9f4af36f0349c10fec4607aaa18d34c8 diff --git a/poetry.lock b/poetry.lock index 26f5d1c..b2f2e1c 100644 --- a/poetry.lock +++ b/poetry.lock @@ -17,17 +17,9 @@ yarl = ">=1.0,<2.0" [package.extras] speedups = ["aiodns", "brotlipy", "cchardet"] -[[package]] -name = "appdirs" -version = "1.4.4" -description = "A small Python module for determining appropriate platform-specific dirs, e.g. a \"user data dir\"." -category = "main" -optional = false -python-versions = "" - [[package]] name = "apscheduler" -version = "3.7.0" +version = "3.8.0" description = "In-process task scheduler with Cron-like capabilities" category = "main" optional = false @@ -51,6 +43,14 @@ tornado = ["tornado (>=4.3)"] twisted = ["twisted"] zookeeper = ["kazoo"] +[[package]] +name = "asn1crypto" +version = "1.4.0" +description = "Fast ASN.1 parser and serializer with definitions for private keys, public keys, certificates, CRL, OCSP, CMS, PKCS#3, PKCS#7, PKCS#8, PKCS#12, PKCS#5, X.509 and TSP" +category = "main" +optional = false +python-versions = "" + [[package]] name = "async-timeout" version = "3.0.1" @@ -81,31 +81,48 @@ docs = ["furo", "sphinx", "zope.interface", "sphinx-notfound-page"] tests = ["coverage[toml] (>=5.0.2)", "hypothesis", "pympler", "pytest (>=4.3.0)", "six", "mypy", "pytest-mypy-plugins", "zope.interface"] tests_no_zope = ["coverage[toml] (>=5.0.2)", "hypothesis", "pympler", "pytest (>=4.3.0)", "six", "mypy", "pytest-mypy-plugins"] +[[package]] +name = "backports.entry-points-selectable" +version = "1.1.0" +description = "Compatibility shim providing selectable entry points for older implementations" +category = "main" +optional = false +python-versions = ">=2.7" + +[package.extras] +docs = ["sphinx", "jaraco.packaging (>=8.2)", "rst.linker (>=1.9)"] +testing = ["pytest (>=4.6)", "pytest-flake8", "pytest-cov", "pytest-black (>=0.3.7)", "pytest-mypy", "pytest-checkdocs (>=2.4)", "pytest-enabler (>=1.0.1)"] + [[package]] name = "black" -version = "21.5b2" +version = "21.9b0" description = "The uncompromising code formatter." category = "main" optional = false python-versions = ">=3.6.2" [package.dependencies] -appdirs = "" click = ">=7.1.2" mypy-extensions = ">=0.4.3" -pathspec = ">=0.8.1,<1" +pathspec = ">=0.9.0,<1" +platformdirs = ">=2" regex = ">=2020.1.8" -toml = ">=0.10.1" +tomli = ">=0.2.6,<2.0.0" +typing-extensions = [ + {version = ">=3.10.0.0", markers = "python_version < \"3.10\""}, + {version = "!=3.10.0.1", markers = "python_version >= \"3.10\""}, +] [package.extras] colorama = ["colorama (>=0.4.3)"] d = ["aiohttp (>=3.6.0)", "aiohttp-cors (>=0.4.0)"] +jupyter = ["ipython (>=7.8.0)", "tokenize-rt (>=3.2.0)"] python2 = ["typed-ast (>=1.4.2)"] uvloop = ["uvloop (>=0.15.2)"] [[package]] name = "certifi" -version = "2021.5.30" +version = "2021.10.8" description = "Python package for providing Mozilla's CA Bundle." category = "main" optional = false @@ -113,7 +130,7 @@ python-versions = "" [[package]] name = "cffi" -version = "1.14.5" +version = "1.15.0" description = "Foreign Function Interface for Python calling C code." category = "main" optional = false @@ -124,7 +141,7 @@ pycparser = "" [[package]] name = "cfgv" -version = "3.3.0" +version = "3.3.1" description = "Validate configuration and produce human readable error messages." category = "main" optional = false @@ -138,9 +155,20 @@ category = "main" optional = false python-versions = ">=2.7, !=3.0., !=3.1., !=3.2., !=3.3., !=3.4." +[[package]] +name = "charset-normalizer" +version = "2.0.7" +description = "The Real First Universal Charset Detector. Open, modern and actively maintained alternative to Chardet." +category = "main" +optional = false +python-versions = ">=3.5.0" + +[package.extras] +unicode_backport = ["unicodedata2"] + [[package]] name = "click" -version = "8.0.1" +version = "8.0.3" description = "Composable command line interface toolkit" category = "main" optional = false @@ -159,26 +187,39 @@ python-versions = ">=2.7, !=3.0., !=3.1., !=3.2., !=3.3., !=3.4." [[package]] name = "cryptography" -version = "3.4.7" +version = "2.3" description = "cryptography is a package which provides cryptographic recipes and primitives to Python developers." category = "main" optional = false -python-versions = ">=3.6" +python-versions = ">=2.7,!=3.0.,!=3.1.,!=3.2.,!=3.3." [package.dependencies] -cffi = ">=1.12" +asn1crypto = ">=0.21.0" +cffi = ">=1.7,<1.11.3 \|\| >1.11.3" +idna = ">=2.1" +six = ">=1.4.1" [package.extras] -docs = ["sphinx (>=1.6.5,!=1.8.0,!=3.1.0,!=3.1.1)", "sphinx-rtd-theme"] -docstest = ["doc8", "pyenchant (>=1.6.11)", "twine (>=1.12.0)", "sphinxcontrib-spelling (>=4.0.1)"] -pep8test = ["black", "flake8", "flake8-import-order", "pep8-naming"] -sdist = ["setuptools-rust (>=0.11.4)"] -ssh = ["bcrypt (>=3.1.5)"] -test = ["pytest (>=6.0)", "pytest-cov", "pytest-subtests", "pytest-xdist", "pretend", "iso8601", "pytz", "hypothesis (>=1.11.4,!=3.79.2)"] +docs = ["sphinx (>=1.6.5)", "sphinx-rtd-theme"] +docstest = ["doc8", "pyenchant (>=1.6.11)", "readme_renderer (>=16.0)", "sphinxcontrib-spelling (>=4.0.1)"] +pep8test = ["flake8", "flake8-import-order", "pep8-naming"] +test = ["pytest (>=3.2.1,!=3.3.0)", "pretend", "iso8601", "pytz", "hypothesis (>=1.11.4)"] + +[[package]] +name = "dateutils" +version = "0.6.12" +description = "Various utilities for working with date and datetime objects" +category = "main" +optional = false +python-versions = "" + +[package.dependencies] +python-dateutil = "" +pytz = "" [[package]] name = "distlib" -version = "0.3.2" +version = "0.3.3" description = "Distribution utilities" category = "main" optional = false @@ -197,11 +238,15 @@ future = ">=0.16,<0.19" [[package]] name = "filelock" -version = "3.0.12" +version = "3.3.1" description = "A platform independent file lock." category = "main" optional = false -python-versions = "" +python-versions = ">=3.6" + +[package.extras] +docs = ["furo (>=2021.8.17b43)", "sphinx (>=4.1)", "sphinx-autodoc-typehints (>=1.12)"] +testing = ["covdefaults (>=1.2.0)", "coverage (>=4)", "pytest (>=4)", "pytest-cov", "pytest-timeout (>=1.4.2)"] [[package]] name = "future" @@ -211,9 +256,50 @@ category = "main" optional = false python-versions = ">=2.6, !=3.0., !=3.1., !=3.2." +[[package]] +name = "gevent" +version = "20.6.1" +description = "Coroutine-based network library" +category = "main" +optional = false +python-versions = ">=2.7,!=3.0.,!=3.1.,!=3.2.,!=3.3.,!=3.4." + +[package.dependencies] +cffi = {version = ">=1.12.2", markers = "platform_python_implementation == \"CPython\" and sys_platform == \"win32\""} +greenlet = {version = ">=0.4.16", markers = "platform_python_implementation == \"CPython\""} +"zope.event" = "" +"zope.interface" = "" + +[package.extras] +dnspython = ["dnspython (>=1.16.0)", "idna"] +docs = ["repoze.sphinx.autointerface", "sphinxcontrib-programoutput"] +monitor = ["psutil (>=5.6.1)", "psutil (==5.6.3)"] +recommended = ["dnspython (>=1.16.0)", "idna", "cffi (>=1.12.2)", "psutil (>=5.6.1)", "psutil (==5.6.3)", "selectors2", "backports.socketpair"] +test = ["dnspython (>=1.16.0)", "idna", "requests", "objgraph", "cffi (>=1.12.2)", "psutil (>=5.6.1)", "psutil (==5.6.3)", "selectors2", "futures", "mock", "backports.socketpair", "contextvars (==2.4)", "coverage (<5.0)", "coveralls (>=1.7.0)"] + +[[package]] +name = "greenlet" +version = "0.4.16" +description = "Lightweight in-process concurrent programming" +category = "main" +optional = false +python-versions = "" + +[[package]] +name = "grequests" +version = "0.6.0" +description = "Requests + Gevent" +category = "main" +optional = false +python-versions = "" + +[package.dependencies] +gevent = "" +requests = "" + [[package]] name = "identify" -version = "2.2.7" +version = "2.3.0" description = "File identification library for Python" category = "main" optional = false @@ -224,11 +310,11 @@ license = ["editdistance-s"] [[package]] name = "idna" -version = "3.2" +version = "2.10" description = "Internationalized Domain Names in Applications (IDNA)" category = "main" optional = false -python-versions = ">=3.5" +python-versions = ">=2.7, !=3.0., !=3.1., !=3.2., !=3.3." [[package]] name = "iniconfig" @@ -240,7 +326,7 @@ python-versions = "" [[package]] name = "jinja2" -version = "3.0.1" +version = "3.0.2" description = "A very fast and expressive template engine." category = "main" optional = false @@ -276,7 +362,7 @@ python-versions = ">=3.6" [[package]] name = "multidict" -version = "5.1.0" +version = "5.2.0" description = "multidict implementation" category = "main" optional = false @@ -316,11 +402,12 @@ python-versions = "" [[package]] name = "openleadr" -version = "0.5.23" -description = "Python3 library for building OpenADR Clients (VENs) and Servers (VTNs)" +version = "0.5.24" +description = "" category = "main" optional = false python-versions = ">=3.7.0" +develop = true [package.dependencies] aiohttp = "" @@ -329,39 +416,56 @@ jinja2 = "" signxml-openadr = "2.9.1" xmltodict = "" +[package.source] +type = "directory" +url = "openleadr" + [[package]] name = "packaging" -version = "20.9" +version = "21.0" description = "Core utilities for Python packages" category = "dev" optional = false -python-versions = ">=2.7, !=3.0., !=3.1., !=3.2., !=3.3." +python-versions = ">=3.6" [package.dependencies] pyparsing = ">=2.0.2" [[package]] name = "pathspec" -version = "0.8.1" +version = "0.9.0" description = "Utility library for gitignore style pattern matching of file paths." category = "main" optional = false -python-versions = ">=2.7, !=3.0., !=3.1., !=3.2., !=3.3., !=3.4." +python-versions = "!=3.0.,!=3.1.,!=3.2.,!=3.3.,!=3.4.,>=2.7" + +[[package]] +name = "platformdirs" +version = "2.4.0" +description = "A small Python module for determining appropriate platform-specific dirs, e.g. a \"user data dir\"." +category = "main" +optional = false +python-versions = ">=3.6" + +[package.extras] +docs = ["Sphinx (>=4)", "furo (>=2021.7.5b38)", "proselint (>=0.10.2)", "sphinx-autodoc-typehints (>=1.12)"] +test = ["appdirs (==1.4.4)", "pytest (>=6)", "pytest-cov (>=2.7)", "pytest-mock (>=3.6)"] [[package]] name = "pluggy" -version = "0.13.1" +version = "1.0.0" description = "plugin and hook calling mechanisms for python" category = "dev" optional = false -python-versions = ">=2.7, !=3.0., !=3.1., !=3.2., !=3.3." +python-versions = ">=3.6" [package.extras] dev = ["pre-commit", "tox"] +testing = ["pytest", "pytest-benchmark"] [[package]] name = "pre-commit" -version = "2.13.0" +version = "2.15.0" description = "A framework for managing and maintaining multi-language pre-commit hooks." category = "main" optional = false @@ -375,11 +479,22 @@ pyyaml = ">=5.1" toml = "" virtualenv = ">=20.0.8" +[[package]] +name = "psutil" +version = "5.8.0" +description = "Cross-platform lib for process and system monitoring in Python." +category = "main" +optional = false +python-versions = ">=2.6, !=3.0., !=3.1., !=3.2., !=3.3." + +[package.extras] +test = ["ipaddress", "mock", "unittest2", "enum34", "pywin32", "wmi"] + [[package]] name = "py" version = "1.10.0" description = "library with cross-python path, ini-parsing, io, code, log facilities" -category = "dev" +category = "main" optional = false python-versions = ">=2.7, !=3.0., !=3.1., !=3.2., !=3.3." @@ -393,14 +508,14 @@ python-versions = ">=2.7, !=3.0., !=3.1., !=3.2., !=3.3." [[package]] name = "pyopenssl" -version = "19.1.0" +version = "19.0.0" description = "Python wrapper module around the OpenSSL library" category = "main" optional = false python-versions = "" [package.dependencies] -cryptography = ">=2.8" +cryptography = ">=2.3" six = ">=1.5.2" [package.extras] @@ -417,7 +532,7 @@ python-versions = ">=2.6, !=3.0., !=3.1., !=3.2." [[package]] name = "pytest" -version = "6.2.4" +version = "6.2.5" description = "pytest: simple powerful testing with Python" category = "dev" optional = false @@ -429,16 +544,27 @@ attrs = ">=19.2.0" colorama = {version = "", markers = "sys_platform == \"win32\""} iniconfig = "" packaging = "" -pluggy = ">=0.12,<1.0.0a1" +pluggy = ">=0.12,<2.0" py = ">=1.8.2" toml = "" [package.extras] testing = ["argcomplete", "hypothesis (>=3.56)", "mock", "nose", "requests", "xmlschema"] +[[package]] +name = "python-dateutil" +version = "2.8.2" +description = "Extensions to the standard Python datetime module" +category = "main" +optional = false +python-versions = "!=3.0.,!=3.1.,!=3.2.,>=2.7" + +[package.dependencies] +six = ">=1.5" + [[package]] name = "pytz" -version = "2021.1" +version = "2021.3" description = "World timezone definitions, modern and historical" category = "main" optional = false @@ -452,14 +578,44 @@ category = "main" optional = false python-versions = ">=2.7, !=3.0., !=3.1., !=3.2., !=3.3., !=3.4., !=3.5." +[[package]] +name = "pyzmq" +version = "22.3.0" +description = "Python bindings for 0MQ" +category = "main" +optional = false +python-versions = ">=3.6" + +[package.dependencies] +cffi = {version = "", markers = "implementation_name == \"pypy\""} +py = {version = "", markers = "implementation_name == \"pypy\""} + [[package]] name = "regex" -version = "2021.4.4" +version = "2021.10.21" description = "Alternative regular expression module, to replace re." category = "main" optional = false python-versions = "" +[[package]] +name = "requests" +version = "2.26.0" +description = "Python HTTP for Humans." +category = "main" +optional = false +python-versions = ">=2.7, !=3.0., !=3.1., !=3.2., !=3.3., !=3.4., !=3.5." + +[package.dependencies] +certifi = ">=2017.4.17" +charset-normalizer = {version = ">=2.0.0,<2.1.0", markers = "python_version >= \"3\""} +idna = {version = ">=2.5,<4", markers = "python_version >= \"3\""} +urllib3 = ">=1.21.1,<1.27" + +[package.extras] +socks = ["PySocks (>=1.5.6,!=1.5.7)", "win-inet-pton"] +use_chardet_on_py3 = ["chardet (>=3.0.2,<5)"] + [[package]] name = "signxml-openadr" version = "2.9.1" @@ -491,6 +647,14 @@ category = "main" optional = false python-versions = ">=2.6, !=3.0., !=3.1., !=3.2." +[[package]] +name = "tomli" +version = "1.2.1" +description = "A lil' TOML parser" +category = "main" +optional = false +python-versions = ">=3.6" + [[package]] name = "typed-ast" version = "1.4.3" @@ -501,7 +665,7 @@ python-versions = "" [[package]] name = "typing-extensions" -version = "3.10.0.0" +version = "3.10.0.2" description = "Backported and Experimental Type Hints for Python 3.5+" category = "main" optional = false @@ -518,23 +682,96 @@ python-versions = "" [package.dependencies] pytz = "" +[[package]] +name = "urllib3" +version = "1.26.7" +description = "HTTP library with thread-safe connection pooling, file post, and more." +category = "main" +optional = false +python-versions = ">=2.7, !=3.0., !=3.1., !=3.2., !=3.3., !=3.4., <4" + +[package.extras] +brotli = ["brotlipy (>=0.6.0)"] +secure = ["pyOpenSSL (>=0.14)", "cryptography (>=1.3.4)", "idna (>=2.0.0)", "certifi", "ipaddress"] +socks = ["PySocks (>=1.5.6,!=1.5.7,<2.0)"] + [[package]] name = "virtualenv" -version = "20.4.7" +version = "20.8.1" description = "Virtual Python Environment builder" category = "main" optional = false -python-versions = "!=3.0.,!=3.1.,!=3.2.,!=3.3.,>=2.7" +python-versions = "!=3.0.,!=3.1.,!=3.2.,!=3.3.,!=3.4.,>=2.7" [package.dependencies] -appdirs = ">=1.4.3,<2" +"backports.entry-points-selectable" = ">=1.0.4" distlib = ">=0.3.1,<1" filelock = ">=3.0.0,<4" +platformdirs = ">=2,<3" six = ">=1.9.0,<2" [package.extras] docs = ["proselint (>=0.10.2)", "sphinx (>=3)", "sphinx-argparse (>=0.2.5)", "sphinx-rtd-theme (>=0.4.3)", "towncrier (>=19.9.0rc1)"] -testing = ["coverage (>=4)", "coverage-enable-subprocess (>=1)", "flaky (>=3)", "pytest (>=4)", "pytest-env (>=0.6.2)", "pytest-freezegun (>=0.4.1)", "pytest-mock (>=2)", "pytest-randomly (>=1)", "pytest-timeout (>=1)", "packaging (>=20.0)", "xonsh (>=0.9.16)"] +testing = ["coverage (>=4)", "coverage-enable-subprocess (>=1)", "flaky (>=3)", "pytest (>=4)", "pytest-env (>=0.6.2)", "pytest-freezegun (>=0.4.1)", "pytest-mock (>=2)", "pytest-randomly (>=1)", "pytest-timeout (>=1)", "packaging (>=20.0)"] + +[[package]] +name = "volttron-client" +version = "0.3.4" +description = "Client for connecting to a volttron server" +category = "main" +optional = false +python-versions = ">=3.6.2,<4" + +[package.dependencies] +dateutils = ">=0.6,<0.7" +grequests = ">=0.6.0,<0.7.0" +idna = ">=2.5,<3" +requests = ">=2.26.0,<3.0.0" +toml = ">=0.10.2,<0.11.0" +volttron-utils = ">=0.3,<0.4" + +[[package]] +name = "volttron-utils" +version = "0.3.2" +description = "Utility classes and methods for VOLTTRON" +category = "main" +optional = false +python-versions = ">=3.6.2,<4.0" + +[package.dependencies] +cryptography = "2.3" +dateutils = ">=0.6.7,<0.7.0" +gevent = "20.6.1" +greenlet = "0.4.16" +psutil = ">=5.7.0,<6.0.0" +pyOpenSSL = "19.0.0" +PyYAML = ">=5.4.1,<6.0.0" +tzlocal = "2.1" +watchdog-gevent = ">=0.1.1,<0.2.0" +zmq = ">=0.0.0,<0.0.1" + +[[package]] +name = "watchdog" +version = "2.1.6" +description = "Filesystem events monitoring" +category = "main" +optional = false +python-versions = ">=3.6" + +[package.extras] +watchmedo = ["PyYAML (>=3.10)"] + +[[package]] +name = "watchdog-gevent" +version = "0.1.1" +description = "A gevent-based observer for watchdog." +category = "main" +optional = false +python-versions = "" + +[package.dependencies] +gevent = ">=1.1" +watchdog = ">=0.8" [[package]] name = "xmltodict" @@ -546,7 +783,7 @@ python-versions = ">=2.7, !=3.0., !=3.1., !=3.2., !=3.3." [[package]] name = "yarl" -version = "1.6.3" +version = "1.7.0" description = "Yet another URL library" category = "main" optional = false @@ -556,10 +793,46 @@ python-versions = ">=3.6" idna = ">=2.0" multidict = ">=4.0" +[[package]] +name = "zmq" +version = "0.0.0" +description = "You are probably looking for pyzmq." +category = "main" +optional = false +python-versions = "" + +[package.dependencies] +pyzmq = "" + +[[package]] +name = "zope.event" +version = "4.5.0" +description = "Very basic event publishing system" +category = "main" +optional = false +python-versions = "" + +[package.extras] +docs = ["sphinx"] +test = ["zope.testrunner"] + +[[package]] +name = "zope.interface" +version = "5.4.0" +description = "Interfaces for Python" +category = "main" +optional = false +python-versions = ">=2.7, !=3.0., !=3.1., !=3.2., !=3.3., !=3.4." + +[package.extras] +docs = ["sphinx", "repoze.sphinx.autointerface"] +test = ["coverage (>=5.0.3)", "zope.event", "zope.testing"] +testing = ["coverage (>=5.0.3)", "zope.event", "zope.testing"] + [metadata] lock-version = "1.1" python-versions = "^3.8" -content-hash = "2a1dd13cfcde4f16340cb5425979909af99a26d9e7091cd2163a7e118ca1004b" +content-hash = "05aefd0e959b8da3ef1df6dbd3f3ec194bfe8b7cbc8d49f393a5919cd78c075e" [metadata.files] aiohttp = [ @@ -601,13 +874,13 @@ aiohttp = [ {file = "aiohttp-3.7.4.post0-cp39-cp39-win_amd64.whl", hash = "sha256:02f46fc0e3c5ac58b80d4d56eb0a7c7d97fcef69ace9326289fb9f1955e65cfe"}, {file = "aiohttp-3.7.4.post0.tar.gz", hash = "sha256:493d3299ebe5f5a7c66b9819eacdcfbbaaf1a8e84911ddffcdc48888497afecf"}, ] -appdirs = [ - {file = "appdirs-1.4.4-py2.py3-none-any.whl", hash = "sha256:a841dacd6b99318a741b166adb07e19ee71a274450e68237b4650ca1055ab128"}, - {file = "appdirs-1.4.4.tar.gz", hash = "sha256:7d5d0167b2b1ba821647616af46a749d1c653740dd0d2415100fe26e27afdf41"}, -] apscheduler = [ - {file = "APScheduler-3.7.0-py2.py3-none-any.whl", hash = "sha256:c06cc796d5bb9eb3c4f77727f6223476eb67749e7eea074d1587550702a7fbe3"}, - {file = "APScheduler-3.7.0.tar.gz", hash = "sha256:1cab7f2521e107d07127b042155b632b7a1cd5e02c34be5a28ff62f77c900c6a"}, + {file = "APScheduler-3.8.0-py2.py3-none-any.whl", hash = "sha256:82d6d21b2f0343510d07bb35618333a794653439a265635555b12935647b460a"}, + {file = "APScheduler-3.8.0.tar.gz", hash = "sha256:793b2d37c52ece53e34626619e6142e99b20b59a12155f39e1e6932e324f079d"}, +] +asn1crypto = [ + {file = "asn1crypto-1.4.0-py2.py3-none-any.whl", hash = "sha256:4bcdf33c861c7d40bdcd74d8e4dd7661aac320fcdf40b9a3f95b4ee12fde2fa8"}, + {file = "asn1crypto-1.4.0.tar.gz", hash = "sha256:f4f6e119474e58e04a2b1af817eb585b4fd72bdd89b998624712b5c99be7641c"}, ] async-timeout = [ {file = "async-timeout-3.0.1.tar.gz", hash = "sha256:0c3c816a028d47f659d6ff5c745cb2acf1f966da1fe5c19c77a70282b25f4c5f"}, @@ -621,125 +894,193 @@ attrs = [ {file = "attrs-21.2.0-py2.py3-none-any.whl", hash = "sha256:149e90d6d8ac20db7a955ad60cf0e6881a3f20d37096140088356da6c716b0b1"}, {file = "attrs-21.2.0.tar.gz", hash = "sha256:ef6aaac3ca6cd92904cdd0d83f629a15f18053ec84e6432106f7a4d04ae4f5fb"}, ] +"backports.entry-points-selectable" = [ + {file = "backports.entry_points_selectable-1.1.0-py2.py3-none-any.whl", hash = "sha256:a6d9a871cde5e15b4c4a53e3d43ba890cc6861ec1332c9c2428c92f977192acc"}, + {file = "backports.entry_points_selectable-1.1.0.tar.gz", hash = "sha256:988468260ec1c196dab6ae1149260e2f5472c9110334e5d51adcb77867361f6a"}, +] black = [ - {file = "black-21.5b2-py3-none-any.whl", hash = "sha256:e5cf21ebdffc7a9b29d73912b6a6a9a4df4ce70220d523c21647da2eae0751ef"}, - {file = "black-21.5b2.tar.gz", hash = "sha256:1fc0e0a2c8ae7d269dfcf0c60a89afa299664f3e811395d40b1922dff8f854b5"}, + {file = "black-21.9b0-py3-none-any.whl", hash = "sha256:380f1b5da05e5a1429225676655dddb96f5ae8c75bdf91e53d798871b902a115"}, + {file = "black-21.9b0.tar.gz", hash = "sha256:7de4cfc7eb6b710de325712d40125689101d21d25283eed7e9998722cf10eb91"}, ] certifi = [ - {file = "certifi-2021.5.30-py2.py3-none-any.whl", hash = "sha256:50b1e4f8446b06f41be7dd6338db18e0990601dce795c2b1686458aa7e8fa7d8"}, - {file = "certifi-2021.5.30.tar.gz", hash = "sha256:2bbf76fd432960138b3ef6dda3dde0544f27cbf8546c458e60baf371917ba9ee"}, + {file = "certifi-2021.10.8-py2.py3-none-any.whl", hash = "sha256:d62a0163eb4c2344ac042ab2bdf75399a71a2d8c7d47eac2e2ee91b9d6339569"}, + {file = "certifi-2021.10.8.tar.gz", hash = "sha256:78884e7c1d4b00ce3cea67b44566851c4343c120abd683433ce934a68ea58872"}, ] cffi = [ - {file = "cffi-1.14.5-cp27-cp27m-macosx_10_9_x86_64.whl", hash = "sha256:bb89f306e5da99f4d922728ddcd6f7fcebb3241fc40edebcb7284d7514741991"}, - {file = "cffi-1.14.5-cp27-cp27m-manylinux1_i686.whl", hash = "sha256:34eff4b97f3d982fb93e2831e6750127d1355a923ebaeeb565407b3d2f8d41a1"}, - {file = "cffi-1.14.5-cp27-cp27m-manylinux1_x86_64.whl", hash = "sha256:99cd03ae7988a93dd00bcd9d0b75e1f6c426063d6f03d2f90b89e29b25b82dfa"}, - {file = "cffi-1.14.5-cp27-cp27m-win32.whl", hash = "sha256:65fa59693c62cf06e45ddbb822165394a288edce9e276647f0046e1ec26920f3"}, - {file = "cffi-1.14.5-cp27-cp27m-win_amd64.whl", hash = "sha256:51182f8927c5af975fece87b1b369f722c570fe169f9880764b1ee3bca8347b5"}, - {file = "cffi-1.14.5-cp27-cp27mu-manylinux1_i686.whl", hash = "sha256:43e0b9d9e2c9e5d152946b9c5fe062c151614b262fda2e7b201204de0b99e482"}, - {file = "cffi-1.14.5-cp27-cp27mu-manylinux1_x86_64.whl", hash = "sha256:cbde590d4faaa07c72bf979734738f328d239913ba3e043b1e98fe9a39f8b2b6"}, - {file = "cffi-1.14.5-cp35-cp35m-macosx_10_9_x86_64.whl", hash = "sha256:5de7970188bb46b7bf9858eb6890aad302577a5f6f75091fd7cdd3ef13ef3045"}, - {file = "cffi-1.14.5-cp35-cp35m-manylinux1_i686.whl", hash = "sha256:a465da611f6fa124963b91bf432d960a555563efe4ed1cc403ba5077b15370aa"}, - {file = "cffi-1.14.5-cp35-cp35m-manylinux1_x86_64.whl", hash = "sha256:d42b11d692e11b6634f7613ad8df5d6d5f8875f5d48939520d351007b3c13406"}, - {file = "cffi-1.14.5-cp35-cp35m-win32.whl", hash = "sha256:72d8d3ef52c208ee1c7b2e341f7d71c6fd3157138abf1a95166e6165dd5d4369"}, - {file = "cffi-1.14.5-cp35-cp35m-win_amd64.whl", hash = "sha256:29314480e958fd8aab22e4a58b355b629c59bf5f2ac2492b61e3dc06d8c7a315"}, - {file = "cffi-1.14.5-cp36-cp36m-macosx_10_9_x86_64.whl", hash = "sha256:3d3dd4c9e559eb172ecf00a2a7517e97d1e96de2a5e610bd9b68cea3925b4892"}, - {file = "cffi-1.14.5-cp36-cp36m-manylinux1_i686.whl", hash = "sha256:48e1c69bbacfc3d932221851b39d49e81567a4d4aac3b21258d9c24578280058"}, - {file = "cffi-1.14.5-cp36-cp36m-manylinux1_x86_64.whl", hash = "sha256:69e395c24fc60aad6bb4fa7e583698ea6cc684648e1ffb7fe85e3c1ca131a7d5"}, - {file = "cffi-1.14.5-cp36-cp36m-manylinux2014_aarch64.whl", hash = "sha256:9e93e79c2551ff263400e1e4be085a1210e12073a31c2011dbbda14bda0c6132"}, - {file = "cffi-1.14.5-cp36-cp36m-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:24ec4ff2c5c0c8f9c6b87d5bb53555bf267e1e6f70e52e5a9740d32861d36b6f"}, - {file = "cffi-1.14.5-cp36-cp36m-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:3c3f39fa737542161d8b0d680df2ec249334cd70a8f420f71c9304bd83c3cbed"}, - {file = "cffi-1.14.5-cp36-cp36m-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:681d07b0d1e3c462dd15585ef5e33cb021321588bebd910124ef4f4fb71aef55"}, - {file = "cffi-1.14.5-cp36-cp36m-win32.whl", hash = "sha256:58e3f59d583d413809d60779492342801d6e82fefb89c86a38e040c16883be53"}, - {file = "cffi-1.14.5-cp36-cp36m-win_amd64.whl", hash = "sha256:005a36f41773e148deac64b08f233873a4d0c18b053d37da83f6af4d9087b813"}, - {file = "cffi-1.14.5-cp37-cp37m-macosx_10_9_x86_64.whl", hash = "sha256:2894f2df484ff56d717bead0a5c2abb6b9d2bf26d6960c4604d5c48bbc30ee73"}, - {file = "cffi-1.14.5-cp37-cp37m-manylinux1_i686.whl", hash = "sha256:0857f0ae312d855239a55c81ef453ee8fd24136eaba8e87a2eceba644c0d4c06"}, - {file = "cffi-1.14.5-cp37-cp37m-manylinux1_x86_64.whl", hash = "sha256:cd2868886d547469123fadc46eac7ea5253ea7fcb139f12e1dfc2bbd406427d1"}, - {file = "cffi-1.14.5-cp37-cp37m-manylinux2014_aarch64.whl", hash = "sha256:35f27e6eb43380fa080dccf676dece30bef72e4a67617ffda586641cd4508d49"}, - {file = "cffi-1.14.5-cp37-cp37m-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:06d7cd1abac2ffd92e65c0609661866709b4b2d82dd15f611e602b9b188b0b69"}, - {file = "cffi-1.14.5-cp37-cp37m-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:0f861a89e0043afec2a51fd177a567005847973be86f709bbb044d7f42fc4e05"}, - {file = "cffi-1.14.5-cp37-cp37m-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:cc5a8e069b9ebfa22e26d0e6b97d6f9781302fe7f4f2b8776c3e1daea35f1adc"}, - {file = "cffi-1.14.5-cp37-cp37m-win32.whl", hash = "sha256:9ff227395193126d82e60319a673a037d5de84633f11279e336f9c0f189ecc62"}, - {file = "cffi-1.14.5-cp37-cp37m-win_amd64.whl", hash = "sha256:9cf8022fb8d07a97c178b02327b284521c7708d7c71a9c9c355c178ac4bbd3d4"}, - {file = "cffi-1.14.5-cp38-cp38-macosx_10_9_x86_64.whl", hash = "sha256:8b198cec6c72df5289c05b05b8b0969819783f9418e0409865dac47288d2a053"}, - {file = "cffi-1.14.5-cp38-cp38-manylinux1_i686.whl", hash = "sha256:ad17025d226ee5beec591b52800c11680fca3df50b8b29fe51d882576e039ee0"}, - {file = "cffi-1.14.5-cp38-cp38-manylinux1_x86_64.whl", hash = "sha256:6c97d7350133666fbb5cf4abdc1178c812cb205dc6f41d174a7b0f18fb93337e"}, - {file = "cffi-1.14.5-cp38-cp38-manylinux2014_aarch64.whl", hash = "sha256:8ae6299f6c68de06f136f1f9e69458eae58f1dacf10af5c17353eae03aa0d827"}, - {file = "cffi-1.14.5-cp38-cp38-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:04c468b622ed31d408fea2346bec5bbffba2cc44226302a0de1ade9f5ea3d373"}, - {file = "cffi-1.14.5-cp38-cp38-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:06db6321b7a68b2bd6df96d08a5adadc1fa0e8f419226e25b2a5fbf6ccc7350f"}, - {file = "cffi-1.14.5-cp38-cp38-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:293e7ea41280cb28c6fcaaa0b1aa1f533b8ce060b9e701d78511e1e6c4a1de76"}, - {file = "cffi-1.14.5-cp38-cp38-win32.whl", hash = "sha256:b85eb46a81787c50650f2392b9b4ef23e1f126313b9e0e9013b35c15e4288e2e"}, - {file = "cffi-1.14.5-cp38-cp38-win_amd64.whl", hash = "sha256:1f436816fc868b098b0d63b8920de7d208c90a67212546d02f84fe78a9c26396"}, - {file = "cffi-1.14.5-cp39-cp39-macosx_10_9_x86_64.whl", hash = "sha256:1071534bbbf8cbb31b498d5d9db0f274f2f7a865adca4ae429e147ba40f73dea"}, - {file = "cffi-1.14.5-cp39-cp39-manylinux1_i686.whl", hash = "sha256:9de2e279153a443c656f2defd67769e6d1e4163952b3c622dcea5b08a6405322"}, - {file = "cffi-1.14.5-cp39-cp39-manylinux1_x86_64.whl", hash = "sha256:6e4714cc64f474e4d6e37cfff31a814b509a35cb17de4fb1999907575684479c"}, - {file = "cffi-1.14.5-cp39-cp39-manylinux2014_aarch64.whl", hash = "sha256:158d0d15119b4b7ff6b926536763dc0714313aa59e320ddf787502c70c4d4bee"}, - {file = "cffi-1.14.5-cp39-cp39-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:1bf1ac1984eaa7675ca8d5745a8cb87ef7abecb5592178406e55858d411eadc0"}, - {file = "cffi-1.14.5-cp39-cp39-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:df5052c5d867c1ea0b311fb7c3cd28b19df469c056f7fdcfe88c7473aa63e333"}, - {file = "cffi-1.14.5-cp39-cp39-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:24a570cd11895b60829e941f2613a4f79df1a27344cbbb82164ef2e0116f09c7"}, - {file = "cffi-1.14.5-cp39-cp39-win32.whl", hash = "sha256:afb29c1ba2e5a3736f1c301d9d0abe3ec8b86957d04ddfa9d7a6a42b9367e396"}, - {file = "cffi-1.14.5-cp39-cp39-win_amd64.whl", hash = "sha256:f2d45f97ab6bb54753eab54fffe75aaf3de4ff2341c9daee1987ee1837636f1d"}, - {file = "cffi-1.14.5.tar.gz", hash = "sha256:fd78e5fee591709f32ef6edb9a015b4aa1a5022598e36227500c8f4e02328d9c"}, + {file = "cffi-1.15.0-cp27-cp27m-macosx_10_9_x86_64.whl", hash = "sha256:c2502a1a03b6312837279c8c1bd3ebedf6c12c4228ddbad40912d671ccc8a962"}, + {file = "cffi-1.15.0-cp27-cp27m-manylinux1_i686.whl", hash = "sha256:23cfe892bd5dd8941608f93348c0737e369e51c100d03718f108bf1add7bd6d0"}, + {file = "cffi-1.15.0-cp27-cp27m-manylinux1_x86_64.whl", hash = "sha256:41d45de54cd277a7878919867c0f08b0cf817605e4eb94093e7516505d3c8d14"}, + {file = "cffi-1.15.0-cp27-cp27m-win32.whl", hash = "sha256:4a306fa632e8f0928956a41fa8e1d6243c71e7eb59ffbd165fc0b41e316b2474"}, + {file = "cffi-1.15.0-cp27-cp27m-win_amd64.whl", hash = "sha256:e7022a66d9b55e93e1a845d8c9eba2a1bebd4966cd8bfc25d9cd07d515b33fa6"}, + {file = "cffi-1.15.0-cp27-cp27mu-manylinux1_i686.whl", hash = "sha256:14cd121ea63ecdae71efa69c15c5543a4b5fbcd0bbe2aad864baca0063cecf27"}, + {file = "cffi-1.15.0-cp27-cp27mu-manylinux1_x86_64.whl", hash = "sha256:d4d692a89c5cf08a8557fdeb329b82e7bf609aadfaed6c0d79f5a449a3c7c023"}, + {file = "cffi-1.15.0-cp310-cp310-macosx_10_9_x86_64.whl", hash = "sha256:0104fb5ae2391d46a4cb082abdd5c69ea4eab79d8d44eaaf79f1b1fd806ee4c2"}, + {file = "cffi-1.15.0-cp310-cp310-macosx_11_0_arm64.whl", hash = "sha256:91ec59c33514b7c7559a6acda53bbfe1b283949c34fe7440bcf917f96ac0723e"}, + {file = "cffi-1.15.0-cp310-cp310-manylinux_2_12_i686.manylinux2010_i686.whl", hash = "sha256:f5c7150ad32ba43a07c4479f40241756145a1f03b43480e058cfd862bf5041c7"}, + {file = "cffi-1.15.0-cp310-cp310-manylinux_2_12_x86_64.manylinux2010_x86_64.whl", hash = "sha256:00c878c90cb53ccfaae6b8bc18ad05d2036553e6d9d1d9dbcf323bbe83854ca3"}, + {file = "cffi-1.15.0-cp310-cp310-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:abb9a20a72ac4e0fdb50dae135ba5e77880518e742077ced47eb1499e29a443c"}, + {file = "cffi-1.15.0-cp310-cp310-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:a5263e363c27b653a90078143adb3d076c1a748ec9ecc78ea2fb916f9b861962"}, + {file = "cffi-1.15.0-cp310-cp310-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:f54a64f8b0c8ff0b64d18aa76675262e1700f3995182267998c31ae974fbc382"}, + {file = "cffi-1.15.0-cp310-cp310-win32.whl", hash = "sha256:c21c9e3896c23007803a875460fb786118f0cdd4434359577ea25eb556e34c55"}, + {file = "cffi-1.15.0-cp310-cp310-win_amd64.whl", hash = "sha256:5e069f72d497312b24fcc02073d70cb989045d1c91cbd53979366077959933e0"}, + {file = "cffi-1.15.0-cp36-cp36m-macosx_10_9_x86_64.whl", hash = "sha256:64d4ec9f448dfe041705426000cc13e34e6e5bb13736e9fd62e34a0b0c41566e"}, + {file = "cffi-1.15.0-cp36-cp36m-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:2756c88cbb94231c7a147402476be2c4df2f6078099a6f4a480d239a8817ae39"}, + {file = "cffi-1.15.0-cp36-cp36m-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:3b96a311ac60a3f6be21d2572e46ce67f09abcf4d09344c49274eb9e0bf345fc"}, + {file = "cffi-1.15.0-cp36-cp36m-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:75e4024375654472cc27e91cbe9eaa08567f7fbdf822638be2814ce059f58032"}, + {file = "cffi-1.15.0-cp36-cp36m-manylinux_2_5_i686.manylinux1_i686.whl", hash = "sha256:59888172256cac5629e60e72e86598027aca6bf01fa2465bdb676d37636573e8"}, + {file = "cffi-1.15.0-cp36-cp36m-manylinux_2_5_x86_64.manylinux1_x86_64.whl", hash = "sha256:27c219baf94952ae9d50ec19651a687b826792055353d07648a5695413e0c605"}, + {file = "cffi-1.15.0-cp36-cp36m-win32.whl", hash = "sha256:4958391dbd6249d7ad855b9ca88fae690783a6be9e86df65865058ed81fc860e"}, + {file = "cffi-1.15.0-cp36-cp36m-win_amd64.whl", hash = "sha256:f6f824dc3bce0edab5f427efcfb1d63ee75b6fcb7282900ccaf925be84efb0fc"}, + {file = "cffi-1.15.0-cp37-cp37m-macosx_10_9_x86_64.whl", hash = "sha256:06c48159c1abed75c2e721b1715c379fa3200c7784271b3c46df01383b593636"}, + {file = "cffi-1.15.0-cp37-cp37m-manylinux_2_12_i686.manylinux2010_i686.whl", hash = "sha256:c2051981a968d7de9dd2d7b87bcb9c939c74a34626a6e2f8181455dd49ed69e4"}, + {file = "cffi-1.15.0-cp37-cp37m-manylinux_2_12_x86_64.manylinux2010_x86_64.whl", hash = "sha256:fd8a250edc26254fe5b33be00402e6d287f562b6a5b2152dec302fa15bb3e997"}, + {file = "cffi-1.15.0-cp37-cp37m-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:91d77d2a782be4274da750752bb1650a97bfd8f291022b379bb8e01c66b4e96b"}, + {file = "cffi-1.15.0-cp37-cp37m-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:45db3a33139e9c8f7c09234b5784a5e33d31fd6907800b316decad50af323ff2"}, + {file = "cffi-1.15.0-cp37-cp37m-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:263cc3d821c4ab2213cbe8cd8b355a7f72a8324577dc865ef98487c1aeee2bc7"}, + {file = "cffi-1.15.0-cp37-cp37m-win32.whl", hash = "sha256:17771976e82e9f94976180f76468546834d22a7cc404b17c22df2a2c81db0c66"}, + {file = "cffi-1.15.0-cp37-cp37m-win_amd64.whl", hash = "sha256:3415c89f9204ee60cd09b235810be700e993e343a408693e80ce7f6a40108029"}, + {file = "cffi-1.15.0-cp38-cp38-macosx_10_9_x86_64.whl", hash = "sha256:4238e6dab5d6a8ba812de994bbb0a79bddbdf80994e4ce802b6f6f3142fcc880"}, + {file = "cffi-1.15.0-cp38-cp38-manylinux_2_12_i686.manylinux2010_i686.whl", hash = "sha256:0808014eb713677ec1292301ea4c81ad277b6cdf2fdd90fd540af98c0b101d20"}, + {file = "cffi-1.15.0-cp38-cp38-manylinux_2_12_x86_64.manylinux2010_x86_64.whl", hash = "sha256:57e9ac9ccc3101fac9d6014fba037473e4358ef4e89f8e181f8951a2c0162024"}, + {file = "cffi-1.15.0-cp38-cp38-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:8b6c2ea03845c9f501ed1313e78de148cd3f6cad741a75d43a29b43da27f2e1e"}, + {file = "cffi-1.15.0-cp38-cp38-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:10dffb601ccfb65262a27233ac273d552ddc4d8ae1bf93b21c94b8511bffe728"}, + {file = "cffi-1.15.0-cp38-cp38-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:786902fb9ba7433aae840e0ed609f45c7bcd4e225ebb9c753aa39725bb3e6ad6"}, + {file = "cffi-1.15.0-cp38-cp38-win32.whl", hash = "sha256:da5db4e883f1ce37f55c667e5c0de439df76ac4cb55964655906306918e7363c"}, + {file = "cffi-1.15.0-cp38-cp38-win_amd64.whl", hash = "sha256:181dee03b1170ff1969489acf1c26533710231c58f95534e3edac87fff06c443"}, + {file = "cffi-1.15.0-cp39-cp39-macosx_10_9_x86_64.whl", hash = "sha256:45e8636704eacc432a206ac7345a5d3d2c62d95a507ec70d62f23cd91770482a"}, + {file = "cffi-1.15.0-cp39-cp39-macosx_11_0_arm64.whl", hash = "sha256:31fb708d9d7c3f49a60f04cf5b119aeefe5644daba1cd2a0fe389b674fd1de37"}, + {file = "cffi-1.15.0-cp39-cp39-manylinux_2_12_i686.manylinux2010_i686.whl", hash = "sha256:6dc2737a3674b3e344847c8686cf29e500584ccad76204efea14f451d4cc669a"}, + {file = "cffi-1.15.0-cp39-cp39-manylinux_2_12_x86_64.manylinux2010_x86_64.whl", hash = "sha256:74fdfdbfdc48d3f47148976f49fab3251e550a8720bebc99bf1483f5bfb5db3e"}, + {file = "cffi-1.15.0-cp39-cp39-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:ffaa5c925128e29efbde7301d8ecaf35c8c60ffbcd6a1ffd3a552177c8e5e796"}, + {file = "cffi-1.15.0-cp39-cp39-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:3f7d084648d77af029acb79a0ff49a0ad7e9d09057a9bf46596dac9514dc07df"}, + {file = "cffi-1.15.0-cp39-cp39-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:ef1f279350da2c586a69d32fc8733092fd32cc8ac95139a00377841f59a3f8d8"}, + {file = "cffi-1.15.0-cp39-cp39-win32.whl", hash = "sha256:2a23af14f408d53d5e6cd4e3d9a24ff9e05906ad574822a10563efcef137979a"}, + {file = "cffi-1.15.0-cp39-cp39-win_amd64.whl", hash = "sha256:3773c4d81e6e818df2efbc7dd77325ca0dcb688116050fb2b3011218eda36139"}, + {file = "cffi-1.15.0.tar.gz", hash = "sha256:920f0d66a896c2d99f0adbb391f990a84091179542c205fa53ce5787aff87954"}, ] cfgv = [ - {file = "cfgv-3.3.0-py2.py3-none-any.whl", hash = "sha256:b449c9c6118fe8cca7fa5e00b9ec60ba08145d281d52164230a69211c5d597a1"}, - {file = "cfgv-3.3.0.tar.gz", hash = "sha256:9e600479b3b99e8af981ecdfc80a0296104ee610cab48a5ae4ffd0b668650eb1"}, + {file = "cfgv-3.3.1-py2.py3-none-any.whl", hash = "sha256:c6a0883f3917a037485059700b9e75da2464e6c27051014ad85ba6aaa5884426"}, + {file = "cfgv-3.3.1.tar.gz", hash = "sha256:f5a830efb9ce7a445376bb66ec94c638a9787422f96264c98edc6bdeed8ab736"}, ] chardet = [ {file = "chardet-4.0.0-py2.py3-none-any.whl", hash = "sha256:f864054d66fd9118f2e67044ac8981a54775ec5b67aed0441892edb553d21da5"}, {file = "chardet-4.0.0.tar.gz", hash = "sha256:0d6f53a15db4120f2b08c94f11e7d93d2c911ee118b6b30a04ec3ee8310179fa"}, ] +charset-normalizer = [ + {file = "charset-normalizer-2.0.7.tar.gz", hash = "sha256:e019de665e2bcf9c2b64e2e5aa025fa991da8720daa3c1138cadd2fd1856aed0"}, + {file = "charset_normalizer-2.0.7-py3-none-any.whl", hash = "sha256:f7af805c321bfa1ce6714c51f254e0d5bb5e5834039bc17db7ebe3a4cec9492b"}, +] click = [ - {file = "click-8.0.1-py3-none-any.whl", hash = "sha256:fba402a4a47334742d782209a7c79bc448911afe1149d07bdabdf480b3e2f4b6"}, - {file = "click-8.0.1.tar.gz", hash = "sha256:8c04c11192119b1ef78ea049e0a6f0463e4c48ef00a30160c704337586f3ad7a"}, + {file = "click-8.0.3-py3-none-any.whl", hash = "sha256:353f466495adaeb40b6b5f592f9f91cb22372351c84caeb068132442a4518ef3"}, + {file = "click-8.0.3.tar.gz", hash = "sha256:410e932b050f5eed773c4cda94de75971c89cdb3155a72a0831139a79e5ecb5b"}, ] colorama = [ {file = "colorama-0.4.4-py2.py3-none-any.whl", hash = "sha256:9f47eda37229f68eee03b24b9748937c7dc3868f906e8ba69fbcbdd3bc5dc3e2"}, {file = "colorama-0.4.4.tar.gz", hash = "sha256:5941b2b48a20143d2267e95b1c2a7603ce057ee39fd88e7329b0c292aa16869b"}, ] cryptography = [ - {file = "cryptography-3.4.7-cp36-abi3-macosx_10_10_x86_64.whl", hash = "sha256:3d8427734c781ea5f1b41d6589c293089704d4759e34597dce91014ac125aad1"}, - {file = "cryptography-3.4.7-cp36-abi3-macosx_11_0_arm64.whl", hash = "sha256:8e56e16617872b0957d1c9742a3f94b43533447fd78321514abbe7db216aa250"}, - {file = "cryptography-3.4.7-cp36-abi3-manylinux2010_x86_64.whl", hash = "sha256:37340614f8a5d2fb9aeea67fd159bfe4f5f4ed535b1090ce8ec428b2f15a11f2"}, - {file = "cryptography-3.4.7-cp36-abi3-manylinux2014_aarch64.whl", hash = "sha256:240f5c21aef0b73f40bb9f78d2caff73186700bf1bc6b94285699aff98cc16c6"}, - {file = "cryptography-3.4.7-cp36-abi3-manylinux2014_x86_64.whl", hash = "sha256:1e056c28420c072c5e3cb36e2b23ee55e260cb04eee08f702e0edfec3fb51959"}, - {file = "cryptography-3.4.7-cp36-abi3-win32.whl", hash = "sha256:0f1212a66329c80d68aeeb39b8a16d54ef57071bf22ff4e521657b27372e327d"}, - {file = "cryptography-3.4.7-cp36-abi3-win_amd64.whl", hash = "sha256:de4e5f7f68220d92b7637fc99847475b59154b7a1b3868fb7385337af54ac9ca"}, - {file = "cryptography-3.4.7-pp36-pypy36_pp73-manylinux2010_x86_64.whl", hash = "sha256:26965837447f9c82f1855e0bc8bc4fb910240b6e0d16a664bb722df3b5b06873"}, - {file = "cryptography-3.4.7-pp36-pypy36_pp73-manylinux2014_x86_64.whl", hash = "sha256:eb8cc2afe8b05acbd84a43905832ec78e7b3873fb124ca190f574dca7389a87d"}, - {file = "cryptography-3.4.7-pp37-pypy37_pp73-manylinux2010_x86_64.whl", hash = "sha256:7ec5d3b029f5fa2b179325908b9cd93db28ab7b85bb6c1db56b10e0b54235177"}, - {file = "cryptography-3.4.7-pp37-pypy37_pp73-manylinux2014_x86_64.whl", hash = "sha256:ee77aa129f481be46f8d92a1a7db57269a2f23052d5f2433b4621bb457081cc9"}, - {file = "cryptography-3.4.7.tar.gz", hash = "sha256:3d10de8116d25649631977cb37da6cbdd2d6fa0e0281d014a5b7d337255ca713"}, + {file = "cryptography-2.3-cp27-cp27m-macosx_10_6_intel.whl", hash = "sha256:87d092a7c2a44e5f7414ab02fb4145723ebba411425e1a99773531dd4c0e9b8d"}, + {file = "cryptography-2.3-cp27-cp27m-manylinux1_i686.whl", hash = "sha256:67f7f57eae8dede577f3f7775957f5bec93edd6bdb6ce597bb5b28e1bdf3d4fb"}, + {file = "cryptography-2.3-cp27-cp27m-manylinux1_x86_64.whl", hash = "sha256:27bb401a20a838d6d0ea380f08c6ead3ccd8c9d8a0232dc9adcc0e4994576a66"}, + {file = "cryptography-2.3-cp27-cp27m-win32.whl", hash = "sha256:7ba834564daef87557e7fcd35c3c3183a4147b0b3a57314e53317360b9b201b3"}, + {file = "cryptography-2.3-cp27-cp27m-win_amd64.whl", hash = "sha256:9449f5d4d7c516a6118fa9210c4a00f34384cb1d2028672100ee0c6cce49d7f6"}, + {file = "cryptography-2.3-cp27-cp27mu-manylinux1_i686.whl", hash = "sha256:c3d945b7b577f07a477700f618f46cbc287af3a9222cd73035c6ef527ef2c363"}, + {file = "cryptography-2.3-cp27-cp27mu-manylinux1_x86_64.whl", hash = "sha256:bc2301170986ad82d9349a91eb8884e0e191209c45f5541b16aa7c0cfb135978"}, + {file = "cryptography-2.3-cp34-abi3-macosx_10_6_intel.whl", hash = "sha256:d01dfc5c2b3495184f683574e03c70022674ca9a7be88589c5aba130d835ea90"}, + {file = "cryptography-2.3-cp34-abi3-manylinux1_i686.whl", hash = "sha256:2a35b7570d8f247889784010aac8b384fd2e4a47b33e15c4a60b45a7c1944120"}, + {file = "cryptography-2.3-cp34-abi3-manylinux1_x86_64.whl", hash = "sha256:5eb86f03f9c4f0ac2336ac5431271072ddf7ecc76b338e26366732cfac58aa19"}, + {file = "cryptography-2.3-cp34-cp34m-win32.whl", hash = "sha256:29720c4253263cff9aea64585adbbe85013ba647f6e98367efff9db2d7193ded"}, + {file = "cryptography-2.3-cp34-cp34m-win_amd64.whl", hash = "sha256:82409f5150e529d699e5c33fa8fd85e965104db03bc564f5f4b6a9199e591f7c"}, + {file = "cryptography-2.3-cp35-cp35m-win32.whl", hash = "sha256:7d7f084cbe1fdb82be5a0545062b59b1ad3637bc5a48612ac2eb428ff31b31ea"}, + {file = "cryptography-2.3-cp35-cp35m-win_amd64.whl", hash = "sha256:6ec84edcbc966ae460560a51a90046503ff0b5b66157a9efc61515c68059f6c8"}, + {file = "cryptography-2.3-cp36-cp36m-win32.whl", hash = "sha256:8c56ef989342e42b9fcaba7c74b446f0cc9bed546dd00034fa7ad66fc00307ef"}, + {file = "cryptography-2.3-cp36-cp36m-win_amd64.whl", hash = "sha256:42c531a6a354407f42ee07fda5c2c0dc822cf6d52744949c182f2b295fbd4183"}, + {file = "cryptography-2.3-cp37-cp37m-win32.whl", hash = "sha256:21af753934f2f6d1a10fe8f4c0a64315af209ef6adeaee63ca349797d747d687"}, + {file = "cryptography-2.3-cp37-cp37m-win_amd64.whl", hash = "sha256:cee18beb4c807b5c0b178f4fa2fae03cef9d51821a358c6890f8b23465b7e5d2"}, + {file = "cryptography-2.3.tar.gz", hash = "sha256:c132bab45d4bd0fff1d3fe294d92b0a6eb8404e93337b3127bdec9f21de117e6"}, +] +dateutils = [ + {file = "dateutils-0.6.12-py2.py3-none-any.whl", hash = "sha256:f33b6ab430fa4166e7e9cb8b21ee9f6c9843c48df1a964466f52c79b2a8d53b3"}, + {file = "dateutils-0.6.12.tar.gz", hash = "sha256:03dd90bcb21541bd4eb4b013637e4f1b5f944881c46cc6e4b67a6059e370e3f1"}, ] distlib = [ - {file = "distlib-0.3.2-py2.py3-none-any.whl", hash = "sha256:23e223426b28491b1ced97dc3bbe183027419dfc7982b4fa2f05d5f3ff10711c"}, - {file = "distlib-0.3.2.zip", hash = "sha256:106fef6dc37dd8c0e2c0a60d3fca3e77460a48907f335fa28420463a6f799736"}, + {file = "distlib-0.3.3-py2.py3-none-any.whl", hash = "sha256:c8b54e8454e5bf6237cc84c20e8264c3e991e824ef27e8f1e81049867d861e31"}, + {file = "distlib-0.3.3.zip", hash = "sha256:d982d0751ff6eaaab5e2ec8e691d949ee80eddf01a62eaa96ddb11531fe16b05"}, ] eight = [ {file = "eight-1.0.1-py2.py3-none-any.whl", hash = "sha256:bd6967ec7cf9e61ab18b55751236ed89a8363cfcb69127aace336869615245d5"}, {file = "eight-1.0.1.tar.gz", hash = "sha256:6465970d3c12a1d759151690629b10b50ffb124c9b62e912999ee9233378a562"}, ] filelock = [ - {file = "filelock-3.0.12-py3-none-any.whl", hash = "sha256:929b7d63ec5b7d6b71b0fa5ac14e030b3f70b75747cef1b10da9b879fef15836"}, - {file = "filelock-3.0.12.tar.gz", hash = "sha256:18d82244ee114f543149c66a6e0c14e9c4f8a1044b5cdaadd0f82159d6a6ff59"}, + {file = "filelock-3.3.1-py3-none-any.whl", hash = "sha256:2b5eb3589e7fdda14599e7eb1a50e09b4cc14f34ed98b8ba56d33bfaafcbef2f"}, + {file = "filelock-3.3.1.tar.gz", hash = "sha256:34a9f35f95c441e7b38209775d6e0337f9a3759f3565f6c5798f19618527c76f"}, ] future = [ {file = "future-0.18.2.tar.gz", hash = "sha256:b1bead90b70cf6ec3f0710ae53a525360fa360d306a86583adc6bf83a4db537d"}, ] +gevent = [ + {file = "gevent-20.6.1-cp27-cp27m-macosx_10_15_x86_64.whl", hash = "sha256:12565278b65f958686199fe7d06e3c2aed5f0587be2e66e62e90b2df0a62a6c7"}, + {file = "gevent-20.6.1-cp27-cp27m-manylinux2010_x86_64.whl", hash = "sha256:2fb496f3b4de6591a293a65d04a14b32881d9e4e424defaa7193769489bc7c6d"}, + {file = "gevent-20.6.1-cp27-cp27m-win32.whl", hash = "sha256:edc4c7213f9c4ee88ad5d03a8f24f3d80b4decd09129f4d1eb660d3068906a69"}, + {file = "gevent-20.6.1-cp27-cp27m-win_amd64.whl", hash = "sha256:261413e577a6c8ca41b9fdf5ab5370bd470e3b32e176ad18e786cc45175e3af4"}, + {file = "gevent-20.6.1-cp27-cp27mu-manylinux2010_x86_64.whl", hash = "sha256:9d33866360dfd9457f509aa249746deba14f8790ca3544c0f1317fda773e3495"}, + {file = "gevent-20.6.1-cp35-cp35m-manylinux2010_x86_64.whl", hash = "sha256:f1dc96268bc3357cfb1d490eaf322c29a3649e63b672b718658e1797bd38cd64"}, + {file = "gevent-20.6.1-cp35-cp35m-win32.whl", hash = "sha256:1f5407c045fcdcf16f3c1e032c09e868eab4e2971a13003cdc38944cfef4a661"}, + {file = "gevent-20.6.1-cp35-cp35m-win_amd64.whl", hash = "sha256:2d36ea41da5347e165f8988718dca481f12711d62d999db953101516852d63fa"}, + {file = "gevent-20.6.1-cp36-cp36m-macosx_10_15_x86_64.whl", hash = "sha256:efdb51bcde9d604ab76d11e08c1adc62a5e87944b5435dcfad030d4cbbfe9a0f"}, + {file = "gevent-20.6.1-cp36-cp36m-manylinux2010_x86_64.whl", hash = "sha256:0fd5fbdaf5de55c38c45584489f73f263fbb4e06dd22496e9ba75ff42f469437"}, + {file = "gevent-20.6.1-cp36-cp36m-win32.whl", hash = "sha256:9164c8d0c217cf1a290d549c51a06f1352cd2cc169dac36305f377169cd7828b"}, + {file = "gevent-20.6.1-cp36-cp36m-win_amd64.whl", hash = "sha256:ad47b503b590dd972f791f01539f8806af33dfe854cae1d577846f9689deeecd"}, + {file = "gevent-20.6.1-cp37-cp37m-macosx_10_15_x86_64.whl", hash = "sha256:50af4b6a54d3b2b631a73c5c0100e2366e66c6d48ad9634f3c8fbdc0fa537964"}, + {file = "gevent-20.6.1-cp37-cp37m-manylinux2010_x86_64.whl", hash = "sha256:4070b229638ddba19850e9e648727462c0c8f0dc4277a753ae84faa7ac489712"}, + {file = "gevent-20.6.1-cp37-cp37m-win32.whl", hash = "sha256:7cb2bef61e59da4c4a6a3520b88ce8b7d5c3f1006657bc1355a693490a5d597a"}, + {file = "gevent-20.6.1-cp37-cp37m-win_amd64.whl", hash = "sha256:75fcbe5b50ba3b55a7a0683f58f02ece099714429e2120bfaa62f1d70d9dbfa5"}, + {file = "gevent-20.6.1-cp38-cp38-macosx_10_15_x86_64.whl", hash = "sha256:db00eb55398844ff4a6dc78f4db80a7dd6acb5cdbd5a0de0d6acaddc52e9c82f"}, + {file = "gevent-20.6.1-cp38-cp38-manylinux2010_x86_64.whl", hash = "sha256:4776da6ba4a784d80b6d9b6055929ebc2e09c1d2a5876f38b32c7d6435b59eb6"}, + {file = "gevent-20.6.1-cp38-cp38-win32.whl", hash = "sha256:68cc23bbf79f80f5313f94be0d5679c6862ce5bc45b34eb927472a327798f691"}, + {file = "gevent-20.6.1-cp38-cp38-win_amd64.whl", hash = "sha256:48ab796f200d2730fded2669ca52e36b68a57d114420ef36adf301bc3f593398"}, + {file = "gevent-20.6.1-cp39-cp39-manylinux2010_x86_64.whl", hash = "sha256:a498902a736382406d401d174df66ba99124b064d5e30db0e61e8bf2aeaf7bb1"}, + {file = "gevent-20.6.1-pp27-pypy_73-win32.whl", hash = "sha256:691e8f96b6357657eb6cd4d76affdac6e1648b3be69e3841b0dc331270c3835c"}, + {file = "gevent-20.6.1.tar.gz", hash = "sha256:53fdc591806b1ed3a747a8a27839f7c6ae229155aad309e265682f0f398eb488"}, +] +greenlet = [ + {file = "greenlet-0.4.16-cp27-cp27m-manylinux1_x86_64.whl", hash = "sha256:80cb0380838bf4e48da6adedb0c7cd060c187bb4a75f67a5aa9ec33689b84872"}, + {file = "greenlet-0.4.16-cp27-cp27m-win32.whl", hash = "sha256:df7de669cbf21de4b04a3ffc9920bc8426cab4c61365fa84d79bf97401a8bef7"}, + {file = "greenlet-0.4.16-cp27-cp27m-win_amd64.whl", hash = "sha256:1429dc183b36ec972055e13250d96e174491559433eb3061691b446899b87384"}, + {file = "greenlet-0.4.16-cp27-cp27mu-manylinux1_x86_64.whl", hash = "sha256:5ea034d040e6ab1d2ae04ab05a3f37dbd719c4dee3804b13903d4cc794b1336e"}, + {file = "greenlet-0.4.16-cp35-cp35m-manylinux1_x86_64.whl", hash = "sha256:c196a5394c56352e21cb7224739c6dd0075b69dd56f758505951d1d8d68cf8a9"}, + {file = "greenlet-0.4.16-cp35-cp35m-win32.whl", hash = "sha256:1000038ba0ea9032948e2156a9c15f5686f36945e8f9906e6b8db49f358e7b52"}, + {file = "greenlet-0.4.16-cp35-cp35m-win_amd64.whl", hash = "sha256:1b805231bfb7b2900a16638c3c8b45c694334c811f84463e52451e00c9412691"}, + {file = "greenlet-0.4.16-cp36-cp36m-manylinux1_x86_64.whl", hash = "sha256:e5db19d4a7d41bbeb3dd89b49fc1bc7e6e515b51bbf32589c618655a0ebe0bf0"}, + {file = "greenlet-0.4.16-cp36-cp36m-win32.whl", hash = "sha256:eac2a3f659d5f41d6bbfb6a97733bc7800ea5e906dc873732e00cebb98cec9e4"}, + {file = "greenlet-0.4.16-cp36-cp36m-win_amd64.whl", hash = "sha256:7eed31f4efc8356e200568ba05ad645525f1fbd8674f1e5be61a493e715e3873"}, + {file = "greenlet-0.4.16-cp37-cp37m-manylinux1_x86_64.whl", hash = "sha256:682328aa576ec393c1872615bcb877cf32d800d4a2f150e1a5dc7e56644010b1"}, + {file = "greenlet-0.4.16-cp37-cp37m-win32.whl", hash = "sha256:3a35e33902b2e6079949feed7a2dafa5ac6f019da97bd255842bb22de3c11bf5"}, + {file = "greenlet-0.4.16-cp37-cp37m-win_amd64.whl", hash = "sha256:b0b2a984bbfc543d144d88caad6cc7ff4a71be77102014bd617bd88cfb038727"}, + {file = "greenlet-0.4.16-cp38-cp38-manylinux1_x86_64.whl", hash = "sha256:d83c1d38658b0f81c282b41238092ed89d8f93c6e342224ab73fb39e16848721"}, + {file = "greenlet-0.4.16-cp38-cp38-win32.whl", hash = "sha256:e695ac8c3efe124d998230b219eb51afb6ef10524a50b3c45109c4b77a8a3a92"}, + {file = "greenlet-0.4.16-cp38-cp38-win_amd64.whl", hash = "sha256:133ba06bad4e5f2f8bf6a0ac434e0fd686df749a86b3478903b92ec3a9c0c90b"}, + {file = "greenlet-0.4.16.tar.gz", hash = "sha256:6e06eac722676797e8fce4adb8ad3dc57a1bb3adfb0dd3fdf8306c055a38456c"}, +] +grequests = [ + {file = "grequests-0.6.0-py3-none-any.whl", hash = "sha256:6eff964416021bb1dee4182a56cc2b551bfa42d37820e6aee6f2efa00d43a061"}, + {file = "grequests-0.6.0.tar.gz", hash = "sha256:7dec890c6668e6755a1ea968565535867956639301268394d24df67b478df666"}, +] identify = [ - {file = "identify-2.2.7-py2.py3-none-any.whl", hash = "sha256:92d6ad08eca19ceb17576733759944b94c0761277ddc3acf65e75e57ef190e32"}, - {file = "identify-2.2.7.tar.gz", hash = "sha256:c29e74c3671fe9537715cb695148231d777170ca1498e1c30c675d4ea782afe9"}, + {file = "identify-2.3.0-py2.py3-none-any.whl", hash = "sha256:d1e82c83d063571bb88087676f81261a4eae913c492dafde184067c584bc7c05"}, + {file = "identify-2.3.0.tar.gz", hash = "sha256:fd08c97f23ceee72784081f1ce5125c8f53a02d3f2716dde79a6ab8f1039fea5"}, ] idna = [ - {file = "idna-3.2-py3-none-any.whl", hash = "sha256:14475042e284991034cb48e06f6851428fb14c4dc953acd9be9a5e95c7b6dd7a"}, - {file = "idna-3.2.tar.gz", hash = "sha256:467fbad99067910785144ce333826c71fb0e63a425657295239737f7ecd125f3"}, + {file = "idna-2.10-py2.py3-none-any.whl", hash = "sha256:b97d804b1e9b523befed77c48dacec60e6dcb0b5391d57af6a65a312a90648c0"}, + {file = "idna-2.10.tar.gz", hash = "sha256:b307872f855b18632ce0c21c5e45be78c0ea7ae4c15c828c20788b26921eb3f6"}, ] iniconfig = [ {file = "iniconfig-1.1.1-py2.py3-none-any.whl", hash = "sha256:011e24c64b7f47f6ebd835bb12a743f2fbe9a26d4cecaa7f53bc4f35ee9da8b3"}, {file = "iniconfig-1.1.1.tar.gz", hash = "sha256:bc3af051d7d14b2ee5ef9969666def0cd1a000e121eaea580d4a313df4b37f32"}, ] jinja2 = [ - {file = "Jinja2-3.0.1-py3-none-any.whl", hash = "sha256:1f06f2da51e7b56b8f238affdd6b4e2c61e39598a378cc49345bc1bd42a978a4"}, - {file = "Jinja2-3.0.1.tar.gz", hash = "sha256:703f484b47a6af502e743c9122595cc812b0271f661722403114f71a79d0f5a4"}, + {file = "Jinja2-3.0.2-py3-none-any.whl", hash = "sha256:8569982d3f0889eed11dd620c706d39b60c36d6d25843961f33f77fb6bc6b20c"}, + {file = "Jinja2-3.0.2.tar.gz", hash = "sha256:827a0e32839ab1600d4eb1c4c33ec5a8edfbc5cb42dafa13b81f182f97784b45"}, ] lxml = [ {file = "lxml-4.6.3-cp27-cp27m-macosx_10_9_x86_64.whl", hash = "sha256:df7c53783a46febb0e70f6b05df2ba104610f2fb0d27023409734a3ecbb78fb2"}, @@ -749,6 +1090,8 @@ lxml = [ {file = "lxml-4.6.3-cp27-cp27m-win_amd64.whl", hash = "sha256:8157dadbb09a34a6bd95a50690595e1fa0af1a99445e2744110e3dca7831c4ee"}, {file = "lxml-4.6.3-cp27-cp27mu-manylinux1_i686.whl", hash = "sha256:7728e05c35412ba36d3e9795ae8995e3c86958179c9770e65558ec3fdfd3724f"}, {file = "lxml-4.6.3-cp27-cp27mu-manylinux1_x86_64.whl", hash = "sha256:4bff24dfeea62f2e56f5bab929b4428ae6caba2d1eea0c2d6eb618e30a71e6d4"}, + {file = "lxml-4.6.3-cp310-cp310-manylinux_2_12_i686.manylinux2010_i686.manylinux_2_24_i686.whl", hash = "sha256:64812391546a18896adaa86c77c59a4998f33c24788cadc35789e55b727a37f4"}, + {file = "lxml-4.6.3-cp310-cp310-manylinux_2_17_x86_64.manylinux2014_x86_64.manylinux_2_24_x86_64.whl", hash = "sha256:c1a40c06fd5ba37ad39caa0b3144eb3772e813b5fb5b084198a985431c2f1e8d"}, {file = "lxml-4.6.3-cp35-cp35m-manylinux1_i686.whl", hash = "sha256:74f7d8d439b18fa4c385f3f5dfd11144bb87c1da034a466c5b5577d23a1d9b51"}, {file = "lxml-4.6.3-cp35-cp35m-manylinux1_x86_64.whl", hash = "sha256:f90ba11136bfdd25cae3951af8da2e95121c9b9b93727b1b896e3fa105b2f586"}, {file = "lxml-4.6.3-cp35-cp35m-manylinux2010_i686.whl", hash = "sha256:4c61b3a0db43a1607d6264166b230438f85bfed02e8cff20c22e564d0faff354"}, @@ -790,12 +1133,22 @@ lxml = [ {file = "lxml-4.6.3.tar.gz", hash = "sha256:39b78571b3b30645ac77b95f7c69d1bffc4cf8c3b157c435a34da72e78c82468"}, ] markupsafe = [ + {file = "MarkupSafe-2.0.1-cp310-cp310-macosx_10_9_universal2.whl", hash = "sha256:d8446c54dc28c01e5a2dbac5a25f071f6653e6e40f3a8818e8b45d790fe6ef53"}, + {file = "MarkupSafe-2.0.1-cp310-cp310-macosx_10_9_x86_64.whl", hash = "sha256:36bc903cbb393720fad60fc28c10de6acf10dc6cc883f3e24ee4012371399a38"}, + {file = "MarkupSafe-2.0.1-cp310-cp310-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:2d7d807855b419fc2ed3e631034685db6079889a1f01d5d9dac950f764da3dad"}, + {file = "MarkupSafe-2.0.1-cp310-cp310-manylinux_2_5_i686.manylinux1_i686.manylinux_2_12_i686.manylinux2010_i686.whl", hash = "sha256:add36cb2dbb8b736611303cd3bfcee00afd96471b09cda130da3581cbdc56a6d"}, + {file = "MarkupSafe-2.0.1-cp310-cp310-manylinux_2_5_x86_64.manylinux1_x86_64.manylinux_2_12_x86_64.manylinux2010_x86_64.whl", hash = "sha256:168cd0a3642de83558a5153c8bd34f175a9a6e7f6dc6384b9655d2697312a646"}, + {file = "MarkupSafe-2.0.1-cp310-cp310-win32.whl", hash = "sha256:99df47edb6bda1249d3e80fdabb1dab8c08ef3975f69aed437cb69d0a5de1e28"}, + {file = "MarkupSafe-2.0.1-cp310-cp310-win_amd64.whl", hash = "sha256:e0f138900af21926a02425cf736db95be9f4af72ba1bb21453432a07f6082134"}, {file = "MarkupSafe-2.0.1-cp36-cp36m-macosx_10_9_x86_64.whl", hash = "sha256:f9081981fe268bd86831e5c75f7de206ef275defcb82bc70740ae6dc507aee51"}, {file = "MarkupSafe-2.0.1-cp36-cp36m-manylinux1_i686.whl", hash = "sha256:0955295dd5eec6cb6cc2fe1698f4c6d84af2e92de33fbcac4111913cd100a6ff"}, {file = "MarkupSafe-2.0.1-cp36-cp36m-manylinux1_x86_64.whl", hash = "sha256:0446679737af14f45767963a1a9ef7620189912317d095f2d9ffa183a4d25d2b"}, {file = "MarkupSafe-2.0.1-cp36-cp36m-manylinux2010_i686.whl", hash = "sha256:f826e31d18b516f653fe296d967d700fddad5901ae07c622bb3705955e1faa94"}, {file = "MarkupSafe-2.0.1-cp36-cp36m-manylinux2010_x86_64.whl", hash = "sha256:fa130dd50c57d53368c9d59395cb5526eda596d3ffe36666cd81a44d56e48872"}, {file = "MarkupSafe-2.0.1-cp36-cp36m-manylinux2014_aarch64.whl", hash = "sha256:905fec760bd2fa1388bb5b489ee8ee5f7291d692638ea5f67982d968366bef9f"}, + {file = "MarkupSafe-2.0.1-cp36-cp36m-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:bf5d821ffabf0ef3533c39c518f3357b171a1651c1ff6827325e4489b0e46c3c"}, + {file = "MarkupSafe-2.0.1-cp36-cp36m-manylinux_2_5_i686.manylinux1_i686.manylinux_2_12_i686.manylinux2010_i686.whl", hash = "sha256:0d4b31cc67ab36e3392bbf3862cfbadac3db12bdd8b02a2731f509ed5b829724"}, + {file = "MarkupSafe-2.0.1-cp36-cp36m-manylinux_2_5_x86_64.manylinux1_x86_64.manylinux_2_12_x86_64.manylinux2010_x86_64.whl", hash = "sha256:baa1a4e8f868845af802979fcdbf0bb11f94f1cb7ced4c4b8a351bb60d108145"}, {file = "MarkupSafe-2.0.1-cp36-cp36m-win32.whl", hash = "sha256:6c4ca60fa24e85fe25b912b01e62cb969d69a23a5d5867682dd3e80b5b02581d"}, {file = "MarkupSafe-2.0.1-cp36-cp36m-win_amd64.whl", hash = "sha256:b2f4bf27480f5e5e8ce285a8c8fd176c0b03e93dcc6646477d4630e83440c6a9"}, {file = "MarkupSafe-2.0.1-cp37-cp37m-macosx_10_9_x86_64.whl", hash = "sha256:0717a7390a68be14b8c793ba258e075c6f4ca819f15edfc2a3a027c823718567"}, @@ -804,14 +1157,21 @@ markupsafe = [ {file = "MarkupSafe-2.0.1-cp37-cp37m-manylinux2010_i686.whl", hash = "sha256:d7f9850398e85aba693bb640262d3611788b1f29a79f0c93c565694658f4071f"}, {file = "MarkupSafe-2.0.1-cp37-cp37m-manylinux2010_x86_64.whl", hash = "sha256:6a7fae0dd14cf60ad5ff42baa2e95727c3d81ded453457771d02b7d2b3f9c0c2"}, {file = "MarkupSafe-2.0.1-cp37-cp37m-manylinux2014_aarch64.whl", hash = "sha256:b7f2d075102dc8c794cbde1947378051c4e5180d52d276987b8d28a3bd58c17d"}, + {file = "MarkupSafe-2.0.1-cp37-cp37m-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:e9936f0b261d4df76ad22f8fee3ae83b60d7c3e871292cd42f40b81b70afae85"}, + {file = "MarkupSafe-2.0.1-cp37-cp37m-manylinux_2_5_i686.manylinux1_i686.manylinux_2_12_i686.manylinux2010_i686.whl", hash = "sha256:2a7d351cbd8cfeb19ca00de495e224dea7e7d919659c2841bbb7f420ad03e2d6"}, + {file = "MarkupSafe-2.0.1-cp37-cp37m-manylinux_2_5_x86_64.manylinux1_x86_64.manylinux_2_12_x86_64.manylinux2010_x86_64.whl", hash = "sha256:60bf42e36abfaf9aff1f50f52644b336d4f0a3fd6d8a60ca0d054ac9f713a864"}, {file = "MarkupSafe-2.0.1-cp37-cp37m-win32.whl", hash = "sha256:a30e67a65b53ea0a5e62fe23682cfe22712e01f453b95233b25502f7c61cb415"}, {file = "MarkupSafe-2.0.1-cp37-cp37m-win_amd64.whl", hash = "sha256:611d1ad9a4288cf3e3c16014564df047fe08410e628f89805e475368bd304914"}, + {file = "MarkupSafe-2.0.1-cp38-cp38-macosx_10_9_universal2.whl", hash = "sha256:5bb28c636d87e840583ee3adeb78172efc47c8b26127267f54a9c0ec251d41a9"}, {file = "MarkupSafe-2.0.1-cp38-cp38-macosx_10_9_x86_64.whl", hash = "sha256:be98f628055368795d818ebf93da628541e10b75b41c559fdf36d104c5787066"}, {file = "MarkupSafe-2.0.1-cp38-cp38-manylinux1_i686.whl", hash = "sha256:1d609f577dc6e1aa17d746f8bd3c31aa4d258f4070d61b2aa5c4166c1539de35"}, {file = "MarkupSafe-2.0.1-cp38-cp38-manylinux1_x86_64.whl", hash = "sha256:7d91275b0245b1da4d4cfa07e0faedd5b0812efc15b702576d103293e252af1b"}, {file = "MarkupSafe-2.0.1-cp38-cp38-manylinux2010_i686.whl", hash = "sha256:01a9b8ea66f1658938f65b93a85ebe8bc016e6769611be228d797c9d998dd298"}, {file = "MarkupSafe-2.0.1-cp38-cp38-manylinux2010_x86_64.whl", hash = "sha256:47ab1e7b91c098ab893b828deafa1203de86d0bc6ab587b160f78fe6c4011f75"}, {file = "MarkupSafe-2.0.1-cp38-cp38-manylinux2014_aarch64.whl", hash = "sha256:97383d78eb34da7e1fa37dd273c20ad4320929af65d156e35a5e2d89566d9dfb"}, + {file = "MarkupSafe-2.0.1-cp38-cp38-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:6fcf051089389abe060c9cd7caa212c707e58153afa2c649f00346ce6d260f1b"}, + {file = "MarkupSafe-2.0.1-cp38-cp38-manylinux_2_5_i686.manylinux1_i686.manylinux_2_12_i686.manylinux2010_i686.whl", hash = "sha256:5855f8438a7d1d458206a2466bf82b0f104a3724bf96a1c781ab731e4201731a"}, + {file = "MarkupSafe-2.0.1-cp38-cp38-manylinux_2_5_x86_64.manylinux1_x86_64.manylinux_2_12_x86_64.manylinux2010_x86_64.whl", hash = "sha256:3dd007d54ee88b46be476e293f48c85048603f5f516008bee124ddd891398ed6"}, {file = "MarkupSafe-2.0.1-cp38-cp38-win32.whl", hash = "sha256:023cb26ec21ece8dc3907c0e8320058b2e0cb3c55cf9564da612bc325bed5e64"}, {file = "MarkupSafe-2.0.1-cp38-cp38-win_amd64.whl", hash = "sha256:984d76483eb32f1bcb536dc27e4ad56bba4baa70be32fa87152832cdd9db0833"}, {file = "MarkupSafe-2.0.1-cp39-cp39-macosx_10_9_universal2.whl", hash = "sha256:2ef54abee730b502252bcdf31b10dacb0a416229b72c18b19e24a4509f273d26"}, @@ -821,48 +1181,86 @@ markupsafe = [ {file = "MarkupSafe-2.0.1-cp39-cp39-manylinux2010_i686.whl", hash = "sha256:4efca8f86c54b22348a5467704e3fec767b2db12fc39c6d963168ab1d3fc9135"}, {file = "MarkupSafe-2.0.1-cp39-cp39-manylinux2010_x86_64.whl", hash = "sha256:ab3ef638ace319fa26553db0624c4699e31a28bb2a835c5faca8f8acf6a5a902"}, {file = "MarkupSafe-2.0.1-cp39-cp39-manylinux2014_aarch64.whl", hash = "sha256:f8ba0e8349a38d3001fae7eadded3f6606f0da5d748ee53cc1dab1d6527b9509"}, + {file = "MarkupSafe-2.0.1-cp39-cp39-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:c47adbc92fc1bb2b3274c4b3a43ae0e4573d9fbff4f54cd484555edbf030baf1"}, + {file = "MarkupSafe-2.0.1-cp39-cp39-manylinux_2_5_i686.manylinux1_i686.manylinux_2_12_i686.manylinux2010_i686.whl", hash = "sha256:37205cac2a79194e3750b0af2a5720d95f786a55ce7df90c3af697bfa100eaac"}, + {file = "MarkupSafe-2.0.1-cp39-cp39-manylinux_2_5_x86_64.manylinux1_x86_64.manylinux_2_12_x86_64.manylinux2010_x86_64.whl", hash = "sha256:1f2ade76b9903f39aa442b4aadd2177decb66525062db244b35d71d0ee8599b6"}, {file = "MarkupSafe-2.0.1-cp39-cp39-win32.whl", hash = "sha256:10f82115e21dc0dfec9ab5c0223652f7197feb168c940f3ef61563fc2d6beb74"}, {file = "MarkupSafe-2.0.1-cp39-cp39-win_amd64.whl", hash = "sha256:693ce3f9e70a6cf7d2fb9e6c9d8b204b6b39897a2c4a1aa65728d5ac97dcc1d8"}, {file = "MarkupSafe-2.0.1.tar.gz", hash = "sha256:594c67807fb16238b30c44bdf74f36c02cdf22d1c8cda91ef8a0ed8dabf5620a"}, ] multidict = [ - {file = "multidict-5.1.0-cp36-cp36m-macosx_10_14_x86_64.whl", hash = "sha256:b7993704f1a4b204e71debe6095150d43b2ee6150fa4f44d6d966ec356a8d61f"}, - {file = "multidict-5.1.0-cp36-cp36m-manylinux1_i686.whl", hash = "sha256:9dd6e9b1a913d096ac95d0399bd737e00f2af1e1594a787e00f7975778c8b2bf"}, - {file = "multidict-5.1.0-cp36-cp36m-manylinux2014_aarch64.whl", hash = "sha256:f21756997ad8ef815d8ef3d34edd98804ab5ea337feedcd62fb52d22bf531281"}, - {file = "multidict-5.1.0-cp36-cp36m-manylinux2014_i686.whl", hash = "sha256:1ab820665e67373de5802acae069a6a05567ae234ddb129f31d290fc3d1aa56d"}, - {file = "multidict-5.1.0-cp36-cp36m-manylinux2014_ppc64le.whl", hash = "sha256:9436dc58c123f07b230383083855593550c4d301d2532045a17ccf6eca505f6d"}, - {file = "multidict-5.1.0-cp36-cp36m-manylinux2014_s390x.whl", hash = "sha256:830f57206cc96ed0ccf68304141fec9481a096c4d2e2831f311bde1c404401da"}, - {file = "multidict-5.1.0-cp36-cp36m-manylinux2014_x86_64.whl", hash = "sha256:2e68965192c4ea61fff1b81c14ff712fc7dc15d2bd120602e4a3494ea6584224"}, - {file = "multidict-5.1.0-cp36-cp36m-win32.whl", hash = "sha256:2f1a132f1c88724674271d636e6b7351477c27722f2ed789f719f9e3545a3d26"}, - {file = "multidict-5.1.0-cp36-cp36m-win_amd64.whl", hash = "sha256:3a4f32116f8f72ecf2a29dabfb27b23ab7cdc0ba807e8459e59a93a9be9506f6"}, - {file = "multidict-5.1.0-cp37-cp37m-macosx_10_14_x86_64.whl", hash = "sha256:46c73e09ad374a6d876c599f2328161bcd95e280f84d2060cf57991dec5cfe76"}, - {file = "multidict-5.1.0-cp37-cp37m-manylinux1_i686.whl", hash = "sha256:018132dbd8688c7a69ad89c4a3f39ea2f9f33302ebe567a879da8f4ca73f0d0a"}, - {file = "multidict-5.1.0-cp37-cp37m-manylinux2014_aarch64.whl", hash = "sha256:4b186eb7d6ae7c06eb4392411189469e6a820da81447f46c0072a41c748ab73f"}, - {file = "multidict-5.1.0-cp37-cp37m-manylinux2014_i686.whl", hash = "sha256:3a041b76d13706b7fff23b9fc83117c7b8fe8d5fe9e6be45eee72b9baa75f348"}, - {file = "multidict-5.1.0-cp37-cp37m-manylinux2014_ppc64le.whl", hash = "sha256:051012ccee979b2b06be928a6150d237aec75dd6bf2d1eeeb190baf2b05abc93"}, - {file = "multidict-5.1.0-cp37-cp37m-manylinux2014_s390x.whl", hash = "sha256:6a4d5ce640e37b0efcc8441caeea8f43a06addace2335bd11151bc02d2ee31f9"}, - {file = "multidict-5.1.0-cp37-cp37m-manylinux2014_x86_64.whl", hash = "sha256:5cf3443199b83ed9e955f511b5b241fd3ae004e3cb81c58ec10f4fe47c7dce37"}, - {file = "multidict-5.1.0-cp37-cp37m-win32.whl", hash = "sha256:f200755768dc19c6f4e2b672421e0ebb3dd54c38d5a4f262b872d8cfcc9e93b5"}, - {file = "multidict-5.1.0-cp37-cp37m-win_amd64.whl", hash = "sha256:05c20b68e512166fddba59a918773ba002fdd77800cad9f55b59790030bab632"}, - {file = "multidict-5.1.0-cp38-cp38-macosx_10_14_x86_64.whl", hash = "sha256:54fd1e83a184e19c598d5e70ba508196fd0bbdd676ce159feb412a4a6664f952"}, - {file = "multidict-5.1.0-cp38-cp38-manylinux1_i686.whl", hash = "sha256:0e3c84e6c67eba89c2dbcee08504ba8644ab4284863452450520dad8f1e89b79"}, - {file = "multidict-5.1.0-cp38-cp38-manylinux2014_aarch64.whl", hash = "sha256:dc862056f76443a0db4509116c5cd480fe1b6a2d45512a653f9a855cc0517456"}, - {file = "multidict-5.1.0-cp38-cp38-manylinux2014_i686.whl", hash = "sha256:0e929169f9c090dae0646a011c8b058e5e5fb391466016b39d21745b48817fd7"}, - {file = "multidict-5.1.0-cp38-cp38-manylinux2014_ppc64le.whl", hash = "sha256:d81eddcb12d608cc08081fa88d046c78afb1bf8107e6feab5d43503fea74a635"}, - {file = "multidict-5.1.0-cp38-cp38-manylinux2014_s390x.whl", hash = "sha256:585fd452dd7782130d112f7ddf3473ffdd521414674c33876187e101b588738a"}, - {file = "multidict-5.1.0-cp38-cp38-manylinux2014_x86_64.whl", hash = "sha256:37e5438e1c78931df5d3c0c78ae049092877e5e9c02dd1ff5abb9cf27a5914ea"}, - {file = "multidict-5.1.0-cp38-cp38-win32.whl", hash = "sha256:07b42215124aedecc6083f1ce6b7e5ec5b50047afa701f3442054373a6deb656"}, - {file = "multidict-5.1.0-cp38-cp38-win_amd64.whl", hash = "sha256:929006d3c2d923788ba153ad0de8ed2e5ed39fdbe8e7be21e2f22ed06c6783d3"}, - {file = "multidict-5.1.0-cp39-cp39-macosx_10_14_x86_64.whl", hash = "sha256:b797515be8743b771aa868f83563f789bbd4b236659ba52243b735d80b29ed93"}, - {file = "multidict-5.1.0-cp39-cp39-manylinux1_i686.whl", hash = "sha256:d5c65bdf4484872c4af3150aeebe101ba560dcfb34488d9a8ff8dbcd21079647"}, - {file = "multidict-5.1.0-cp39-cp39-manylinux2014_aarch64.whl", hash = "sha256:b47a43177a5e65b771b80db71e7be76c0ba23cc8aa73eeeb089ed5219cdbe27d"}, - {file = "multidict-5.1.0-cp39-cp39-manylinux2014_i686.whl", hash = "sha256:806068d4f86cb06af37cd65821554f98240a19ce646d3cd24e1c33587f313eb8"}, - {file = "multidict-5.1.0-cp39-cp39-manylinux2014_ppc64le.whl", hash = "sha256:46dd362c2f045095c920162e9307de5ffd0a1bfbba0a6e990b344366f55a30c1"}, - {file = "multidict-5.1.0-cp39-cp39-manylinux2014_s390x.whl", hash = "sha256:ace010325c787c378afd7f7c1ac66b26313b3344628652eacd149bdd23c68841"}, - {file = "multidict-5.1.0-cp39-cp39-manylinux2014_x86_64.whl", hash = "sha256:ecc771ab628ea281517e24fd2c52e8f31c41e66652d07599ad8818abaad38cda"}, - {file = "multidict-5.1.0-cp39-cp39-win32.whl", hash = "sha256:fc13a9524bc18b6fb6e0dbec3533ba0496bbed167c56d0aabefd965584557d80"}, - {file = "multidict-5.1.0-cp39-cp39-win_amd64.whl", hash = "sha256:7df80d07818b385f3129180369079bd6934cf70469f99daaebfac89dca288359"}, - {file = "multidict-5.1.0.tar.gz", hash = "sha256:25b4e5f22d3a37ddf3effc0710ba692cfc792c2b9edfb9c05aefe823256e84d5"}, + {file = "multidict-5.2.0-cp310-cp310-macosx_10_9_universal2.whl", hash = "sha256:3822c5894c72e3b35aae9909bef66ec83e44522faf767c0ad39e0e2de11d3b55"}, + {file = "multidict-5.2.0-cp310-cp310-macosx_10_9_x86_64.whl", hash = "sha256:28e6d883acd8674887d7edc896b91751dc2d8e87fbdca8359591a13872799e4e"}, + {file = "multidict-5.2.0-cp310-cp310-macosx_11_0_arm64.whl", hash = "sha256:b61f85101ef08cbbc37846ac0e43f027f7844f3fade9b7f6dd087178caedeee7"}, + {file = "multidict-5.2.0-cp310-cp310-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:d9b668c065968c5979fe6b6fa6760bb6ab9aeb94b75b73c0a9c1acf6393ac3bf"}, + {file = "multidict-5.2.0-cp310-cp310-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:517d75522b7b18a3385726b54a081afd425d4f41144a5399e5abd97ccafdf36b"}, + {file = "multidict-5.2.0-cp310-cp310-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:1b4ac3ba7a97b35a5ccf34f41b5a8642a01d1e55454b699e5e8e7a99b5a3acf5"}, + {file = "multidict-5.2.0-cp310-cp310-manylinux_2_5_i686.manylinux1_i686.manylinux_2_12_i686.manylinux2010_i686.whl", hash = "sha256:df23c83398715b26ab09574217ca21e14694917a0c857e356fd39e1c64f8283f"}, + {file = "multidict-5.2.0-cp310-cp310-manylinux_2_5_x86_64.manylinux1_x86_64.manylinux_2_12_x86_64.manylinux2010_x86_64.whl", hash = "sha256:e58a9b5cc96e014ddf93c2227cbdeca94b56a7eb77300205d6e4001805391747"}, + {file = "multidict-5.2.0-cp310-cp310-musllinux_1_1_aarch64.whl", hash = "sha256:f76440e480c3b2ca7f843ff8a48dc82446b86ed4930552d736c0bac507498a52"}, + {file = "multidict-5.2.0-cp310-cp310-musllinux_1_1_i686.whl", hash = "sha256:cfde464ca4af42a629648c0b0d79b8f295cf5b695412451716531d6916461628"}, + {file = "multidict-5.2.0-cp310-cp310-musllinux_1_1_ppc64le.whl", hash = "sha256:0fed465af2e0eb6357ba95795d003ac0bdb546305cc2366b1fc8f0ad67cc3fda"}, + {file = "multidict-5.2.0-cp310-cp310-musllinux_1_1_s390x.whl", hash = "sha256:b70913cbf2e14275013be98a06ef4b412329fe7b4f83d64eb70dce8269ed1e1a"}, + {file = "multidict-5.2.0-cp310-cp310-musllinux_1_1_x86_64.whl", hash = "sha256:a5635bcf1b75f0f6ef3c8a1ad07b500104a971e38d3683167b9454cb6465ac86"}, + {file = "multidict-5.2.0-cp310-cp310-win32.whl", hash = "sha256:77f0fb7200cc7dedda7a60912f2059086e29ff67cefbc58d2506638c1a9132d7"}, + {file = "multidict-5.2.0-cp310-cp310-win_amd64.whl", hash = "sha256:9416cf11bcd73c861267e88aea71e9fcc35302b3943e45e1dbb4317f91a4b34f"}, + {file = "multidict-5.2.0-cp36-cp36m-macosx_10_9_x86_64.whl", hash = "sha256:fd77c8f3cba815aa69cb97ee2b2ef385c7c12ada9c734b0f3b32e26bb88bbf1d"}, + {file = "multidict-5.2.0-cp36-cp36m-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:98ec9aea6223adf46999f22e2c0ab6cf33f5914be604a404f658386a8f1fba37"}, + {file = "multidict-5.2.0-cp36-cp36m-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:e5283c0a00f48e8cafcecadebfa0ed1dac8b39e295c7248c44c665c16dc1138b"}, + {file = "multidict-5.2.0-cp36-cp36m-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:5f79c19c6420962eb17c7e48878a03053b7ccd7b69f389d5831c0a4a7f1ac0a1"}, + {file = "multidict-5.2.0-cp36-cp36m-manylinux_2_5_i686.manylinux1_i686.manylinux_2_12_i686.manylinux2010_i686.whl", hash = "sha256:e4a67f1080123de76e4e97a18d10350df6a7182e243312426d508712e99988d4"}, + {file = "multidict-5.2.0-cp36-cp36m-manylinux_2_5_x86_64.manylinux1_x86_64.manylinux_2_12_x86_64.manylinux2010_x86_64.whl", hash = "sha256:94b117e27efd8e08b4046c57461d5a114d26b40824995a2eb58372b94f9fca02"}, + {file = "multidict-5.2.0-cp36-cp36m-musllinux_1_1_aarch64.whl", hash = "sha256:2e77282fd1d677c313ffcaddfec236bf23f273c4fba7cdf198108f5940ae10f5"}, + {file = "multidict-5.2.0-cp36-cp36m-musllinux_1_1_i686.whl", hash = "sha256:116347c63ba049c1ea56e157fa8aa6edaf5e92925c9b64f3da7769bdfa012858"}, + {file = "multidict-5.2.0-cp36-cp36m-musllinux_1_1_ppc64le.whl", hash = "sha256:dc3a866cf6c13d59a01878cd806f219340f3e82eed514485e094321f24900677"}, + {file = "multidict-5.2.0-cp36-cp36m-musllinux_1_1_s390x.whl", hash = "sha256:ac42181292099d91217a82e3fa3ce0e0ddf3a74fd891b7c2b347a7f5aa0edded"}, + {file = "multidict-5.2.0-cp36-cp36m-musllinux_1_1_x86_64.whl", hash = "sha256:f0bb0973f42ffcb5e3537548e0767079420aefd94ba990b61cf7bb8d47f4916d"}, + {file = "multidict-5.2.0-cp36-cp36m-win32.whl", hash = "sha256:ea21d4d5104b4f840b91d9dc8cbc832aba9612121eaba503e54eaab1ad140eb9"}, + {file = "multidict-5.2.0-cp36-cp36m-win_amd64.whl", hash = "sha256:e6453f3cbeb78440747096f239d282cc57a2997a16b5197c9bc839099e1633d0"}, + {file = "multidict-5.2.0-cp37-cp37m-macosx_10_9_x86_64.whl", hash = "sha256:d3def943bfd5f1c47d51fd324df1e806d8da1f8e105cc7f1c76a1daf0f7e17b0"}, + {file = "multidict-5.2.0-cp37-cp37m-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:35591729668a303a02b06e8dba0eb8140c4a1bfd4c4b3209a436a02a5ac1de11"}, + {file = "multidict-5.2.0-cp37-cp37m-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:ce8cacda0b679ebc25624d5de66c705bc53dcc7c6f02a7fb0f3ca5e227d80422"}, + {file = "multidict-5.2.0-cp37-cp37m-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:baf1856fab8212bf35230c019cde7c641887e3fc08cadd39d32a421a30151ea3"}, + {file = "multidict-5.2.0-cp37-cp37m-manylinux_2_5_i686.manylinux1_i686.manylinux_2_12_i686.manylinux2010_i686.whl", hash = "sha256:a43616aec0f0d53c411582c451f5d3e1123a68cc7b3475d6f7d97a626f8ff90d"}, + {file = "multidict-5.2.0-cp37-cp37m-manylinux_2_5_x86_64.manylinux1_x86_64.manylinux_2_12_x86_64.manylinux2010_x86_64.whl", hash = "sha256:25cbd39a9029b409167aa0a20d8a17f502d43f2efebfe9e3ac019fe6796c59ac"}, + {file = "multidict-5.2.0-cp37-cp37m-musllinux_1_1_aarch64.whl", hash = "sha256:0a2cbcfbea6dc776782a444db819c8b78afe4db597211298dd8b2222f73e9cd0"}, + {file = "multidict-5.2.0-cp37-cp37m-musllinux_1_1_i686.whl", hash = "sha256:3d2d7d1fff8e09d99354c04c3fd5b560fb04639fd45926b34e27cfdec678a704"}, + {file = "multidict-5.2.0-cp37-cp37m-musllinux_1_1_ppc64le.whl", hash = "sha256:a37e9a68349f6abe24130846e2f1d2e38f7ddab30b81b754e5a1fde32f782b23"}, + {file = "multidict-5.2.0-cp37-cp37m-musllinux_1_1_s390x.whl", hash = "sha256:637c1896497ff19e1ee27c1c2c2ddaa9f2d134bbb5e0c52254361ea20486418d"}, + {file = "multidict-5.2.0-cp37-cp37m-musllinux_1_1_x86_64.whl", hash = "sha256:9815765f9dcda04921ba467957be543423e5ec6a1136135d84f2ae092c50d87b"}, + {file = "multidict-5.2.0-cp37-cp37m-win32.whl", hash = "sha256:8b911d74acdc1fe2941e59b4f1a278a330e9c34c6c8ca1ee21264c51ec9b67ef"}, + {file = "multidict-5.2.0-cp37-cp37m-win_amd64.whl", hash = "sha256:380b868f55f63d048a25931a1632818f90e4be71d2081c2338fcf656d299949a"}, + {file = "multidict-5.2.0-cp38-cp38-macosx_10_9_universal2.whl", hash = "sha256:e7d81ce5744757d2f05fc41896e3b2ae0458464b14b5a2c1e87a6a9d69aefaa8"}, + {file = "multidict-5.2.0-cp38-cp38-macosx_10_9_x86_64.whl", hash = "sha256:2d1d55cdf706ddc62822d394d1df53573d32a7a07d4f099470d3cb9323b721b6"}, + {file = "multidict-5.2.0-cp38-cp38-macosx_11_0_arm64.whl", hash = "sha256:a4771d0d0ac9d9fe9e24e33bed482a13dfc1256d008d101485fe460359476065"}, + {file = "multidict-5.2.0-cp38-cp38-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:da7d57ea65744d249427793c042094c4016789eb2562576fb831870f9c878d9e"}, + {file = "multidict-5.2.0-cp38-cp38-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:cdd68778f96216596218b4e8882944d24a634d984ee1a5a049b300377878fa7c"}, + {file = "multidict-5.2.0-cp38-cp38-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:ecc99bce8ee42dcad15848c7885197d26841cb24fa2ee6e89d23b8993c871c64"}, + {file = "multidict-5.2.0-cp38-cp38-manylinux_2_5_i686.manylinux1_i686.manylinux_2_12_i686.manylinux2010_i686.whl", hash = "sha256:067150fad08e6f2dd91a650c7a49ba65085303fcc3decbd64a57dc13a2733031"}, + {file = "multidict-5.2.0-cp38-cp38-manylinux_2_5_x86_64.manylinux1_x86_64.manylinux_2_12_x86_64.manylinux2010_x86_64.whl", hash = "sha256:78c106b2b506b4d895ddc801ff509f941119394b89c9115580014127414e6c2d"}, + {file = "multidict-5.2.0-cp38-cp38-musllinux_1_1_aarch64.whl", hash = "sha256:e6c4fa1ec16e01e292315ba76eb1d012c025b99d22896bd14a66628b245e3e01"}, + {file = "multidict-5.2.0-cp38-cp38-musllinux_1_1_i686.whl", hash = "sha256:b227345e4186809d31f22087d0265655114af7cda442ecaf72246275865bebe4"}, + {file = "multidict-5.2.0-cp38-cp38-musllinux_1_1_ppc64le.whl", hash = "sha256:06560fbdcf22c9387100979e65b26fba0816c162b888cb65b845d3def7a54c9b"}, + {file = "multidict-5.2.0-cp38-cp38-musllinux_1_1_s390x.whl", hash = "sha256:7878b61c867fb2df7a95e44b316f88d5a3742390c99dfba6c557a21b30180cac"}, + {file = "multidict-5.2.0-cp38-cp38-musllinux_1_1_x86_64.whl", hash = "sha256:246145bff76cc4b19310f0ad28bd0769b940c2a49fc601b86bfd150cbd72bb22"}, + {file = "multidict-5.2.0-cp38-cp38-win32.whl", hash = "sha256:c30ac9f562106cd9e8071c23949a067b10211917fdcb75b4718cf5775356a940"}, + {file = "multidict-5.2.0-cp38-cp38-win_amd64.whl", hash = "sha256:f19001e790013ed580abfde2a4465388950728861b52f0da73e8e8a9418533c0"}, + {file = "multidict-5.2.0-cp39-cp39-macosx_10_9_universal2.whl", hash = "sha256:c1ff762e2ee126e6f1258650ac641e2b8e1f3d927a925aafcfde943b77a36d24"}, + {file = "multidict-5.2.0-cp39-cp39-macosx_10_9_x86_64.whl", hash = "sha256:bd6c9c50bf2ad3f0448edaa1a3b55b2e6866ef8feca5d8dbec10ec7c94371d21"}, + {file = "multidict-5.2.0-cp39-cp39-macosx_11_0_arm64.whl", hash = "sha256:fc66d4016f6e50ed36fb39cd287a3878ffcebfa90008535c62e0e90a7ab713ae"}, + {file = "multidict-5.2.0-cp39-cp39-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:a9acb76d5f3dd9421874923da2ed1e76041cb51b9337fd7f507edde1d86535d6"}, + {file = "multidict-5.2.0-cp39-cp39-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:dfc924a7e946dd3c6360e50e8f750d51e3ef5395c95dc054bc9eab0f70df4f9c"}, + {file = "multidict-5.2.0-cp39-cp39-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:32fdba7333eb2351fee2596b756d730d62b5827d5e1ab2f84e6cbb287cc67fe0"}, + {file = "multidict-5.2.0-cp39-cp39-manylinux_2_5_i686.manylinux1_i686.manylinux_2_12_i686.manylinux2010_i686.whl", hash = "sha256:b9aad49466b8d828b96b9e3630006234879c8d3e2b0a9d99219b3121bc5cdb17"}, + {file = "multidict-5.2.0-cp39-cp39-manylinux_2_5_x86_64.manylinux1_x86_64.manylinux_2_12_x86_64.manylinux2010_x86_64.whl", hash = "sha256:93de39267c4c676c9ebb2057e98a8138bade0d806aad4d864322eee0803140a0"}, + {file = "multidict-5.2.0-cp39-cp39-musllinux_1_1_aarch64.whl", hash = "sha256:f9bef5cff994ca3026fcc90680e326d1a19df9841c5e3d224076407cc21471a1"}, + {file = "multidict-5.2.0-cp39-cp39-musllinux_1_1_i686.whl", hash = "sha256:5f841c4f14331fd1e36cbf3336ed7be2cb2a8f110ce40ea253e5573387db7621"}, + {file = "multidict-5.2.0-cp39-cp39-musllinux_1_1_ppc64le.whl", hash = "sha256:38ba256ee9b310da6a1a0f013ef4e422fca30a685bcbec86a969bd520504e341"}, + {file = "multidict-5.2.0-cp39-cp39-musllinux_1_1_s390x.whl", hash = "sha256:3bc3b1621b979621cee9f7b09f024ec76ec03cc365e638126a056317470bde1b"}, + {file = "multidict-5.2.0-cp39-cp39-musllinux_1_1_x86_64.whl", hash = "sha256:6ee908c070020d682e9b42c8f621e8bb10c767d04416e2ebe44e37d0f44d9ad5"}, + {file = "multidict-5.2.0-cp39-cp39-win32.whl", hash = "sha256:1c7976cd1c157fa7ba5456ae5d31ccdf1479680dc9b8d8aa28afabc370df42b8"}, + {file = "multidict-5.2.0-cp39-cp39-win_amd64.whl", hash = "sha256:c9631c642e08b9fff1c6255487e62971d8b8e821808ddd013d8ac058087591ac"}, + {file = "multidict-5.2.0.tar.gz", hash = "sha256:0dd1c93edb444b33ba2274b66f63def8a327d607c6c790772f448a53b6ea59ce"}, ] mypy = [ {file = "mypy-0.812-cp35-cp35m-macosx_10_9_x86_64.whl", hash = "sha256:a26f8ec704e5a7423c8824d425086705e381b4f1dfdef6e3a1edab7ba174ec49"}, @@ -896,25 +1294,56 @@ nodeenv = [ {file = "nodeenv-1.6.0-py2.py3-none-any.whl", hash = "sha256:621e6b7076565ddcacd2db0294c0381e01fd28945ab36bcf00f41c5daf63bef7"}, {file = "nodeenv-1.6.0.tar.gz", hash = "sha256:3ef13ff90291ba2a4a7a4ff9a979b63ffdd00a464dbe04acf0ea6471517a4c2b"}, ] -openleadr = [ - {file = "openleadr-0.5.23-py3-none-any.whl", hash = "sha256:863522020f7489d67827c69c1b3d6e75531ff75637238c87ecf69a4b08af439c"}, - {file = "openleadr-0.5.23.tar.gz", hash = "sha256:d6a77576120e51db471fcd1f70df50163363989399a058082e43e5a51a0f7c67"}, -] +openleadr = [] packaging = [ - {file = "packaging-20.9-py2.py3-none-any.whl", hash = "sha256:67714da7f7bc052e064859c05c595155bd1ee9f69f76557e21f051443c20947a"}, - {file = "packaging-20.9.tar.gz", hash = "sha256:5b327ac1320dc863dca72f4514ecc086f31186744b84a230374cc1fd776feae5"}, + {file = "packaging-21.0-py3-none-any.whl", hash = "sha256:c86254f9220d55e31cc94d69bade760f0847da8000def4dfe1c6b872fd14ff14"}, + {file = "packaging-21.0.tar.gz", hash = "sha256:7dc96269f53a4ccec5c0670940a4281106dd0bb343f47b7471f779df49c2fbe7"}, ] pathspec = [ - {file = "pathspec-0.8.1-py2.py3-none-any.whl", hash = "sha256:aa0cb481c4041bf52ffa7b0d8fa6cd3e88a2ca4879c533c9153882ee2556790d"}, - {file = "pathspec-0.8.1.tar.gz", hash = "sha256:86379d6b86d75816baba717e64b1a3a3469deb93bb76d613c9ce79edc5cb68fd"}, + {file = "pathspec-0.9.0-py2.py3-none-any.whl", hash = "sha256:7d15c4ddb0b5c802d161efc417ec1a2558ea2653c2e8ad9c19098201dc1c993a"}, + {file = "pathspec-0.9.0.tar.gz", hash = "sha256:e564499435a2673d586f6b2130bb5b95f04a3ba06f81b8f895b651a3c76aabb1"}, +] +platformdirs = [ + {file = "platformdirs-2.4.0-py3-none-any.whl", hash = "sha256:8868bbe3c3c80d42f20156f22e7131d2fb321f5bc86a2a345375c6481a67021d"}, + {file = "platformdirs-2.4.0.tar.gz", hash = "sha256:367a5e80b3d04d2428ffa76d33f124cf11e8fff2acdaa9b43d545f5c7d661ef2"}, ] pluggy = [ - {file = "pluggy-0.13.1-py2.py3-none-any.whl", hash = "sha256:966c145cd83c96502c3c3868f50408687b38434af77734af1e9ca461a4081d2d"}, - {file = "pluggy-0.13.1.tar.gz", hash = "sha256:15b2acde666561e1298d71b523007ed7364de07029219b604cf808bfa1c765b0"}, + {file = "pluggy-1.0.0-py2.py3-none-any.whl", hash = "sha256:74134bbf457f031a36d68416e1509f34bd5ccc019f0bcc952c7b909d06b37bd3"}, + {file = "pluggy-1.0.0.tar.gz", hash = "sha256:4224373bacce55f955a878bf9cfa763c1e360858e330072059e10bad68531159"}, ] pre-commit = [ - {file = "pre_commit-2.13.0-py2.py3-none-any.whl", hash = "sha256:b679d0fddd5b9d6d98783ae5f10fd0c4c59954f375b70a58cbe1ce9bcf9809a4"}, - {file = "pre_commit-2.13.0.tar.gz", hash = "sha256:764972c60693dc668ba8e86eb29654ec3144501310f7198742a767bec385a378"}, + {file = "pre_commit-2.15.0-py2.py3-none-any.whl", hash = "sha256:a4ed01000afcb484d9eb8d504272e642c4c4099bbad3a6b27e519bd6a3e928a6"}, + {file = "pre_commit-2.15.0.tar.gz", hash = "sha256:3c25add78dbdfb6a28a651780d5c311ac40dd17f160eb3954a0c59da40a505a7"}, +] +psutil = [ + {file = "psutil-5.8.0-cp27-cp27m-macosx_10_9_x86_64.whl", hash = "sha256:0066a82f7b1b37d334e68697faba68e5ad5e858279fd6351c8ca6024e8d6ba64"}, + {file = "psutil-5.8.0-cp27-cp27m-manylinux2010_i686.whl", hash = "sha256:0ae6f386d8d297177fd288be6e8d1afc05966878704dad9847719650e44fc49c"}, + {file = "psutil-5.8.0-cp27-cp27m-manylinux2010_x86_64.whl", hash = "sha256:12d844996d6c2b1d3881cfa6fa201fd635971869a9da945cf6756105af73d2df"}, + {file = "psutil-5.8.0-cp27-cp27mu-manylinux2010_i686.whl", hash = "sha256:02b8292609b1f7fcb34173b25e48d0da8667bc85f81d7476584d889c6e0f2131"}, + {file = "psutil-5.8.0-cp27-cp27mu-manylinux2010_x86_64.whl", hash = "sha256:6ffe81843131ee0ffa02c317186ed1e759a145267d54fdef1bc4ea5f5931ab60"}, + {file = "psutil-5.8.0-cp27-none-win32.whl", hash = "sha256:ea313bb02e5e25224e518e4352af4bf5e062755160f77e4b1767dd5ccb65f876"}, + {file = "psutil-5.8.0-cp27-none-win_amd64.whl", hash = "sha256:5da29e394bdedd9144c7331192e20c1f79283fb03b06e6abd3a8ae45ffecee65"}, + {file = "psutil-5.8.0-cp36-cp36m-macosx_10_9_x86_64.whl", hash = "sha256:74fb2557d1430fff18ff0d72613c5ca30c45cdbfcddd6a5773e9fc1fe9364be8"}, + {file = "psutil-5.8.0-cp36-cp36m-manylinux2010_i686.whl", hash = "sha256:74f2d0be88db96ada78756cb3a3e1b107ce8ab79f65aa885f76d7664e56928f6"}, + {file = "psutil-5.8.0-cp36-cp36m-manylinux2010_x86_64.whl", hash = "sha256:99de3e8739258b3c3e8669cb9757c9a861b2a25ad0955f8e53ac662d66de61ac"}, + {file = "psutil-5.8.0-cp36-cp36m-win32.whl", hash = "sha256:36b3b6c9e2a34b7d7fbae330a85bf72c30b1c827a4366a07443fc4b6270449e2"}, + {file = "psutil-5.8.0-cp36-cp36m-win_amd64.whl", hash = "sha256:52de075468cd394ac98c66f9ca33b2f54ae1d9bff1ef6b67a212ee8f639ec06d"}, + {file = "psutil-5.8.0-cp37-cp37m-macosx_10_9_x86_64.whl", hash = "sha256:c6a5fd10ce6b6344e616cf01cc5b849fa8103fbb5ba507b6b2dee4c11e84c935"}, + {file = "psutil-5.8.0-cp37-cp37m-manylinux2010_i686.whl", hash = "sha256:61f05864b42fedc0771d6d8e49c35f07efd209ade09a5afe6a5059e7bb7bf83d"}, + {file = "psutil-5.8.0-cp37-cp37m-manylinux2010_x86_64.whl", hash = "sha256:0dd4465a039d343925cdc29023bb6960ccf4e74a65ad53e768403746a9207023"}, + {file = "psutil-5.8.0-cp37-cp37m-win32.whl", hash = "sha256:1bff0d07e76114ec24ee32e7f7f8d0c4b0514b3fae93e3d2aaafd65d22502394"}, + {file = "psutil-5.8.0-cp37-cp37m-win_amd64.whl", hash = "sha256:fcc01e900c1d7bee2a37e5d6e4f9194760a93597c97fee89c4ae51701de03563"}, + {file = "psutil-5.8.0-cp38-cp38-macosx_10_9_x86_64.whl", hash = "sha256:6223d07a1ae93f86451d0198a0c361032c4c93ebd4bf6d25e2fb3edfad9571ef"}, + {file = "psutil-5.8.0-cp38-cp38-manylinux2010_i686.whl", hash = "sha256:d225cd8319aa1d3c85bf195c4e07d17d3cd68636b8fc97e6cf198f782f99af28"}, + {file = "psutil-5.8.0-cp38-cp38-manylinux2010_x86_64.whl", hash = "sha256:28ff7c95293ae74bf1ca1a79e8805fcde005c18a122ca983abf676ea3466362b"}, + {file = "psutil-5.8.0-cp38-cp38-win32.whl", hash = "sha256:ce8b867423291cb65cfc6d9c4955ee9bfc1e21fe03bb50e177f2b957f1c2469d"}, + {file = "psutil-5.8.0-cp38-cp38-win_amd64.whl", hash = "sha256:90f31c34d25b1b3ed6c40cdd34ff122b1887a825297c017e4cbd6796dd8b672d"}, + {file = "psutil-5.8.0-cp39-cp39-macosx_10_9_x86_64.whl", hash = "sha256:6323d5d845c2785efb20aded4726636546b26d3b577aded22492908f7c1bdda7"}, + {file = "psutil-5.8.0-cp39-cp39-manylinux2010_i686.whl", hash = "sha256:245b5509968ac0bd179287d91210cd3f37add77dad385ef238b275bad35fa1c4"}, + {file = "psutil-5.8.0-cp39-cp39-manylinux2010_x86_64.whl", hash = "sha256:90d4091c2d30ddd0a03e0b97e6a33a48628469b99585e2ad6bf21f17423b112b"}, + {file = "psutil-5.8.0-cp39-cp39-win32.whl", hash = "sha256:ea372bcc129394485824ae3e3ddabe67dc0b118d262c568b4d2602a7070afdb0"}, + {file = "psutil-5.8.0-cp39-cp39-win_amd64.whl", hash = "sha256:f4634b033faf0d968bb9220dd1c793b897ab7f1189956e1aa9eae752527127d3"}, + {file = "psutil-5.8.0.tar.gz", hash = "sha256:0c9ccb99ab76025f2f0bbecf341d4656e9c1351db8cc8a03ccd62e318ab4b5c6"}, ] py = [ {file = "py-1.10.0-py2.py3-none-any.whl", hash = "sha256:3b80836aa6d1feeaa108e046da6423ab8f6ceda6468545ae8d02d9d58d18818a"}, @@ -925,20 +1354,24 @@ pycparser = [ {file = "pycparser-2.20.tar.gz", hash = "sha256:2d475327684562c3a96cc71adf7dc8c4f0565175cf86b6d7a404ff4c771f15f0"}, ] pyopenssl = [ - {file = "pyOpenSSL-19.1.0-py2.py3-none-any.whl", hash = "sha256:621880965a720b8ece2f1b2f54ea2071966ab00e2970ad2ce11d596102063504"}, - {file = "pyOpenSSL-19.1.0.tar.gz", hash = "sha256:9a24494b2602aaf402be5c9e30a0b82d4a5c67528fe8fb475e3f3bc00dd69507"}, + {file = "pyOpenSSL-19.0.0-py2.py3-none-any.whl", hash = "sha256:c727930ad54b10fc157015014b666f2d8b41f70c0d03e83ab67624fd3dd5d1e6"}, + {file = "pyOpenSSL-19.0.0.tar.gz", hash = "sha256:aeca66338f6de19d1aa46ed634c3b9ae519a64b458f8468aec688e7e3c20f200"}, ] pyparsing = [ {file = "pyparsing-2.4.7-py2.py3-none-any.whl", hash = "sha256:ef9d7589ef3c200abe66653d3f1ab1033c3c419ae9b9bdb1240a85b024efc88b"}, {file = "pyparsing-2.4.7.tar.gz", hash = "sha256:c203ec8783bf771a155b207279b9bccb8dea02d8f0c9e5f8ead507bc3246ecc1"}, ] pytest = [ - {file = "pytest-6.2.4-py3-none-any.whl", hash = "sha256:91ef2131a9bd6be8f76f1f08eac5c5317221d6ad1e143ae03894b862e8976890"}, - {file = "pytest-6.2.4.tar.gz", hash = "sha256:50bcad0a0b9c5a72c8e4e7c9855a3ad496ca6a881a3641b4260605450772c54b"}, + {file = "pytest-6.2.5-py3-none-any.whl", hash = "sha256:7310f8d27bc79ced999e760ca304d69f6ba6c6649c0b60fb0e04a4a77cacc134"}, + {file = "pytest-6.2.5.tar.gz", hash = "sha256:131b36680866a76e6781d13f101efb86cf674ebb9762eb70d3082b6f29889e89"}, +] +python-dateutil = [ + {file = "python-dateutil-2.8.2.tar.gz", hash = "sha256:0123cacc1627ae19ddf3c27a5de5bd67ee4586fbdd6440d9748f8abb483d3e86"}, + {file = "python_dateutil-2.8.2-py2.py3-none-any.whl", hash = "sha256:961d03dc3453ebbc59dbdea9e4e11c5651520a876d0f4db161e8674aae935da9"}, ] pytz = [ - {file = "pytz-2021.1-py2.py3-none-any.whl", hash = "sha256:eb10ce3e7736052ed3623d49975ce333bcd712c7bb19a58b9e2089d4057d0798"}, - {file = "pytz-2021.1.tar.gz", hash = "sha256:83a4a90894bf38e243cf052c8b58f381bfe9a7a483f6a9cab140bc7f702ac4da"}, + {file = "pytz-2021.3-py2.py3-none-any.whl", hash = "sha256:3672058bc3453457b622aab7a1c3bfd5ab0bdae451512f6cf25f64ed37f5b87c"}, + {file = "pytz-2021.3.tar.gz", hash = "sha256:acad2d8b20a1af07d4e4c9d2e9285c5ed9104354062f275f3fcd88dcef4f1326"}, ] pyyaml = [ {file = "PyYAML-5.4.1-cp27-cp27m-macosx_10_9_x86_64.whl", hash = "sha256:3b2b1824fe7112845700f815ff6a489360226a5609b96ec2190a45e62a9fc922"}, @@ -971,48 +1404,86 @@ pyyaml = [ {file = "PyYAML-5.4.1-cp39-cp39-win_amd64.whl", hash = "sha256:c20cfa2d49991c8b4147af39859b167664f2ad4561704ee74c1de03318e898db"}, {file = "PyYAML-5.4.1.tar.gz", hash = "sha256:607774cbba28732bfa802b54baa7484215f530991055bb562efbed5b2f20a45e"}, ] +pyzmq = [ + {file = "pyzmq-22.3.0-cp310-cp310-macosx_10_15_universal2.whl", hash = "sha256:6b217b8f9dfb6628f74b94bdaf9f7408708cb02167d644edca33f38746ca12dd"}, + {file = "pyzmq-22.3.0-cp310-cp310-macosx_10_9_x86_64.whl", hash = "sha256:2841997a0d85b998cbafecb4183caf51fd19c4357075dfd33eb7efea57e4c149"}, + {file = "pyzmq-22.3.0-cp310-cp310-manylinux_2_12_i686.manylinux2010_i686.whl", hash = "sha256:f89468059ebc519a7acde1ee50b779019535db8dcf9b8c162ef669257fef7a93"}, + {file = "pyzmq-22.3.0-cp310-cp310-manylinux_2_12_x86_64.manylinux2010_x86_64.whl", hash = "sha256:ea12133df25e3a6918718fbb9a510c6ee5d3fdd5a346320421aac3882f4feeea"}, + {file = "pyzmq-22.3.0-cp310-cp310-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:76c532fd68b93998aab92356be280deec5de8f8fe59cd28763d2cc8a58747b7f"}, + {file = "pyzmq-22.3.0-cp310-cp310-win32.whl", hash = "sha256:67db33bea0a29d03e6eeec55a8190e033318cee3cbc732ba8fd939617cbf762d"}, + {file = "pyzmq-22.3.0-cp310-cp310-win_amd64.whl", hash = "sha256:7661fc1d5cb73481cf710a1418a4e1e301ed7d5d924f91c67ba84b2a1b89defd"}, + {file = "pyzmq-22.3.0-cp36-cp36m-macosx_10_9_x86_64.whl", hash = "sha256:79244b9e97948eaf38695f4b8e6fc63b14b78cc37f403c6642ba555517ac1268"}, + {file = "pyzmq-22.3.0-cp36-cp36m-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:ab888624ed68930442a3f3b0b921ad7439c51ba122dbc8c386e6487a658e4a4e"}, + {file = "pyzmq-22.3.0-cp36-cp36m-manylinux_2_5_i686.manylinux1_i686.whl", hash = "sha256:18cd854b423fce44951c3a4d3e686bac8f1243d954f579e120a1714096637cc0"}, + {file = "pyzmq-22.3.0-cp36-cp36m-manylinux_2_5_x86_64.manylinux1_x86_64.whl", hash = "sha256:de8df0684398bd74ad160afdc2a118ca28384ac6f5e234eb0508858d8d2d9364"}, + {file = "pyzmq-22.3.0-cp36-cp36m-win32.whl", hash = "sha256:3c1895c95be92600233e476fe283f042e71cf8f0b938aabf21b7aafa62a8dac9"}, + {file = "pyzmq-22.3.0-cp36-cp36m-win_amd64.whl", hash = "sha256:851977788b9caa8ed011f5f643d3ee8653af02c5fc723fa350db5125abf2be7b"}, + {file = "pyzmq-22.3.0-cp37-cp37m-macosx_10_9_x86_64.whl", hash = "sha256:b4ebed0977f92320f6686c96e9e8dd29eed199eb8d066936bac991afc37cbb70"}, + {file = "pyzmq-22.3.0-cp37-cp37m-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:42abddebe2c6a35180ca549fadc7228d23c1e1f76167c5ebc8a936b5804ea2df"}, + {file = "pyzmq-22.3.0-cp37-cp37m-manylinux_2_5_i686.manylinux1_i686.whl", hash = "sha256:c1e41b32d6f7f9c26bc731a8b529ff592f31fc8b6ef2be9fa74abd05c8a342d7"}, + {file = "pyzmq-22.3.0-cp37-cp37m-manylinux_2_5_x86_64.manylinux1_x86_64.whl", hash = "sha256:be4e0f229cf3a71f9ecd633566bd6f80d9fa6afaaff5489492be63fe459ef98c"}, + {file = "pyzmq-22.3.0-cp37-cp37m-win32.whl", hash = "sha256:7c58f598d9fcc52772b89a92d72bf8829c12d09746a6d2c724c5b30076c1f11d"}, + {file = "pyzmq-22.3.0-cp37-cp37m-win_amd64.whl", hash = "sha256:2b97502c16a5ec611cd52410bdfaab264997c627a46b0f98d3f666227fd1ea2d"}, + {file = "pyzmq-22.3.0-cp38-cp38-macosx_10_9_x86_64.whl", hash = "sha256:d728b08448e5ac3e4d886b165385a262883c34b84a7fe1166277fe675e1c197a"}, + {file = "pyzmq-22.3.0-cp38-cp38-manylinux_2_12_i686.manylinux2010_i686.whl", hash = "sha256:480b9931bfb08bf8b094edd4836271d4d6b44150da051547d8c7113bf947a8b0"}, + {file = "pyzmq-22.3.0-cp38-cp38-manylinux_2_12_x86_64.manylinux2010_x86_64.whl", hash = "sha256:7dc09198e4073e6015d9a8ea093fc348d4e59de49382476940c3dd9ae156fba8"}, + {file = "pyzmq-22.3.0-cp38-cp38-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:0ca6cd58f62a2751728016d40082008d3b3412a7f28ddfb4a2f0d3c130f69e74"}, + {file = "pyzmq-22.3.0-cp38-cp38-win32.whl", hash = "sha256:c0f84360dcca3481e8674393bdf931f9f10470988f87311b19d23cda869bb6b7"}, + {file = "pyzmq-22.3.0-cp38-cp38-win_amd64.whl", hash = "sha256:f762442bab706fd874064ca218b33a1d8e40d4938e96c24dafd9b12e28017f45"}, + {file = "pyzmq-22.3.0-cp39-cp39-macosx_10_15_universal2.whl", hash = "sha256:954e73c9cd4d6ae319f1c936ad159072b6d356a92dcbbabfd6e6204b9a79d356"}, + {file = "pyzmq-22.3.0-cp39-cp39-macosx_10_9_x86_64.whl", hash = "sha256:f43b4a2e6218371dd4f41e547bd919ceeb6ebf4abf31a7a0669cd11cd91ea973"}, + {file = "pyzmq-22.3.0-cp39-cp39-manylinux_2_12_i686.manylinux2010_i686.whl", hash = "sha256:acebba1a23fb9d72b42471c3771b6f2f18dcd46df77482612054bd45c07dfa36"}, + {file = "pyzmq-22.3.0-cp39-cp39-manylinux_2_12_x86_64.manylinux2010_x86_64.whl", hash = "sha256:cf98fd7a6c8aaa08dbc699ffae33fd71175696d78028281bc7b832b26f00ca57"}, + {file = "pyzmq-22.3.0-cp39-cp39-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:d072f7dfbdb184f0786d63bda26e8a0882041b1e393fbe98940395f7fab4c5e2"}, + {file = "pyzmq-22.3.0-cp39-cp39-win32.whl", hash = "sha256:e6a02cf7271ee94674a44f4e62aa061d2d049001c844657740e156596298b70b"}, + {file = "pyzmq-22.3.0-cp39-cp39-win_amd64.whl", hash = "sha256:d3dcb5548ead4f1123851a5ced467791f6986d68c656bc63bfff1bf9e36671e2"}, + {file = "pyzmq-22.3.0-pp37-pypy37_pp73-macosx_10_9_x86_64.whl", hash = "sha256:3a4c9886d61d386b2b493377d980f502186cd71d501fffdba52bd2a0880cef4f"}, + {file = "pyzmq-22.3.0-pp37-pypy37_pp73-manylinux_2_12_i686.manylinux2010_i686.whl", hash = "sha256:80e043a89c6cadefd3a0712f8a1322038e819ebe9dbac7eca3bce1721bcb63bf"}, + {file = "pyzmq-22.3.0-pp37-pypy37_pp73-manylinux_2_12_x86_64.manylinux2010_x86_64.whl", hash = "sha256:1621e7a2af72cced1f6ec8ca8ca91d0f76ac236ab2e8828ac8fe909512d566cb"}, + {file = "pyzmq-22.3.0-pp37-pypy37_pp73-win_amd64.whl", hash = "sha256:d6157793719de168b199194f6b6173f0ccd3bf3499e6870fac17086072e39115"}, + {file = "pyzmq-22.3.0.tar.gz", hash = "sha256:8eddc033e716f8c91c6a2112f0a8ebc5e00532b4a6ae1eb0ccc48e027f9c671c"}, +] regex = [ - {file = "regex-2021.4.4-cp36-cp36m-macosx_10_9_x86_64.whl", hash = "sha256:619d71c59a78b84d7f18891fe914446d07edd48dc8328c8e149cbe0929b4e000"}, - {file = "regex-2021.4.4-cp36-cp36m-manylinux1_i686.whl", hash = "sha256:47bf5bf60cf04d72bf6055ae5927a0bd9016096bf3d742fa50d9bf9f45aa0711"}, - {file = "regex-2021.4.4-cp36-cp36m-manylinux1_x86_64.whl", hash = "sha256:281d2fd05555079448537fe108d79eb031b403dac622621c78944c235f3fcf11"}, - {file = "regex-2021.4.4-cp36-cp36m-manylinux2010_i686.whl", hash = "sha256:bd28bc2e3a772acbb07787c6308e00d9626ff89e3bfcdebe87fa5afbfdedf968"}, - {file = "regex-2021.4.4-cp36-cp36m-manylinux2010_x86_64.whl", hash = "sha256:7c2a1af393fcc09e898beba5dd59196edaa3116191cc7257f9224beaed3e1aa0"}, - {file = "regex-2021.4.4-cp36-cp36m-manylinux2014_aarch64.whl", hash = "sha256:c38c71df845e2aabb7fb0b920d11a1b5ac8526005e533a8920aea97efb8ec6a4"}, - {file = "regex-2021.4.4-cp36-cp36m-manylinux2014_i686.whl", hash = "sha256:96fcd1888ab4d03adfc9303a7b3c0bd78c5412b2bfbe76db5b56d9eae004907a"}, - {file = "regex-2021.4.4-cp36-cp36m-manylinux2014_x86_64.whl", hash = "sha256:ade17eb5d643b7fead300a1641e9f45401c98eee23763e9ed66a43f92f20b4a7"}, - {file = "regex-2021.4.4-cp36-cp36m-win32.whl", hash = "sha256:e8e5b509d5c2ff12f8418006d5a90e9436766133b564db0abaec92fd27fcee29"}, - {file = "regex-2021.4.4-cp36-cp36m-win_amd64.whl", hash = "sha256:11d773d75fa650cd36f68d7ca936e3c7afaae41b863b8c387a22aaa78d3c5c79"}, - {file = "regex-2021.4.4-cp37-cp37m-macosx_10_9_x86_64.whl", hash = "sha256:d3029c340cfbb3ac0a71798100ccc13b97dddf373a4ae56b6a72cf70dfd53bc8"}, - {file = "regex-2021.4.4-cp37-cp37m-manylinux1_i686.whl", hash = "sha256:18c071c3eb09c30a264879f0d310d37fe5d3a3111662438889ae2eb6fc570c31"}, - {file = "regex-2021.4.4-cp37-cp37m-manylinux1_x86_64.whl", hash = "sha256:4c557a7b470908b1712fe27fb1ef20772b78079808c87d20a90d051660b1d69a"}, - {file = "regex-2021.4.4-cp37-cp37m-manylinux2010_i686.whl", hash = "sha256:01afaf2ec48e196ba91b37451aa353cb7eda77efe518e481707e0515025f0cd5"}, - {file = "regex-2021.4.4-cp37-cp37m-manylinux2010_x86_64.whl", hash = "sha256:3a9cd17e6e5c7eb328517969e0cb0c3d31fd329298dd0c04af99ebf42e904f82"}, - {file = "regex-2021.4.4-cp37-cp37m-manylinux2014_aarch64.whl", hash = "sha256:90f11ff637fe8798933fb29f5ae1148c978cccb0452005bf4c69e13db951e765"}, - {file = "regex-2021.4.4-cp37-cp37m-manylinux2014_i686.whl", hash = "sha256:919859aa909429fb5aa9cf8807f6045592c85ef56fdd30a9a3747e513db2536e"}, - {file = "regex-2021.4.4-cp37-cp37m-manylinux2014_x86_64.whl", hash = "sha256:339456e7d8c06dd36a22e451d58ef72cef293112b559010db3d054d5560ef439"}, - {file = "regex-2021.4.4-cp37-cp37m-win32.whl", hash = "sha256:67bdb9702427ceddc6ef3dc382455e90f785af4c13d495f9626861763ee13f9d"}, - {file = "regex-2021.4.4-cp37-cp37m-win_amd64.whl", hash = "sha256:32e65442138b7b76dd8173ffa2cf67356b7bc1768851dded39a7a13bf9223da3"}, - {file = "regex-2021.4.4-cp38-cp38-macosx_10_9_x86_64.whl", hash = "sha256:1e1c20e29358165242928c2de1482fb2cf4ea54a6a6dea2bd7a0e0d8ee321500"}, - {file = "regex-2021.4.4-cp38-cp38-manylinux1_i686.whl", hash = "sha256:314d66636c494ed9c148a42731b3834496cc9a2c4251b1661e40936814542b14"}, - {file = "regex-2021.4.4-cp38-cp38-manylinux1_x86_64.whl", hash = "sha256:6d1b01031dedf2503631d0903cb563743f397ccaf6607a5e3b19a3d76fc10480"}, - {file = "regex-2021.4.4-cp38-cp38-manylinux2010_i686.whl", hash = "sha256:741a9647fcf2e45f3a1cf0e24f5e17febf3efe8d4ba1281dcc3aa0459ef424dc"}, - {file = "regex-2021.4.4-cp38-cp38-manylinux2010_x86_64.whl", hash = "sha256:4c46e22a0933dd783467cf32b3516299fb98cfebd895817d685130cc50cd1093"}, - {file = "regex-2021.4.4-cp38-cp38-manylinux2014_aarch64.whl", hash = "sha256:e512d8ef5ad7b898cdb2d8ee1cb09a8339e4f8be706d27eaa180c2f177248a10"}, - {file = "regex-2021.4.4-cp38-cp38-manylinux2014_i686.whl", hash = "sha256:980d7be47c84979d9136328d882f67ec5e50008681d94ecc8afa8a65ed1f4a6f"}, - {file = "regex-2021.4.4-cp38-cp38-manylinux2014_x86_64.whl", hash = "sha256:ce15b6d103daff8e9fee13cf7f0add05245a05d866e73926c358e871221eae87"}, - {file = "regex-2021.4.4-cp38-cp38-win32.whl", hash = "sha256:a91aa8619b23b79bcbeb37abe286f2f408d2f2d6f29a17237afda55bb54e7aac"}, - {file = "regex-2021.4.4-cp38-cp38-win_amd64.whl", hash = "sha256:c0502c0fadef0d23b128605d69b58edb2c681c25d44574fc673b0e52dce71ee2"}, - {file = "regex-2021.4.4-cp39-cp39-macosx_10_9_x86_64.whl", hash = "sha256:598585c9f0af8374c28edd609eb291b5726d7cbce16be6a8b95aa074d252ee17"}, - {file = "regex-2021.4.4-cp39-cp39-manylinux1_i686.whl", hash = "sha256:ee54ff27bf0afaf4c3b3a62bcd016c12c3fdb4ec4f413391a90bd38bc3624605"}, - {file = "regex-2021.4.4-cp39-cp39-manylinux1_x86_64.whl", hash = "sha256:7d9884d86dd4dd489e981d94a65cd30d6f07203d90e98f6f657f05170f6324c9"}, - {file = "regex-2021.4.4-cp39-cp39-manylinux2010_i686.whl", hash = "sha256:bf5824bfac591ddb2c1f0a5f4ab72da28994548c708d2191e3b87dd207eb3ad7"}, - {file = "regex-2021.4.4-cp39-cp39-manylinux2010_x86_64.whl", hash = "sha256:563085e55b0d4fb8f746f6a335893bda5c2cef43b2f0258fe1020ab1dd874df8"}, - {file = "regex-2021.4.4-cp39-cp39-manylinux2014_aarch64.whl", hash = "sha256:b9c3db21af35e3b3c05764461b262d6f05bbca08a71a7849fd79d47ba7bc33ed"}, - {file = "regex-2021.4.4-cp39-cp39-manylinux2014_i686.whl", hash = "sha256:3916d08be28a1149fb97f7728fca1f7c15d309a9f9682d89d79db75d5e52091c"}, - {file = "regex-2021.4.4-cp39-cp39-manylinux2014_x86_64.whl", hash = "sha256:fd45ff9293d9274c5008a2054ecef86a9bfe819a67c7be1afb65e69b405b3042"}, - {file = "regex-2021.4.4-cp39-cp39-win32.whl", hash = "sha256:fa4537fb4a98fe8fde99626e4681cc644bdcf2a795038533f9f711513a862ae6"}, - {file = "regex-2021.4.4-cp39-cp39-win_amd64.whl", hash = "sha256:97f29f57d5b84e73fbaf99ab3e26134e6687348e95ef6b48cfd2c06807005a07"}, - {file = "regex-2021.4.4.tar.gz", hash = "sha256:52ba3d3f9b942c49d7e4bc105bb28551c44065f139a65062ab7912bef10c9afb"}, + {file = "regex-2021.10.21-cp310-cp310-macosx_10_9_x86_64.whl", hash = "sha256:edff4e31d159672a7b9d70164b21289e4b53b239ce1dc945bf9643d266537573"}, + {file = "regex-2021.10.21-cp310-cp310-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:d6432daf42f2c487b357e1aa0bdc43193f050ff53a3188bfab20b88202b53027"}, + {file = "regex-2021.10.21-cp310-cp310-manylinux_2_5_i686.manylinux1_i686.manylinux_2_12_i686.manylinux2010_i686.whl", hash = "sha256:468de52dd3f20187ab5ca4fd265c1bea61a5346baef01ad0333a5e89fa9fad29"}, + {file = "regex-2021.10.21-cp310-cp310-manylinux_2_5_i686.manylinux1_i686.manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:c5a2ac760f2fc13a1c58131ec217779911890899ce1a0a63c9409bd23fecde6f"}, + {file = "regex-2021.10.21-cp310-cp310-manylinux_2_5_x86_64.manylinux1_x86_64.manylinux_2_12_x86_64.manylinux2010_x86_64.whl", hash = "sha256:ad1fedca001fefc3030d1e9022b038af429e58dc06a7e9c55e40bd1f834582ec"}, + {file = "regex-2021.10.21-cp310-cp310-win32.whl", hash = "sha256:9c613d797a3790f6b12e78a61e1cd29df7fc88135218467cf8b0891353292b9c"}, + {file = "regex-2021.10.21-cp310-cp310-win_amd64.whl", hash = "sha256:678d9a4ce79e1eaa4ebe88bc9769df52919eb30c597576a0deba1f3cf2360e65"}, + {file = "regex-2021.10.21-cp36-cp36m-macosx_10_9_x86_64.whl", hash = "sha256:2044174af237bb9c56ecc07294cf38623ee379e8dca14b01e970f8b015c71917"}, + {file = "regex-2021.10.21-cp36-cp36m-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:98743a2d827a135bf3390452be18d95839b947a099734d53c17e09a64fc09480"}, + {file = "regex-2021.10.21-cp36-cp36m-manylinux_2_5_i686.manylinux1_i686.manylinux_2_12_i686.manylinux2010_i686.whl", hash = "sha256:f1b23304855303bd97b5954edab63b8ddd56c91c41c6d4eba408228c0bae95f3"}, + {file = "regex-2021.10.21-cp36-cp36m-manylinux_2_5_i686.manylinux1_i686.manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:19c4fd59747236423016ccd89b9a6485d958bf1aa7a8a902a6ba28029107a87f"}, + {file = "regex-2021.10.21-cp36-cp36m-manylinux_2_5_x86_64.manylinux1_x86_64.manylinux_2_12_x86_64.manylinux2010_x86_64.whl", hash = "sha256:130a002fa386c976615a2f6d6dff0fcc25da24858994a36b14d2e3129dce7de2"}, + {file = "regex-2021.10.21-cp36-cp36m-win32.whl", hash = "sha256:8bd83d9b8ee125350cd666b55294f4bc9993c4f0d9b1be9344a318d0762e94cc"}, + {file = "regex-2021.10.21-cp36-cp36m-win_amd64.whl", hash = "sha256:98fe0e1b07a314f0a86dc58af4e717c379d48a403eddd8d966ab9b8bf91ce164"}, + {file = "regex-2021.10.21-cp37-cp37m-macosx_10_9_x86_64.whl", hash = "sha256:ded4748c7be6f31fb207387ee83a3a0f625e700defe32f268cb1d350ed6e4a66"}, + {file = "regex-2021.10.21-cp37-cp37m-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:f3da121de36a9ead0f32b44ea720ee8c87edbb59dca6bb980d18377d84ad58a3"}, + {file = "regex-2021.10.21-cp37-cp37m-manylinux_2_5_i686.manylinux1_i686.manylinux_2_12_i686.manylinux2010_i686.whl", hash = "sha256:b9dfba513eae785e3d868803f5a7e21a032cb2b038fa4a1ea7ec691037426ad3"}, + {file = "regex-2021.10.21-cp37-cp37m-manylinux_2_5_i686.manylinux1_i686.manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:2ff91696888755e96230138355cbe8ce2965d930d967d6cff7c636082d038c78"}, + {file = "regex-2021.10.21-cp37-cp37m-manylinux_2_5_x86_64.manylinux1_x86_64.manylinux_2_12_x86_64.manylinux2010_x86_64.whl", hash = "sha256:0f82de529d7595011a40573cc0f27422e59cafa94943b64a4d17d966d75f2c01"}, + {file = "regex-2021.10.21-cp37-cp37m-win32.whl", hash = "sha256:164e51ace4d00f07c519f85ec2209e8faaeab18bc77be6b35685c18d4ac1c22a"}, + {file = "regex-2021.10.21-cp37-cp37m-win_amd64.whl", hash = "sha256:e39eafa854e469d7225066c806c76b9a0acba5ff5ce36c82c0224b75e24888f2"}, + {file = "regex-2021.10.21-cp38-cp38-macosx_10_9_x86_64.whl", hash = "sha256:740a28580520b099b804776db1e919360fcbf30a734a14c5985d5e39a39e7237"}, + {file = "regex-2021.10.21-cp38-cp38-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:de7dbf72ae80f06e79444ff9614fb5e3a7956645d513b0e12d1bbe6f3ccebd11"}, + {file = "regex-2021.10.21-cp38-cp38-manylinux_2_5_i686.manylinux1_i686.manylinux_2_12_i686.manylinux2010_i686.whl", hash = "sha256:dc1a9bedf389bf3d3627a4d2b21cbdc5fe5e0f029d1f465972f4437833dcc946"}, + {file = "regex-2021.10.21-cp38-cp38-manylinux_2_5_i686.manylinux1_i686.manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:9cd14f22425beecf727f6dbdf5c893e46ecbc5ff16197c16a6f38a9066f2d4d5"}, + {file = "regex-2021.10.21-cp38-cp38-manylinux_2_5_x86_64.manylinux1_x86_64.manylinux_2_12_x86_64.manylinux2010_x86_64.whl", hash = "sha256:5b75a3db3aab0bfa51b6af3f820760779d360eb79f59e32c88c7fba648990b4f"}, + {file = "regex-2021.10.21-cp38-cp38-win32.whl", hash = "sha256:f68c71aabb10b1352a06515e25a425a703ba85660ae04cf074da5eb91c0af5e5"}, + {file = "regex-2021.10.21-cp38-cp38-win_amd64.whl", hash = "sha256:c0f49f1f03be3e4a5faaadc35db7afa2b83a871943b889f9f7bba56e0e2e8bd5"}, + {file = "regex-2021.10.21-cp39-cp39-macosx_10_9_x86_64.whl", hash = "sha256:201890fdc8a65396cfb6aa4493201353b2a6378e27d2de65234446f8329233cb"}, + {file = "regex-2021.10.21-cp39-cp39-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:fd1bfc6b7347de9f0ae1fb6f9080426bed6a9ca55b5766fa4fdf7b3a29ccae9c"}, + {file = "regex-2021.10.21-cp39-cp39-manylinux_2_5_i686.manylinux1_i686.manylinux_2_12_i686.manylinux2010_i686.whl", hash = "sha256:72a0b98d41c4508ed23a96eef41090f78630b44ba746e28cd621ecbe961e0a16"}, + {file = "regex-2021.10.21-cp39-cp39-manylinux_2_5_i686.manylinux1_i686.manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:3b5a0660a63b0703380758a7141b96cc1c1a13dee2b8e9c280a2522962fd12af"}, + {file = "regex-2021.10.21-cp39-cp39-manylinux_2_5_x86_64.manylinux1_x86_64.manylinux_2_12_x86_64.manylinux2010_x86_64.whl", hash = "sha256:f82d3adde46ac9188db3aa7e6e1690865ebb6448d245df5a3ea22284f70d9e46"}, + {file = "regex-2021.10.21-cp39-cp39-win32.whl", hash = "sha256:bc4637390235f1e3e2fcdd3e904ca0b42aa655ae28a78072248b2992b4ad4c08"}, + {file = "regex-2021.10.21-cp39-cp39-win_amd64.whl", hash = "sha256:74d03c256cf0aed81997e87be8e24297b5792c9718f3a735f5055ddfad392f06"}, + {file = "regex-2021.10.21.tar.gz", hash = "sha256:4832736b3f24617e63dc919ce8c4215680ba94250a5d9e710fcc0c5f457b5028"}, +] +requests = [ + {file = "requests-2.26.0-py2.py3-none-any.whl", hash = "sha256:6c1246513ecd5ecd4528a0906f910e8f0f9c6b8ec72030dc9fd154dc1a6efd24"}, + {file = "requests-2.26.0.tar.gz", hash = "sha256:b8aa58f8cf793ffd8782d3d8cb19e66ef36f7aba4353eec859e74678b01b07a7"}, ] signxml-openadr = [ {file = "signxml-openadr-2.9.1.tar.gz", hash = "sha256:93bf41824247a36fa09d8ac23435f5675aaa9e1e58fcb3809a549fba9a905c23"}, @@ -1026,6 +1497,10 @@ toml = [ {file = "toml-0.10.2-py2.py3-none-any.whl", hash = "sha256:806143ae5bfb6a3c6e736a764057db0e6a0e05e338b5630894a5f779cabb4f9b"}, {file = "toml-0.10.2.tar.gz", hash = "sha256:b3bda1d108d5dd99f4a20d24d9c348e91c4db7ab1b749200bded2f839ccbe68f"}, ] +tomli = [ + {file = "tomli-1.2.1-py3-none-any.whl", hash = "sha256:8dd0e9524d6f386271a36b41dbf6c57d8e32fd96fd22b6584679dc569d20899f"}, + {file = "tomli-1.2.1.tar.gz", hash = "sha256:a5b75cb6f3968abb47af1b40c1819dc519ea82bcc065776a866e8d74c5ca9442"}, +] typed-ast = [ {file = "typed_ast-1.4.3-cp35-cp35m-manylinux1_i686.whl", hash = "sha256:2068531575a125b87a41802130fa7e29f26c09a2833fea68d9a40cf33902eba6"}, {file = "typed_ast-1.4.3-cp35-cp35m-manylinux1_x86_64.whl", hash = "sha256:c907f561b1e83e93fad565bac5ba9c22d96a54e7ea0267c708bffe863cbe4075"}, @@ -1059,58 +1534,195 @@ typed-ast = [ {file = "typed_ast-1.4.3.tar.gz", hash = "sha256:fb1bbeac803adea29cedd70781399c99138358c26d05fcbd23c13016b7f5ec65"}, ] typing-extensions = [ - {file = "typing_extensions-3.10.0.0-py2-none-any.whl", hash = "sha256:0ac0f89795dd19de6b97debb0c6af1c70987fd80a2d62d1958f7e56fcc31b497"}, - {file = "typing_extensions-3.10.0.0-py3-none-any.whl", hash = "sha256:779383f6086d90c99ae41cf0ff39aac8a7937a9283ce0a414e5dd782f4c94a84"}, - {file = "typing_extensions-3.10.0.0.tar.gz", hash = "sha256:50b6f157849174217d0656f99dc82fe932884fb250826c18350e159ec6cdf342"}, + {file = "typing_extensions-3.10.0.2-py2-none-any.whl", hash = "sha256:d8226d10bc02a29bcc81df19a26e56a9647f8b0a6d4a83924139f4a8b01f17b7"}, + {file = "typing_extensions-3.10.0.2-py3-none-any.whl", hash = "sha256:f1d25edafde516b146ecd0613dabcc61409817af4766fbbcfb8d1ad4ec441a34"}, + {file = "typing_extensions-3.10.0.2.tar.gz", hash = "sha256:49f75d16ff11f1cd258e1b988ccff82a3ca5570217d7ad8c5f48205dd99a677e"}, ] tzlocal = [ {file = "tzlocal-2.1-py2.py3-none-any.whl", hash = "sha256:e2cb6c6b5b604af38597403e9852872d7f534962ae2954c7f35efcb1ccacf4a4"}, {file = "tzlocal-2.1.tar.gz", hash = "sha256:643c97c5294aedc737780a49d9df30889321cbe1204eac2c2ec6134035a92e44"}, ] +urllib3 = [ + {file = "urllib3-1.26.7-py2.py3-none-any.whl", hash = "sha256:c4fdf4019605b6e5423637e01bc9fe4daef873709a7973e195ceba0a62bbc844"}, + {file = "urllib3-1.26.7.tar.gz", hash = "sha256:4987c65554f7a2dbf30c18fd48778ef124af6fab771a377103da0585e2336ece"}, +] virtualenv = [ - {file = "virtualenv-20.4.7-py2.py3-none-any.whl", hash = "sha256:2b0126166ea7c9c3661f5b8e06773d28f83322de7a3ff7d06f0aed18c9de6a76"}, - {file = "virtualenv-20.4.7.tar.gz", hash = "sha256:14fdf849f80dbb29a4eb6caa9875d476ee2a5cf76a5f5415fa2f1606010ab467"}, + {file = "virtualenv-20.8.1-py2.py3-none-any.whl", hash = "sha256:10062e34c204b5e4ec5f62e6ef2473f8ba76513a9a617e873f1f8fb4a519d300"}, + {file = "virtualenv-20.8.1.tar.gz", hash = "sha256:bcc17f0b3a29670dd777d6f0755a4c04f28815395bca279cdcb213b97199a6b8"}, +] +volttron-client = [ + {file = "volttron-client-0.3.4.tar.gz", hash = "sha256:a426efd8c5d10c5f574b78501082c3ac76714903c29d4fe9c8c9d335abc03d47"}, + {file = "volttron_client-0.3.4-py3-none-any.whl", hash = "sha256:60ba299b75431e5d249ff0cf871e1978ab75e65814a28b6252a5f7612f1ec658"}, +] +volttron-utils = [ + {file = "volttron-utils-0.3.2.tar.gz", hash = "sha256:abb35d335ba60ecc06e9e1bbcbb5dd0b3dbc9445e58a366dfaa6071e4ce5ce80"}, + {file = "volttron_utils-0.3.2-py3-none-any.whl", hash = "sha256:240e6f84ef164d7740b0c035b01efbd00c194c1b5b12bb3b73fe28386fff6f5f"}, +] +watchdog = [ + {file = "watchdog-2.1.6-cp310-cp310-macosx_10_9_universal2.whl", hash = "sha256:9693f35162dc6208d10b10ddf0458cc09ad70c30ba689d9206e02cd836ce28a3"}, + {file = "watchdog-2.1.6-cp310-cp310-macosx_10_9_x86_64.whl", hash = "sha256:aba5c812f8ee8a3ff3be51887ca2d55fb8e268439ed44110d3846e4229eb0e8b"}, + {file = "watchdog-2.1.6-cp310-cp310-macosx_11_0_arm64.whl", hash = "sha256:4ae38bf8ba6f39d5b83f78661273216e7db5b00f08be7592062cb1fc8b8ba542"}, + {file = "watchdog-2.1.6-cp36-cp36m-macosx_10_9_x86_64.whl", hash = "sha256:ad6f1796e37db2223d2a3f302f586f74c72c630b48a9872c1e7ae8e92e0ab669"}, + {file = "watchdog-2.1.6-cp37-cp37m-macosx_10_9_x86_64.whl", hash = "sha256:922a69fa533cb0c793b483becaaa0845f655151e7256ec73630a1b2e9ebcb660"}, + {file = "watchdog-2.1.6-cp38-cp38-macosx_10_9_universal2.whl", hash = "sha256:b2fcf9402fde2672545b139694284dc3b665fd1be660d73eca6805197ef776a3"}, + {file = "watchdog-2.1.6-cp38-cp38-macosx_10_9_x86_64.whl", hash = "sha256:3386b367e950a11b0568062b70cc026c6f645428a698d33d39e013aaeda4cc04"}, + {file = "watchdog-2.1.6-cp38-cp38-macosx_11_0_arm64.whl", hash = "sha256:8f1c00aa35f504197561060ca4c21d3cc079ba29cf6dd2fe61024c70160c990b"}, + {file = "watchdog-2.1.6-cp39-cp39-macosx_10_9_universal2.whl", hash = "sha256:b52b88021b9541a60531142b0a451baca08d28b74a723d0c99b13c8c8d48d604"}, + {file = "watchdog-2.1.6-cp39-cp39-macosx_10_9_x86_64.whl", hash = "sha256:8047da932432aa32c515ec1447ea79ce578d0559362ca3605f8e9568f844e3c6"}, + {file = "watchdog-2.1.6-cp39-cp39-macosx_11_0_arm64.whl", hash = "sha256:e92c2d33858c8f560671b448205a268096e17870dcf60a9bb3ac7bfbafb7f5f9"}, + {file = "watchdog-2.1.6-pp37-pypy37_pp73-macosx_10_9_x86_64.whl", hash = "sha256:b7d336912853d7b77f9b2c24eeed6a5065d0a0cc0d3b6a5a45ad6d1d05fb8cd8"}, + {file = "watchdog-2.1.6-py3-none-manylinux2014_aarch64.whl", hash = "sha256:cca7741c0fcc765568350cb139e92b7f9f3c9a08c4f32591d18ab0a6ac9e71b6"}, + {file = "watchdog-2.1.6-py3-none-manylinux2014_armv7l.whl", hash = "sha256:25fb5240b195d17de949588628fdf93032ebf163524ef08933db0ea1f99bd685"}, + {file = "watchdog-2.1.6-py3-none-manylinux2014_i686.whl", hash = "sha256:be9be735f827820a06340dff2ddea1fb7234561fa5e6300a62fe7f54d40546a0"}, + {file = "watchdog-2.1.6-py3-none-manylinux2014_ppc64.whl", hash = "sha256:d0d19fb2441947b58fbf91336638c2b9f4cc98e05e1045404d7a4cb7cddc7a65"}, + {file = "watchdog-2.1.6-py3-none-manylinux2014_ppc64le.whl", hash = "sha256:3becdb380d8916c873ad512f1701f8a92ce79ec6978ffde92919fd18d41da7fb"}, + {file = "watchdog-2.1.6-py3-none-manylinux2014_s390x.whl", hash = "sha256:ae67501c95606072aafa865b6ed47343ac6484472a2f95490ba151f6347acfc2"}, + {file = "watchdog-2.1.6-py3-none-manylinux2014_x86_64.whl", hash = "sha256:e0f30db709c939cabf64a6dc5babb276e6d823fd84464ab916f9b9ba5623ca15"}, + {file = "watchdog-2.1.6-py3-none-win32.whl", hash = "sha256:e02794ac791662a5eafc6ffeaf9bcc149035a0e48eb0a9d40a8feb4622605a3d"}, + {file = "watchdog-2.1.6-py3-none-win_amd64.whl", hash = "sha256:bd9ba4f332cf57b2c1f698be0728c020399ef3040577cde2939f2e045b39c1e5"}, + {file = "watchdog-2.1.6-py3-none-win_ia64.whl", hash = "sha256:a0f1c7edf116a12f7245be06120b1852275f9506a7d90227648b250755a03923"}, + {file = "watchdog-2.1.6.tar.gz", hash = "sha256:a36e75df6c767cbf46f61a91c70b3ba71811dfa0aca4a324d9407a06a8b7a2e7"}, +] +watchdog-gevent = [ + {file = "watchdog_gevent-0.1.1-py3-none-any.whl", hash = "sha256:dcebd07668b472790ad0e7a7b40ad8c365e156c40cf54ffd5499c65c65c4f66f"}, + {file = "watchdog_gevent-0.1.1.tar.gz", hash = "sha256:71dced5ec451b3deafd1dc73baaf52a7ce4d54d9a1eb606f9370235125f2424a"}, ] xmltodict = [ {file = "xmltodict-0.12.0-py2.py3-none-any.whl", hash = "sha256:8bbcb45cc982f48b2ca8fe7e7827c5d792f217ecf1792626f808bf41c3b86051"}, {file = "xmltodict-0.12.0.tar.gz", hash = "sha256:50d8c638ed7ecb88d90561beedbf720c9b4e851a9fa6c47ebd64e99d166d8a21"}, ] yarl = [ - {file = "yarl-1.6.3-cp36-cp36m-macosx_10_14_x86_64.whl", hash = "sha256:0355a701b3998dcd832d0dc47cc5dedf3874f966ac7f870e0f3a6788d802d434"}, - {file = "yarl-1.6.3-cp36-cp36m-manylinux1_i686.whl", hash = "sha256:bafb450deef6861815ed579c7a6113a879a6ef58aed4c3a4be54400ae8871478"}, - {file = "yarl-1.6.3-cp36-cp36m-manylinux2014_aarch64.whl", hash = "sha256:547f7665ad50fa8563150ed079f8e805e63dd85def6674c97efd78eed6c224a6"}, - {file = "yarl-1.6.3-cp36-cp36m-manylinux2014_i686.whl", hash = "sha256:63f90b20ca654b3ecc7a8d62c03ffa46999595f0167d6450fa8383bab252987e"}, - {file = "yarl-1.6.3-cp36-cp36m-manylinux2014_ppc64le.whl", hash = "sha256:97b5bdc450d63c3ba30a127d018b866ea94e65655efaf889ebeabc20f7d12406"}, - {file = "yarl-1.6.3-cp36-cp36m-manylinux2014_s390x.whl", hash = "sha256:d8d07d102f17b68966e2de0e07bfd6e139c7c02ef06d3a0f8d2f0f055e13bb76"}, - {file = "yarl-1.6.3-cp36-cp36m-manylinux2014_x86_64.whl", hash = "sha256:15263c3b0b47968c1d90daa89f21fcc889bb4b1aac5555580d74565de6836366"}, - {file = "yarl-1.6.3-cp36-cp36m-win32.whl", hash = "sha256:b5dfc9a40c198334f4f3f55880ecf910adebdcb2a0b9a9c23c9345faa9185721"}, - {file = "yarl-1.6.3-cp36-cp36m-win_amd64.whl", hash = "sha256:b2e9a456c121e26d13c29251f8267541bd75e6a1ccf9e859179701c36a078643"}, - {file = "yarl-1.6.3-cp37-cp37m-macosx_10_14_x86_64.whl", hash = "sha256:ce3beb46a72d9f2190f9e1027886bfc513702d748047b548b05dab7dfb584d2e"}, - {file = "yarl-1.6.3-cp37-cp37m-manylinux1_i686.whl", hash = "sha256:2ce4c621d21326a4a5500c25031e102af589edb50c09b321049e388b3934eec3"}, - {file = "yarl-1.6.3-cp37-cp37m-manylinux2014_aarch64.whl", hash = "sha256:d26608cf178efb8faa5ff0f2d2e77c208f471c5a3709e577a7b3fd0445703ac8"}, - {file = "yarl-1.6.3-cp37-cp37m-manylinux2014_i686.whl", hash = "sha256:4c5bcfc3ed226bf6419f7a33982fb4b8ec2e45785a0561eb99274ebbf09fdd6a"}, - {file = "yarl-1.6.3-cp37-cp37m-manylinux2014_ppc64le.whl", hash = "sha256:4736eaee5626db8d9cda9eb5282028cc834e2aeb194e0d8b50217d707e98bb5c"}, - {file = "yarl-1.6.3-cp37-cp37m-manylinux2014_s390x.whl", hash = "sha256:68dc568889b1c13f1e4745c96b931cc94fdd0defe92a72c2b8ce01091b22e35f"}, - {file = "yarl-1.6.3-cp37-cp37m-manylinux2014_x86_64.whl", hash = "sha256:7356644cbed76119d0b6bd32ffba704d30d747e0c217109d7979a7bc36c4d970"}, - {file = "yarl-1.6.3-cp37-cp37m-win32.whl", hash = "sha256:00d7ad91b6583602eb9c1d085a2cf281ada267e9a197e8b7cae487dadbfa293e"}, - {file = "yarl-1.6.3-cp37-cp37m-win_amd64.whl", hash = "sha256:69ee97c71fee1f63d04c945f56d5d726483c4762845400a6795a3b75d56b6c50"}, - {file = "yarl-1.6.3-cp38-cp38-macosx_10_14_x86_64.whl", hash = "sha256:e46fba844f4895b36f4c398c5af062a9808d1f26b2999c58909517384d5deda2"}, - {file = "yarl-1.6.3-cp38-cp38-manylinux1_i686.whl", hash = "sha256:31ede6e8c4329fb81c86706ba8f6bf661a924b53ba191b27aa5fcee5714d18ec"}, - {file = "yarl-1.6.3-cp38-cp38-manylinux2014_aarch64.whl", hash = "sha256:fcbb48a93e8699eae920f8d92f7160c03567b421bc17362a9ffbbd706a816f71"}, - {file = "yarl-1.6.3-cp38-cp38-manylinux2014_i686.whl", hash = "sha256:72a660bdd24497e3e84f5519e57a9ee9220b6f3ac4d45056961bf22838ce20cc"}, - {file = "yarl-1.6.3-cp38-cp38-manylinux2014_ppc64le.whl", hash = "sha256:324ba3d3c6fee56e2e0b0d09bf5c73824b9f08234339d2b788af65e60040c959"}, - {file = "yarl-1.6.3-cp38-cp38-manylinux2014_s390x.whl", hash = "sha256:e6b5460dc5ad42ad2b36cca524491dfcaffbfd9c8df50508bddc354e787b8dc2"}, - {file = "yarl-1.6.3-cp38-cp38-manylinux2014_x86_64.whl", hash = "sha256:6d6283d8e0631b617edf0fd726353cb76630b83a089a40933043894e7f6721e2"}, - {file = "yarl-1.6.3-cp38-cp38-win32.whl", hash = "sha256:9ede61b0854e267fd565e7527e2f2eb3ef8858b301319be0604177690e1a3896"}, - {file = "yarl-1.6.3-cp38-cp38-win_amd64.whl", hash = "sha256:f0b059678fd549c66b89bed03efcabb009075bd131c248ecdf087bdb6faba24a"}, - {file = "yarl-1.6.3-cp39-cp39-macosx_10_14_x86_64.whl", hash = "sha256:329412812ecfc94a57cd37c9d547579510a9e83c516bc069470db5f75684629e"}, - {file = "yarl-1.6.3-cp39-cp39-manylinux1_i686.whl", hash = "sha256:c49ff66d479d38ab863c50f7bb27dee97c6627c5fe60697de15529da9c3de724"}, - {file = "yarl-1.6.3-cp39-cp39-manylinux2014_aarch64.whl", hash = "sha256:f040bcc6725c821a4c0665f3aa96a4d0805a7aaf2caf266d256b8ed71b9f041c"}, - {file = "yarl-1.6.3-cp39-cp39-manylinux2014_i686.whl", hash = "sha256:d5c32c82990e4ac4d8150fd7652b972216b204de4e83a122546dce571c1bdf25"}, - {file = "yarl-1.6.3-cp39-cp39-manylinux2014_ppc64le.whl", hash = "sha256:d597767fcd2c3dc49d6eea360c458b65643d1e4dbed91361cf5e36e53c1f8c96"}, - {file = "yarl-1.6.3-cp39-cp39-manylinux2014_s390x.whl", hash = "sha256:8aa3decd5e0e852dc68335abf5478a518b41bf2ab2f330fe44916399efedfae0"}, - {file = "yarl-1.6.3-cp39-cp39-manylinux2014_x86_64.whl", hash = "sha256:73494d5b71099ae8cb8754f1df131c11d433b387efab7b51849e7e1e851f07a4"}, - {file = "yarl-1.6.3-cp39-cp39-win32.whl", hash = "sha256:5b883e458058f8d6099e4420f0cc2567989032b5f34b271c0827de9f1079a424"}, - {file = "yarl-1.6.3-cp39-cp39-win_amd64.whl", hash = "sha256:4953fb0b4fdb7e08b2f3b3be80a00d28c5c8a2056bb066169de00e6501b986b6"}, - {file = "yarl-1.6.3.tar.gz", hash = "sha256:8a9066529240171b68893d60dca86a763eae2139dd42f42106b03cf4b426bf10"}, + {file = "yarl-1.7.0-cp310-cp310-macosx_10_9_universal2.whl", hash = "sha256:e35d8230e4b08d86ea65c32450533b906a8267a87b873f2954adeaecede85169"}, + {file = "yarl-1.7.0-cp310-cp310-macosx_10_9_x86_64.whl", hash = "sha256:eb4b3f277880c314e47720b4b6bb2c85114ab3c04c5442c9bc7006b3787904d8"}, + {file = "yarl-1.7.0-cp310-cp310-macosx_11_0_arm64.whl", hash = "sha256:c7015dcedb91d90a138eebdc7e432aec8966e0147ab2a55f2df27b1904fa7291"}, + {file = "yarl-1.7.0-cp310-cp310-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:bb3e478175e15e00d659fb0354a6a8db71a7811a2a5052aed98048bc972e5d2b"}, + {file = "yarl-1.7.0-cp310-cp310-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:8b8c409aa3a7966647e7c1c524846b362a6bcbbe120bf8a176431f940d2b9a2e"}, + {file = "yarl-1.7.0-cp310-cp310-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:b22ea41c7e98170474a01e3eded1377d46b2dfaef45888a0005c683eaaa49285"}, + {file = "yarl-1.7.0-cp310-cp310-manylinux_2_5_i686.manylinux1_i686.manylinux_2_12_i686.manylinux2010_i686.whl", hash = "sha256:a7dfc46add4cfe5578013dbc4127893edc69fe19132d2836ff2f6e49edc5ecd6"}, + {file = "yarl-1.7.0-cp310-cp310-manylinux_2_5_x86_64.manylinux1_x86_64.manylinux_2_12_x86_64.manylinux2010_x86_64.whl", hash = "sha256:82ff6f85f67500a4f74885d81659cd270eb24dfe692fe44e622b8a2fd57e7279"}, + {file = "yarl-1.7.0-cp310-cp310-musllinux_1_1_aarch64.whl", hash = "sha256:f3cd2158b2ed0fb25c6811adfdcc47224efe075f2d68a750071dacc03a7a66e4"}, + {file = "yarl-1.7.0-cp310-cp310-musllinux_1_1_i686.whl", hash = "sha256:59c0f13f9592820c51280d1cf811294d753e4a18baf90f0139d1dc93d4b6fc5f"}, + {file = "yarl-1.7.0-cp310-cp310-musllinux_1_1_ppc64le.whl", hash = "sha256:7f7655ad83d1a8afa48435a449bf2f3009293da1604f5dd95b5ddcf5f673bd69"}, + {file = "yarl-1.7.0-cp310-cp310-musllinux_1_1_s390x.whl", hash = "sha256:aa9f0d9b62d15182341b3e9816582f46182cab91c1a57b2d308b9a3c4e2c4f78"}, + {file = "yarl-1.7.0-cp310-cp310-musllinux_1_1_x86_64.whl", hash = "sha256:fdd1b90c225a653b1bd1c0cae8edf1957892b9a09c8bf7ee6321eeb8208eac0f"}, + {file = "yarl-1.7.0-cp310-cp310-win32.whl", hash = "sha256:7c8d0bb76eabc5299db203e952ec55f8f4c53f08e0df4285aac8c92bd9e12675"}, + {file = "yarl-1.7.0-cp310-cp310-win_amd64.whl", hash = "sha256:622a36fa779efb4ff9eff5fe52730ff17521431379851a31e040958fc251670c"}, + {file = "yarl-1.7.0-cp36-cp36m-macosx_10_9_x86_64.whl", hash = "sha256:3d461b7a8e139b9e4b41f62eb417ffa0b98d1c46d4caf14c845e6a3b349c0bb1"}, + {file = "yarl-1.7.0-cp36-cp36m-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:81cfacdd1e40bc931b5519499342efa388d24d262c30a3d31187bfa04f4a7001"}, + {file = "yarl-1.7.0-cp36-cp36m-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:821b978f2152be7695d4331ef0621d207aedf9bbd591ba23a63412a3efc29a01"}, + {file = "yarl-1.7.0-cp36-cp36m-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:b64bd24c8c9a487f4a12260dc26732bf41028816dbf0c458f17864fbebdb3131"}, + {file = "yarl-1.7.0-cp36-cp36m-manylinux_2_5_i686.manylinux1_i686.manylinux_2_12_i686.manylinux2010_i686.whl", hash = "sha256:98c9ddb92b60a83c21be42c776d3d9d5ec632a762a094c41bda37b7dfbd2cd83"}, + {file = "yarl-1.7.0-cp36-cp36m-manylinux_2_5_x86_64.manylinux1_x86_64.manylinux_2_12_x86_64.manylinux2010_x86_64.whl", hash = "sha256:a532d75ca74431c053a88a802e161fb3d651b8bf5821a3440bc3616e38754583"}, + {file = "yarl-1.7.0-cp36-cp36m-musllinux_1_1_aarch64.whl", hash = "sha256:053e09817eafb892e94e172d05406c1b3a22a93bc68f6eff5198363a3d764459"}, + {file = "yarl-1.7.0-cp36-cp36m-musllinux_1_1_i686.whl", hash = "sha256:98c51f02d542945d306c8e934aa2c1e66ba5e9c1c86b5bf37f3a51c8a747067e"}, + {file = "yarl-1.7.0-cp36-cp36m-musllinux_1_1_ppc64le.whl", hash = "sha256:15ec41a5a5fdb7bace6d7b16701f9440007a82734f69127c0fbf6d87e10f4a1e"}, + {file = "yarl-1.7.0-cp36-cp36m-musllinux_1_1_s390x.whl", hash = "sha256:a7f08819dba1e1255d6991ed37448a1bf4b1352c004bcd899b9da0c47958513d"}, + {file = "yarl-1.7.0-cp36-cp36m-musllinux_1_1_x86_64.whl", hash = "sha256:8e3ffab21db0542ffd1887f3b9575ddd58961f2cf61429cb6458afc00c4581e0"}, + {file = "yarl-1.7.0-cp36-cp36m-win32.whl", hash = "sha256:50127634f519b2956005891507e3aa4ac345f66a7ea7bbc2d7dcba7401f41898"}, + {file = "yarl-1.7.0-cp36-cp36m-win_amd64.whl", hash = "sha256:36ec44f15193f6d5288d42ebb8e751b967ebdfb72d6830983838d45ab18edb4f"}, + {file = "yarl-1.7.0-cp37-cp37m-macosx_10_9_x86_64.whl", hash = "sha256:ec1b5a25a25c880c976d0bb3d107def085bb08dbb3db7f4442e0a2b980359d24"}, + {file = "yarl-1.7.0-cp37-cp37m-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:b36f5a63c891f813c6f04ef19675b382efc190fd5ce7e10ab19386d2548bca06"}, + {file = "yarl-1.7.0-cp37-cp37m-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:38173b8c3a29945e7ecade9a3f6ff39581eee8201338ee6a2c8882db5df3e806"}, + {file = "yarl-1.7.0-cp37-cp37m-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:8ba402f32184f0b405fb281b93bd0d8ab7e3257735b57b62a6ed2e94cdf4fe50"}, + {file = "yarl-1.7.0-cp37-cp37m-manylinux_2_5_i686.manylinux1_i686.manylinux_2_12_i686.manylinux2010_i686.whl", hash = "sha256:be52bc5208d767cdd8308a9e93059b3b36d1e048fecbea0e0346d0d24a76adc0"}, + {file = "yarl-1.7.0-cp37-cp37m-manylinux_2_5_x86_64.manylinux1_x86_64.manylinux_2_12_x86_64.manylinux2010_x86_64.whl", hash = "sha256:08c2044a956f4ef30405f2f433ce77f1f57c2c773bf81ae43201917831044d5a"}, + {file = "yarl-1.7.0-cp37-cp37m-musllinux_1_1_aarch64.whl", hash = "sha256:484d61c047c45670ef5967653a1d0783e232c54bf9dd786a7737036828fa8d54"}, + {file = "yarl-1.7.0-cp37-cp37m-musllinux_1_1_i686.whl", hash = "sha256:b7de92a4af85cfcaf4081f8aa6165b1d63ee5de150af3ee85f954145f93105a7"}, + {file = "yarl-1.7.0-cp37-cp37m-musllinux_1_1_ppc64le.whl", hash = "sha256:376e41775aab79c5575534924a386c8e0f1a5d91db69fc6133fd27a489bcaf10"}, + {file = "yarl-1.7.0-cp37-cp37m-musllinux_1_1_s390x.whl", hash = "sha256:8a8b10d0e7bac154f959b709fcea593cda527b234119311eb950096653816a86"}, + {file = "yarl-1.7.0-cp37-cp37m-musllinux_1_1_x86_64.whl", hash = "sha256:f46cd4c43e6175030e2a56def8f1d83b64e6706eeb2bb9ab0ef4756f65eab23f"}, + {file = "yarl-1.7.0-cp37-cp37m-win32.whl", hash = "sha256:b28cfb46140efe1a6092b8c5c4994a1fe70dc83c38fbcea4992401e0c6fb9cce"}, + {file = "yarl-1.7.0-cp37-cp37m-win_amd64.whl", hash = "sha256:9624154ec9c02a776802da1086eed7f5034bd1971977f5146233869c2ac80297"}, + {file = "yarl-1.7.0-cp38-cp38-macosx_10_9_universal2.whl", hash = "sha256:69945d13e1bbf81784a9bc48824feb9cd66491e6a503d4e83f6cd7c7cc861361"}, + {file = "yarl-1.7.0-cp38-cp38-macosx_10_9_x86_64.whl", hash = "sha256:46a742ed9e363bd01be64160ce7520e92e11989bd4cb224403cfd31c101cc83d"}, + {file = "yarl-1.7.0-cp38-cp38-macosx_11_0_arm64.whl", hash = "sha256:cb4ff1ac7cb4500f43581b3f4cbd627d702143aa6be1fdc1fa3ebffaf4dc1be5"}, + {file = "yarl-1.7.0-cp38-cp38-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:3ad51e17cd65ea3debb0e10f0120cf8dd987c741fe423ed2285087368090b33d"}, + {file = "yarl-1.7.0-cp38-cp38-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:7e37786ea89a5d3ffbbf318ea9790926f8dfda83858544f128553c347ad143c6"}, + {file = "yarl-1.7.0-cp38-cp38-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:c63c1e208f800daad71715786bfeb1cecdc595d87e2e9b1cd234fd6e597fd71d"}, + {file = "yarl-1.7.0-cp38-cp38-manylinux_2_5_i686.manylinux1_i686.manylinux_2_12_i686.manylinux2010_i686.whl", hash = "sha256:91cbe24300c11835ef186436363352b3257db7af165e0a767f4f17aa25761388"}, + {file = "yarl-1.7.0-cp38-cp38-manylinux_2_5_x86_64.manylinux1_x86_64.manylinux_2_12_x86_64.manylinux2010_x86_64.whl", hash = "sha256:e510dbec7c59d32eaa61ffa48173d5e3d7170a67f4a03e8f5e2e9e3971aca622"}, + {file = "yarl-1.7.0-cp38-cp38-musllinux_1_1_aarch64.whl", hash = "sha256:3def6e681cc02397e5d8141ee97b41d02932b2bcf0fb34532ad62855eab7c60e"}, + {file = "yarl-1.7.0-cp38-cp38-musllinux_1_1_i686.whl", hash = "sha256:263c81b94e6431942b27f6f671fa62f430a0a5c14bb255f2ab69eeb9b2b66ff7"}, + {file = "yarl-1.7.0-cp38-cp38-musllinux_1_1_ppc64le.whl", hash = "sha256:e78c91faefe88d601ddd16e3882918dbde20577a2438e2320f8239c8b7507b8f"}, + {file = "yarl-1.7.0-cp38-cp38-musllinux_1_1_s390x.whl", hash = "sha256:22b2430c49713bfb2f0a0dd4a8d7aab218b28476ba86fd1c78ad8899462cbcf2"}, + {file = "yarl-1.7.0-cp38-cp38-musllinux_1_1_x86_64.whl", hash = "sha256:2e7ad9db939082f5d0b9269cfd92c025cb8f2fbbb1f1b9dc5a393c639db5bd92"}, + {file = "yarl-1.7.0-cp38-cp38-win32.whl", hash = "sha256:3a31e4a8dcb1beaf167b7e7af61b88cb961b220db8d3ba1c839723630e57eef7"}, + {file = "yarl-1.7.0-cp38-cp38-win_amd64.whl", hash = "sha256:d579957439933d752358c6a300c93110f84aae67b63dd0c19dde6ecbf4056f6b"}, + {file = "yarl-1.7.0-cp39-cp39-macosx_10_9_universal2.whl", hash = "sha256:87721b549505a546eb003252185103b5ec8147de6d3ad3714d148a5a67b6fe53"}, + {file = "yarl-1.7.0-cp39-cp39-macosx_10_9_x86_64.whl", hash = "sha256:a1fa866fa24d9f4108f9e58ea8a2135655419885cdb443e36b39a346e1181532"}, + {file = "yarl-1.7.0-cp39-cp39-macosx_11_0_arm64.whl", hash = "sha256:1d3b8449dfedfe94eaff2b77954258b09b24949f6818dfa444b05dbb05ae1b7e"}, + {file = "yarl-1.7.0-cp39-cp39-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:db2372e350794ce8b9f810feb094c606b7e0e4aa6807141ac4fadfe5ddd75bb0"}, + {file = "yarl-1.7.0-cp39-cp39-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:a06d9d0b9a97fa99b84fee71d9dd11e69e21ac8a27229089f07b5e5e50e8d63c"}, + {file = "yarl-1.7.0-cp39-cp39-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:a3455c2456d6307bcfa80bc1157b8603f7d93573291f5bdc7144489ca0df4628"}, + {file = "yarl-1.7.0-cp39-cp39-manylinux_2_5_i686.manylinux1_i686.manylinux_2_12_i686.manylinux2010_i686.whl", hash = "sha256:d30d67e3486aea61bb2cbf7cf81385364c2e4f7ce7469a76ed72af76a5cdfe6b"}, + {file = "yarl-1.7.0-cp39-cp39-manylinux_2_5_x86_64.manylinux1_x86_64.manylinux_2_12_x86_64.manylinux2010_x86_64.whl", hash = "sha256:c18a4b286e8d780c3a40c31d7b79836aa93b720f71d5743f20c08b7e049ca073"}, + {file = "yarl-1.7.0-cp39-cp39-musllinux_1_1_aarch64.whl", hash = "sha256:d54c925396e7891666cabc0199366ca55b27d003393465acef63fd29b8b7aa92"}, + {file = "yarl-1.7.0-cp39-cp39-musllinux_1_1_i686.whl", hash = "sha256:64773840952de17851a1c7346ad7f71688c77e74248d1f0bc230e96680f84028"}, + {file = "yarl-1.7.0-cp39-cp39-musllinux_1_1_ppc64le.whl", hash = "sha256:acbf1756d9dc7cd0ae943d883be72e84e04396f6c2ff93a6ddeca929d562039f"}, + {file = "yarl-1.7.0-cp39-cp39-musllinux_1_1_s390x.whl", hash = "sha256:2e48f27936aa838939c798f466c851ba4ae79e347e8dfce43b009c64b930df12"}, + {file = "yarl-1.7.0-cp39-cp39-musllinux_1_1_x86_64.whl", hash = "sha256:1beef4734ca1ad40a9d8c6b20a76ab46e3a2ed09f38561f01e4aa2ea82cafcef"}, + {file = "yarl-1.7.0-cp39-cp39-win32.whl", hash = "sha256:8ee78c9a5f3c642219d4607680a4693b59239c27a3aa608b64ef79ddc9698039"}, + {file = "yarl-1.7.0-cp39-cp39-win_amd64.whl", hash = "sha256:d750503682605088a14d29a4701548c15c510da4f13c8b17409c4097d5b04c52"}, + {file = "yarl-1.7.0.tar.gz", hash = "sha256:8e7ebaf62e19c2feb097ffb7c94deb0f0c9fab52590784c8cd679d30ab009162"}, +] +zmq = [ + {file = "zmq-0.0.0.tar.gz", hash = "sha256:6b1a1de53338646e8c8405803cffb659e8eb7bb02fff4c9be62a7acfac8370c9"}, + {file = "zmq-0.0.0.zip", hash = "sha256:21cfc6be254c9bc25e4dabb8a3b2006a4227966b7b39a637426084c8dc6901f7"}, +] +"zope.event" = [ + {file = "zope.event-4.5.0-py2.py3-none-any.whl", hash = "sha256:2666401939cdaa5f4e0c08cf7f20c9b21423b95e88f4675b1443973bdb080c42"}, + {file = "zope.event-4.5.0.tar.gz", hash = "sha256:5e76517f5b9b119acf37ca8819781db6c16ea433f7e2062c4afc2b6fbedb1330"}, +] +"zope.interface" = [ + {file = "zope.interface-5.4.0-cp27-cp27m-macosx_10_14_x86_64.whl", hash = "sha256:7df1e1c05304f26faa49fa752a8c690126cf98b40b91d54e6e9cc3b7d6ffe8b7"}, + {file = "zope.interface-5.4.0-cp27-cp27m-manylinux1_i686.whl", hash = "sha256:2c98384b254b37ce50eddd55db8d381a5c53b4c10ee66e1e7fe749824f894021"}, + {file = "zope.interface-5.4.0-cp27-cp27m-manylinux1_x86_64.whl", hash = "sha256:08f9636e99a9d5410181ba0729e0408d3d8748026ea938f3b970a0249daa8192"}, + {file = "zope.interface-5.4.0-cp27-cp27m-manylinux2010_i686.whl", hash = "sha256:0ea1d73b7c9dcbc5080bb8aaffb776f1c68e807767069b9ccdd06f27a161914a"}, + {file = "zope.interface-5.4.0-cp27-cp27m-manylinux2010_x86_64.whl", hash = "sha256:273f158fabc5ea33cbc936da0ab3d4ba80ede5351babc4f577d768e057651531"}, + {file = "zope.interface-5.4.0-cp27-cp27m-win32.whl", hash = "sha256:a1e6e96217a0f72e2b8629e271e1b280c6fa3fe6e59fa8f6701bec14e3354325"}, + {file = "zope.interface-5.4.0-cp27-cp27m-win_amd64.whl", hash = "sha256:877473e675fdcc113c138813a5dd440da0769a2d81f4d86614e5d62b69497155"}, + {file = "zope.interface-5.4.0-cp27-cp27mu-manylinux1_i686.whl", hash = "sha256:f7ee479e96f7ee350db1cf24afa5685a5899e2b34992fb99e1f7c1b0b758d263"}, + {file = "zope.interface-5.4.0-cp27-cp27mu-manylinux1_x86_64.whl", hash = "sha256:b0297b1e05fd128d26cc2460c810d42e205d16d76799526dfa8c8ccd50e74959"}, + {file = "zope.interface-5.4.0-cp27-cp27mu-manylinux2010_i686.whl", hash = "sha256:af310ec8335016b5e52cae60cda4a4f2a60a788cbb949a4fbea13d441aa5a09e"}, + {file = "zope.interface-5.4.0-cp27-cp27mu-manylinux2010_x86_64.whl", hash = "sha256:9a9845c4c6bb56e508651f005c4aeb0404e518c6f000d5a1123ab077ab769f5c"}, + {file = "zope.interface-5.4.0-cp35-cp35m-manylinux1_i686.whl", hash = "sha256:0b465ae0962d49c68aa9733ba92a001b2a0933c317780435f00be7ecb959c702"}, + {file = "zope.interface-5.4.0-cp35-cp35m-manylinux1_x86_64.whl", hash = "sha256:5dd9ca406499444f4c8299f803d4a14edf7890ecc595c8b1c7115c2342cadc5f"}, + {file = "zope.interface-5.4.0-cp35-cp35m-manylinux2010_i686.whl", hash = "sha256:469e2407e0fe9880ac690a3666f03eb4c3c444411a5a5fddfdabc5d184a79f05"}, + {file = "zope.interface-5.4.0-cp35-cp35m-manylinux2010_x86_64.whl", hash = "sha256:52de7fc6c21b419078008f697fd4103dbc763288b1406b4562554bd47514c004"}, + {file = "zope.interface-5.4.0-cp35-cp35m-manylinux2014_aarch64.whl", hash = "sha256:3dd4952748521205697bc2802e4afac5ed4b02909bb799ba1fe239f77fd4e117"}, + {file = "zope.interface-5.4.0-cp35-cp35m-win32.whl", hash = "sha256:dd93ea5c0c7f3e25335ab7d22a507b1dc43976e1345508f845efc573d3d779d8"}, + {file = "zope.interface-5.4.0-cp35-cp35m-win_amd64.whl", hash = "sha256:3748fac0d0f6a304e674955ab1365d515993b3a0a865e16a11ec9d86fb307f63"}, + {file = "zope.interface-5.4.0-cp36-cp36m-macosx_10_14_x86_64.whl", hash = "sha256:66c0061c91b3b9cf542131148ef7ecbecb2690d48d1612ec386de9d36766058f"}, + {file = "zope.interface-5.4.0-cp36-cp36m-manylinux1_i686.whl", hash = "sha256:d0c1bc2fa9a7285719e5678584f6b92572a5b639d0e471bb8d4b650a1a910920"}, + {file = "zope.interface-5.4.0-cp36-cp36m-manylinux1_x86_64.whl", hash = "sha256:2876246527c91e101184f63ccd1d716ec9c46519cc5f3d5375a3351c46467c46"}, + {file = "zope.interface-5.4.0-cp36-cp36m-manylinux2010_i686.whl", hash = "sha256:334701327f37c47fa628fc8b8d28c7d7730ce7daaf4bda1efb741679c2b087fc"}, + {file = "zope.interface-5.4.0-cp36-cp36m-manylinux2010_x86_64.whl", hash = "sha256:71aace0c42d53abe6fc7f726c5d3b60d90f3c5c055a447950ad6ea9cec2e37d9"}, + {file = "zope.interface-5.4.0-cp36-cp36m-manylinux2014_aarch64.whl", hash = "sha256:5bb3489b4558e49ad2c5118137cfeaf59434f9737fa9c5deefc72d22c23822e2"}, + {file = "zope.interface-5.4.0-cp36-cp36m-win32.whl", hash = "sha256:1c0e316c9add0db48a5b703833881351444398b04111188069a26a61cfb4df78"}, + {file = "zope.interface-5.4.0-cp36-cp36m-win_amd64.whl", hash = "sha256:6f0c02cbb9691b7c91d5009108f975f8ffeab5dff8f26d62e21c493060eff2a1"}, + {file = "zope.interface-5.4.0-cp37-cp37m-macosx_10_14_x86_64.whl", hash = "sha256:7d97a4306898b05404a0dcdc32d9709b7d8832c0c542b861d9a826301719794e"}, + {file = "zope.interface-5.4.0-cp37-cp37m-manylinux1_i686.whl", hash = "sha256:867a5ad16892bf20e6c4ea2aab1971f45645ff3102ad29bd84c86027fa99997b"}, + {file = "zope.interface-5.4.0-cp37-cp37m-manylinux1_x86_64.whl", hash = "sha256:5f931a1c21dfa7a9c573ec1f50a31135ccce84e32507c54e1ea404894c5eb96f"}, + {file = "zope.interface-5.4.0-cp37-cp37m-manylinux2010_i686.whl", hash = "sha256:194d0bcb1374ac3e1e023961610dc8f2c78a0f5f634d0c737691e215569e640d"}, + {file = "zope.interface-5.4.0-cp37-cp37m-manylinux2010_x86_64.whl", hash = "sha256:8270252effc60b9642b423189a2fe90eb6b59e87cbee54549db3f5562ff8d1b8"}, + {file = "zope.interface-5.4.0-cp37-cp37m-manylinux2014_aarch64.whl", hash = "sha256:15e7d1f7a6ee16572e21e3576d2012b2778cbacf75eb4b7400be37455f5ca8bf"}, + {file = "zope.interface-5.4.0-cp37-cp37m-win32.whl", hash = "sha256:8892f89999ffd992208754851e5a052f6b5db70a1e3f7d54b17c5211e37a98c7"}, + {file = "zope.interface-5.4.0-cp37-cp37m-win_amd64.whl", hash = "sha256:2e5a26f16503be6c826abca904e45f1a44ff275fdb7e9d1b75c10671c26f8b94"}, + {file = "zope.interface-5.4.0-cp38-cp38-macosx_10_14_x86_64.whl", hash = "sha256:0f91b5b948686659a8e28b728ff5e74b1be6bf40cb04704453617e5f1e945ef3"}, + {file = "zope.interface-5.4.0-cp38-cp38-manylinux1_i686.whl", hash = "sha256:4de4bc9b6d35c5af65b454d3e9bc98c50eb3960d5a3762c9438df57427134b8e"}, + {file = "zope.interface-5.4.0-cp38-cp38-manylinux1_x86_64.whl", hash = "sha256:bf68f4b2b6683e52bec69273562df15af352e5ed25d1b6641e7efddc5951d1a7"}, + {file = "zope.interface-5.4.0-cp38-cp38-manylinux2010_i686.whl", hash = "sha256:63b82bb63de7c821428d513607e84c6d97d58afd1fe2eb645030bdc185440120"}, + {file = "zope.interface-5.4.0-cp38-cp38-manylinux2010_x86_64.whl", hash = "sha256:db1fa631737dab9fa0b37f3979d8d2631e348c3b4e8325d6873c2541d0ae5a48"}, + {file = "zope.interface-5.4.0-cp38-cp38-manylinux2014_aarch64.whl", hash = "sha256:f44e517131a98f7a76696a7b21b164bcb85291cee106a23beccce454e1f433a4"}, + {file = "zope.interface-5.4.0-cp38-cp38-win32.whl", hash = "sha256:a9506a7e80bcf6eacfff7f804c0ad5350c8c95b9010e4356a4b36f5322f09abb"}, + {file = "zope.interface-5.4.0-cp38-cp38-win_amd64.whl", hash = "sha256:3c02411a3b62668200910090a0dff17c0b25aaa36145082a5a6adf08fa281e54"}, + {file = "zope.interface-5.4.0-cp39-cp39-macosx_10_14_x86_64.whl", hash = "sha256:0cee5187b60ed26d56eb2960136288ce91bcf61e2a9405660d271d1f122a69a4"}, + {file = "zope.interface-5.4.0-cp39-cp39-manylinux1_i686.whl", hash = "sha256:a8156e6a7f5e2a0ff0c5b21d6bcb45145efece1909efcbbbf48c56f8da68221d"}, + {file = "zope.interface-5.4.0-cp39-cp39-manylinux1_x86_64.whl", hash = "sha256:205e40ccde0f37496904572035deea747390a8b7dc65146d30b96e2dd1359a83"}, + {file = "zope.interface-5.4.0-cp39-cp39-manylinux2010_i686.whl", hash = "sha256:3f24df7124c323fceb53ff6168da70dbfbae1442b4f3da439cd441681f54fe25"}, + {file = "zope.interface-5.4.0-cp39-cp39-manylinux2010_x86_64.whl", hash = "sha256:5208ebd5152e040640518a77827bdfcc73773a15a33d6644015b763b9c9febc1"}, + {file = "zope.interface-5.4.0-cp39-cp39-manylinux2014_aarch64.whl", hash = "sha256:17776ecd3a1fdd2b2cd5373e5ef8b307162f581c693575ec62e7c5399d80794c"}, + {file = "zope.interface-5.4.0-cp39-cp39-win32.whl", hash = "sha256:d4d9d6c1a455d4babd320203b918ccc7fcbefe308615c521062bc2ba1aa4d26e"}, + {file = "zope.interface-5.4.0-cp39-cp39-win_amd64.whl", hash = "sha256:0cba8477e300d64a11a9789ed40ee8932b59f9ee05f85276dbb4b59acee5dd09"}, + {file = "zope.interface-5.4.0.tar.gz", hash = "sha256:5dba5f530fec3f0988d83b78cc591b58c0b6eb8431a85edd1569a0539a8a5a0e"}, ] diff --git a/pyproject.toml b/pyproject.toml index 0343e3c..3f39fdf 100644 --- a/pyproject.toml +++ b/pyproject.toml @@ -6,10 +6,13 @@ authors = ["Mark Bonicillo <[email protected]>"] [tool.poetry.dependencies] python = "^3.8" -openleadr = "^0.5.23" black = "^21.5b2" mypy = "^0.812" pre-commit = "^2.13.0" +volttron-client = "^0.3.4" + +# local dependencies +openleadr= {path="./openleadr", develop=true} [tool.poetry.dev-dependencies] pytest = "^6.2.4" @@ -17,3 +20,19 @@ pytest = "^6.2.4" [build-system] requires = ["poetry-core>=1.0.0"] build-backend = "poetry.core.masonry.api" + +[tool.black] +line-length = 79 +include = '\.pyi?$' +exclude = ''' +/( + \.git + \| \.hg + \| \.mypy_cache + \| \.tox + \| \.venv + \| _build + \| buck-out + \| build + \| dist +)/''' diff --git a/volttron_client-0.1.2-py3-none-any.whl b/volttron_client-0.1.2-py3-none-any.whl deleted file mode 100644 index 870e627..0000000 Binary files a/volttron_client-0.1.2-py3-none-any.whl and /dev/null differ diff --git a/volttron_openadr_ven/agent.py b/volttron_openadr_ven/agent.py index c79e9de..ef2c3e2 100644 --- a/volttron_openadr_ven/agent.py +++ b/volttron_openadr_ven/agent.py @@ -42,22 +42,38 @@ import openleadr.enums from pprint import pformat -from typing import Dict from datetime import datetime -from openleadr.enums import OPT -from openleadr.objects import Event +from openleadr.enums import OPT, REPORT_NAME, MEASUREMENTS from openleadr import OpenADRClient -from volttron.platform.agent import utils -from volttron.platform.vip.agent import Agent, Core -from volttron.platform.messaging import topics, headers -from volttron.utils import jsonapi - -utils.setup_logging() +from volttron.utils import ( + get_aware_utc_now, + jsonapi, + setup_logging, + vip_main, + format_timestamp, + load_config, +) +from volttron.client.vip.agent import Agent, Core +from volttron.client.messaging import topics, headers + +setup_logging(level=logging.DEBUG) _log = logging.getLogger(__name__) __version__ = "1.0" +VEN_NAME = "ven_name" +VTN_URL = "vtn_url" +DEBUG = "debug" +CERT = "cert" +KEY = "key" +PASSPHRASE = "passphrase" +VTN_FINGERPRINT = "vtn_fingerprint" +SHOW_FINGERPRINT = "show_fingerprint" +CA_FILE = "ca_file" +VEN_ID = "ven_id" +DISABLE_SIGNATURE = "disable_signature" + class OpenADRVenAgent(Agent): """ @@ -72,106 +88,116 @@ def __init__( self, ven_name: str, vtn_url: str, - debug: bool, - cert: str, - key: str, - fingerprint: str, - show_fingerprint: str, - ca_file: str, + debug: bool = False, + cert: str = None, + key: str = None, + passphrase: str = None, + vtn_fingerprint: str = None, + show_fingerprint: str = None, + ca_file: str = None, + ven_id: str = None, + disable_signature: bool = None, kwargs, ) -> None: + """ + Initialize the agent's configuration. Create an OpenADR Client using OpenLeadr. + Configuration parameters (see config for a sample config file): + + str ven_name: The name for this VEN + str vtn_url: The URL of the VTN (Server) to connect to + bool debug: Whether or not to print debugging messages + str cert: The path to a PEM-formatted Certificate file to use + for signing messages. + str key: The path to a PEM-formatted Private Key file to use + for signing messages. + str passphrase: The passphrase for the Private Key + str vtn_fingerprint: The fingerprint for the VTN's certificate to + verify incomnig messages + str show_fingerprint: Whether to print your own fingerprint + on startup. Defaults to True. + str ca_file: The path to the PEM-formatted CA file for validating the VTN server's + certificate. + str ven_id: The ID for this VEN. If you leave this blank, a VEN_ID will be assigned by the VTN. + bool disable_signature: Whether to disable digital signatures + """ super(OpenADRVenAgent, self).__init__(enable_web=True, kwargs) - self.ven_name = None - self.vtn_url = None - self.debug = None - self.cert = None - self.key = None - self.fingerprint = None - self.show_fingerprint = None - self.ca_file = None - self.ven_id = None + # client will be initialized in configure_agent() + self.ven_client: OpenADRClient self.default_config = { - "ven_name": ven_name, - "vtn_url": vtn_url, - "debug": debug, - "cert": cert, - "key": key, - "fingerprint": fingerprint, - "show_fingerprint": show_fingerprint, - "ca_file": ca_file, + VEN_NAME: ven_name, + VTN_URL: vtn_url, + DEBUG: debug, + CERT: cert, + KEY: key, + PASSPHRASE: passphrase, + VTN_FINGERPRINT: vtn_fingerprint, + SHOW_FINGERPRINT: show_fingerprint, + CA_FILE: ca_file, + VEN_ID: ven_id, + DISABLE_SIGNATURE: disable_signature, } + + # SubSystem/ConfigStore self.vip.config.set_default("config", self.default_config) self.vip.config.subscribe( self._configure, actions=["NEW", "UPDATE"], pattern="config" ) - # properties of the this class will be initialized in initialize_config - # self.ven_client is an instance of OpenLeadr's OpenADRClient, an implementation of OpenADR; - # OpenADRClient will be instantiated in initialize_config - self.initialize_config(self.default_config) + # initialize all attributes of this class + self.configure_agent(self.default_config) - def initialize_config(self, config: Dict) -> None: + def configure_agent(self, config: dict) -> None: """ Initialize the agent's configuration. Create an OpenADR Client using OpenLeadr. - Configuration parameters (see config for a sample config file): - - str ven_name: The name for this VEN - str vtn_url: The URL of the VTN (Server) to connect to - bool debug: Whether or not to print debugging messages - str cert: The path to a PEM-formatted Certificate file to use - for signing messages. - str key: The path to a PEM-formatted Private Key file to use - for signing messages. - str fingerprint: The fingerprint for the VTN's certificate to - verify incomnig messages - str show_fingerprint: Whether to print your own fingerprint - on startup. Defaults to True. - str ca_file: The path to the PEM-formatted CA file for validating the VTN server's - certificate. """ - _log.debug("Configuring agent") - self.ven_name = config.get("ven_name") - self.vtn_url = config.get("vtn_url") - self.debug = config.get("debug", False) - self.cert = config.get("cert") - self.key = config.get("key") - self.fingerprint = config.get("fingerprint") - self.show_fingerprint = config.get("show_fingerprint", True) - self.ca_file = config.get("ca_file") - - _log.info("Configuration parameters:") - _log.info(pformat(config)) + _log.debug(f"Configuring agent with: \n {pformat(config)} ") # instantiate and add handlers to the OpenADR Client + _log.info("Creating OpenLeadrVen Client...") self.ven_client = OpenADRClient( - ven_name=self.ven_name, - vtn_url=self.vtn_url, - debug=self.debug, - cert=self.cert, - key=self.key, - show_fingerprint=self.show_fingerprint, - ca_file=self.ca_file, + ven_name=config.get(VEN_NAME), + vtn_url=config.get(VTN_URL), + debug=config.get(DEBUG), + cert=config.get(CERT), + key=config.get(KEY), + passphrase=config.get(PASSPHRASE), + vtn_fingerprint=config.get(VTN_FINGERPRINT), + show_fingerprint=config.get(SHOW_FINGERPRINT), + ca_file=config.get(CA_FILE), + ven_id=config.get(VEN_ID), + disable_signature=config.get(DISABLE_SIGNATURE), ) - # if you want to add more handlers on a specific event, you must create a coroutine - # in this class and then add it as the second input for 'self.ven_client.add_handler(<some event>, <coroutine>)' + # Add event handling capability to the client + # if you want to add more handlers on a specific event, you must create a coroutine in this class + # and then add it as the second input for 'self.ven_client.add_handler(<some event>, <coroutine>)' self.ven_client.add_handler("on_event", self.handle_event) - def _configure(self, contents: Dict) -> None: + # Add the report capability to the client + # the following is a report to be sent to a IPKeys test VTN + self.ven_client.add_report( + callback=self.collect_report_value, + report_name=REPORT_NAME.TELEMETRY_USAGE, + resource_id="device001", + measurement=MEASUREMENTS.VOLTAGE, + ) + + _log.info("Configuration complete.") + + def _configure(self, config_name, action, contents: dict) -> None: """The agent's config may have changed. Re-initialize it.""" config = self.default_config.copy() config.update(contents) - self.initialize_config(config) + self.configure_agent(config) # *************** Methods for Managing the Agent on Volttron **************** @Core.receiver("onstart") - def onstart(self) -> None: + def onstart(self, sender) -> None: """The agent has started.""" - _log.debug("Starting agent") - - # Running the ven_client in the Python AsyncIO Event Loop + _log.info(f"Sender {sender}") + _log.info("Starting agent...") loop = asyncio.get_event_loop() loop.create_task(self.ven_client.run()) loop.run_forever() @@ -197,13 +223,12 @@ def onstart(self) -> None: # ************* Methods for Servicing VTN Requests **************** - async def handle_event(self, event: openleadr.objects.Event) -> openleadr.enums.OPT: + async def handle_event(self, event: dict) -> openleadr.enums.OPT: """ Publish event to the Volttron message bus. Return OPT response. This coroutine will be called when there is an event to be handled. - :param event: Event - For more info on shape of 'event', see https://openleadr.org/docs/client.html#dealing-with-events and https://openleadr.org/docs/api/openleadr.html#openleadr.objects.Event + :param event: dict Example of event: { 'event_id': '123786-129831', @@ -225,51 +250,63 @@ async def handle_event(self, event: openleadr.objects.Event) -> openleadr.enums. } :return: openleadr.enums.OPT """ - _log.info(f"Event: {pformat(event)}") + try: + _log.debug("Received event. Processing event now...") + signal = event.get("event_signals")[0] # type: ignore[index] + _log.info(f"Event signal:\n {pformat(signal)}") + except IndexError as e: + _log.debug( + f"Event signals is empty. {e} \n Showing whole event: {pformat(event)}" + ) + pass + self.publish_event(event) + return OPT.OPT_IN + async def collect_report_value(self): + # This callback is called when you need to collect a value for your Report + # below is dummy code; replace with your business logic + return 1.23 + # ************* VOLTTRON Pub/Sub Requests **************** - def publish_event(self, event: openleadr.objects.Event) -> None: + def publish_event(self, event: dict) -> None: """Publish an event to the Volttron message bus. When an event is created/updated, it is published to the VOLTTRON bus with a topic that includes 'openadr/event_update'. - :param event: + :param event: dict :return: None """ # OADR rule 6: If testEvent is present and != "false", handle the event as a test event. - if event.event_descriptor.test_event != "false": - _log.debug(f"Suppressing publication of test event {event}") - return - - _log.debug("Publishing event {}".format(event)) + try: + if event["event_descriptor"]["test_event"]: + _log.debug("Suppressing publication of test event") + return + except KeyError as e: + _log.debug(f"Key error: {e}") + pass + + _log.debug(f"Publishing real/non-test event \n {pformat(event)}") self.vip.pubsub.publish( peer="pubsub", - topic=f"{topics.OPENADR_EVENT}/{self.ven_id}", - headers={ - headers.TIMESTAMP: utils.format_timestamp(utils.get_aware_utc_now()) - }, - message=self._json_object(event.__dict__), + topic=f"{topics.OPENADR_EVENT}/{self.ven_client.ven_name}", + headers={headers.TIMESTAMP: format_timestamp(get_aware_utc_now())}, + message=self._json_object(event), ) + return # ************* Helper methods **************** - def _json_object(self, obj: Dict): + def _json_object(self, obj: dict): """Ensure that an object is valid JSON by dumping it with json_converter and then reloading it.""" obj_string = jsonapi.dumps( - obj, default=lambda x: x.__str__() if isinstance(x, datetime) else None + obj, + default=lambda x: x.__str__() if isinstance(x, datetime) else None, ) return jsonapi.loads(obj_string) def main(): """Main method called to start the agent.""" - utils.vip_main(ven_agent, version=__version__) - - -if __name__ == "__main__": - try: - sys.exit(main()) - except KeyboardInterrupt: - pass + vip_main(ven_agent, version=__version__) def ven_agent(config_path: str, kwargs) -> OpenADRVenAgent: @@ -280,10 +317,13 @@ def ven_agent(config_path: str, kwargs) -> OpenADRVenAgent: :param config_path: (str) Path to a configuration file. - :returns: OpenADRVenAgent instance + :returns: OpenADRVenAgent """ try: - config = utils.load_config(config_path) + config = load_config(config_path) + except NameError as err: + _log.exception(err) + raise except Exception as err: _log.error("Error loading configuration: {}".format(err)) config = {} @@ -291,31 +331,42 @@ def ven_agent(config_path: str, kwargs) -> OpenADRVenAgent: if not config: raise Exception("Configuration cannot be empty.") - ven_name = config.get("ven_name") + ven_name = config.get(VEN_NAME) if not ven_name: - raise Exception("ven_name is required.") - vtn_url = config.get("vtn_url") + raise Exception(f"{VEN_NAME} is required.") + vtn_url = config.get(VTN_URL) if not vtn_url: raise Exception( - "vtn_url is required. Ensure vtn_url is given a URL to the VTN." + f"{VTN_URL} is required. Ensure {VTN_URL} is given a URL to the VTN." ) - - # below are optional configurations - debug = config.get("debug") - cert = config.get("cert") - key = config.get("key") - fingerprint = config.get("fingerprint") - show_fingerprint = config.get("show_fingerprint") - ca_file = config.get("ca_file") + debug = config.get(DEBUG) + cert = config.get(CERT) + key = config.get(KEY) + passphrase = config.get(PASSPHRASE) + vtn_fingerprint = config.get(VTN_FINGERPRINT) + show_fingerprint = config.get(SHOW_FINGERPRINT) + ca_file = config.get(CA_FILE) + ven_id = config.get(VEN_ID) + disable_signature = bool(config.get(DISABLE_SIGNATURE)) return OpenADRVenAgent( ven_name, vtn_url, - debug, - cert, - key, - fingerprint, - show_fingerprint, - ca_file, + debug=debug, + cert=cert, + key=key, + passphrase=passphrase, + vtn_fingerprint=vtn_fingerprint, + show_fingerprint=show_fingerprint, + ca_file=ca_file, + ven_id=ven_id, + disable_signature=disable_signature, **kwargs, ) + + +if __name__ == "__main__": + try: + sys.exit(main()) + except KeyboardInterrupt: + pass diff --git a/volttron_openadr_ven/config.json b/volttron_openadr_ven/config.json new file mode 100644 index 0000000..acba83b --- /dev/null +++ b/volttron_openadr_ven/config.json @@ -0,0 +1,8 @@ +{ + "ven_name": "PNNLVEN", + "vtn_url": "https://eiss2demo.ipkeys.com/oadr2/OpenADR2/Simple/2.0b", + "cert": "secret/", + "key": "secret/", + "debug": true, + "disable_signature": true +} diff --git a/volttron_utils-0.1.1-py3-none-any.whl b/volttron_utils-0.1.1-py3-none-any.whl deleted file mode 100644 index 6512925..0000000 Binary files a/volttron_utils-0.1.1-py3-none-any.whl and /dev/null differ	Work with openleadr in a script with certificates Verify working with openleadr allows the security and usage we are going to need. https://pypi.org/project/openleadr/		2021-10-21T21:44:33	[]	[]
eclipse-volttron/volttron-openadr-ven	eclipse-volttron__volttron-openadr-ven-8	a3f3883b84c59f609ea9d5f81f17e3343f678f2c	diff --git a/.gitignore b/.gitignore index b6e4761..b97131e 100644 --- a/.gitignore +++ b/.gitignore @@ -127,3 +127,5 @@ dmypy.json # Pyre type checker .pyre/ + +.idea/ diff --git a/.pre-commit-config.yaml b/.pre-commit-config.yaml new file mode 100644 index 0000000..218ffbf --- /dev/null +++ b/.pre-commit-config.yaml @@ -0,0 +1,15 @@ +repos: +- repo: https://github.com/pre-commit/pre-commit-hooks + rev: v2.3.0 + hooks: + - id: check-yaml + - id: end-of-file-fixer + - id: trailing-whitespace +- repo: https://github.com/psf/black + rev: 19.3b0 + hooks: + - id: black +- repo: https://github.com/pre-commit/mirrors-mypy + rev: v0.812 + hooks: + - id: mypy diff --git a/README.md b/README.md index d59628d..0191cf6 100644 --- a/README.md +++ b/README.md @@ -1,1 +1,95 @@ -# volttron-openadr-ven \ No newline at end of file +# Prerequisites + +* Python >=3.7 +* Pip >=20.1 +* Poetry >=1.16 + +# Development + +## Environment setup + +Note: This repo uses [Poetry](https://python-poetry.org/), a dependency management and packaging tool for Python. If you don't have Poetry installed on your machine, follow [these steps](https://python-poetry.org/docs/#installation) to install it on your machine. +To check if Poetry is installed, run ```poetry --version```. If your receive 'command not found: poetry' error, add the following line to your '~/.bashrc' script: ```export PATH=$PATH:$HOME/.poetry/bin```. + +To setup your development environment, do the following: + +1. Create virtual environment + +Run the following command on the CLI. By default, poetry creates a virtual environment in {cache-dir}/virtualenvs +({cache-dir}\virtualenvs on Windows). For more info on creating virtual environments through Poetry, see [this](https://python-poetry.org/docs/basic-usage/#using-your-virtual-environment). + +```shell +$ poetry shell +``` + +NOTE: Poetry needs the command 'python' in order to create a new shell. On Linux, if you only have python3 installed, run the following to link 'python' to 'python3': +```shell +$ sudo apt install python-is-python3 +``` + +2. Install requirements + +```shell +(env) $ poetry install +``` + + +3. Install Volttron-specific dependencies for this agent. + +```shell +(env) $ pip install volttron_utils-0.1.1-py3-none-any.whl volttron_client-0.1.2-py3-none-any.whl +``` + + +## Code Quality and Standardization + +## Git Hooks and Pre-commit + +This repo provides pre-commit hooks to enforce code quality and standardization before pushing any commit to the repo. To use this tool locally, +run the following commands: + +```shell +(env) $ pre-commit install + +# To check that pre-commit scripts are installed and properly run, run pre-commit on all the files +(env) $ pre-commit run --all +``` + +With pre-commit setup, every time you run "git commit", git will check your commit against a set of code linters. For details, see .pre-commit-config.yaml + +## Black + +This repo uses [Black](https://pypi.org/project/black/), an opinionated Python code formatter. To use Black to check and automagically reformat your code, +run the following command on your chosen Python file: + +```shell +(env) $ black <path to file> +``` + +## MyPy + +This repo uses [MyPy](https://mypy.readthedocs.io/en/stable/), a static type checker for Python. To use MyPy on a Python file, run the following command: +```shell +(env) $ mypy <path to file> +``` + +# CI/CD + +TODO: + +# Building Wheel and Publishing to PyPi + +TODO: + +Build wheel +```python setup.py -q sdist bdist_wheel``` + +Check wheel +```check-wheel-contents dist/.whl``` +```tar tzf <>``` + +Test uploading +```twine upload --repository-url https://test.pypi.org/legacy/ dist/``` + +Upload to PyPi +```python -m twin upload dist/``` diff --git a/mypy.ini b/mypy.ini new file mode 100644 index 0000000..03aefad --- /dev/null +++ b/mypy.ini @@ -0,0 +1,4 @@ +[mypy] +show_error_context = True +pretty = True +show_column_numbers = True diff --git a/poetry.lock b/poetry.lock new file mode 100644 index 0000000..26f5d1c --- /dev/null +++ b/poetry.lock @@ -0,0 +1,1116 @@ +[[package]] +name = "aiohttp" +version = "3.7.4.post0" +description = "Async http client/server framework (asyncio)" +category = "main" +optional = false +python-versions = ">=3.6" + +[package.dependencies] +async-timeout = ">=3.0,<4.0" +attrs = ">=17.3.0" +chardet = ">=2.0,<5.0" +multidict = ">=4.5,<7.0" +typing-extensions = ">=3.6.5" +yarl = ">=1.0,<2.0" + +[package.extras] +speedups = ["aiodns", "brotlipy", "cchardet"] + +[[package]] +name = "appdirs" +version = "1.4.4" +description = "A small Python module for determining appropriate platform-specific dirs, e.g. a \"user data dir\"." +category = "main" +optional = false +python-versions = "" + +[[package]] +name = "apscheduler" +version = "3.7.0" +description = "In-process task scheduler with Cron-like capabilities" +category = "main" +optional = false +python-versions = ">=2.7, !=3.0., !=3.1., !=3.2., !=3.3., !=3.4., <4" + +[package.dependencies] +pytz = "" +six = ">=1.4.0" +tzlocal = ">=2.0,<3.0" + +[package.extras] +asyncio = ["trollius"] +doc = ["sphinx", "sphinx-rtd-theme"] +gevent = ["gevent"] +mongodb = ["pymongo (>=3.0)"] +redis = ["redis (>=3.0)"] +rethinkdb = ["rethinkdb (>=2.4.0)"] +sqlalchemy = ["sqlalchemy (>=0.8)"] +testing = ["pytest (<6)", "pytest-cov", "pytest-tornado5", "mock", "pytest-asyncio (<0.6)", "pytest-asyncio"] +tornado = ["tornado (>=4.3)"] +twisted = ["twisted"] +zookeeper = ["kazoo"] + +[[package]] +name = "async-timeout" +version = "3.0.1" +description = "Timeout context manager for asyncio programs" +category = "main" +optional = false +python-versions = ">=3.5.3" + +[[package]] +name = "atomicwrites" +version = "1.4.0" +description = "Atomic file writes." +category = "dev" +optional = false +python-versions = ">=2.7, !=3.0., !=3.1., !=3.2., !=3.3." + +[[package]] +name = "attrs" +version = "21.2.0" +description = "Classes Without Boilerplate" +category = "main" +optional = false +python-versions = ">=2.7, !=3.0., !=3.1., !=3.2., !=3.3., !=3.4." + +[package.extras] +dev = ["coverage[toml] (>=5.0.2)", "hypothesis", "pympler", "pytest (>=4.3.0)", "six", "mypy", "pytest-mypy-plugins", "zope.interface", "furo", "sphinx", "sphinx-notfound-page", "pre-commit"] +docs = ["furo", "sphinx", "zope.interface", "sphinx-notfound-page"] +tests = ["coverage[toml] (>=5.0.2)", "hypothesis", "pympler", "pytest (>=4.3.0)", "six", "mypy", "pytest-mypy-plugins", "zope.interface"] +tests_no_zope = ["coverage[toml] (>=5.0.2)", "hypothesis", "pympler", "pytest (>=4.3.0)", "six", "mypy", "pytest-mypy-plugins"] + +[[package]] +name = "black" +version = "21.5b2" +description = "The uncompromising code formatter." +category = "main" +optional = false +python-versions = ">=3.6.2" + +[package.dependencies] +appdirs = "" +click = ">=7.1.2" +mypy-extensions = ">=0.4.3" +pathspec = ">=0.8.1,<1" +regex = ">=2020.1.8" +toml = ">=0.10.1" + +[package.extras] +colorama = ["colorama (>=0.4.3)"] +d = ["aiohttp (>=3.6.0)", "aiohttp-cors (>=0.4.0)"] +python2 = ["typed-ast (>=1.4.2)"] +uvloop = ["uvloop (>=0.15.2)"] + +[[package]] +name = "certifi" +version = "2021.5.30" +description = "Python package for providing Mozilla's CA Bundle." +category = "main" +optional = false +python-versions = "" + +[[package]] +name = "cffi" +version = "1.14.5" +description = "Foreign Function Interface for Python calling C code." +category = "main" +optional = false +python-versions = "" + +[package.dependencies] +pycparser = "" + +[[package]] +name = "cfgv" +version = "3.3.0" +description = "Validate configuration and produce human readable error messages." +category = "main" +optional = false +python-versions = ">=3.6.1" + +[[package]] +name = "chardet" +version = "4.0.0" +description = "Universal encoding detector for Python 2 and 3" +category = "main" +optional = false +python-versions = ">=2.7, !=3.0., !=3.1., !=3.2., !=3.3., !=3.4." + +[[package]] +name = "click" +version = "8.0.1" +description = "Composable command line interface toolkit" +category = "main" +optional = false +python-versions = ">=3.6" + +[package.dependencies] +colorama = {version = "", markers = "platform_system == \"Windows\""} + +[[package]] +name = "colorama" +version = "0.4.4" +description = "Cross-platform colored terminal text." +category = "main" +optional = false +python-versions = ">=2.7, !=3.0., !=3.1., !=3.2., !=3.3., !=3.4." + +[[package]] +name = "cryptography" +version = "3.4.7" +description = "cryptography is a package which provides cryptographic recipes and primitives to Python developers." +category = "main" +optional = false +python-versions = ">=3.6" + +[package.dependencies] +cffi = ">=1.12" + +[package.extras] +docs = ["sphinx (>=1.6.5,!=1.8.0,!=3.1.0,!=3.1.1)", "sphinx-rtd-theme"] +docstest = ["doc8", "pyenchant (>=1.6.11)", "twine (>=1.12.0)", "sphinxcontrib-spelling (>=4.0.1)"] +pep8test = ["black", "flake8", "flake8-import-order", "pep8-naming"] +sdist = ["setuptools-rust (>=0.11.4)"] +ssh = ["bcrypt (>=3.1.5)"] +test = ["pytest (>=6.0)", "pytest-cov", "pytest-subtests", "pytest-xdist", "pretend", "iso8601", "pytz", "hypothesis (>=1.11.4,!=3.79.2)"] + +[[package]] +name = "distlib" +version = "0.3.2" +description = "Distribution utilities" +category = "main" +optional = false +python-versions = "" + +[[package]] +name = "eight" +version = "1.0.1" +description = "Python 2 to the power of 3. A lightweight porting helper library." +category = "main" +optional = false +python-versions = "" + +[package.dependencies] +future = ">=0.16,<0.19" + +[[package]] +name = "filelock" +version = "3.0.12" +description = "A platform independent file lock." +category = "main" +optional = false +python-versions = "" + +[[package]] +name = "future" +version = "0.18.2" +description = "Clean single-source support for Python 3 and 2" +category = "main" +optional = false +python-versions = ">=2.6, !=3.0., !=3.1., !=3.2." + +[[package]] +name = "identify" +version = "2.2.7" +description = "File identification library for Python" +category = "main" +optional = false +python-versions = ">=3.6.1" + +[package.extras] +license = ["editdistance-s"] + +[[package]] +name = "idna" +version = "3.2" +description = "Internationalized Domain Names in Applications (IDNA)" +category = "main" +optional = false +python-versions = ">=3.5" + +[[package]] +name = "iniconfig" +version = "1.1.1" +description = "iniconfig: brain-dead simple config-ini parsing" +category = "dev" +optional = false +python-versions = "" + +[[package]] +name = "jinja2" +version = "3.0.1" +description = "A very fast and expressive template engine." +category = "main" +optional = false +python-versions = ">=3.6" + +[package.dependencies] +MarkupSafe = ">=2.0" + +[package.extras] +i18n = ["Babel (>=2.7)"] + +[[package]] +name = "lxml" +version = "4.6.3" +description = "Powerful and Pythonic XML processing library combining libxml2/libxslt with the ElementTree API." +category = "main" +optional = false +python-versions = ">=2.7, !=3.0., !=3.1., !=3.2., !=3.3., != 3.4." + +[package.extras] +cssselect = ["cssselect (>=0.7)"] +html5 = ["html5lib"] +htmlsoup = ["beautifulsoup4"] +source = ["Cython (>=0.29.7)"] + +[[package]] +name = "markupsafe" +version = "2.0.1" +description = "Safely add untrusted strings to HTML/XML markup." +category = "main" +optional = false +python-versions = ">=3.6" + +[[package]] +name = "multidict" +version = "5.1.0" +description = "multidict implementation" +category = "main" +optional = false +python-versions = ">=3.6" + +[[package]] +name = "mypy" +version = "0.812" +description = "Optional static typing for Python" +category = "main" +optional = false +python-versions = ">=3.5" + +[package.dependencies] +mypy-extensions = ">=0.4.3,<0.5.0" +typed-ast = ">=1.4.0,<1.5.0" +typing-extensions = ">=3.7.4" + +[package.extras] +dmypy = ["psutil (>=4.0)"] + +[[package]] +name = "mypy-extensions" +version = "0.4.3" +description = "Experimental type system extensions for programs checked with the mypy typechecker." +category = "main" +optional = false +python-versions = "" + +[[package]] +name = "nodeenv" +version = "1.6.0" +description = "Node.js virtual environment builder" +category = "main" +optional = false +python-versions = "" + +[[package]] +name = "openleadr" +version = "0.5.23" +description = "Python3 library for building OpenADR Clients (VENs) and Servers (VTNs)" +category = "main" +optional = false +python-versions = ">=3.7.0" + +[package.dependencies] +aiohttp = "" +apscheduler = "" +jinja2 = "" +signxml-openadr = "2.9.1" +xmltodict = "" + +[[package]] +name = "packaging" +version = "20.9" +description = "Core utilities for Python packages" +category = "dev" +optional = false +python-versions = ">=2.7, !=3.0., !=3.1., !=3.2., !=3.3." + +[package.dependencies] +pyparsing = ">=2.0.2" + +[[package]] +name = "pathspec" +version = "0.8.1" +description = "Utility library for gitignore style pattern matching of file paths." +category = "main" +optional = false +python-versions = ">=2.7, !=3.0., !=3.1., !=3.2., !=3.3., !=3.4." + +[[package]] +name = "pluggy" +version = "0.13.1" +description = "plugin and hook calling mechanisms for python" +category = "dev" +optional = false +python-versions = ">=2.7, !=3.0., !=3.1., !=3.2., !=3.3." + +[package.extras] +dev = ["pre-commit", "tox"] + +[[package]] +name = "pre-commit" +version = "2.13.0" +description = "A framework for managing and maintaining multi-language pre-commit hooks." +category = "main" +optional = false +python-versions = ">=3.6.1" + +[package.dependencies] +cfgv = ">=2.0.0" +identify = ">=1.0.0" +nodeenv = ">=0.11.1" +pyyaml = ">=5.1" +toml = "" +virtualenv = ">=20.0.8" + +[[package]] +name = "py" +version = "1.10.0" +description = "library with cross-python path, ini-parsing, io, code, log facilities" +category = "dev" +optional = false +python-versions = ">=2.7, !=3.0., !=3.1., !=3.2., !=3.3." + +[[package]] +name = "pycparser" +version = "2.20" +description = "C parser in Python" +category = "main" +optional = false +python-versions = ">=2.7, !=3.0., !=3.1., !=3.2., !=3.3." + +[[package]] +name = "pyopenssl" +version = "19.1.0" +description = "Python wrapper module around the OpenSSL library" +category = "main" +optional = false +python-versions = "" + +[package.dependencies] +cryptography = ">=2.8" +six = ">=1.5.2" + +[package.extras] +docs = ["sphinx", "sphinx-rtd-theme"] +test = ["flaky", "pretend", "pytest (>=3.0.1)"] + +[[package]] +name = "pyparsing" +version = "2.4.7" +description = "Python parsing module" +category = "dev" +optional = false +python-versions = ">=2.6, !=3.0., !=3.1., !=3.2." + +[[package]] +name = "pytest" +version = "6.2.4" +description = "pytest: simple powerful testing with Python" +category = "dev" +optional = false +python-versions = ">=3.6" + +[package.dependencies] +atomicwrites = {version = ">=1.0", markers = "sys_platform == \"win32\""} +attrs = ">=19.2.0" +colorama = {version = "", markers = "sys_platform == \"win32\""} +iniconfig = "" +packaging = "" +pluggy = ">=0.12,<1.0.0a1" +py = ">=1.8.2" +toml = "" + +[package.extras] +testing = ["argcomplete", "hypothesis (>=3.56)", "mock", "nose", "requests", "xmlschema"] + +[[package]] +name = "pytz" +version = "2021.1" +description = "World timezone definitions, modern and historical" +category = "main" +optional = false +python-versions = "" + +[[package]] +name = "pyyaml" +version = "5.4.1" +description = "YAML parser and emitter for Python" +category = "main" +optional = false +python-versions = ">=2.7, !=3.0., !=3.1., !=3.2., !=3.3., !=3.4., !=3.5." + +[[package]] +name = "regex" +version = "2021.4.4" +description = "Alternative regular expression module, to replace re." +category = "main" +optional = false +python-versions = "" + +[[package]] +name = "signxml-openadr" +version = "2.9.1" +description = "Python XML Signature library" +category = "main" +optional = false +python-versions = "" + +[package.dependencies] +certifi = ">=2018.1.18" +cryptography = ">=2.1.4,<4" +eight = ">=0.4.2,<2" +lxml = ">=4.2.1,<5" +pyOpenSSL = ">=17.5.0,<20" + +[[package]] +name = "six" +version = "1.16.0" +description = "Python 2 and 3 compatibility utilities" +category = "main" +optional = false +python-versions = ">=2.7, !=3.0., !=3.1., !=3.2." + +[[package]] +name = "toml" +version = "0.10.2" +description = "Python Library for Tom's Obvious, Minimal Language" +category = "main" +optional = false +python-versions = ">=2.6, !=3.0., !=3.1., !=3.2." + +[[package]] +name = "typed-ast" +version = "1.4.3" +description = "a fork of Python 2 and 3 ast modules with type comment support" +category = "main" +optional = false +python-versions = "" + +[[package]] +name = "typing-extensions" +version = "3.10.0.0" +description = "Backported and Experimental Type Hints for Python 3.5+" +category = "main" +optional = false +python-versions = "" + +[[package]] +name = "tzlocal" +version = "2.1" +description = "tzinfo object for the local timezone" +category = "main" +optional = false +python-versions = "" + +[package.dependencies] +pytz = "" + +[[package]] +name = "virtualenv" +version = "20.4.7" +description = "Virtual Python Environment builder" +category = "main" +optional = false +python-versions = "!=3.0.,!=3.1.,!=3.2.,!=3.3.,>=2.7" + +[package.dependencies] +appdirs = ">=1.4.3,<2" +distlib = ">=0.3.1,<1" +filelock = ">=3.0.0,<4" +six = ">=1.9.0,<2" + +[package.extras] +docs = ["proselint (>=0.10.2)", "sphinx (>=3)", "sphinx-argparse (>=0.2.5)", "sphinx-rtd-theme (>=0.4.3)", "towncrier (>=19.9.0rc1)"] +testing = ["coverage (>=4)", "coverage-enable-subprocess (>=1)", "flaky (>=3)", "pytest (>=4)", "pytest-env (>=0.6.2)", "pytest-freezegun (>=0.4.1)", "pytest-mock (>=2)", "pytest-randomly (>=1)", "pytest-timeout (>=1)", "packaging (>=20.0)", "xonsh (>=0.9.16)"] + +[[package]] +name = "xmltodict" +version = "0.12.0" +description = "Makes working with XML feel like you are working with JSON" +category = "main" +optional = false +python-versions = ">=2.7, !=3.0., !=3.1., !=3.2., !=3.3." + +[[package]] +name = "yarl" +version = "1.6.3" +description = "Yet another URL library" +category = "main" +optional = false +python-versions = ">=3.6" + +[package.dependencies] +idna = ">=2.0" +multidict = ">=4.0" + +[metadata] +lock-version = "1.1" +python-versions = "^3.8" +content-hash = "2a1dd13cfcde4f16340cb5425979909af99a26d9e7091cd2163a7e118ca1004b" + +[metadata.files] +aiohttp = [ + {file = "aiohttp-3.7.4.post0-cp36-cp36m-macosx_10_14_x86_64.whl", hash = "sha256:3cf75f7cdc2397ed4442594b935a11ed5569961333d49b7539ea741be2cc79d5"}, + {file = "aiohttp-3.7.4.post0-cp36-cp36m-manylinux1_i686.whl", hash = "sha256:4b302b45040890cea949ad092479e01ba25911a15e648429c7c5aae9650c67a8"}, + {file = "aiohttp-3.7.4.post0-cp36-cp36m-manylinux2014_aarch64.whl", hash = "sha256:fe60131d21b31fd1a14bd43e6bb88256f69dfc3188b3a89d736d6c71ed43ec95"}, + {file = "aiohttp-3.7.4.post0-cp36-cp36m-manylinux2014_i686.whl", hash = "sha256:393f389841e8f2dfc86f774ad22f00923fdee66d238af89b70ea314c4aefd290"}, + {file = "aiohttp-3.7.4.post0-cp36-cp36m-manylinux2014_ppc64le.whl", hash = "sha256:c6e9dcb4cb338d91a73f178d866d051efe7c62a7166653a91e7d9fb18274058f"}, + {file = "aiohttp-3.7.4.post0-cp36-cp36m-manylinux2014_s390x.whl", hash = "sha256:5df68496d19f849921f05f14f31bd6ef53ad4b00245da3195048c69934521809"}, + {file = "aiohttp-3.7.4.post0-cp36-cp36m-manylinux2014_x86_64.whl", hash = "sha256:0563c1b3826945eecd62186f3f5c7d31abb7391fedc893b7e2b26303b5a9f3fe"}, + {file = "aiohttp-3.7.4.post0-cp36-cp36m-win32.whl", hash = "sha256:3d78619672183be860b96ed96f533046ec97ca067fd46ac1f6a09cd9b7484287"}, + {file = "aiohttp-3.7.4.post0-cp36-cp36m-win_amd64.whl", hash = "sha256:f705e12750171c0ab4ef2a3c76b9a4024a62c4103e3a55dd6f99265b9bc6fcfc"}, + {file = "aiohttp-3.7.4.post0-cp37-cp37m-macosx_10_14_x86_64.whl", hash = "sha256:230a8f7e24298dea47659251abc0fd8b3c4e38a664c59d4b89cca7f6c09c9e87"}, + {file = "aiohttp-3.7.4.post0-cp37-cp37m-manylinux1_i686.whl", hash = "sha256:2e19413bf84934d651344783c9f5e22dee452e251cfd220ebadbed2d9931dbf0"}, + {file = "aiohttp-3.7.4.post0-cp37-cp37m-manylinux2014_aarch64.whl", hash = "sha256:e4b2b334e68b18ac9817d828ba44d8fcb391f6acb398bcc5062b14b2cbeac970"}, + {file = "aiohttp-3.7.4.post0-cp37-cp37m-manylinux2014_i686.whl", hash = "sha256:d012ad7911653a906425d8473a1465caa9f8dea7fcf07b6d870397b774ea7c0f"}, + {file = "aiohttp-3.7.4.post0-cp37-cp37m-manylinux2014_ppc64le.whl", hash = "sha256:40eced07f07a9e60e825554a31f923e8d3997cfc7fb31dbc1328c70826e04cde"}, + {file = "aiohttp-3.7.4.post0-cp37-cp37m-manylinux2014_s390x.whl", hash = "sha256:209b4a8ee987eccc91e2bd3ac36adee0e53a5970b8ac52c273f7f8fd4872c94c"}, + {file = "aiohttp-3.7.4.post0-cp37-cp37m-manylinux2014_x86_64.whl", hash = "sha256:14762875b22d0055f05d12abc7f7d61d5fd4fe4642ce1a249abdf8c700bf1fd8"}, + {file = "aiohttp-3.7.4.post0-cp37-cp37m-win32.whl", hash = "sha256:7615dab56bb07bff74bc865307aeb89a8bfd9941d2ef9d817b9436da3a0ea54f"}, + {file = "aiohttp-3.7.4.post0-cp37-cp37m-win_amd64.whl", hash = "sha256:d9e13b33afd39ddeb377eff2c1c4f00544e191e1d1dee5b6c51ddee8ea6f0cf5"}, + {file = "aiohttp-3.7.4.post0-cp38-cp38-macosx_10_14_x86_64.whl", hash = "sha256:547da6cacac20666422d4882cfcd51298d45f7ccb60a04ec27424d2f36ba3eaf"}, + {file = "aiohttp-3.7.4.post0-cp38-cp38-manylinux1_i686.whl", hash = "sha256:af9aa9ef5ba1fd5b8c948bb11f44891968ab30356d65fd0cc6707d989cd521df"}, + {file = "aiohttp-3.7.4.post0-cp38-cp38-manylinux2014_aarch64.whl", hash = "sha256:64322071e046020e8797117b3658b9c2f80e3267daec409b350b6a7a05041213"}, + {file = "aiohttp-3.7.4.post0-cp38-cp38-manylinux2014_i686.whl", hash = "sha256:bb437315738aa441251214dad17428cafda9cdc9729499f1d6001748e1d432f4"}, + {file = "aiohttp-3.7.4.post0-cp38-cp38-manylinux2014_ppc64le.whl", hash = "sha256:e54962802d4b8b18b6207d4a927032826af39395a3bd9196a5af43fc4e60b009"}, + {file = "aiohttp-3.7.4.post0-cp38-cp38-manylinux2014_s390x.whl", hash = "sha256:a00bb73540af068ca7390e636c01cbc4f644961896fa9363154ff43fd37af2f5"}, + {file = "aiohttp-3.7.4.post0-cp38-cp38-manylinux2014_x86_64.whl", hash = "sha256:79ebfc238612123a713a457d92afb4096e2148be17df6c50fb9bf7a81c2f8013"}, + {file = "aiohttp-3.7.4.post0-cp38-cp38-win32.whl", hash = "sha256:515dfef7f869a0feb2afee66b957cc7bbe9ad0cdee45aec7fdc623f4ecd4fb16"}, + {file = "aiohttp-3.7.4.post0-cp38-cp38-win_amd64.whl", hash = "sha256:114b281e4d68302a324dd33abb04778e8557d88947875cbf4e842c2c01a030c5"}, + {file = "aiohttp-3.7.4.post0-cp39-cp39-macosx_10_14_x86_64.whl", hash = "sha256:7b18b97cf8ee5452fa5f4e3af95d01d84d86d32c5e2bfa260cf041749d66360b"}, + {file = "aiohttp-3.7.4.post0-cp39-cp39-manylinux1_i686.whl", hash = "sha256:15492a6368d985b76a2a5fdd2166cddfea5d24e69eefed4630cbaae5c81d89bd"}, + {file = "aiohttp-3.7.4.post0-cp39-cp39-manylinux2014_aarch64.whl", hash = "sha256:bdb230b4943891321e06fc7def63c7aace16095be7d9cf3b1e01be2f10fba439"}, + {file = "aiohttp-3.7.4.post0-cp39-cp39-manylinux2014_i686.whl", hash = "sha256:cffe3ab27871bc3ea47df5d8f7013945712c46a3cc5a95b6bee15887f1675c22"}, + {file = "aiohttp-3.7.4.post0-cp39-cp39-manylinux2014_ppc64le.whl", hash = "sha256:f881853d2643a29e643609da57b96d5f9c9b93f62429dcc1cbb413c7d07f0e1a"}, + {file = "aiohttp-3.7.4.post0-cp39-cp39-manylinux2014_s390x.whl", hash = "sha256:a5ca29ee66f8343ed336816c553e82d6cade48a3ad702b9ffa6125d187e2dedb"}, + {file = "aiohttp-3.7.4.post0-cp39-cp39-manylinux2014_x86_64.whl", hash = "sha256:17c073de315745a1510393a96e680d20af8e67e324f70b42accbd4cb3315c9fb"}, + {file = "aiohttp-3.7.4.post0-cp39-cp39-win32.whl", hash = "sha256:932bb1ea39a54e9ea27fc9232163059a0b8855256f4052e776357ad9add6f1c9"}, + {file = "aiohttp-3.7.4.post0-cp39-cp39-win_amd64.whl", hash = "sha256:02f46fc0e3c5ac58b80d4d56eb0a7c7d97fcef69ace9326289fb9f1955e65cfe"}, + {file = "aiohttp-3.7.4.post0.tar.gz", hash = "sha256:493d3299ebe5f5a7c66b9819eacdcfbbaaf1a8e84911ddffcdc48888497afecf"}, +] +appdirs = [ + {file = "appdirs-1.4.4-py2.py3-none-any.whl", hash = "sha256:a841dacd6b99318a741b166adb07e19ee71a274450e68237b4650ca1055ab128"}, + {file = "appdirs-1.4.4.tar.gz", hash = "sha256:7d5d0167b2b1ba821647616af46a749d1c653740dd0d2415100fe26e27afdf41"}, +] +apscheduler = [ + {file = "APScheduler-3.7.0-py2.py3-none-any.whl", hash = "sha256:c06cc796d5bb9eb3c4f77727f6223476eb67749e7eea074d1587550702a7fbe3"}, + {file = "APScheduler-3.7.0.tar.gz", hash = "sha256:1cab7f2521e107d07127b042155b632b7a1cd5e02c34be5a28ff62f77c900c6a"}, +] +async-timeout = [ + {file = "async-timeout-3.0.1.tar.gz", hash = "sha256:0c3c816a028d47f659d6ff5c745cb2acf1f966da1fe5c19c77a70282b25f4c5f"}, + {file = "async_timeout-3.0.1-py3-none-any.whl", hash = "sha256:4291ca197d287d274d0b6cb5d6f8f8f82d434ed288f962539ff18cc9012f9ea3"}, +] +atomicwrites = [ + {file = "atomicwrites-1.4.0-py2.py3-none-any.whl", hash = "sha256:6d1784dea7c0c8d4a5172b6c620f40b6e4cbfdf96d783691f2e1302a7b88e197"}, + {file = "atomicwrites-1.4.0.tar.gz", hash = "sha256:ae70396ad1a434f9c7046fd2dd196fc04b12f9e91ffb859164193be8b6168a7a"}, +] +attrs = [ + {file = "attrs-21.2.0-py2.py3-none-any.whl", hash = "sha256:149e90d6d8ac20db7a955ad60cf0e6881a3f20d37096140088356da6c716b0b1"}, + {file = "attrs-21.2.0.tar.gz", hash = "sha256:ef6aaac3ca6cd92904cdd0d83f629a15f18053ec84e6432106f7a4d04ae4f5fb"}, +] +black = [ + {file = "black-21.5b2-py3-none-any.whl", hash = "sha256:e5cf21ebdffc7a9b29d73912b6a6a9a4df4ce70220d523c21647da2eae0751ef"}, + {file = "black-21.5b2.tar.gz", hash = "sha256:1fc0e0a2c8ae7d269dfcf0c60a89afa299664f3e811395d40b1922dff8f854b5"}, +] +certifi = [ + {file = "certifi-2021.5.30-py2.py3-none-any.whl", hash = "sha256:50b1e4f8446b06f41be7dd6338db18e0990601dce795c2b1686458aa7e8fa7d8"}, + {file = "certifi-2021.5.30.tar.gz", hash = "sha256:2bbf76fd432960138b3ef6dda3dde0544f27cbf8546c458e60baf371917ba9ee"}, +] +cffi = [ + {file = "cffi-1.14.5-cp27-cp27m-macosx_10_9_x86_64.whl", hash = "sha256:bb89f306e5da99f4d922728ddcd6f7fcebb3241fc40edebcb7284d7514741991"}, + {file = "cffi-1.14.5-cp27-cp27m-manylinux1_i686.whl", hash = "sha256:34eff4b97f3d982fb93e2831e6750127d1355a923ebaeeb565407b3d2f8d41a1"}, + {file = "cffi-1.14.5-cp27-cp27m-manylinux1_x86_64.whl", hash = "sha256:99cd03ae7988a93dd00bcd9d0b75e1f6c426063d6f03d2f90b89e29b25b82dfa"}, + {file = "cffi-1.14.5-cp27-cp27m-win32.whl", hash = "sha256:65fa59693c62cf06e45ddbb822165394a288edce9e276647f0046e1ec26920f3"}, + {file = "cffi-1.14.5-cp27-cp27m-win_amd64.whl", hash = "sha256:51182f8927c5af975fece87b1b369f722c570fe169f9880764b1ee3bca8347b5"}, + {file = "cffi-1.14.5-cp27-cp27mu-manylinux1_i686.whl", hash = "sha256:43e0b9d9e2c9e5d152946b9c5fe062c151614b262fda2e7b201204de0b99e482"}, + {file = "cffi-1.14.5-cp27-cp27mu-manylinux1_x86_64.whl", hash = "sha256:cbde590d4faaa07c72bf979734738f328d239913ba3e043b1e98fe9a39f8b2b6"}, + {file = "cffi-1.14.5-cp35-cp35m-macosx_10_9_x86_64.whl", hash = "sha256:5de7970188bb46b7bf9858eb6890aad302577a5f6f75091fd7cdd3ef13ef3045"}, + {file = "cffi-1.14.5-cp35-cp35m-manylinux1_i686.whl", hash = "sha256:a465da611f6fa124963b91bf432d960a555563efe4ed1cc403ba5077b15370aa"}, + {file = "cffi-1.14.5-cp35-cp35m-manylinux1_x86_64.whl", hash = "sha256:d42b11d692e11b6634f7613ad8df5d6d5f8875f5d48939520d351007b3c13406"}, + {file = "cffi-1.14.5-cp35-cp35m-win32.whl", hash = "sha256:72d8d3ef52c208ee1c7b2e341f7d71c6fd3157138abf1a95166e6165dd5d4369"}, + {file = "cffi-1.14.5-cp35-cp35m-win_amd64.whl", hash = "sha256:29314480e958fd8aab22e4a58b355b629c59bf5f2ac2492b61e3dc06d8c7a315"}, + {file = "cffi-1.14.5-cp36-cp36m-macosx_10_9_x86_64.whl", hash = "sha256:3d3dd4c9e559eb172ecf00a2a7517e97d1e96de2a5e610bd9b68cea3925b4892"}, + {file = "cffi-1.14.5-cp36-cp36m-manylinux1_i686.whl", hash = "sha256:48e1c69bbacfc3d932221851b39d49e81567a4d4aac3b21258d9c24578280058"}, + {file = "cffi-1.14.5-cp36-cp36m-manylinux1_x86_64.whl", hash = "sha256:69e395c24fc60aad6bb4fa7e583698ea6cc684648e1ffb7fe85e3c1ca131a7d5"}, + {file = "cffi-1.14.5-cp36-cp36m-manylinux2014_aarch64.whl", hash = "sha256:9e93e79c2551ff263400e1e4be085a1210e12073a31c2011dbbda14bda0c6132"}, + {file = "cffi-1.14.5-cp36-cp36m-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:24ec4ff2c5c0c8f9c6b87d5bb53555bf267e1e6f70e52e5a9740d32861d36b6f"}, + {file = "cffi-1.14.5-cp36-cp36m-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:3c3f39fa737542161d8b0d680df2ec249334cd70a8f420f71c9304bd83c3cbed"}, + {file = "cffi-1.14.5-cp36-cp36m-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:681d07b0d1e3c462dd15585ef5e33cb021321588bebd910124ef4f4fb71aef55"}, + {file = "cffi-1.14.5-cp36-cp36m-win32.whl", hash = "sha256:58e3f59d583d413809d60779492342801d6e82fefb89c86a38e040c16883be53"}, + {file = "cffi-1.14.5-cp36-cp36m-win_amd64.whl", hash = "sha256:005a36f41773e148deac64b08f233873a4d0c18b053d37da83f6af4d9087b813"}, + {file = "cffi-1.14.5-cp37-cp37m-macosx_10_9_x86_64.whl", hash = "sha256:2894f2df484ff56d717bead0a5c2abb6b9d2bf26d6960c4604d5c48bbc30ee73"}, + {file = "cffi-1.14.5-cp37-cp37m-manylinux1_i686.whl", hash = "sha256:0857f0ae312d855239a55c81ef453ee8fd24136eaba8e87a2eceba644c0d4c06"}, + {file = "cffi-1.14.5-cp37-cp37m-manylinux1_x86_64.whl", hash = "sha256:cd2868886d547469123fadc46eac7ea5253ea7fcb139f12e1dfc2bbd406427d1"}, + {file = "cffi-1.14.5-cp37-cp37m-manylinux2014_aarch64.whl", hash = "sha256:35f27e6eb43380fa080dccf676dece30bef72e4a67617ffda586641cd4508d49"}, + {file = "cffi-1.14.5-cp37-cp37m-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:06d7cd1abac2ffd92e65c0609661866709b4b2d82dd15f611e602b9b188b0b69"}, + {file = "cffi-1.14.5-cp37-cp37m-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:0f861a89e0043afec2a51fd177a567005847973be86f709bbb044d7f42fc4e05"}, + {file = "cffi-1.14.5-cp37-cp37m-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:cc5a8e069b9ebfa22e26d0e6b97d6f9781302fe7f4f2b8776c3e1daea35f1adc"}, + {file = "cffi-1.14.5-cp37-cp37m-win32.whl", hash = "sha256:9ff227395193126d82e60319a673a037d5de84633f11279e336f9c0f189ecc62"}, + {file = "cffi-1.14.5-cp37-cp37m-win_amd64.whl", hash = "sha256:9cf8022fb8d07a97c178b02327b284521c7708d7c71a9c9c355c178ac4bbd3d4"}, + {file = "cffi-1.14.5-cp38-cp38-macosx_10_9_x86_64.whl", hash = "sha256:8b198cec6c72df5289c05b05b8b0969819783f9418e0409865dac47288d2a053"}, + {file = "cffi-1.14.5-cp38-cp38-manylinux1_i686.whl", hash = "sha256:ad17025d226ee5beec591b52800c11680fca3df50b8b29fe51d882576e039ee0"}, + {file = "cffi-1.14.5-cp38-cp38-manylinux1_x86_64.whl", hash = "sha256:6c97d7350133666fbb5cf4abdc1178c812cb205dc6f41d174a7b0f18fb93337e"}, + {file = "cffi-1.14.5-cp38-cp38-manylinux2014_aarch64.whl", hash = "sha256:8ae6299f6c68de06f136f1f9e69458eae58f1dacf10af5c17353eae03aa0d827"}, + {file = "cffi-1.14.5-cp38-cp38-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:04c468b622ed31d408fea2346bec5bbffba2cc44226302a0de1ade9f5ea3d373"}, + {file = "cffi-1.14.5-cp38-cp38-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:06db6321b7a68b2bd6df96d08a5adadc1fa0e8f419226e25b2a5fbf6ccc7350f"}, + {file = "cffi-1.14.5-cp38-cp38-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:293e7ea41280cb28c6fcaaa0b1aa1f533b8ce060b9e701d78511e1e6c4a1de76"}, + {file = "cffi-1.14.5-cp38-cp38-win32.whl", hash = "sha256:b85eb46a81787c50650f2392b9b4ef23e1f126313b9e0e9013b35c15e4288e2e"}, + {file = "cffi-1.14.5-cp38-cp38-win_amd64.whl", hash = "sha256:1f436816fc868b098b0d63b8920de7d208c90a67212546d02f84fe78a9c26396"}, + {file = "cffi-1.14.5-cp39-cp39-macosx_10_9_x86_64.whl", hash = "sha256:1071534bbbf8cbb31b498d5d9db0f274f2f7a865adca4ae429e147ba40f73dea"}, + {file = "cffi-1.14.5-cp39-cp39-manylinux1_i686.whl", hash = "sha256:9de2e279153a443c656f2defd67769e6d1e4163952b3c622dcea5b08a6405322"}, + {file = "cffi-1.14.5-cp39-cp39-manylinux1_x86_64.whl", hash = "sha256:6e4714cc64f474e4d6e37cfff31a814b509a35cb17de4fb1999907575684479c"}, + {file = "cffi-1.14.5-cp39-cp39-manylinux2014_aarch64.whl", hash = "sha256:158d0d15119b4b7ff6b926536763dc0714313aa59e320ddf787502c70c4d4bee"}, + {file = "cffi-1.14.5-cp39-cp39-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:1bf1ac1984eaa7675ca8d5745a8cb87ef7abecb5592178406e55858d411eadc0"}, + {file = "cffi-1.14.5-cp39-cp39-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:df5052c5d867c1ea0b311fb7c3cd28b19df469c056f7fdcfe88c7473aa63e333"}, + {file = "cffi-1.14.5-cp39-cp39-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:24a570cd11895b60829e941f2613a4f79df1a27344cbbb82164ef2e0116f09c7"}, + {file = "cffi-1.14.5-cp39-cp39-win32.whl", hash = "sha256:afb29c1ba2e5a3736f1c301d9d0abe3ec8b86957d04ddfa9d7a6a42b9367e396"}, + {file = "cffi-1.14.5-cp39-cp39-win_amd64.whl", hash = "sha256:f2d45f97ab6bb54753eab54fffe75aaf3de4ff2341c9daee1987ee1837636f1d"}, + {file = "cffi-1.14.5.tar.gz", hash = "sha256:fd78e5fee591709f32ef6edb9a015b4aa1a5022598e36227500c8f4e02328d9c"}, +] +cfgv = [ + {file = "cfgv-3.3.0-py2.py3-none-any.whl", hash = "sha256:b449c9c6118fe8cca7fa5e00b9ec60ba08145d281d52164230a69211c5d597a1"}, + {file = "cfgv-3.3.0.tar.gz", hash = "sha256:9e600479b3b99e8af981ecdfc80a0296104ee610cab48a5ae4ffd0b668650eb1"}, +] +chardet = [ + {file = "chardet-4.0.0-py2.py3-none-any.whl", hash = "sha256:f864054d66fd9118f2e67044ac8981a54775ec5b67aed0441892edb553d21da5"}, + {file = "chardet-4.0.0.tar.gz", hash = "sha256:0d6f53a15db4120f2b08c94f11e7d93d2c911ee118b6b30a04ec3ee8310179fa"}, +] +click = [ + {file = "click-8.0.1-py3-none-any.whl", hash = "sha256:fba402a4a47334742d782209a7c79bc448911afe1149d07bdabdf480b3e2f4b6"}, + {file = "click-8.0.1.tar.gz", hash = "sha256:8c04c11192119b1ef78ea049e0a6f0463e4c48ef00a30160c704337586f3ad7a"}, +] +colorama = [ + {file = "colorama-0.4.4-py2.py3-none-any.whl", hash = "sha256:9f47eda37229f68eee03b24b9748937c7dc3868f906e8ba69fbcbdd3bc5dc3e2"}, + {file = "colorama-0.4.4.tar.gz", hash = "sha256:5941b2b48a20143d2267e95b1c2a7603ce057ee39fd88e7329b0c292aa16869b"}, +] +cryptography = [ + {file = "cryptography-3.4.7-cp36-abi3-macosx_10_10_x86_64.whl", hash = "sha256:3d8427734c781ea5f1b41d6589c293089704d4759e34597dce91014ac125aad1"}, + {file = "cryptography-3.4.7-cp36-abi3-macosx_11_0_arm64.whl", hash = "sha256:8e56e16617872b0957d1c9742a3f94b43533447fd78321514abbe7db216aa250"}, + {file = "cryptography-3.4.7-cp36-abi3-manylinux2010_x86_64.whl", hash = "sha256:37340614f8a5d2fb9aeea67fd159bfe4f5f4ed535b1090ce8ec428b2f15a11f2"}, + {file = "cryptography-3.4.7-cp36-abi3-manylinux2014_aarch64.whl", hash = "sha256:240f5c21aef0b73f40bb9f78d2caff73186700bf1bc6b94285699aff98cc16c6"}, + {file = "cryptography-3.4.7-cp36-abi3-manylinux2014_x86_64.whl", hash = "sha256:1e056c28420c072c5e3cb36e2b23ee55e260cb04eee08f702e0edfec3fb51959"}, + {file = "cryptography-3.4.7-cp36-abi3-win32.whl", hash = "sha256:0f1212a66329c80d68aeeb39b8a16d54ef57071bf22ff4e521657b27372e327d"}, + {file = "cryptography-3.4.7-cp36-abi3-win_amd64.whl", hash = "sha256:de4e5f7f68220d92b7637fc99847475b59154b7a1b3868fb7385337af54ac9ca"}, + {file = "cryptography-3.4.7-pp36-pypy36_pp73-manylinux2010_x86_64.whl", hash = "sha256:26965837447f9c82f1855e0bc8bc4fb910240b6e0d16a664bb722df3b5b06873"}, + {file = "cryptography-3.4.7-pp36-pypy36_pp73-manylinux2014_x86_64.whl", hash = "sha256:eb8cc2afe8b05acbd84a43905832ec78e7b3873fb124ca190f574dca7389a87d"}, + {file = "cryptography-3.4.7-pp37-pypy37_pp73-manylinux2010_x86_64.whl", hash = "sha256:7ec5d3b029f5fa2b179325908b9cd93db28ab7b85bb6c1db56b10e0b54235177"}, + {file = "cryptography-3.4.7-pp37-pypy37_pp73-manylinux2014_x86_64.whl", hash = "sha256:ee77aa129f481be46f8d92a1a7db57269a2f23052d5f2433b4621bb457081cc9"}, + {file = "cryptography-3.4.7.tar.gz", hash = "sha256:3d10de8116d25649631977cb37da6cbdd2d6fa0e0281d014a5b7d337255ca713"}, +] +distlib = [ + {file = "distlib-0.3.2-py2.py3-none-any.whl", hash = "sha256:23e223426b28491b1ced97dc3bbe183027419dfc7982b4fa2f05d5f3ff10711c"}, + {file = "distlib-0.3.2.zip", hash = "sha256:106fef6dc37dd8c0e2c0a60d3fca3e77460a48907f335fa28420463a6f799736"}, +] +eight = [ + {file = "eight-1.0.1-py2.py3-none-any.whl", hash = "sha256:bd6967ec7cf9e61ab18b55751236ed89a8363cfcb69127aace336869615245d5"}, + {file = "eight-1.0.1.tar.gz", hash = "sha256:6465970d3c12a1d759151690629b10b50ffb124c9b62e912999ee9233378a562"}, +] +filelock = [ + {file = "filelock-3.0.12-py3-none-any.whl", hash = "sha256:929b7d63ec5b7d6b71b0fa5ac14e030b3f70b75747cef1b10da9b879fef15836"}, + {file = "filelock-3.0.12.tar.gz", hash = "sha256:18d82244ee114f543149c66a6e0c14e9c4f8a1044b5cdaadd0f82159d6a6ff59"}, +] +future = [ + {file = "future-0.18.2.tar.gz", hash = "sha256:b1bead90b70cf6ec3f0710ae53a525360fa360d306a86583adc6bf83a4db537d"}, +] +identify = [ + {file = "identify-2.2.7-py2.py3-none-any.whl", hash = "sha256:92d6ad08eca19ceb17576733759944b94c0761277ddc3acf65e75e57ef190e32"}, + {file = "identify-2.2.7.tar.gz", hash = "sha256:c29e74c3671fe9537715cb695148231d777170ca1498e1c30c675d4ea782afe9"}, +] +idna = [ + {file = "idna-3.2-py3-none-any.whl", hash = "sha256:14475042e284991034cb48e06f6851428fb14c4dc953acd9be9a5e95c7b6dd7a"}, + {file = "idna-3.2.tar.gz", hash = "sha256:467fbad99067910785144ce333826c71fb0e63a425657295239737f7ecd125f3"}, +] +iniconfig = [ + {file = "iniconfig-1.1.1-py2.py3-none-any.whl", hash = "sha256:011e24c64b7f47f6ebd835bb12a743f2fbe9a26d4cecaa7f53bc4f35ee9da8b3"}, + {file = "iniconfig-1.1.1.tar.gz", hash = "sha256:bc3af051d7d14b2ee5ef9969666def0cd1a000e121eaea580d4a313df4b37f32"}, +] +jinja2 = [ + {file = "Jinja2-3.0.1-py3-none-any.whl", hash = "sha256:1f06f2da51e7b56b8f238affdd6b4e2c61e39598a378cc49345bc1bd42a978a4"}, + {file = "Jinja2-3.0.1.tar.gz", hash = "sha256:703f484b47a6af502e743c9122595cc812b0271f661722403114f71a79d0f5a4"}, +] +lxml = [ + {file = "lxml-4.6.3-cp27-cp27m-macosx_10_9_x86_64.whl", hash = "sha256:df7c53783a46febb0e70f6b05df2ba104610f2fb0d27023409734a3ecbb78fb2"}, + {file = "lxml-4.6.3-cp27-cp27m-manylinux1_i686.whl", hash = "sha256:1b7584d421d254ab86d4f0b13ec662a9014397678a7c4265a02a6d7c2b18a75f"}, + {file = "lxml-4.6.3-cp27-cp27m-manylinux1_x86_64.whl", hash = "sha256:079f3ae844f38982d156efce585bc540c16a926d4436712cf4baee0cce487a3d"}, + {file = "lxml-4.6.3-cp27-cp27m-win32.whl", hash = "sha256:bc4313cbeb0e7a416a488d72f9680fffffc645f8a838bd2193809881c67dd106"}, + {file = "lxml-4.6.3-cp27-cp27m-win_amd64.whl", hash = "sha256:8157dadbb09a34a6bd95a50690595e1fa0af1a99445e2744110e3dca7831c4ee"}, + {file = "lxml-4.6.3-cp27-cp27mu-manylinux1_i686.whl", hash = "sha256:7728e05c35412ba36d3e9795ae8995e3c86958179c9770e65558ec3fdfd3724f"}, + {file = "lxml-4.6.3-cp27-cp27mu-manylinux1_x86_64.whl", hash = "sha256:4bff24dfeea62f2e56f5bab929b4428ae6caba2d1eea0c2d6eb618e30a71e6d4"}, + {file = "lxml-4.6.3-cp35-cp35m-manylinux1_i686.whl", hash = "sha256:74f7d8d439b18fa4c385f3f5dfd11144bb87c1da034a466c5b5577d23a1d9b51"}, + {file = "lxml-4.6.3-cp35-cp35m-manylinux1_x86_64.whl", hash = "sha256:f90ba11136bfdd25cae3951af8da2e95121c9b9b93727b1b896e3fa105b2f586"}, + {file = "lxml-4.6.3-cp35-cp35m-manylinux2010_i686.whl", hash = "sha256:4c61b3a0db43a1607d6264166b230438f85bfed02e8cff20c22e564d0faff354"}, + {file = "lxml-4.6.3-cp35-cp35m-manylinux2014_x86_64.whl", hash = "sha256:5c8c163396cc0df3fd151b927e74f6e4acd67160d6c33304e805b84293351d16"}, + {file = "lxml-4.6.3-cp35-cp35m-win32.whl", hash = "sha256:f2380a6376dfa090227b663f9678150ef27543483055cc327555fb592c5967e2"}, + {file = "lxml-4.6.3-cp35-cp35m-win_amd64.whl", hash = "sha256:c4f05c5a7c49d2fb70223d0d5bcfbe474cf928310ac9fa6a7c6dddc831d0b1d4"}, + {file = "lxml-4.6.3-cp36-cp36m-macosx_10_9_x86_64.whl", hash = "sha256:d2e35d7bf1c1ac8c538f88d26b396e73dd81440d59c1ef8522e1ea77b345ede4"}, + {file = "lxml-4.6.3-cp36-cp36m-manylinux1_i686.whl", hash = "sha256:289e9ca1a9287f08daaf796d96e06cb2bc2958891d7911ac7cae1c5f9e1e0ee3"}, + {file = "lxml-4.6.3-cp36-cp36m-manylinux1_x86_64.whl", hash = "sha256:bccbfc27563652de7dc9bdc595cb25e90b59c5f8e23e806ed0fd623755b6565d"}, + {file = "lxml-4.6.3-cp36-cp36m-manylinux2010_i686.whl", hash = "sha256:d916d31fd85b2f78c76400d625076d9124de3e4bda8b016d25a050cc7d603f24"}, + {file = "lxml-4.6.3-cp36-cp36m-manylinux2014_aarch64.whl", hash = "sha256:820628b7b3135403540202e60551e741f9b6d3304371712521be939470b454ec"}, + {file = "lxml-4.6.3-cp36-cp36m-manylinux2014_x86_64.whl", hash = "sha256:c47ff7e0a36d4efac9fd692cfa33fbd0636674c102e9e8d9b26e1b93a94e7617"}, + {file = "lxml-4.6.3-cp36-cp36m-win32.whl", hash = "sha256:5a0a14e264069c03e46f926be0d8919f4105c1623d620e7ec0e612a2e9bf1c04"}, + {file = "lxml-4.6.3-cp36-cp36m-win_amd64.whl", hash = "sha256:92e821e43ad382332eade6812e298dc9701c75fe289f2a2d39c7960b43d1e92a"}, + {file = "lxml-4.6.3-cp37-cp37m-macosx_10_9_x86_64.whl", hash = "sha256:efd7a09678fd8b53117f6bae4fa3825e0a22b03ef0a932e070c0bdbb3a35e654"}, + {file = "lxml-4.6.3-cp37-cp37m-manylinux1_i686.whl", hash = "sha256:efac139c3f0bf4f0939f9375af4b02c5ad83a622de52d6dfa8e438e8e01d0eb0"}, + {file = "lxml-4.6.3-cp37-cp37m-manylinux1_x86_64.whl", hash = "sha256:0fbcf5565ac01dff87cbfc0ff323515c823081c5777a9fc7703ff58388c258c3"}, + {file = "lxml-4.6.3-cp37-cp37m-manylinux2010_i686.whl", hash = "sha256:36108c73739985979bf302006527cf8a20515ce444ba916281d1c43938b8bb96"}, + {file = "lxml-4.6.3-cp37-cp37m-manylinux2014_aarch64.whl", hash = "sha256:122fba10466c7bd4178b07dba427aa516286b846b2cbd6f6169141917283aae2"}, + {file = "lxml-4.6.3-cp37-cp37m-manylinux2014_x86_64.whl", hash = "sha256:cdaf11d2bd275bf391b5308f86731e5194a21af45fbaaaf1d9e8147b9160ea92"}, + {file = "lxml-4.6.3-cp37-cp37m-win32.whl", hash = "sha256:3439c71103ef0e904ea0a1901611863e51f50b5cd5e8654a151740fde5e1cade"}, + {file = "lxml-4.6.3-cp37-cp37m-win_amd64.whl", hash = "sha256:4289728b5e2000a4ad4ab8da6e1db2e093c63c08bdc0414799ee776a3f78da4b"}, + {file = "lxml-4.6.3-cp38-cp38-macosx_10_9_x86_64.whl", hash = "sha256:b007cbb845b28db4fb8b6a5cdcbf65bacb16a8bd328b53cbc0698688a68e1caa"}, + {file = "lxml-4.6.3-cp38-cp38-manylinux1_i686.whl", hash = "sha256:76fa7b1362d19f8fbd3e75fe2fb7c79359b0af8747e6f7141c338f0bee2f871a"}, + {file = "lxml-4.6.3-cp38-cp38-manylinux1_x86_64.whl", hash = "sha256:26e761ab5b07adf5f555ee82fb4bfc35bf93750499c6c7614bd64d12aaa67927"}, + {file = "lxml-4.6.3-cp38-cp38-manylinux2010_i686.whl", hash = "sha256:e1cbd3f19a61e27e011e02f9600837b921ac661f0c40560eefb366e4e4fb275e"}, + {file = "lxml-4.6.3-cp38-cp38-manylinux2014_aarch64.whl", hash = "sha256:66e575c62792c3f9ca47cb8b6fab9e35bab91360c783d1606f758761810c9791"}, + {file = "lxml-4.6.3-cp38-cp38-manylinux2014_x86_64.whl", hash = "sha256:1b38116b6e628118dea5b2186ee6820ab138dbb1e24a13e478490c7db2f326ae"}, + {file = "lxml-4.6.3-cp38-cp38-win32.whl", hash = "sha256:89b8b22a5ff72d89d48d0e62abb14340d9e99fd637d046c27b8b257a01ffbe28"}, + {file = "lxml-4.6.3-cp38-cp38-win_amd64.whl", hash = "sha256:2a9d50e69aac3ebee695424f7dbd7b8c6d6eb7de2a2eb6b0f6c7db6aa41e02b7"}, + {file = "lxml-4.6.3-cp39-cp39-macosx_10_9_x86_64.whl", hash = "sha256:ce256aaa50f6cc9a649c51be3cd4ff142d67295bfc4f490c9134d0f9f6d58ef0"}, + {file = "lxml-4.6.3-cp39-cp39-manylinux1_i686.whl", hash = "sha256:7610b8c31688f0b1be0ef882889817939490a36d0ee880ea562a4e1399c447a1"}, + {file = "lxml-4.6.3-cp39-cp39-manylinux1_x86_64.whl", hash = "sha256:f8380c03e45cf09f8557bdaa41e1fa7c81f3ae22828e1db470ab2a6c96d8bc23"}, + {file = "lxml-4.6.3-cp39-cp39-manylinux2010_i686.whl", hash = "sha256:3082c518be8e97324390614dacd041bb1358c882d77108ca1957ba47738d9d59"}, + {file = "lxml-4.6.3-cp39-cp39-manylinux2014_aarch64.whl", hash = "sha256:884ab9b29feaca361f7f88d811b1eea9bfca36cf3da27768d28ad45c3ee6f969"}, + {file = "lxml-4.6.3-cp39-cp39-manylinux2014_x86_64.whl", hash = "sha256:6f12e1427285008fd32a6025e38e977d44d6382cf28e7201ed10d6c1698d2a9a"}, + {file = "lxml-4.6.3-cp39-cp39-win32.whl", hash = "sha256:33bb934a044cf32157c12bfcfbb6649807da20aa92c062ef51903415c704704f"}, + {file = "lxml-4.6.3-cp39-cp39-win_amd64.whl", hash = "sha256:542d454665a3e277f76954418124d67516c5f88e51a900365ed54a9806122b83"}, + {file = "lxml-4.6.3.tar.gz", hash = "sha256:39b78571b3b30645ac77b95f7c69d1bffc4cf8c3b157c435a34da72e78c82468"}, +] +markupsafe = [ + {file = "MarkupSafe-2.0.1-cp36-cp36m-macosx_10_9_x86_64.whl", hash = "sha256:f9081981fe268bd86831e5c75f7de206ef275defcb82bc70740ae6dc507aee51"}, + {file = "MarkupSafe-2.0.1-cp36-cp36m-manylinux1_i686.whl", hash = "sha256:0955295dd5eec6cb6cc2fe1698f4c6d84af2e92de33fbcac4111913cd100a6ff"}, + {file = "MarkupSafe-2.0.1-cp36-cp36m-manylinux1_x86_64.whl", hash = "sha256:0446679737af14f45767963a1a9ef7620189912317d095f2d9ffa183a4d25d2b"}, + {file = "MarkupSafe-2.0.1-cp36-cp36m-manylinux2010_i686.whl", hash = "sha256:f826e31d18b516f653fe296d967d700fddad5901ae07c622bb3705955e1faa94"}, + {file = "MarkupSafe-2.0.1-cp36-cp36m-manylinux2010_x86_64.whl", hash = "sha256:fa130dd50c57d53368c9d59395cb5526eda596d3ffe36666cd81a44d56e48872"}, + {file = "MarkupSafe-2.0.1-cp36-cp36m-manylinux2014_aarch64.whl", hash = "sha256:905fec760bd2fa1388bb5b489ee8ee5f7291d692638ea5f67982d968366bef9f"}, + {file = "MarkupSafe-2.0.1-cp36-cp36m-win32.whl", hash = "sha256:6c4ca60fa24e85fe25b912b01e62cb969d69a23a5d5867682dd3e80b5b02581d"}, + {file = "MarkupSafe-2.0.1-cp36-cp36m-win_amd64.whl", hash = "sha256:b2f4bf27480f5e5e8ce285a8c8fd176c0b03e93dcc6646477d4630e83440c6a9"}, + {file = "MarkupSafe-2.0.1-cp37-cp37m-macosx_10_9_x86_64.whl", hash = "sha256:0717a7390a68be14b8c793ba258e075c6f4ca819f15edfc2a3a027c823718567"}, + {file = "MarkupSafe-2.0.1-cp37-cp37m-manylinux1_i686.whl", hash = "sha256:6557b31b5e2c9ddf0de32a691f2312a32f77cd7681d8af66c2692efdbef84c18"}, + {file = "MarkupSafe-2.0.1-cp37-cp37m-manylinux1_x86_64.whl", hash = "sha256:49e3ceeabbfb9d66c3aef5af3a60cc43b85c33df25ce03d0031a608b0a8b2e3f"}, + {file = "MarkupSafe-2.0.1-cp37-cp37m-manylinux2010_i686.whl", hash = "sha256:d7f9850398e85aba693bb640262d3611788b1f29a79f0c93c565694658f4071f"}, + {file = "MarkupSafe-2.0.1-cp37-cp37m-manylinux2010_x86_64.whl", hash = "sha256:6a7fae0dd14cf60ad5ff42baa2e95727c3d81ded453457771d02b7d2b3f9c0c2"}, + {file = "MarkupSafe-2.0.1-cp37-cp37m-manylinux2014_aarch64.whl", hash = "sha256:b7f2d075102dc8c794cbde1947378051c4e5180d52d276987b8d28a3bd58c17d"}, + {file = "MarkupSafe-2.0.1-cp37-cp37m-win32.whl", hash = "sha256:a30e67a65b53ea0a5e62fe23682cfe22712e01f453b95233b25502f7c61cb415"}, + {file = "MarkupSafe-2.0.1-cp37-cp37m-win_amd64.whl", hash = "sha256:611d1ad9a4288cf3e3c16014564df047fe08410e628f89805e475368bd304914"}, + {file = "MarkupSafe-2.0.1-cp38-cp38-macosx_10_9_x86_64.whl", hash = "sha256:be98f628055368795d818ebf93da628541e10b75b41c559fdf36d104c5787066"}, + {file = "MarkupSafe-2.0.1-cp38-cp38-manylinux1_i686.whl", hash = "sha256:1d609f577dc6e1aa17d746f8bd3c31aa4d258f4070d61b2aa5c4166c1539de35"}, + {file = "MarkupSafe-2.0.1-cp38-cp38-manylinux1_x86_64.whl", hash = "sha256:7d91275b0245b1da4d4cfa07e0faedd5b0812efc15b702576d103293e252af1b"}, + {file = "MarkupSafe-2.0.1-cp38-cp38-manylinux2010_i686.whl", hash = "sha256:01a9b8ea66f1658938f65b93a85ebe8bc016e6769611be228d797c9d998dd298"}, + {file = "MarkupSafe-2.0.1-cp38-cp38-manylinux2010_x86_64.whl", hash = "sha256:47ab1e7b91c098ab893b828deafa1203de86d0bc6ab587b160f78fe6c4011f75"}, + {file = "MarkupSafe-2.0.1-cp38-cp38-manylinux2014_aarch64.whl", hash = "sha256:97383d78eb34da7e1fa37dd273c20ad4320929af65d156e35a5e2d89566d9dfb"}, + {file = "MarkupSafe-2.0.1-cp38-cp38-win32.whl", hash = "sha256:023cb26ec21ece8dc3907c0e8320058b2e0cb3c55cf9564da612bc325bed5e64"}, + {file = "MarkupSafe-2.0.1-cp38-cp38-win_amd64.whl", hash = "sha256:984d76483eb32f1bcb536dc27e4ad56bba4baa70be32fa87152832cdd9db0833"}, + {file = "MarkupSafe-2.0.1-cp39-cp39-macosx_10_9_universal2.whl", hash = "sha256:2ef54abee730b502252bcdf31b10dacb0a416229b72c18b19e24a4509f273d26"}, + {file = "MarkupSafe-2.0.1-cp39-cp39-macosx_10_9_x86_64.whl", hash = "sha256:3c112550557578c26af18a1ccc9e090bfe03832ae994343cfdacd287db6a6ae7"}, + {file = "MarkupSafe-2.0.1-cp39-cp39-manylinux1_i686.whl", hash = "sha256:53edb4da6925ad13c07b6d26c2a852bd81e364f95301c66e930ab2aef5b5ddd8"}, + {file = "MarkupSafe-2.0.1-cp39-cp39-manylinux1_x86_64.whl", hash = "sha256:f5653a225f31e113b152e56f154ccbe59eeb1c7487b39b9d9f9cdb58e6c79dc5"}, + {file = "MarkupSafe-2.0.1-cp39-cp39-manylinux2010_i686.whl", hash = "sha256:4efca8f86c54b22348a5467704e3fec767b2db12fc39c6d963168ab1d3fc9135"}, + {file = "MarkupSafe-2.0.1-cp39-cp39-manylinux2010_x86_64.whl", hash = "sha256:ab3ef638ace319fa26553db0624c4699e31a28bb2a835c5faca8f8acf6a5a902"}, + {file = "MarkupSafe-2.0.1-cp39-cp39-manylinux2014_aarch64.whl", hash = "sha256:f8ba0e8349a38d3001fae7eadded3f6606f0da5d748ee53cc1dab1d6527b9509"}, + {file = "MarkupSafe-2.0.1-cp39-cp39-win32.whl", hash = "sha256:10f82115e21dc0dfec9ab5c0223652f7197feb168c940f3ef61563fc2d6beb74"}, + {file = "MarkupSafe-2.0.1-cp39-cp39-win_amd64.whl", hash = "sha256:693ce3f9e70a6cf7d2fb9e6c9d8b204b6b39897a2c4a1aa65728d5ac97dcc1d8"}, + {file = "MarkupSafe-2.0.1.tar.gz", hash = "sha256:594c67807fb16238b30c44bdf74f36c02cdf22d1c8cda91ef8a0ed8dabf5620a"}, +] +multidict = [ + {file = "multidict-5.1.0-cp36-cp36m-macosx_10_14_x86_64.whl", hash = "sha256:b7993704f1a4b204e71debe6095150d43b2ee6150fa4f44d6d966ec356a8d61f"}, + {file = "multidict-5.1.0-cp36-cp36m-manylinux1_i686.whl", hash = "sha256:9dd6e9b1a913d096ac95d0399bd737e00f2af1e1594a787e00f7975778c8b2bf"}, + {file = "multidict-5.1.0-cp36-cp36m-manylinux2014_aarch64.whl", hash = "sha256:f21756997ad8ef815d8ef3d34edd98804ab5ea337feedcd62fb52d22bf531281"}, + {file = "multidict-5.1.0-cp36-cp36m-manylinux2014_i686.whl", hash = "sha256:1ab820665e67373de5802acae069a6a05567ae234ddb129f31d290fc3d1aa56d"}, + {file = "multidict-5.1.0-cp36-cp36m-manylinux2014_ppc64le.whl", hash = "sha256:9436dc58c123f07b230383083855593550c4d301d2532045a17ccf6eca505f6d"}, + {file = "multidict-5.1.0-cp36-cp36m-manylinux2014_s390x.whl", hash = "sha256:830f57206cc96ed0ccf68304141fec9481a096c4d2e2831f311bde1c404401da"}, + {file = "multidict-5.1.0-cp36-cp36m-manylinux2014_x86_64.whl", hash = "sha256:2e68965192c4ea61fff1b81c14ff712fc7dc15d2bd120602e4a3494ea6584224"}, + {file = "multidict-5.1.0-cp36-cp36m-win32.whl", hash = "sha256:2f1a132f1c88724674271d636e6b7351477c27722f2ed789f719f9e3545a3d26"}, + {file = "multidict-5.1.0-cp36-cp36m-win_amd64.whl", hash = "sha256:3a4f32116f8f72ecf2a29dabfb27b23ab7cdc0ba807e8459e59a93a9be9506f6"}, + {file = "multidict-5.1.0-cp37-cp37m-macosx_10_14_x86_64.whl", hash = "sha256:46c73e09ad374a6d876c599f2328161bcd95e280f84d2060cf57991dec5cfe76"}, + {file = "multidict-5.1.0-cp37-cp37m-manylinux1_i686.whl", hash = "sha256:018132dbd8688c7a69ad89c4a3f39ea2f9f33302ebe567a879da8f4ca73f0d0a"}, + {file = "multidict-5.1.0-cp37-cp37m-manylinux2014_aarch64.whl", hash = "sha256:4b186eb7d6ae7c06eb4392411189469e6a820da81447f46c0072a41c748ab73f"}, + {file = "multidict-5.1.0-cp37-cp37m-manylinux2014_i686.whl", hash = "sha256:3a041b76d13706b7fff23b9fc83117c7b8fe8d5fe9e6be45eee72b9baa75f348"}, + {file = "multidict-5.1.0-cp37-cp37m-manylinux2014_ppc64le.whl", hash = "sha256:051012ccee979b2b06be928a6150d237aec75dd6bf2d1eeeb190baf2b05abc93"}, + {file = "multidict-5.1.0-cp37-cp37m-manylinux2014_s390x.whl", hash = "sha256:6a4d5ce640e37b0efcc8441caeea8f43a06addace2335bd11151bc02d2ee31f9"}, + {file = "multidict-5.1.0-cp37-cp37m-manylinux2014_x86_64.whl", hash = "sha256:5cf3443199b83ed9e955f511b5b241fd3ae004e3cb81c58ec10f4fe47c7dce37"}, + {file = "multidict-5.1.0-cp37-cp37m-win32.whl", hash = "sha256:f200755768dc19c6f4e2b672421e0ebb3dd54c38d5a4f262b872d8cfcc9e93b5"}, + {file = "multidict-5.1.0-cp37-cp37m-win_amd64.whl", hash = "sha256:05c20b68e512166fddba59a918773ba002fdd77800cad9f55b59790030bab632"}, + {file = "multidict-5.1.0-cp38-cp38-macosx_10_14_x86_64.whl", hash = "sha256:54fd1e83a184e19c598d5e70ba508196fd0bbdd676ce159feb412a4a6664f952"}, + {file = "multidict-5.1.0-cp38-cp38-manylinux1_i686.whl", hash = "sha256:0e3c84e6c67eba89c2dbcee08504ba8644ab4284863452450520dad8f1e89b79"}, + {file = "multidict-5.1.0-cp38-cp38-manylinux2014_aarch64.whl", hash = "sha256:dc862056f76443a0db4509116c5cd480fe1b6a2d45512a653f9a855cc0517456"}, + {file = "multidict-5.1.0-cp38-cp38-manylinux2014_i686.whl", hash = "sha256:0e929169f9c090dae0646a011c8b058e5e5fb391466016b39d21745b48817fd7"}, + {file = "multidict-5.1.0-cp38-cp38-manylinux2014_ppc64le.whl", hash = "sha256:d81eddcb12d608cc08081fa88d046c78afb1bf8107e6feab5d43503fea74a635"}, + {file = "multidict-5.1.0-cp38-cp38-manylinux2014_s390x.whl", hash = "sha256:585fd452dd7782130d112f7ddf3473ffdd521414674c33876187e101b588738a"}, + {file = "multidict-5.1.0-cp38-cp38-manylinux2014_x86_64.whl", hash = "sha256:37e5438e1c78931df5d3c0c78ae049092877e5e9c02dd1ff5abb9cf27a5914ea"}, + {file = "multidict-5.1.0-cp38-cp38-win32.whl", hash = "sha256:07b42215124aedecc6083f1ce6b7e5ec5b50047afa701f3442054373a6deb656"}, + {file = "multidict-5.1.0-cp38-cp38-win_amd64.whl", hash = "sha256:929006d3c2d923788ba153ad0de8ed2e5ed39fdbe8e7be21e2f22ed06c6783d3"}, + {file = "multidict-5.1.0-cp39-cp39-macosx_10_14_x86_64.whl", hash = "sha256:b797515be8743b771aa868f83563f789bbd4b236659ba52243b735d80b29ed93"}, + {file = "multidict-5.1.0-cp39-cp39-manylinux1_i686.whl", hash = "sha256:d5c65bdf4484872c4af3150aeebe101ba560dcfb34488d9a8ff8dbcd21079647"}, + {file = "multidict-5.1.0-cp39-cp39-manylinux2014_aarch64.whl", hash = "sha256:b47a43177a5e65b771b80db71e7be76c0ba23cc8aa73eeeb089ed5219cdbe27d"}, + {file = "multidict-5.1.0-cp39-cp39-manylinux2014_i686.whl", hash = "sha256:806068d4f86cb06af37cd65821554f98240a19ce646d3cd24e1c33587f313eb8"}, + {file = "multidict-5.1.0-cp39-cp39-manylinux2014_ppc64le.whl", hash = "sha256:46dd362c2f045095c920162e9307de5ffd0a1bfbba0a6e990b344366f55a30c1"}, + {file = "multidict-5.1.0-cp39-cp39-manylinux2014_s390x.whl", hash = "sha256:ace010325c787c378afd7f7c1ac66b26313b3344628652eacd149bdd23c68841"}, + {file = "multidict-5.1.0-cp39-cp39-manylinux2014_x86_64.whl", hash = "sha256:ecc771ab628ea281517e24fd2c52e8f31c41e66652d07599ad8818abaad38cda"}, + {file = "multidict-5.1.0-cp39-cp39-win32.whl", hash = "sha256:fc13a9524bc18b6fb6e0dbec3533ba0496bbed167c56d0aabefd965584557d80"}, + {file = "multidict-5.1.0-cp39-cp39-win_amd64.whl", hash = "sha256:7df80d07818b385f3129180369079bd6934cf70469f99daaebfac89dca288359"}, + {file = "multidict-5.1.0.tar.gz", hash = "sha256:25b4e5f22d3a37ddf3effc0710ba692cfc792c2b9edfb9c05aefe823256e84d5"}, +] +mypy = [ + {file = "mypy-0.812-cp35-cp35m-macosx_10_9_x86_64.whl", hash = "sha256:a26f8ec704e5a7423c8824d425086705e381b4f1dfdef6e3a1edab7ba174ec49"}, + {file = "mypy-0.812-cp35-cp35m-manylinux1_x86_64.whl", hash = "sha256:28fb5479c494b1bab244620685e2eb3c3f988d71fd5d64cc753195e8ed53df7c"}, + {file = "mypy-0.812-cp35-cp35m-manylinux2010_x86_64.whl", hash = "sha256:9743c91088d396c1a5a3c9978354b61b0382b4e3c440ce83cf77994a43e8c521"}, + {file = "mypy-0.812-cp35-cp35m-win_amd64.whl", hash = "sha256:d7da2e1d5f558c37d6e8c1246f1aec1e7349e4913d8fb3cb289a35de573fe2eb"}, + {file = "mypy-0.812-cp36-cp36m-macosx_10_9_x86_64.whl", hash = "sha256:4eec37370483331d13514c3f55f446fc5248d6373e7029a29ecb7b7494851e7a"}, + {file = "mypy-0.812-cp36-cp36m-manylinux1_x86_64.whl", hash = "sha256:d65cc1df038ef55a99e617431f0553cd77763869eebdf9042403e16089fe746c"}, + {file = "mypy-0.812-cp36-cp36m-manylinux2010_x86_64.whl", hash = "sha256:61a3d5b97955422964be6b3baf05ff2ce7f26f52c85dd88db11d5e03e146a3a6"}, + {file = "mypy-0.812-cp36-cp36m-win_amd64.whl", hash = "sha256:25adde9b862f8f9aac9d2d11971f226bd4c8fbaa89fb76bdadb267ef22d10064"}, + {file = "mypy-0.812-cp37-cp37m-macosx_10_9_x86_64.whl", hash = "sha256:552a815579aa1e995f39fd05dde6cd378e191b063f031f2acfe73ce9fb7f9e56"}, + {file = "mypy-0.812-cp37-cp37m-manylinux1_x86_64.whl", hash = "sha256:499c798053cdebcaa916eef8cd733e5584b5909f789de856b482cd7d069bdad8"}, + {file = "mypy-0.812-cp37-cp37m-manylinux2010_x86_64.whl", hash = "sha256:5873888fff1c7cf5b71efbe80e0e73153fe9212fafdf8e44adfe4c20ec9f82d7"}, + {file = "mypy-0.812-cp37-cp37m-win_amd64.whl", hash = "sha256:9f94aac67a2045ec719ffe6111df543bac7874cee01f41928f6969756e030564"}, + {file = "mypy-0.812-cp38-cp38-macosx_10_9_x86_64.whl", hash = "sha256:d23e0ea196702d918b60c8288561e722bf437d82cb7ef2edcd98cfa38905d506"}, + {file = "mypy-0.812-cp38-cp38-manylinux1_x86_64.whl", hash = "sha256:674e822aa665b9fd75130c6c5f5ed9564a38c6cea6a6432ce47eafb68ee578c5"}, + {file = "mypy-0.812-cp38-cp38-manylinux2010_x86_64.whl", hash = "sha256:abf7e0c3cf117c44d9285cc6128856106183938c68fd4944763003decdcfeb66"}, + {file = "mypy-0.812-cp38-cp38-win_amd64.whl", hash = "sha256:0d0a87c0e7e3a9becdfbe936c981d32e5ee0ccda3e0f07e1ef2c3d1a817cf73e"}, + {file = "mypy-0.812-cp39-cp39-macosx_10_9_x86_64.whl", hash = "sha256:7ce3175801d0ae5fdfa79b4f0cfed08807af4d075b402b7e294e6aa72af9aa2a"}, + {file = "mypy-0.812-cp39-cp39-manylinux1_x86_64.whl", hash = "sha256:b09669bcda124e83708f34a94606e01b614fa71931d356c1f1a5297ba11f110a"}, + {file = "mypy-0.812-cp39-cp39-manylinux2010_x86_64.whl", hash = "sha256:33f159443db0829d16f0a8d83d94df3109bb6dd801975fe86bacb9bf71628e97"}, + {file = "mypy-0.812-cp39-cp39-win_amd64.whl", hash = "sha256:3f2aca7f68580dc2508289c729bd49ee929a436208d2b2b6aab15745a70a57df"}, + {file = "mypy-0.812-py3-none-any.whl", hash = "sha256:2f9b3407c58347a452fc0736861593e105139b905cca7d097e413453a1d650b4"}, + {file = "mypy-0.812.tar.gz", hash = "sha256:cd07039aa5df222037005b08fbbfd69b3ab0b0bd7a07d7906de75ae52c4e3119"}, +] +mypy-extensions = [ + {file = "mypy_extensions-0.4.3-py2.py3-none-any.whl", hash = "sha256:090fedd75945a69ae91ce1303b5824f428daf5a028d2f6ab8a299250a846f15d"}, + {file = "mypy_extensions-0.4.3.tar.gz", hash = "sha256:2d82818f5bb3e369420cb3c4060a7970edba416647068eb4c5343488a6c604a8"}, +] +nodeenv = [ + {file = "nodeenv-1.6.0-py2.py3-none-any.whl", hash = "sha256:621e6b7076565ddcacd2db0294c0381e01fd28945ab36bcf00f41c5daf63bef7"}, + {file = "nodeenv-1.6.0.tar.gz", hash = "sha256:3ef13ff90291ba2a4a7a4ff9a979b63ffdd00a464dbe04acf0ea6471517a4c2b"}, +] +openleadr = [ + {file = "openleadr-0.5.23-py3-none-any.whl", hash = "sha256:863522020f7489d67827c69c1b3d6e75531ff75637238c87ecf69a4b08af439c"}, + {file = "openleadr-0.5.23.tar.gz", hash = "sha256:d6a77576120e51db471fcd1f70df50163363989399a058082e43e5a51a0f7c67"}, +] +packaging = [ + {file = "packaging-20.9-py2.py3-none-any.whl", hash = "sha256:67714da7f7bc052e064859c05c595155bd1ee9f69f76557e21f051443c20947a"}, + {file = "packaging-20.9.tar.gz", hash = "sha256:5b327ac1320dc863dca72f4514ecc086f31186744b84a230374cc1fd776feae5"}, +] +pathspec = [ + {file = "pathspec-0.8.1-py2.py3-none-any.whl", hash = "sha256:aa0cb481c4041bf52ffa7b0d8fa6cd3e88a2ca4879c533c9153882ee2556790d"}, + {file = "pathspec-0.8.1.tar.gz", hash = "sha256:86379d6b86d75816baba717e64b1a3a3469deb93bb76d613c9ce79edc5cb68fd"}, +] +pluggy = [ + {file = "pluggy-0.13.1-py2.py3-none-any.whl", hash = "sha256:966c145cd83c96502c3c3868f50408687b38434af77734af1e9ca461a4081d2d"}, + {file = "pluggy-0.13.1.tar.gz", hash = "sha256:15b2acde666561e1298d71b523007ed7364de07029219b604cf808bfa1c765b0"}, +] +pre-commit = [ + {file = "pre_commit-2.13.0-py2.py3-none-any.whl", hash = "sha256:b679d0fddd5b9d6d98783ae5f10fd0c4c59954f375b70a58cbe1ce9bcf9809a4"}, + {file = "pre_commit-2.13.0.tar.gz", hash = "sha256:764972c60693dc668ba8e86eb29654ec3144501310f7198742a767bec385a378"}, +] +py = [ + {file = "py-1.10.0-py2.py3-none-any.whl", hash = "sha256:3b80836aa6d1feeaa108e046da6423ab8f6ceda6468545ae8d02d9d58d18818a"}, + {file = "py-1.10.0.tar.gz", hash = "sha256:21b81bda15b66ef5e1a777a21c4dcd9c20ad3efd0b3f817e7a809035269e1bd3"}, +] +pycparser = [ + {file = "pycparser-2.20-py2.py3-none-any.whl", hash = "sha256:7582ad22678f0fcd81102833f60ef8d0e57288b6b5fb00323d101be910e35705"}, + {file = "pycparser-2.20.tar.gz", hash = "sha256:2d475327684562c3a96cc71adf7dc8c4f0565175cf86b6d7a404ff4c771f15f0"}, +] +pyopenssl = [ + {file = "pyOpenSSL-19.1.0-py2.py3-none-any.whl", hash = "sha256:621880965a720b8ece2f1b2f54ea2071966ab00e2970ad2ce11d596102063504"}, + {file = "pyOpenSSL-19.1.0.tar.gz", hash = "sha256:9a24494b2602aaf402be5c9e30a0b82d4a5c67528fe8fb475e3f3bc00dd69507"}, +] +pyparsing = [ + {file = "pyparsing-2.4.7-py2.py3-none-any.whl", hash = "sha256:ef9d7589ef3c200abe66653d3f1ab1033c3c419ae9b9bdb1240a85b024efc88b"}, + {file = "pyparsing-2.4.7.tar.gz", hash = "sha256:c203ec8783bf771a155b207279b9bccb8dea02d8f0c9e5f8ead507bc3246ecc1"}, +] +pytest = [ + {file = "pytest-6.2.4-py3-none-any.whl", hash = "sha256:91ef2131a9bd6be8f76f1f08eac5c5317221d6ad1e143ae03894b862e8976890"}, + {file = "pytest-6.2.4.tar.gz", hash = "sha256:50bcad0a0b9c5a72c8e4e7c9855a3ad496ca6a881a3641b4260605450772c54b"}, +] +pytz = [ + {file = "pytz-2021.1-py2.py3-none-any.whl", hash = "sha256:eb10ce3e7736052ed3623d49975ce333bcd712c7bb19a58b9e2089d4057d0798"}, + {file = "pytz-2021.1.tar.gz", hash = "sha256:83a4a90894bf38e243cf052c8b58f381bfe9a7a483f6a9cab140bc7f702ac4da"}, +] +pyyaml = [ + {file = "PyYAML-5.4.1-cp27-cp27m-macosx_10_9_x86_64.whl", hash = "sha256:3b2b1824fe7112845700f815ff6a489360226a5609b96ec2190a45e62a9fc922"}, + {file = "PyYAML-5.4.1-cp27-cp27m-win32.whl", hash = "sha256:129def1b7c1bf22faffd67b8f3724645203b79d8f4cc81f674654d9902cb4393"}, + {file = "PyYAML-5.4.1-cp27-cp27m-win_amd64.whl", hash = "sha256:4465124ef1b18d9ace298060f4eccc64b0850899ac4ac53294547536533800c8"}, + {file = "PyYAML-5.4.1-cp27-cp27mu-manylinux1_x86_64.whl", hash = "sha256:bb4191dfc9306777bc594117aee052446b3fa88737cd13b7188d0e7aa8162185"}, + {file = "PyYAML-5.4.1-cp36-cp36m-macosx_10_9_x86_64.whl", hash = "sha256:6c78645d400265a062508ae399b60b8c167bf003db364ecb26dcab2bda048253"}, + {file = "PyYAML-5.4.1-cp36-cp36m-manylinux1_x86_64.whl", hash = "sha256:4e0583d24c881e14342eaf4ec5fbc97f934b999a6828693a99157fde912540cc"}, + {file = "PyYAML-5.4.1-cp36-cp36m-manylinux2014_aarch64.whl", hash = "sha256:72a01f726a9c7851ca9bfad6fd09ca4e090a023c00945ea05ba1638c09dc3347"}, + {file = "PyYAML-5.4.1-cp36-cp36m-manylinux2014_s390x.whl", hash = "sha256:895f61ef02e8fed38159bb70f7e100e00f471eae2bc838cd0f4ebb21e28f8541"}, + {file = "PyYAML-5.4.1-cp36-cp36m-win32.whl", hash = "sha256:3bd0e463264cf257d1ffd2e40223b197271046d09dadf73a0fe82b9c1fc385a5"}, + {file = "PyYAML-5.4.1-cp36-cp36m-win_amd64.whl", hash = "sha256:e4fac90784481d221a8e4b1162afa7c47ed953be40d31ab4629ae917510051df"}, + {file = "PyYAML-5.4.1-cp37-cp37m-macosx_10_9_x86_64.whl", hash = "sha256:5accb17103e43963b80e6f837831f38d314a0495500067cb25afab2e8d7a4018"}, + {file = "PyYAML-5.4.1-cp37-cp37m-manylinux1_x86_64.whl", hash = "sha256:e1d4970ea66be07ae37a3c2e48b5ec63f7ba6804bdddfdbd3cfd954d25a82e63"}, + {file = "PyYAML-5.4.1-cp37-cp37m-manylinux2014_aarch64.whl", hash = "sha256:cb333c16912324fd5f769fff6bc5de372e9e7a202247b48870bc251ed40239aa"}, + {file = "PyYAML-5.4.1-cp37-cp37m-manylinux2014_s390x.whl", hash = "sha256:fe69978f3f768926cfa37b867e3843918e012cf83f680806599ddce33c2c68b0"}, + {file = "PyYAML-5.4.1-cp37-cp37m-win32.whl", hash = "sha256:dd5de0646207f053eb0d6c74ae45ba98c3395a571a2891858e87df7c9b9bd51b"}, + {file = "PyYAML-5.4.1-cp37-cp37m-win_amd64.whl", hash = "sha256:08682f6b72c722394747bddaf0aa62277e02557c0fd1c42cb853016a38f8dedf"}, + {file = "PyYAML-5.4.1-cp38-cp38-macosx_10_9_x86_64.whl", hash = "sha256:d2d9808ea7b4af864f35ea216be506ecec180628aced0704e34aca0b040ffe46"}, + {file = "PyYAML-5.4.1-cp38-cp38-manylinux1_x86_64.whl", hash = "sha256:8c1be557ee92a20f184922c7b6424e8ab6691788e6d86137c5d93c1a6ec1b8fb"}, + {file = "PyYAML-5.4.1-cp38-cp38-manylinux2014_aarch64.whl", hash = "sha256:fd7f6999a8070df521b6384004ef42833b9bd62cfee11a09bda1079b4b704247"}, + {file = "PyYAML-5.4.1-cp38-cp38-manylinux2014_s390x.whl", hash = "sha256:bfb51918d4ff3d77c1c856a9699f8492c612cde32fd3bcd344af9be34999bfdc"}, + {file = "PyYAML-5.4.1-cp38-cp38-win32.whl", hash = "sha256:fa5ae20527d8e831e8230cbffd9f8fe952815b2b7dae6ffec25318803a7528fc"}, + {file = "PyYAML-5.4.1-cp38-cp38-win_amd64.whl", hash = "sha256:0f5f5786c0e09baddcd8b4b45f20a7b5d61a7e7e99846e3c799b05c7c53fa696"}, + {file = "PyYAML-5.4.1-cp39-cp39-macosx_10_9_x86_64.whl", hash = "sha256:294db365efa064d00b8d1ef65d8ea2c3426ac366c0c4368d930bf1c5fb497f77"}, + {file = "PyYAML-5.4.1-cp39-cp39-manylinux1_x86_64.whl", hash = "sha256:74c1485f7707cf707a7aef42ef6322b8f97921bd89be2ab6317fd782c2d53183"}, + {file = "PyYAML-5.4.1-cp39-cp39-manylinux2014_aarch64.whl", hash = "sha256:d483ad4e639292c90170eb6f7783ad19490e7a8defb3e46f97dfe4bacae89122"}, + {file = "PyYAML-5.4.1-cp39-cp39-manylinux2014_s390x.whl", hash = "sha256:fdc842473cd33f45ff6bce46aea678a54e3d21f1b61a7750ce3c498eedfe25d6"}, + {file = "PyYAML-5.4.1-cp39-cp39-win32.whl", hash = "sha256:49d4cdd9065b9b6e206d0595fee27a96b5dd22618e7520c33204a4a3239d5b10"}, + {file = "PyYAML-5.4.1-cp39-cp39-win_amd64.whl", hash = "sha256:c20cfa2d49991c8b4147af39859b167664f2ad4561704ee74c1de03318e898db"}, + {file = "PyYAML-5.4.1.tar.gz", hash = "sha256:607774cbba28732bfa802b54baa7484215f530991055bb562efbed5b2f20a45e"}, +] +regex = [ + {file = "regex-2021.4.4-cp36-cp36m-macosx_10_9_x86_64.whl", hash = "sha256:619d71c59a78b84d7f18891fe914446d07edd48dc8328c8e149cbe0929b4e000"}, + {file = "regex-2021.4.4-cp36-cp36m-manylinux1_i686.whl", hash = "sha256:47bf5bf60cf04d72bf6055ae5927a0bd9016096bf3d742fa50d9bf9f45aa0711"}, + {file = "regex-2021.4.4-cp36-cp36m-manylinux1_x86_64.whl", hash = "sha256:281d2fd05555079448537fe108d79eb031b403dac622621c78944c235f3fcf11"}, + {file = "regex-2021.4.4-cp36-cp36m-manylinux2010_i686.whl", hash = "sha256:bd28bc2e3a772acbb07787c6308e00d9626ff89e3bfcdebe87fa5afbfdedf968"}, + {file = "regex-2021.4.4-cp36-cp36m-manylinux2010_x86_64.whl", hash = "sha256:7c2a1af393fcc09e898beba5dd59196edaa3116191cc7257f9224beaed3e1aa0"}, + {file = "regex-2021.4.4-cp36-cp36m-manylinux2014_aarch64.whl", hash = "sha256:c38c71df845e2aabb7fb0b920d11a1b5ac8526005e533a8920aea97efb8ec6a4"}, + {file = "regex-2021.4.4-cp36-cp36m-manylinux2014_i686.whl", hash = "sha256:96fcd1888ab4d03adfc9303a7b3c0bd78c5412b2bfbe76db5b56d9eae004907a"}, + {file = "regex-2021.4.4-cp36-cp36m-manylinux2014_x86_64.whl", hash = "sha256:ade17eb5d643b7fead300a1641e9f45401c98eee23763e9ed66a43f92f20b4a7"}, + {file = "regex-2021.4.4-cp36-cp36m-win32.whl", hash = "sha256:e8e5b509d5c2ff12f8418006d5a90e9436766133b564db0abaec92fd27fcee29"}, + {file = "regex-2021.4.4-cp36-cp36m-win_amd64.whl", hash = "sha256:11d773d75fa650cd36f68d7ca936e3c7afaae41b863b8c387a22aaa78d3c5c79"}, + {file = "regex-2021.4.4-cp37-cp37m-macosx_10_9_x86_64.whl", hash = "sha256:d3029c340cfbb3ac0a71798100ccc13b97dddf373a4ae56b6a72cf70dfd53bc8"}, + {file = "regex-2021.4.4-cp37-cp37m-manylinux1_i686.whl", hash = "sha256:18c071c3eb09c30a264879f0d310d37fe5d3a3111662438889ae2eb6fc570c31"}, + {file = "regex-2021.4.4-cp37-cp37m-manylinux1_x86_64.whl", hash = "sha256:4c557a7b470908b1712fe27fb1ef20772b78079808c87d20a90d051660b1d69a"}, + {file = "regex-2021.4.4-cp37-cp37m-manylinux2010_i686.whl", hash = "sha256:01afaf2ec48e196ba91b37451aa353cb7eda77efe518e481707e0515025f0cd5"}, + {file = "regex-2021.4.4-cp37-cp37m-manylinux2010_x86_64.whl", hash = "sha256:3a9cd17e6e5c7eb328517969e0cb0c3d31fd329298dd0c04af99ebf42e904f82"}, + {file = "regex-2021.4.4-cp37-cp37m-manylinux2014_aarch64.whl", hash = "sha256:90f11ff637fe8798933fb29f5ae1148c978cccb0452005bf4c69e13db951e765"}, + {file = "regex-2021.4.4-cp37-cp37m-manylinux2014_i686.whl", hash = "sha256:919859aa909429fb5aa9cf8807f6045592c85ef56fdd30a9a3747e513db2536e"}, + {file = "regex-2021.4.4-cp37-cp37m-manylinux2014_x86_64.whl", hash = "sha256:339456e7d8c06dd36a22e451d58ef72cef293112b559010db3d054d5560ef439"}, + {file = "regex-2021.4.4-cp37-cp37m-win32.whl", hash = "sha256:67bdb9702427ceddc6ef3dc382455e90f785af4c13d495f9626861763ee13f9d"}, + {file = "regex-2021.4.4-cp37-cp37m-win_amd64.whl", hash = "sha256:32e65442138b7b76dd8173ffa2cf67356b7bc1768851dded39a7a13bf9223da3"}, + {file = "regex-2021.4.4-cp38-cp38-macosx_10_9_x86_64.whl", hash = "sha256:1e1c20e29358165242928c2de1482fb2cf4ea54a6a6dea2bd7a0e0d8ee321500"}, + {file = "regex-2021.4.4-cp38-cp38-manylinux1_i686.whl", hash = "sha256:314d66636c494ed9c148a42731b3834496cc9a2c4251b1661e40936814542b14"}, + {file = "regex-2021.4.4-cp38-cp38-manylinux1_x86_64.whl", hash = "sha256:6d1b01031dedf2503631d0903cb563743f397ccaf6607a5e3b19a3d76fc10480"}, + {file = "regex-2021.4.4-cp38-cp38-manylinux2010_i686.whl", hash = "sha256:741a9647fcf2e45f3a1cf0e24f5e17febf3efe8d4ba1281dcc3aa0459ef424dc"}, + {file = "regex-2021.4.4-cp38-cp38-manylinux2010_x86_64.whl", hash = "sha256:4c46e22a0933dd783467cf32b3516299fb98cfebd895817d685130cc50cd1093"}, + {file = "regex-2021.4.4-cp38-cp38-manylinux2014_aarch64.whl", hash = "sha256:e512d8ef5ad7b898cdb2d8ee1cb09a8339e4f8be706d27eaa180c2f177248a10"}, + {file = "regex-2021.4.4-cp38-cp38-manylinux2014_i686.whl", hash = "sha256:980d7be47c84979d9136328d882f67ec5e50008681d94ecc8afa8a65ed1f4a6f"}, + {file = "regex-2021.4.4-cp38-cp38-manylinux2014_x86_64.whl", hash = "sha256:ce15b6d103daff8e9fee13cf7f0add05245a05d866e73926c358e871221eae87"}, + {file = "regex-2021.4.4-cp38-cp38-win32.whl", hash = "sha256:a91aa8619b23b79bcbeb37abe286f2f408d2f2d6f29a17237afda55bb54e7aac"}, + {file = "regex-2021.4.4-cp38-cp38-win_amd64.whl", hash = "sha256:c0502c0fadef0d23b128605d69b58edb2c681c25d44574fc673b0e52dce71ee2"}, + {file = "regex-2021.4.4-cp39-cp39-macosx_10_9_x86_64.whl", hash = "sha256:598585c9f0af8374c28edd609eb291b5726d7cbce16be6a8b95aa074d252ee17"}, + {file = "regex-2021.4.4-cp39-cp39-manylinux1_i686.whl", hash = "sha256:ee54ff27bf0afaf4c3b3a62bcd016c12c3fdb4ec4f413391a90bd38bc3624605"}, + {file = "regex-2021.4.4-cp39-cp39-manylinux1_x86_64.whl", hash = "sha256:7d9884d86dd4dd489e981d94a65cd30d6f07203d90e98f6f657f05170f6324c9"}, + {file = "regex-2021.4.4-cp39-cp39-manylinux2010_i686.whl", hash = "sha256:bf5824bfac591ddb2c1f0a5f4ab72da28994548c708d2191e3b87dd207eb3ad7"}, + {file = "regex-2021.4.4-cp39-cp39-manylinux2010_x86_64.whl", hash = "sha256:563085e55b0d4fb8f746f6a335893bda5c2cef43b2f0258fe1020ab1dd874df8"}, + {file = "regex-2021.4.4-cp39-cp39-manylinux2014_aarch64.whl", hash = "sha256:b9c3db21af35e3b3c05764461b262d6f05bbca08a71a7849fd79d47ba7bc33ed"}, + {file = "regex-2021.4.4-cp39-cp39-manylinux2014_i686.whl", hash = "sha256:3916d08be28a1149fb97f7728fca1f7c15d309a9f9682d89d79db75d5e52091c"}, + {file = "regex-2021.4.4-cp39-cp39-manylinux2014_x86_64.whl", hash = "sha256:fd45ff9293d9274c5008a2054ecef86a9bfe819a67c7be1afb65e69b405b3042"}, + {file = "regex-2021.4.4-cp39-cp39-win32.whl", hash = "sha256:fa4537fb4a98fe8fde99626e4681cc644bdcf2a795038533f9f711513a862ae6"}, + {file = "regex-2021.4.4-cp39-cp39-win_amd64.whl", hash = "sha256:97f29f57d5b84e73fbaf99ab3e26134e6687348e95ef6b48cfd2c06807005a07"}, + {file = "regex-2021.4.4.tar.gz", hash = "sha256:52ba3d3f9b942c49d7e4bc105bb28551c44065f139a65062ab7912bef10c9afb"}, +] +signxml-openadr = [ + {file = "signxml-openadr-2.9.1.tar.gz", hash = "sha256:93bf41824247a36fa09d8ac23435f5675aaa9e1e58fcb3809a549fba9a905c23"}, + {file = "signxml_openadr-2.9.1-py2.py3-none-any.whl", hash = "sha256:da393d73f2f9739b3dcc6b7cbf3c32e22406370ce714cc490ec303d1db822c48"}, +] +six = [ + {file = "six-1.16.0-py2.py3-none-any.whl", hash = "sha256:8abb2f1d86890a2dfb989f9a77cfcfd3e47c2a354b01111771326f8aa26e0254"}, + {file = "six-1.16.0.tar.gz", hash = "sha256:1e61c37477a1626458e36f7b1d82aa5c9b094fa4802892072e49de9c60c4c926"}, +] +toml = [ + {file = "toml-0.10.2-py2.py3-none-any.whl", hash = "sha256:806143ae5bfb6a3c6e736a764057db0e6a0e05e338b5630894a5f779cabb4f9b"}, + {file = "toml-0.10.2.tar.gz", hash = "sha256:b3bda1d108d5dd99f4a20d24d9c348e91c4db7ab1b749200bded2f839ccbe68f"}, +] +typed-ast = [ + {file = "typed_ast-1.4.3-cp35-cp35m-manylinux1_i686.whl", hash = "sha256:2068531575a125b87a41802130fa7e29f26c09a2833fea68d9a40cf33902eba6"}, + {file = "typed_ast-1.4.3-cp35-cp35m-manylinux1_x86_64.whl", hash = "sha256:c907f561b1e83e93fad565bac5ba9c22d96a54e7ea0267c708bffe863cbe4075"}, + {file = "typed_ast-1.4.3-cp35-cp35m-manylinux2014_aarch64.whl", hash = "sha256:1b3ead4a96c9101bef08f9f7d1217c096f31667617b58de957f690c92378b528"}, + {file = "typed_ast-1.4.3-cp35-cp35m-win32.whl", hash = "sha256:dde816ca9dac1d9c01dd504ea5967821606f02e510438120091b84e852367428"}, + {file = "typed_ast-1.4.3-cp35-cp35m-win_amd64.whl", hash = "sha256:777a26c84bea6cd934422ac2e3b78863a37017618b6e5c08f92ef69853e765d3"}, + {file = "typed_ast-1.4.3-cp36-cp36m-macosx_10_9_x86_64.whl", hash = "sha256:f8afcf15cc511ada719a88e013cec87c11aff7b91f019295eb4530f96fe5ef2f"}, + {file = "typed_ast-1.4.3-cp36-cp36m-manylinux1_i686.whl", hash = "sha256:52b1eb8c83f178ab787f3a4283f68258525f8d70f778a2f6dd54d3b5e5fb4341"}, + {file = "typed_ast-1.4.3-cp36-cp36m-manylinux1_x86_64.whl", hash = "sha256:01ae5f73431d21eead5015997ab41afa53aa1fbe252f9da060be5dad2c730ace"}, + {file = "typed_ast-1.4.3-cp36-cp36m-manylinux2014_aarch64.whl", hash = "sha256:c190f0899e9f9f8b6b7863debfb739abcb21a5c054f911ca3596d12b8a4c4c7f"}, + {file = "typed_ast-1.4.3-cp36-cp36m-win32.whl", hash = "sha256:398e44cd480f4d2b7ee8d98385ca104e35c81525dd98c519acff1b79bdaac363"}, + {file = "typed_ast-1.4.3-cp36-cp36m-win_amd64.whl", hash = "sha256:bff6ad71c81b3bba8fa35f0f1921fb24ff4476235a6e94a26ada2e54370e6da7"}, + {file = "typed_ast-1.4.3-cp37-cp37m-macosx_10_9_x86_64.whl", hash = "sha256:0fb71b8c643187d7492c1f8352f2c15b4c4af3f6338f21681d3681b3dc31a266"}, + {file = "typed_ast-1.4.3-cp37-cp37m-manylinux1_i686.whl", hash = "sha256:760ad187b1041a154f0e4d0f6aae3e40fdb51d6de16e5c99aedadd9246450e9e"}, + {file = "typed_ast-1.4.3-cp37-cp37m-manylinux1_x86_64.whl", hash = "sha256:5feca99c17af94057417d744607b82dd0a664fd5e4ca98061480fd8b14b18d04"}, + {file = "typed_ast-1.4.3-cp37-cp37m-manylinux2014_aarch64.whl", hash = "sha256:95431a26309a21874005845c21118c83991c63ea800dd44843e42a916aec5899"}, + {file = "typed_ast-1.4.3-cp37-cp37m-win32.whl", hash = "sha256:aee0c1256be6c07bd3e1263ff920c325b59849dc95392a05f258bb9b259cf39c"}, + {file = "typed_ast-1.4.3-cp37-cp37m-win_amd64.whl", hash = "sha256:9ad2c92ec681e02baf81fdfa056fe0d818645efa9af1f1cd5fd6f1bd2bdfd805"}, + {file = "typed_ast-1.4.3-cp38-cp38-macosx_10_9_x86_64.whl", hash = "sha256:b36b4f3920103a25e1d5d024d155c504080959582b928e91cb608a65c3a49e1a"}, + {file = "typed_ast-1.4.3-cp38-cp38-manylinux1_i686.whl", hash = "sha256:067a74454df670dcaa4e59349a2e5c81e567d8d65458d480a5b3dfecec08c5ff"}, + {file = "typed_ast-1.4.3-cp38-cp38-manylinux1_x86_64.whl", hash = "sha256:7538e495704e2ccda9b234b82423a4038f324f3a10c43bc088a1636180f11a41"}, + {file = "typed_ast-1.4.3-cp38-cp38-manylinux2014_aarch64.whl", hash = "sha256:af3d4a73793725138d6b334d9d247ce7e5f084d96284ed23f22ee626a7b88e39"}, + {file = "typed_ast-1.4.3-cp38-cp38-win32.whl", hash = "sha256:f2362f3cb0f3172c42938946dbc5b7843c2a28aec307c49100c8b38764eb6927"}, + {file = "typed_ast-1.4.3-cp38-cp38-win_amd64.whl", hash = "sha256:dd4a21253f42b8d2b48410cb31fe501d32f8b9fbeb1f55063ad102fe9c425e40"}, + {file = "typed_ast-1.4.3-cp39-cp39-macosx_10_9_x86_64.whl", hash = "sha256:f328adcfebed9f11301eaedfa48e15bdece9b519fb27e6a8c01aa52a17ec31b3"}, + {file = "typed_ast-1.4.3-cp39-cp39-manylinux1_i686.whl", hash = "sha256:2c726c276d09fc5c414693a2de063f521052d9ea7c240ce553316f70656c84d4"}, + {file = "typed_ast-1.4.3-cp39-cp39-manylinux1_x86_64.whl", hash = "sha256:cae53c389825d3b46fb37538441f75d6aecc4174f615d048321b716df2757fb0"}, + {file = "typed_ast-1.4.3-cp39-cp39-manylinux2014_aarch64.whl", hash = "sha256:b9574c6f03f685070d859e75c7f9eeca02d6933273b5e69572e5ff9d5e3931c3"}, + {file = "typed_ast-1.4.3-cp39-cp39-win32.whl", hash = "sha256:209596a4ec71d990d71d5e0d312ac935d86930e6eecff6ccc7007fe54d703808"}, + {file = "typed_ast-1.4.3-cp39-cp39-win_amd64.whl", hash = "sha256:9c6d1a54552b5330bc657b7ef0eae25d00ba7ffe85d9ea8ae6540d2197a3788c"}, + {file = "typed_ast-1.4.3.tar.gz", hash = "sha256:fb1bbeac803adea29cedd70781399c99138358c26d05fcbd23c13016b7f5ec65"}, +] +typing-extensions = [ + {file = "typing_extensions-3.10.0.0-py2-none-any.whl", hash = "sha256:0ac0f89795dd19de6b97debb0c6af1c70987fd80a2d62d1958f7e56fcc31b497"}, + {file = "typing_extensions-3.10.0.0-py3-none-any.whl", hash = "sha256:779383f6086d90c99ae41cf0ff39aac8a7937a9283ce0a414e5dd782f4c94a84"}, + {file = "typing_extensions-3.10.0.0.tar.gz", hash = "sha256:50b6f157849174217d0656f99dc82fe932884fb250826c18350e159ec6cdf342"}, +] +tzlocal = [ + {file = "tzlocal-2.1-py2.py3-none-any.whl", hash = "sha256:e2cb6c6b5b604af38597403e9852872d7f534962ae2954c7f35efcb1ccacf4a4"}, + {file = "tzlocal-2.1.tar.gz", hash = "sha256:643c97c5294aedc737780a49d9df30889321cbe1204eac2c2ec6134035a92e44"}, +] +virtualenv = [ + {file = "virtualenv-20.4.7-py2.py3-none-any.whl", hash = "sha256:2b0126166ea7c9c3661f5b8e06773d28f83322de7a3ff7d06f0aed18c9de6a76"}, + {file = "virtualenv-20.4.7.tar.gz", hash = "sha256:14fdf849f80dbb29a4eb6caa9875d476ee2a5cf76a5f5415fa2f1606010ab467"}, +] +xmltodict = [ + {file = "xmltodict-0.12.0-py2.py3-none-any.whl", hash = "sha256:8bbcb45cc982f48b2ca8fe7e7827c5d792f217ecf1792626f808bf41c3b86051"}, + {file = "xmltodict-0.12.0.tar.gz", hash = "sha256:50d8c638ed7ecb88d90561beedbf720c9b4e851a9fa6c47ebd64e99d166d8a21"}, +] +yarl = [ + {file = "yarl-1.6.3-cp36-cp36m-macosx_10_14_x86_64.whl", hash = "sha256:0355a701b3998dcd832d0dc47cc5dedf3874f966ac7f870e0f3a6788d802d434"}, + {file = "yarl-1.6.3-cp36-cp36m-manylinux1_i686.whl", hash = "sha256:bafb450deef6861815ed579c7a6113a879a6ef58aed4c3a4be54400ae8871478"}, + {file = "yarl-1.6.3-cp36-cp36m-manylinux2014_aarch64.whl", hash = "sha256:547f7665ad50fa8563150ed079f8e805e63dd85def6674c97efd78eed6c224a6"}, + {file = "yarl-1.6.3-cp36-cp36m-manylinux2014_i686.whl", hash = "sha256:63f90b20ca654b3ecc7a8d62c03ffa46999595f0167d6450fa8383bab252987e"}, + {file = "yarl-1.6.3-cp36-cp36m-manylinux2014_ppc64le.whl", hash = "sha256:97b5bdc450d63c3ba30a127d018b866ea94e65655efaf889ebeabc20f7d12406"}, + {file = "yarl-1.6.3-cp36-cp36m-manylinux2014_s390x.whl", hash = "sha256:d8d07d102f17b68966e2de0e07bfd6e139c7c02ef06d3a0f8d2f0f055e13bb76"}, + {file = "yarl-1.6.3-cp36-cp36m-manylinux2014_x86_64.whl", hash = "sha256:15263c3b0b47968c1d90daa89f21fcc889bb4b1aac5555580d74565de6836366"}, + {file = "yarl-1.6.3-cp36-cp36m-win32.whl", hash = "sha256:b5dfc9a40c198334f4f3f55880ecf910adebdcb2a0b9a9c23c9345faa9185721"}, + {file = "yarl-1.6.3-cp36-cp36m-win_amd64.whl", hash = "sha256:b2e9a456c121e26d13c29251f8267541bd75e6a1ccf9e859179701c36a078643"}, + {file = "yarl-1.6.3-cp37-cp37m-macosx_10_14_x86_64.whl", hash = "sha256:ce3beb46a72d9f2190f9e1027886bfc513702d748047b548b05dab7dfb584d2e"}, + {file = "yarl-1.6.3-cp37-cp37m-manylinux1_i686.whl", hash = "sha256:2ce4c621d21326a4a5500c25031e102af589edb50c09b321049e388b3934eec3"}, + {file = "yarl-1.6.3-cp37-cp37m-manylinux2014_aarch64.whl", hash = "sha256:d26608cf178efb8faa5ff0f2d2e77c208f471c5a3709e577a7b3fd0445703ac8"}, + {file = "yarl-1.6.3-cp37-cp37m-manylinux2014_i686.whl", hash = "sha256:4c5bcfc3ed226bf6419f7a33982fb4b8ec2e45785a0561eb99274ebbf09fdd6a"}, + {file = "yarl-1.6.3-cp37-cp37m-manylinux2014_ppc64le.whl", hash = "sha256:4736eaee5626db8d9cda9eb5282028cc834e2aeb194e0d8b50217d707e98bb5c"}, + {file = "yarl-1.6.3-cp37-cp37m-manylinux2014_s390x.whl", hash = "sha256:68dc568889b1c13f1e4745c96b931cc94fdd0defe92a72c2b8ce01091b22e35f"}, + {file = "yarl-1.6.3-cp37-cp37m-manylinux2014_x86_64.whl", hash = "sha256:7356644cbed76119d0b6bd32ffba704d30d747e0c217109d7979a7bc36c4d970"}, + {file = "yarl-1.6.3-cp37-cp37m-win32.whl", hash = "sha256:00d7ad91b6583602eb9c1d085a2cf281ada267e9a197e8b7cae487dadbfa293e"}, + {file = "yarl-1.6.3-cp37-cp37m-win_amd64.whl", hash = "sha256:69ee97c71fee1f63d04c945f56d5d726483c4762845400a6795a3b75d56b6c50"}, + {file = "yarl-1.6.3-cp38-cp38-macosx_10_14_x86_64.whl", hash = "sha256:e46fba844f4895b36f4c398c5af062a9808d1f26b2999c58909517384d5deda2"}, + {file = "yarl-1.6.3-cp38-cp38-manylinux1_i686.whl", hash = "sha256:31ede6e8c4329fb81c86706ba8f6bf661a924b53ba191b27aa5fcee5714d18ec"}, + {file = "yarl-1.6.3-cp38-cp38-manylinux2014_aarch64.whl", hash = "sha256:fcbb48a93e8699eae920f8d92f7160c03567b421bc17362a9ffbbd706a816f71"}, + {file = "yarl-1.6.3-cp38-cp38-manylinux2014_i686.whl", hash = "sha256:72a660bdd24497e3e84f5519e57a9ee9220b6f3ac4d45056961bf22838ce20cc"}, + {file = "yarl-1.6.3-cp38-cp38-manylinux2014_ppc64le.whl", hash = "sha256:324ba3d3c6fee56e2e0b0d09bf5c73824b9f08234339d2b788af65e60040c959"}, + {file = "yarl-1.6.3-cp38-cp38-manylinux2014_s390x.whl", hash = "sha256:e6b5460dc5ad42ad2b36cca524491dfcaffbfd9c8df50508bddc354e787b8dc2"}, + {file = "yarl-1.6.3-cp38-cp38-manylinux2014_x86_64.whl", hash = "sha256:6d6283d8e0631b617edf0fd726353cb76630b83a089a40933043894e7f6721e2"}, + {file = "yarl-1.6.3-cp38-cp38-win32.whl", hash = "sha256:9ede61b0854e267fd565e7527e2f2eb3ef8858b301319be0604177690e1a3896"}, + {file = "yarl-1.6.3-cp38-cp38-win_amd64.whl", hash = "sha256:f0b059678fd549c66b89bed03efcabb009075bd131c248ecdf087bdb6faba24a"}, + {file = "yarl-1.6.3-cp39-cp39-macosx_10_14_x86_64.whl", hash = "sha256:329412812ecfc94a57cd37c9d547579510a9e83c516bc069470db5f75684629e"}, + {file = "yarl-1.6.3-cp39-cp39-manylinux1_i686.whl", hash = "sha256:c49ff66d479d38ab863c50f7bb27dee97c6627c5fe60697de15529da9c3de724"}, + {file = "yarl-1.6.3-cp39-cp39-manylinux2014_aarch64.whl", hash = "sha256:f040bcc6725c821a4c0665f3aa96a4d0805a7aaf2caf266d256b8ed71b9f041c"}, + {file = "yarl-1.6.3-cp39-cp39-manylinux2014_i686.whl", hash = "sha256:d5c32c82990e4ac4d8150fd7652b972216b204de4e83a122546dce571c1bdf25"}, + {file = "yarl-1.6.3-cp39-cp39-manylinux2014_ppc64le.whl", hash = "sha256:d597767fcd2c3dc49d6eea360c458b65643d1e4dbed91361cf5e36e53c1f8c96"}, + {file = "yarl-1.6.3-cp39-cp39-manylinux2014_s390x.whl", hash = "sha256:8aa3decd5e0e852dc68335abf5478a518b41bf2ab2f330fe44916399efedfae0"}, + {file = "yarl-1.6.3-cp39-cp39-manylinux2014_x86_64.whl", hash = "sha256:73494d5b71099ae8cb8754f1df131c11d433b387efab7b51849e7e1e851f07a4"}, + {file = "yarl-1.6.3-cp39-cp39-win32.whl", hash = "sha256:5b883e458058f8d6099e4420f0cc2567989032b5f34b271c0827de9f1079a424"}, + {file = "yarl-1.6.3-cp39-cp39-win_amd64.whl", hash = "sha256:4953fb0b4fdb7e08b2f3b3be80a00d28c5c8a2056bb066169de00e6501b986b6"}, + {file = "yarl-1.6.3.tar.gz", hash = "sha256:8a9066529240171b68893d60dca86a763eae2139dd42f42106b03cf4b426bf10"}, +] diff --git a/pyproject.toml b/pyproject.toml new file mode 100644 index 0000000..0343e3c --- /dev/null +++ b/pyproject.toml @@ -0,0 +1,19 @@ +[tool.poetry] +name = "volttron-openadr-ven" +version = "0.1.0" +description = "A Volttron agent that acts as a Virtual End Node (VEN) within the OpenADR 2.0b specification." +authors = ["Mark Bonicillo <[email protected]>"] + +[tool.poetry.dependencies] +python = "^3.8" +openleadr = "^0.5.23" +black = "^21.5b2" +mypy = "^0.812" +pre-commit = "^2.13.0" + +[tool.poetry.dev-dependencies] +pytest = "^6.2.4" + +[build-system] +requires = ["poetry-core>=1.0.0"] +build-backend = "poetry.core.masonry.api" diff --git a/volttron_client-0.1.2-py3-none-any.whl b/volttron_client-0.1.2-py3-none-any.whl new file mode 100644 index 0000000..870e627 Binary files /dev/null and b/volttron_client-0.1.2-py3-none-any.whl differ diff --git a/volttron_openadr_ven/agent.py b/volttron_openadr_ven/agent.py new file mode 100644 index 0000000..c79e9de --- /dev/null +++ b/volttron_openadr_ven/agent.py @@ -0,0 +1,321 @@ +# -- coding: utf-8 -- {{{ +# vim: set fenc=utf-8 ft=python sw=4 ts=4 sts=4 et: +# +# Copyright 2020, Battelle Memorial Institute. +# +# Licensed under the Apache License, Version 2.0 (the "License"); +# you may not use this file except in compliance with the License. +# You may obtain a copy of the License at +# +# http://www.apache.org/licenses/LICENSE-2.0 +# +# Unless required by applicable law or agreed to in writing, software +# distributed under the License is distributed on an "AS IS" BASIS, +# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. +# See the License for the specific language governing permissions and +# limitations under the License. +# +# This material was prepared as an account of work sponsored by an agency of +# the United States Government. Neither the United States Government nor the +# United States Department of Energy, nor Battelle, nor any of their +# employees, nor any jurisdiction or organization that has cooperated in the +# development of these materials, makes any warranty, express or +# implied, or assumes any legal liability or responsibility for the accuracy, +# completeness, or usefulness or any information, apparatus, product, +# software, or process disclosed, or represents that its use would not infringe +# privately owned rights. Reference herein to any specific commercial product, +# process, or service by trade name, trademark, manufacturer, or otherwise +# does not necessarily constitute or imply its endorsement, recommendation, or +# favoring by the United States Government or any agency thereof, or +# Battelle Memorial Institute. The views and opinions of authors expressed +# herein do not necessarily state or reflect those of the +# United States Government or any agency thereof. +# +# PACIFIC NORTHWEST NATIONAL LABORATORY operated by +# BATTELLE for the UNITED STATES DEPARTMENT OF ENERGY +# under Contract DE-AC05-76RL01830 +# }}} + +import logging +import asyncio +import sys +import openleadr.enums + +from pprint import pformat +from typing import Dict +from datetime import datetime + +from openleadr.enums import OPT +from openleadr.objects import Event +from openleadr import OpenADRClient + +from volttron.platform.agent import utils +from volttron.platform.vip.agent import Agent, Core +from volttron.platform.messaging import topics, headers +from volttron.utils import jsonapi + +utils.setup_logging() +_log = logging.getLogger(__name__) +__version__ = "1.0" + + +class OpenADRVenAgent(Agent): + """ + OpenADR (Automated Demand Response) is a standard for alerting and responding + to the need to adjust electric power consumption in response to fluctuations in grid demand. + OpenADR communications are conducted between Virtual Top Nodes (VTNs) and Virtual End Nodes (VENs). + This agent is a wrapper around OpenLEADR, an open-source implementation of OpenADR 2.0.b for both servers, VTN, and clients, VEN. + This agent creates an instance of OpenLEADR's VEN client, which is used to communicated with a VTN. + """ + + def __init__( + self, + ven_name: str, + vtn_url: str, + debug: bool, + cert: str, + key: str, + fingerprint: str, + show_fingerprint: str, + ca_file: str, + kwargs, + ) -> None: + super(OpenADRVenAgent, self).__init__(enable_web=True, kwargs) + self.ven_name = None + self.vtn_url = None + self.debug = None + self.cert = None + self.key = None + self.fingerprint = None + self.show_fingerprint = None + self.ca_file = None + self.ven_id = None + + self.default_config = { + "ven_name": ven_name, + "vtn_url": vtn_url, + "debug": debug, + "cert": cert, + "key": key, + "fingerprint": fingerprint, + "show_fingerprint": show_fingerprint, + "ca_file": ca_file, + } + self.vip.config.set_default("config", self.default_config) + self.vip.config.subscribe( + self._configure, actions=["NEW", "UPDATE"], pattern="config" + ) + + # properties of the this class will be initialized in initialize_config + # self.ven_client is an instance of OpenLeadr's OpenADRClient, an implementation of OpenADR; + # OpenADRClient will be instantiated in initialize_config + self.initialize_config(self.default_config) + + def initialize_config(self, config: Dict) -> None: + """ + Initialize the agent's configuration. Create an OpenADR Client using OpenLeadr. + Configuration parameters (see config for a sample config file): + + str ven_name: The name for this VEN + str vtn_url: The URL of the VTN (Server) to connect to + bool debug: Whether or not to print debugging messages + str cert: The path to a PEM-formatted Certificate file to use + for signing messages. + str key: The path to a PEM-formatted Private Key file to use + for signing messages. + str fingerprint: The fingerprint for the VTN's certificate to + verify incomnig messages + str show_fingerprint: Whether to print your own fingerprint + on startup. Defaults to True. + str ca_file: The path to the PEM-formatted CA file for validating the VTN server's + certificate. + """ + _log.debug("Configuring agent") + self.ven_name = config.get("ven_name") + self.vtn_url = config.get("vtn_url") + self.debug = config.get("debug", False) + self.cert = config.get("cert") + self.key = config.get("key") + self.fingerprint = config.get("fingerprint") + self.show_fingerprint = config.get("show_fingerprint", True) + self.ca_file = config.get("ca_file") + + _log.info("Configuration parameters:") + _log.info(pformat(config)) + + # instantiate and add handlers to the OpenADR Client + self.ven_client = OpenADRClient( + ven_name=self.ven_name, + vtn_url=self.vtn_url, + debug=self.debug, + cert=self.cert, + key=self.key, + show_fingerprint=self.show_fingerprint, + ca_file=self.ca_file, + ) + + # if you want to add more handlers on a specific event, you must create a coroutine + # in this class and then add it as the second input for 'self.ven_client.add_handler(<some event>, <coroutine>)' + self.ven_client.add_handler("on_event", self.handle_event) + + def _configure(self, contents: Dict) -> None: + """The agent's config may have changed. Re-initialize it.""" + config = self.default_config.copy() + config.update(contents) + self.initialize_config(config) + + # ************** Methods for Managing the Agent on Volttron **************** + + @Core.receiver("onstart") + def onstart(self) -> None: + """The agent has started.""" + _log.debug("Starting agent") + + # Running the ven_client in the Python AsyncIO Event Loop + loop = asyncio.get_event_loop() + loop.create_task(self.ven_client.run()) + loop.run_forever() + + # # TODO: Identify actions needed to be done before shutdown + # @Core.receiver("onstop") + # def onstop(self, sender, kwargs): + # """ + # This method is called when the Agent is about to shutdown, but before it disconnects from + # the message bus. + # """ + # pass + # + # # TODO: Identify what, if at all, RPC methods should be available to other Agents + # @RPC.export + # def rpc_method(self, arg1, arg2, kwarg1=None, kwarg2=None): + # """ + # RPC method + # + # May be called from another agent via self.vip.rpc.call + # """ + # pass + + # *************** Methods for Servicing VTN Requests **************** + + async def handle_event(self, event: openleadr.objects.Event) -> openleadr.enums.OPT: + """ + Publish event to the Volttron message bus. Return OPT response. + This coroutine will be called when there is an event to be handled. + + :param event: Event + For more info on shape of 'event', see https://openleadr.org/docs/client.html#dealing-with-events and https://openleadr.org/docs/api/openleadr.html#openleadr.objects.Event + Example of event: + { + 'event_id': '123786-129831', + 'active_period': {'dtstart': datetime.datetime(2020, 1, 1, 12, 0, 0, tzinfo=timezone.utc), + 'duration': datetime.timedelta(minutes=30)} + 'event_signals': [{'signal_name': 'simple', + 'signal_type': 'level', + 'intervals': [{'dtstart': datetime.datetime(2020, 1, 1, 12, 0, 0, tzinfo=timezone.utc), + 'duration': datetime.timedelta(minutes=10), + 'signal_payload': 1}, + {'dtstart': datetime.datetime(2020, 1, 1, 12, 10, 0, tzinfo=timezone.utc), + 'duration': datetime.timedelta(minutes=10), + 'signal_payload': 0}, + {'dtstart': datetime.datetime(2020, 1, 1, 12, 20, 0, tzinfo=timezone.utc), + 'duration': datetime.timedelta(minutes=10), + 'signal_payload': 1}], + 'targets': [{'resource_id': 'Device001'}], + 'targets_by_type': {'resource_id': ['Device001']} + } + :return: openleadr.enums.OPT + """ + _log.info(f"Event: {pformat(event)}") + self.publish_event(event) + return OPT.OPT_IN + + # ************* VOLTTRON Pub/Sub Requests **************** + def publish_event(self, event: openleadr.objects.Event) -> None: + """Publish an event to the Volttron message bus. When an event is created/updated, it is published to the VOLTTRON bus with a topic that includes 'openadr/event_update'. + :param event: + :return: None + """ + # OADR rule 6: If testEvent is present and != "false", handle the event as a test event. + if event.event_descriptor.test_event != "false": + _log.debug(f"Suppressing publication of test event {event}") + return + + _log.debug("Publishing event {}".format(event)) + self.vip.pubsub.publish( + peer="pubsub", + topic=f"{topics.OPENADR_EVENT}/{self.ven_id}", + headers={ + headers.TIMESTAMP: utils.format_timestamp(utils.get_aware_utc_now()) + }, + message=self._json_object(event.__dict__), + ) + return + + # ************* Helper methods **************** + def _json_object(self, obj: Dict): + """Ensure that an object is valid JSON by dumping it with json_converter and then reloading it.""" + obj_string = jsonapi.dumps( + obj, default=lambda x: x.__str__() if isinstance(x, datetime) else None + ) + return jsonapi.loads(obj_string) + + +def main(): + """Main method called to start the agent.""" + utils.vip_main(ven_agent, version=__version__) + + +if __name__ == "__main__": + try: + sys.exit(main()) + except KeyboardInterrupt: + pass + + +def ven_agent(config_path: str, kwargs) -> OpenADRVenAgent: + """ + Parse the OpenADRVenAgent configuration file and return an instance of + the agent that has been created using that configuration. + See initialize_config() method documentation for a description of each configurable parameter. + + :param config_path: (str) Path to a configuration file. + + :returns: OpenADRVenAgent instance + """ + try: + config = utils.load_config(config_path) + except Exception as err: + _log.error("Error loading configuration: {}".format(err)) + config = {} + + if not config: + raise Exception("Configuration cannot be empty.") + + ven_name = config.get("ven_name") + if not ven_name: + raise Exception("ven_name is required.") + vtn_url = config.get("vtn_url") + if not vtn_url: + raise Exception( + "vtn_url is required. Ensure vtn_url is given a URL to the VTN." + ) + + # below are optional configurations + debug = config.get("debug") + cert = config.get("cert") + key = config.get("key") + fingerprint = config.get("fingerprint") + show_fingerprint = config.get("show_fingerprint") + ca_file = config.get("ca_file") + + return OpenADRVenAgent( + ven_name, + vtn_url, + debug, + cert, + key, + fingerprint, + show_fingerprint, + ca_file, + **kwargs, + ) diff --git a/volttron_utils-0.1.1-py3-none-any.whl b/volttron_utils-0.1.1-py3-none-any.whl new file mode 100644 index 0000000..6512925 Binary files /dev/null and b/volttron_utils-0.1.1-py3-none-any.whl differ	Create Skeleton Agent Structure This skeleton will have volttron agent properties as well as being able to start the async loop and be able to call the openleadr source code https://pypi.org/project/openleadr/	Additional Notes: * This agent will take the following dependencies: * https://github.com/VOLTTRON/volttron-client/tree/main/volttron/platform * https://github.com/VOLTTRON/volttron-utils * This agent is a pared-down version of OpenADRAgent in volttron core; see https://github.com/VOLTTRON/volttron/blob/develop/services/core/OpenADRVenAgent/openadrven/agent.py#L414 * Agent should have the following methods * publish_event * process_event * process_report * Agent should ideally be a gevent-based agent that can make asyn calls; see example from openleader: https://openleadr.org/docs/#getting-started * Agent should be packaged and able to be installed via wheel on a volttron platform using `vctl install`	2021-05-30T19:53:10	[]	[]
octue/django-gcp	octue__django-gcp-40	99d1932865eb9bb8ea765bbab57c5c6e3b9416e4	diff --git a/.devcontainer/devcontainer.json b/.devcontainer/devcontainer.json index 0bfdd36..89c9842 100644 --- a/.devcontainer/devcontainer.json +++ b/.devcontainer/devcontainer.json @@ -71,7 +71,8 @@ "thebarkman.vscode-djaneiro", "trond-snekvik.simple-rst", "ms-azuretools.vscode-docker", - "ryanluker.vscode-coverage-gutters" + "ryanluker.vscode-coverage-gutters", + "ms-python.black-formatter" ] } }, diff --git a/.pre-commit-config.yaml b/.pre-commit-config.yaml index eada270..395bc36 100644 --- a/.pre-commit-config.yaml +++ b/.pre-commit-config.yaml @@ -20,7 +20,7 @@ repos: - id: check-json5 - repo: https://github.com/psf/black - rev: 22.3.0 + rev: 22.10.0 hooks: - id: black args: ['--line-length', '120'] diff --git a/django_gcp/__init__.py b/django_gcp/__init__.py index 3dd48a2..d73c30f 100644 --- a/django_gcp/__init__.py +++ b/django_gcp/__init__.py @@ -1,1 +1,1 @@ -default_app_config = "django_gcp.apps.DjangoGcpAppConfig" +default_app_config = "django_gcp.apps.DjangoGCPAppConfig" diff --git a/django_gcp/storage/fields.py b/django_gcp/storage/fields.py index 76dd6f0..7aec214 100644 --- a/django_gcp/storage/fields.py +++ b/django_gcp/storage/fields.py @@ -59,8 +59,8 @@ class BlobField(models.JSONField): We do this to support a clear and maintainable implementation. :param ingress_to: A string defining the path within the bucket to which direct uploads - will be ingressed, if temporary_ingress=True. These uploaded files will be moved to their - ultimate path in the store on save of the model. + will be ingressed. These uploaded files will be moved to their ultimate path in the store on save of the model. + :param accept_mimetype: A string passed to widgets, suitable for use in the `accept` attribute of an html filepicker. This will allow you to narrow down, eg to `image/` or an even more specific mimetype. No validation is done at the field level that objects actually are of this @@ -77,7 +77,7 @@ class BlobField(models.JSONField): :param max_size_bytes: The maximum size in bytes of files that can be uploaded. - :overwrite_mode: One of `OVERWRITE_MODES` determining the circumstances under which overwrite + :param overwrite_mode: One of `OVERWRITE_MODES` determining the circumstances under which overwrite is allowed. overwrite mode behaves as follows: - never: Never allows overwrite - update: Only when updating an object @@ -86,6 +86,10 @@ class BlobField(models.JSONField): - add-versioned: Only when adding an object to a versioned bucket (again, for completeness) - add-update: Always allow (ie when adding or updating an object) - add-update-versioned: When adding or updating an object to a versioned bucket + + :param on_change: A callable that will be executed on change of the field value. This will be called + on commit of the transaction (ie once any file upload is ingressed to its final location) and allows you, + for example, to dispatch a worker task to further process the uploaded blob. """ description = _("A GCP Cloud Storage object") @@ -99,6 +103,7 @@ def __init__( accept_mimetype=DEFAULT_ACCEPT_MIMETYPE, max_size_bytes=DEFAULT_MAX_SIZE_BYTES, overwrite_mode=DEFAULT_OVERWRITE_MODE, + on_change=None, kwargs, ): self._versioning_enabled = None @@ -112,6 +117,7 @@ def __init__( self.store_key = store_key self.accept_mimetype = accept_mimetype self.max_size_bytes = max_size_bytes + self.on_change = on_change kwargs["default"] = kwargs.pop("default", None) kwargs["help_text"] = kwargs.pop("help_text", "GCP cloud storage object") @@ -144,6 +150,7 @@ def check(self, kwargs): self._check_get_destination_path(), self._check_ingress_to(), self._check_overwrite_mode(), + self._check_on_change(), ] def deconstruct(self): @@ -153,6 +160,7 @@ def deconstruct(self): kwargs["accept_mimetype"] = self.accept_mimetype kwargs["overwrite_mode"] = self.overwrite_mode kwargs["get_destination_path"] = self.get_destination_path + kwargs["on_change"] = self.on_change return name, path, args, kwargs def formfield(self, kwargs): @@ -172,7 +180,11 @@ def formfield(self, kwargs): @property def override_blobfield_value(self): """Shortcut to access the GCP_STORAGE_OVERRIDE_BLOBFIELD_VALUE setting""" - return getattr(settings, "GCP_STORAGE_OVERRIDE_BLOBFIELD_VALUE", DEFAULT_OVERRIDE_BLOBFIELD_VALUE) + return getattr( + settings, + "GCP_STORAGE_OVERRIDE_BLOBFIELD_VALUE", + DEFAULT_OVERRIDE_BLOBFIELD_VALUE, + ) def pre_save(self, model_instance, add): """Return field's value just before saving.""" @@ -184,7 +196,12 @@ def pre_save(self, model_instance, add): # explicitly for the purpose of data migration and manipulation. You should never allow an untrusted # client to set paths directly, because knowing the path of a pre-existing object allows you to assume # access to it. Tip: You can use django's override_settings context manager to set this temporarily. - if getattr(settings, "GCP_STORAGE_OVERRIDE_BLOBFIELD_VALUE", DEFAULT_OVERRIDE_BLOBFIELD_VALUE): + # Note that you'll have to execute any on_changed + if getattr( + settings, + "GCP_STORAGE_OVERRIDE_BLOBFIELD_VALUE", + DEFAULT_OVERRIDE_BLOBFIELD_VALUE, + ): logger.warning( "Overriding %s value to %s", self.__class__.__name__, @@ -206,8 +223,15 @@ def pre_save(self, model_instance, add): elif adding_blank or updating_valid_to_blank: new_value = None - elif adding_valid or updating_blank_to_valid or updating_valid_to_valid: + # Trigger the on_change callback at the end of the commit when we know the + # database transaction will work + def on_commit_blank(): + if self.on_change is not None: + self.on_change(new_value, instance=model_instance) + + transaction.on_commit(on_commit_blank) + elif adding_valid or updating_blank_to_valid or updating_valid_to_valid: new_value = {} new_value["path"], allow_overwrite = self.get_destination_path( instance=model_instance, @@ -228,8 +252,8 @@ def pre_save(self, model_instance, add): # capture the dual edge cases of the file not moving correctly, and the database # row not saving (eg due to validation errors in other model fields). # https://stackoverflow.com/questions/33180727/trigering-post-save-signal-only-after-transaction-has-completed - transaction.on_commit( - lambda: copy_blob( + def on_commit_valid(): + copy_blob( self.storage.bucket, value["_tmp_path"], self.storage.bucket, @@ -238,9 +262,15 @@ def pre_save(self, model_instance, add): overwrite=allow_overwrite, attributes=value.get("attributes", None), ) - ) + if self.on_change is not None: + self.on_change(new_value, instance=model_instance) + + transaction.on_commit(on_commit_valid) + logger.info( - "Registered move of %s to %s to happen on transaction commit", value["_tmp_path"], new_value["path"] + "Registered move of %s to %s to happen on transaction commit", + value["_tmp_path"], + new_value["path"], ) else: @@ -407,6 +437,17 @@ def _check_get_destination_path(self): ] return [] + def _check_on_change(self): + if self.on_change is not None and not callable(self.on_change): + return [ + checks.Error( + f"'on_change' argument in {self.__class__.__name__} must be or a callable function, or None", + obj=self, + id="fields.E201", + ) + ] + return [] + def _get_allow_overwrite(self, add): """Return a boolean determining if overwrite should be allowed for this operation""" mode_map = { diff --git a/django_gcp/storage/forms.py b/django_gcp/storage/forms.py index 6748e67..a16a0dc 100644 --- a/django_gcp/storage/forms.py +++ b/django_gcp/storage/forms.py @@ -12,59 +12,3 @@ class CloudObjectFormField(JSONField): empty_value = None widget = CloudObjectWidget - - -# from django.forms import FileField -# from django.core.files.uploadedfile import UploadedFile -# from django.core.exceptions import ValidationError - -# from .direct_upload_widgets import DirectUploadWidget -# class DirectUploadedFile(UploadedFile): -# """Represents a file that was uploaded direct to bucket""" - - -# class DirectUploadFormField(FileField): -# """Overrides the usual forms.FileField with a widget that handles direct uploads""" - -# widget = DirectUploadWidget - -# def to_python(self, data): -# """Override the FileField to_python method to accept data as a string file name""" - -# if data in self.empty_values: -# return None - -# # A file was attached as an indirect upload (the normal way) -# if isinstance(data, UploadedFile): -# try: -# file_name = data.name -# file_size = data.size -# except AttributeError: -# # pylint: disable=raise-missing-from -# raise ValidationError(self.error_messages["invalid"], code="invalid") -# # pylint: enable=raise-missing-from - -# if not self.allow_empty_file and not file_size: -# raise ValidationError(self.error_messages["empty"], code="empty") - -# # A file was uploaded direct-to-bucket, data only contains its name -# elif isinstance(data, str): -# file_name = data -# file_size = None -# data = DirectUploadedFile(name=file_name, size=None) -# if not self.allow_empty_file: -# logger.warning( -# "`allow_empty_file` is set but file has been directly uploaded. Cannot validate emptiness of file." -# ) - -# else: -# raise ValueError(f"Unknown data of class {data.__class__}: {data}") -# print("FILE NAME", file_name) -# if self.max_length is not None and len(file_name) > self.max_length: -# params = {"max": self.max_length, "length": len(file_name)} -# raise ValidationError(self.error_messages["max_length"], code="max_length", params=params) - -# if not file_name: -# raise ValidationError(self.error_messages["invalid"], code="invalid") - -# return data diff --git a/django_gcp/storage/gcloud.py b/django_gcp/storage/gcloud.py index 251a153..108586b 100644 --- a/django_gcp/storage/gcloud.py +++ b/django_gcp/storage/gcloud.py @@ -1,6 +1,7 @@ import logging import mimetypes from tempfile import SpooledTemporaryFile +from django.conf import settings from django.core.exceptions import SuspiciousOperation from django.core.files.base import File from django.core.files.storage import Storage @@ -12,7 +13,7 @@ from .compress import CompressedFileMixin, CompressStorageMixin from .settings import StorageSettings -from .utils import clean_name, get_available_overwrite_name, safe_join, setting, to_bytes +from .utils import clean_name, get_available_overwrite_name, safe_join, to_bytes CONTENT_ENCODING = "content_encoding" @@ -44,7 +45,7 @@ def _get_file(self): self._file = SpooledTemporaryFile( max_size=self._storage.settings.max_memory_size, suffix=".GSStorageFile", - dir=setting("FILE_UPLOAD_TEMP_DIR"), + dir=getattr(settings, "FILE_UPLOAD_TEMP_DIR"), ) if "r" in self._mode: self._is_dirty = False @@ -286,7 +287,7 @@ def get_modified_time(self, name): name = self._normalize_name(clean_name(name)) blob = self._get_blob(name) updated = blob.updated - return updated if setting("USE_TZ") else timezone.make_naive(updated) + return updated if getattr(settings, "USE_TZ") else timezone.make_naive(updated) def get_created_time(self, name): """ @@ -296,7 +297,7 @@ def get_created_time(self, name): name = self._normalize_name(clean_name(name)) blob = self._get_blob(name) created = blob.time_created - return created if setting("USE_TZ") else timezone.make_naive(created) + return created if getattr(settings, "USE_TZ") else timezone.make_naive(created) def url(self, name): """ diff --git a/django_gcp/storage/utils.py b/django_gcp/storage/utils.py index 8420b54..e36d141 100644 --- a/django_gcp/storage/utils.py +++ b/django_gcp/storage/utils.py @@ -1,6 +1,5 @@ import os import posixpath -from django.conf import settings from django.core.exceptions import SuspiciousFileOperation from django.utils.encoding import force_bytes @@ -13,37 +12,24 @@ def to_bytes(content): return force_bytes(content) -def setting(name, default=None): - """ - Helper function to get a Django setting by name. If setting doesn't exists - it will return a default. - - :param name: Name of setting - :type name: str - :param default: Value if setting is unfound - :returns: Setting's value - """ - return getattr(settings, name, default) - - def clean_name(name): """ Cleans the name so that Windows style paths work """ # Normalize Windows style paths - clean_name = posixpath.normpath(name).replace("\\", "/") + cleaned = posixpath.normpath(name).replace("\\", "/") # os.path.normpath() can strip trailing slashes so we implement # a workaround here. - if name.endswith("/") and not clean_name.endswith("/"): + if name.endswith("/") and not cleaned.endswith("/"): # Add a trailing slash as it was stripped. - clean_name = clean_name + "/" + cleaned = cleaned + "/" # Given an empty string, os.path.normpath() will return ., which we don't want - if clean_name == ".": - clean_name = "" + if cleaned == ".": + cleaned = "" - return clean_name + return cleaned def safe_join(base, paths): @@ -77,12 +63,13 @@ def safe_join(base, *paths): # the base path is /. base_path_len = len(base_path) if not final_path.startswith(base_path) or final_path[base_path_len] != "/": - raise ValueError("the joined path is located outside of the base path" " component") + raise ValueError("the joined path is located outside of the base path component") return final_path.lstrip("/") def get_available_overwrite_name(name, max_length): + """Truncate a filename to obey max_length limit""" if max_length is None or len(name) <= max_length: return name @@ -94,8 +81,6 @@ def get_available_overwrite_name(name, max_length): file_root = file_root[:-truncation] if not file_root: raise SuspiciousFileOperation( - 'Storage tried to truncate away entire filename "%s". ' - "Please make sure that the corresponding file field " - 'allows sufficient "max_length".' % name + f'Storage tried to truncate away entire filename "{name}". Please make sure that the corresponding file field allows sufficient "max_length".' ) - return os.path.join(dir_name, "{}{}".format(file_root, file_ext)) + return os.path.join(dir_name, f"{file_root}{file_ext}") diff --git a/pyproject.toml b/pyproject.toml index 317cb20..f03918f 100644 --- a/pyproject.toml +++ b/pyproject.toml @@ -1,6 +1,6 @@ [tool.poetry] name = "django-gcp" -version = "0.9.3" +version = "0.10.0" description = "Utilities to run Django on Google Cloud Platform" authors = ["Tom Clark"] license = "MIT" diff --git a/tox.ini b/tox.ini index 85e4686..4842010 100644 --- a/tox.ini +++ b/tox.ini @@ -5,7 +5,7 @@ isolated_build = True [testenv] whitelist_externals = poetry setenv = - DJANGO_SETTINGS_MODULE=tests.server.settings + DJANGO_SETTINGS_MODULE=tests.server.settings_ci PYTHONPATH = {toxinidir}:{toxinidir}/django_gcp sqlite: DATABASE_ENGINE=sqlite postgres: DATABASE_ENGINE=postgres	Incorrect Default App Config ## Bug report ### What is the current behavior? The `default_app_config` value in `__init__.py` is incorrecly set to `django_gcp.apps.DjangoGcpAppConfig` when the class name is `DjangoGCPAppConfig`. This results in the following (largely confusing) error: ``` Exception in thread django-main-thread: Traceback (most recent call last): File "/usr/local/lib/python3.9/threading.py", line 980, in _bootstrap_inner self.run() File "/usr/local/lib/python3.9/threading.py", line 917, in run self._target(self._args, self._kwargs) File "/usr/local/lib/python3.9/site-packages/django/utils/autoreload.py", line 64, in wrapper fn(args, *kwargs) File "/usr/local/lib/python3.9/site-packages/django/core/management/commands/runserver.py", line 110, in inner_run autoreload.raise_last_exception() File "/usr/local/lib/python3.9/site-packages/django/utils/autoreload.py", line 87, in raise_last_exception raise _exception[1] File "/usr/local/lib/python3.9/site-packages/django/core/management/__init__.py", line 375, in execute autoreload.check_errors(django.setup)() File "/usr/local/lib/python3.9/site-packages/django/utils/autoreload.py", line 64, in wrapper fn(args, **kwargs) File "/usr/local/lib/python3.9/site-packages/django/__init__.py", line 24, in setup apps.populate(settings.INSTALLED_APPS) File "/usr/local/lib/python3.9/site-packages/django/apps/registry.py", line 91, in populate app_config = AppConfig.create(entry) File "/usr/local/lib/python3.9/site-packages/django/apps/config.py", line 228, in create if not issubclass(app_config_class, AppConfig): TypeError: issubclass() arg 1 must be a class ``` ### Workaround Set the full path in your django installed apps: ```py INSTALLED_APPS = [ # ... "django_gcp.apps.DjangoGCPAppConfig", # ... ] ``` ### Proposed Solution Fix the naming.		2023-06-12T16:25:50	[]	[]
scikit-hep/cuda-histogram	scikit-hep__cuda-histogram-13	3a747b6fb7cda2e5d1ffe41016b95fab58e78255	diff --git a/README.md b/README.md index 6aa53dc..5ff37b5 100644 --- a/README.md +++ b/README.md @@ -13,10 +13,10 @@ [![Conda latest release][conda-version]][conda-link] [![LICENSE][license-badge]][license-link] [![Scikit-HEP][sk-badge]][sk-link] -`cuda-histogram` is a histogram filling, transformation, and plotting package -for GPUs. The package follows [UHI](https://uhi.readthedocs.io) and keeps its -API similar to [boost-histogram](https://github.com/scikit-hep/boost-histogram) -and [hist](https://github.com/scikit-hep/hist). +`cuda-histogram` is a histogram filling package for GPUs. The package follows +[UHI](https://uhi.readthedocs.io) and keeps its API similar to +[boost-histogram](https://github.com/scikit-hep/boost-histogram) and +[hist](https://github.com/scikit-hep/hist). The package is under active development at the moment. diff --git a/docs/api/cuda_histogram.plot.rst b/docs/api/cuda_histogram.plot.rst deleted file mode 100644 index bc175e5..0000000 --- a/docs/api/cuda_histogram.plot.rst +++ /dev/null @@ -1,8 +0,0 @@ -cuda_histogram.plot module -========================== - -.. automodule:: cuda_histogram.plot - :members: - :undoc-members: - :show-inheritance: - :private-members: diff --git a/docs/api/cuda_histogram.rst b/docs/api/cuda_histogram.rst index 65f032a..dd7d1ad 100644 --- a/docs/api/cuda_histogram.rst +++ b/docs/api/cuda_histogram.rst @@ -8,4 +8,3 @@ Submodules :maxdepth: 3 cuda_histogram.hist_tools - cuda_histogram.plot diff --git a/noxfile.py b/noxfile.py index a3108fe..8936386 100644 --- a/noxfile.py +++ b/noxfile.py @@ -30,7 +30,7 @@ def tests(session: nox.Session) -> None: """ Run the unit and regular tests. """ - session.install(".[test,dev,plot,gpu]") + session.install(".[test,dev]") session.run("pytest", session.posargs) diff --git a/pyproject.toml b/pyproject.toml index 5b91e5b..35cb87e 100644 --- a/pyproject.toml +++ b/pyproject.toml @@ -41,15 +41,6 @@ dependencies = [ "boost-histogram", ] [project.optional-dependencies] -plot = [ - "matplotlib>=3", - "mplhep>=0.1.18", - "bokeh!=3.0.,>=2.4.2", -] -test = [ - "uproot>=5.3.0,!=5.3.3,!=5.3.4,!=5.3.5", - "coffea>=2024.6.0" -] dev = [ "pre-commit", "pytest>=6", diff --git a/src/cuda_histogram/__init__.py b/src/cuda_histogram/__init__.py index e0e1068..768e57f 100644 --- a/src/cuda_histogram/__init__.py +++ b/src/cuda_histogram/__init__.py @@ -1,5 +1,5 @@ """ -`cuda-histogram` is a histogram filling, transformation, and plotting package for GPUs. +`cuda-histogram` is a histogram filling package for GPUs. The package follows `UHI <https://uhi.readthedocs.io>`__ and keeps its API similar to `boost-histogram <https://github.com/scikit-hep/boost-histogram>`__ @@ -9,30 +9,14 @@ from __future__ import annotations from cuda_histogram.hist_tools import Bin, Cat, Hist, Interval, StringBin -from cuda_histogram.plot import ( - clopper_pearson_interval, - normal_interval, - plot1d, - plot2d, - plotgrid, - plotratio, - poisson_interval, -) from ._version import version as __version__ -__all__ = [ +__all__: list[str] = [ "Hist", "Bin", "Interval", "Cat", "StringBin", - "poisson_interval", - "clopper_pearson_interval", - "normal_interval", - "plot1d", - "plotratio", - "plot2d", - "plotgrid", "__version__", ] diff --git a/src/cuda_histogram/plot.py b/src/cuda_histogram/plot.py deleted file mode 100644 index 6184817..0000000 --- a/src/cuda_histogram/plot.py +++ /dev/null @@ -1,1065 +0,0 @@ -from __future__ import annotations - -import numbers -import warnings - -import numpy as np -import scipy.stats - -from .hist_tools import DenseAxis, Interval, SparseAxis, overflow_behavior - -# Plotting is always terrible -# Let's try our best to follow matplotlib idioms -# https://matplotlib.org/tutorials/introductory/usage.html#coding-styles - -_coverage1sd = scipy.stats.norm.cdf(1) - scipy.stats.norm.cdf(-1) - - -def poisson_interval(sumw, sumw2, coverage=_coverage1sd): - """Frequentist coverage interval for Poisson-distributed observations - - Parameters - ---------- - sumw : np.ndarray - Sum of weights vector - sumw2 : np.ndarray - Sum weights squared vector - coverage : float, optional - Central coverage interval, defaults to 68% - - Calculates the so-called 'Garwood' interval, - c.f. https://www.ine.pt/revstat/pdf/rs120203.pdf or - http://ms.mcmaster.ca/peter/s743/poissonalpha.html - For weighted data, this approximates the observed count by ``sumw2/sumw2``, which - effectively scales the unweighted poisson interval by the average weight. - This may not be the optimal solution: see https://arxiv.org/pdf/1309.1287.pdf for a proper treatment. - When a bin is zero, the scale of the nearest nonzero bin is substituted to scale the nominal upper bound. - If all bins zero, a warning is generated and interval is set to ``sumw``. - """ - scale = np.empty_like(sumw) - scale[sumw != 0] = sumw2[sumw != 0] / sumw[sumw != 0] - if np.sum(sumw == 0) > 0: - missing = np.where(sumw == 0) - available = np.nonzero(sumw) - if len(available[0]) == 0: - warnings.warn( - "All sumw are zero! Cannot compute meaningful error bars", - RuntimeWarning, - ) - return np.vstack([sumw, sumw]) - nearest = sum( - [np.subtract.outer(d, d0) 2 for d, d0 in zip(available, missing)] - ).argmin(axis=0) - argnearest = tuple(dim[nearest] for dim in available) - scale[missing] = scale[argnearest] - counts = sumw / scale - lo = scale * scipy.stats.chi2.ppf((1 - coverage) / 2, 2 * counts) / 2.0 - hi = scale * scipy.stats.chi2.ppf((1 + coverage) / 2, 2 * (counts + 1)) / 2.0 - interval = np.array([lo, hi]) - interval[interval == np.nan] = 0.0 # chi2.ppf produces nan for counts=0 - return interval - - -def clopper_pearson_interval(num, denom, coverage=_coverage1sd): - """Compute Clopper-Pearson coverage interval for a binomial distribution - - Parameters - ---------- - num : np.ndarray - Numerator, or number of successes, vectorized - denom : np.ndarray - Denominator or number of trials, vectorized - coverage : float, optional - Central coverage interval, defaults to 68% - - c.f. http://en.wikipedia.org/wiki/Binomial_proportion_confidence_interval - """ - if np.any(num > denom): - raise ValueError( - "Found numerator larger than denominator while calculating binomial uncertainty" - ) - lo = scipy.stats.beta.ppf((1 - coverage) / 2, num, denom - num + 1) - hi = scipy.stats.beta.ppf((1 + coverage) / 2, num + 1, denom - num) - interval = np.array([lo, hi]) - interval[:, num == 0.0] = 0.0 - interval[1, num == denom] = 1.0 - return interval - - -def normal_interval(pw, tw, pw2, tw2, coverage=_coverage1sd): - """Compute errors based on the expansion of pass/(pass + fail), possibly weighted - - Parameters - ---------- - pw : np.ndarray - Numerator, or number of (weighted) successes, vectorized - tw : np.ndarray - Denominator or number of (weighted) trials, vectorized - pw2 : np.ndarray - Numerator sum of weights squared, vectorized - tw2 : np.ndarray - Denominator sum of weights squared, vectorized - coverage : float, optional - Central coverage interval, defaults to 68% - - c.f. https://root.cern.ch/doc/master/TEfficiency_8cxx_source.html#l02515 - """ - - eff = pw / tw - - variance = (pw2 * (1 - 2 * eff) + tw2 * eff2) / (tw2) - sigma = np.sqrt(variance) - - prob = 0.5 * (1 - coverage) - delta = np.zeros_like(sigma) - delta[sigma != 0] = scipy.stats.norm.ppf(prob, scale=sigma[sigma != 0]) - - lo = eff - np.minimum(eff + delta, np.ones_like(eff)) - hi = np.maximum(eff - delta, np.zeros_like(eff)) - eff - - return np.array([lo, hi]) - - -def plot1d( - hist, - ax=None, - clear=True, - overlay=None, - stack=False, - overflow="none", - line_opts=None, - fill_opts=None, - error_opts=None, - legend_opts=None, - overlay_overflow="none", - density=False, - binwnorm=None, - order=None, -): - """Create a 1D plot from a 1D or 2D `Hist` object - - Parameters - ---------- - hist : Hist - Histogram with maximum of two dimensions - ax : matplotlib.axes.Axes, optional - Axes object (if None, one is created) - clear : bool, optional - Whether to clear Axes before drawing (if passed); if False, this function will skip drawing the legend - overlay : str, optional - In the case that ``hist`` is 2D, specify the axis of hist to overlay (remaining axis will be x axis) - stack : bool, optional - Whether to stack or overlay non-axis dimension (if it exists) - order : list, optional - How to order when stacking. Take a list of identifiers. - overflow : str, optional - If overflow behavior is not 'none', extra bins will be drawn on either end of the nominal - axis range, to represent the contents of the overflow bins. See `Hist.sum` documentation - for a description of the options. - line_opts : dict, optional - A dictionary of options to pass to the matplotlib - `ax.step <https://matplotlib.org/3.1.1/api/_as_gen/matplotlib.pyplot.step.html>`_ call - internal to this function. Leave blank for defaults. - fill_opts : dict, optional - A dictionary of options to pass to the matplotlib - `ax.fill_between <https://matplotlib.org/3.1.1/api/_as_gen/matplotlib.axes.Axes.fill_between.html>`_ call - internal to this function. Leave blank for defaults. - error_opts : dict, optional - A dictionary of options to pass to the matplotlib - `ax.errorbar <https://matplotlib.org/3.1.1/api/_as_gen/matplotlib.pyplot.errorbar.html>`_ call - internal to this function. Leave blank for defaults. Some special options are interpreted by - this function and not passed to matplotlib: 'emarker' (default: '') specifies the marker type - to place at cap of the errorbar. - legend_opts : dict, optional - A dictionary of options to pass to the matplotlib - `ax.legend <https://matplotlib.org/3.1.1/api/_as_gen/matplotlib.pyplot.legend.html>`_ call - internal to this function. Leave blank for defaults. - overlay_overflow : str, optional - If overflow behavior is not 'none', extra bins in the overlay axis will be overlaid or stacked, - to represent the contents of the overflow bins. See `Hist.sum` documentation for a description of the options. - density : bool, optional - If true, convert sum weights to probability density (i.e. integrates to 1 over domain of axis) - (Note: this option conflicts with ``binwnorm``) - binwnorm : float, optional - If true, convert sum weights to bin-width-normalized, with unit equal to supplied value (usually you want to specify 1.) - - - Returns - ------- - ax : matplotlib.axes.Axes - A matplotlib `Axes <https://matplotlib.org/3.1.1/api/axes_api.html>`_ object - """ - import matplotlib.pyplot as plt - import mplhep as hep - - if legend_opts is None: - legend_opts = {} - - if ax is None: - ax = plt.gca() - else: - if not isinstance(ax, plt.Axes): - raise ValueError("ax must be a matplotlib Axes object") - if clear: - ax.clear() - if hist.dim() > 2: - raise ValueError( - "plot1d() can only support up to two dimensions (one for axis, one to stack or overlay)" - ) - if overlay is None and hist.sparse_dim() == 1 and hist.dense_dim() == 1: - overlay = hist.sparse_axes()[0].name - elif overlay is None and hist.dim() > 1: - raise ValueError( - "plot1d() can only support one dimension without an overlay axis chosen" - ) - if density and binwnorm is not None: - raise ValueError("Cannot use density and binwnorm at the same time!") - if binwnorm is not None and not isinstance(binwnorm, numbers.Number): - raise ValueError( - f"Bin width normalization not a number, but a {binwnorm.__class__!r}" - ) - if line_opts is None and fill_opts is None and error_opts is None: - if stack: - fill_opts = {} - else: - line_opts = {} - error_opts = {} - - axis = hist.axes()[0] - if overlay is not None: - overlay = hist.axis(overlay) - if axis == overlay: - axis = hist.axes()[1] - if isinstance(axis, SparseAxis): - raise NotImplementedError("Plot a sparse axis (e.g. bar chart)") - elif isinstance(axis, DenseAxis): - ax.set_xlabel(axis.label) - ax.set_ylabel(hist.label) - edges = axis.edges(overflow=overflow).get() - if order is None: - identifiers = ( - hist.identifiers(overlay, overflow=overlay_overflow) - if overlay is not None - else [None] - ) - else: - identifiers = order - plot_info = { - "identifier": identifiers, - "label": list(map(str, identifiers)), - "sumw": [], - "sumw2": [], - } - for i, identifier in enumerate(identifiers): - if identifier is None: - sumw, sumw2 = hist.values(sumw2=True, overflow=overflow)[()] - elif isinstance(overlay, SparseAxis): - sumw, sumw2 = hist.integrate(overlay, identifier).values( - sumw2=True, overflow=overflow - )[()] - else: - sumw, sumw2 = hist.values(sumw2=True, overflow="allnan")[()] - the_slice = ( - i if overflow_behavior(overlay_overflow).start is None else i + 1, - overflow_behavior(overflow), - ) - if hist._idense(overlay) == 1: - the_slice = (the_slice[1], the_slice[0]) - sumw = sumw[the_slice] - sumw2 = sumw2[the_slice] - plot_info["sumw"].append(sumw.get()) - plot_info["sumw2"].append(sumw2.get()) - - def w2err(sumw, sumw2): - err = [] - for a, b in zip(sumw, sumw2): - err.append(np.abs(poisson_interval(a, b) - a)) - return err - - kwargs = None - if line_opts is not None and error_opts is None: - _error = None - else: - _error = w2err(plot_info["sumw"], plot_info["sumw2"]) - if fill_opts is not None: - histtype = "fill" - kwargs = fill_opts - elif error_opts is not None and line_opts is None: - histtype = "errorbar" - kwargs = error_opts - else: - histtype = "step" - kwargs = line_opts - if kwargs is None: - kwargs = {} - - hep.histplot( - plot_info["sumw"], - edges, - label=plot_info["label"], - yerr=_error, - histtype=histtype, - ax=ax, - density=density, - binwnorm=binwnorm, - stack=stack, - *kwargs, - ) - - if stack and error_opts is not None: - stack_sumw = np.sum(plot_info["sumw"], axis=0) - stack_sumw2 = np.sum(plot_info["sumw2"], axis=0) - err = poisson_interval(stack_sumw, stack_sumw2) - if binwnorm is not None: - err = binwnorm / np.diff(edges)[None, :] - opts = { - "step": "post", - "label": "Sum unc.", - "hatch": "///", - "facecolor": "none", - "edgecolor": (0, 0, 0, 0.5), - "linewidth": 0, - } - opts.update(error_opts) - ax.fill_between( - x=edges, - y1=np.r_[err[0, :], err[0, -1]], - y2=np.r_[err[1, :], err[1, -1]], - opts, - ) - - if legend_opts is not None: - _label = overlay.label if overlay is not None else "" - ax.legend(title=_label, legend_opts) - else: - ax.legend(title=_label) - ax.autoscale(axis="x", tight=True) - ax.set_ylim(0, None) - - return ax - - -def plotratio( - num, - denom, - ax=None, - clear=True, - overflow="none", - xerr=False, - error_opts=None, - denom_fill_opts=None, - guide_opts=None, - unc="clopper-pearson", - label=None, -): - """Create a ratio plot, dividing two compatible histograms - - Parameters - ---------- - num : Hist - Numerator, a single-axis histogram - denom : Hist - Denominator, a single-axis histogram - ax : matplotlib.axes.Axes, optional - Axes object (if None, one is created) - clear : bool, optional - Whether to clear Axes before drawing (if passed); if False, this function will skip drawing the legend - overflow : str, optional - If overflow behavior is not 'none', extra bins will be drawn on either end of the nominal - axis range, to represent the contents of the overflow bins. See `Hist.sum` documentation - for a description of the options. - xerr: bool, optional - If true, then error bars are drawn for x-axis to indicate the size of the bin. - error_opts : dict, optional - A dictionary of options to pass to the matplotlib - `ax.errorbar <https://matplotlib.org/3.1.1/api/_as_gen/matplotlib.pyplot.errorbar.html>`_ call - internal to this function. Leave blank for defaults. Some special options are interpreted by - this function and not passed to matplotlib: 'emarker' (default: '') specifies the marker type - to place at cap of the errorbar. - denom_fill_opts : dict, optional - A dictionary of options to pass to the matplotlib - `ax.fill_between <https://matplotlib.org/3.1.1/api/_as_gen/matplotlib.axes.Axes.fill_between.html>`_ call - internal to this function, filling the denominator uncertainty band. Leave blank for defaults. - guide_opts : dict, optional - A dictionary of options to pass to the matplotlib - `ax.axhline <https://matplotlib.org/3.1.1/api/_as_gen/matplotlib.axes.Axes.axhline.html>`_ call - internal to this function, to plot a horizontal guide line at ratio of 1. Leave blank for defaults. - unc : str, optional - Uncertainty calculation option: 'clopper-pearson' interval for efficiencies; 'poisson-ratio' interval - for ratio of poisson distributions; 'num' poisson interval of numerator scaled by denominator value - (common for data/mc, for better or worse). - label : str, optional - Associate a label to this entry (note: y axis label set by ``num.label``) - - Returns - ------- - ax : matplotlib.axes.Axes - A matplotlib `Axes <https://matplotlib.org/3.1.1/api/axes_api.html>`_ object - """ - import matplotlib.pyplot as plt - - if ax is None: - fig, ax = plt.subplots(1, 1) - else: - if not isinstance(ax, plt.Axes): - raise ValueError("ax must be a matplotlib Axes object") - if clear: - ax.clear() - if not num.compatible(denom): - raise ValueError( - "numerator and denominator histograms have incompatible axis definitions" - ) - if num.dim() > 1: - raise ValueError("plotratio() can only support one-dimensional histograms") - if error_opts is None and denom_fill_opts is None and guide_opts is None: - error_opts = {} - denom_fill_opts = {} - - axis = num.axes()[0] - if isinstance(axis, SparseAxis): - raise NotImplementedError("Ratio for sparse axes (labeled axis with errorbars)") - elif isinstance(axis, DenseAxis): - ax.set_xlabel(axis.label) - ax.set_ylabel(num.label) - edges = axis.edges(overflow=overflow).get() - centers = axis.centers(overflow=overflow).get() - ranges = (edges[1:] - edges[:-1]) / 2 if xerr else None - - sumw_num, sumw2_num = num.values(sumw2=True, overflow=overflow)[()] - sumw_denom, sumw2_denom = denom.values(sumw2=True, overflow=overflow)[()] - sumw_num, sumw2_num = sumw_num.get(), sumw2_num.get() - sumw_denom, sumw2_denom = sumw_denom.get(), sumw2_denom.get() - - rsumw = sumw_num / sumw_denom - if unc == "clopper-pearson": - rsumw_err = np.abs(clopper_pearson_interval(sumw_num, sumw_denom) - rsumw) - elif unc == "poisson-ratio": - # poisson ratio n/m is equivalent to binomial n/(n+m) - rsumw_err = np.abs( - clopper_pearson_interval(sumw_num, sumw_num + sumw_denom) - rsumw - ) - elif unc == "num": - rsumw_err = np.abs( - poisson_interval(rsumw, sumw2_num / sumw_denom2) - rsumw - ) - elif unc == "normal": - rsumw_err = np.abs( - normal_interval(sumw_num, sumw_denom, sumw2_num, sumw2_denom) - ) - else: - raise ValueError(f"Unrecognized uncertainty option: {unc!r}") - - if error_opts is not None: - opts = {"label": label, "linestyle": "none"} - opts.update(error_opts) - emarker = opts.pop("emarker", "") - errbar = ax.errorbar( - x=centers, y=rsumw, xerr=ranges, yerr=rsumw_err, opts - ) - plt.setp(errbar[1], "marker", emarker) - if denom_fill_opts is not None: - unity = np.ones_like(sumw_denom) - denom_unc = poisson_interval(unity, sumw2_denom / sumw_denom2) - opts = {"step": "post", "facecolor": (0, 0, 0, 0.3), "linewidth": 0} - opts.update(denom_fill_opts) - ax.fill_between( - edges, - np.r_[denom_unc[0], denom_unc[0, -1]], - np.r_[denom_unc[1], denom_unc[1, -1]], - opts, - ) - if guide_opts is not None: - opts = {"linestyle": "--", "color": (0, 0, 0, 0.5), "linewidth": 1} - opts.update(guide_opts) - ax.axhline(1.0, *opts) - - if clear: - ax.autoscale(axis="x", tight=True) - ax.set_ylim(0, None) - - return ax - - -def plot2d( - hist, - xaxis, - ax=None, - clear=True, - xoverflow="none", - yoverflow="none", - patch_opts=None, - text_opts=None, - density=False, - binwnorm=None, -): - """Create a 2D plot from a 2D `Hist` object - - Parameters - ---------- - hist : Hist - Histogram with two dimensions - xaxis : str or Axis - Which of the two dimensions to use as an x axis - ax : matplotlib.axes.Axes, optional - Axes object (if None, one is created) - clear : bool, optional - Whether to clear Axes before drawing (if passed); if False, this function will skip drawing the legend - xoverflow : str, optional - If overflow behavior is not 'none', extra bins will be drawn on either end of the nominal x - axis range, to represent the contents of the overflow bins. See `Hist.sum` documentation - for a description of the options. - yoverflow : str, optional - Similar to ``xoverflow`` - patch_opts : dict, optional - Options passed to the matplotlib `pcolormesh <https://matplotlib.org/api/_as_gen/matplotlib.pyplot.pcolormesh.html>`_ - call internal to this function, to plot a rectangular grid of patches colored according to the bin values. - Leave empty for defaults. - text_opts : dict, optional - Options passed to the matplotlib `text <https://matplotlib.org/api/text_api.html#matplotlib.text.Text>`_ - call internal to this function, to place a text label at each bin center with the bin value. Special - options interpreted by this function and not passed to matplotlib: 'format': printf-style float format - , default '%.2g'. - density : bool, optional - If true, convert sum weights to probability density (i.e. integrates to 1 over domain of axis) - (Note: this option conflicts with ``binwnorm``) - binwnorm : float, optional - If true, convert sum weights to bin-width-normalized, with unit equal to supplied value (usually you want to specify 1.) - - Returns - ------- - ax : matplotlib.axes.Axes - A matplotlib `Axes <https://matplotlib.org/3.1.1/api/axes_api.html>`_ object - """ - import matplotlib.pyplot as plt - - if ax is None: - fig, ax = plt.subplots(1, 1) - else: - if not isinstance(ax, plt.Axes): - raise ValueError("ax must be a matplotlib Axes object") - if clear: - ax.clear() - fig = ax.figure - if hist.dim() != 2: - raise ValueError("plot2d() can only support exactly two dimensions") - if density and binwnorm is not None: - raise ValueError("Cannot use density and binwnorm at the same time!") - if binwnorm is not None and not isinstance(binwnorm, numbers.Number): - raise ValueError( - f"Bin width normalization not a number, but a {binwnorm.__class__!r}" - ) - if patch_opts is None and text_opts is None: - patch_opts = {} - - xaxis = hist.axis(xaxis) - yaxis = hist.axes()[1] - transpose = False - if yaxis == xaxis: - yaxis = hist.axes()[0] - transpose = True - if isinstance(xaxis, SparseAxis) or isinstance(yaxis, SparseAxis): - raise NotImplementedError("Plot a sparse axis (e.g. bar chart or labeled bins)") - else: - xedges = xaxis.edges(overflow=xoverflow).get() - yedges = yaxis.edges(overflow=yoverflow).get() - sumw, sumw2 = hist.values(sumw2=True, overflow="allnan")[()] - sumw, sumw2 = sumw.get(), sumw2.get() - if transpose: - sumw = sumw.T - sumw2 = sumw2.T - # no support for different overflow behavior per axis, do it ourselves - sumw = sumw[overflow_behavior(xoverflow), overflow_behavior(yoverflow)] - sumw2 = sumw2[overflow_behavior(xoverflow), overflow_behavior(yoverflow)] - if (density or binwnorm is not None) and np.sum(sumw) > 0: - overallnorm = np.sum(sumw) binwnorm if binwnorm is not None else 1.0 - areas = np.multiply.outer(np.diff(xedges), np.diff(yedges)) - binnorms = overallnorm / (areas * np.sum(sumw)) - sumw = sumw * binnorms - sumw2 = sumw2 * binnorms2 - - if patch_opts is not None: - opts = {"cmap": "viridis"} - opts.update(patch_opts) - pc = ax.pcolormesh(xedges, yedges, sumw.T, opts) - ax.add_collection(pc) - if clear: - fig.colorbar(pc, ax=ax, label=hist.label) - if text_opts is not None: - for ix, xcenter in enumerate(xaxis.centers()): - for iy, ycenter in enumerate(yaxis.centers()): - opts = { - "horizontalalignment": "center", - "verticalalignment": "center", - } - if patch_opts is not None: - opts["color"] = ( - "black" if pc.norm(sumw[ix, iy]) > 0.5 else "lightgrey" - ) - opts.update(text_opts) - txtformat = opts.pop("format", r"%.2g") - ax.text(xcenter, ycenter, txtformat % sumw[ix, iy], opts) - - if clear: - ax.set_xlabel(xaxis.label) - ax.set_ylabel(yaxis.label) - ax.set_xlim(xedges[0], xedges[-1]) - ax.set_ylim(yedges[0], yedges[-1]) - - return ax - - -def plotgrid( - h, - figure=None, - row=None, - col=None, - overlay=None, - row_overflow="none", - col_overflow="none", - plot_opts, -): - """Create a grid of plots, enumerating identifiers on up to 3 axes - - Parameters - ---------- - h : Hist - A histogram with up to 3 axes - figure : matplotlib.figure.Figure, optional - If omitted, a new figure is created. Otherwise, the axes will be redrawn on this existing figure. - row : str - Name of row axis - col : str - Name of column axis - overlay : str - name of overlay axis - row_overflow : str, optional - If overflow behavior is not 'none', extra bins will be drawn on either end of the nominal x - axis range, to represent the contents of the overflow bins. See `Hist.sum` documentation - for a description of the options. - col_overflow : str, optional - Similar to ``row_overflow`` - ``*plot_opts`` : kwargs - The remaining axis of the histogram, after removing any of ``row,col,overlay`` specified, - will be the plot axis, with ``plot_opts`` passed to the `plot1d` call. - - Returns - ------- - axes : np.ndarray - An array of matplotlib `Axes <https://matplotlib.org/3.1.1/api/axes_api.html>`_ objects - """ - import matplotlib.pyplot as plt - - haxes = {ax.name for ax in h.axes()} - nrow, ncol = 1, 1 - if row: - row_identifiers = h.identifiers(row, overflow=row_overflow) - nrow = len(row_identifiers) - haxes.remove(row) - if col: - col_identifiers = h.identifiers(col, overflow=col_overflow) - ncol = len(col_identifiers) - haxes.remove(col) - if overlay: - haxes.remove(overlay) - if len(haxes) > 1: - raise ValueError( - "More than one dimension left: {}".format(",".join(ax for ax in haxes)) - ) - elif len(haxes) == 0: - raise ValueError(f"Not enough dimensions available in {h!r}") - - figsize = plt.rcParams["figure.figsize"] - figsize = figsize[0] max(ncol, 1), figsize[1] * max(nrow, 1) - if figure is None: - fig, axes = plt.subplots( - nrow, ncol, figsize=figsize, squeeze=False, sharex=True, sharey=True - ) - else: - fig = figure - shape = (0, 0) - lastax = fig.get_children()[-1] - if isinstance(lastax, plt.Axes): - shape = lastax.rowNum + 1, lastax.colNum + 1 - if shape[0] == nrow and shape[1] == ncol: - axes = np.array(fig.axes).reshape(shape) - else: - fig.clear() - # fig.set_size_inches(figsize) - axes = fig.subplots(nrow, ncol, squeeze=False, sharex=True, sharey=True) - - for icol in range(ncol): - hcol = h - coltitle = None - if col: - vcol = col_identifiers[icol] - hcol = h.integrate(col, vcol) - coltitle = str(vcol) - if isinstance(vcol, Interval) and vcol.label is None: - coltitle = f"{h.axis(col).label} ∈ {coltitle}" - for irow in range(nrow): - ax = axes[irow, icol] - hplot = hcol - rowtitle = None - if row: - vrow = row_identifiers[irow] - hplot = hcol.integrate(row, vrow) - rowtitle = str(vrow) - if isinstance(vrow, Interval) and vrow.label is None: - rowtitle = f"{h.axis(row).label} ∈ {rowtitle}" - - plot1d(hplot, ax=ax, overlay=overlay, *plot_opts) - if row is not None and col is not None: - ax.set_title(f"{rowtitle}, {coltitle}") - elif row is not None: - ax.set_title(rowtitle) - elif col is not None: - ax.set_title(coltitle) - - for ax in axes.flatten(): - ax.autoscale(axis="y") - ax.set_ylim(0, None) - - return axes - - -def isnotebook(): - try: - shell = get_ipython().__class__.__name__ - if shell == "ZMQInteractiveShell": - return True # Jupyter notebook or qtconsole - elif shell == "TerminalInteractiveShell": - return False # Terminal running IPython - else: - return False # Other type (?) - except NameError: - return False - - -def bokeh_plot(histo, jup_url="http://127.0.0.1:8889"): - if not isnotebook(): - raise NotImplementedError("Only usable in jupyter notebook") - import bokeh.plotting.figure as bk_figure - from bokeh import palettes - - # init bokeh - from bokeh.application import Application - from bokeh.application.handlers import FunctionHandler - from bokeh.core.validation import silence - from bokeh.core.validation.warnings import EMPTY_LAYOUT - from bokeh.io import ( - output_notebook, # enables plot interface in J notebook - show, - ) - from bokeh.layouts import column, row - from bokeh.models import ColumnDataSource - from bokeh.models.widgets import ( - CheckboxButtonGroup, - Div, - RadioButtonGroup, - RangeSlider, - ) - - silence(EMPTY_LAYOUT, True) - - output_notebook() - - # Set up widgets - cfg_labels = ["Ghost"] - wi_config = CheckboxButtonGroup(labels=cfg_labels, active=[0]) - wi_dense_select = RadioButtonGroup( - labels=[ax.name for ax in histo.dense_axes()], active=0 - ) - wi_sparse_select = RadioButtonGroup( - labels=[ax.name for ax in histo.sparse_axes()], active=0 - ) - - # Dense widgets - sliders = {} - for ax in histo.dense_axes(): - edge_vals = (histo.axis(ax.name).edges()[0], histo.axis(ax.name).edges()[-1]) - _smallest_bin = np.min(np.diff(histo.axis(ax.name).edges())) - sliders[ax.name] = RangeSlider( - title=ax.name, - value=edge_vals, - start=edge_vals[0], - end=edge_vals[1], - step=_smallest_bin, - name=ax.name, - ) - - # Cat widgets - togglers = {} - for ax in histo.sparse_axes(): - togglers[ax.name] = CheckboxButtonGroup( - labels=[i.name for i in ax.identifiers()], active=[0], name=ax.name - ) - - # Toggles for all widgets - configers = {} - for ax in histo.sparse_axes(): - configers[ax.name] = CheckboxButtonGroup( - labels=["Display", "Ghost"], active=[0, 1], name=ax.name - ) - for ax in histo.dense_axes(): - configers[ax.name] = CheckboxButtonGroup( - labels=["Display"], active=[0], name=ax.name - ) - - # Figure - fig = bk_figure( - title="1D Projection", - plot_width=500, - plot_height=500, - min_border=20, - toolbar_location=None, - ) - fig.yaxis.axis_label = "N" - fig.xaxis.axis_label = "Quantity" - - # Iterate over possible overlays - _max_idents = 0 # Max number of simultaneou histograms - for ax in histo.sparse_axes(): - _max_idents = max(_max_idents, len([i.name for i in ax.identifiers()])) - - # Data source list - sources = [] - sources_ghost = [] - for _ in range(_max_idents): - sources.append( - ColumnDataSource({"left": [], "right": [], "top": [], "bottom": []}) - ) - sources_ghost.append( - ColumnDataSource({"left": [], "right": [], "top": [], "bottom": []}) - ) - - # Hist list - hists = [] - hists_ghost = [] - for i in range(_max_idents): - if _max_idents < 10: - _color = palettes.Category10[min(max(3, _max_idents), 10)][i] - else: - _color = palettes.magma(_max_idents)[i] - hists.append( - fig.quad( - left="left", - right="right", - top="top", - bottom="bottom", - source=sources[i], - alpha=0.9, - color=_color, - ) - ) - hists_ghost.append( - fig.quad( - left="left", - right="right", - top="top", - bottom="bottom", - source=sources_ghost[i], - alpha=0.05, - color=_color, - ) - ) - - def update_data(attrname, old, new): # noqa: ARG001 - sparse_active = wi_sparse_select.active - sparse_name = [ax.name for ax in histo.sparse_axes()][sparse_active] - sparse_other = [ax.name for ax in histo.sparse_axes() if ax.name != sparse_name] - - dense_active = wi_dense_select.active - dense_name = [ax.name for ax in histo.dense_axes()][dense_active] - dense_other = [ax.name for ax in histo.dense_axes() if ax.name != dense_name] - - # Apply cuts in projections - _h = histo.copy() - for proj_ax in sparse_other: - _idents = histo.axis(proj_ax).identifiers() - _labels = [ident.name for ident in _idents] - if 0 in configers[proj_ax].active: - _h = _h.integrate( - proj_ax, [_labels[i] for i in togglers[proj_ax].active] - ) - else: - _h = _h.integrate(proj_ax) - - for proj_ax in dense_other: - _h = _h.integrate( - proj_ax, slice(sliders[proj_ax].value[0], sliders[proj_ax].value[1]) - ) - - for cat_ix in range(_max_idents): - # Update histo for each toggled overlay - if cat_ix in togglers[sparse_name].active: - cat_value = histo.axis(sparse_name).identifiers()[cat_ix] - h1d = _h.integrate(sparse_name, cat_value) - - # Get shown histogram - values = h1d.project(dense_name).values() - if values != {}: - h = values[()] - bins = h1d.axis(dense_name).edges() - - # Apply cuts on shown axis - bin_los = bins[:-1][bins[:-1] > sliders[dense_name].value[0]] - bin_his = bins[1:][bins[1:] < sliders[dense_name].value[1]] - new_bins = np.intersect1d(bin_los, bin_his) - bin_ixs = np.searchsorted(bins, new_bins)[:-1] - h = h[bin_ixs] - - sources[cat_ix].data = { - "left": new_bins[:-1], - "right": new_bins[1:], - "top": h, - "bottom": np.zeros_like(h), - } - else: - sources[cat_ix].data = { - "left": [], - "right": [], - "top": [], - "bottom": [], - } - - # Add ghosts - if 0 in wi_config.active: - h1d = histo.integrate(sparse_name, cat_value) - for proj_ax in sparse_other: - _idents = histo.axis(proj_ax).identifiers() - _labels = [ident.name for ident in _idents] - if 1 not in configers[proj_ax].active: - h1d = h1d.integrate( - proj_ax, [_labels[i] for i in togglers[proj_ax].active] - ) - else: - h1d = h1d.integrate(proj_ax) - values = h1d.project(dense_name).values() - if values != {}: - h = h1d.project(dense_name).values()[()] - bins = h1d.axis(dense_name).edges() - sources_ghost[cat_ix].data = { - "left": bins[:-1], - "right": bins[1:], - "top": h, - "bottom": np.zeros_like(h), - } - else: - sources_ghost[cat_ix].data = { - "left": [], - "right": [], - "top": [], - "bottom": [], - } - else: - sources[cat_ix].data = { - "left": [], - "right": [], - "top": [], - "bottom": [], - } - sources_ghost[cat_ix].data = { - "left": [], - "right": [], - "top": [], - "bottom": [], - } - - # Cosmetics - fig.xaxis.axis_label = dense_name - - for _, slider in sliders.items(): - slider.on_change("value", update_data) - for _, toggler in togglers.items(): - toggler.on_change("active", update_data) - for _, configurer in configers.items(): - configurer.on_change("active", update_data) - # Button - for w in [wi_dense_select, wi_sparse_select, wi_config]: - w.on_change("active", update_data) - - from bokeh.models.widgets import Panel, Tabs - - layout = row( - fig, - column( - Div( - text="<b>Overlay Axis:</b>", - style={"font-size": "100%", "color": "black"}, - ), - wi_sparse_select, - Div( - text="<b>Plot Axis:</b>", style={"font-size": "100%", "color": "black"} - ), - wi_dense_select, - Div( - text="<b>Categorical Cuts:</b>", - style={"font-size": "100%", "color": "black"}, - ), - [toggler for name, toggler in togglers.items()], - Div( - text="<b>Dense Cuts:</b>", style={"font-size": "100%", "color": "black"} - ), - *[slider for name, slider in sliders.items()], - ), - ) - - # Config prep - incl_lists = [[], [], []] - for i, key in enumerate(list(configers.keys())): - incl_lists[i // max(5, len(list(configers.keys())) / 3)].append( - Div( - text=f"<b>{key}:</b>", - style={"font-size": "70%", "color": "black"}, - ) - ) - incl_lists[i // max(5, len(list(configers.keys())) / 3)].append(configers[key]) - - layout_cfgs = column( - row( - column( - Div( - text="<b>Configs:</b>", - style={"font-size": "100%", "color": "black"}, - ), - wi_config, - ) - ), - Div( - text="<b>Axis togglers:</b>", style={"font-size": "100%", "color": "black"} - ), - row( - column(incl_lists[0]), - column(incl_lists[1]), - column(incl_lists[2]), - ), - ) - - # Update active buttons - def update_layout(attrname, old, new): # noqa: ARG001 - active_axes = [None] - for name, wi in configers.items(): - if 0 in wi.active: - active_axes.append(name) - for child in layout.children[1].children: - if child.name not in active_axes: - child.visible = False - else: - child.visible = True - - for _, configurer in configers.items(): - configurer.on_change("active", update_layout) - - tab1 = Panel(child=layout, title="Projection") - tab2 = Panel(child=layout_cfgs, title="Configs") - tabs = Tabs(tabs=[tab1, tab2]) - - def modify_doc(doc): - doc.add_root(row(tabs, width=800)) - doc.title = "Sliders" - - handler = FunctionHandler(modify_doc) - app = Application(handler) - - show(app, notebook_url=jup_url) - update_data("", "", "")	Figure out testing on GPUs (failing tests in GH Actions)		2024-07-22T17:07:59	[]	[]
RudolfCardinal/pythonlib	RudolfCardinal__pythonlib-16	15f5c245129befef8537506b742273a017f66a75	diff --git a/.gitignore b/.gitignore index fab9b37..c02d158 100644 --- a/.gitignore +++ b/.gitignore @@ -9,3 +9,6 @@ private_testing/* # Silver searcher .ignore + +# pytest coverage report +.coverage diff --git a/cardinal_pythonlib/pdf.py b/cardinal_pythonlib/pdf.py index 7ae502f..40aefc1 100644 --- a/cardinal_pythonlib/pdf.py +++ b/cardinal_pythonlib/pdf.py @@ -36,12 +36,13 @@ import tempfile from typing import Any, Dict, Iterable, Union -from cardinal_pythonlib.logs import get_brace_style_log_with_null_handler - # noinspection PyProtectedMember -from PyPDF2 import PdfFileMerger, PdfFileReader, PdfFileWriter +from pypdf import PdfMerger, PdfReader, PdfWriter from semantic_version import Version +from cardinal_pythonlib.logs import get_brace_style_log_with_null_handler + + # ============================================================================= # Conditional/optional imports # ============================================================================= @@ -171,13 +172,13 @@ def __init__( self.filename = filename def add_to_writer( - self, writer: PdfFileWriter, start_recto: bool = True + self, writer: PdfWriter, start_recto: bool = True ) -> None: """ Add the PDF described by this class to a PDF writer. Args: - writer: a :class:`PyPDF2.PdfFileWriter` + writer: a :class:`pypdf.PdfWriter` start_recto: start a new right-hand page? """ @@ -191,10 +192,10 @@ def add_to_writer( ) append_memory_pdf_to_writer(pdf, writer, start_recto=start_recto) elif self.is_filename: - if start_recto and writer.getNumPages() % 2 != 0: - writer.addBlankPage() - writer.appendPagesFromReader( - PdfFileReader(open(self.filename, "rb")) + if start_recto and len(writer.pages) % 2 != 0: + writer.add_blank_page() + writer.append_pages_from_reader( + PdfReader(open(self.filename, "rb")) ) else: raise AssertionError("PdfPlan: shouldn't get here!") @@ -547,9 +548,9 @@ def make_pdf_on_disk_from_html( return result -def pdf_from_writer(writer: Union[PdfFileWriter, PdfFileMerger]) -> bytes: +def pdf_from_writer(writer: Union[PdfWriter, PdfMerger]) -> bytes: """ - Extracts a PDF (as binary data) from a PyPDF2 writer or merger object. + Extracts a PDF (as binary data) from a pypdf writer or merger object. """ memfile = io.BytesIO() writer.write(memfile) @@ -577,18 +578,18 @@ def serve_pdf_to_stdout(pdf: bytes) -> None: sys.stdout.buffer.write(pdf) -def make_pdf_writer() -> PdfFileWriter: +def make_pdf_writer() -> PdfWriter: """ - Creates and returns a PyPDF2 writer. + Creates and returns a pypdf writer. """ - return PdfFileWriter() + return PdfWriter() def append_memory_pdf_to_writer( - input_pdf: bytes, writer: PdfFileWriter, start_recto: bool = True + input_pdf: bytes, writer: PdfWriter, start_recto: bool = True ) -> None: """ - Appends a PDF (as bytes in memory) to a PyPDF2 writer. + Appends a PDF (as bytes in memory) to a pypdf writer. Args: input_pdf: the PDF, as ``bytes`` @@ -597,16 +598,16 @@ def append_memory_pdf_to_writer( """ if not input_pdf: return - if start_recto and writer.getNumPages() % 2 != 0: - writer.addBlankPage() + if start_recto and len(writer.pages) % 2 != 0: + writer.add_blank_page() # ... suitable for double-sided printing infile = io.BytesIO(input_pdf) - reader = PdfFileReader(infile) - for page_num in range(reader.numPages): - writer.addPage(reader.getPage(page_num)) + reader = PdfReader(infile) + for page in reader.pages: + writer.add_page(page) -def append_pdf(input_pdf: bytes, output_writer: PdfFileWriter): +def append_pdf(input_pdf: bytes, output_writer: PdfWriter): """ Appends a PDF to a pyPDF writer. Legacy interface. """ @@ -630,13 +631,13 @@ def append_pdf(input_pdf: bytes, output_writer: PdfFileWriter): # def append_disk_pdf_to_writer(filename, writer): # """Appends a PDF from disk to a pyPDF writer.""" -# if writer.getNumPages() % 2 != 0: -# writer.addBlankPage() +# if len(writer.pages) % 2 != 0: +# writer.add_blank_page() # # ... keeps final result suitable for double-sided printing # with open(filename, mode='rb') as infile: -# reader = PdfFileReader(infile) -# for page_num in range(reader.numPages): -# writer.addPage(reader.getPage(page_num)) +# reader = PdfReader(infile) +# for page in reader.pages: +# writer.add_page(page) def get_concatenated_pdf_from_disk( @@ -656,17 +657,17 @@ def get_concatenated_pdf_from_disk( # https://stackoverflow.com/questions/17104926/pypdf-merging-multiple-pdf-files-into-one-pdf # noqa # https://en.wikipedia.org/wiki/Recto_and_verso if start_recto: - writer = PdfFileWriter() + writer = PdfWriter() for filename in filenames: if filename: - if writer.getNumPages() % 2 != 0: - writer.addBlankPage() - writer.appendPagesFromReader( - PdfFileReader(open(filename, "rb")) + if len(writer.pages) % 2 != 0: + writer.add_blank_page() + writer.append_pages_from_reader( + PdfReader(open(filename, "rb")) ) return pdf_from_writer(writer) else: - merger = PdfFileMerger() + merger = PdfMerger() for filename in filenames: if filename: merger.append(open(filename, "rb")) @@ -687,7 +688,7 @@ def get_concatenated_pdf_in_memory( concatenated PDF, as ``bytes`` """ - writer = PdfFileWriter() + writer = PdfWriter() for pdfplan in pdf_plans: pdfplan.add_to_writer(writer, start_recto=start_recto) return pdf_from_writer(writer) diff --git a/docs/docs_requirements.txt b/docs/docs_requirements.txt index 2ed1df9..bcf81ad 100644 --- a/docs/docs_requirements.txt +++ b/docs/docs_requirements.txt @@ -7,6 +7,7 @@ dogpile.cache==0.9.2 # CRATE is on 3.2 Django<4.0 libChEBIpy +pdfkit pyramid==1.10.8 pytest # CamCOPS and CRATE both on this version diff --git a/docs/source/autodoc/_index.rst b/docs/source/autodoc/_index.rst index d2e2706..888322b 100644 --- a/docs/source/autodoc/_index.rst +++ b/docs/source/autodoc/_index.rst @@ -174,6 +174,7 @@ Automatic documentation of source code tests/dogpile_cache_tests.py.rst tests/interval_tests.py.rst tests/lists_tests.py.rst + tests/pdf_tests.py.rst tests/rate_limiting_tests.py.rst tests/rounding_tests.py.rst tests/rpm_tests.py.rst diff --git a/docs/source/changelog.rst b/docs/source/changelog.rst index e746c93..7b924a6 100644 --- a/docs/source/changelog.rst +++ b/docs/source/changelog.rst @@ -768,3 +768,6 @@ Quick links: ``wkhtmltopdf_options``; this prevents a bug where calls using e.g. a temporary file as footer HTML then make the next call, with no footer, fail (because the footer filename was written back to the dict). + +- ``pdf.py`` updated to use pypdf instead of PyPDF2, which is no longer + supported. diff --git a/setup.py b/setup.py index 35aebf3..d86604a 100644 --- a/setup.py +++ b/setup.py @@ -61,7 +61,7 @@ "psutil", "pygments", "pyparsing", - "PyPDF2", + "pypdf>=3.1.0", "python-dateutil", # "import dateutil" "scipy", "semantic-version",	PyPDF2 is no longer maintained According to https://pypi.org/project/PyPDF2/ : "NOTE: The PyPDF2 project is going back to its roots. PyPDF2==3.0.X will be the last version of PyPDF2. Development will continue with [pypdf==3.1.0](https://pypi.org/project/pyPdf/)." There is a migration guide at https://pypdf.readthedocs.io/en/latest/user/migration-1-to-2.html I don't know if there are changes between 2 and 3.	There is also a security advisory for 1.27.5, the version currently used by CRATE.	2023-01-27T17:05:48	[]	[]
RudolfCardinal/pythonlib	RudolfCardinal__pythonlib-14	6e0e850b2c0372dcb120ebbce7c182dba2fe94e1	diff --git a/.github/scripts/python_checks.sh b/.github/scripts/python_checks.sh new file mode 100755 index 0000000..0f13bf9 --- /dev/null +++ b/.github/scripts/python_checks.sh @@ -0,0 +1,17 @@ +#!/usr/bin/env bash + +# Run from .github/workflows/python_checks.yml + +set -eux -o pipefail + +python3 -m venv "${HOME}/venv" +source "${HOME}/venv/bin/activate" +python -m site +python -m pip install -U pip setuptools + +echo installing pip packages +python -m pip install -e . +python -m pip install pre-commit + +echo running pre-commit checks +pre-commit run --all-files diff --git a/.github/workflows/python_checks.yml b/.github/workflows/python_checks.yml new file mode 100644 index 0000000..384b7d4 --- /dev/null +++ b/.github/workflows/python_checks.yml @@ -0,0 +1,23 @@ +--- +# yamllint disable rule:line-length +name: Python checks +# yamllint disable-line rule:truthy +on: + push: + paths: + - '*.py' + - .github/workflows/python-checks.yml + - .github/scripts/python_checks.sh +jobs: + python-checks: + runs-on: ubuntu-latest + strategy: + matrix: + python-version: ["3.7", "3.8", "3.9", "3.10"] + steps: + - uses: actions/checkout@v3 + - uses: actions/setup-python@v4 + with: + python-version: ${{ matrix.python-version }} + - name: Python checks + run: ${GITHUB_WORKSPACE}/.github/scripts/python_checks.sh diff --git a/.pre-commit-config.yaml b/.pre-commit-config.yaml new file mode 100644 index 0000000..fc00319 --- /dev/null +++ b/.pre-commit-config.yaml @@ -0,0 +1,20 @@ +# See https://pre-commit.com for more information +# See https://pre-commit.com/hooks.html for more hooks +repos: +- repo: https://github.com/pre-commit/pre-commit-hooks + rev: v3.2.0 + hooks: + - id: check-added-large-files + - id: check-json + - id: check-yaml + - id: check-xml +- repo: https://github.com/psf/black + rev: 22.8.0 + hooks: + - id: black + args: + - --line-length=79 +- repo: https://github.com/PyCQA/flake8 + rev: 5.0.4 + hooks: + - id: flake8 diff --git a/cardinal_pythonlib/__init__.py b/cardinal_pythonlib/__init__.py index 1e14c7a..4730460 100644 --- a/cardinal_pythonlib/__init__.py +++ b/cardinal_pythonlib/__init__.py @@ -29,4 +29,6 @@ # Create cardinal_pythonlib.__version__: # noinspection PyPep8Naming -from cardinal_pythonlib.version_string import VERSION_STRING as __version__ +from cardinal_pythonlib.version_string import ( # noqa: F401 + VERSION_STRING as __version__, +) diff --git a/cardinal_pythonlib/athena_ohdsi.py b/cardinal_pythonlib/athena_ohdsi.py index 38843c7..4801368 100644 --- a/cardinal_pythonlib/athena_ohdsi.py +++ b/cardinal_pythonlib/athena_ohdsi.py @@ -388,42 +388,42 @@ def gen_rows() -> Generator[AthenaConceptRow, None, None]: yield concept def filter_vocab( - concepts_: Iterable[AthenaConceptRow] + concepts_: Iterable[AthenaConceptRow], ) -> Generator[AthenaConceptRow, None, None]: for concept in concepts_: if concept.vocabulary_id in vocabulary_ids: yield concept def filter_code( - concepts_: Iterable[AthenaConceptRow] + concepts_: Iterable[AthenaConceptRow], ) -> Generator[AthenaConceptRow, None, None]: for concept in concepts_: if concept.concept_code in concept_codes: yield concept def filter_id( - concepts_: Iterable[AthenaConceptRow] + concepts_: Iterable[AthenaConceptRow], ) -> Generator[AthenaConceptRow, None, None]: for concept in concepts_: if concept.concept_id in concept_ids: yield concept def filter_not_vocab( - concepts_: Iterable[AthenaConceptRow] + concepts_: Iterable[AthenaConceptRow], ) -> Generator[AthenaConceptRow, None, None]: for concept in concepts_: if concept.vocabulary_id not in not_vocabulary_ids: yield concept def filter_not_code( - concepts_: Iterable[AthenaConceptRow] + concepts_: Iterable[AthenaConceptRow], ) -> Generator[AthenaConceptRow, None, None]: for concept in concepts_: if concept.concept_code not in not_concept_codes: yield concept def filter_not_id( - concepts_: Iterable[AthenaConceptRow] + concepts_: Iterable[AthenaConceptRow], ) -> Generator[AthenaConceptRow, None, None]: for concept in concepts_: if concept.concept_id not in not_concept_ids: @@ -526,42 +526,42 @@ def gen_rows() -> Generator[AthenaConceptRelationshipRow, None, None]: yield rel def filter_rel( - rels: Iterable[AthenaConceptRelationshipRow] + rels: Iterable[AthenaConceptRelationshipRow], ) -> Generator[AthenaConceptRelationshipRow, None, None]: for rel in rels: if rel.relationship_id in relationship_id_values: yield rel def filter_c1( - rels: Iterable[AthenaConceptRelationshipRow] + rels: Iterable[AthenaConceptRelationshipRow], ) -> Generator[AthenaConceptRelationshipRow, None, None]: for rel in rels: if rel.concept_id_1 in concept_id_1_values: yield rel def filter_c2( - rels: Iterable[AthenaConceptRelationshipRow] + rels: Iterable[AthenaConceptRelationshipRow], ) -> Generator[AthenaConceptRelationshipRow, None, None]: for rel in rels: if rel.concept_id_2 in concept_id_2_values: yield rel def filter_not_rel( - rels: Iterable[AthenaConceptRelationshipRow] + rels: Iterable[AthenaConceptRelationshipRow], ) -> Generator[AthenaConceptRelationshipRow, None, None]: for rel in rels: if rel.relationship_id not in not_relationship_id_values: yield rel def filter_not_c1( - rels: Iterable[AthenaConceptRelationshipRow] + rels: Iterable[AthenaConceptRelationshipRow], ) -> Generator[AthenaConceptRelationshipRow, None, None]: for rel in rels: if rel.concept_id_1 not in not_concept_id_1_values: yield rel def filter_not_c2( - rels: Iterable[AthenaConceptRelationshipRow] + rels: Iterable[AthenaConceptRelationshipRow], ) -> Generator[AthenaConceptRelationshipRow, None, None]: for rel in rels: if rel.concept_id_2 not in not_concept_id_2_values: diff --git a/cardinal_pythonlib/bulk_email/main.py b/cardinal_pythonlib/bulk_email/main.py index 32afb24..f4a1699 100644 --- a/cardinal_pythonlib/bulk_email/main.py +++ b/cardinal_pythonlib/bulk_email/main.py @@ -164,7 +164,7 @@ def add_job( # ------------------------------------------------------------------------- log.info(f"Reading recipients from: {recipients_filename}") - recipient_addresses = set() # type: Set[str] + recipient_addresses: Set[str] = set() recipients_failed = False for recipient_addr in gen_lines_without_comments(recipients_filename): if not is_email_valid(recipient_addr): @@ -175,7 +175,7 @@ def add_job( if recipients_failed: raise ValueError("Errors in recipient addresses") - recipients = [] # type: List[Recipient] + recipients: List[Recipient] = [] for recipient_addr in sorted(recipient_addresses): recipient = Recipient.fetch_or_add(recipient_addr, session) session.add(recipient) @@ -434,7 +434,9 @@ def main() -> None: "--recipients_file", type=str, required=True, - help="Name of a file containing one recipient e-mail address per line.", + help=( + "Name of a file containing one recipient e-mail address per line." + ), ) parser_add_job.add_argument( "--from_addr", type=str, required=True, help="From address." diff --git a/cardinal_pythonlib/bulk_email/models.py b/cardinal_pythonlib/bulk_email/models.py index 8ee0565..51f38ce 100644 --- a/cardinal_pythonlib/bulk_email/models.py +++ b/cardinal_pythonlib/bulk_email/models.py @@ -153,7 +153,7 @@ class Config(Base): # https://dev.mysql.com/doc/refman/8.0/en/create-table-check-constraints.html # noqa primary_key=True, autoincrement=True, - comment=f"Primary key.", + comment="Primary key.", ) database_creation_datetime = Column( DateTime, comment="Date/time this database was created." diff --git a/cardinal_pythonlib/colander_utils.py b/cardinal_pythonlib/colander_utils.py index 909a704..924584e 100644 --- a/cardinal_pythonlib/colander_utils.py +++ b/cardinal_pythonlib/colander_utils.py @@ -468,8 +468,8 @@ def schema_type() -> SchemaType: OptionalPendulumNode = ( - OptionalPendulumNodeLocalTZ -) # synonym for back-compatibility + OptionalPendulumNodeLocalTZ # synonym for back-compatibility +) class OptionalPendulumNodeUTC(SchemaNode): diff --git a/cardinal_pythonlib/datetimefunc.py b/cardinal_pythonlib/datetimefunc.py index 95e58d9..f6cb499 100644 --- a/cardinal_pythonlib/datetimefunc.py +++ b/cardinal_pythonlib/datetimefunc.py @@ -63,8 +63,6 @@ from pendulum.tz import local_timezone from pendulum.tz.timezone import Timezone -from cardinal_pythonlib.logs import main_only_quicksetup_rootlogger - if Arrow is not None: PotentialDatetimeType = Union[ None, datetime.datetime, datetime.date, DateTime, str, Arrow @@ -651,8 +649,8 @@ def duration_to_iso( if permit_years_months: # Watch out here. Before Pendulum 2.1.1, d.total_seconds() ignored # year/month components. But from Pendulum 2.1.1, it incorporates - # year/month information with the assumption that a year is 365 days and - # a month is 30 days, which is perhaps a bit iffy. + # year/month information with the assumption that a year is 365 days + # and a month is 30 days, which is perhaps a bit iffy. y = d.years m = d.months s = get_pendulum_duration_nonyear_nonmonth_seconds(d) @@ -676,7 +674,7 @@ def duration_to_iso( def truncate_date_to_first_of_month( - dt: Optional[DateLikeType] + dt: Optional[DateLikeType], ) -> Optional[DateLikeType]: """ Change the day to the first of the month. diff --git a/cardinal_pythonlib/debugging.py b/cardinal_pythonlib/debugging.py index 0dd1ce2..7f916f6 100644 --- a/cardinal_pythonlib/debugging.py +++ b/cardinal_pythonlib/debugging.py @@ -28,7 +28,6 @@ import ctypes import inspect -from inspect import FrameInfo import logging import pdb import sys @@ -54,7 +53,7 @@ def pdb_run(func: Callable, args: Any, *kwargs: Any) -> Any: # noinspection PyBroadException try: return func(args, *kwargs) - except: # nopep8 + except Exception: type_, value, tb = sys.exc_info() traceback.print_exc() pdb.post_mortem(tb) diff --git a/cardinal_pythonlib/django/fields/isodatetimetz.py b/cardinal_pythonlib/django/fields/isodatetimetz.py index a7a8b96..572aec6 100644 --- a/cardinal_pythonlib/django/fields/isodatetimetz.py +++ b/cardinal_pythonlib/django/fields/isodatetimetz.py @@ -56,7 +56,7 @@ def iso_string_to_python_datetime( - isostring: str + isostring: str, ) -> Optional[datetime.datetime]: """ Takes an ISO-8601 string and returns a ``datetime``. @@ -67,7 +67,7 @@ def iso_string_to_python_datetime( def python_utc_datetime_to_sqlite_strftime_string( - value: datetime.datetime + value: datetime.datetime, ) -> str: """ Converts a Python datetime to a string literal compatible with SQLite, @@ -246,8 +246,8 @@ def get_prep_value(self, value): Convert Python value to database value for QUERYING. We query with UTC, so this function converts datetime values to UTC. - Calls to this function are followed by calls to ``get_db_prep_value()``, - which is for backend-specific conversions. + Calls to this function are followed by calls to + ``get_db_prep_value()``, which is for backend-specific conversions. """ log.debug("get_prep_value: {}, {}", value, type(value)) if not value: diff --git a/cardinal_pythonlib/django/function_cache.py b/cardinal_pythonlib/django/function_cache.py index f663b19..a994ecb 100644 --- a/cardinal_pythonlib/django/function_cache.py +++ b/cardinal_pythonlib/django/function_cache.py @@ -87,8 +87,8 @@ def make_cache_key(call_signature: str, debug_cache: bool = False) -> str: is ``'a.b.c'`` a ``TableId``, or is it a ``ColumnId`` with no ``'db'`` field? - We could use ``repr()``: sometimes that gives us helpful things that - could in principle be passed to ``eval()``, in which case ``repr()`` would - be fine, but sometimes it doesn't, and gives unhelpful things like + could in principle be passed to ``eval()``, in which case ``repr()`` + would be fine, but sometimes it doesn't, and gives unhelpful things like ``'<__main__.Thing object at 0x7ff3093ebda0>'``. - However, if something encodes to JSON, that representation should be reversible and thus contain the right sort of information. diff --git a/cardinal_pythonlib/django/tools/generate_new_django_secret_key.py b/cardinal_pythonlib/django/tools/generate_new_django_secret_key.py index f3b0026..96759d4 100644 --- a/cardinal_pythonlib/django/tools/generate_new_django_secret_key.py +++ b/cardinal_pythonlib/django/tools/generate_new_django_secret_key.py @@ -35,10 +35,10 @@ def main() -> None: """ Generates a new Django secret key and prints it to stdout. """ - chars = 'abcdefghijklmnopqrstuvwxyz0123456789!@#$%^&(-_=+)' # valid chars + chars = "abcdefghijklmnopqrstuvwxyz0123456789!@#$%^&(-_=+)" # valid chars key = get_random_string(50, chars) print(key) -if __name__ == '__main__': +if __name__ == "__main__": main() diff --git a/cardinal_pythonlib/email/sendmail.py b/cardinal_pythonlib/email/sendmail.py index 891c50d..d90c73c 100755 --- a/cardinal_pythonlib/email/sendmail.py +++ b/cardinal_pythonlib/email/sendmail.py @@ -66,21 +66,24 @@ # Make e-mail message # ============================================================================= -def make_email(from_addr: str, - date: str = None, - sender: str = "", - reply_to: Union[str, List[str]] = "", - to: Union[str, List[str]] = "", - cc: Union[str, List[str]] = "", - bcc: Union[str, List[str]] = "", - subject: str = "", - body: str = "", - content_type: str = CONTENT_TYPE_TEXT, - charset: str = UTF8, - attachment_filenames: Sequence[str] = None, - attachment_binaries: Sequence[bytes] = None, - attachment_binary_filenames: Sequence[str] = None, - verbose: bool = False) -> email.mime.multipart.MIMEMultipart: + +def make_email( + from_addr: str, + date: str = None, + sender: str = "", + reply_to: Union[str, List[str]] = "", + to: Union[str, List[str]] = "", + cc: Union[str, List[str]] = "", + bcc: Union[str, List[str]] = "", + subject: str = "", + body: str = "", + content_type: str = CONTENT_TYPE_TEXT, + charset: str = UTF8, + attachment_filenames: Sequence[str] = None, + attachment_binaries: Sequence[bytes] = None, + attachment_binary_filenames: Sequence[str] = None, + verbose: bool = False, +) -> email.mime.multipart.MIMEMultipart: """ Makes an e-mail message. @@ -129,6 +132,7 @@ def make_email(from_addr: str, :exc:`AssertionError`, :exc:`ValueError` """ + def _csv_list_to_list(x: str) -> List[str]: stripped = [item.strip() for item in x.split(COMMA)] return [item for item in stripped if item] @@ -137,9 +141,9 @@ def _assert_nocomma(x: Union[str, List[str]]) -> None: if isinstance(x, str): x = [x] for _addr in x: - assert COMMA not in _addr, ( - f"Commas not allowed in e-mail addresses: {_addr!r}" - ) + assert ( + COMMA not in _addr + ), f"Commas not allowed in e-mail addresses: {_addr!r}" # ------------------------------------------------------------------------- # Arguments @@ -151,9 +155,9 @@ def _assert_nocomma(x: Union[str, List[str]]) -> None: f"(for Python sendmail, not RFC 2822); was {from_addr!r}" ) _assert_nocomma(from_addr) - assert isinstance(sender, str), ( - f"'Sender:' can only be a single address; was {sender!r}" - ) + assert isinstance( + sender, str + ), f"'Sender:' can only be a single address; was {sender!r}" _assert_nocomma(sender) if isinstance(reply_to, str): reply_to = [reply_to] if reply_to else [] # type: List[str] @@ -169,11 +173,13 @@ def _assert_nocomma(x: Union[str, List[str]]) -> None: _assert_nocomma(cc) _assert_nocomma(bcc) attachment_filenames = attachment_filenames or [] # type: List[str] - assert all(attachment_filenames), ( - f"Missing attachment filenames: {attachment_filenames!r}" - ) + assert all( + attachment_filenames + ), f"Missing attachment filenames: {attachment_filenames!r}" attachment_binaries = attachment_binaries or [] # type: List[bytes] - attachment_binary_filenames = attachment_binary_filenames or [] # type: List[str] # noqa + attachment_binary_filenames = ( + attachment_binary_filenames or [] + ) # type: List[str] # noqa assert len(attachment_binaries) == len(attachment_binary_filenames), ( "If you specify attachment_binaries or attachment_binary_filenames, " "they must be iterables of the same length." @@ -229,8 +235,8 @@ def _assert_nocomma(x: Union[str, List[str]]) -> None: part.set_payload(open(f, "rb").read()) email.encoders.encode_base64(part) part.add_header( - 'Content-Disposition', - 'attachment; filename="%s"' % os.path.basename(f) + "Content-Disposition", + 'attachment; filename="%s"' % os.path.basename(f), ) msg.attach(part) if attachment_binaries: @@ -238,8 +244,10 @@ def _assert_nocomma(x: Union[str, List[str]]) -> None: # Binary attachments, which have a notional filename # ----------------------------------------------------------------- if verbose: - log.debug("attachment_binary_filenames: {}", - attachment_binary_filenames) + log.debug( + "attachment_binary_filenames: {}", + attachment_binary_filenames, + ) for i in range(len(attachment_binaries)): blob = attachment_binaries[i] filename = attachment_binary_filenames[i] @@ -247,8 +255,9 @@ def _assert_nocomma(x: Union[str, List[str]]) -> None: part.set_payload(blob) email.encoders.encode_base64(part) part.add_header( - 'Content-Disposition', - 'attachment; filename="%s"' % filename) + "Content-Disposition", + 'attachment; filename="%s"' % filename, + ) msg.attach(part) except Exception as e: raise ValueError(f"send_email: Failed to attach files: {e}") @@ -260,15 +269,18 @@ def _assert_nocomma(x: Union[str, List[str]]) -> None: # Send message # ============================================================================= -def send_msg(from_addr: str, - to_addrs: Union[str, List[str]], - host: str, - user: str, - password: str, - port: int = None, - use_tls: bool = True, - msg: email.mime.multipart.MIMEMultipart = None, - msg_string: str = None) -> None: + +def send_msg( + from_addr: str, + to_addrs: Union[str, List[str]], + host: str, + user: str, + password: str, + port: int = None, + use_tls: bool = True, + msg: email.mime.multipart.MIMEMultipart = None, + msg_string: str = None, +) -> None: """ Sends a pre-built e-mail message. @@ -303,7 +315,8 @@ def send_msg(from_addr: str, # e.g. ConnectionRefusedError when creating the socket # SMTPException is a subclass of OSError since 3.4 raise RuntimeError( - f"send_msg: Failed to connect to host {host}, port {port}: {e}") + f"send_msg: Failed to connect to host {host}, port {port}: {e}" + ) try: session.ehlo() except smtplib.SMTPException as e: @@ -322,7 +335,8 @@ def send_msg(from_addr: str, session.login(user, password) except smtplib.SMTPException as e: raise RuntimeError( - f"send_msg: Failed to login as user {user}: {e}") + f"send_msg: Failed to login as user {user}: {e}" + ) else: log.debug("Not using SMTP AUTH; no user specified") # For systems with... lax... security requirements @@ -341,26 +355,29 @@ def send_msg(from_addr: str, # Send e-mail # ============================================================================= -def send_email(from_addr: str, - host: str, - user: str, - password: str, - port: int = None, - use_tls: bool = True, - date: str = None, - sender: str = "", - reply_to: Union[str, List[str]] = "", - to: Union[str, List[str]] = "", - cc: Union[str, List[str]] = "", - bcc: Union[str, List[str]] = "", - subject: str = "", - body: str = "", - content_type: str = CONTENT_TYPE_TEXT, - charset: str = UTF8, - attachment_filenames: Sequence[str] = None, - attachment_binaries: Sequence[bytes] = None, - attachment_binary_filenames: Sequence[str] = None, - verbose: bool = False) -> Tuple[bool, str]: + +def send_email( + from_addr: str, + host: str, + user: str, + password: str, + port: int = None, + use_tls: bool = True, + date: str = None, + sender: str = "", + reply_to: Union[str, List[str]] = "", + to: Union[str, List[str]] = "", + cc: Union[str, List[str]] = "", + bcc: Union[str, List[str]] = "", + subject: str = "", + body: str = "", + content_type: str = CONTENT_TYPE_TEXT, + charset: str = UTF8, + attachment_filenames: Sequence[str] = None, + attachment_binaries: Sequence[bytes] = None, + attachment_binary_filenames: Sequence[str] = None, + verbose: bool = False, +) -> Tuple[bool, str]: """ Sends an e-mail in text/html format using SMTP via TLS. @@ -546,6 +563,7 @@ def get_email_domain(email_: str) -> str: # Parse command line # ============================================================================= + def main() -> NoReturn: """ Command-line processor. See ``--help`` for details. @@ -553,29 +571,25 @@ def main() -> NoReturn: logging.basicConfig() log.setLevel(logging.DEBUG) parser = argparse.ArgumentParser( - description="Send an e-mail from the command line.") - parser.add_argument("sender", action="store", - help="Sender's e-mail address") - parser.add_argument("host", action="store", - help="SMTP server hostname") - parser.add_argument("user", action="store", - help="SMTP username") - parser.add_argument("password", action="store", - help="SMTP password") - parser.add_argument("recipient", action="append", - help="Recipient e-mail address(es)") - parser.add_argument("subject", action="store", - help="Message subject") - parser.add_argument("body", action="store", - help="Message body") - parser.add_argument("--attach", nargs="", - help="Filename(s) to attach") - parser.add_argument("--tls", action="store_false", - help="Use TLS connection security") - parser.add_argument("--verbose", action="store_true", - help="Be verbose") - parser.add_argument("-h --help", action="help", - help="Prints this help") + description="Send an e-mail from the command line." + ) + parser.add_argument( + "sender", action="store", help="Sender's e-mail address" + ) + parser.add_argument("host", action="store", help="SMTP server hostname") + parser.add_argument("user", action="store", help="SMTP username") + parser.add_argument("password", action="store", help="SMTP password") + parser.add_argument( + "recipient", action="append", help="Recipient e-mail address(es)" + ) + parser.add_argument("subject", action="store", help="Message subject") + parser.add_argument("body", action="store", help="Message body") + parser.add_argument("--attach", nargs="", help="Filename(s) to attach") + parser.add_argument( + "--tls", action="store_false", help="Use TLS connection security" + ) + parser.add_argument("--verbose", action="store_true", help="Be verbose") + parser.add_argument("-h --help", action="help", help="Prints this help") args = parser.parse_args() (result, msg) = send_email( from_addr=args.sender, @@ -597,5 +611,5 @@ def main() -> NoReturn: sys.exit(0 if result else 1) -if __name__ == '__main__': +if __name__ == "__main__": main() diff --git a/cardinal_pythonlib/extract_text.py b/cardinal_pythonlib/extract_text.py index 04d0784..1209ce7 100755 --- a/cardinal_pythonlib/extract_text.py +++ b/cardinal_pythonlib/extract_text.py @@ -1574,13 +1574,13 @@ def main() -> None: help="Use semi-plain format (re e.g. table layouts)", ) parser.add_argument( - "--width", type=int, default=DEFAULT_WIDTH, help=f"Word wrapping width" + "--width", type=int, default=DEFAULT_WIDTH, help="Word wrapping width" ) parser.add_argument( "--min-col-width", type=int, default=DEFAULT_MIN_COL_WIDTH, - help=f"Minimum column width for tables", + help="Minimum column width for tables", ) args = parser.parse_args() if args.availability: diff --git a/cardinal_pythonlib/file_io.py b/cardinal_pythonlib/file_io.py index 3ac558f..3bbc5f2 100644 --- a/cardinal_pythonlib/file_io.py +++ b/cardinal_pythonlib/file_io.py @@ -219,7 +219,7 @@ def gen_lines_without_comments( def gen_textfiles_from_filenames( - filenames: Iterable[str] + filenames: Iterable[str], ) -> Generator[TextIO, None, None]: """ Generates file-like objects from a list of filenames. @@ -237,7 +237,7 @@ def gen_textfiles_from_filenames( def gen_lines_from_textfiles( - files: Iterable[TextIO] + files: Iterable[TextIO], ) -> Generator[str, None, None]: """ Generates lines from file-like objects. diff --git a/cardinal_pythonlib/getch.py b/cardinal_pythonlib/getch.py index 973fe0c..fb3befd 100644 --- a/cardinal_pythonlib/getch.py +++ b/cardinal_pythonlib/getch.py @@ -41,7 +41,7 @@ except ImportError: msvcrt = None import termios # Unix only - import tty # requires termios, so Unix only + import tty # requires termios, so Unix only # noqa: F401 # ============================================================================= diff --git a/cardinal_pythonlib/hash.py b/cardinal_pythonlib/hash.py index 5362e85..105214a 100644 --- a/cardinal_pythonlib/hash.py +++ b/cardinal_pythonlib/hash.py @@ -73,7 +73,7 @@ # https://docs.python.org/3/library/platform.html#platform.architecture -IS_64_BIT = sys.maxsize > 2 * 32 +IS_64_BIT = sys.maxsize > 232 TIMING_HASH = "hash" @@ -485,7 +485,7 @@ def murmur3_64(data: Union[bytes, bytearray], seed: int = 19820125) -> int: m = 0xC6A4A7935BD1E995 r = 47 - mask = 2 64 - 1 + mask = 264 - 1 length = len(data) diff --git a/cardinal_pythonlib/interval.py b/cardinal_pythonlib/interval.py index b5d1294..4372044 100644 --- a/cardinal_pythonlib/interval.py +++ b/cardinal_pythonlib/interval.py @@ -59,9 +59,9 @@ datetime.datetime.strptime(x, "%Y-%m-%d").date() for x in [ # https://www.gov.uk/bank-holidays - # All bank holiday dates vary, even the date-based ones; e.g. if Christmas - # Day is a Sunday, then the Christmas Day substitute bank holiday is Tue 27 - # Dec, after the Boxing Day Monday bank holiday. + # All bank holiday dates vary, even the date-based ones; e.g. if + # Christmas Day is a Sunday, then the Christmas Day substitute bank + # holiday is Tue 27 Dec, after the Boxing Day Monday bank holiday. # 2014 "2014-01-01", # New Year's Day "2014-04-18", # Good Friday diff --git a/cardinal_pythonlib/lists.py b/cardinal_pythonlib/lists.py index fc7046a..38823a1 100644 --- a/cardinal_pythonlib/lists.py +++ b/cardinal_pythonlib/lists.py @@ -186,13 +186,13 @@ def count_bool(blist: Iterable[Any]) -> int: def delete_elements_by_index(x: List[Any], indices: Sequence[int]) -> None: """ Deletes (in place) objects from a list, by (zero-based) index values. - + Args: x: list to modify indices: zero-based index values at which to delete elements of x - + Note: - If you pass duplicate values, unexpected results will occur. - If you pass out-of-bound indices, :exc:`IndexError` will be raised. diff --git a/cardinal_pythonlib/logs.py b/cardinal_pythonlib/logs.py index d6fd5e5..2704492 100644 --- a/cardinal_pythonlib/logs.py +++ b/cardinal_pythonlib/logs.py @@ -593,7 +593,7 @@ def emit(self, record: logging.LogRecord) -> None: try: html = self.format(record) self.logfunction(html) - except: # nopep8 + except Exception: self.handleError(record) diff --git a/cardinal_pythonlib/maths_numpy.py b/cardinal_pythonlib/maths_numpy.py index f26c748..1c46358 100644 --- a/cardinal_pythonlib/maths_numpy.py +++ b/cardinal_pythonlib/maths_numpy.py @@ -229,7 +229,7 @@ def _test_softmax() -> None: [1, 1, 1, 1, 1.01], [1, 2, 3, 4, 5], [1, 2, 3, 4, 1000], - [1, 2, 3, 4, 5.0 400], + [1, 2, 3, 4, 5.0*400], ] betas = [0, 0.5, 1, 2, 10, 100] for x in arrays: diff --git a/cardinal_pythonlib/module_version.py b/cardinal_pythonlib/module_version.py index 9e21988..ea9a67b 100644 --- a/cardinal_pythonlib/module_version.py +++ b/cardinal_pythonlib/module_version.py @@ -40,7 +40,7 @@ from semantic_version import Version # noinspection PyUnresolvedReferences -import cardinal_pythonlib.ensure_test_executed_correctly +import cardinal_pythonlib.ensure_test_executed_correctly # noqa: F401 if sys.version_info > (3, 8): # Python 3.8 or higher diff --git a/cardinal_pythonlib/nhs.py b/cardinal_pythonlib/nhs.py index 31c7a3d..dfd690b 100644 --- a/cardinal_pythonlib/nhs.py +++ b/cardinal_pythonlib/nhs.py @@ -163,7 +163,7 @@ def test_nhs_rng(n: int = 100) -> None: def generate_nhs_number_from_first_9_digits( - first9digits: str + first9digits: str, ) -> Optional[int]: """ Returns a valid NHS number, as an ``int``, given the first 9 digits. diff --git a/cardinal_pythonlib/openxml/find_recovered_openxml.py b/cardinal_pythonlib/openxml/find_recovered_openxml.py index ff5f193..deca194 100644 --- a/cardinal_pythonlib/openxml/find_recovered_openxml.py +++ b/cardinal_pythonlib/openxml/find_recovered_openxml.py @@ -123,7 +123,8 @@ (SOURCE_STDOUT, " or ENTER (try reading this split again): ", "q\n"), ( SOURCE_STDERR, - "zip: malloc.c:2394: sysmalloc: Assertion `(old_top == initial_top (av) " + "zip: malloc.c:2394: sysmalloc: " + "Assertion `(old_top == initial_top (av) " "&& old_size == 0) \|\| ((unsigned long) (old_size) >= MINSIZE && " "prev_inuse (old_top) && ((unsigned long) old_end & (pagesize - 1)) " "== 0)' failed.", @@ -546,7 +547,9 @@ def main() -> None: "filename": filename, "filetypes": filetypes, "move_to": args.move_to, - "delete_if_not_specified_file_type": args.delete_if_not_specified_file_type, + "delete_if_not_specified_file_type": ( + args.delete_if_not_specified_file_type + ), "show_zip_output": args.show_zip_output, } # log.critical("start") diff --git a/cardinal_pythonlib/platformfunc.py b/cardinal_pythonlib/platformfunc.py index 816d952..a7e0acd 100644 --- a/cardinal_pythonlib/platformfunc.py +++ b/cardinal_pythonlib/platformfunc.py @@ -48,39 +48,6 @@ def fix_windows_utf8_output() -> None: # Python 3 only now, so nothing to do return - # if six.PY3: - # return - # reload_module(sys) - # # noinspection PyUnresolvedReferences - # sys.setdefaultencoding('utf-8') - # # print sys.getdefaultencoding() - # - # if sys.platform == 'win32': - # try: - # import win32console - # except ImportError: - # win32console = None - # print( - # "Python Win32 Extensions module is required.\n " - # "You can download it from " - # "https://sourceforge.net/projects/pywin32/ " - # "(x86 and x64 builds are available)\n") - # exit(-1) - # # win32console implementation of SetConsoleCP does not return a value - # # CP_UTF8 = 65001 - # win32console.SetConsoleCP(65001) - # if win32console.GetConsoleCP() != 65001: - # raise RuntimeError("Cannot set console codepage to 65001 (UTF-8)") - # win32console.SetConsoleOutputCP(65001) - # if win32console.GetConsoleOutputCP() != 65001: - # raise RuntimeError("Cannot set console output codepage to 65001 " - # "(UTF-8)") - # - # sys.stdout = codecs.getwriter('utf8')(sys.stdout) - # sys.stderr = codecs.getwriter('utf8')(sys.stderr) - # # CHECK: does that modify the "global" sys.stdout? - # # You can't use "global sys.stdout"; that raises an error - def test_windows_utf8_output() -> None: """ @@ -262,7 +229,7 @@ def handle_line(line: str) -> Tuple[str, str]: # RNC: as function lines = gen_lines() # RNC # consume whatever output occurs until the tag is reached - consume(itertools.takewhile(lambda l: tag not in l, lines)) + consume(itertools.takewhile(lambda line: tag not in line, lines)) # ... RNC: note that itertools.takewhile() generates values not matching # the condition, but then consumes the condition value itself. So the # tag's already gone. Example: diff --git a/cardinal_pythonlib/probability.py b/cardinal_pythonlib/probability.py index 9f9efcc..19f32b6 100644 --- a/cardinal_pythonlib/probability.py +++ b/cardinal_pythonlib/probability.py @@ -87,7 +87,7 @@ def ln(x: float) -> float: # return math_ln(x) if x != 0 else MINUS_INFINITY timeit.timeit('(ln(x) for x in range(1, 100))', number=10000) # 0.012411762960255146 - + # try... except ValueError... timeit.timeit('(ln(x) for x in range(1, 100))', number=10000) # 0.007783170789480209 diff --git a/cardinal_pythonlib/process.py b/cardinal_pythonlib/process.py index 95b9e7b..bfbac32 100644 --- a/cardinal_pythonlib/process.py +++ b/cardinal_pythonlib/process.py @@ -26,6 +26,7 @@ """ +import os import shlex import subprocess import sys @@ -81,7 +82,7 @@ def get_pipe_series_output( # than the maximum specified. # print commands - processes = [] # type: List[subprocess.Popen] + processes: List[subprocess.Popen] = [] for i in range(len(commands)): if i == 0: # first processes processes.append( diff --git a/cardinal_pythonlib/psychiatry/drugs.py b/cardinal_pythonlib/psychiatry/drugs.py index ce1b9e5..0cffc75 100644 --- a/cardinal_pythonlib/psychiatry/drugs.py +++ b/cardinal_pythonlib/psychiatry/drugs.py @@ -560,8 +560,8 @@ def regex_to_sql_like( Args: regex_text: regular expression text to work with single_wildcard: SQL single wildcard, typically an underscore - zero_or_more_wildcard: SQL "zero/one/many" wildcard, probably always - a percent symbol + zero_or_more_wildcard: SQL "zero/one/many" wildcard, probably + always a percent symbol Returns: string for an SQL string literal @@ -1468,7 +1468,7 @@ def all_drugs_where( print([d.generic_name for d in non_ssri_antidep]) conventional_antidep = all_drugs_where(conventional_antidepressant=True) print([d.generic_name for d in conventional_antidep]) - """ + """ # noqa: E501 matching_drugs = [] # type: List[Drug] for drug in DRUGS: if drug.category_not_drug and not include_categories: diff --git a/cardinal_pythonlib/psychiatry/simhelpers.py b/cardinal_pythonlib/psychiatry/simhelpers.py index 4f2311d..46ade62 100644 --- a/cardinal_pythonlib/psychiatry/simhelpers.py +++ b/cardinal_pythonlib/psychiatry/simhelpers.py @@ -58,7 +58,8 @@ def _central_value(values: Union[List, Tuple], name: str = "?") -> Any: """ if not isinstance(values, (list, tuple)): raise ValueError( - f"Parameter {name!r} should have been list/tuple but was {values!r}" + f"Parameter {name!r} should have been list/tuple but was " + f"{values!r}" ) n = len(values) if n == 0: @@ -96,13 +97,13 @@ def gen_params_around_centre( Collins, 2019, PMID 31769410, Box 2/Figure 1B), because simulating a subject that chooses based on calculated probabilities introduces random noise. You will want to simulate a range of parameter combinations. - + You might want to explore the entire parameter space (in which case, see :func:`cardinal_pythonlib.iterhelp.product_dict`), but that may be computationally unfeasible -- creating the simulations takes time and energy, and analysing them to recover those parameters takes time and energy. - + An alternative strategy is to take central values of all parameters, and then vary one parameter at a time about those central values. For example, if you have n = 6 parameters with k = 5 possible values each, the diff --git a/cardinal_pythonlib/psychiatry/timeline.py b/cardinal_pythonlib/psychiatry/timeline.py index 4f4783e..97265cf 100755 --- a/cardinal_pythonlib/psychiatry/timeline.py +++ b/cardinal_pythonlib/psychiatry/timeline.py @@ -223,7 +223,7 @@ # reticulate_1.11.1-9000 [see sessionInfo()]... # ... yes, that fixes it. -""" +""" # noqa: E501 # ============================================================================= # Imports diff --git a/cardinal_pythonlib/psychiatry/treatment_resistant_depression.py b/cardinal_pythonlib/psychiatry/treatment_resistant_depression.py index 5396349..0a58508 100755 --- a/cardinal_pythonlib/psychiatry/treatment_resistant_depression.py +++ b/cardinal_pythonlib/psychiatry/treatment_resistant_depression.py @@ -467,7 +467,7 @@ def _get_patient_result(_patient_id: str) -> Optional[DataFrame]: def test_two_antidepressant_episodes( - n_sets: int = TEST_PATIENT_MULTIPLE + n_sets: int = TEST_PATIENT_MULTIPLE, ) -> None: fluox = "fluoxetine" cital = "citalopram" diff --git a/cardinal_pythonlib/pyramid/requests.py b/cardinal_pythonlib/pyramid/requests.py index 8afdefb..d444d3c 100644 --- a/cardinal_pythonlib/pyramid/requests.py +++ b/cardinal_pythonlib/pyramid/requests.py @@ -63,7 +63,7 @@ def gen_accept_encoding_definitions( - accept_encoding: str + accept_encoding: str, ) -> Generator[str, None, None]: """ For a given HTTP ``Accept-Encoding`` field value, generate encoding @@ -150,7 +150,7 @@ def gen_content_encodings(request: Request) -> Generator[str, None, None]: def gen_content_encodings_reversed( - request: Request + request: Request, ) -> Generator[str, None, None]: """ Generates content encodings in reverse order -- that is, in the order diff --git a/cardinal_pythonlib/rate_limiting.py b/cardinal_pythonlib/rate_limiting.py index 8fc733c..0752c29 100644 --- a/cardinal_pythonlib/rate_limiting.py +++ b/cardinal_pythonlib/rate_limiting.py @@ -30,8 +30,6 @@ from time import perf_counter, sleep from typing import Any, Callable, Optional, Union -from cardinal_pythonlib.logs import main_only_quicksetup_rootlogger - log = logging.getLogger(__name__) FuncType = Callable[..., Any] diff --git a/cardinal_pythonlib/rnc_db.py b/cardinal_pythonlib/rnc_db.py index 9d9f554..0eaf3f2 100644 --- a/cardinal_pythonlib/rnc_db.py +++ b/cardinal_pythonlib/rnc_db.py @@ -188,7 +188,7 @@ ) from cardinal_pythonlib.logs import get_brace_style_log_with_null_handler -from cardinal_pythonlib.sql.validation import +from cardinal_pythonlib.sql.validation import split_long_sqltype # 1. An ODBC driver try: @@ -959,7 +959,7 @@ def get_sql_select_all_non_pk_fields_by_pk( table: str, fieldlist: Sequence[str], delims: Tuple[str, str] = ("", "") ) -> str: """Returns SQL: - SELECT [all but the first field] WHERE [the first field] = ? + SELECT [all but the first field] WHERE [the first field] = ? """ return ( "SELECT " @@ -979,7 +979,7 @@ def get_sql_select_all_fields_by_key( delims: Tuple[str, str] = ("", ""), ) -> str: """Returns SQL: - SELECT [all fields in the fieldlist] WHERE [keyname] = ? + SELECT [all fields in the fieldlist] WHERE [keyname] = ? """ return ( "SELECT " @@ -1010,7 +1010,8 @@ def get_sql_insert( def get_sql_insert_or_update( table: str, fieldlist: Sequence[str], delims: Tuple[str, str] = ("", "") ) -> str: - """Returns ?-marked SQL for an INSERT-or-if-duplicate-key-UPDATE statement. + """ + Returns ?-marked SQL for an INSERT-or-if-duplicate-key-UPDATE statement. """ # https://stackoverflow.com/questions/4205181 return """ @@ -1104,8 +1105,8 @@ def debug_object(obj: T) -> str: """Prints key/value pairs for an object's dictionary.""" pairs = [] for k, v in vars(obj).items(): - pairs.append(u"{}={}".format(k, v)) - return u", ".join(pairs) + pairs.append("{}={}".format(k, v)) + return ", ".join(pairs) def dump_database_object(obj: T, fieldlist: Iterable[str]) -> None: @@ -1113,7 +1114,7 @@ def dump_database_object(obj: T, fieldlist: Iterable[str]) -> None: log.info(_LINE_EQUALS) log.info("DUMP OF: {}", obj) for f in fieldlist: - log.info(u"{f}: {v}", f=f, v=getattr(obj, f)) + log.info("{f}: {v}", f=f, v=getattr(obj, f)) log.info(_LINE_EQUALS) @@ -1138,8 +1139,8 @@ def create_object_from_list( *kwargs ) -> T: """ - Create an object by instantiating ``cls(args, *kwargs)`` and assigning the - values in ``valuelist`` to the fields in ``fieldlist``. + Create an object by instantiating ``cls(args, *kwargs)`` and assigning + the values in ``valuelist`` to the fields in ``fieldlist``. If ``construct_with_pk`` is ``True``, initialize with ``cls(valuelist[0], args, *kwargs)`` @@ -1293,7 +1294,7 @@ def reconfigure_jaydebeapi() -> None: log.warning("Old jaydebeapi version") # noinspection PyUnresolvedReferences,PyProtectedMember converters = jaydebeapi.dbapi2._DEFAULT_CONVERTERS - except: # nopep8 + except Exception: raise AssertionError( "Don't know how to hook into this version of JayDeBeApi" ) @@ -1499,7 +1500,7 @@ def get_database( autocommit=autocommit, # if False, need to commit ) return db - except: # nopep8 + except Exception: if securely: raise NoDatabaseError( "Problem opening or reading from database {}; details " @@ -1517,7 +1518,7 @@ def get_database_from_configparser( dbc = DatabaseConfig(parser, section) db = dbc.get_database(securely=securely) return db - except: # nopep8 + except Exception: if securely: raise NoDatabaseError( "Problem opening or reading from database {}; details " @@ -1573,8 +1574,8 @@ def connect( use_unicode: bool = True, ) -> bool: """ - engine: access, mysql, sqlserver - interface: mysql, odbc, jdbc + engine: access, mysql, sqlserver + interface: mysql, odbc, jdbc """ # Catch all exceptions, so the error-catcher never shows a password. @@ -1811,8 +1812,10 @@ def _connect( dr=driver, s=host, db=database, u=user ) ) - connectstring = "DRIVER={};SERVER={};DATABASE={};UID={};PWD={}".format( - driver, host, database, user, password + connectstring = ( + "DRIVER={};SERVER={};DATABASE={};UID={};PWD={}".format( + driver, host, database, user, password + ) ) # noinspection PyUnresolvedReferences self.db = pyodbc.connect(connectstring, unicode_results=True) @@ -2117,7 +2120,7 @@ def insert_record( new_pk = get_pk_of_last_insert(cursor) log.debug("Record inserted.") return new_pk - except: # nopep8 + except Exception: log.exception("insert_record: Failed to insert record.") raise @@ -2165,7 +2168,7 @@ def insert_record_by_dict( new_pk = get_pk_of_last_insert(cursor) log.debug("Record inserted.") return new_pk - except: # nopep8 + except Exception: log.exception("insert_record_by_dict: Failed to insert record.") raise @@ -2191,7 +2194,7 @@ def insert_multiple_records( # http://www.python.org/dev/peps/pep-0249/ log.debug("Records inserted.") return cursor.rowcount - except: # nopep8 + except Exception: log.exception("insert_multiple_records: Failed to insert records.") raise @@ -2203,7 +2206,7 @@ def db_exec_with_cursor(self, cursor, sql: str, args) -> int: debug_sql(sql, args) cursor.execute(sql, args) return cursor.rowcount - except: # nopep8 + except Exception: log.exception("db_exec_with_cursor: SQL was: " + sql) raise # MySQLdb: @@ -2233,7 +2236,7 @@ def db_exec_literal(self, sql: str) -> int: try: cursor.execute(sql) return cursor.rowcount - except: # nopep8 + except Exception: log.exception("db_exec_literal: SQL was: " + sql) raise @@ -2265,7 +2268,7 @@ def fetchone(self, sql: str, args) -> Optional[Sequence[Any]]: self.db_exec_with_cursor(cursor, sql, args) try: return cursor.fetchone() - except: # nopep8 + except Exception: log.exception("fetchone: SQL was: " + sql) raise @@ -2277,7 +2280,7 @@ def fetchall(self, sql: str, args) -> Sequence[Sequence[Any]]: try: rows = cursor.fetchall() return rows - except: # nopep8 + except Exception: log.exception("fetchall: SQL was: " + sql) raise @@ -2291,7 +2294,7 @@ def gen_fetchall(self, sql: str, args) -> Iterator[Sequence[Any]]: while row is not None: yield row row = cursor.fetchone() - except: # nopep8 + except Exception: log.exception("gen_fetchall: SQL was: " + sql) raise @@ -2305,7 +2308,7 @@ def gen_fetchfirst(self, sql: str, args) -> Iterator[Any]: while row is not None: yield row[0] row = cursor.fetchone() - except: # nopep8 + except Exception: log.exception("gen_fetchfirst: SQL was: " + sql) raise @@ -2320,7 +2323,7 @@ def fetchall_with_fieldnames( rows = cursor.fetchall() fieldnames = [i[0] for i in cursor.description] return rows, fieldnames - except: # nopep8 + except Exception: log.exception("fetchall_with_fieldnames: SQL was: " + sql) raise @@ -2337,7 +2340,7 @@ def fetchall_as_dictlist(self, sql: str, args) -> List[Dict[str, Any]]: for r in rows: dictlist.append(dict(zip(fieldnames, r))) return dictlist - except: # nopep8 + except Exception: log.exception("fetchall_as_dictlist: SQL was: " + sql) raise @@ -2353,7 +2356,7 @@ def fetch_fieldnames(self, sql: str, args) -> List[str]: self.db_exec_with_cursor(cursor, sql, args) try: return [i[0] for i in cursor.description] - except: # nopep8 + except Exception: log.exception("fetch_fieldnames: SQL was: " + sql) raise @@ -2696,7 +2699,7 @@ def create_fulltext_index( @staticmethod def fieldnames_from_fieldspeclist( - fieldspeclist: FIELDSPECLIST_TYPE + fieldspeclist: FIELDSPECLIST_TYPE, ) -> List[str]: """Returns fieldnames from a field specification list.""" return [x["name"] for x in fieldspeclist] diff --git a/cardinal_pythonlib/snomed.py b/cardinal_pythonlib/snomed.py index adf8185..7d2063b 100644 --- a/cardinal_pythonlib/snomed.py +++ b/cardinal_pythonlib/snomed.py @@ -195,7 +195,7 @@ def test(s): ret.append(r"\\") elif ord(c) < 32: # two-digit hex format, e.g. \x1F for ASCII 31 - ret.append(fr"\x{ord(c):02X}") + ret.append(rf"\x{ord(c):02X}") else: ret.append(c) ret.append(dquote) diff --git a/cardinal_pythonlib/sphinxtools.py b/cardinal_pythonlib/sphinxtools.py index fd6a2d4..3d196cb 100644 --- a/cardinal_pythonlib/sphinxtools.py +++ b/cardinal_pythonlib/sphinxtools.py @@ -47,15 +47,18 @@ sep, splitext, ) -from typing import Dict, Iterable, List, NoReturn, Union +from typing import Dict, Iterable, List, NoReturn, TYPE_CHECKING, Union from cardinal_pythonlib.fileops import mkdir_p, relative_filename_within_dir from cardinal_pythonlib.logs import BraceStyleAdapter from cardinal_pythonlib.reprfunc import auto_repr -from pygments.lexer import Lexer from pygments.lexers import get_lexer_for_filename from pygments.util import ClassNotFound +if TYPE_CHECKING: + from pygments.lexer import Lexer + + log = BraceStyleAdapter(logging.getLogger(__name__)) @@ -172,6 +175,7 @@ def filename_matches_glob(filename: str, globtext: str) -> bool: # Otherwise: return False + # ============================================================================= # FileToAutodocument # ============================================================================= @@ -771,7 +775,7 @@ def add_source_files( source_filename ), method=method, - source_rst_title_style_python=source_rst_title_style_python, + source_rst_title_style_python=source_rst_title_style_python, # noqa: E501 pygments_language_override=pygments_language_override, ) ) diff --git a/cardinal_pythonlib/spreadsheets.py b/cardinal_pythonlib/spreadsheets.py index 5558d12..b72743e 100644 --- a/cardinal_pythonlib/spreadsheets.py +++ b/cardinal_pythonlib/spreadsheets.py @@ -892,8 +892,8 @@ def __init__(self, sheetholder: SheetHolder, row: int) -> None: self.sheetholder = sheetholder self.row = row # zero-based index of our row self._next_col = ( - 0 - ) # zero-based column index of the next column to read # noqa + 0 # zero-based column index of the next column to read # noqa + ) # ------------------------------------------------------------------------- # Information diff --git a/cardinal_pythonlib/sql/literals.py b/cardinal_pythonlib/sql/literals.py index ef6b8dc..beece64 100644 --- a/cardinal_pythonlib/sql/literals.py +++ b/cardinal_pythonlib/sql/literals.py @@ -42,8 +42,8 @@ def sql_string_literal(text: str) -> str: """ - Transforms text into its ANSI SQL-quoted version, e.g. (in Python ``repr()`` - format): + Transforms text into its ANSI SQL-quoted version, e.g. (in Python + ``repr()`` format): .. code-block:: none diff --git a/cardinal_pythonlib/sql/sql_grammar_mssql.py b/cardinal_pythonlib/sql/sql_grammar_mssql.py index 4bcf1ba..c7f8b44 100644 --- a/cardinal_pythonlib/sql/sql_grammar_mssql.py +++ b/cardinal_pythonlib/sql/sql_grammar_mssql.py @@ -393,7 +393,8 @@ class SqlGrammarMSSQLServer(SqlGrammar): opAssoc.LEFT, ), ((BETWEEN, AND), TERNARY_OP, opAssoc.LEFT), - # CASE handled above (hoping precedence is not too much of a problem) + # CASE handled above (hoping precedence is not too much of a + # problem) (NOT, UNARY_OP, opAssoc.RIGHT), (AND \| "&&" \| OR \| "\|\|" \| ":=", BINARY_OP, opAssoc.LEFT), ], diff --git a/cardinal_pythonlib/sql/sql_grammar_mysql.py b/cardinal_pythonlib/sql/sql_grammar_mysql.py index 7467c71..8d9e1c0 100644 --- a/cardinal_pythonlib/sql/sql_grammar_mysql.py +++ b/cardinal_pythonlib/sql/sql_grammar_mysql.py @@ -380,8 +380,9 @@ class SqlGrammarMySQL(SqlGrammar): argument_list = ( delimitedList(expr).setName("arglist").setParseAction(", ".join) ) - # ... we don't care about sub-parsing the argument list, so use combine=True - # or setParseAction: https://stackoverflow.com/questions/37926516 + # ... we don't care about sub-parsing the argument list, so use + # combine=True or setParseAction: + # https://stackoverflow.com/questions/37926516 function_call = Combine(function_name + LPAR) + argument_list + RPAR # http://dev.mysql.com/doc/refman/5.7/en/partitioning-selection.html @@ -516,7 +517,8 @@ class SqlGrammarMySQL(SqlGrammar): opAssoc.LEFT, ), ((BETWEEN, AND), TERNARY_OP, opAssoc.LEFT), - # CASE handled above (hoping precedence is not too much of a problem) + # CASE handled above (hoping precedence is not too much of a + # problem) (NOT, UNARY_OP, opAssoc.RIGHT), (AND \| "&&" \| XOR \| OR \| "\|\|" \| ":=", BINARY_OP, opAssoc.LEFT), ], @@ -577,7 +579,7 @@ class SqlGrammarMySQL(SqlGrammar): \| BIT_OR \| BIT_XOR \| COUNT - \| GROUP_CONCAT # also: special handling for COUNT(DISTINCT ...), see below + \| GROUP_CONCAT # also: special handling for COUNT(DISTINCT ...), see below # noqa: E501 \| MAX \| MIN \| STD diff --git a/cardinal_pythonlib/sqlalchemy/alembic_func.py b/cardinal_pythonlib/sqlalchemy/alembic_func.py index da7356d..7d54dac 100644 --- a/cardinal_pythonlib/sqlalchemy/alembic_func.py +++ b/cardinal_pythonlib/sqlalchemy/alembic_func.py @@ -289,51 +289,51 @@ def create_database_migration_numbered_style( n_sequence_chars: int = 4, ) -> None: """ - Create a new Alembic migration script. + Create a new Alembic migration script. - Alembic compares the state of the database to the state of the - metadata, and generates a migration that brings the former up to the - latter. (It does not compare the most recent revision to the current - metadata, so make sure your database is up to date with the most recent - revision before running this!) + Alembic compares the state of the database to the state of the + metadata, and generates a migration that brings the former up to the + latter. (It does not compare the most recent revision to the current + metadata, so make sure your database is up to date with the most recent + revision before running this!) - You must check that the autogenerated revisions are sensible. + You must check that the autogenerated revisions are sensible. - How does it know where to look for the database? + How does it know where to look for the database? - 1. This function changes into the directory of the Alembic ``.ini`` - file and calls the external program + 1. This function changes into the directory of the Alembic ``.ini`` + file and calls the external program - .. code-block:: bash + .. code-block:: bash - alembic -c ALEMBIC_INI_FILE revision --autogenerate -m MESSAGE --rev-id REVISION_ID + alembic -c ALEMBIC_INI_FILE revision --autogenerate -m MESSAGE --rev-id REVISION_ID - 2. The Alembic ``.ini`` file points (via the ``script_location`` - variable) to a directory containing your ``env.py``. Alembic loads - this script. + 2. The Alembic ``.ini`` file points (via the ``script_location`` + variable) to a directory containing your ``env.py``. Alembic loads + this script. - 3. That script typically works out the database URL and calls further - into the Alembic code. + 3. That script typically works out the database URL and calls further + into the Alembic code. - See https://alembic.zzzcomputing.com/en/latest/autogenerate.html. + See https://alembic.zzzcomputing.com/en/latest/autogenerate.html. - Regarding filenames: the default ``n_sequence_chars`` of 4 is like Django - and gives files with names like + Regarding filenames: the default ``n_sequence_chars`` of 4 is like Django + and gives files with names like - .. code-block:: none + .. code-block:: none - 0001_x.py, 0002_y.py, ... + 0001_x.py, 0002_y.py, ... - NOTE THAT TO USE A NON-STANDARD ALEMBIC VERSION TABLE, YOU MUST SPECIFY - THAT IN YOUR ``env.py`` (see e.g. CamCOPS). + NOTE THAT TO USE A NON-STANDARD ALEMBIC VERSION TABLE, YOU MUST SPECIFY + THAT IN YOUR ``env.py`` (see e.g. CamCOPS). - Args: - alembic_ini_file: filename of Alembic ``alembic.ini`` file - alembic_versions_dir: directory in which you keep your Python scripts, - one per Alembic revision - message: message to be associated with this revision - n_sequence_chars: number of numerical sequence characters to use in the - filename/revision (see above). + Args: + alembic_ini_file: filename of Alembic ``alembic.ini`` file + alembic_versions_dir: directory in which you keep your Python scripts, + one per Alembic revision + message: message to be associated with this revision + n_sequence_chars: number of numerical sequence characters to use in the + filename/revision (see above). """ # noqa file_regex = r"\d{" + str(n_sequence_chars) + r"}_\S\.py$" diff --git a/cardinal_pythonlib/sqlalchemy/logs.py b/cardinal_pythonlib/sqlalchemy/logs.py index 67496a9..4a41f30 100644 --- a/cardinal_pythonlib/sqlalchemy/logs.py +++ b/cardinal_pythonlib/sqlalchemy/logs.py @@ -44,9 +44,9 @@ def pre_disable_sqlalchemy_extra_echo_log() -> None: sqlalchemy.engine.base.Engine.__init__() sqlalchemy.log.instance_logger() sqlalchemy.log.InstanceLogger.__init__() - # ... which checks that the logger has no handlers and if not calls: + # ... which checks that the logger has no handlers and if not calls: sqlalchemy.log._add_default_handler() - # ... which adds a handler to sys.stdout + # ... which adds a handler to sys.stdout """ log = logging.getLogger("sqlalchemy.engine.base.Engine") log.addHandler(logging.NullHandler()) diff --git a/cardinal_pythonlib/sqlalchemy/orm_inspect.py b/cardinal_pythonlib/sqlalchemy/orm_inspect.py index a422408..293e656 100644 --- a/cardinal_pythonlib/sqlalchemy/orm_inspect.py +++ b/cardinal_pythonlib/sqlalchemy/orm_inspect.py @@ -96,7 +96,7 @@ class type. This function ensures that such classes are converted to """ if isinstance(coltype, TypeEngine): return coltype - return coltype() # type: TypeEngine + return coltype() # ============================================================================= @@ -627,7 +627,7 @@ def get_pk_attrnames(obj) -> List[str]: def gen_columns_for_uninstrumented_class( - cls: Type + cls: Type, ) -> Generator[Tuple[str, Column], None, None]: """ Generate ``(attr_name, Column)`` tuples from an UNINSTRUMENTED class, i.e. @@ -675,7 +675,7 @@ def colname_to_attrname_dict(cls) -> Dict[str, str]: def gen_relationships( - obj + obj, ) -> Generator[Tuple[str, RelationshipProperty, Type], None, None]: """ Yields tuples of ``(attrname, RelationshipProperty, related_class)`` diff --git a/cardinal_pythonlib/sqlalchemy/schema.py b/cardinal_pythonlib/sqlalchemy/schema.py index 5dc3e07..5f9b9b7 100644 --- a/cardinal_pythonlib/sqlalchemy/schema.py +++ b/cardinal_pythonlib/sqlalchemy/schema.py @@ -120,8 +120,8 @@ class SqlaColumnInspectionInfo(object): """ Class to represent information from inspecting a database column. - A clearer way of getting information than the plain ``dict`` that SQLAlchemy - uses. + A clearer way of getting information than the plain ``dict`` that + SQLAlchemy uses. """ def __init__(self, sqla_info_dict: Dict[str, Any]) -> None: @@ -333,9 +333,8 @@ def mssql_get_pk_index_name( AND s.name = :schemaname """ ).bindparams(tablename=tablename, schemaname=schemaname) - with contextlib.closing( - engine.execute(query) - ) as result: # type: ResultProxy + with contextlib.closing(engine.execute(query)) as result: + result: ResultProxy row = result.fetchone() return row[0] if row else "" @@ -849,7 +848,7 @@ def get_sqla_coltype_from_dialect_str( except TypeError: return cls() - except: + except Exception: # noinspection PyUnresolvedReferences raise ValueError( f"Failed to convert SQL type {coltype!r} in dialect " diff --git a/cardinal_pythonlib/tools/explore_clang_format_config.py b/cardinal_pythonlib/tools/explore_clang_format_config.py index 5fdf657..92c7f28 100644 --- a/cardinal_pythonlib/tools/explore_clang_format_config.py +++ b/cardinal_pythonlib/tools/explore_clang_format_config.py @@ -81,7 +81,7 @@ def clang_format( subprocess.check_call( [ clang_format_exe, - f"-style=file", # read ./.clang-format; see above + "-style=file", # read ./.clang-format; see above "-i", # edit in place dest, ] @@ -144,7 +144,7 @@ def explore_clang_format( log.info(f"Source = {source_filenames}") _ = monitor_diff([working_src, current_cpp, last_cpp], meld_exe) watcher = FileWatcher([config_filename] + source_filenames) - log.info(f"Watching config/source files; press Ctrl-C to abort") + log.info("Watching config/source files; press Ctrl-C to abort") while True: time.sleep(sleep_time_s) changed_filenames = watcher.changed() @@ -186,24 +186,30 @@ def main() -> None: ) parser.add_argument( "config", - help=f"YAML clang-format config file. Create one with e.g.: " - f"clang-format --style=WebKit --dump-config > myconfig.yaml", + help=( + "YAML clang-format config file. Create one with e.g.: " + "clang-format --style=WebKit --dump-config > myconfig.yaml" + ), ) parser.add_argument( "source", nargs="+", - help=f"C++ source (.cpp, .h) file(s) to play with (if you specify " - f"several, they will be concatenated)", + help=( + "C++ source (.cpp, .h) file(s) to play with (if you specify " + "several, they will be concatenated)" + ), ) parser.add_argument( "--clangformat", - help=f"clang-format executable (see https://apt.llvm.org/ and " - f"consider: 'sudo apt install clang-format-14')", + help=( + "clang-format executable (see https://apt.llvm.org/ and " + "consider: 'sudo apt install clang-format-14')" + ), default=shutil.which("clang-format"), ) parser.add_argument( "--meld", - help=f"meld executable (consider: 'sudo apt install meld')", + help="meld executable (consider: 'sudo apt install meld')", default=shutil.which("meld"), ) args = parser.parse_args() diff --git a/cardinal_pythonlib/ui.py b/cardinal_pythonlib/ui.py index 0059e0b..91f9bcd 100644 --- a/cardinal_pythonlib/ui.py +++ b/cardinal_pythonlib/ui.py @@ -30,7 +30,10 @@ # For back compatibility (these functions used to be in this file): # noinspection PyUnresolvedReferences -from cardinal_pythonlib.ui_commandline import ask_user, ask_user_password +from cardinal_pythonlib.ui_commandline import ( # noqa: F401 + ask_user, + ask_user_password, +) try: import tkinter diff --git a/cardinal_pythonlib/wsgi/request_logging_mw.py b/cardinal_pythonlib/wsgi/request_logging_mw.py index 101a4e2..8854ad5 100644 --- a/cardinal_pythonlib/wsgi/request_logging_mw.py +++ b/cardinal_pythonlib/wsgi/request_logging_mw.py @@ -114,14 +114,16 @@ def __call__( ) except KeyError: forwarded_for = "" - request_details = '{remote}{fwd}: "{method} {path}{qmark}{query} {proto}"'.format( - remote=environ.get(WsgiEnvVar.REMOTE_ADDR, ""), - fwd=forwarded_for, - method=environ.get(WsgiEnvVar.REQUEST_METHOD, ""), - path=environ.get(WsgiEnvVar.PATH_INFO, ""), - qmark="?" if query_string else "", - query=query_string, - proto=environ.get(WsgiEnvVar.SERVER_PROTOCOL, ""), + request_details = ( + '{remote}{fwd}: "{method} {path}{qmark}{query} {proto}"'.format( + remote=environ.get(WsgiEnvVar.REMOTE_ADDR, ""), + fwd=forwarded_for, + method=environ.get(WsgiEnvVar.REQUEST_METHOD, ""), + path=environ.get(WsgiEnvVar.PATH_INFO, ""), + qmark="?" if query_string else "", + query=query_string, + proto=environ.get(WsgiEnvVar.SERVER_PROTOCOL, ""), + ) ) msg_parts = [] # type: List[str] if self.show_request_immediately: diff --git a/docs/create_all_autodocs.py b/docs/create_all_autodocs.py index afd3d52..1bfca4b 100755 --- a/docs/create_all_autodocs.py +++ b/docs/create_all_autodocs.py @@ -75,8 +75,11 @@ def make_autodoc(make: bool, destroy_first: bool) -> None: log.info("Deleting directory {!r}", AUTODOC_DIR) rmtree(AUTODOC_DIR) else: - log.warning("Would delete directory {!r} (not doing so as in mock " - "mode)", AUTODOC_DIR) + log.warning( + "Would delete directory {!r} (not doing so as in mock " + "mode)", + AUTODOC_DIR, + ) top_idx = AutodocIndex( autodoc_rst_root_dir=AUTODOC_DIR, highest_code_dir=CODE_ROOT_DIR, @@ -100,22 +103,24 @@ def make_autodoc(make: bool, destroy_first: bool) -> None: def main() -> None: parser = argparse.ArgumentParser() parser.add_argument( - "--make", action="store_true", - help="Do things! Otherwise will just show its intent.") - parser.add_argument( - "--destroy_first", action="store_true", - help="Destroy all existing autodocs first") + "--make", + action="store_true", + help="Do things! Otherwise will just show its intent.", + ) parser.add_argument( - "--verbose", action="store_true", - help="Be verbose") + "--destroy_first", + action="store_true", + help="Destroy all existing autodocs first", + ) + parser.add_argument("--verbose", action="store_true", help="Be verbose") args = parser.parse_args() main_only_quicksetup_rootlogger( - level=logging.DEBUG if args.verbose else logging.INFO) + level=logging.DEBUG if args.verbose else logging.INFO + ) - make_autodoc(make=args.make, - destroy_first=args.destroy_first) + make_autodoc(make=args.make, destroy_first=args.destroy_first) -if __name__ == '__main__': +if __name__ == "__main__": main() diff --git a/docs/rebuild_docs.py b/docs/rebuild_docs.py index 5073420..4b37cbb 100755 --- a/docs/rebuild_docs.py +++ b/docs/rebuild_docs.py @@ -28,6 +28,7 @@ import shutil import subprocess import sys + if sys.version_info[0] < 3: raise AssertionError("Need Python 3") @@ -37,7 +38,7 @@ DEST_DIRS = [] -if __name__ == '__main__': +if __name__ == "__main__": # Remove anything old for destdir in [BUILD_HTML_DIR] + DEST_DIRS: print(f"Deleting directory {destdir!r}") diff --git a/docs/source/conf.py b/docs/source/conf.py index 4813d02..3a34f55 100644 --- a/docs/source/conf.py +++ b/docs/source/conf.py @@ -18,6 +18,7 @@ try: # noinspection PyUnresolvedReferences import django + # noinspection PyUnresolvedReferences from django.conf import settings except ImportError: @@ -29,10 +30,10 @@ # -- Project information ----------------------------------------------------- -project = 'cardinal_pythonlib' +project = "cardinal_pythonlib" # noinspection PyShadowingBuiltins -copyright = '2009-2020, Rudolf Cardinal' -author = 'Rudolf Cardinal' +copyright = "2009-2020, Rudolf Cardinal" +author = "Rudolf Cardinal" # The short X.Y version version = VERSION_STRING @@ -50,26 +51,26 @@ # extensions coming with Sphinx (named 'sphinx.ext.') or your custom # ones. extensions = [ - 'sphinx.ext.autodoc', - 'sphinx.ext.doctest', - 'sphinx.ext.todo', - 'sphinx.ext.imgmath', - 'sphinx.ext.viewcode', - 'sphinx.ext.napoleon', - 'sphinx_paramlinks', + "sphinx.ext.autodoc", + "sphinx.ext.doctest", + "sphinx.ext.todo", + "sphinx.ext.imgmath", + "sphinx.ext.viewcode", + "sphinx.ext.napoleon", + "sphinx_paramlinks", ] # Add any paths that contain templates here, relative to this directory. -templates_path = ['_templates'] +templates_path = ["_templates"] # The suffix(es) of source filenames. # You can specify multiple suffix as a list of string: # # source_suffix = ['.rst', '.md'] -source_suffix = '.rst' +source_suffix = ".rst" # The master toctree document. -master_doc = 'index' +master_doc = "index" # The language for content autogenerated by Sphinx. Refer to documentation # for a list of supported languages. @@ -84,7 +85,7 @@ exclude_patterns = [] # The name of the Pygments (syntax highlighting) style to use. -pygments_style = 'sphinx' +pygments_style = "sphinx" # -- Options for HTML output ------------------------------------------------- @@ -93,7 +94,7 @@ # a list of builtin themes. # html_theme = 'classic' -html_theme = 'sphinx_rtd_theme' # pip install sphinx_rtd_theme +html_theme = "sphinx_rtd_theme" # pip install sphinx_rtd_theme # Theme options are theme-specific and customize the look and feel of a theme # further. For a list of options available for each theme, see the @@ -120,7 +121,7 @@ # -- Options for HTMLHelp output --------------------------------------------- # Output file base name for HTML help builder. -htmlhelp_basename = 'cardinal_pythonlibdoc' +htmlhelp_basename = "cardinal_pythonlibdoc" # -- Options for LaTeX output ------------------------------------------------ @@ -129,15 +130,12 @@ # The paper size ('letterpaper' or 'a4paper'). # # 'papersize': 'letterpaper', - # The font size ('10pt', '11pt' or '12pt'). # # 'pointsize': '10pt', - # Additional stuff for the LaTeX preamble. # # 'preamble': '', - # Latex figure (float) alignment # # 'figure_align': 'htbp', @@ -147,8 +145,13 @@ # (source start file, target name, title, # author, documentclass [howto, manual, or own class]). latex_documents = [ - (master_doc, 'cardinal_pythonlib.tex', 'cardinal\\_pythonlib Documentation', - 'Rudolf Cardinal', 'manual'), + ( + master_doc, + "cardinal_pythonlib.tex", + "cardinal\\_pythonlib Documentation", + "Rudolf Cardinal", + "manual", + ), ] @@ -157,8 +160,13 @@ # One entry per manual page. List of tuples # (source start file, name, description, authors, manual section). man_pages = [ - (master_doc, 'cardinal_pythonlib', 'cardinal_pythonlib Documentation', - [author], 1) + ( + master_doc, + "cardinal_pythonlib", + "cardinal_pythonlib Documentation", + [author], + 1, + ) ] @@ -168,9 +176,15 @@ # (source start file, target name, title, author, # dir menu entry, description, category) texinfo_documents = [ - (master_doc, 'cardinal_pythonlib', 'cardinal_pythonlib Documentation', - author, 'cardinal_pythonlib', 'One line description of project.', - 'Miscellaneous'), + ( + master_doc, + "cardinal_pythonlib", + "cardinal_pythonlib Documentation", + author, + "cardinal_pythonlib", + "One line description of project.", + "Miscellaneous", + ), ] @@ -184,7 +198,7 @@ # -- Options for autodoc extension ------------------------------------------- -autoclass_content = 'both' +autoclass_content = "both" # https://stackoverflow.com/questions/5599254/how-to-use-sphinxs-autodoc-to-document-a-classs-init-self-method # noqa diff --git a/setup.cfg b/setup.cfg index a40f10d..7985ef3 100644 --- a/setup.cfg +++ b/setup.cfg @@ -7,3 +7,18 @@ filterwarnings = testpaths = cardinal_pythonlib + +[flake8] +linters=pycodestyle,pyflakes +max-line-length=79 +# Not compatible with Black and not PEP8 apparently +# E203: Whitespace before ':' +extend-ignore = E203 + +[mypy] +# MyPy is a static type checker. It will not execute the code! +# pip install mypy +no_strict_optional = True +allow_redefinition = True +disallow_untyped_defs = True +disallow_incomplete_defs = True	Code style Use Black, as per https://github.com/RudolfCardinal/crate		2023-01-24T18:59:16	[]	[]
soda-inria/sklearn-numba-dpex	soda-inria__sklearn-numba-dpex-67	3d306c12ccd7fc55b7f0305dca61ec6acb8d9918	diff --git a/.github/workflows/run_code_format_checks.yaml b/.github/workflows/run_code_format_checks.yaml index b2c67d7..5c5eada 100644 --- a/.github/workflows/run_code_format_checks.yaml +++ b/.github/workflows/run_code_format_checks.yaml @@ -18,4 +18,4 @@ jobs: pre-commit install - name: Run checks - run: pre-commit run + run: pre-commit run --all-files diff --git a/README.md b/README.md index 4a496d8..59fc676 100644 --- a/README.md +++ b/README.md @@ -9,13 +9,13 @@ workable state at the moment. This requires working with the following branch of scikit-learn which is itself not yet in a working state at the moment: -- `wip-engines` branch on https://github.com/ogrisel/scikit-learn +- `wip-engines` branch on https://github.com/ogrisel/scikit-learn ## List of Included Engines - `sklearn.cluster.KMeans` for the standard LLoyd's algorithm on dense data arrays. The k-means++ initialization scheme is not implemented yet but this is planned. - + No implementation of k-means for sparse input data or for the "Elkan" algorithm is included at the moment. @@ -201,7 +201,7 @@ To run the `scikit-learn` tests with the `sklearn_numba_dpex` engine you can run pytest --sklearn-engine-provider sklearn_numba_dpex --pyargs sklearn.cluster.tests.test_k_means ``` -(change the `--pyargs` option accordingly to select other test suites). +(change the `--pyargs` option accordingly to select other test suites). The `--sklearn-engine-provider sklearn_numba_dpex` option offered by the sklearn pytest plugin will automatically activate the `sklearn_numba_dpex` engine for all tests. diff --git a/benchmark/kmeans.py b/benchmark/kmeans.py index 10c964f..957493a 100644 --- a/benchmark/kmeans.py +++ b/benchmark/kmeans.py @@ -131,7 +131,8 @@ def _check_same_fit(self, estimator, name, max_iter, assert_allclose): if n_iter != max_iter: raise RuntimeError( f"The estimator {name} only ran {n_iter} iterations instead of " - f"max_iter={max_iter} iterations. " + runtime_error_message + f"max_iter={max_iter} iterations. " + + runtime_error_message ) if not assert_allclose: diff --git a/sklearn_numba_dpex/common/kernels.py b/sklearn_numba_dpex/common/kernels.py index 0b801dd..a491585 100644 --- a/sklearn_numba_dpex/common/kernels.py +++ b/sklearn_numba_dpex/common/kernels.py @@ -20,13 +20,12 @@ import math from functools import lru_cache -import numpy as np import dpctl.tensor as dpt import numba_dpex as dpex +import numpy as np from sklearn_numba_dpex.common._utils import check_power_of_2 - zero_idx = np.int64(0) @@ -114,7 +113,7 @@ def half_l2_norm( data, # IN (size0, size1) result, # OUT (size1,) ): - # fmt: on + # fmt: on col_idx = dpex.get_global_id(zero_idx) if col_idx >= size1: @@ -265,7 +264,7 @@ def partial_sum_reduction( summands, # IN (n_rows, sum_axis_size) result, # OUT (n_rows, math.ceil(size / (2 * work_group_size),) ): - # fmt: on + # fmt: on # NB: This kernel only perform a partial reduction group_id = dpex.get_group_id(zero_idx) local_work_id = dpex.get_local_id(zero_idx) @@ -294,7 +293,11 @@ def partial_sum_reduction( # We discard half of the remaining active work items at each iteration current_n_work_items = current_n_work_items // two_as_a_long if local_work_id < current_n_work_items: - add_cols_inplace(local_values, local_work_id + current_n_work_items, local_work_id) + add_cols_inplace( + local_values, + local_work_id + current_n_work_items, + local_work_id, + ) dpex.barrier(dpex.CLK_LOCAL_MEM_FENCE) @@ -320,8 +323,8 @@ def partial_sum_reduction( # NB: here memory for partial results is allocated ahead of time and will only # be garbage collected when the instance of `sum_reduction` is garbage # collected. Thus it can be more efficient to re-use a same instance of - # `sum_reduction` (e.g within iterations of a loop) since it avoid deallocation - # and reallocation every time. + # `sum_reduction` (e.g within iterations of a loop) since it avoid + # deallocation and reallocation every time. result = dpt.empty(result_shape, dtype=dtype, device=device) kernels_and_empty_tensors_pairs.append((kernel, result)) @@ -337,7 +340,8 @@ def sum_reduction(summands): @lru_cache def make_argmin_reduction_1d_kernel(size, work_group_size, device, dtype): - """Implement 1d argmin with the same strategy than for make_sum_reduction_2d_axis1_kernel.""" + """Implement 1d argmin with the same strategy than for + make_sum_reduction_2d_axis1_kernel.""" check_power_of_2(work_group_size) # Number of iteration in each execution of the kernel: @@ -361,14 +365,16 @@ def partial_argmin_reduction( # / (2 * work_group_size),) # )) ): - # fmt: on + # fmt: on group_id = dpex.get_group_id(zero_idx) local_work_id = dpex.get_local_id(zero_idx) first_work_id = local_work_id == zero_idx previous_result_size = previous_result.shape[zero_idx] has_previous_result = previous_result_size > one_idx - current_size = previous_result_size if has_previous_result else values.shape[zero_idx] + current_size = ( + previous_result_size if has_previous_result else values.shape[zero_idx] + ) local_argmin = dpex.local.array(work_group_size, dtype=np.int32) local_values = dpex.local.array(work_group_size, dtype=dtype) @@ -395,7 +401,7 @@ def partial_argmin_reduction( y = values[y_idx] if x < y or (x == y and x_idx < y_idx): local_argmin[local_work_id] = x_idx - local_values[local_work_id] = x + local_values[local_work_id] = x else: local_argmin[local_work_id] = y_idx local_values[local_work_id] = y @@ -408,10 +414,10 @@ def partial_argmin_reduction( local_x_idx = local_work_id local_y_idx = local_work_id + current_n_work_items - x= local_values[local_x_idx] - y= local_values[local_y_idx] + x = local_values[local_x_idx] + y = local_values[local_y_idx] - if x> y: + if x > y: local_values[local_x_idx] = y local_argmin[local_x_idx] = local_argmin[local_y_idx] diff --git a/sklearn_numba_dpex/common/random.py b/sklearn_numba_dpex/common/random.py index 36c63a2..6379568 100644 --- a/sklearn_numba_dpex/common/random.py +++ b/sklearn_numba_dpex/common/random.py @@ -1,20 +1,19 @@ -import warnings import random +import warnings from functools import lru_cache -from numba import float32, float64, uint32, int64, uint64 -import numpy as np - import dpctl import dpctl.tensor as dpt import numba_dpex as dpex +import numpy as np +from numba import float32, float64, int64, uint32, uint64 # This code is largely inspired from the numba.cuda.random module and the # numba/cuda/random.py where it's defined (v<0.57), and by the implementation of the # same algorithm in the package `randomgen`. # numba.cuda.random: https://github.com/numba/numba/blob/0.56.3/numba/cuda/random.py -# randomgen: https://github.com/bashtage/randomgen/blob/v1.23.1/randomgen/xoroshiro128.pyx +# randomgen: https://github.com/bashtage/randomgen/blob/v1.23.1/randomgen/xoroshiro128.pyx # noqa # NB1: we implement xoroshiro128++ rather than just xoroshiro128+, which is preferred. # Reference resource about PRNG: https://prng.di.unimi.it/ @@ -284,7 +283,7 @@ def _xoroshiro128pp_jump(states, state_idx): # result = s0 + s1 # s1 ^= s0 - # states[state_idx, zero_idx] = _rotl(s0, next_rot_p_1) ^ s1 ^ (s1 << next_rot_p_2) + # states[state_idx, zero_idx] = _rotl(s0, next_rot_p_1) ^ s1 ^ (s1 << next_rot_p_2) # noqa # states[state_idx, one_idx] = _rotl(s1, next_rot_p_3) # return result @@ -300,8 +299,7 @@ def _xoroshiro128pp_jump(states, state_idx): from_cpu_to_device = True except dpctl.SyclDeviceCreationError: warnings.warn( - "No CPU found, falling back random initiatlization to default " - "device." + "No CPU found, falling back random initiatlization to default device." ) states = dpt.empty( diff --git a/sklearn_numba_dpex/kmeans/drivers.py b/sklearn_numba_dpex/kmeans/drivers.py index 74434fb..c794d90 100644 --- a/sklearn_numba_dpex/kmeans/drivers.py +++ b/sklearn_numba_dpex/kmeans/drivers.py @@ -1,37 +1,35 @@ -import warnings import math +import warnings -import numpy as np import dpctl import dpctl.tensor as dpt import dpnp - -from sklearn_numba_dpex.common.random import ( - get_random_raw, - create_xoroshiro128pp_states, -) +import numpy as np from sklearn_numba_dpex.common.kernels import ( - make_initialize_to_zeros_2d_kernel, - make_initialize_to_zeros_3d_kernel, + make_argmin_reduction_1d_kernel, make_broadcast_division_1d_2d_kernel, make_half_l2_norm_2d_axis0_kernel, + make_initialize_to_zeros_2d_kernel, + make_initialize_to_zeros_3d_kernel, make_sum_reduction_2d_axis1_kernel, - make_argmin_reduction_1d_kernel, ) - +from sklearn_numba_dpex.common.random import ( + create_xoroshiro128pp_states, + get_random_raw, +) from sklearn_numba_dpex.kmeans.kernels import ( - make_lloyd_single_step_fixed_window_kernel, + make_centroid_shifts_kernel, make_compute_euclidean_distances_fixed_window_kernel, - make_label_assignment_fixed_window_kernel, make_compute_inertia_kernel, make_kmeansplusplus_init_kernel, - make_sample_center_candidates_kernel, make_kmeansplusplus_single_step_fixed_window_kernel, + make_label_assignment_fixed_window_kernel, + make_lloyd_single_step_fixed_window_kernel, + make_reduce_centroid_data_kernel, make_relocate_empty_clusters_kernel, + make_sample_center_candidates_kernel, make_select_samples_far_from_centroid_kernel, - make_centroid_shifts_kernel, - make_reduce_centroid_data_kernel, ) @@ -67,7 +65,7 @@ def lloyd( return_assignments=bool(verbose), sub_group_size=sub_group_size, global_mem_cache_size=global_mem_cache_size, - centroids_private_copies_max_cache_occupancy=centroids_private_copies_max_cache_occupancy, + centroids_private_copies_max_cache_occupancy=centroids_private_copies_max_cache_occupancy, # noqa work_group_size=max_work_group_size, dtype=compute_dtype, ) diff --git a/sklearn_numba_dpex/kmeans/engine.py b/sklearn_numba_dpex/kmeans/engine.py index 41338e5..ed82ff2 100644 --- a/sklearn_numba_dpex/kmeans/engine.py +++ b/sklearn_numba_dpex/kmeans/engine.py @@ -1,14 +1,13 @@ import warnings +from typing import Any, Dict -import numpy as np import dpctl import dpctl.tensor as dpt - +import numpy as np from sklearn.cluster._kmeans import KMeansCythonEngine +from sklearn.exceptions import DataConversionWarning, NotSupportedByEngineError -from sklearn.exceptions import NotSupportedByEngineError, DataConversionWarning - -from .drivers import lloyd, get_labels_inertia, get_euclidean_distances, kmeans_plusplus +from .drivers import get_euclidean_distances, get_labels_inertia, kmeans_plusplus, lloyd class _IgnoreSampleWeight: @@ -47,7 +46,7 @@ class KMeansEngine(KMeansCythonEngine): """ - _CONFIG = dict() + _CONFIG: Dict[str, Any] = dict() def __init__(self, estimator): self.device = dpctl.SyclDevice(self._CONFIG.get("device")) @@ -75,13 +74,15 @@ def prepare_fit(self, X, y=None, sample_weight=None): ) except Exception as e: raise NotSupportedByEngineError( - "The sklearn_nunmba_dpex engine for KMeans does not support the format of the inputed data." + "The sklearn_nunmba_dpex engine for KMeans does not support the format" + " of the inputed data." ) from e algorithm = estimator.algorithm if algorithm not in ("lloyd", "auto", "full"): raise NotSupportedByEngineError( - f"The sklearn_nunmba_dpex engine for KMeans only support the Lloyd algorithm, {algorithm} is not supported." + "The sklearn_nunmba_dpex engine for KMeans only support the Lloyd" + f" algorithm, {algorithm} is not supported." ) self.sample_weight = sample_weight @@ -226,15 +227,15 @@ def _set_dtype(self, X, sample_weight, cluster_centers): if (compute_dtype != np.float32) and (compute_dtype != np.float64): text = ( f"KMeans has been set to compute with type {compute_dtype} but only " - f"the types float32 and float64 are supported. The computations and " - f"outputs will default back to float32 type." + "the types float32 and float64 are supported. The computations and " + "outputs will default back to float32 type." ) output_dtype = compute_dtype = np.float32 elif (compute_dtype == np.float64) and not self.device.has_aspect_fp64: text = ( f"KMeans is set to compute with type {compute_dtype} but this type is " f"not supported by the device {self.device.name}. The computations " - f"will default back to float32 type." + "will default back to float32 type." ) compute_dtype = np.float32 @@ -244,8 +245,8 @@ def _set_dtype(self, X, sample_weight, cluster_centers): if copy: text += ( f" A copy of the data casted to type {compute_dtype} will be created. " - f"To save memory and suppress this warning, ensure that the dtype of " - f"the input data matches the dtype required for computations." + "To save memory and suppress this warning, ensure that the dtype of " + "the input data matches the dtype required for computations." ) warnings.warn(text, DataConversionWarning) # TODO: instead of triggering a copy on the host side, we could use the @@ -262,7 +263,7 @@ def _set_dtype(self, X, sample_weight, cluster_centers): f"The centers have been passed with type {cluster_centers_dtype} but " f"type {compute_dtype} is expected. A copy will be created with the " f"correct type {compute_dtype}. Ensure that the centers are passed " - f"with the correct dtype to save memory and suppress this warning.", + "with the correct dtype to save memory and suppress this warning.", DataConversionWarning, ) cluster_centers = cluster_centers.astype(compute_dtype) @@ -274,7 +275,7 @@ def _set_dtype(self, X, sample_weight, cluster_centers): f"sample_weight has been passed with type {sample_weight.dtype} but " f"type {compute_dtype} is expected. A copy will be created with the " f"correct type {compute_dtype}. Ensure that sample_weight is passed " - f"with the correct dtype to save memory and suppress this warning.", + "with the correct dtype to save memory and suppress this warning.", DataConversionWarning, ) sample_weight = sample_weight.astype(compute_dtype) diff --git a/sklearn_numba_dpex/kmeans/kernels/__init__.py b/sklearn_numba_dpex/kmeans/kernels/__init__.py index b0444d3..c4949d2 100644 --- a/sklearn_numba_dpex/kmeans/kernels/__init__.py +++ b/sklearn_numba_dpex/kmeans/kernels/__init__.py @@ -1,22 +1,21 @@ -from .lloyd_single_step import make_lloyd_single_step_fixed_window_kernel from .compute_euclidean_distances import ( make_compute_euclidean_distances_fixed_window_kernel, ) -from .compute_labels import make_label_assignment_fixed_window_kernel from .compute_inertia import make_compute_inertia_kernel +from .compute_labels import make_label_assignment_fixed_window_kernel from .kmeans_plusplus import ( make_kmeansplusplus_init_kernel, - make_sample_center_candidates_kernel, make_kmeansplusplus_single_step_fixed_window_kernel, + make_sample_center_candidates_kernel, ) +from .lloyd_single_step import make_lloyd_single_step_fixed_window_kernel from .utils import ( - make_relocate_empty_clusters_kernel, - make_select_samples_far_from_centroid_kernel, make_centroid_shifts_kernel, make_reduce_centroid_data_kernel, + make_relocate_empty_clusters_kernel, + make_select_samples_far_from_centroid_kernel, ) - __all__ = ( "make_lloyd_single_step_fixed_window_kernel", "make_compute_euclidean_distances_fixed_window_kernel", diff --git a/sklearn_numba_dpex/kmeans/kernels/_base_kmeans_kernel_funcs.py b/sklearn_numba_dpex/kmeans/kernels/_base_kmeans_kernel_funcs.py index f990484..ce6d0bc 100644 --- a/sklearn_numba_dpex/kmeans/kernels/_base_kmeans_kernel_funcs.py +++ b/sklearn_numba_dpex/kmeans/kernels/_base_kmeans_kernel_funcs.py @@ -210,7 +210,7 @@ def _initialize_window_of_centroids( window_of_centroids_half_l2_norms, # OUT results, # OUT ): - # fmt: on + # fmt: on _initialize_results(results) # The first `window_n_centroids` work items cooperate on loading the @@ -240,18 +240,18 @@ def _load_window_of_centroids_and_features( current_centroids_t, # IN centroids_window, # OUT ): - # fmt: on + # fmt: on # The work items in the work group cooperatively load the values in shared # memory. At each iteration, the work item loads one value and adjacent work # items load adjacent values. loading_feature_idx = first_feature_idx + window_loading_feature_offset - + if (loading_feature_idx < n_features) and ( loading_centroid_idx < n_clusters ): - value = current_centroids_t[loading_feature_idx, loading_centroid_idx] + value = current_centroids_t[loading_feature_idx, loading_centroid_idx] else: - value = zero + value = zero centroids_window[ window_loading_feature_offset, window_loading_centroid_idx @@ -276,7 +276,7 @@ def _accumulate_sum_of_ops( centroids_window, # IN result, # OUT ): - # fmt: on + # fmt: on for window_feature_idx in range(window_n_features): feature_idx = window_feature_idx + first_feature_idx @@ -289,7 +289,9 @@ def _accumulate_sum_of_ops( # For this given feature, loop on all centroids in the current # window and accumulate the partial results for window_centroid_idx in range(window_n_centroids): - centroid_value = centroids_window[window_feature_idx, window_centroid_idx] + centroid_value = ( + centroids_window[window_feature_idx, window_centroid_idx] + ) if accumulate_dot_product: result[window_centroid_idx] += centroid_value * X_value else: @@ -351,7 +353,7 @@ def update_closest_centroid( window_of_centroids_half_l2_norms, # IN dot_products, # IN ): - # fmt: on + # fmt: on for i in range(window_n_centroids): current_sample_pseudo_inertia = ( window_of_centroids_half_l2_norms[i] - dot_products[i] diff --git a/sklearn_numba_dpex/kmeans/kernels/compute_euclidean_distances.py b/sklearn_numba_dpex/kmeans/kernels/compute_euclidean_distances.py index 9cb8cc5..f8980fd 100644 --- a/sklearn_numba_dpex/kmeans/kernels/compute_euclidean_distances.py +++ b/sklearn_numba_dpex/kmeans/kernels/compute_euclidean_distances.py @@ -1,8 +1,8 @@ import math from functools import lru_cache -import numpy as np import numba_dpex as dpex +import numpy as np from ._base_kmeans_kernel_funcs import make_pairwise_ops_base_kernel_funcs @@ -60,7 +60,7 @@ def compute_distances( current_centroids_t, # IN READ-ONLY (n_features, n_clusters) euclidean_distances_t, # OUT (n_clusters, n_samples) ): - # fmt: on + # fmt: on sample_idx = dpex.get_global_id(zero_idx) local_work_id = dpex.get_local_id(zero_idx) @@ -112,10 +112,12 @@ def compute_distances( for i in range(window_n_centroids): centroid_idx = first_centroid_idx + i if centroid_idx < n_clusters: - euclidean_distances_t[first_centroid_idx + i, sample_idx] = math.sqrt(sq_distances[i]) + euclidean_distances_t[first_centroid_idx + i, sample_idx] = ( + math.sqrt(sq_distances[i]) + ) first_centroid_idx += window_n_centroids - + dpex.barrier(dpex.CLK_LOCAL_MEM_FENCE) global_size = (math.ceil(n_samples / work_group_size)) * (work_group_size) diff --git a/sklearn_numba_dpex/kmeans/kernels/compute_inertia.py b/sklearn_numba_dpex/kmeans/kernels/compute_inertia.py index 52244cb..59184dc 100644 --- a/sklearn_numba_dpex/kmeans/kernels/compute_inertia.py +++ b/sklearn_numba_dpex/kmeans/kernels/compute_inertia.py @@ -1,8 +1,8 @@ import math from functools import lru_cache -import numpy as np import numba_dpex as dpex +import numpy as np @lru_cache @@ -20,7 +20,7 @@ def compute_inertia( assignments_idx, # IN READ-ONLY (n_samples,) per_sample_inertia, # OUT (n_samples,) ): - # fmt: on + # fmt: on sample_idx = dpex.get_global_id(zero_idx) diff --git a/sklearn_numba_dpex/kmeans/kernels/compute_labels.py b/sklearn_numba_dpex/kmeans/kernels/compute_labels.py index 059a2cf..f1bdba1 100644 --- a/sklearn_numba_dpex/kmeans/kernels/compute_labels.py +++ b/sklearn_numba_dpex/kmeans/kernels/compute_labels.py @@ -1,8 +1,8 @@ import math from functools import lru_cache -import numpy as np import numba_dpex as dpex +import numpy as np from ._base_kmeans_kernel_funcs import ( make_pairwise_ops_base_kernel_funcs, @@ -68,7 +68,7 @@ def assignment( centroids_half_l2_norm, # IN (n_clusters,) assignments_idx, # OUT (n_samples,) ): - # fmt: on + # fmt: on sample_idx = dpex.get_global_id(zero_idx) local_work_id = dpex.get_local_id(zero_idx) @@ -125,7 +125,7 @@ def assignment( is_last_feature_window, is_last_centroid_window ) - + first_feature_idx += centroids_window_height dpex.barrier(dpex.CLK_LOCAL_MEM_FENCE) @@ -140,7 +140,7 @@ def assignment( ) first_centroid_idx += window_n_centroids - + dpex.barrier(dpex.CLK_LOCAL_MEM_FENCE) # No update step, only store min_idx in the output array diff --git a/sklearn_numba_dpex/kmeans/kernels/kmeans_plusplus.py b/sklearn_numba_dpex/kmeans/kernels/kmeans_plusplus.py index 498858b..e922611 100644 --- a/sklearn_numba_dpex/kmeans/kernels/kmeans_plusplus.py +++ b/sklearn_numba_dpex/kmeans/kernels/kmeans_plusplus.py @@ -1,13 +1,13 @@ import math from functools import lru_cache -import numpy as np import numba_dpex as dpex - -from ._base_kmeans_kernel_funcs import make_pairwise_ops_base_kernel_funcs +import numpy as np from sklearn_numba_dpex.common.random import make_rand_uniform_kernel_func +from ._base_kmeans_kernel_funcs import make_pairwise_ops_base_kernel_funcs + # NB: refer to the definition of the main lloyd function for a more comprehensive # inline commenting of the kernel. @@ -32,7 +32,7 @@ def kmeansplusplus_init( center_indices, # OUT (n_clusters,) closest_dist_sq, # OUT (n_samples,) ): - # fmt: on + # fmt: on sample_idx = dpex.get_global_id(zero_idx) if sample_idx >= n_samples: return @@ -81,7 +81,7 @@ def sample_center_candidates( random_state, # INOUT (n_local_trials, 2) candidates_id, # OUT (n_local_trials,) ): - # fmt: on + # fmt: on local_trial_idx = dpex.get_global_id(zero_idx) if local_trial_idx >= n_local_trials: return @@ -90,12 +90,15 @@ def sample_center_candidates( cumulative_potential = zero_init candidate_id = one_decr - while (random_value > cumulative_potential) and (candidate_id < max_candidate_id): + while ( + (random_value > cumulative_potential) and + (candidate_id < max_candidate_id) + ): candidate_id += one_incr cumulative_potential += closest_dist_sq[candidate_id] candidates_id[local_trial_idx] = candidate_id - global_size = (math.ceil(n_local_trials / work_group_size)) * (work_group_size) + global_size = (math.ceil(n_local_trials / work_group_size)) * work_group_size return sample_center_candidates[global_size, work_group_size] @@ -151,7 +154,7 @@ def kmeansplusplus_single_step( closest_dist_sq, # IN (n_samples,) sq_distances_t, # OUT (n_candidates, n_samples) ): - # fmt: on + # fmt: on sample_idx = dpex.get_global_id(zero_idx) local_work_id = dpex.get_local_id(zero_idx) @@ -211,8 +214,11 @@ def kmeansplusplus_single_step( if candidate_idx < n_candidates: sq_distance_i = min( sq_distances[i] * sample_weight_, - closest_dist_sq_) - sq_distances_t[first_candidate_idx + i, sample_idx] = sq_distance_i + closest_dist_sq_ + ) + sq_distances_t[first_candidate_idx + i, sample_idx] = ( + sq_distance_i + ) first_candidate_idx += window_n_candidates diff --git a/sklearn_numba_dpex/kmeans/kernels/lloyd_single_step.py b/sklearn_numba_dpex/kmeans/kernels/lloyd_single_step.py index 55dd295..17caf70 100644 --- a/sklearn_numba_dpex/kmeans/kernels/lloyd_single_step.py +++ b/sklearn_numba_dpex/kmeans/kernels/lloyd_single_step.py @@ -1,8 +1,8 @@ import math from functools import lru_cache -import numpy as np import numba_dpex as dpex +import numpy as np from ._base_kmeans_kernel_funcs import ( make_pairwise_ops_base_kernel_funcs, @@ -129,7 +129,7 @@ def fused_lloyd_single_step( new_centroids_t_private_copies, # OUT (n_private_copies, n_features, n_clusters) # noqa cluster_sizes_private_copies, # OUT (n_private_copies, n_clusters) # noqa ): - # fmt: on + # fmt: on """One full iteration of LLoyd's k-means. The kernel is meant to be invoked on a 1D grid spanning the data samples of @@ -252,7 +252,6 @@ def fused_lloyd_single_step( # operations in the current iteration to be over for all work items. dpex.barrier(dpex.CLK_LOCAL_MEM_FENCE) - # End of inner loop. The pseudo inertia is now computed for all centroids # in the window, we can coalesce it to the accumulation of the min pseudo # inertia for the current sample. @@ -267,13 +266,11 @@ def fused_lloyd_single_step( first_centroid_idx += window_n_centroids - # When the next iteration starts work items will overwrite shared memory # with new values, so before that we must wait for all reading # operations in the current iteration to be over for all work items. dpex.barrier(dpex.CLK_LOCAL_MEM_FENCE) - # End of outer loop. By now min_idx and min_sample_pseudo_inertia # contains the expected values. @@ -300,7 +297,7 @@ def fused_lloyd_single_step( # which we mitigate with a strategy of "privatization" for reducing the # probability of collisions. The array of centroids is duplicated in global # memory as many time as possible and each sub-group of work items of size - # `sub_group_size` is assigned to a different duplicata and update the values + # `sub_group_size` is assigned to a different duplicata and update the values # of this single duplicata. # The resulting copies of centroids updates will then need to be reduced to a diff --git a/sklearn_numba_dpex/kmeans/kernels/utils.py b/sklearn_numba_dpex/kmeans/kernels/utils.py index ea33a12..35d4ba9 100644 --- a/sklearn_numba_dpex/kmeans/kernels/utils.py +++ b/sklearn_numba_dpex/kmeans/kernels/utils.py @@ -1,9 +1,8 @@ import math from functools import lru_cache -import numpy as np import numba_dpex as dpex - +import numpy as np zero_idx = np.int64(0) @@ -30,8 +29,8 @@ def relocate_empty_clusters( per_sample_inertia, # INOUT (n_samples,) centroids_t, # INOUT (n_features, n_clusters) cluster_sizes # INOUT (n_clusters,) - ): - # fmt: on + ): + # fmt: on """NB: because of how the array was created, samples_far_from_center values are located between (n_selected_gt_threshold - relocated_idx) and (relocated_idx) indices. @@ -39,22 +38,25 @@ def relocate_empty_clusters( group_idx = dpex.get_group_id(zero_idx) item_idx = dpex.get_local_id(zero_idx) relocated_idx = group_idx // n_work_groups_for_cluster - feature_idx = ((group_idx % n_work_groups_for_cluster) * work_group_size) + item_idx + feature_idx = ( + ((group_idx % n_work_groups_for_cluster) * work_group_size) + item_idx + ) if feature_idx >= n_features: return relocated_cluster_idx = empty_clusters_list[relocated_idx] - new_location_X_idx = samples_far_from_center[n_selected_gt_threshold - relocated_idx] + new_location_X_idx = ( + samples_far_from_center[n_selected_gt_threshold - relocated_idx] + ) new_location_previous_assignment = assignments_idx[new_location_X_idx] new_centroid_value = X_t[feature_idx, new_location_X_idx] new_location_weight = sample_weight[new_location_X_idx] X_centroid_addend = new_centroid_value * new_location_weight - - # Cancel the contribution to new_location_previous_assignment, the previously - # updated centroids of the sample. This contribution will be assigned to the + # Cancel the contribution to new_location_previous_assignment, the previously + # updated centroids of the sample. This contribution will be assigned to the # centroid of the clusters that relocates to this sample. dpex.atomic.sub( centroids_t, @@ -66,7 +68,7 @@ def relocate_empty_clusters( # sample the cluster has been relocated to. centroids_t[feature_idx, relocated_cluster_idx] = X_centroid_addend - # Likewise, we update the weights in the clusters. This is done once using + # Likewise, we update the weights in the clusters. This is done once using # `feature_idx`'s value. if feature_idx == zero_idx: per_sample_inertia[new_location_X_idx] = zero @@ -98,8 +100,8 @@ def select_samples_far_from_centroid( selected_samples_idx, # OUT (n_samples,) n_selected_gt_threshold, # OUT (1,) n_selected_eq_threshold, # OUT (1,) - ): - # fmt: on + ): + # fmt: on """This kernel writes in selected_samples_idx the idx of the top n_selected items in distance_to_centroid with highest values. @@ -122,8 +124,10 @@ def select_samples_far_from_centroid( n_selected_gt_threshold_ = n_selected_gt_threshold[zero_idx] n_selected_eq_threshold_ = n_selected_eq_threshold[zero_idx] - if ((n_selected_gt_threshold_ == max_n_selected_gt_threshold) - and (n_selected_eq_threshold_ >= min_n_selected_eq_threshold)): + if ( + (n_selected_gt_threshold_ == max_n_selected_gt_threshold) and + (n_selected_eq_threshold_ >= min_n_selected_eq_threshold) + ): return distance_to_centroid_ = distance_to_centroid[sample_idx] @@ -166,7 +170,7 @@ def centroid_shifts( new_centroids_t, # IN (n_features, n_clusters) centroid_shifts, # OUT (n_clusters,) ): - # fmt: on + # fmt: on sample_idx = dpex.get_global_id(zero_idx) if sample_idx >= n_clusters: @@ -213,7 +217,7 @@ def reduce_centroid_data( empty_clusters_list, # OUT (n_clusters,) n_empty_clusters, # OUT (1,) ): - # fmt: on + # fmt: on group_idx = dpex.get_group_id(zero_idx) item_idx = dpex.get_local_id(zero_idx)	Code formatting pipeline doesn't seem to work Example of output: https://github.com/soda-inria/sklearn-numba-dpex/actions/runs/3541364642/jobs/5945541586 Configuration needed to point to the folders that should be scanned ?	Indeed, the checked aren't run on the CI. Adding the `--all-files` option force it to runs everywhere: ```bash pre-commit run --all-files ``` I will do a PR.	2022-11-28T10:49:45	[]	[]
soda-inria/sklearn-numba-dpex	soda-inria__sklearn-numba-dpex-25	3d28986e375cb47d981629cd6fac47e911336275	diff --git a/benchmark/ext_helpers/daal4py.py b/benchmark/ext_helpers/daal4py.py index 7dde6f0..a0b1721 100644 --- a/benchmark/ext_helpers/daal4py.py +++ b/benchmark/ext_helpers/daal4py.py @@ -22,6 +22,9 @@ from daal4py.sklearn.cluster._k_means_0_23 import KMeans +from sklearn.exceptions import NotSupportedByEngineError + + # TODO: instead of relying on monkey patching, find a way to register a distinct entry # point that can load a distinct engine outside of setup.py (impossible ?) @@ -37,6 +40,11 @@ def daal4py_kmeans_single(self, X, sample_weight, centers_init): f"but got `estimator._algorithm = {self.estimator._algorithm}` instead " ) + if sample_weight is not None and any(sample_weight != sample_weight[0]): + raise NotSupportedByEngineError( + "Non unary sample_weight is not supported by daal4py." + ) + est = KMeans( n_clusters=centers_init.shape[0], init=centers_init, diff --git a/benchmark/kmeans.py b/benchmark/kmeans.py index 295a6dd..d729e19 100644 --- a/benchmark/kmeans.py +++ b/benchmark/kmeans.py @@ -1,9 +1,11 @@ from time import perf_counter +from inspect import signature import numpy as np import sklearn from numpy.testing import assert_array_equal from sklearn.cluster import KMeans +from sklearn.exceptions import NotSupportedByEngineError # TODO: maybe this class could be abstracted to use the same strategy with other @@ -35,12 +37,19 @@ def __init__( max_iter, skip_slow=False, run_consistency_checks=True, + data_initialization_kwargs, ): ( self.X, - self.sample_weight, + self.sample_weight_, self.centers_init, - ) = data_initialization_fn() + ) = data_initialization_fn(data_initialization_kwargs) + + self.sample_weight = data_initialization_kwargs.get( + "sample_weight", + signature(data_initialization_fn).parameters["sample_weight"].default, + ) + self.max_iter = max_iter self.run_consistency_checks = run_consistency_checks self.results = None @@ -67,7 +76,7 @@ def timeit(self, name, engine_provider=None, is_slow=False): centers_init = self.centers_init.copy() X = self.X.copy() sample_weight = ( - None if self.sample_weight is None else self.sample_weight.copy() + None if self.sample_weight_ is None else self.sample_weight_.copy() ) est_kwargs = dict( @@ -85,14 +94,22 @@ def timeit(self, name, engine_provider=None, is_slow=False): estimator = KMeans(est_kwargs) - print(f"Running {name} with parameters max_iter={max_iter} tol={tol} ...") + print( + f"Running {name} with parameters sample_weight={self.sample_weight} " + f"n_clusters={n_clusters} data_shape={X.shape} max_iter={max_iter} " + "tol={tol} ..." + ) with sklearn.config_context(engine_provider=engine_provider): # dry run, to be fair for JIT implementations - KMeans(est_kwargs).set_params(max_iter=1).fit( - X, sample_weight=sample_weight - ) + try: + KMeans(*est_kwargs).set_params(max_iter=1).fit( + X, sample_weight=sample_weight + ) + except NotSupportedByEngineError as e: + print((repr(e) + "\n")) + return t0 = perf_counter() estimator.fit(X, sample_weight=sample_weight) @@ -154,6 +171,7 @@ def _check_same_fit(self, estimator, name, max_iter, assert_allclose): # TODO: expose CLI args. random_state = 123 + sample_weight = "random" # None, "unary", or "random" n_clusters = 127 max_iter = 100 tol = 0 @@ -163,7 +181,11 @@ def _check_same_fit(self, estimator, name, max_iter, assert_allclose): # close results but with significant differences for a few elements. run_consistency_checks = False - def benchmark_data_initialization(random_state=random_state, n_clusters=n_clusters): + def benchmark_data_initialization( + random_state, + n_clusters, + sample_weight, + ): X, _ = fetch_openml(name="spoken-arabic-digit", return_X_y=True, parser="auto") X = X.astype(dtype) scaler_x = MinMaxScaler() @@ -174,10 +196,29 @@ def benchmark_data_initialization(random_state=random_state, n_clusters=n_cluste # X = np.vstack([X for _ in range(20)]) rng = default_rng(random_state) init = np.array(rng.choice(X, n_clusters, replace=False), dtype=np.float32) - return X, None, init + + if sample_weight is None: + pass + elif sample_weight == "unary": + sample_weight = np.ones(len(X), dtype=dtype) + elif sample_weight == "random": + sample_weight = rng.random(size=len(X)).astype(dtype) + else: + raise ValueError( + 'Expected sample_weight in {None, "unary", "random"}, got' + f" {sample_weight} instead." + ) + + return X, sample_weight, init kmeans_timer = KMeansLloydTimeit( - benchmark_data_initialization, max_iter, skip_slow, run_consistency_checks + benchmark_data_initialization, + max_iter, + skip_slow, + run_consistency_checks, + random_state=random_state, + n_clusters=n_clusters, + sample_weight=sample_weight, ) kmeans_timer.timeit( @@ -199,10 +240,12 @@ def benchmark_data_initialization(random_state=random_state, n_clusters=n_cluste kmeans_timer.timeit( name="daal4py lloyd GPU", engine_provider="sklearn_numba_dpex", - is_sloww=True, + is_slow=True, ) - for multiplier in [1, 2, 4, 8]: + work_group_size_multipliers = [1, 2, 4, 8] + + for multiplier in work_group_size_multipliers: with override_attr_context( KMeansEngine, diff --git a/sklearn_numba_dpex/kmeans/drivers.py b/sklearn_numba_dpex/kmeans/drivers.py index b11adb6..227e50d 100644 --- a/sklearn_numba_dpex/kmeans/drivers.py +++ b/sklearn_numba_dpex/kmeans/drivers.py @@ -16,10 +16,8 @@ make_compute_euclidean_distances_fixed_window_kernel, make_centroid_shifts_kernel, make_reduce_centroid_data_kernel, - make_initialize_to_zeros_1d_kernel, make_initialize_to_zeros_2d_kernel, make_initialize_to_zeros_3d_kernel, - make_copyto_2d_kernel, make_broadcast_division_1d_2d_kernel, make_half_l2_norm_2d_axis0_kernel, make_sum_reduction_1d_kernel, @@ -224,6 +222,7 @@ def _lloyd( ( X, + sample_weight, cluster_centers, compute_dtype, output_dtype, @@ -231,7 +230,7 @@ def _lloyd( n_features, n_samples, n_clusters, - ) = self._check_inputs(X, centers_init, sample_weight) + ) = self._check_inputs(X, sample_weight, centers_init) # Create a set of kernels ( @@ -241,6 +240,7 @@ def _lloyd( n_samples, n_features, n_clusters, + return_inertia=bool(verbose), preferred_work_group_size_multiple=self.preferred_work_group_size_multiple, global_mem_cache_size=self.global_mem_cache_size, centroids_window_width_multiplier=self.centroids_window_width_multiplier, @@ -261,10 +261,6 @@ def _lloyd( dtype=compute_dtype, ) - reset_per_sample_inertia_kernel = make_initialize_to_zeros_1d_kernel( - size=n_samples, work_group_size=work_group_size, dtype=compute_dtype - ) - reset_cluster_sizes_private_copies_kernel = make_initialize_to_zeros_2d_kernel( size0=n_centroids_private_copies, size1=n_clusters, @@ -280,10 +276,6 @@ def _lloyd( dtype=compute_dtype, ) - copyto_kernel = make_copyto_2d_kernel( - size0=n_features, size1=n_clusters, work_group_size=work_group_size - ) - broadcast_division_kernel = make_broadcast_division_1d_2d_kernel( size0=n_features, size1=n_clusters, @@ -326,11 +318,15 @@ def _lloyd( dtype=compute_dtype, ) - X_t, centroids_t = self._load_transposed_data_to_device(X, cluster_centers) + X_t, sample_weight, centroids_t = self._load_transposed_data_to_device( + X, + sample_weight, + cluster_centers, + ) # Allocate the necessary memory in the device global memory new_centroids_t = dpctl.tensor.empty_like(centroids_t, device=self.device) - best_centroids_t = dpctl.tensor.empty_like(new_centroids_t, device=self.device) + centroids_half_l2_norm = dpctl.tensor.empty( n_clusters, dtype=compute_dtype, device=self.device ) @@ -340,13 +336,12 @@ def _lloyd( centroid_shifts = dpctl.tensor.empty( n_clusters, dtype=compute_dtype, device=self.device ) - per_sample_inertia = per_sample_pseudo_inertia = dpctl.tensor.empty( + per_sample_inertia = dpctl.tensor.empty( n_samples, dtype=compute_dtype, device=self.device ) assignments_idx = dpctl.tensor.empty( n_samples, dtype=np.uint32, device=self.device ) - new_centroids_t_private_copies = dpctl.tensor.empty( (n_centroids_private_copies, n_features, n_clusters), dtype=compute_dtype, @@ -367,25 +362,12 @@ def _lloyd( # The loop n_iteration = 0 centroid_shifts_sum = np.inf - best_pseudo_inertia = np.inf - - copyto_kernel(centroids_t, best_centroids_t) # TODO: Investigate possible speedup with a custom dpctl queue with a custom # DAG of events and a final single "wait" - # TODO: should we conform to sklearn vanilla kmeans strategy and avoid - # computing an extra centroid update in the last iteration that will not be - # used ? - # It's not a significant difference anyway since when performance matters - # usually we have n_iteration > 100 - # TODO: there's an implicit choice in this code that will zero-out a centroid - # if its cluster is empty. It is not a good choice (a better choice would be to - # replace the centroid with a point in the dataset that is far for its closest - # centroid) - while (n_iteration <= max_iter) and (centroid_shifts_sum > tol): + while (n_iteration < max_iter) and (centroid_shifts_sum > tol): half_l2_norm_kernel(centroids_t, centroids_half_l2_norm) - reset_per_sample_inertia_kernel(per_sample_pseudo_inertia) reset_cluster_sizes_private_copies_kernel(cluster_sizes_private_copies) reset_centroids_private_copies_kernel(new_centroids_t_private_copies) nb_empty_clusters[0] = np.int32(0) @@ -393,9 +375,10 @@ def _lloyd( # TODO: implement special case where only one copy is needed fused_lloyd_fixed_window_single_step_kernel( X_t, + sample_weight, centroids_t, centroids_half_l2_norm, - per_sample_pseudo_inertia, + per_sample_inertia, new_centroids_t_private_copies, cluster_sizes_private_copies, ) @@ -426,17 +409,34 @@ def _lloyd( reduce_centroid_shifts_kernel(centroid_shifts) ) - pseudo_inertia, _ = dpctl.tensor.asnumpy( - reduce_inertia_kernel(per_sample_pseudo_inertia) - ) - - # TODO: should we drop this check ? - # (it should not be needed because we have theoritical guarantee that - # inertia decreases, if it doesn't it should only be because of rounding - # errors) - if pseudo_inertia < best_pseudo_inertia: - best_pseudo_inertia = pseudo_inertia - copyto_kernel(centroids_t, best_centroids_t) + if verbose: + # ???: verbosity comes at the cost of performance since it triggers + # computing exact inertia at each iteration. Shouldn't this be + # documented ? + inertia, _ = dpctl.tensor.asnumpy( + reduce_inertia_kernel(per_sample_inertia) + ) + print(f"Iteration {n_iteration}, inertia {inertia}") + + # ???: unlike sklearn, sklearn_intelex checks that pseudo_inertia decreases + # and keep an additional copy of centroids that is updated only if the + # value of the pseudo_inertia is smaller than all the past values. + # + # (the check should not be needed because we have theoritical guarantee + # that inertia decreases, if it doesn't it should only be because of + # rounding errors ?) + # + # To this purpose this code could be inserted here: + # + # if pseudo_inertia < best_pseudo_inertia: + # best_pseudo_inertia = pseudo_inertia + # copyto_kernel(centroids_t, best_centroids_t) + # + # Note that what that is saved as "best centroid" is the array before the + # update, to which the pseudo inertia that is computed at this iteration + # refers. For this reason, this strategy is not compatible with sklearn + # unit tests, that consider new_centroids_t (the array after the update) + # to be the best centroids at each iteration. centroids_t, new_centroids_t = ( new_centroids_t, @@ -445,14 +445,25 @@ def _lloyd( n_iteration += 1 + if verbose: + converged_at = n_iteration - 1 + if centroid_shifts_sum == 0: # NB: possible if tol = 0 + print(f"Converged at iteration {converged_at}: strict convergence.") + + elif centroid_shifts_sum <= tol: + print( + f"Converged at iteration {converged_at}: center shift " + f"{centroid_shifts_sum} within tolerance {tol}." + ) + # Finally, run an assignment kernel to compute the assignments to the best # centroids found, along with the exact inertia. - half_l2_norm_kernel(best_centroids_t, centroids_half_l2_norm) - reset_per_sample_inertia_kernel(per_sample_inertia) + half_l2_norm_kernel(centroids_t, centroids_half_l2_norm) fixed_window_assignment_kernel( X_t, - best_centroids_t, + sample_weight, + centroids_t, centroids_half_l2_norm, per_sample_inertia, assignments_idx, @@ -466,8 +477,8 @@ def _lloyd( return ( assignments_idx, inertia, - best_centroids_t.T, - n_iteration - 1, # substract 1 to conform with vanilla sklearn count + centroids_t.T, + n_iteration, output_dtype, ) @@ -477,13 +488,12 @@ def get_labels( sample_weight, centers, ): - labels = self._get_labels(X, sample_weight, centers) + labels = self._get_labels(X, centers) return dpctl.tensor.asnumpy(labels).astype(np.int32) def _get_labels( self, X, - sample_weight, centers, ): """This call is expected to accept the same inputs than @@ -492,6 +502,7 @@ def _get_labels( """ ( X, + _, centers, compute_dtype, _, @@ -499,7 +510,11 @@ def _get_labels( n_features, n_samples, n_clusters, - ) = self._check_inputs(X, centers, sample_weight) + ) = self._check_inputs( + X, + sample_weight=_IgnoreSampleWeight, + cluster_centers=centers, + ) # Create a set of kernels label_assignment_fixed_window_kernel = make_label_assignment_fixed_window_kernel( @@ -520,7 +535,7 @@ def _get_labels( dtype=compute_dtype, ) - X_t, centroids_t = self._load_transposed_data_to_device(X, centers) + X_t, _, centroids_t = self._load_transposed_data_to_device(X, None, centers) centroids_half_l2_norm = dpctl.tensor.empty( n_clusters, dtype=compute_dtype, device=self.device @@ -552,6 +567,7 @@ def get_inertia( """ ( X, + sample_weight, centers, compute_dtype, output_dtype, @@ -559,7 +575,7 @@ def get_inertia( n_features, n_samples, n_clusters, - ) = self._check_inputs(X, centers, sample_weight) + ) = self._check_inputs(X, sample_weight, centers) # Create a set of kernels compute_inertia_fixed_window_kernel = make_compute_inertia_fixed_window_kernel( @@ -587,7 +603,9 @@ def get_inertia( dtype=compute_dtype, ) - X_t, centroids_t = self._load_transposed_data_to_device(X, centers) + X_t, sample_weight, centroids_t = self._load_transposed_data_to_device( + X, sample_weight, centers + ) centroids_half_l2_norm = dpctl.tensor.empty( n_clusters, dtype=compute_dtype, device=self.device @@ -600,6 +618,7 @@ def get_inertia( compute_inertia_fixed_window_kernel( X_t, + sample_weight, centroids_t, centroids_half_l2_norm, per_sample_inertia, @@ -622,9 +641,9 @@ def get_euclidean_distances(self, X, Y): return dpctl.tensor.asnumpy(euclidean_distances).astype(output_dtype) def _get_euclidean_distances(self, X, Y): - # Input validation ( X, + _, Y, compute_dtype, output_dtype, @@ -632,7 +651,11 @@ def _get_euclidean_distances(self, X, Y): n_features, n_samples, n_clusters, - ) = self._check_inputs(X, Y, sample_weight=None) + ) = self._check_inputs( + X, + sample_weight=_IgnoreSampleWeight, + cluster_centers=Y, + ) label_assignment_fixed_window_kernel = make_compute_euclidean_distances_fixed_window_kernel( n_samples, @@ -652,7 +675,7 @@ def _get_euclidean_distances(self, X, Y): dtype=compute_dtype, ) - X_t, Y_t = self._load_transposed_data_to_device(X, Y) + X_t, _, Y_t = self._load_transposed_data_to_device(X, None, Y) Y_half_l2_norm = dpctl.tensor.empty( n_clusters, dtype=compute_dtype, device=self.device @@ -671,7 +694,7 @@ def _get_euclidean_distances(self, X, Y): ) return euclidean_distances_t.T, output_dtype() - def _set_dtype(self, X, centers_init): + def _set_dtype(self, X, sample_weight, centers_init): input_dtype = output_dtype = np.dtype(X.dtype).type compute_dtype = np.dtype(self.dtype or input_dtype).type copy = True @@ -701,7 +724,7 @@ def _set_dtype(self, X, centers_init): if copy: text += ( f" A copy of the data casted to type {compute_dtype} will be created. " - f"To save memory and prevent this warning, ensure that the dtype of " + f"To save memory and suppress this warning, ensure that the dtype of " f"the input data matches the dtype required for computations." ) warnings.warn(text, DataConversionWarning) @@ -718,26 +741,37 @@ def _set_dtype(self, X, centers_init): f"The centers have been passed with type {centers_init_dtype} but " f"type {compute_dtype} is expected. A copy will be created with the " f"correct type {compute_dtype}. Ensure that the centers are passed " - f"with the correct dtype to save memory and disable this warning.", + f"with the correct dtype to save memory and suppress this warning.", DataConversionWarning, ) centers_init = centers_init.astype(compute_dtype) - return X, centers_init, compute_dtype, output_dtype - - def _check_inputs(self, X, cluster_centers, sample_weight): - if (sample_weight is not None) and ((sample_weight != sample_weight[0]).any()): - raise NotSupportedByEngineError( - "The sklearn_numba_dpex engine for KMeans does not support " - "non-uniform sample weights at this time, but it is a planned " - "feature. More information about the progress toward support is " - "available on the issue tracker at: " - "https://github.com/soda-inria/sklearn-numba-dpex/issues/22 ." + if (sample_weight is not _IgnoreSampleWeight) and ( + sample_weight.dtype != compute_dtype + ): + warnings.warn( + f"sample_weight has been passed with type {sample_weight.dtype} but " + f"type {compute_dtype} is expected. A copy will be created with the " + f"correct type {compute_dtype}. Ensure that sample_weight is passed " + f"with the correct dtype to save memory and suppress this warning.", + DataConversionWarning, ) + sample_weight = sample_weight.astype(compute_dtype) - X, cluster_centers, compute_dtype, output_dtype = self._set_dtype( - X, cluster_centers - ) + return X, sample_weight, centers_init, compute_dtype, output_dtype + + def _check_inputs(self, X, sample_weight, cluster_centers): + + if sample_weight is None: + sample_weight = np.ones(len(X), dtype=(self.dtype or X.dtype)) + + ( + X, + sample_weight, + cluster_centers, + compute_dtype, + output_dtype, + ) = self._set_dtype(X, sample_weight, cluster_centers) work_group_size = ( self.work_group_size_multiplier self.preferred_work_group_size_multiple @@ -748,6 +782,7 @@ def _check_inputs(self, X, cluster_centers, sample_weight): n_clusters = cluster_centers.shape[0] return ( X, + sample_weight, cluster_centers, compute_dtype, output_dtype, @@ -757,7 +792,7 @@ def _check_inputs(self, X, cluster_centers, sample_weight): n_clusters, ) - def _load_transposed_data_to_device(self, X, cluster_centers): + def _load_transposed_data_to_device(self, X, sample_weight, cluster_centers): # Transfer the input data to device memory, # TODO: let the user pass directly dpctl.tensor or dpnp arrays to avoid copies. if self.X_layout == "C": @@ -777,4 +812,16 @@ def _load_transposed_data_to_device(self, X, cluster_centers): raise ValueError( f"Expected X_layout to be equal to 'C' or 'F', but got {self.X_layout} ." ) - return X_t, dpctl.tensor.from_numpy(cluster_centers.T, device=self.device) + + sample_weight = dpctl.tensor.from_numpy(sample_weight, device=self.device) + cluster_centers = dpctl.tensor.from_numpy(cluster_centers.T, device=self.device) + + return ( + X_t, + sample_weight, + cluster_centers, + ) + + +class _IgnoreSampleWeight: + pass diff --git a/sklearn_numba_dpex/kmeans/kernels/__init__.py b/sklearn_numba_dpex/kmeans/kernels/__init__.py index b049ebc..329ea58 100644 --- a/sklearn_numba_dpex/kmeans/kernels/__init__.py +++ b/sklearn_numba_dpex/kmeans/kernels/__init__.py @@ -8,10 +8,8 @@ from .utils import ( make_centroid_shifts_kernel, make_reduce_centroid_data_kernel, - make_initialize_to_zeros_1d_kernel, make_initialize_to_zeros_2d_kernel, make_initialize_to_zeros_3d_kernel, - make_copyto_2d_kernel, make_broadcast_division_1d_2d_kernel, make_half_l2_norm_2d_axis0_kernel, make_sum_reduction_1d_kernel, @@ -26,10 +24,8 @@ "make_compute_labels_inertia_fixed_window_kernel", "make_centroid_shifts_kernel", "make_reduce_centroid_data_kernel", - "make_initialize_to_zeros_1d_kernel", "make_initialize_to_zeros_2d_kernel", "make_initialize_to_zeros_3d_kernel", - "make_copyto_2d_kernel", "make_broadcast_division_1d_2d_kernel", "make_half_l2_norm_2d_axis0_kernel", "make_sum_reduction_1d_kernel", diff --git a/sklearn_numba_dpex/kmeans/kernels/compute_euclidean_distances.py b/sklearn_numba_dpex/kmeans/kernels/compute_euclidean_distances.py index 31b482c..96cc1ef 100644 --- a/sklearn_numba_dpex/kmeans/kernels/compute_euclidean_distances.py +++ b/sklearn_numba_dpex/kmeans/kernels/compute_euclidean_distances.py @@ -5,7 +5,6 @@ from ._base_kmeans_kernel_funcs import ( _make_initialize_window_kernel_funcs, - _make_update_closest_centroid_kernel_func, _make_accumulate_dot_products_kernel_func, ) diff --git a/sklearn_numba_dpex/kmeans/kernels/compute_inertia.py b/sklearn_numba_dpex/kmeans/kernels/compute_inertia.py index f12b81e..e9db5fc 100644 --- a/sklearn_numba_dpex/kmeans/kernels/compute_inertia.py +++ b/sklearn_numba_dpex/kmeans/kernels/compute_inertia.py @@ -75,6 +75,7 @@ def make_compute_inertia_fixed_window_kernel( # fmt: off def compute_inertia( X_t, # IN READ-ONLY (n_features, n_samples) + sample_weight, # IN READ-ONLY (n_features,) centroids_t, # IN READ-ONLY (n_features, n_clusters) centroids_half_l2_norm, # IN (n_clusters,) per_sample_inertia, # OUT (n_samples,) @@ -159,7 +160,7 @@ def compute_inertia( if sample_idx >= n_samples: return - per_sample_inertia[sample_idx] = X_l2_norm + (two * min_sample_pseudo_inertia) + per_sample_inertia[sample_idx] = X_l2_norm + (two * min_sample_pseudo_inertia) * sample_weight[sample_idx] global_size = (math.ceil(n_samples / work_group_size)) * (work_group_size) return compute_inertia[global_size, work_group_size] diff --git a/sklearn_numba_dpex/kmeans/kernels/compute_labels_inertia.py b/sklearn_numba_dpex/kmeans/kernels/compute_labels_inertia.py index d086da4..03404c1 100644 --- a/sklearn_numba_dpex/kmeans/kernels/compute_labels_inertia.py +++ b/sklearn_numba_dpex/kmeans/kernels/compute_labels_inertia.py @@ -74,6 +74,7 @@ def make_compute_labels_inertia_fixed_window_kernel( # fmt: off def labels_inertia( X_t, # IN READ-ONLY (n_features, n_samples) + sample_weight, # IN READ-ONLY (n_features,) centroids_t, # IN (n_features, n_clusters) centroids_half_l2_norm, # IN (n_clusters,) per_sample_inertia, # OUT (n_samples,) @@ -171,7 +172,7 @@ def labels_inertia( return assignments_idx[sample_idx] = min_idx - per_sample_inertia[sample_idx] = X_l2_norm + (two * min_sample_pseudo_inertia) + per_sample_inertia[sample_idx] = (X_l2_norm + (two * min_sample_pseudo_inertia)) * sample_weight[sample_idx] global_size = (math.ceil(n_samples / work_group_size)) * (work_group_size) return labels_inertia[global_size, work_group_size] diff --git a/sklearn_numba_dpex/kmeans/kernels/lloyd_single_step.py b/sklearn_numba_dpex/kmeans/kernels/lloyd_single_step.py index 1e70ca8..3895531 100644 --- a/sklearn_numba_dpex/kmeans/kernels/lloyd_single_step.py +++ b/sklearn_numba_dpex/kmeans/kernels/lloyd_single_step.py @@ -34,6 +34,7 @@ def make_lloyd_single_step_fixed_window_kernel( n_samples, n_features, n_clusters, + return_inertia, preferred_work_group_size_multiple, global_mem_cache_size, centroids_window_width_multiplier, @@ -71,6 +72,19 @@ def make_lloyd_single_step_fixed_window_kernel( dtype=dtype, ) + if return_inertia: + _accumulate_dot_products_and_X_l2_norm = ( + _make_accumulate_dot_products_kernel_func( + n_samples, + n_features, + centroids_window_height, + window_n_centroids, + with_X_l2_norm=True, + dtype=dtype, + ) + ) + two = dtype(2) + n_windows_per_feature = math.ceil(n_clusters / window_n_centroids) n_windows_per_centroid = math.ceil(n_features / centroids_window_height) @@ -86,7 +100,6 @@ def make_lloyd_single_step_fixed_window_kernel( // n_cluster_bytes ) - one = dtype(1.0) inf = dtype(math.inf) # TODO: currently, constant memory is not supported by numba_dpex, but for read-only @@ -104,9 +117,10 @@ def make_lloyd_single_step_fixed_window_kernel( # fmt: off def fused_lloyd_single_step( X_t, # IN READ-ONLY (n_features, n_samples) + sample_weight, # IN READ-ONLY (n_features,) current_centroids_t, # IN (n_features, n_clusters) centroids_half_l2_norm, # IN (n_clusters,) - per_sample_pseudo_inertia, # OUT (n_samples,) + per_sample_inertia, # OUT (n_samples,) new_centroids_t_private_copies, # OUT (n_private_copies, n_features, n_clusters) # noqa cluster_sizes_private_copies, # OUT (n_private_copies, n_clusters) # noqa ): @@ -213,13 +227,22 @@ def fused_lloyd_single_step( # before going forward dpex.barrier(dpex.CLK_LOCAL_MEM_FENCE) - _accumulate_dot_products( - sample_idx, - first_feature_idx, - X_t, - centroids_window, - dot_products, - ) + if return_inertia and _0 == 0: + X_l2_norm = _accumulate_dot_products_and_X_l2_norm( + sample_idx, + first_feature_idx, + X_t, + centroids_window, + dot_products, + ) + else: + _accumulate_dot_products( + sample_idx, + first_feature_idx, + X_t, + centroids_window, + dot_products, + ) # When the next iteration starts work items will overwrite shared memory # with new values, so before that we must wait for all reading @@ -249,6 +272,20 @@ def fused_lloyd_single_step( # End of outer loop. By now min_idx and min_sample_pseudo_inertia # contains the expected values. + # NB: this check can't be moved at the top at the kernel, because if a work item + # exits early with a `return` it will never reach the barriers, thus causing a + # deadlock. Early returns are only possible when there are no barriers within + # the remaining set of instructions for the running kernel. + if sample_idx >= n_samples: + return + + weight = sample_weight[sample_idx] + + if return_inertia: + inertia = min_sample_pseudo_inertia + inertia = (X_l2_norm + (two * min_sample_pseudo_inertia)) * weight + per_sample_inertia[sample_idx] = inertia + # STEP 2: update centroids. # Each work item updates n_features values in global memory for the centroid @@ -278,15 +315,6 @@ def fused_lloyd_single_step( # The privatization strategy also applies to the updates of the centroid # counts. - # NB: this check can't be moved at the top at the kernel, because if a work item - # exits early with a `return` it will never reach the barriers, thus causing a - # deadlock. Early returns are only possible when there are no barriers within - # the remaining set of instructions for the running kernel. - if sample_idx >= n_samples: - return - - per_sample_pseudo_inertia[sample_idx] = min_sample_pseudo_inertia - # each work item is assigned an array of centroids in a round robin manner privatization_idx = ( sample_idx // preferred_work_group_size_multiple @@ -295,14 +323,14 @@ def fused_lloyd_single_step( dpex.atomic.add( cluster_sizes_private_copies, (privatization_idx, min_idx), - one, + weight ) for feature_idx in range(n_features): dpex.atomic.add( new_centroids_t_private_copies, (privatization_idx, feature_idx, min_idx), - X_t[feature_idx, sample_idx], + X_t[feature_idx, sample_idx] * weight, ) global_size = (math.ceil(n_samples / work_group_size)) * (work_group_size) return ( diff --git a/sklearn_numba_dpex/kmeans/kernels/utils.py b/sklearn_numba_dpex/kmeans/kernels/utils.py index 7a2bbe5..81134d0 100644 --- a/sklearn_numba_dpex/kmeans/kernels/utils.py +++ b/sklearn_numba_dpex/kmeans/kernels/utils.py @@ -121,24 +121,6 @@ def reduce_centroid_data( return reduce_centroid_data[global_size, work_group_size] -@lru_cache -def make_initialize_to_zeros_1d_kernel(size, work_group_size, dtype): - - global_size = math.ceil(size / work_group_size) * work_group_size - zero = dtype(0) - - @dpex.kernel - def initialize_to_zeros(data): - item_idx = dpex.get_global_id(0) - - if item_idx >= size: - return - - data[item_idx] = zero - - return initialize_to_zeros[global_size, work_group_size] - - @lru_cache def make_initialize_to_zeros_2d_kernel(size0, size1, work_group_size, dtype): @@ -186,25 +168,6 @@ def initialize_to_zeros(data): return initialize_to_zeros[global_size, work_group_size] -@lru_cache -def make_copyto_2d_kernel(size0, size1, work_group_size): - global_size = math.ceil(size1 / work_group_size) * work_group_size - - # Optimized for C-contiguous array and for - # size1 >> preferred_work_group_size_multiple - @dpex.kernel - def copyto_kernel(input_array, target_array): - col_idx = dpex.get_global_id(0) - - if col_idx >= size1: - return - - for row_idx in range(size0): - target_array[row_idx, col_idx] = input_array[row_idx, col_idx] - - return copyto_kernel[global_size, work_group_size] - - @lru_cache def make_broadcast_division_1d_2d_kernel(size0, size1, work_group_size): global_size = math.ceil(size1 / work_group_size) * work_group_size	The KMeans engine does not support non-uniform sample_weight Trying to use the KMeans engine in `sklearn_numba_dpex` currently results in a `NotSupportedByEngineError` exception.		2022-09-29T16:15:29	[]	[]
erikrose/parsimonious	erikrose__parsimonious-211	69d8ab8ddfcaf67a35a3cc0f8738a5b0d47e3689	diff --git a/README.rst b/README.rst index afa55a6..fab1b16 100644 --- a/README.rst +++ b/README.rst @@ -443,6 +443,7 @@ Version History =============== (Next release) + * Improve error message in left-recursive rules. (lucaswiman) * Add support for range ``{min,max}`` repetition expressions (righthandabacus) * Fix bug in ```` and ``+`` for token grammars (lucaswiman) Add support for grammars on bytestrings (lucaswiman) diff --git a/parsimonious/exceptions.py b/parsimonious/exceptions.py index d0f9375..a03a549 100644 --- a/parsimonious/exceptions.py +++ b/parsimonious/exceptions.py @@ -1,3 +1,4 @@ +from textwrap import dedent from parsimonious.utils import StrAndRepr @@ -44,6 +45,20 @@ def column(self): return self.pos + 1 +class LeftRecursionError(ParseError): + def __str__(self): + rule_name = self.expr.name if self.expr.name else str(self.expr) + window = self.text[self.pos:self.pos + 20] + return dedent(f""" + Left recursion in rule {rule_name!r} at {window!r} (line {self.line()}, column {self.column()}). + + Parsimonious is a packrat parser, so it can't handle left recursion. + See https://en.wikipedia.org/wiki/Parsing_expression_grammar#Indirect_left_recursion + for how to rewrite your grammar into a rule that does not use left-recursion. + """ + ).strip() + + class IncompleteParseError(ParseError): """A call to ``parse()`` matched a whole Expression but did not consume the entire text.""" diff --git a/parsimonious/expressions.py b/parsimonious/expressions.py index 9200365..f93f2c6 100644 --- a/parsimonious/expressions.py +++ b/parsimonious/expressions.py @@ -10,7 +10,7 @@ from inspect import getfullargspec, isfunction, ismethod, ismethoddescriptor import regex as re -from parsimonious.exceptions import ParseError, IncompleteParseError +from parsimonious.exceptions import ParseError, IncompleteParseError, LeftRecursionError from parsimonious.nodes import Node, RegexNode from parsimonious.utils import StrAndRepr @@ -96,6 +96,9 @@ def _as_rhs(self): return AdHocExpression(name=rule_name) +IN_PROGRESS = object() + + class Expression(StrAndRepr): """A thing that can be matched against a piece of text""" @@ -186,10 +189,10 @@ def match_core(self, text, pos, cache, error): node = expr_cache[pos] else: # TODO: Set default value to prevent infinite recursion in left-recursive rules. - node = expr_cache[pos] = self._uncached_match(text, - pos, - cache, - error) + expr_cache[pos] = IN_PROGRESS # Mark as in progress + node = expr_cache[pos] = self._uncached_match(text, pos, cache, error) + if node is IN_PROGRESS: + raise LeftRecursionError(text, pos=-1, expr=self) # Record progress for error reporting: if node is None and pos >= error.pos and (	Don't allow recursion if no characters are being used up. I was trying to write left associative operators in grammar, but an issue that kept coming up was infinite recursion. This could be solved by making every operator right associative, because the parser would cut out the option of just going down the same node over and over. Making it so that the parser can only visit one node twice if characters have been used up would completely solve this issue.	After looking through the code I found that there is already a todo in the code. `expressions.py` line 183. This would be really nice to get solved. @joranmulderij could you post the example that's causing the infinite recursion, or a simplified version of it? That would make addressing the bug much easier because it can be turned into a test case. Thanks! Here is a pretty minimal example of what would cause this issue. This is a stripped down version of my grammar. ``` from parsimonious.grammar import Grammar language_grammar = r""" expression = operator_expression / non_operator_expression non_operator_expression = number_expression operator_expression = expression "+" non_operator_expression number_expression = ~"[0-9]+" """ grammar = Grammar(language_grammar) grammar.parse('1 + 2') ``` When this script is run, infinite recursion occurs.	2022-05-07T20:59:50	[]	[]
erikrose/parsimonious	erikrose__parsimonious-203	aa032364ffac8b0c1893f9c29dfdca7dd70298f6	diff --git a/README.rst b/README.rst index a6c8e07..80a41bf 100644 --- a/README.rst +++ b/README.rst @@ -282,6 +282,15 @@ Syntax Reference ``(things)`` Parentheses are used for grouping, like in every other language. + +``thing{n}`` Exactly ``n`` repetitions of ``thing``. + +``thing{n,m}`` Between ``n`` and ``m`` repititions (inclusive.) + +``thing{,m}`` At most ``m`` repetitions of ``thing``. + +``thing{n,}`` At least ``n`` repetitions of ``thing``. + ==================== ======================================================== .. _flags: https://docs.python.org/3/howto/regex.html#compilation @@ -434,6 +443,8 @@ Version History =============== (Next release) + * Add support for range ``{min,max}`` repetition expressions (righthandabacus) + * Fix bug in ```` and ``+`` for token grammars (lucaswiman) Add support for grammars on bytestrings (lucaswiman) .. warning:: diff --git a/parsimonious/exceptions.py b/parsimonious/exceptions.py index b51c336..29cdef9 100644 --- a/parsimonious/exceptions.py +++ b/parsimonious/exceptions.py @@ -30,14 +30,17 @@ def line(self): match.""" # This is a method rather than a property in case we ever wanted to # pass in which line endings we want to use. - return self.text.count('\n', 0, self.pos) + 1 + if isinstance(self.text, list): # TokenGrammar + return None + else: + return self.text.count('\n', 0, self.pos) + 1 def column(self): """Return the 1-based column where the expression ceased to match.""" # We choose 1-based because that's what Python does with SyntaxErrors. try: return self.pos - self.text.rindex('\n', 0, self.pos) - except ValueError: + except (ValueError, AttributeError): return self.pos + 1 diff --git a/parsimonious/expressions.py b/parsimonious/expressions.py index 0679527..bf10656 100644 --- a/parsimonious/expressions.py +++ b/parsimonious/expressions.py @@ -375,105 +375,71 @@ class Lookahead(Compound): """An expression which consumes nothing, even if its contained expression succeeds""" - # TODO: Merge this and Not for better cache hit ratios and less code. - # Downside: pretty-printed grammars might be spelled differently than what - # went in. That doesn't bother me. + __slots__ = ['negativity'] - def _uncached_match(self, text, pos, cache, error): - node = self.members[0].match_core(text, pos, cache, error) - if node is not None: - return Node(self, text, pos, pos) - - def _as_rhs(self): - return u'&%s' % self._unicode_members()[0] + def __init__(self, member, , negative=False, kwargs): + super(Lookahead, self).__init__(member, kwargs) + self.negativity = bool(negative) - -class Not(Compound): - """An expression that succeeds only if the expression within it doesn't - - In any case, it never consumes any characters; it's a negative lookahead. - - """ def _uncached_match(self, text, pos, cache, error): - # FWIW, the implementation in Parsing Techniques in Figure 15.29 does - # not bother to cache NOTs directly. node = self.members[0].match_core(text, pos, cache, error) - if node is None: + if (node is None) == self.negativity: # negative lookahead == match only if not found return Node(self, text, pos, pos) def _as_rhs(self): - # TODO: Make sure this parenthesizes the member properly if it's an OR - # or AND. - return u'!%s' % self._unicode_members()[0] + return u'%s%s' % ('!' if self.negativity else '&', self._unicode_members()[0]) +def Not(term): + return Lookahead(term, negative=True) # Quantifiers. None of these is strictly necessary, but they're darn handy. -class Optional(Compound): - """An expression that succeeds whether or not the contained one does - - If the contained expression succeeds, it goes ahead and consumes what it - consumes. Otherwise, it consumes nothing. - - """ - def _uncached_match(self, text, pos, cache, error): - node = self.members[0].match_core(text, pos, cache, error) - return (Node(self, text, pos, pos) if node is None else - Node(self, text, pos, node.end, children=[node])) - - def _as_rhs(self): - return u'%s?' % self._unicode_members()[0] +class Quantifier(Compound): + """An expression wrapper like the /+/?/{n,m} quantifier in regexes.""" + __slots__ = ['min', 'max'] -# TODO: Merge with OneOrMore. -class ZeroOrMore(Compound): - """An expression wrapper like the * quantifier in regexes.""" - - def _uncached_match(self, text, pos, cache, error): - new_pos = pos - children = [] - while True: - node = self.members[0].match_core(text, new_pos, cache, error) - if node is None or not (node.end - node.start): - # Node was None or 0 length. 0 would otherwise loop infinitely. - return Node(self, text, pos, new_pos, children) - children.append(node) - new_pos += node.end - node.start - - def _as_rhs(self): - return u'%s' % self._unicode_members()[0] - - -class OneOrMore(Compound): - """An expression wrapper like the + quantifier in regexes. - - You can also pass in an alternate minimum to make this behave like "2 or - more", "3 or more", etc. - - """ - __slots__ = ['min'] - - # TODO: Add max. It should probably succeed if there are more than the max - # --just not consume them. - - def __init__(self, member, name='', min=1): - super(OneOrMore, self).__init__(member, name=name) + def __init__(self, member, , min=0, max=float('inf'), name='', kwargs): + super(Quantifier, self).__init__(member, name=name, kwargs) self.min = min + self.max = max def _uncached_match(self, text, pos, cache, error): new_pos = pos children = [] - while True: + size = len(text) + while new_pos < size and len(children) < self.max: node = self.members[0].match_core(text, new_pos, cache, error) if node is None: - break + break # no more matches children.append(node) length = node.end - node.start - if length == 0: # Don't loop infinitely. + if len(children) >= self.min and length == 0: # Don't loop infinitely break new_pos += length if len(children) >= self.min: return Node(self, text, pos, new_pos, children) def _as_rhs(self): - return u'%s+' % self._unicode_members()[0] + if self.min == 0 and self.max == 1: + qualifier = '?' + elif self.min == 0 and self.max == float('inf'): + qualifier = '' + elif self.min == 1 and self.max == float('inf'): + qualifier = '+' + elif self.max == float('inf'): + qualifier = '{%d,}' % self.min + elif self.min == 0: + qualifier = '{,%d}' % self.max + else: + qualifier = '{%d,%d}' % (self.min, self.max) + return '%s%s' % (self._unicode_members()[0], qualifier) + +def ZeroOrMore(member, name=''): + return Quantifier(member, name=name, min=0, max=float('inf')) + +def OneOrMore(member, name='', min=1): + return Quantifier(member, name=name, min=min, max=float('inf')) + +def Optional(member, name=''): + return Quantifier(member, name=name, min=0, max=1) diff --git a/parsimonious/grammar.py b/parsimonious/grammar.py index 617f4d1..e38f1d0 100644 --- a/parsimonious/grammar.py +++ b/parsimonious/grammar.py @@ -10,7 +10,7 @@ from parsimonious.exceptions import BadGrammar, UndefinedLabel from parsimonious.expressions import (Literal, Regex, Sequence, OneOf, - Lookahead, Optional, ZeroOrMore, OneOrMore, Not, TokenMatcher, + Lookahead, Quantifier, Optional, ZeroOrMore, OneOrMore, Not, TokenMatcher, expression, is_callable) from parsimonious.nodes import NodeVisitor from parsimonious.utils import evaluate_string @@ -241,7 +241,7 @@ def _expressions_from_rules(self, rule_syntax, custom_rules): atom = reference / literal / regex / parenthesized regex = "~" spaceless_literal ~"[ilmsuxa]"i _ parenthesized = "(" _ expression ")" _ - quantifier = ~"[+?]" _ + quantifier = ~r"[+?]\|\{\d,\d+\}\|\{\d+,\d\}\|\{\d+\}" _ reference = label !equals # A subsequent equal sign is the only thing that distinguishes a label @@ -305,7 +305,17 @@ def visit_quantifier(self, node, quantifier): def visit_quantified(self, node, quantified): atom, quantifier = quantified - return self.quantifier_classes[quantifier.text](atom) + try: + return self.quantifier_classes[quantifier.text](atom) + except KeyError: + # This should pass: assert re.full_match("\{(\d)(,(\d))?\}", quantifier) + quantifier = quantifier.text[1:-1].split(",") + if len(quantifier) == 1: + min_match = max_match = int(quantifier[0]) + else: + min_match = int(quantifier[0]) if quantifier[0] else 0 + max_match = int(quantifier[1]) if quantifier[1] else float('inf') + return Quantifier(atom, min=min_match, max=max_match) def visit_lookahead_term(self, node, lookahead_term): ampersand, term, _ = lookahead_term	Range and some-of operators ### Range operators ~~Stolen~~ Adapted from POSIX regexes: ``` foo = bar{2} # expands into foo = bar bar foo = bar{2,4} # expands into foo = bar bar bar? bar? # this one isn't part of POSIX, but it's a fairly obvious extension foo = bar{,3} # expands into foo = bar? bar? bar? ``` Further, suffixing the upper bound with `=` prevents more instances of `bar` from existing: ``` foo = bar{,3=} # expands into foo = bar? bar? bar? !bar ``` This might be worth making the default? I'm not sure. It shouldn't be usable without the prefixed comma, since that sends conflicting messages about whether the number is a lower or upper bound. ### Some-of operators I have no idea what this syntax should look like. This is just what first came to mind. ``` foo = ~(bar baz bluh){2} # expands into foo = (bar baz) / (bar bluh) / (baz bluh) foo = ~(bar baz bluh){2,3} # expands into foo = (bar baz) / (bar bluh) / (baz bluh) / (bar baz bluh) ``` There isn't much of a point to a lower bound of 0, as that would be equivalent to chaining `?`, but it might as well be allowed anyway for simplicity. Note that reordering the parts isn't permitted here; it would be preferable to allow that, but the resulting implicit alternatives would be confusing: ``` foo = ~(bar baz bluh){2} # could expand into foo = (bar baz) / (bar bluh) / (baz bar) / (baz bluh) / (bluh bar) / (bluh baz) # but what does that parse things as? we just don't know # (it's still well-defined, of course, just weird) ``` If we introduce syntax to do this, then a lower bound of 0 does become meaningful. And, again, `=` is possible: ``` foo = ~(bar baz bluh){,2=} # expands into foo = ((bar baz) / (bar bluh) / (baz bluh)) (!bar !baz !bluh) ``	[Lojban's machine grammar]( http://lojban.org/publications/cll/cll_v1.1_xhtml-chapter-chunks/chapter-grammars.htm) uses `&` for my `~(...){}`. This doesn't allow for the `{}` minimum-maximum addition, but it's a lead to follow, if nothing else.	2022-04-11T06:25:42	[]	[]
atsb/nose-py3	atsb__nose-py3-5	069782a1bbec38236d5ee376e3c76ed364754269	diff --git a/nose/case.py b/nose/case.py index abef702..35f61ed 100644 --- a/nose/case.py +++ b/nose/case.py @@ -333,7 +333,7 @@ def __init__(self, method, test=None, arg=tuple(), descriptor=None): if isfunction(method): raise ValueError( "Unbound methods must be wrapped using pyversion.unbound_method before passing to MethodTestCase") - self.cls = method.im_class + self.cls = method.__self__.__class__ self.inst = self.cls() if self.test is None: method_name = self.method.__name__ diff --git a/nose/config.py b/nose/config.py index bcff992..fb15e9e 100644 --- a/nose/config.py +++ b/nose/config.py @@ -636,8 +636,8 @@ def __nonzero__(self): def user_config_files(): """Return path to any existing user config files """ - return filter(os.path.exists, - map(os.path.expanduser, config_files)) + return list(filter(os.path.exists, + map(os.path.expanduser, config_files))) def all_config_files(): diff --git a/nose/core.py b/nose/core.py index 0989236..e03d49c 100644 --- a/nose/core.py +++ b/nose/core.py @@ -224,32 +224,26 @@ def add_option(self, arg, kw): v = self.config.verbosity self.config.plugins.sort() for p in self.config.plugins: - print - "Plugin %s" % p.name + print("Plugin %s" % p.name) if v >= 2: - print - " score: %s" % p.score - print - '\n'.join(textwrap.wrap(p.help().strip(), - initial_indent=' ', - subsequent_indent=' ')) + print(" score: %s" % p.score) + print('\n'.join(textwrap.wrap(p.help().strip(), + initial_indent=' ', + subsequent_indent=' '))) if v >= 3: parser = DummyParser() p.addOptions(parser) if len(parser.options): - print - print - " Options:" + print() + print(" Options:") for opts, help in parser.options: - print - ' %s' % (', '.join(opts)) + print(' %s' % (', '.join(opts))) if help: - print - '\n'.join( + print('\n'.join( textwrap.wrap(help.strip(), initial_indent=' ', - subsequent_indent=' ')) - print + subsequent_indent=' '))) + print() def usage(cls): import nose diff --git a/nose/ext/dtcompat.py b/nose/ext/dtcompat.py index 8bebb61..5b73c7f 100644 --- a/nose/ext/dtcompat.py +++ b/nose/ext/dtcompat.py @@ -9,7 +9,7 @@ # # Modified for inclusion in nose to provide support for DocFileTest # -# - all doctests removed from module (they fail under 2.3 and 2.5) +# - all doctests removed from module (they fail under 2.3 and 2.5) # - now handles the $py.class extension when ran under Jython r"""Module doctest -- a framework for running examples in docstrings. @@ -905,8 +905,7 @@ def _find(self, tests, obj, name, module, source_lines, globs, seen): add them to `tests`. """ if self._verbose: - print - 'Finding tests in %s' % name + print('Finding tests in %s' % name) # If we've already processed this object, then ignore it. if id(obj) in seen: diff --git a/nose/failure.py b/nose/failure.py index ecef903..d955d9c 100644 --- a/nose/failure.py +++ b/nose/failure.py @@ -15,7 +15,7 @@ class Failure(unittest.TestCase): A Failure case is placed in a test suite to indicate the presence of a test that could not be loaded or executed. A common example is a test module that fails to import. - + """ __test__ = False # do not collect @@ -37,7 +37,7 @@ def address(self): def runTest(self): if self.tb is not None: if is_base_exception(self.exc_val): - raise self.exc_val(None).with_traceback(self.tb) + raise self.exc_class(None).with_traceback(self.tb) raise self.exc_class(self.exc_val).with_traceback(self.tb) else: raise self.exc_class(self.exc_val) diff --git a/nose/loader.py b/nose/loader.py index 605e76e..f57fb9b 100644 --- a/nose/loader.py +++ b/nose/loader.py @@ -112,7 +112,7 @@ def wanted(attr, cls=testCaseClass, sel=self.selector): return False return sel.wantMethod(item) - cases = filter(wanted, dir(testCaseClass)) + cases = list(filter(wanted, dir(testCaseClass))) # add runTest if nothing else picked if not cases and hasattr(testCaseClass, 'runTest'): @@ -225,7 +225,7 @@ def loadTestsFromFile(self, filename): # Plugins can yield False to indicate that they were # unable to load tests from a file, but it was not an # error -- the file just had no tests to load. - tests = filter(None, tests) + tests = list(filter(None, tests)) return self.suiteClass(tests) else: # Nothing was able to even try to load from this file @@ -276,8 +276,8 @@ def loadTestsFromGeneratorMethod(self, generator, cls): """ # convert the unbound generator method # into a bound method so it can be called below - if hasattr(generator, 'im_class'): - cls = generator.im_class + if hasattr(generator, '__self__'): + cls = generator.__self__.__class__ inst = cls() method = generator.__name__ generator = getattr(inst, method) @@ -330,10 +330,12 @@ def loadTestsFromModule(self, module, path=None, discovered=False): test_classes.append(test) elif isfunction(test) and self.selector.wantFunction(test): test_funcs.append(test) + test_classes = sort_list(test_classes, lambda x: x.__name__) + tests.extend(self.makeTest(t, parent=module) for t in test_classes) + test_funcs = sort_list(test_funcs, func_lineno) - tests = map(lambda t: self.makeTest(t, parent=module), - test_classes + test_funcs) + tests.extend(self.makeTest(t, parent=module) for t in test_funcs) # Now, descend into packages # FIXME can or should this be lazy? diff --git a/nose/plugins/attrib.py b/nose/plugins/attrib.py index 217f1e5..5ab0c63 100644 --- a/nose/plugins/attrib.py +++ b/nose/plugins/attrib.py @@ -125,7 +125,7 @@ def wrap_ob(ob): def get_method_attr(method, cls, attr_name, default=False): - """Look up an attribute on a method/ function. + """Look up an attribute on a method/ function. If the attribute isn't found there, looking it up in the method's class, if any. """ @@ -140,7 +140,7 @@ def get_method_attr(method, cls, attr_name, default=False): class ContextHelper: """Object that can act as context dictionary for eval and looks up - names as attributes on a method/ function and its class. + names as attributes on a method/ function and its class. """ def __init__(self, method, cls): @@ -283,7 +283,7 @@ def wantMethod(self, method): """Accept the method if its attributes match. """ try: - cls = method.im_class + cls = method.__self__.__class__ except AttributeError: return False return self.validateAttrib(method, cls) diff --git a/nose/plugins/xunit.py b/nose/plugins/xunit.py index 7500620..1d51e2f 100644 --- a/nose/plugins/xunit.py +++ b/nose/plugins/xunit.py @@ -55,7 +55,7 @@ from io import StringIO -from numpy import unicode +from numpy.core import unicode from nose.plugins.base import Plugin from nose.pyversion import force_unicode, format_exception diff --git a/nose/pyversion.py b/nose/pyversion.py index a241d73..c7de0ae 100644 --- a/nose/pyversion.py +++ b/nose/pyversion.py @@ -6,12 +6,12 @@ import sys import traceback -from numpy import unicode +from numpy.core import unicode import nose.util -__all__ = ['make_instancemethod', 'cmp_to_key', 'sort_list', 'ClassType', - 'TypeType', 'UNICODE_STRINGS', 'unbound_method', 'ismethod', +__all__ = ['make_instancemethod', 'cmp_to_key', 'sort_list', + 'UNICODE_STRINGS', 'unbound_method', 'ismethod', 'bytes_', 'is_base_exception', 'force_unicode', 'exc_to_unicode', 'format_exception', 'isgenerator'] @@ -61,13 +61,6 @@ def sort_list(l, key, reverse=False): return sorted_list -# In Python 3.x, all objects are "new style" objects descended from 'type', and -# thus types.ClassType and types.TypeType don't exist anymore. For -# compatibility, we make sure they still work. -ClassType = type -TypeType = type - - # The following emulates the behavior (we need) of an 'unbound method' under # Python 3.x (namely, the ability to have a class associated with a function # definition so that things can do stuff based on its associated class) @@ -78,8 +71,6 @@ def __init__(self, cls, func): self.__dict__ = func.__dict__.copy() self._func = func self.__self__ = UnboundSelf(cls) - if sys.version_info < (3, 0): - self.im_class = cls self.__doc__ = getattr(func, '__doc__', None) def address(self): diff --git a/nose/suite.py b/nose/suite.py index 9ee91d5..5dfd961 100644 --- a/nose/suite.py +++ b/nose/suite.py @@ -436,9 +436,7 @@ def ancestry(self, context): # Methods include reference to module they are defined in, we # don't want that, instead want the module the class is in now # (classes are re-ancestored elsewhere). - if hasattr(context, 'im_class'): - context = context.im_class - elif hasattr(context, '__self__'): + if hasattr(context, '__self__'): context = context.__self__.__class__ if hasattr(context, '__module__'): ancestors = context.__module__.split('.') diff --git a/nose/util.py b/nose/util.py index 0eb3d2d..b5f325f 100644 --- a/nose/util.py +++ b/nose/util.py @@ -10,14 +10,13 @@ import types import unittest -from numpy import unicode +from numpy.core import unicode -from nose.pyversion import ClassType, TypeType, isgenerator +from nose.pyversion import isgenerator log = logging.getLogger('nose') ident_re = re.compile(r'^[A-Za-z_][A-Za-z0-9_.]$') -class_types = (ClassType, TypeType) skip_pattern = r"(?:\.svn)\|(?:[^.]+\.py[co])\|(?:.~)\|(?:.\$py\.class)\|(?:__pycache__)" set = set @@ -157,7 +156,7 @@ def isclass(obj): for objects that can't be subclasses of anything. """ obj_type = type(obj) - return obj_type in class_types or issubclass(obj_type, type) + return obj_type is type or issubclass(obj_type, type) # backwards compat (issue #64) @@ -398,7 +397,7 @@ def test_address(test): file = getattr(test, '__file__', None) module = getattr(test, '__name__', None) return src(file), module, call - if t == types.FunctionType or issubclass(t, type) or t == types.ClassType: + if t == types.FunctionType or t is type or issubclass(t, type): module = getattr(test, '__module__', None) if module is not None: m = sys.modules[module] @@ -408,7 +407,7 @@ def test_address(test): call = getattr(test, '__name__', None) return src(file), module, call if t == types.MethodType: - cls_adr = test_address(test.im_class) + cls_adr = test_address(test.__self__.__class__) return (src(cls_adr[0]), cls_adr[1], "%s.%s" % (cls_adr[2], test.__name__)) # handle unittest.TestCase instances @@ -449,15 +448,16 @@ def try_run(obj, names): if type(obj) == types.ModuleType: # py.test compatibility if isinstance(func, types.FunctionType): - args, varargs, varkw, defaults = \ - inspect.getfullargspec(func) + spec = inspect.getfullargspec(func) + args, varargs, varkw, defaults = spec[:4] + else: # Not a function. If it's callable, call it anyway if hasattr(func, '__call__') and not inspect.ismethod(func): func = func.__call__ try: - args, varargs, varkw, defaults = \ - inspect.getfullargspec(func) + spec = inspect.getfullargspec(func) + args, varargs, varkw, defaults = spec[:4] args.pop(0) # pop the self off except TypeError: raise TypeError("Attribute %s of %r is not a python "	AttributeError since upgrade from 1.4.0 -> 1.4.3 Hi, I get the exception below when upgrading nose-py3 from 1.4.0 to 1.4.3: ``` (daproject) [dauser@desk004 django]$ python --version Python 3.10.3 (daproject) [dauser@desk004 django]$ pip freeze \| grep -i nose django-nose==1.4.7 nose-py3==1.4.3 (daproject) [dauser@desk004 django]$ nosetests positions --verbose --nocapture --verbosity=2 Traceback (most recent call last): File "/home/dauser/projects/daproject/venv/daproject/bin/nosetests", line 8, in <module> sys.exit(run_exit()) File "/home/dauser/projects/daproject/venv/daproject/lib/python3.10/site-packages/nose/core.py", line 117, in __init__ unittest.TestProgram.__init__( File "/home/dauser/installed/fedora_f35/python3.10.3/usr/local/lib/python3.10/unittest/main.py", line 100, in __init__ self.parseArgs(argv) File "/home/dauser/projects/daproject/venv/daproject/lib/python3.10/site-packages/nose/core.py", line 178, in parseArgs self.createTests() File "/home/dauser/projects/daproject/venv/daproject/lib/python3.10/site-packages/nose/core.py", line 192, in createTests self.test = self.testLoader.loadTestsFromNames(self.testNames) File "/home/dauser/projects/daproject/venv/daproject/lib/python3.10/site-packages/nose/loader.py", line 484, in loadTestsFromNames return unittest.TestLoader.loadTestsFromNames(self, names, module) File "/home/dauser/installed/fedora_f35/python3.10.3/usr/local/lib/python3.10/unittest/loader.py", line 220, in loadTestsFromNames suites = [self.loadTestsFromName(name, module) for name in names] File "/home/dauser/installed/fedora_f35/python3.10.3/usr/local/lib/python3.10/unittest/loader.py", line 220, in <listcomp> suites = [self.loadTestsFromName(name, module) for name in names] File "/home/dauser/projects/daproject/venv/daproject/lib/python3.10/site-packages/nose/loader.py", line 434, in loadTestsFromName return self.loadTestsFromModule( File "/home/dauser/projects/daproject/venv/daproject/lib/python3.10/site-packages/nose/loader.py", line 357, in loadTestsFromModule tests.extend(self.loadTestsFromDir(module_path)) AttributeError: 'map' object has no attribute 'extend' ```	Hmmm this is really strange. I gave 1.4.3 a good bashing on my system before I uploaded the package and had no issues with it not running tests. I am assuming that these tests worked previously on 1.4.0? > I am assuming that these tests worked previously on 1.4.0? Yes, rolling back to 1.4.0 removes this exception for me. To be precise: I usually run this using django's testing facility by running `python manage.py test positions --keepdb --verbose --nocapture` The first line of output of that statement is `nosetests positions --verbose --nocapture --verbosity=2` Using 1.4.0 it then proceeds as expected, whereas using 1.4.3 leads to the above exception with a slightly longer stacktrace: ``` (daproject) [dauser@desk004 django]$ python manage.py test positions --keepdb --verbose --nocapture nosetests positions --verbose --nocapture --verbosity=2 Traceback (most recent call last): File "/home/dauser/projects/daproject/src/django/manage.py", line 10, in <module> execute_from_command_line(sys.argv) File "/home/dauser/projects/daproject/venv/daproject/lib/python3.10/site-packages/django/core/management/__init__.py", line 446, in execute_from_command_line utility.execute() File "/home/dauser/projects/daproject/venv/daproject/lib/python3.10/site-packages/django/core/management/__init__.py", line 440, in execute self.fetch_command(subcommand).run_from_argv(self.argv) File "/home/dauser/projects/daproject/venv/daproject/lib/python3.10/site-packages/django/core/management/commands/test.py", line 24, in run_from_argv super().run_from_argv(argv) File "/home/dauser/projects/daproject/venv/daproject/lib/python3.10/site-packages/django/core/management/base.py", line 414, in run_from_argv self.execute(args, cmd_options) File "/home/dauser/projects/daproject/venv/daproject/lib/python3.10/site-packages/django/core/management/base.py", line 460, in execute output = self.handle(args, *options) File "/home/dauser/projects/daproject/venv/daproject/lib/python3.10/site-packages/django/core/management/commands/test.py", line 68, in handle failures = test_runner.run_tests(test_labels) File "/home/dauser/projects/daproject/venv/daproject/lib/python3.10/site-packages/django_nose/runner.py", line 308, in run_tests result = self.run_suite(nose_argv) File "/home/dauser/projects/daproject/venv/daproject/lib/python3.10/site-packages/django_nose/runner.py", line 244, in run_suite nose.core.TestProgram(argv=nose_argv, exit=False, File "/home/dauser/projects/daproject/venv/daproject/lib/python3.10/site-packages/nose/core.py", line 117, in __init__ unittest.TestProgram.__init__( File "/home/dauser/installed/fedora_f35/python3.10.3/usr/local/lib/python3.10/unittest/main.py", line 100, in __init__ self.parseArgs(argv) File "/home/dauser/projects/daproject/venv/daproject/lib/python3.10/site-packages/nose/core.py", line 178, in parseArgs self.createTests() File "/home/dauser/projects/daproject/venv/daproject/lib/python3.10/site-packages/nose/core.py", line 192, in createTests self.test = self.testLoader.loadTestsFromNames(self.testNames) File "/home/dauser/projects/daproject/venv/daproject/lib/python3.10/site-packages/nose/loader.py", line 484, in loadTestsFromNames return unittest.TestLoader.loadTestsFromNames(self, names, module) File "/home/dauser/installed/fedora_f35/python3.10.3/usr/local/lib/python3.10/unittest/loader.py", line 220, in loadTestsFromNames suites = [self.loadTestsFromName(name, module) for name in names] File "/home/dauser/installed/fedora_f35/python3.10.3/usr/local/lib/python3.10/unittest/loader.py", line 220, in <listcomp> suites = [self.loadTestsFromName(name, module) for name in names] File "/home/dauser/projects/daproject/venv/daproject/lib/python3.10/site-packages/nose/loader.py", line 434, in loadTestsFromName return self.loadTestsFromModule( File "/home/dauser/projects/daproject/venv/daproject/lib/python3.10/site-packages/nose/loader.py", line 357, in loadTestsFromModule tests.extend(self.loadTestsFromDir(module_path)) AttributeError: 'map' object has no attribute 'extend' ``` If just for fun I then execute `nosetests positions --verbose --nocapture --verbosity=2` without using `manage.py` that comes with django, I get the result in the original description above. Using 1.4.0 without django leads to some other errors that seem to be django related: \| \| 1.4.0 \| 1.4.3\| \| --------------- \| --------------- \| --------------- \| \|django \| OK \| error1 above \| \|nosetest only \| error2 \| error1 above \| The error1 prototype seems to be the following: ``` $ python Python 3.10.3 (main, Mar 22 2022, 13:00:21) [GCC 11.2.1 20220127 (Red Hat 11.2.1-9)] on linux Type "help", "copyright", "credits" or "license" for more information. >>> a = map(lambda x: x + 1, [2,3]) >>> a.extend Traceback (most recent call last): File "<stdin>", line 1, in <module> AttributeError: 'map' object has no attribute 'extend' ``` Which presumably could arise from line 335 in loader.py, where it says ``` tests = map(lambda t: self.makeTest(t, parent=module), test_classes + test_funcs) ``` followed by `tests.extend(self.loadTestsFromDir(module_path))` at line 357, which maybe expects `tests` to be a `list` rather than a `map` Replacing line 335 with ``` tests = list(map(lambda t: self.makeTest(t, parent=module), test_classes + test_funcs)) ``` seems to remove this exception, but then I get another one: ``` ====================================================================== ERROR: Failure: AttributeError ('function' object has no attribute 'im_class') ---------------------------------------------------------------------- Traceback (most recent call last): File "/home/dauser/projects/daproject/venv/daproject/lib/python3.10/site-packages/nose/failure.py", line 40, in runTest raise self.exc_val(None).with_traceback(self.tb) TypeError: --------------------- >> end captured logging << --------------------- ``` I'll stay with 1.4.0 for the time being ;-) I just stumbled upon the same AttributeError* with the latest 1.4.3 release. Pinning the version to 1.4.0 seems to work fine. As I only need this package to migrate some old legacy Python 2 code written with nose tests to Python 3 which will not use nose any more, this is no big deal here and therefore I did no further research on this topic. I've looked into this quite a bit but have not found the reason. It works fine for my projects on three separate installations, however this could be because I wipe the previous version and install fresh. There were quite a few changes from 1.4.0 to 1.4.3 and upgrading from that could indeed cause issues. In my case, I explicitly installed the nose-py3 package in a clean CI environment here on GitHub Actions, so in theory there should not be such a conflict. Adding on the comment from @odmi-frijt, the offending map call is limited to `not discovered or self.selector.wantModule(module)`, so you might run into a different path inside your projects and never call map there. If you want to further debug this: My test repository is a fork of https://github.com/jwilk-archive/ocrodjvu, which I tested with exactly the code of the base repository. This code is still Python2-only, but this should not be the problem, as I got the aforementioned error when simply doing `python -m pip install nose-py3 && make test` while on Python 3. Thanks! I'll get on that	2022-12-23T13:22:08	[]	[]
scrapinghub/dateparser	scrapinghub__dateparser-1224	f65936408e24ede5a4e3c05f624323bfc474c585	diff --git a/dateparser/parser.py b/dateparser/parser.py index 16751188b..40aa8ff7b 100644 --- a/dateparser/parser.py +++ b/dateparser/parser.py @@ -598,10 +598,13 @@ def _correct_for_month(self, dateobj): relative_base_month = ( relative_base.month if hasattr(relative_base, "month") else relative_base ) - if getattr(self, "_token_month", None) or relative_base_month: + + if getattr(self, "_token_month", None): return dateobj - dateobj = set_correct_month_from_settings(dateobj, self.settings) + dateobj = set_correct_month_from_settings( + dateobj, self.settings, relative_base_month + ) return dateobj @classmethod @@ -613,11 +616,13 @@ def parse(cls, datestring, settings, tz=None): # correction for past, future if applicable dateobj = po._correct_for_time_frame(dateobj, tz) + # correction for preference of month: beginning, current, end + # must happen before day so that day is derived from the correct month + dateobj = po._correct_for_month(dateobj) + # correction for preference of day: beginning, current, end dateobj = po._correct_for_day(dateobj) - # correction for preference of month: beginning, current, end - dateobj = po._correct_for_month(dateobj) period = po._get_period() return dateobj, period	Parsing using `{"PREFER_DAY_OF_MONTH": "last", "PREFER_MONTH_OF_YEAR": "last"}` doesn't use last day of the last month Hi, I noticed that when parsing a date with missing month and day of month with `{"PREFER_DAY_OF_MONTH": "last", "PREFER_MONTH_OF_YEAR": "last"}` settings, we default to the last day of the current month instead of the last day of the last month. For example, when running this code in February, it will return `2022/12/28` instead of `2022/12/31`, as the last day of feb is 28. ```python >>> import dateparser >>> dateparser.parse("2022", settings={"PREFER_DAY_OF_MONTH": "last", "PREFER_MONTH_OF_YEAR": "last"}) datetime.datetime(2022, 12, 28, 0, 0) ```	+1; this bit me because it correctly returned the 31st when I wrote my tests in March, then started failing in April :( I would guess it should resolve the month before the day	2024-04-02T15:56:26	[]	[]
scrapinghub/dateparser	scrapinghub__dateparser-1219	1d4b05875d2ee35987607f8e33efa4c8df74dfb2	diff --git a/dateparser/data/date_translation_data/en.py b/dateparser/data/date_translation_data/en.py index b70975e1b..1409dcadb 100644 --- a/dateparser/data/date_translation_data/en.py +++ b/dateparser/data/date_translation_data/en.py @@ -800,7 +800,7 @@ "(?:12\\s+)?midnight": "00:00" }, { - "(\\d+[.,]?\\d)h(\\d+[.,]?\\d)m?": "\\1:\\2" + "(\\d+[.,]?\\d)h(\\d+[.,]?\\d)": "\\1:\\2" }, { "(?<=from\\s+)now": "in" diff --git a/dateparser_data/supplementary_language_data/date_translation_data/en.yaml b/dateparser_data/supplementary_language_data/date_translation_data/en.yaml index d103bc577..38b31297d 100644 --- a/dateparser_data/supplementary_language_data/date_translation_data/en.yaml +++ b/dateparser_data/supplementary_language_data/date_translation_data/en.yaml @@ -67,7 +67,7 @@ simplifications: - a: '1' - (?:12\s+)?noon: '12:00' - (?:12\s+)?midnight: '00:00' - - (\d+[.,]?\d)h(\d+[.,]?\d)m?: \1:\2 + - (\d+[.,]?\d)h(\d+[.,]?\d): \1:\2 - (?<=from\s+)now: in - less than 1 minute ago: 45 second ago - (\d+[.,]?\d*) (decade\|year\|month\|week\|day\|hour\|minute\|second)s? later: in \1 \2	Formats like "1h20m" should be treated as relative but are actually absolute When parsing a string like `"1h20m"`, it should be treated as a relative date; i.e. the same as it would treat the string `"1 hour and 20 minutes"`. However this is not the actual behaviour. ```py >>> from datetime import datetime >>> from dateparser import parse >>> datetime.now() datetime.datetime(2021, 10, 28, 18, 17, 11, 563442) >>> parse("1h20m") datetime.datetime(2021, 10, 28, 1, 20) >>> parse("in 1h20m") datetime.datetime(2021, 10, 28, 1, 20) >>> parse("in 1 hour and 20 minutes") datetime.datetime(2021, 10, 28, 19, 37, 34, 355178) ```	I agree that “in 1h20m” needs fixing, I’m less sure about the one without the “in”. I feel like for the interpretation of that one we should have a setting that could indicate if it should be interpreted as time in the future, time in the past, or absolute time (which I would keep as default). Even with the setting to prefer in the future it still treats "1h20m" as absolute. Also, I feel it would definitely make the most sense for "1h20m" and "1 hour 20 minutes" to give the same results as one is a literal abbreviation of the other. Note that a space makes a difference: ```py >>> parse("1h20m") datetime.datetime(2021, 10, 28, 1, 20) >>> parse("1h 20m") datetime.datetime(2021, 10, 28, 16, 46, 56, 916241) ``` That is really counter intuitive. I think it would make most sense to make all those as relative. A possible workaround for the time being: ```py string = re.sub(r"([dhms])(\d)", r"\1 \2", "1h20m") parse(string) ``` This will add a space which will cause it to be treated relatively	2024-02-15T20:36:12	[]	[]
scrapinghub/dateparser	scrapinghub__dateparser-1183	3955530ad9a910bad5a463dc986fe383bb185f87	diff --git a/dateparser/parser.py b/dateparser/parser.py index 297e73f3b..dcd8a194d 100644 --- a/dateparser/parser.py +++ b/dateparser/parser.py @@ -563,18 +563,15 @@ def _correct_for_time_frame(self, dateobj, tz): # Convert dateobj to utc time to compare with self.now try: tz = tz or get_timezone_from_tz_string(self.settings.TIMEZONE) + tz_offset = tz.utcoffset(dateobj) except pytz.UnknownTimeZoneError: - tz = None + tz_offset = timedelta(hours=0) - if tz: - dateobj_time = (dateobj - tz.utcoffset(dateobj)).time() - else: - dateobj_time = dateobj.time() if "past" in self.settings.PREFER_DATES_FROM: - if self.now.time() < dateobj_time: + if self.now < dateobj - tz_offset: dateobj = dateobj + timedelta(days=-1) if "future" in self.settings.PREFER_DATES_FROM: - if self.now.time() > dateobj_time: + if self.now > dateobj - tz_offset: dateobj = dateobj + timedelta(days=1) # Reset dateobj to the original value, thus removing any offset awareness that may	Returned datetime skips a day I reported in #1092 that the datetime is returned correctly only when supplying the timezone in the argument but not via settings, which is still the behaviour as of 1.1.7. As of 1.1.8, both options give the same result again, however the date is wrong. See an example below. Note this is a different example than the one provided in #1092, the original example now works correctly on 1.1.8. This bug seems to occur when the time is already the next day in UTC. Versions: - Ubuntu 22.04.1 LTS - Python 3.10.6 - dateparser (1.1.8) - python-dateutil (2.8.2) - pytz (2022.6) - regex (2022.3.2) - tzlocal (4.2) Minimum working example ``` #!/usr/bin/env python3 import dateparser import datetime parse_settings={ 'PREFER_DATES_FROM': 'future', 'TO_TIMEZONE': 'etc/utc', 'RETURN_AS_TIMEZONE_AWARE': False, 'RELATIVE_BASE': datetime.datetime(2023, 5, 21, 15, 0) } time = dateparser.parse('7pm EDT', settings=parse_settings) print(time) # 2023-05-21 23:00:00 time = dateparser.parse('9pm EDT', settings=parse_settings) print(time) # 2023-05-23 01:00:00 parse_settings['TIMEZONE'] = 'america/new_york' time = dateparser.parse('7pm', settings=parse_settings) print(time) # 2023-05-21 23:00:00 time = dateparser.parse('9pm', settings=parse_settings) print(time) # 2023-05-23 01:00:00 ```		2023-09-14T02:39:40	[]	[]
scrapinghub/dateparser	scrapinghub__dateparser-1129	7cd3b80369ea85aaf9b544991dfc7e2e122b8b08	diff --git a/dateparser/data/date_translation_data/ru.py b/dateparser/data/date_translation_data/ru.py index eade5dc92..1561b02b2 100644 --- a/dateparser/data/date_translation_data/ru.py +++ b/dateparser/data/date_translation_data/ru.py @@ -431,6 +431,12 @@ }, { "((?<=(через\|спустя\|в течение)\\s+)года?\|(?<=[^\\d]\\s+\|^)год(?=(\\s+назад)))": "1 год" + }, + { + "(\\d{3,}1)\\sгод\\s$": "\\1" + }, + { + "(\\d[02-9])\\sгод\\b": "\\1" } ] } diff --git a/dateparser_data/supplementary_language_data/date_translation_data/ru.yaml b/dateparser_data/supplementary_language_data/date_translation_data/ru.yaml index b6004c57e..3112ffa1c 100644 --- a/dateparser_data/supplementary_language_data/date_translation_data/ru.yaml +++ b/dateparser_data/supplementary_language_data/date_translation_data/ru.yaml @@ -103,3 +103,5 @@ simplifications: - ((?<=(через\|спустя\|в течение)\s+)недел[юи]\|(?<=[^\d]\s+\|^)неделю(?=(\s+назад))): 1 неделю - ((?<=(через\|спустя\|в течение)\s+)месяца?\|(?<=[^\d]\s+\|^)месяц(?=(\s+назад))): 1 месяц - ((?<=(через\|спустя\|в течение)\s+)года?\|(?<=[^\d]\s+\|^)год(?=(\s+назад))): 1 год + - (\d{3,}1)\sгод\s$: \1 + - (\d[02-9])\sгод\b: \1	possible fix of 'ago' problem in Russian '_2000 год_' in Russian is '_year 2000_' Before was parsed as '_2000 years ago_' ``` parse('2000 год') datetime.datetime(17, 8, 3, 18, 44, 48, 222615) ``` After ``` parse('2000 год') datetime.datetime(2000, 8, 3, 0, 0) ```	# [Codecov](https://codecov.io/gh/scrapinghub/dateparser/pull/340?src=pr&el=h1) Report > Merging [#340](https://codecov.io/gh/scrapinghub/dateparser/pull/340?src=pr&el=desc) into [master](https://codecov.io/gh/scrapinghub/dateparser/commit/53084340d11f442b1a681fb13f023b28a879857c?src=pr&el=desc) will increase coverage by `<.01%`. > The diff coverage is `100%`. [![Impacted file tree graph](https://codecov.io/gh/scrapinghub/dateparser/pull/340/graphs/tree.svg?height=150&width=650&token=XbZaFXS3m4&src=pr)](https://codecov.io/gh/scrapinghub/dateparser/pull/340?src=pr&el=tree) ```diff @@ Coverage Diff @@ ## master #340 +/- ## ========================================== + Coverage 97.61% 97.61% +<.01% ========================================== Files 20 20 Lines 1674 1677 +3 ========================================== + Hits 1634 1637 +3 Misses 40 40 ``` \| [Impacted Files](https://codecov.io/gh/scrapinghub/dateparser/pull/340?src=pr&el=tree) \| Coverage Δ \| \| \|---\|---\|---\| \| [dateparser/freshness\_date\_parser.py](https://codecov.io/gh/scrapinghub/dateparser/pull/340/diff?src=pr&el=tree#diff-ZGF0ZXBhcnNlci9mcmVzaG5lc3NfZGF0ZV9wYXJzZXIucHk=) \| `98.97% <100%> (+0.03%)` \| :arrow_up: \| ------ [Continue to review full report at Codecov](https://codecov.io/gh/scrapinghub/dateparser/pull/340?src=pr&el=continue). > Legend - [Click here to learn more](https://docs.codecov.io/docs/codecov-delta) > `Δ = absolute <relative> (impact)`, `ø = not affected`, `? = missing data` > Powered by [Codecov](https://codecov.io/gh/scrapinghub/dateparser/pull/340?src=pr&el=footer). Last update [5308434...3cf105a](https://codecov.io/gh/scrapinghub/dateparser/pull/340?src=pr&el=lastupdated). Read the [comment docs](https://docs.codecov.io/docs/pull-request-comments). Please, see https://dateparser.readthedocs.io/en/latest/contributing.html#guidelines-for-editing-translation-data The reason behind this is that he `relative-time` parser is applied before the `absolute-time` parser... so it could be fixed by using the `PARSERS` settings, either deleting the `relative-time` parser or putting it in the last position. i.e.: ``` dateparser.parse('2000 год', settings={'PARSERS': ['timestamp', 'custom-formats', 'absolute-time', 'base-formats', 'relative-time']}) ``` Your approach is not good because we can't hardcode that regex there and would have side effects for other languages. However, the issue is still valid. We could probably add a regex into the ru.yaml file. Maybe a simplification as `- (\d+)\s*год: \1` would work, however, I can't speak Russian and I don't know if we should add something more, as it could be that with other following or preceding words (as "ago" or "in") this simplification shouldn't be applied.	2023-01-24T15:49:19	[]	[]
scrapinghub/dateparser	scrapinghub__dateparser-1121	78dc29866aca496642742f6c7aeffbc1bb508043	diff --git a/dateparser/data/date_translation_data/en.py b/dateparser/data/date_translation_data/en.py index 5c5b18acf..a90dfed21 100644 --- a/dateparser/data/date_translation_data/en.py +++ b/dateparser/data/date_translation_data/en.py @@ -93,7 +93,6 @@ "month": [ "mo", "month", - "mon", "months" ], "week": [ diff --git a/dateparser_data/supplementary_language_data/date_translation_data/en.yaml b/dateparser_data/supplementary_language_data/date_translation_data/en.yaml index 6885c5562..a2736f213 100644 --- a/dateparser_data/supplementary_language_data/date_translation_data/en.yaml +++ b/dateparser_data/supplementary_language_data/date_translation_data/en.yaml @@ -16,7 +16,6 @@ year: - y - years month: - - mon - months week: - weeks	Is not parsing properly the Last-Modified header value when is a Monday Hi, Just in case, in the previous version we were able to parse the following 'Mon, 05 Dec 2022 19:43:03 GMT' without any problem and now is returning None. This is happening only when is Monday. It could be related to this commit: https://github.com/scrapinghub/dateparser/commit/89a5b29517a358e2784cc818687b7b6bed343159 ? Thank you! Best, Nicolas.	I have confirmed https://github.com/scrapinghub/dateparser/commit/89a5b29517a358e2784cc818687b7b6bed343159 as the cause.	2023-01-10T15:51:06	[]	[]
scrapinghub/dateparser	scrapinghub__dateparser-1103	a51d9bd31c9c28f43bf069f87b42e0cd3454001e	diff --git a/dateparser/data/date_translation_data/en.py b/dateparser/data/date_translation_data/en.py index df187f684..5c5b18acf 100644 --- a/dateparser/data/date_translation_data/en.py +++ b/dateparser/data/date_translation_data/en.py @@ -87,11 +87,13 @@ "year": [ "year", "yr", + "y", "years" ], "month": [ "mo", "month", + "mon", "months" ], "week": [ @@ -101,6 +103,7 @@ ], "day": [ "day", + "d", "days" ], "hour": [ diff --git a/dateparser_data/supplementary_language_data/date_translation_data/en.yaml b/dateparser_data/supplementary_language_data/date_translation_data/en.yaml index 69ef6c515..6885c5562 100644 --- a/dateparser_data/supplementary_language_data/date_translation_data/en.yaml +++ b/dateparser_data/supplementary_language_data/date_translation_data/en.yaml @@ -13,12 +13,15 @@ decade: - decade - decades year: + - y - years month: + - mon - months week: - weeks day: + - d - days hour: - h	Cannot parse short days ago. Exp 1d ago Hi, I cannot parse some type of short days ago. Some examples - `1d ago` - `1mon ago` - `2y ago` ```python from dateparser import parse parse('1mon ago') >> None ```	another issue I find connected to this one: other languages sometimes uses 'ago' and for french the dateparser cannot parse it. ```python from dateparser import parse parse('4 hours ago') >> datetime.datetime(2022, 5, 31, 8, 25, 46, 392093) ``` ```python from dateparser import parse parse('4 heures ago') >> None ``` ```python from dateparser import parse parse('il y a 4 heures') >> datetime.datetime(2022, 5, 31, 8, 29, 7, 442826) ``` I guess it should be really easy to fix this issue, unless the short days as @hoangchunghien presented are a feature dateparser still doesn't includes	2022-11-24T13:55:50	[]	[]
scrapinghub/dateparser	scrapinghub__dateparser-1084	78dc29866aca496642742f6c7aeffbc1bb508043	diff --git a/dateparser/languages/locale.py b/dateparser/languages/locale.py index a64c7e79b..e3b43b6be 100644 --- a/dateparser/languages/locale.py +++ b/dateparser/languages/locale.py @@ -263,7 +263,7 @@ def _sentence_split(self, string, settings): splitters_dict = {1: r'[\.!?;…\r\n]+(?:\s\|$)', # most European, Tagalog, Hebrew, Georgian, # Indonesian, Vietnamese - 2: r'(?:[¡¿]+\|[\.!?;…\r\n]+(?:\s\|$))+', # Spanish + 2: r'[\.!?;…\r\n]+(\s[¡¿]*\|$)\|[¡¿]+', # Spanish 3: r'[\|!?;\r\n]+(?:\s\|$)+', # Hindi and Bangla 4: r'[。…‥\.!?？！;\r\n]+(?:\s\|$)+', # Japanese and Chinese 5: r'[\r\n]+', # Thai @@ -275,9 +275,7 @@ def _sentence_split(self, string, settings): split_reg = abbreviation_string + splitters_dict[self.info['sentence_splitter_group']] sentences = re.split(split_reg, string) - for i in sentences: - if not i: - sentences.remove(i) + sentences = filter(None, sentences) return sentences def _simplify_split_align(self, original, settings):	SECURITY: bad regex pattern in 'dateparser/languages/locale.py' will cause 'ReDos' security problem. #### Problem description i found one bad regex pattern in 'dateparser/languages/locale.py' ``` r'(?:[¡¿]+\|[\.!?;…\r\n]+(?:\s\|$))+', # Spanish ``` those pattern will cause 'ReDos' security problem, proof of code like below ``` import re p = re.compile(r'(?:[¡¿]+\|[\.!?;…\r\n]+(?:\s\|$))+') re.findall(p, "?"100000+"x") ``` run the above code, cpu utilization will be 100% in a very long period. more detail about 'ReDos' please see [owasp](https://owasp.org/www-community/attacks/Regular_expression_Denial_of_Service_-_ReDoS). #### effect of this security problem some api will call the pattern，like below ``` from dateparser.search import search_dates search_dates("?"100000+"x") ```	Hi @leveryd I tried running `%time search_dates("x"100000+"x")` (note that the first character is an "x") and I got: ``` CPU times: user 3min 27s, sys: 6.04 s, total: 3min 33s Wall time: 3 min 38s ``` The CPU was at 99% :sweat_smile: but it didn't hang :stuck_out_tongue:. Then I tried with your example, changing the first character to "?": `%time search_dates("?"100000+"x")`. And this is what I found: ``` CPU times: user 9min 23s, sys 9.61 s, total: 9 min 32s Wall time: 9min 37s ``` So it seems that it's definitely having some impact, but it's not so critical as I initially thought. After this I introduced the mentioned regex to this reDoS checker: http://redos-checker.surge.sh/ and it wasn't detected as a vulnerable regex (though it could be wrong). Before finishing doing my tests, I tried this: `%time search_dates("?"10000+"x")` (4 zeroes) and I got: ``` Wall time: 4.54 s ``` and then I tried switching from "?" to "x" (`%time search_dates("x"10000+"x")`) and I got this: ``` Wall time: 20 s ``` That's definitely suspicious, and I'm not sure if it's a vulnerable regex or not... I would appreciate any help from people with more experience with reDoS. regexploit also reports this as exponential: ``` Redos(starriness=11, prefix_sequence=SEQ{ [27:'] }, redos_sequence=SEQ{ [21:!,2026,0a,0d,2e:.,3b:;,3f:?]{1+}{1+} [27:'] }, repeated_character=[21:!,2026,0a,0d,2e:.,3b:;,3f:?], killer=None) Worst-case complexity: 11 ⭐⭐⭐⭐⭐⭐⭐⭐⭐⭐ (exponential) Repeated character: [21:!,2026,0a,0d,2e:.,3b:;,3f:?] Example: '\'' + '!' * 3456 ``` btw: snyk reports this as vulnerability of this library: https://security.snyk.io/vuln/SNYK-PYTHON-DATEPARSER-1063229 Unfortunately my organization's new security policies have bumped into that SNYK vulnerability where we have a couple of projects that use this library. I won't claim to be a regexpert, but it seems like a quick fix might be to limit the quantifiers to a range one could "reasonably" expect as a sentence separator. E.g., instead of: `(?:[¡¿]+\|[\.!?;…\r\n]+(?:\s\|$))+` Something like: `(?:[¡¿]{1,50}\|[\.!?;…\r\n]{1,50}(?:\s\|$))+` (I can't imagine anyone using more than 50 punctuation marks to end a sentence but just having that limit seems to cut the processing time significantly) Would that be a satisfactory resolution to get the SNYK report closed? Also, regexploit no longer seems to report the original regex as vulnerable, I'm not sure what changed there. Any updates on that?	2022-10-12T10:31:07	[]	[]
scrapinghub/dateparser	scrapinghub__dateparser-1028	4490ca6de44f2e9771e2017e0695f7d9f2d7207e	diff --git a/dateparser/date.py b/dateparser/date.py index ac8283ba2..1a91764e4 100644 --- a/dateparser/date.py +++ b/dateparser/date.py @@ -35,7 +35,7 @@ RE_SANITIZE_SKIP = re.compile(r'\t\|\n\|\r\|\u00bb\|,\s\u0432\b\|\u200e\|\xb7\|\u200f\|\u064e\|\u064f', flags=re.M) RE_SANITIZE_RUSSIAN = re.compile(r'([\W\d])\u0433\.', flags=re.I \| re.U) -RE_SANITIZE_PERIOD = re.compile(r'(?<=\D+)\.', flags=re.U) +RE_SANITIZE_PERIOD = re.compile(r'(?<=[^0-9\s])\.', flags=re.U) RE_SANITIZE_ON = re.compile(r'^.?on:\s+(.)') RE_SANITIZE_APOSTROPHE = re.compile('\|'.join(APOSTROPHE_LOOK_ALIKE_CHARS)) diff --git a/dateparser/parser.py b/dateparser/parser.py index e2df66b03..30536cab7 100644 --- a/dateparser/parser.py +++ b/dateparser/parser.py @@ -223,7 +223,7 @@ class _parser: def __init__(self, tokens, settings): self.settings = settings - self.tokens = list(tokens) + self.tokens = [(t[0].strip(), t[1]) for t in list(tokens)] self.filtered_tokens = [(t[0], t[1], i) for i, t in enumerate(self.tokens) if t[1] <= 1] self.unset_tokens = []	Regression dates with dots and spaces Since upgrading from `0.7.6` to `1.0.0` or `1.1.0`, dates with dots and spaces are not parsed correctly: ```python from datetime import datetime from dateparser import parse # both assert pass in `0.7.6` but not in `1.0.0` or `1.1.0` assert parse("26. 10.21") == datetime(2021, 10, 26) assert parse("26. 10.21", settings={"STRICT_PARSING": True}) == datetime(2021, 10, 26) ``` Is this intended behavior or a bug with the new version? This might be related to https://github.com/scrapinghub/dateparser/pull/741 Thanks!		2021-12-18T16:25:51	[]	[]
scrapinghub/dateparser	scrapinghub__dateparser-1005	8af4f296a41e5ca9b2e3cde7deccb26a0652b70a	diff --git a/dateparser/data/date_translation_data/en.py b/dateparser/data/date_translation_data/en.py index 62e2649c2..a156bf899 100644 --- a/dateparser/data/date_translation_data/en.py +++ b/dateparser/data/date_translation_data/en.py @@ -126,7 +126,8 @@ ], "relative-type": { "0 day ago": [ - "today" + "today", + "till date" ], "0 hour ago": [ "this hour" diff --git a/dateparser_data/supplementary_language_data/date_translation_data/en.yaml b/dateparser_data/supplementary_language_data/date_translation_data/en.yaml index 600cf325c..b21e70b14 100644 --- a/dateparser_data/supplementary_language_data/date_translation_data/en.yaml +++ b/dateparser_data/supplementary_language_data/date_translation_data/en.yaml @@ -47,6 +47,8 @@ relative-type: - this decade in 1 decade: - next decade + 0 day ago: + - till date relative-type-regex: in \1 decade:	Is the string 'tilldate' supported in the parser? Hi! Is the string `tilldate`/`till date` supported? This is what i've tried: ``` >>> from dateparser import parse, search >>> parsed_date = parse('tilldate') >>> print(parsed_date) None >>> parsed_date = parse('till date') >>> print(parsed_date) None >>> parsed_date = search.search_dates('tilldate') >>> print(parsed_date) None >>> parsed_date = search.search_dates('till date') >>> print(parsed_date) None ``` It returns `None` but i was expecting it to return `datetime.datetime(2021, 1, 6, 0, 0)` instead. Thanks!	Looks like a valid expression: https://english.stackexchange.com/a/146771/87187 Not sure if we support any of the alternative expressions, but I guess it makes sense to eventually support it. Hi @Gallaecio, thanks for a really quick response. Good to know that the expression would be supported in future. Is the project open for contributions regarding this? Cheers! I _think_ so, specially if it’s doable by modifying the corresponding YAML file. See https://dateparser.readthedocs.io/en/latest/contributing.html#guidelines-for-editing-translation-data	2021-10-23T16:29:27	[]	[]
oeg-upm/yatter	oeg-upm__yatter-85	7421fc9fc07f38cbcc2a2730a3110bcb6c747c14	diff --git a/VERSION b/VERSION index e25d8d9..ea710ab 100644 --- a/VERSION +++ b/VERSION @@ -1,1 +1,1 @@ -1.1.5 +1.2 \ No newline at end of file diff --git a/src/yatter/__init__.py b/src/yatter/__init__.py index f586a89..e32272d 100644 --- a/src/yatter/__init__.py +++ b/src/yatter/__init__.py @@ -17,13 +17,14 @@ def translate(yarrrml_data, mapping_format=RML_URI): rml_mapping.extend(generate_database_connections(yarrrml_data, list_initial_sources)) rml_mapping.extend(add_logical_targets(yarrrml_data)) rml_mapping.extend(add_functions(yarrrml_data)) + external_refs = [] try: mappings, mapping_format = get_non_asserted_mappings(yarrrml_data, mapping_format) for mapping in yarrrml_data.get(YARRRML_MAPPINGS): if mapping_format == R2RML_URI: source_list = add_table(yarrrml_data, mapping, list_initial_sources) else: - source_list = add_source(yarrrml_data, mapping, list_initial_sources) + source_list, external_refs = add_source(yarrrml_data, mapping, list_initial_sources) subject_list = add_subject(yarrrml_data, mapping, mapping_format) pred = add_predicate_object_maps(yarrrml_data, mapping, mapping_format) it = 0 @@ -37,6 +38,9 @@ def translate(yarrrml_data, mapping_format=RML_URI): rml_mapping[len(rml_mapping) - 1] = rml_mapping[len(rml_mapping) - 1][:-2] rml_mapping.append(".\n\n\n") it = it + 1 + external_refs = list(dict.fromkeys(external_refs)) + for ref in external_refs: + rml_mapping.append(ref) logger.info("RML content is created!") rml_mapping_string = "".join(rml_mapping) diff --git a/src/yatter/constants.py b/src/yatter/constants.py index c5aad0a..6f363dd 100644 --- a/src/yatter/constants.py +++ b/src/yatter/constants.py @@ -21,6 +21,7 @@ VOID_URI = 'http://rdfs.org/ns/void#' FNML_URI = 'http://semweb.mmlab.be/ns/fnml#' GREL_URI = 'http://users.ugent.be/~bjdmeest/function/grel.ttl#' +SD_URI = 'https://w3id.org/okn/o/sd/' RML_PREFIX = '@prefix' RML_BASE = '@base' @@ -28,6 +29,7 @@ RML_LOGICAL_SOURCE = 'rml:logicalSource' RML_SOURCE = 'rml:source' RML_REFERENCE_FORMULATION = 'rml:referenceFormulation' +RML_REFERENCE_FORMULATION_CLASS = 'rml:ReferenceFormulation' RML_ITERATOR = 'rml:iterator' RML_REFERENCE = 'rml:reference' RML_LANGUAGE_MAP = 'rml:languageMap' @@ -103,7 +105,16 @@ D2RQ_USER = 'd2rq:username' D2RQ_PASS = 'd2rq:password' - +############################################################################## +############################# SD CONSTANTS ########################### +############################################################################## +SD_DATASET_SPEC = 'sd:DatasetSpecification' +SD_NAME = 'sd:name' +SD_HAS_DATA_TRANSFORMATION = 'sd:hasDataTransformation' +SD_HAS_SOFTWARE_REQUIREMENTS = 'sd:hasSoftwareRequirements' +SD_HAS_SOURCE_CODE= 'sd:hasSourceCode' +SD_PROGRAMMING_LANGUAGE = 'sd:programmingLanguage' +KG4DI_DEFINED_BY = 'kg4di:definedBy' ############################################################################## ############################# YARRRML CONSTANTS ########################### ############################################################################## @@ -122,6 +133,11 @@ YARRRML_USERNAME = 'username' YARRRML_PASSWORD = 'password' +YARRRML_STRUCTURE_DEFINER = 'structureDefiner' +YARRRML_SOFTWARE_SPECIFICATION = 'softwareSpecification' +YARRRML_PROGRAMMING_LANGUAGE = 'programmingLanguage' +YARRRML_SOFTWARE_REQUIREMENTS = 'softwareRequirements' + YARRRML_MAPPINGS = 'mappings' # used for mappings in conditions and mappings main key YARRRML_MAPPING = 'mapping' YARRRRL_MAPPINGS_SHORTCUT = 'm' diff --git a/src/yatter/predicateobject.py b/src/yatter/predicateobject.py index bc2434b..0e2a059 100644 --- a/src/yatter/predicateobject.py +++ b/src/yatter/predicateobject.py @@ -288,7 +288,7 @@ def ref_mapping(data, mapping, om, yarrrml_key, ref_type_property, mapping_forma else: if mapping_format == STAR_URI: object = STAR_OBJECT - source_list = add_source(data, mapping_join, list_initial_sources) + source_list, external_refs = add_source(data, mapping_join, list_initial_sources) number_joins_rml = len(subject_list) * len(source_list) for i in range(number_joins_rml): diff --git a/src/yatter/source.py b/src/yatter/source.py index f5fb009..222acd5 100644 --- a/src/yatter/source.py +++ b/src/yatter/source.py @@ -30,6 +30,7 @@ def add_source(data, mapping, list_initial_sources): source_template = "\t" + RML_LOGICAL_SOURCE + " [\n\t\ta " + RML_LOGICAL_SOURCE_CLASS + \ ";\n\t\t" + RML_SOURCE + " " final_list = [] + external_references_list = [] sources = get_sources(data, mapping) for source in sources: db_identifier = mapping @@ -40,13 +41,24 @@ def add_source(data, mapping, list_initial_sources): if YARRRML_ACCESS in source: if YARRRML_QUERY in source: final_list.append(source_template + database_source(mapping, source, db_identifier)) + elif YARRRML_STRUCTURE_DEFINER in source: + source, external_references = add_in_memory_source(mapping,source) + final_list.append(source_template + source) + if external_references is not None: + external_references_list.append(external_references) else: final_list.append(source_template + add_source_full(mapping, source)) elif type(source) is list: - final_list.append(source_template + add_source_simplified(mapping, source)) + if "$(" in source[0]: + source, external_references = add_in_memory_source(mapping, source) + final_list.append(source_template + source) + if external_references is not None: + external_references_list.append(external_references) + else: + final_list.append(source_template + add_source_simplified(mapping, source)) else: raise Exception("ERROR: source " + source + " in mapping " + mapping + " not valid") - return final_list + return final_list, external_references_list def add_table(data, mapping, list_initial_sources): @@ -105,6 +117,53 @@ def add_source_simplified(mapping, source): + source[1] + "\";\n\t];\n" return source_rdf +def add_in_memory_source(mapping, source): + external_reference_formulation = None + if type(source) is list: + source = extend_in_memory(source) + source_rdf = "[\n\t\t\ta " + SD_DATASET_SPEC + ";\n\t\t\t" + + access = str(source.get(YARRRML_ACCESS)).replace("$(","").replace(")","") + source_rdf += SD_NAME + " \"" + access + "\";\n" + + if YARRRML_SOFTWARE_SPECIFICATION in source: + source_rdf += "\t\t\t" + SD_HAS_DATA_TRANSFORMATION + "[\n\t\t\t\t" + + if YARRRML_SOFTWARE_REQUIREMENTS in source.get(YARRRML_SOFTWARE_SPECIFICATION): + software_requirements = str(source.get(YARRRML_SOFTWARE_SPECIFICATION)[YARRRML_SOFTWARE_REQUIREMENTS]) + source_rdf += SD_HAS_SOFTWARE_REQUIREMENTS + " \""+ software_requirements +"\";\n\t\t\t\t" + + if YARRRML_PROGRAMMING_LANGUAGE in source.get(YARRRML_SOFTWARE_SPECIFICATION): + programming_language = str(source.get(YARRRML_SOFTWARE_SPECIFICATION)[YARRRML_PROGRAMMING_LANGUAGE]) + source_rdf += SD_HAS_SOURCE_CODE +"[\n\t\t\t\t\t" + SD_PROGRAMMING_LANGUAGE + " \"" + programming_language + "\";" + source_rdf +="\n\t\t\t\t];\n" + + source_rdf += "\t\t\t];\n" + source_rdf += "\t\t];\n" + + if YARRRML_ITERATOR in source: + source_rdf += "\t\t" + RML_ITERATOR + " \"" + source.get(YARRRML_ITERATOR) + "\";\n" + + if YARRRML_REFERENCE_FORMULATION in source: + reference_formulation = str(source.get(YARRRML_REFERENCE_FORMULATION)) + source_rdf += "\t\t" + RML_REFERENCE_FORMULATION + " ql:"+ reference_formulation + ";\n" + external_reference_formulation = "ql:" + reference_formulation + " a " + RML_REFERENCE_FORMULATION_CLASS +";\n" + external_reference_formulation +="\t" + KG4DI_DEFINED_BY +" \""+source.get(YARRRML_STRUCTURE_DEFINER) +"\"." + source_rdf += "\t];\n" + + + + return source_rdf, external_reference_formulation + +def extend_in_memory(source): + features = source[0].split("~") + access = features[0] + defined_by = features[1].split("-")[0] + reference_formulation = features[1].split("-")[1] + extended_source = {"access": access, "structureDefiner": defined_by, "referenceFormulation": reference_formulation} + if len(source) == 2: + extended_source["iterator"] = source[1] + return extended_source def add_source_full(mapping, source): source_rdf = ""	In-memory YARRRML to RML support (new-feature) Support of in-memory extension YARRRML syntax to RML.	Test-cases for testing the solution are added by @TheRazorace in #67	2024-10-29T10:46:57	[]	[]
oeg-upm/yatter	oeg-upm__yatter-84	08a6e3db303122671e102ba560e12b67ce5568bc	diff --git a/src/yatter/__main__.py b/src/yatter/__main__.py index 4b71d8c..5cb9e92 100644 --- a/src/yatter/__main__.py +++ b/src/yatter/__main__.py @@ -5,8 +5,8 @@ from . import translate, inverse_translation, merge_mappings from .constants import * -def write_results(mapping): +def write_results(mapping): if type(mapping) is str: output_file = open(args.output_mapping_path, "w") output_file.write(mapping) @@ -18,6 +18,7 @@ def write_results(mapping): yaml.default_flow_style = False yaml.dump(mapping, f) + def parse_inputs(): input_format = RML_URI yaml = YAML(typ='safe', pure=True) @@ -25,7 +26,8 @@ def parse_inputs(): if args.input_mapping_path.endswith('.yml') or args.input_mapping_path.endswith('.yaml'): with open(args.input_mapping_path) as f: input_data = yaml.load(f) - elif args.input_mapping_path.endswith('.ttl') or args.input_mapping_path.endswith('.rml') or args.input_mapping_path.endswith('.r2rml'): + elif args.input_mapping_path.endswith('.ttl') or args.input_mapping_path.endswith( + '.rml') or args.input_mapping_path.endswith('.r2rml'): input_data = Graph() input_data.parse(args.input_mapping_path, format="turtle") else: diff --git a/src/yatter/predicateobject.py b/src/yatter/predicateobject.py index 7ddb90f..bc2434b 100644 --- a/src/yatter/predicateobject.py +++ b/src/yatter/predicateobject.py @@ -372,15 +372,13 @@ def add_inverse_pom(mapping_id, rdf_mapping, classes, prefixes): logger.error("ERROR: There is POM without predicate map defined") raise Exception("Review your mapping " + str(mapping_id)) - prefix = list({i for i in prefixes if predicate.startswith(prefixes[i])}) - if not predicate.startswith("http") and "{" not in predicate: + if not predicate.startswith("http"): predicate = '$(' + predicate + ')' - elif "{" in predicate and "}" in predicate: + elif predicate.startswith("http") and "{" not in predicate: + predicate = find_prefixes(predicate, prefixes) + else: predicate = predicate.replace('{', '$(').replace('}', ')') - elif prefix: - predicate = tm['predicateValue'].toPython().replace(prefixes[prefix[0]], prefix[0] + ":") - predicate = find_prefixes(predicate,prefixes) if tm['parentTriplesMap']: if tm['child']: diff --git a/src/yatter/subject.py b/src/yatter/subject.py index 3b28037..9934c85 100644 --- a/src/yatter/subject.py +++ b/src/yatter/subject.py @@ -54,7 +54,7 @@ def add_subject(data, mapping, mapping_format): if YARRRML_TARGETS in individual_subject: subject_termmap = subject_termmap[0:-3]+"\t"+RML_LOGICAL_TARGET+" <"+individual_subject[YARRRML_TARGETS]+">\n\t];\n" - if YARRRML_TYPE in individual_subject: + if isinstance(individual_subject, dict) and YARRRML_TYPE in individual_subject: if individual_subject.get(YARRRML_TYPE) == YARRRML_BLANK: subject_termmap = subject_termmap[0:-3] + "\t" + R2RML_TERMTYPE + " " + R2RML_BLANK_NODE +"\n\t];\n"	Package inconsistency between versions. Error: 'str' object has no attribute 'get' in yatter 1.1.4 The following YARRRML excerpt does work with Yatter v1.1.0 (Python 3.9), but does not work with Yatter v1.1.4 (tested with Python 3.9 and Python 3.12), even though the syntax is the same and seems correct, according to documentation. Error: The YARRRML mapping has not been translated. 'str' object has no attribute 'get' ``` prefixes: mlso: http://w3id.org/mlso/ mls: http://www.w3.org/ns/mls# rdfs: http://www.w3.org/2000/01/rdf-schema# sources: evaluations_meta: ['/evaluations_meta.json~jsonpath', '$.'] mappings: evaluationMeasureDatatype: sources: - evaluations_meta s: http://w3id.org/rexpek-kg/thermal_uc_v1/evaluationMeasure/$(name)/characteristic/datatype po: - [a, mlso:EvaluationMeasureCharacteristic] - [rdfs:label, Datatype, rdfs:Literal] - [mls:hasValue, $(datatype), rdfs:Literal] ``` Package inconsistency between versions. Error: 'str' object has no attribute 'get' in yatter 1.1.4 The following YARRRML excerpt does work with Yatter v1.1.0 (Python 3.9), but does not work with Yatter v1.1.4 (tested with Python 3.9 and Python 3.12), even though the syntax is the same and seems correct, according to documentation. Error: The YARRRML mapping has not been translated. 'str' object has no attribute 'get' ``` prefixes: mlso: http://w3id.org/mlso/ mls: http://www.w3.org/ns/mls# rdfs: http://www.w3.org/2000/01/rdf-schema# sources: evaluations_meta: ['/evaluations_meta.json~jsonpath', '$.'] mappings: evaluationMeasureDatatype: sources: - evaluations_meta s: http://w3id.org/rexpek-kg/thermal_uc_v1/evaluationMeasure/$(name)/characteristic/datatype po: - [a, mlso:EvaluationMeasureCharacteristic] - [rdfs:label, Datatype, rdfs:Literal] - [mls:hasValue, $(datatype), rdfs:Literal] ```		2024-08-29T11:20:26	[]	[]
oeg-upm/yatter	oeg-upm__yatter-73	75b4e3ae0cb850f3fa261c1ac99d0584ea06892d	diff --git a/VERSION b/VERSION index 524cb55..45a1b3f 100644 --- a/VERSION +++ b/VERSION @@ -1,1 +1,1 @@ -1.1.1 +1.1.2 diff --git a/src/yatter/mapping.py b/src/yatter/mapping.py index 1edb0a9..4a0fa71 100644 --- a/src/yatter/mapping.py +++ b/src/yatter/mapping.py @@ -1,7 +1,6 @@ from .import * prefixes = {} - def add_mapping(mapping, mappings, it): map_template = "<" + mapping + "_" + str(it) + "> a " if mappings[mapping]: @@ -99,7 +98,7 @@ def add_inverse_prefix(rdf_mapping): def get_non_asserted_mappings(yarrrml_data, mapping_format): - mappings = dict.fromkeys(list(yarrrml_data.get(YARRRML_MAPPINGS).keys())) + mappings = dict.fromkeys(list(yarrrml_data.get(YARRRML_MAPPINGS).keys())) for mapping in yarrrml_data.get(YARRRML_MAPPINGS): keys = yarrrml_data.get(YARRRML_MAPPINGS).get(mapping).keys() for key in keys: diff --git a/src/yatter/predicateobject.py b/src/yatter/predicateobject.py index d8fa5e8..c2d9358 100644 --- a/src/yatter/predicateobject.py +++ b/src/yatter/predicateobject.py @@ -150,7 +150,7 @@ def add_predicate_object(data, mapping, predicate_object, mapping_format=RML_URI execution = True template += generate_rml_termmap(R2RML_PREDICATE, R2RML_PREDICATE_CLASS, pm_value,"\t\t\t") if execution: - template = template.replace(R2RML_CONSTANT + " " + pm_value, RML_EXECUTION + " <" + pm_value + ">") + template = template.replace(R2RML_CONSTANT + " \"" + pm_value + "\"", RML_EXECUTION + " <" + pm_value + ">") if YARRRML_TARGETS in pm: template = template[0:-3] + "\t" + RML_LOGICAL_TARGET + " <" + pm[YARRRML_TARGETS] + ">\n\t\t];\n" @@ -225,7 +225,7 @@ def add_predicate_object(data, mapping, predicate_object, mapping_format=RML_URI object_value, "\t\t\t", mapping_format) elif YARRRML_FUNCTION in om: template += generate_rml_termmap(R2RML_OBJECT, R2RML_OBJECT_CLASS, om[YARRRML_FUNCTION], "\t\t\t", mapping_format) - template = template.replace(R2RML_CONSTANT+" "+om[YARRRML_FUNCTION], RML_EXECUTION + " <" + om.get( + template = template.replace(R2RML_CONSTANT+" \""+om[YARRRML_FUNCTION]+ "\"", RML_EXECUTION + " <" + om.get( YARRRML_FUNCTION) + ">") else: template += generate_rml_termmap(R2RML_OBJECT, R2RML_OBJECT_CLASS, diff --git a/src/yatter/subject.py b/src/yatter/subject.py index d77691d..3b28037 100644 --- a/src/yatter/subject.py +++ b/src/yatter/subject.py @@ -48,7 +48,7 @@ def add_subject(data, mapping, mapping_format): subject_termmap = generate_rml_termmap(R2RML_SUBJECT, R2RML_SUBJECT_CLASS, subject_value, "\t\t", mapping_format) if YARRRML_FUNCTION in individual_subject: - subject_termmap = subject_termmap.replace(R2RML_CONSTANT+" "+subject_value, RML_EXECUTION + " <" + individual_subject.get( + subject_termmap = subject_termmap.replace(R2RML_CONSTANT+" \""+subject_value + "\"", RML_EXECUTION + " <" + individual_subject.get( YARRRML_FUNCTION) + ">") if YARRRML_TARGETS in individual_subject: diff --git a/src/yatter/termmap.py b/src/yatter/termmap.py index ee2334d..8693fe1 100644 --- a/src/yatter/termmap.py +++ b/src/yatter/termmap.py @@ -19,11 +19,16 @@ def get_termmap_type(text, mapping_format): ## Generates a TermMap (subject, predicate, object) based on the property, class and the text def generate_rml_termmap(rml_property, rml_class, text, indentation, mapping_format=RML_URI): - template = indentation[0:-1] + rml_property + " [\n"+indentation+"a " + rml_class + ";\n" + indentation + from .mapping import prefixes + template = indentation[0:-1] + rml_property + " [\n" + indentation + "a " + rml_class + ";\n" + indentation term_map = get_termmap_type(text, mapping_format) if term_map == R2RML_TEMPLATE: text = generate_rml_template(text) text = text.replace('"', r'\"') + if ":" in text: + text_prefix_split = text.split(":") + if text_prefix_split[0] in prefixes: + text = prefixes[text_prefix_split[0]] + text_prefix_split[1] elif term_map == RML_REFERENCE or term_map == R2RML_COLUMN: text = text.replace("$(", "").replace(")", "") text = text.replace('"', r'\"') @@ -41,7 +46,10 @@ def generate_rml_termmap(rml_property, rml_class, text, indentation, mapping_for if text.startswith("http"): template += term_map + " <" + text + ">;\n" + indentation[0:-1] + "];\n" else: - template += term_map + " " + text + ";\n"+indentation[0:-1]+"];\n" + if ":" in text or "<" in text: + template += term_map + " " + text + ";\n"+indentation[0:-1]+"];\n" + else: + template += term_map + " \"" + text + "\";\n" + indentation[0:-1] + "];\n" return template	Resolve prefixes in the generation of SubjectMap Describe the bug Considering a SubjectMap `s: ex:person/$(firstname)` and a prefix `ex`, the prefix is not resolved when generating the SubjectMap. To Reproduce Use this mapping file ``` prefixes: ex: "http://example.com/" mappings: person: sources: - ['persons.json~jsonpath', '$.persons[]'] s: ex:person/$(firstname) po: - [a, foaf:Person] ``` Expected behavior* In the SubjectMap generated the `ex:` is resolved as `http://example.com/`. ``` rr:subjectMap [ a rr:SubjectMap; rr:template "http://example.com/person/{firstname}"; ]; ``` problem translating functions? For the following YARRRML document: ``` prefixes: ex: "http://www.semanticweb.org/tijnd/" ontology: "http://www.semanticweb.org/tijnd/ontology/" schema: "https://schema.org/" idlab-fn: "http://example.com/idlab/function/" grel: "http://users.ugent.be/~bjdmeest/function/grel.ttl#" owl: "http://www.w3.org/2002/07/owl#" rdf: "http://www.w3.org/1999/02/22-rdf-syntax-ns#" rdfs: "http://www.w3.org/2000/01/rdf-schema#" xsd: "http://www.w3.org/2001/XMLSchema#" sources: imdb: access: imdb_video_game_rating_modified.csv referenceFormulation: csv mappings: director: sources: - imdb s: ex:Person/$(title) #Not really the correct subject, only used to simply test it po: - [a, schema:Person] - p: schema:name o: function: grel:string_split parameters: - [grel:p_string_sep, ", "] - [grel:valueParameter, $(directors)] ``` Yatter has problems using the function, mainly with the ", " it gives an error "objectlist expected." The same YARRRML document is properly translated with Matey. The dataset considered for this task is this: https://www.kaggle.com/datasets/nyagami/video-game-ratings-from-imdb (if it matters)	Sorry for the late fix, but now prefixes are substituted in any TermMap with reference and prefixes. This is already solved! Thanks :)	2023-12-19T12:19:13	[]	[]
oeg-upm/yatter	oeg-upm__yatter-70	da6e9c02e810619c58b95071a1be3e6122a90625	diff --git a/VERSION b/VERSION index 9084fa2..524cb55 100644 --- a/VERSION +++ b/VERSION @@ -1,1 +1,1 @@ -1.1.0 +1.1.1 diff --git a/src/yatter/constants.py b/src/yatter/constants.py index df7387a..08d2c9e 100644 --- a/src/yatter/constants.py +++ b/src/yatter/constants.py @@ -81,6 +81,7 @@ R2RML_TERMTYPE = 'rr:termType' R2RML_LANGUAGE = 'rr:language' R2RML_IRI = 'rr:IRI' +R2RML_BLANK_NODE = 'rr:BlankNode' R2RML_LITERAL = 'rr:Literal' R2RML_REFOBJECT_CLASS = 'rr:RefObjectMap' R2RML_PARENT_TRIPLESMAP = 'rr:parentTriplesMap' @@ -92,6 +93,7 @@ R2RML_TABLE_NAME = 'rr:tableName' R2RML_COLUMN = 'rr:column' + ############################################################################## ############################# D2RQ CONSTANTS ########################### ############################################################################## @@ -154,6 +156,7 @@ YARRRML_IRI = '~iri' YARRRML_LANG = '~lang' +YARRRML_BLANK = 'blank' YARRRML_QUOTED = 'quoted' YARRRML_NON_ASSERTED = 'quotedNonAsserted' diff --git a/src/yatter/predicateobject.py b/src/yatter/predicateobject.py index 4972923..d8fa5e8 100644 --- a/src/yatter/predicateobject.py +++ b/src/yatter/predicateobject.py @@ -156,11 +156,15 @@ def add_predicate_object(data, mapping, predicate_object, mapping_format=RML_URI for om in object_list: iri = False + blank = False if type(om) == list: object_value = om[0] if YARRRML_IRI in om[0]: - object_value = om[0].split(YARRRML_IRI)[0] + object_value = om[0].split("~")[0] iri = True + if YARRRML_BLANK in om[0]: + object_value = om[0].split("~")[0] + blank = True if mapping_format == STAR_URI: template += generate_rml_termmap(STAR_OBJECT, R2RML_OBJECT_CLASS, object_value, "\t\t\t", mapping_format) @@ -195,6 +199,9 @@ def add_predicate_object(data, mapping, predicate_object, mapping_format=RML_URI if iri: template = template[0:len(template) - 5] + "\t\t\t" + R2RML_TERMTYPE + " " \ + R2RML_IRI + "\n\t\t];\n" + if blank: + template = template[0:len(template) - 5] + "\t\t\t" + R2RML_TERMTYPE + " " \ + + R2RML_BLANK_NODE + "\n\t\t];\n" elif YARRRML_MAPPING in om or YARRRML_NON_ASSERTED in om or YARRRML_QUOTED in om: if YARRRML_MAPPING in om: template += ref_mapping(data, mapping, om, YARRRML_MAPPING, R2RML_PARENT_TRIPLESMAP, mapping_format) @@ -236,6 +243,9 @@ def add_predicate_object(data, mapping, predicate_object, mapping_format=RML_URI elif om.get(YARRRML_TYPE) == "literal": template = template[0:len(template) - 5] + "\t\t\t" + R2RML_TERMTYPE + " " \ + R2RML_LITERAL + "\n\t\t];\n" + elif om.get(YARRRML_TYPE) == YARRRML_BLANK: + template = template[0:len(template) - 5] + "\t\t\t" + R2RML_TERMTYPE + " " \ + + R2RML_BLANK_NODE + "\n\t\t];\n" if YARRRML_TARGETS in om: template = template[0:len(template) - 5] + "\t\t\t" + RML_LOGICAL_TARGET + " <"+ om.get(YARRRML_TARGETS) + ">\n\t\t];\n" diff --git a/src/yatter/subject.py b/src/yatter/subject.py index cb92703..d77691d 100644 --- a/src/yatter/subject.py +++ b/src/yatter/subject.py @@ -54,6 +54,10 @@ def add_subject(data, mapping, mapping_format): if YARRRML_TARGETS in individual_subject: subject_termmap = subject_termmap[0:-3]+"\t"+RML_LOGICAL_TARGET+" <"+individual_subject[YARRRML_TARGETS]+">\n\t];\n" + if YARRRML_TYPE in individual_subject: + if individual_subject.get(YARRRML_TYPE) == YARRRML_BLANK: + subject_termmap = subject_termmap[0:-3] + "\t" + R2RML_TERMTYPE + " " + R2RML_BLANK_NODE +"\n\t];\n" + rml_subjects.append(subject_termmap) if YARRRML_GRAPHS in data.get(YARRRML_MAPPINGS).get(mapping): diff --git a/src/yatter/termmap.py b/src/yatter/termmap.py index a522707..ee2334d 100644 --- a/src/yatter/termmap.py +++ b/src/yatter/termmap.py @@ -18,8 +18,8 @@ def get_termmap_type(text, mapping_format): ## Generates a TermMap (subject, predicate, object) based on the property, class and the text -def generate_rml_termmap(rml_property, rml_class, text, identation, mapping_format=RML_URI): - template = identation[0:-1] + rml_property + " [\n"+identation+"a " + rml_class + ";\n" + identation +def generate_rml_termmap(rml_property, rml_class, text, indentation, mapping_format=RML_URI): + template = indentation[0:-1] + rml_property + " [\n"+indentation+"a " + rml_class + ";\n" + indentation term_map = get_termmap_type(text, mapping_format) if term_map == R2RML_TEMPLATE: text = generate_rml_template(text) @@ -32,16 +32,16 @@ def generate_rml_termmap(rml_property, rml_class, text, identation, mapping_form if term_map == STAR_QUOTED: if 'quoted' in text: - template += term_map + " <" + text[YARRRML_QUOTED] + "_0>;\n" + identation[0:-1] + "];\n" + template += term_map + " <" + text[YARRRML_QUOTED] + "_0>;\n" + indentation[0:-1] + "];\n" else: - template += term_map + " <" + text[YARRRML_NON_ASSERTED] + "_0>;\n" + identation[0:-1] + "];\n" + template += term_map + " <" + text[YARRRML_NON_ASSERTED] + "_0>;\n" + indentation[0:-1] + "];\n" elif term_map != "rr:constant": - template += term_map + " \"" + text + "\";\n"+identation[0:-1]+"];\n" + template += term_map + " \"" + text + "\";\n"+indentation[0:-1]+"];\n" else: if text.startswith("http"): - template += term_map + " <" + text + ">;\n" + identation[0:-1] + "];\n" + template += term_map + " <" + text + ">;\n" + indentation[0:-1] + "];\n" else: - template += term_map + " " + text + ";\n"+identation[0:-1]+"];\n" + template += term_map + " " + text + ";\n"+indentation[0:-1]+"];\n" return template	Blank Nodes in YARRRML trigger TypeError: argument of type 'NoneType' is not iterable Describe the bug Following the [YARRRML subjects spec](https://rml.io/yarrrml/spec/#subjects), I use blank nodes with `s: null` . This is not supported by yatter To Reproduce Steps to reproduce the behavior (and resources): with the files below, run the yatter library `python3 -m yatter -i yarrrml/yatter-test.yml -o rml/yatter-test_rml.ttl` yatter-test.yml ``` # Define namespace prefixes for commonly used ontologies prefixes: exp: "http://example.org/participant/" ex: "http://example.org/#" rdf: "http://www.w3.org/1999/02/22-rdf-syntax-ns#" rdfs: "http://www.w3.org/2000/01/rdf-schema#" xsd: "http://www.w3.org/2001/XMLSchema#" dcterms: "http://purl.org/dc/terms/" locn: "http://www.w3.org/ns/locn#" grel: "http://users.ugent.be/~bjdmeest/function/grel.ttl#" skos: "http://www.w3.org/2004/02/skos/core#" wegwijs: "https://data.vlaanderen.be/id/organisatie/" adms: "http://www.w3.org/ns/adms#" # Define the mapping rules mappings: # Mapping for 'participants' data participants: # Define sources to extract data from JSON sources: - ['yatter-test.csv~csv'] # Define subject mapping s: exp:$(participantUuid) # Define predicate-object mappings po: - [a, ex:Participant] - [dcterms:created, $(created), xsd:dateTime] - [dcterms:modified, $(modified), xsd:dateTime] - p: adms:identifier o: type: blank mapping: identificator condition: function: equal parameters: - [str1, $(participantUuid), s] - [str2, $(participantUuid), o] # Mapping for 'identificator' data identificator: sources: - ['yatter-test.csv~csv'] # Define subject as null, indicating a blank node will be generated s: null # Define predicate-object mappings po: - [a, adms:Identifier] - [skos:notation, exp:$(participantUuid), xsd:anyURI] - [dcterms:creator, wegwijs:OVO026559~iri] ``` yatter-test.csv ``` participantUuid,created,modified,birthYear,placeHolderStatSect 493b78b8-efa7-45b8-af5f-ca08ec90179e,2023-06-06T15:17:32.954+0000,2023-06-06T15:19:14.358+0000,1964,Wilrijk ea9ba26f-e283-42a2-b57a-55ea3f5571ff,2022-12-28T14:46:03.682+0000,2022-12-28T14:48:23.17+0000,1957,Marcinelle e0f5d6b0-984d-4192-b00f-8a07d2189712,2023-06-06T17:36:44.39+0000,2023-06-06T17:36:44.396+0000,1963,Zingem 20a15803-ca0e-42f9-9abe-93dc5d593a31,2022-12-16T11:39:39.481+0000,2022-12-16T11:43:58.073+0000,1951,Gent ``` Current behavior TypeError stacktrace ```2023-12-05 10:44:58,430 \| INFO: Translating YARRRML mapping to [R2]RML Traceback (most recent call last): File "<frozen runpy>", line 198, in _run_module_as_main File "<frozen runpy>", line 88, in _run_code File "/Library/Frameworks/Python.framework/Versions/3.11/lib/python3.11/site-packages/yatter/__main__.py", line 68, in <module> mapping_content = translate(mapping_data, mapping_format) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "/Library/Frameworks/Python.framework/Versions/3.11/lib/python3.11/site-packages/yatter/__init__.py", line 15, in translate list_initial_sources = get_initial_sources(yarrrml_data) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "/Library/Frameworks/Python.framework/Versions/3.11/lib/python3.11/site-packages/yatter/source.py", line 9, in get_initial_sources if YARRRML_SOURCES in data: ^^^^^^^^^^^^^^^^^^^^^^^ TypeError: argument of type 'NoneType' is not iterable ``` Expected behavior an RML file is produced like the one by YARRRML's matey: [output.rml.ttl.txt](https://github.com/oeg-upm/yatter/files/13558196/output.rml.ttl.txt) output turtle [output (1).txt](https://github.com/oeg-upm/yatter/files/13558264/output.1.txt)	Hi @Stiksels, As there is no consensus in the W3C CG about supporting BN without data references, I would not consider implementing this feature for now. However, I can provide support for BN with data reference on the subject. As there is no example in the spec, I will do it similarly as it's indicated for ObjectMaps, as both (OM and SM) are term maps. It will be something like: ```yaml identificator: sources: - ['yatter-test.csv~csv'] # Define subject as null, indicating a blank node will be generated s: value: $(participantUuid) # could be template, reference or constant type: blank # Define predicate-object mappings po: - [a, adms:Identifier] - [skos:notation, exp:$(participantUuid), xsd:anyURI] - [dcterms:creator, wegwijs:OVO026559~iri] ```	2023-12-05T12:18:50	[]	[]
oeg-upm/yatter	oeg-upm__yatter-45	eee599dcaf6c31abe8af7b5fa641ce9566764fda	diff --git a/README.md b/README.md index 6b3d172..218a8b3 100644 --- a/README.md +++ b/README.md @@ -17,35 +17,62 @@ We are working on including the following features which are not yet implemented pip install yarrrml-translator ``` -## Execution +## Execution from CLI + +### From YARRRML to \[R2\]RML To execute from command line run the following: ```bash -python3 -m yarrrml_translator -i path_to_input_yarrrml.yml -o path_to_output.ttl [-f R2RML] +python3 -m yarrrml_translator -i path_to_input_yarrrml.yml -o path_to_rdf_mapping.ttl [-f R2RML] ``` +`-f R2RML` is an optional parameter for translating input YARRRML into R2RML +### From \[R2\]RML to YARRRML +To execute from command line run the following: +```bash +python3 -m yarrrml_translator -i path_to_input_rdf_mapping.ttl -o path_to_output_yarrrml.yml [-f R2RML] +``` `-f R2RML` is an optional parameter for translating input YARRRML into R2RML +### Merging mappings based on TriplesMap id +To execute from command line run the following (keys of each TriplesMap should be disjoint): +```bash +python3 -m yarrrml_translator -m yarrrml_mapping1.yaml yarrrml_mapping2.yaml [..] -o path_to_output_yarrrml.yml +``` + +## Execution as a library + If you want to include the module in your implementation: -- for translating RML mappings: +- for translating YARRRML mapping to RML mappings (and inverse): ```python import yarrrml_translator import yaml - +# YARRRML to RML rml_content = yarrrml_translator.translate(yaml.safe_load(open("path-to-yarrrml"))) +# RML to YARRRML +yarrrml_content = yarrrml_translator.inverse_translate("rdf_mapping_content") ``` - -- for translating R2RML mappings: +- for translating YARRRML mappings to R2RML mappings (and inverse): ```python import yarrrml_translator import yaml R2RML_URI = 'http://www.w3.org/ns/r2rml#' +# YARRRML to R2RML rml_content = yarrrml_translator.translate(yaml.safe_load(open("path-to-yarrrml")), mapping_format=R2RML_URI) +# R2RML to YARRRML +yarrrml_content = yarrrml_translator.inverse_translate("rdf_mapping_content", mapping_format=R2RML_URI) +``` +- for merging TriplesMap based on id: +```python +import yarrrml_translator +list_yarrrml_mappings = ["content_mapping_yarrrml1", "content_mapping_yarrrml1"] +yarrrml_content = yarrrml_translator.merge_mappings(list_yarrrml_mappings) ``` ## Authors -Ontology Engineering Group - Data Integration: +Ontology Engineering Group: - [David Chaves-Fraga](mailto:[email protected]) +- Marino González García (Final bachelor thesis - Systematic Testing) - Luis López Piñero (Final bachelor thesis - v0.1) diff --git a/VERSION b/VERSION index 39e898a..aec258d 100644 --- a/VERSION +++ b/VERSION @@ -1,1 +1,1 @@ -0.7.1 +0.8 diff --git a/requirements.txt b/requirements.txt index 750f28b..be81d1b 100644 --- a/requirements.txt +++ b/requirements.txt @@ -1,3 +1,3 @@ -pyyaml >= 6.0 +ruamel.yaml >= 0.17.21 rdflib >= 6.2.0 coloredlogs >= 15.0.1 \ No newline at end of file diff --git a/yarrrml-translator/yarrrml_translator/__init__.py b/yarrrml-translator/yarrrml_translator/__init__.py index 5a9a17a..1fc05d6 100644 --- a/yarrrml-translator/yarrrml_translator/__init__.py +++ b/yarrrml-translator/yarrrml_translator/__init__.py @@ -1,26 +1,28 @@ from .constants import * -from .mapping import add_prefix, add_mapping, add_inverse_prefix, get_non_asserted_mappings +from .mapping import add_prefix, add_mapping, add_inverse_prefix, get_non_asserted_mappings, merge_mapping_section_by_key from .source import get_initial_sources, add_source, generate_database_connections, add_table, add_inverse_source from .subject import add_subject, add_inverse_subject from .predicateobject import add_predicate_object_maps, add_inverse_pom import rdflib -import yaml - +import ruamel.yaml as yaml +from ruamel.yaml import YAML +from ruamel.yaml.compat import StringIO +import sys def translate(yarrrml_data, mapping_format=RML_URI): logger.info("Translating YARRRML mapping to [R2]RML") list_initial_sources = get_initial_sources(yarrrml_data) rml_mapping = [add_prefix(yarrrml_data)] - rml_mapping.extend(generate_database_connections(yarrrml_data)) + rml_mapping.extend(generate_database_connections(yarrrml_data, list_initial_sources)) try: - mappings = get_non_asserted_mappings(yarrrml_data, dict.fromkeys(list(yarrrml_data.get(YARRRML_MAPPINGS).keys()))) + mappings, mapping_format = get_non_asserted_mappings(yarrrml_data, dict.fromkeys(list(yarrrml_data.get(YARRRML_MAPPINGS).keys())), mapping_format) for mapping in yarrrml_data.get(YARRRML_MAPPINGS): if mapping_format == R2RML_URI: source_list = add_table(yarrrml_data, mapping, list_initial_sources) else: source_list = add_source(yarrrml_data, mapping, list_initial_sources) - subject_list = add_subject(yarrrml_data, mapping) + subject_list = add_subject(yarrrml_data, mapping, mapping_format) pred = add_predicate_object_maps(yarrrml_data, mapping, mapping_format) it = 0 for source in source_list: @@ -55,22 +57,38 @@ def translate(yarrrml_data, mapping_format=RML_URI): def inverse_translation(rdf_mapping, mapping_format=RML_URI): - yarrrml_mapping = {'prefixes': [], 'mappings': {}} + yarrrml_mapping = {YARRRML_PREFIXES: {}, YARRRML_MAPPINGS: {}} rdf_mapping.bind('rml', rdflib.term.URIRef(RML_URI)) rdf_mapping.bind('rr', rdflib.term.URIRef(R2RML_URI)) rdf_mapping.bind('ql', rdflib.term.URIRef(QL_URI)) - yarrrml_mapping['prefixes'] = add_inverse_prefix(rdf_mapping) + yarrrml_mapping[YARRRML_PREFIXES] = add_inverse_prefix(rdf_mapping) query = f'SELECT ?triplesMap WHERE {{ ?triplesMap {RDF_TYPE} {R2RML_TRIPLES_MAP} . }} ' triples_map = [tm[rdflib.Variable('triplesMap')] for tm in rdf_mapping.query(query).bindings] for tm in triples_map: tm_name = tm.split("/")[-1] - yarrrml_tm = {'sources': [add_inverse_source(tm, rdf_mapping, mapping_format)]} - yarrrml_tm['s'], classes = add_inverse_subject(tm, rdf_mapping) - yarrrml_tm['po'] = add_inverse_pom(tm, rdf_mapping, classes, yarrrml_mapping['prefixes']) - yarrrml_mapping['mappings'][tm_name] = yarrrml_tm + yarrrml_tm = {YARRRML_SOURCE: add_inverse_source(tm, rdf_mapping, mapping_format)} + yarrrml_tm[YARRRML_SHORTCUT_SUBJECTS], classes = add_inverse_subject(tm, rdf_mapping) + yarrrml_tm[YARRRML_SHORTCUT_PREDICATEOBJECT] = add_inverse_pom(tm, rdf_mapping, classes, yarrrml_mapping[YARRRML_PREFIXES]) + yarrrml_mapping[YARRRML_MAPPINGS][tm_name] = yarrrml_tm - string_content = str(yaml.dump(yarrrml_mapping, default_flow_style=None, sort_keys=False)).replace("'\"", - '"').replace( - "\"'", ' " ').replace('\'', '') + string_content = yaml.dump(yarrrml_mapping) return string_content + +def merge_mappings(yarrrrml_list): + combined_mapping = {YARRRML_MAPPINGS:{}} + prefixes_list = [] + for mapping in yarrrrml_list: + prefixes_list.append(mapping[YARRRML_PREFIXES]) + combined_mapping = combined_mapping \| merge_mapping_section_by_key(YARRRML_PREFIXES, prefixes_list) + + triples_map_ids = yarrrrml_list[0][YARRRML_MAPPINGS].keys() + for individual_id in triples_map_ids: + mapping_content_list = [] + for mapping in yarrrrml_list: + if individual_id in mapping[YARRRML_MAPPINGS]: + mapping_content_list.append(mapping[YARRRML_MAPPINGS][individual_id]) + combined_mapping[YARRRML_MAPPINGS] = combined_mapping[YARRRML_MAPPINGS] \| merge_mapping_section_by_key(individual_id, mapping_content_list) + + string_content = yaml.dump(combined_mapping) + return string_content \ No newline at end of file diff --git a/yarrrml-translator/yarrrml_translator/__main__.py b/yarrrml-translator/yarrrml_translator/__main__.py index 3947bc3..f9c3721 100644 --- a/yarrrml-translator/yarrrml_translator/__main__.py +++ b/yarrrml-translator/yarrrml_translator/__main__.py @@ -1,8 +1,8 @@ import sys -import yaml +from ruamel.yaml import YAML import argparse from rdflib import Graph -from . import translate, inverse_translation +from . import translate, inverse_translation, merge_mappings from .constants import * @@ -18,9 +18,9 @@ def parse_inputs(): if args.input_mapping_path and args.output_mapping_path: if args.input_mapping_path.endswith('.yml') or args.input_mapping_path.endswith('.yaml'): with open(args.input_mapping_path) as f: - input_data = yaml.safe_load(f) - elif args.input_mapping_path.endswith('.ttl') or args.input_mapping_path.endswith( - '.rml') or args.input_mapping_path.endswith('.r2rml'): + yaml = YAML(typ='safe', pure=True) + input_data = yaml.load(f) + elif args.input_mapping_path.endswith('.ttl') or args.input_mapping_path.endswith('.rml') or args.input_mapping_path.endswith('.r2rml'): input_data = Graph() input_data.parse(args.input_mapping_path, format="turtle") else: @@ -30,7 +30,11 @@ def parse_inputs(): if args.format is not None and args.format == "R2RML": input_format = R2RML_URI - + elif args.merge_mapping: + input_data = [] + for mapping in args.merge_mapping: + with open(mapping) as f: + input_data.append(yaml.safe_load(f)) else: sys.tracebacklimit = 0 logger.error("No correct arguments, run python3 -m yarrrml_translator -h to see the help)") @@ -41,8 +45,9 @@ def parse_inputs(): def define_args(): parser = argparse.ArgumentParser() - parser.add_argument("-i", "--input_mapping_path", required=True, help="Input mapping path in YARRRML or RML") - parser.add_argument("-o", "--output_mapping_path", required=True, help="Output path for the generated mapping") + parser.add_argument("-i", "--input_mapping_path", required=False, help="Input mapping path in YARRRML or RML") + parser.add_argument("-o", "--output_mapping_path", required=False, help="Output path for the generated mapping") + parser.add_argument("-m", "--merge_mapping", nargs="*", required=False, help="List of mappings to be merged") parser.add_argument("-f", "--format", required=False, help="Mapping input format. Values: R2RML or RML (RML by default)") return parser @@ -53,6 +58,8 @@ def define_args(): mapping_format, mapping_data = parse_inputs() if type(mapping_data) is Graph: mapping_content = inverse_translation(mapping_data, mapping_format) + elif args.merge_mapping: + mapping_content = merge_mappings(mapping_data) else: mapping_content = translate(mapping_data, mapping_format) write_results(mapping_content) diff --git a/yarrrml-translator/yarrrml_translator/constants.py b/yarrrml-translator/yarrrml_translator/constants.py index cb25adc..3ebd6be 100644 --- a/yarrrml-translator/yarrrml_translator/constants.py +++ b/yarrrml-translator/yarrrml_translator/constants.py @@ -24,13 +24,18 @@ RML_REFERENCE_FORMULATION = 'rml:referenceFormulation' RML_ITERATOR = 'rml:iterator' RML_REFERENCE = 'rml:reference' - +RML_LANGUAGE_MAP = 'rml:languageMap' +RML_LANGUAGE_MAP_CLASS = 'rml:LanguageMap' +RML_DATATYPE_MAP = 'rml:datatypeMap' +RML_DATATYPE_MAP_CLASS = 'rml:DatatypeMap' +RML_QUERY = 'rml:query' STAR_CLASS = 'rml:StarMap' STAR_NON_ASSERTED_CLASS = 'rml:NonAssertedTriplesMap' STAR_QUOTED = 'rml:quotedTriplesMap' STAR_SUBJECT = 'rml:subjectMap' STAR_OBJECT = 'rml:objectMap' +STAR_URI = 'https://w3id.org/kg-construct/rml-star' R2RML_TEMPLATE = 'rr:template' @@ -55,6 +60,7 @@ R2RML_TERMTYPE = 'rr:termType' R2RML_LANGUAGE = 'rr:language' R2RML_IRI = 'rr:IRI' +R2RML_LITERAL = 'rr:Literal' R2RML_REFOBJECT_CLASS = 'rr:RefObjectMap' R2RML_PARENT_TRIPLESMAP = 'rr:parentTriplesMap' R2RML_JOIN_CONITION = 'rr:joinCondition' diff --git a/yarrrml-translator/yarrrml_translator/mapping.py b/yarrrml-translator/yarrrml_translator/mapping.py index 2ed12db..43dfc8e 100644 --- a/yarrrml-translator/yarrrml_translator/mapping.py +++ b/yarrrml-translator/yarrrml_translator/mapping.py @@ -2,7 +2,7 @@ def add_mapping(mapping, mappings, it): - map_template = "<#" + mapping + "_" + str(it) + "> a " + map_template = "<" + mapping + "_" + str(it) + "> a " if mappings[mapping]: if mappings[mapping] == "non_asserted": map_template += STAR_NON_ASSERTED_CLASS + ", " @@ -71,15 +71,30 @@ def add_inverse_prefix(rdf_mapping): return prefixes -def get_non_asserted_mappings(yarrrml_data, mappings): +def get_non_asserted_mappings(yarrrml_data, mappings, mapping_format): for mapping in yarrrml_data.get(YARRRML_MAPPINGS): keys = yarrrml_data.get(YARRRML_MAPPINGS).get(mapping).keys() for key in keys: - if type(yarrrml_data.get(YARRRML_MAPPINGS).get(mapping).get(key)) is list: - values_list = yarrrml_data.get(YARRRML_MAPPINGS).get(mapping).get(key) - for value in values_list: - if YARRRML_NON_ASSERTED in value: - mappings[value[YARRRML_NON_ASSERTED]] = "non_asserted" - elif 'o' in value and YARRRML_NON_ASSERTED in value['o'][0]: - mappings[value['o'][0][YARRRML_NON_ASSERTED]] = "non_asserted" - return mappings + if type(yarrrml_data.get(YARRRML_MAPPINGS).get(mapping).get(key)) is dict: + values = yarrrml_data.get(YARRRML_MAPPINGS).get(mapping).get(key) + if YARRRML_NON_ASSERTED in values: + mappings[values[YARRRML_NON_ASSERTED]] = "non_asserted" + mapping_format = STAR_URI + elif YARRRML_QUOTED in values: + mapping_format = STAR_URI + if type(yarrrml_data.get(YARRRML_MAPPINGS).get(mapping).get(key)) is list and key==YARRRML_SHORTCUT_PREDICATEOBJECT: + for value in yarrrml_data.get(YARRRML_MAPPINGS).get(mapping).get(key): + if type(value) is dict and YARRRML_SHORTCUT_OBJECT in value: + if YARRRML_NON_ASSERTED in value[YARRRML_SHORTCUT_OBJECT]: + mappings[value[YARRRML_SHORTCUT_OBJECT][YARRRML_NON_ASSERTED]] = "non_asserted" + mapping_format = STAR_URI + elif YARRRML_QUOTED in value[YARRRML_SHORTCUT_OBJECT]: + mapping_format = STAR_URI + + return mappings, mapping_format + +def merge_mapping_section_by_key(key,yarrrml_list): + output = {key:{}} + for yarrrml_mapping in yarrrml_list: + output[key] = output[key] \| yarrrml_mapping + return output \ No newline at end of file diff --git a/yarrrml-translator/yarrrml_translator/predicateobject.py b/yarrrml-translator/yarrrml_translator/predicateobject.py index 7c3a94d..03735bd 100644 --- a/yarrrml-translator/yarrrml_translator/predicateobject.py +++ b/yarrrml-translator/yarrrml_translator/predicateobject.py @@ -75,7 +75,17 @@ def get_object_list(predicate_object, object_access): for i in range(len(object_maps)): object_list.append([object_maps[i], predicate_object[2]]) else: - object_list.extend(object_maps) + for object in object_maps: + if YARRRML_LANGUAGE in object: + object_list.append([object[YARRRML_VALUE],object[YARRRML_LANGUAGE]+"~lang"]) + elif YARRRML_DATATYPE in object: + object_list.append([object[YARRRML_VALUE], object[YARRRML_DATATYPE]]) + elif YARRRML_TYPE in object: + object_list.append([object[YARRRML_VALUE]+"~"+object[YARRRML_TYPE]]) + elif YARRRML_VALUE in object: + object_list.append([object[YARRRML_VALUE]]) + else: + object_list.append([object]) else: if object_access is None and len(predicate_object) == 3: object_list.append([object_maps, predicate_object[2]]) @@ -124,16 +134,28 @@ def add_predicate_object(data, mapping, predicate_object, mapping_format=RML_URI if YARRRML_IRI in om[0]: object_value = om[0].split(YARRRML_IRI)[0] iri = True - template += generate_rml_termmap(R2RML_OBJECT, R2RML_OBJECT_CLASS, + if mapping_format == STAR_URI: + template += generate_rml_termmap(STAR_OBJECT, R2RML_OBJECT_CLASS, + object_value, "\t\t\t", mapping_format) + else: + template += generate_rml_termmap(R2RML_OBJECT, R2RML_OBJECT_CLASS, object_value, "\t\t\t", mapping_format) if len(om) == 2: types = check_type(om[1]) if types != "error": if types == YARRRML_LANGUAGE: - template = template[0:len(template) - 5] + "\t\t\t" + R2RML_LANGUAGE + " \"" \ + if "$(" in om[1]: + template = template[0:len(template) - 5] + generate_rml_termmap(RML_LANGUAGE_MAP, RML_LANGUAGE_MAP_CLASS, + om[1].replace("~lang",""), "\t\t\t\t", mapping_format) + "\n\t\t];\n" + else: + template = template[0:len(template) - 5] + "\t\t\t" + R2RML_LANGUAGE + " \"" \ + om[1].replace(YARRRML_LANG, "") + "\"\n\t\t];\n" elif types == YARRRML_DATATYPE: - template = template[0:len(template) - 5] + "\t\t\t" + R2RML_DATATYPE + " " \ + if "$(" in om[1]: + template = template[0:len(template) - 5] + generate_rml_termmap(RML_DATATYPE_MAP, RML_DATATYPE_MAP_CLASS, + om[1], "\t\t\t\t", mapping_format) + "\n\t\t];\n" + else: + template = template[0:len(template) - 5] + "\t\t\t" + R2RML_DATATYPE + " " \ + om[1] + "\n\t\t];\n" if iri: template = template[0:len(template) - 5] + "\t\t\t" + R2RML_TERMTYPE + " " \ @@ -142,9 +164,13 @@ def add_predicate_object(data, mapping, predicate_object, mapping_format=RML_URI if YARRRML_MAPPING in om: template += ref_mapping(data, mapping, om, YARRRML_MAPPING, R2RML_PARENT_TRIPLESMAP, mapping_format) elif YARRRML_NON_ASSERTED in om: - template += ref_mapping(data, mapping, om, YARRRML_NON_ASSERTED, STAR_QUOTED, mapping_format) + if YARRRML_CONDITION in om: + template += ref_mapping(data, mapping, om, YARRRML_NON_ASSERTED, STAR_QUOTED, mapping_format) + else: + template += generate_rml_termmap(STAR_OBJECT, STAR_CLASS,om, "\t\t\t", mapping_format) else: - template += ref_mapping(data, mapping, om, YARRRML_QUOTED, STAR_QUOTED, mapping_format) + template += ref_mapping(data, mapping, om, YARRRML_QUOTED, STAR_QUOTED, mapping_format) + # This could be removed else: if YARRRML_VALUE in om: object_value = om.get(YARRRML_VALUE) @@ -153,18 +179,28 @@ def add_predicate_object(data, mapping, predicate_object, mapping_format=RML_URI if YARRRML_IRI in om: object_value = om.split(YARRRML_IRI)[0] iri = True - template += generate_rml_termmap(R2RML_OBJECT, R2RML_OBJECT_CLASS, - object_value, "\t\t\t", mapping_format) - if type(om) == list and YARRRML_DATATYPE in om: + if mapping_format == STAR_URI: + template += generate_rml_termmap(STAR_OBJECT, R2RML_OBJECT_CLASS, + object_value, "\t\t\t", mapping_format) + else: + template += generate_rml_termmap(R2RML_OBJECT, R2RML_OBJECT_CLASS, + object_value, "\t\t\t", mapping_format) + if YARRRML_DATATYPE in om: template = template[0:len(template) - 5] + "\t\t\t" + R2RML_DATATYPE + " " \ + om.get(YARRRML_DATATYPE) + "\n\t\t];\n" - if (type(om) == list and YARRRML_TYPE in om and om.get(YARRRML_TYPE) == "iri") or iri: - template = template[0:len(template) - 5] + "\t\t\t" + R2RML_TERMTYPE + " " \ - + R2RML_IRI + "\n\t\t];\n" - if type(om) == list and YARRRML_LANGUAGE in om: + if YARRRML_LANGUAGE in om: template = template[0:len(template) - 5] + "\t\t\t" + R2RML_LANGUAGE + " \"" \ + om.get(YARRRML_LANGUAGE) + "\"\n\t\t];\n" + if YARRRML_TYPE in om: + if om.get(YARRRML_TYPE) == "iri": + iri = True + elif om.get(YARRRML_TYPE) == "literal": + template = template[0:len(template) - 5] + "\t\t\t" + R2RML_TERMTYPE + " " \ + + R2RML_LITERAL + "\n\t\t];\n" + if iri: + template = template[0:len(template) - 5] + "\t\t\t" + R2RML_TERMTYPE + " " \ + + R2RML_IRI + "\n\t\t];\n" for graph in graph_list: template += generate_rml_termmap(R2RML_GRAPH, R2RML_GRAPH_CLASS, graph, "\t\t\t") @@ -180,7 +216,7 @@ def ref_mapping(data, mapping, om, yarrrml_key, ref_type_property, mapping_forma mapping_join = om.get(yarrrml_key) if mapping_join in list_mappings: - subject_list = add_subject(data, mapping_join) + subject_list = add_subject(data, mapping_join, mapping_format) list_initial_sources = get_initial_sources(data) if mapping_format == R2RML_URI: source_list = add_table(data, mapping_join, list_initial_sources) @@ -191,7 +227,7 @@ def ref_mapping(data, mapping, om, yarrrml_key, ref_type_property, mapping_forma for i in range(number_joins_rml): template += "\t\t" + R2RML_OBJECT + \ " [ \n\t\t\ta " + R2RML_REFOBJECT_CLASS + \ - ";\n\t\t\t" + ref_type_property + " <#" + mapping_join + "_" + str(i) + ">;\n" + ";\n\t\t\t" + ref_type_property + " <" + mapping_join + "_" + str(i) + ">;\n" if YARRRML_CONDITION in om: conditions = om.get(YARRRML_CONDITION) if type(conditions) is not list: @@ -225,8 +261,8 @@ def add_inverse_pom(tm, rdf_mapping, classes, prefixes): for c in classes: yarrrml_poms.append(['rdf:type', c]) - query = f'SELECT ?predicate ?predicateValue ?object ?objectValue ?termtype ?datatype ' \ - f'?parentTriplesMap ?child ?parent' \ + query = f'SELECT ?predicate ?predicateValue ?object ?objectValue ?termtype ?datatype ?datatypeMapValue ' \ + f'?language ?languageMapValue ?parentTriplesMap ?child ?parent' \ f' WHERE {{ ' \ f'<{tm}> {R2RML_PREDICATE_OBJECT_MAP} ?predicateObjectMap . ' \ f'?predicateObjectMap {R2RML_PREDICATE}\|{R2RML_SHORTCUT_PREDICATE} ?predicate .' \ @@ -235,7 +271,14 @@ def add_inverse_pom(tm, rdf_mapping, classes, prefixes): f'OPTIONAL {{ ' \ f'?object {R2RML_TEMPLATE}\|{R2RML_COLUMN}\|{R2RML_CONSTANT}\|{RML_REFERENCE} ?objectValue .' \ f'OPTIONAL {{ ?object {R2RML_TERMTYPE} ?termtype . }}' \ - f'OPTIONAL {{ ?object {R2RML_DATATYPE} ?datatype .}} }} ' \ + f'OPTIONAL {{ ?object {R2RML_DATATYPE} ?datatype .}} ' \ + f'OPTIONAL {{ ' \ + f' ?object {RML_DATATYPE_MAP} ?datatypeMap .' \ + f' ?datatypeMap {R2RML_TEMPLATE}\|{R2RML_CONSTANT}\|{RML_REFERENCE} ?datatypeMapValue .}} ' \ + f'OPTIONAL {{ ?object {R2RML_LANGUAGE} ?language .}} ' \ + f'OPTIONAL {{ ' \ + f' ?object {RML_LANGUAGE_MAP} ?languageMap .' \ + f' ?languageMap {R2RML_TEMPLATE}\|{R2RML_CONSTANT}\|{RML_REFERENCE} ?languageMapValue .}} }} ' \ f'OPTIONAL {{ ?object {R2RML_PARENT_TRIPLESMAP} ?parentTriplesMap .' \ f'?object {R2RML_JOIN_CONITION} ?join_condition .' \ f'?join_condition {R2RML_CHILD} ?child .' \ @@ -262,13 +305,19 @@ def add_inverse_pom(tm, rdf_mapping, classes, prefixes): else: yarrrml_pom.append(predicate) if tm['objectValue']: # we have extended objectMap version - object = tm['objectValue'].replace('{', "$(").replace('}', ')') + object = tm['objectValue'].replace('{', '$(').replace('}', ')') if tm['termtype']: if tm['termtype'] == rdflib.URIRef(R2RML_IRI): object = object + '~iri' yarrrml_pom.append(object) if tm['datatype']: yarrrml_pom.append(tm['datatype']) + elif tm['datatypeMapValue']: + yarrrml_pom.append(tm['datatypeMapValue'].replace('{', '$(').replace('}', ')')) + if tm['language']: + yarrrml_pom.append(tm['language']+"~lang") + elif tm['languageMapValue']: + yarrrml_pom.append(tm['languageMapValue'].replace('{', '$(').replace('}', ')')+"~lang") elif tm['object'] and not tm['parentTriplesMap']: # we have object shortcut yarrrml_pom.append(tm['object'].replace('{', "$(").replace('}', ')')) diff --git a/yarrrml-translator/yarrrml_translator/source.py b/yarrrml-translator/yarrrml_translator/source.py index 6d55204..a39095b 100644 --- a/yarrrml-translator/yarrrml_translator/source.py +++ b/yarrrml-translator/yarrrml_translator/source.py @@ -31,13 +31,14 @@ def add_source(data, mapping, list_initial_sources): final_list = [] sources = get_sources(data, mapping) for source in sources: - + db_identifier = mapping if source in list_initial_sources: + db_identifier = source source = data.get(YARRRML_SOURCES).get(source) if YARRRML_ACCESS in source: if YARRRML_QUERY in source: - final_list.append(source_template + database_source(mapping, source)) + final_list.append(source_template + database_source(mapping, source, db_identifier)) else: final_list.append(source_template + add_source_full(mapping, source)) elif type(source) is list: @@ -55,11 +56,13 @@ def add_table(data, mapping, list_initial_sources): sources = get_sources(data, mapping) for source in sources: sql_version = False + db_identifier = mapping if source in list_initial_sources: + db_identifier = source source = data.get(YARRRML_SOURCES).get(source) if YARRRML_ACCESS in source and YARRRML_QUERY in source: - r2rml_access = database_source(mapping, source) + r2rml_access = database_source(mapping, source, db_identifier) sql_version = True elif YARRRML_QUERY in source: r2rml_access = R2RML_SQL_QUERY + " \"" + source.get(YARRRML_QUERY).replace("\n", " ").replace("\"", @@ -112,8 +115,7 @@ def add_source_full(mapping, source): reference_formulation = str(source.get(YARRRML_REFERENCE_FORMULATION)) format_from_reference = switch_in_reference_formulation(reference_formulation.lower()) ref_formulation_rml = reference_formulation.replace("jsonpath", "JSONPath").replace("csv", "CSV").replace( - "xpath", - "XPath") + "xpath","XPath") if extension != format_from_reference or format_from_reference == "ERROR": raise Exception("ERROR: not referenceFormulation found or mismatch between the format and " "referenceFormulation in source " + access + "in mapping " + mapping) @@ -139,24 +141,20 @@ def add_source_full(mapping, source): return source_rdf -def database_source(mapping, source): +def database_source(mapping, source, db_identifier): source_rdf = "" if YARRRML_ACCESS in source: if YARRRML_CREDENTIALS in source: if YARRRML_TYPE in source: - type = source.get(YARRRML_TYPE) - access = source.get(YARRRML_ACCESS) - username = source.get(YARRRML_CREDENTIALS).get(YARRRML_USERNAME) - password = source.get(YARRRML_CREDENTIALS).get(YARRRML_PASSWORD) - hash_datasource = abs(hash(access + type + username + password)) - source_rdf += "<#DataSource_" + str(hash_datasource) + ">;\n\t\t" + source_rdf += "<DataSource_" + str(db_identifier) + ">;\n\t\t" if YARRRML_QUERY in source: - source_rdf += R2RML_SQL_QUERY + " \"" + source.get(YARRRML_QUERY) + "\"" + source_rdf += RML_QUERY + " \"" + source.get(YARRRML_QUERY).replace("\n", " ").replace("\"", + "\\\"") + "\"" elif YARRRML_TABLE in source: source_rdf += R2RML_TABLE_NAME + " \"" + source.get(YARRRML_TABLE) + "\"" if YARRRML_REFERENCE_FORMULATION in source: source_rdf += ";\n\t\t" + RML_REFERENCE_FORMULATION + " ql:" \ - + switch_in_reference_formulation(source.get(YARRRML_REFERENCE_FORMULATION)) + + switch_in_reference_formulation(source.get(YARRRML_REFERENCE_FORMULATION)).upper() if YARRRML_QUERY_FORMULATION in source: source_rdf += ";\n\t\t" + R2RML_SQL_VERSION + " rr:" + source.get(YARRRML_QUERY_FORMULATION).upper() else: @@ -187,35 +185,43 @@ def switch_in_reference_formulation(value): return switcher -def generate_database_connections(data): +def generate_database_connections(data, list_initial_sources): database = [] - hash_ids = [] + sources_ids = set() for mapping in data.get(YARRRML_MAPPINGS): sources = get_sources(data, mapping) for source in sources: - if YARRRML_QUERY in source and YARRRML_ACCESS in source: - type = source.get(YARRRML_TYPE) - if type == "mysql": - driver = "com.mysql.jdbc.Driver" - elif type == "postgresql": - driver = "org.postgresql.Driver" - elif type == "sqlserver": - driver = "com.microsoft.sqlserver.jdbc.SQLServerDriver" - else: - driver = None - access = source.get(YARRRML_ACCESS) - username = source.get(YARRRML_CREDENTIALS).get(YARRRML_USERNAME) - password = source.get(YARRRML_CREDENTIALS).get(YARRRML_PASSWORD) - hash_datasource = abs(hash(access + type + username + password)) - if not hash_datasource in hash_ids: - hash_ids.append(hash_datasource) + db_identifier = mapping + external = False + number_external_sources = len(sources_ids) + if source in list_initial_sources: + external = True + db_identifier = source + sources_ids.add(source) + source = data.get(YARRRML_SOURCES).get(source) + if (external and len(sources_ids) > number_external_sources) or not external: + if YARRRML_QUERY in source and YARRRML_ACCESS in source: + db_type = source.get(YARRRML_TYPE) + if db_type == "mysql": + driver = "com.mysql.jdbc.Driver" + elif db_type == "postgresql": + driver = "org.postgresql.Driver" + elif db_type == "sqlserver": + driver = "com.microsoft.sqlserver.jdbc.SQLServerDriver" + else: + driver = None + access = source.get(YARRRML_ACCESS) + username = source.get(YARRRML_CREDENTIALS).get(YARRRML_USERNAME) + password = source.get(YARRRML_CREDENTIALS).get(YARRRML_PASSWORD) + + if driver is None: - database.append("<#DataSource_" + str(hash_datasource) + "> a " + D2RQ_DATABASE_CLASS + ";\n\t" + database.append("<DataSource_" + str(db_identifier) + "> a " + D2RQ_DATABASE_CLASS + ";\n\t" + D2RQ_DSN + " \"" + access + "\";\n\t" + D2RQ_USER + " \"" + username + "\";\n\t" + D2RQ_PASS + " \"" + password + "\".\n\n") else: - database.append("<#DataSource_" + str(hash_datasource) + "> a " + D2RQ_DATABASE_CLASS + ";\n\t" + database.append("<DataSource_" + str(db_identifier) + "> a " + D2RQ_DATABASE_CLASS + ";\n\t" + D2RQ_DSN + " \"" + access + "\";\n\t" + D2RQ_DRIVER + " \"" + driver + "\";\n\t" + D2RQ_USER + " \"" + username + "\";\n\t" diff --git a/yarrrml-translator/yarrrml_translator/subject.py b/yarrrml-translator/yarrrml_translator/subject.py index dda6430..4be5f33 100644 --- a/yarrrml-translator/yarrrml_translator/subject.py +++ b/yarrrml-translator/yarrrml_translator/subject.py @@ -4,7 +4,7 @@ ## Adds a subject or set of subjects to the given TriplesMap -def add_subject(data, mapping): +def add_subject(data, mapping, mapping_format): rml_subjects = [] only_one = False @@ -26,9 +26,19 @@ def add_subject(data, mapping): for individual_subject in subject: if YARRRML_QUOTED in individual_subject or YARRRML_NON_ASSERTED in individual_subject: + if YARRRML_CONDITION in individual_subject: + from .predicateobject import ref_mapping + if YARRRML_NON_ASSERTED in individual_subject: + subject_termmap = ref_mapping(data, mapping, individual_subject, YARRRML_NON_ASSERTED, STAR_QUOTED, mapping_format) + else: + subject_termmap = ref_mapping(data, mapping, individual_subject, YARRRML_QUOTED, STAR_QUOTED, mapping_format) + subject_termmap=subject_termmap.replace(R2RML_REFOBJECT_CLASS,STAR_CLASS).replace(R2RML_OBJECT,STAR_SUBJECT).replace("\t\t\t\t","\t\t\t\t\t").replace("\t\t","\t") + else: + subject_termmap = generate_rml_termmap(STAR_SUBJECT, STAR_CLASS, individual_subject, "\t\t") + elif mapping_format == STAR_URI: subject_termmap = generate_rml_termmap(STAR_SUBJECT, R2RML_SUBJECT_CLASS, individual_subject, "\t\t") else: - subject_termmap = generate_rml_termmap(R2RML_SUBJECT, R2RML_SUBJECT_CLASS, individual_subject, "\t\t") + subject_termmap = generate_rml_termmap(R2RML_SUBJECT, R2RML_SUBJECT_CLASS, individual_subject, "\t\t", mapping_format) rml_subjects.append(subject_termmap) if YARRRML_GRAPHS in data.get(YARRRML_MAPPINGS).get(mapping): diff --git a/yarrrml-translator/yarrrml_translator/termmap.py b/yarrrml-translator/yarrrml_translator/termmap.py index 6406cbf..6d4ab76 100644 --- a/yarrrml-translator/yarrrml_translator/termmap.py +++ b/yarrrml-translator/yarrrml_translator/termmap.py @@ -30,9 +30,9 @@ def generate_rml_termmap(rml_property, rml_class, text, identation, mapping_form if term_map == STAR_QUOTED: if 'quoted' in text: - template += term_map + " <#" + text[YARRRML_QUOTED] + "_0>;\n" + identation[0:-1] + "];\n" + template += term_map + " <" + text[YARRRML_QUOTED] + "_0>;\n" + identation[0:-1] + "];\n" else: - template += term_map + " <#" + text[YARRRML_NON_ASSERTED] + "_0>;\n" + identation[0:-1] + "];\n" + template += term_map + " <" + text[YARRRML_NON_ASSERTED] + "_0>;\n" + identation[0:-1] + "];\n" elif term_map != "rr:constant": template += term_map + " \"" + text + "\";\n"+identation[0:-1]+"];\n" else:	Mapping-pattern merge for PPDS Given a set of non-compliant mapping rules in YARRRML (e.g. TMs without subject), merging them based on the mapping id and validate they are correct tfg test case development this pull request intends to merge the test cases that have been developed in my repository to the translator's main repository		2023-01-12T16:45:12	[]	[]
oeg-upm/yatter	oeg-upm__yatter-38	7b3d0d20afd322ce82de37afe7e784459352249d	diff --git a/.github/workflows/python-publish-dev.yml b/.github/workflows/python-publish-dev.yml deleted file mode 100644 index 330a33d..0000000 --- a/.github/workflows/python-publish-dev.yml +++ /dev/null @@ -1,32 +0,0 @@ -# This workflows will upload a Python Package using Twine when a release is created -# For more information see: https://help.github.com/en/actions/language-and-framework-guides/using-python-with-github-actions#publishing-to-package-registries - -name: Upload Python Package (development version) - -on: - push: - branch: master - -jobs: - deploy: - - runs-on: ubuntu-latest - - steps: - - uses: actions/checkout@v2 - - name: Set up Python - uses: actions/setup-python@v2 - with: - python-version: '3.x' - - name: Install dependencies - run: \| - python -m pip install --upgrade pip - pip install setuptools wheel twine - - name: Build and publish - env: - TWINE_USERNAME: ${{ secrets.PYPI_USERNAME }} - TWINE_PASSWORD: ${{ secrets.PYPI_PASSWORD }} - run: \| - cd pretty_yarrrml2rml - python setup.py -k dev sdist bdist_wheel - twine upload dist/* diff --git a/.github/workflows/python-publish-rel.yml b/.github/workflows/python-publish-rel.yml index 99d2aa2..574ace4 100644 --- a/.github/workflows/python-publish-rel.yml +++ b/.github/workflows/python-publish-rel.yml @@ -1,7 +1,7 @@ # This workflows will upload a Python Package using Twine when a release is created # For more information see: https://help.github.com/en/actions/language-and-framework-guides/using-python-with-github-actions#publishing-to-package-registries -name: Upload Python Package (release) +name: Upload Python Package on: release: @@ -27,6 +27,6 @@ jobs: TWINE_USERNAME: ${{ secrets.PYPI_USERNAME }} TWINE_PASSWORD: ${{ secrets.PYPI_PASSWORD }} run: \| - cd pretty_yarrrml2rml + cd yarrrml-translator python setup.py -k rel sdist bdist_wheel twine upload dist/* diff --git a/.gitignore b/.gitignore index 0aa167d..994284d 100644 --- a/.gitignore +++ b/.gitignore @@ -1,4 +1,5 @@ .DS_Store lib/__pychache__ .idea -venv \ No newline at end of file +venv +.pytest_cache \ No newline at end of file diff --git a/README.md b/README.md index 171ce80..6b3d172 100644 --- a/README.md +++ b/README.md @@ -1,26 +1,46 @@ -# Pretty-yarrrml2rml -The tool translates mapping rules from YARRRML in a pretty turtle-based serialization of RML. The translation is based -on [RML]((https://rml.io/specs/rml/ "RML Spec") and [YARRRML]((https://rml.io/yarrrml/spec/ "YARRRML Spec") specifications. +# yarrrml-translator -## Limitations: -We are working on inlcuding the following features which are not yet implemented: -- Translation to [R2RML](https://www.w3.org/TR/r2rml) -- [Named graphs](https://www.w3.org/TR/r2rml/#named-graphs) from R2RML -- Functions included using the [FnO](https://fno.io/) Ontology +![GitHub](https://img.shields.io/github/license/oeg-upm/yarrrml-translator?style=flat) +[![DOI](https://zenodo.org/badge/DOI/10.5281/zenodo.7024501.svg)](https://doi.org/10.5281/zenodo.7024501) +[![PyPI](https://img.shields.io/pypi/v/yarrrml-translator?style=flat)](https://pypi.org/project/yarrrml-translator) +![GitHub Release Date](https://img.shields.io/github/release-date/oeg-upm/yarrrml-translator) + +The tool translates mapping rules from YARRRML in a turtle-based serialization of RML or R2RML. The translation is based +on the [RML](https://rml.io/specs/rml/) and [YARRRML](https://rml.io/yarrrml/spec/) specifications. + +## Limitations +We are working on including the following features which are not yet implemented: +- Functions using the [FnO](https://fno.io/) Ontology + +## Installation: +``` +pip install yarrrml-translator +``` ## Execution -In order to execute the code run you can download de PyPi module: +To execute from command line run the following: ```bash -python3 -m pip install pretty-yarrrml2rml -python3 -m pretty_yarrrml2rml -i path_to_input_yarrrml.yml -o path_to_output_rml.rml +python3 -m yarrrml_translator -i path_to_input_yarrrml.yml -o path_to_output.ttl [-f R2RML] ``` +`-f R2RML` is an optional parameter for translating input YARRRML into R2RML + If you want to include the module in your implementation: +- for translating RML mappings: +```python +import yarrrml_translator +import yaml + +rml_content = yarrrml_translator.translate(yaml.safe_load(open("path-to-yarrrml"))) +``` + +- for translating R2RML mappings: ```python -import pretty_yarrrml2rml +import yarrrml_translator import yaml -rml_content = pretty_yarrrml2rml.translate(yaml.safe_load(open("path-to-yarrrml"))) +R2RML_URI = 'http://www.w3.org/ns/r2rml#' +rml_content = yarrrml_translator.translate(yaml.safe_load(open("path-to-yarrrml")), mapping_format=R2RML_URI) ``` ## Authors diff --git a/VERSION b/VERSION index ee6cdce..39e898a 100644 --- a/VERSION +++ b/VERSION @@ -1,1 +1,1 @@ -0.6.1 +0.7.1 diff --git a/requirements.txt b/requirements.txt index 71e1245..5e28d05 100644 --- a/requirements.txt +++ b/requirements.txt @@ -1,2 +1,3 @@ -pyyaml == 6.0 -rdflib == 6.2.0 +pyyaml >= 6.0 +rdflib >= 6.2.0 +coloredlogs >= 15.0.1 diff --git a/pretty_yarrrml2rml/setup.py b/yarrrml-translator/setup.py similarity index 72% rename from pretty_yarrrml2rml/setup.py rename to yarrrml-translator/setup.py index c44bae0..856696b 100644 --- a/pretty_yarrrml2rml/setup.py +++ b/yarrrml-translator/setup.py @@ -16,25 +16,26 @@ with open("../VERSION", "r") as fh: v = fh.read().replace("\n", "") - if parsed.kind == "rel": - vers_taged = v - else: - vers_taged = v+".dev"+v_time + vers_taged = v with open("../requirements.txt") as r: requirements = list(filter(None, r.read().split("\n")[0:])) setuptools.setup( - name="pretty_yarrrml2rml", + name="yarrrml_translator", version=vers_taged, author="David Chaves-Fraga", author_email="[email protected]", license="Apache 2.0", - description="The tool translates mapping rules in RML from YARRRML serialization to RDF turtle in a pretty and interpreatable way for humans. The translation is based on RML and YARRRML specifications.", + description="A translator from YARRRML to RML mappings.", long_description=long_description, long_description_content_type="text/markdown", - url="https://github.com/oeg-upm/pretty-yarrrml2rml", + url="https://github.com/oeg-upm/yarrrml-translator", + project_urls={ + 'Source code': 'https://github.com/oeg-upm/yarrrml-translator', + 'Issue tracker': 'https://github.com/oeg-upm/yarrrml-translator/issues', + }, include_package_data=True, packages=setuptools.find_packages(), classifiers=[ @@ -42,11 +43,11 @@ "License :: OSI Approved :: Apache Software License", "Operating System :: OS Independent", "Environment :: Console", - "Intended Audience :: Science/Research", + 'Intended Audience :: Information Technology', "Topic :: Utilities", "Topic :: Software Development :: Pre-processors", "Topic :: Scientific/Engineering :: Interface Engine/Protocol Translator" ], install_requires=requirements, - python_requires='>=3.6', + python_requires='>=3.7', ) diff --git a/pretty_yarrrml2rml/pretty_yarrrml2rml/__init__.py b/yarrrml-translator/yarrrml_translator/__init__.py similarity index 54% rename from pretty_yarrrml2rml/pretty_yarrrml2rml/__init__.py rename to yarrrml-translator/yarrrml_translator/__init__.py index b89db22..b8a2452 100644 --- a/pretty_yarrrml2rml/pretty_yarrrml2rml/__init__.py +++ b/yarrrml-translator/yarrrml_translator/__init__.py @@ -1,22 +1,30 @@ +import logging, coloredlogs from .constants import * from .mapping import add_prefix, add_mapping -from .source import get_initial_sources, add_source, generate_database_connections +from .source import get_initial_sources, add_source, generate_database_connections, add_table from .subject import add_subject from .predicateobject import add_predicate_object_maps -import os from rdflib import Graph -def translate(yarrrml_data): - print("------------------------START TRANSLATING YARRRML TO RML-------------------------------") +def translate(yarrrml_data, mapping_format=RML_URI): + # configure logging with colors + logger = logging.getLogger(__name__) + coloredlogs.install(level='DEBUG', fmt='%(asctime)s,%(msecs)03d \| %(levelname)s: %(message)s') + + logger.info("Translating YARRRML mapping to [R2]RML") + list_initial_sources = get_initial_sources(yarrrml_data) rml_mapping = [add_prefix(yarrrml_data)] rml_mapping.extend(generate_database_connections(yarrrml_data)) try: for mapping in yarrrml_data.get(YARRRML_MAPPINGS): - source_list = add_source(yarrrml_data, mapping, list_initial_sources) + if mapping_format == R2RML_URI: + source_list = add_table(yarrrml_data, mapping, list_initial_sources) + else: + source_list = add_source(yarrrml_data, mapping, list_initial_sources) subject_list = add_subject(yarrrml_data, mapping) - pred = add_predicate_object_maps(yarrrml_data, mapping) + pred = add_predicate_object_maps(yarrrml_data, mapping, mapping_format) it = 0 for source in source_list: for subject in subject_list: @@ -29,28 +37,21 @@ def translate(yarrrml_data): rml_mapping.append(".\n\n\n") it = it + 1 - print("RML content successfully created!\nStarting the validation with RDFLib....") + logger.info("RML content is created!") rml_mapping_string = "".join(rml_mapping) try: - rml_output_file = open("tmp.nt", "w") - rml_output_file.write(rml_mapping_string) - rml_output_file.close() graph = Graph() - graph.parse("tmp.nt", format="turtle") - if os.path.exists("tmp.nt"): - os.remove("tmp.nt") - print("The mapping in RML was successfully validated.") + graph.parse(data=rml_mapping_string, format="turtle") + logger.info("Mapping has been syntactically validated.") except Exception as e: - print("------------------------ERROR-------------------------------") - print("Validating the RDF generated: " + str(e)) - if os.path.exists("tmp.nt"): - os.remove("tmp.nt") + logger.error("ERROR: There is a syntactic error in the generated mapping") + logger.error(str(e)) return None except Exception as e: - print("------------------------ERROR-------------------------------") - print("RML content not generated: " + str(e)) + logger.error("ERROR: The YARRRML mapping has not been translated") + logger.error(str(e)) return None - print("------------------------END TRANSLATION-------------------------------") + logger.info("Translation has finished successfully.") - return rml_mapping_string + return rml_mapping_string \ No newline at end of file diff --git a/pretty_yarrrml2rml/pretty_yarrrml2rml/__main__.py b/yarrrml-translator/yarrrml_translator/__main__.py similarity index 61% rename from pretty_yarrrml2rml/pretty_yarrrml2rml/__main__.py rename to yarrrml-translator/yarrrml_translator/__main__.py index 23ea966..4680033 100644 --- a/pretty_yarrrml2rml/pretty_yarrrml2rml/__main__.py +++ b/yarrrml-translator/yarrrml_translator/__main__.py @@ -1,42 +1,45 @@ -import sys -import yaml -from . import translate - - -def write_results(rml_mapping): - if len(sys.argv) == 1: - rml_output_path = input("Name the path for the output file:") - else: - rml_output_path = sys.argv[4] - - rml_output_file = open(rml_output_path, "w") - rml_output_file.write(rml_mapping) - rml_output_file.close() - - print("Output mapping saved") - - -def run_parsing_system_inputs(): - if len(sys.argv) == 1: - file = input("Introduce the .yml file you want to convert to RML") - with open(file) as f: - yaml_data = yaml.safe_load(f) - - elif len(sys.argv) == 5 and sys.argv[1] == "-i" and sys.argv[3] == "-o": - with open(sys.argv[2]) as f: - yaml_data = yaml.safe_load(f) - else: - sys.tracebacklimit = 0 - raise Exception( - "\n####################################\nERROR: Wrong argument input. You can:" - "\n-Use no arguments\n-Use arguments (in this order): -i yarrrml.yml -o mapping.rml.ttl" - "\n####################################\n") - - return yaml_data - - -yarrrml_data = run_parsing_system_inputs() - -rml_content = translate(yarrrml_data) - -write_results(rml_content) +import sys +import yaml +from . import translate +from .constants import RML_URI, R2RML_URI + + +def write_results(rml_mapping): + if len(sys.argv) == 1: + rml_output_path = input("Name the path for the output file:") + else: + rml_output_path = sys.argv[sys.argv.index("-o") + 1] + + rml_output_file = open(rml_output_path, "w") + rml_output_file.write(rml_mapping) + rml_output_file.close() + + +def run_parsing_system_inputs(): + mapping_format = RML_URI + if len(sys.argv) == 1: + file = input("Introduce the .yml file you want to convert to RML") + with open(file) as f: + yaml_data = yaml.safe_load(f) + + elif (len(sys.argv) == 5 or len(sys.argv) == 7) and "-i" in sys.argv and "-o" in sys.argv: + with open(sys.argv[sys.argv.index("-i") + 1]) as f: + yaml_data = yaml.safe_load(f) + if "-f" in sys.argv: + if sys.argv[sys.argv.index("-f") + 1] == "R2RML": + mapping_format = R2RML_URI + else: + sys.tracebacklimit = 0 + raise Exception( + "\n####################################\nERROR: Wrong argument input. You can:" + "\n-Use no arguments\n-Use arguments (in this order): -i yarrrml.yml -o mapping.rml.ttl" + "\n####################################\n") + + return mapping_format, yaml_data + + +mapping_format, yarrrml_data = run_parsing_system_inputs() + +rml_content = translate(yarrrml_data, mapping_format) + +write_results(rml_content) diff --git a/pretty_yarrrml2rml/pretty_yarrrml2rml/constants.py b/yarrrml-translator/yarrrml_translator/constants.py similarity index 93% rename from pretty_yarrrml2rml/pretty_yarrrml2rml/constants.py rename to yarrrml-translator/yarrrml_translator/constants.py index 5095eb4..db075a6 100644 --- a/pretty_yarrrml2rml/pretty_yarrrml2rml/constants.py +++ b/yarrrml-translator/yarrrml_translator/constants.py @@ -13,6 +13,7 @@ EXAMPLE_URI = 'http://example.com/ns#' RDFS_URI = 'http://www.w3.org/2000/01/rdf-schema#' XSD_URI = 'http://www.w3.org/2001/XMLSchema#' +FOAF_URI ='http://xmlns.com/foaf/0.1/' RDF_TYPE = 'rdf:type' RML_PREFIX = '@prefix' @@ -51,6 +52,10 @@ R2RML_JOIN_CONITION = 'rr:joinCondition' R2RML_CHILD = 'rr:child' R2RML_PARENT = 'rr:parent' +R2RML_LOGICAL_TABLE_CLASS = 'rr:LogicalTable' +R2RML_LOGICAL_TABLE = 'rr:logicalTable' +R2RML_TABLE_NAME = 'rr:tableName' +R2RML_COLUMN = 'rr:column' ############################################################################## ############################# D2RQ CONSTANTS ########################### @@ -70,9 +75,11 @@ YARRRML_MAPPINGS = 'mappings' YARRRML_SOURCES = 'sources' YARRRML_SOURCE = 'source' +YARRRML_TABLE = 'table' YARRRML_ACCESS = 'access' YARRRML_QUERY = 'query' YARRRML_REFERENCE_FORMULATION = 'referenceFormulation' +YARRRML_QUERY_FORMULATION = 'queryFormulation' YARRRML_ITERATOR = 'iterator' YARRRML_CREDENTIALS = 'credentials' YARRRML_TYPE = 'type' diff --git a/pretty_yarrrml2rml/pretty_yarrrml2rml/mapping.py b/yarrrml-translator/yarrrml_translator/mapping.py similarity index 94% rename from pretty_yarrrml2rml/pretty_yarrrml2rml/mapping.py rename to yarrrml-translator/yarrrml_translator/mapping.py index c0c5da3..ab85c3e 100644 --- a/pretty_yarrrml2rml/pretty_yarrrml2rml/mapping.py +++ b/yarrrml-translator/yarrrml_translator/mapping.py @@ -30,6 +30,8 @@ def add_prefix(data): template.append(RML_PREFIX + " ql: <" + QL_URI + ">.\n") if "d2rq" not in common_prefixes: template.append(RML_PREFIX + " d2rq: <" + D2RQ_URI + ">.\n") + if "foaf" not in common_prefixes: + template.append(RML_PREFIX + " foaf: <" + FOAF_URI + ">.\n") if "base" not in common_prefixes: template.append(RML_BASE + " <" + EXAMPLE_URI + ">.\n") template.append("\n\n") diff --git a/pretty_yarrrml2rml/pretty_yarrrml2rml/predicateobject.py b/yarrrml-translator/yarrrml_translator/predicateobject.py similarity index 91% rename from pretty_yarrrml2rml/pretty_yarrrml2rml/predicateobject.py rename to yarrrml-translator/yarrrml_translator/predicateobject.py index 73dfe13..476cde3 100644 --- a/pretty_yarrrml2rml/pretty_yarrrml2rml/predicateobject.py +++ b/yarrrml-translator/yarrrml_translator/predicateobject.py @@ -46,15 +46,15 @@ def get_graph_access(predicate_object_map): return graph_access -def add_predicate_object_maps(data, mapping): +def add_predicate_object_maps(data, mapping, mapping_format): po_template = "" pom_text = "\t" + R2RML_PREDICATE_OBJECT_MAP + " [\n" mapping_data = data.get(YARRRML_MAPPINGS).get(mapping) for predicate_object_map in mapping_data.get(get_predicate_object_access(mapping_data)): if type(predicate_object_map) is list: - po_template += pom_text + add_predicate_object(data, mapping, predicate_object_map) + "\n" + po_template += pom_text + add_predicate_object(data, mapping, predicate_object_map, mapping_format) + "\n" else: - po_template += pom_text + add_predicate_object(data, mapping, predicate_object_map, + po_template += pom_text + add_predicate_object(data, mapping, predicate_object_map, mapping_format, get_predicate_access(predicate_object_map), get_object_access(predicate_object_map), get_graph_access(predicate_object_map)) + "\n" @@ -102,7 +102,7 @@ def get_graph_list(predicate_object,graph_access): else: graphs = [] return graphs -def add_predicate_object(data, mapping, predicate_object, predicate_access=None, object_access=None, graph_access=None): +def add_predicate_object(data, mapping, predicate_object, mapping_format=RML_URI, predicate_access=None, object_access=None, graph_access=None): template = "" predicate_list = get_predicate_list(predicate_object, predicate_access) object_list = get_object_list(predicate_object, object_access) @@ -119,7 +119,7 @@ def add_predicate_object(data, mapping, predicate_object, predicate_access=None, object_value = om[0].split(YARRRML_IRI)[0] iri = True template += generate_rml_termmap(R2RML_OBJECT, R2RML_OBJECT_CLASS, - object_value, "\t\t\t") + object_value, "\t\t\t", mapping_format) if len(om) == 2: types = check_type(om[1]) if types != "error": @@ -133,7 +133,7 @@ def add_predicate_object(data, mapping, predicate_object, predicate_access=None, template = template[0:len(template) - 5] + "\t\t\t" + R2RML_TERMTYPE + " " \ + R2RML_IRI + "\n\t\t];\n" elif YARRRML_MAPPING in om: - template += join_mapping(data, mapping, om) + template += join_mapping(data, mapping, om, mapping_format) else: if YARRRML_VALUE in om: object_value = om.get(YARRRML_VALUE) @@ -143,7 +143,7 @@ def add_predicate_object(data, mapping, predicate_object, predicate_access=None, object_value = om.split(YARRRML_IRI)[0] iri = True template += generate_rml_termmap(R2RML_OBJECT, R2RML_OBJECT_CLASS, - object_value, "\t\t\t") + object_value, "\t\t\t", mapping_format) if type(om) == list and YARRRML_DATATYPE in om: template = template[0:len(template) - 5] + "\t\t\t" + R2RML_DATATYPE + " " \ + om.get(YARRRML_DATATYPE) + "\n\t\t];\n" @@ -161,7 +161,7 @@ def add_predicate_object(data, mapping, predicate_object, predicate_access=None, return template + "\t];" -def join_mapping(data, mapping, om): +def join_mapping(data, mapping, om, mapping_format): list_mappings = [] template = "" for mappings in data.get(YARRRML_MAPPINGS): @@ -172,7 +172,10 @@ def join_mapping(data, mapping, om): if mapping_join in list_mappings: subject_list = add_subject(data, mapping_join) list_initial_sources = get_initial_sources(data) - source_list = add_source(data, mapping_join, list_initial_sources) + if mapping_format == R2RML_URI: + source_list = add_table(data, mapping_join, list_initial_sources) + else: + source_list = add_source(data, mapping_join, list_initial_sources) number_joins_rml = len(subject_list) * len(source_list) for i in range(number_joins_rml): @@ -195,7 +198,7 @@ def join_mapping(data, mapping, om): ";\n\t\t\t\t" + R2RML_PARENT + " " + parent + ";\n\t\t\t]; \n" else: - raise Exception("Error: more than two parameters in join condition in mapping1 " + mapping) + raise Exception("Error: more than two parameters in join condition in mapping " + mapping) template += "\t\t];\n" else: template += "\n\t\t]\n" diff --git a/pretty_yarrrml2rml/pretty_yarrrml2rml/source.py b/yarrrml-translator/yarrrml_translator/source.py similarity index 76% rename from pretty_yarrrml2rml/pretty_yarrrml2rml/source.py rename to yarrrml-translator/yarrrml_translator/source.py index 2157742..d852550 100644 --- a/pretty_yarrrml2rml/pretty_yarrrml2rml/source.py +++ b/yarrrml-translator/yarrrml_translator/source.py @@ -15,11 +15,12 @@ def get_sources(data, mapping): sources = data.get(YARRRML_MAPPINGS).get(mapping).get(YARRRML_SOURCES) elif YARRRML_SOURCE in data.get(YARRRML_MAPPINGS).get(mapping): sources = data.get(YARRRML_MAPPINGS).get(mapping).get(YARRRML_SOURCE) - if type(sources) is not list: - sources = [sources] else: raise Exception("ERROR: sources not defined in mapping " + mapping) + if type(sources) is not list: + sources = [sources] + return sources @@ -45,12 +46,42 @@ def add_source(data, mapping, list_initial_sources): return final_list +def add_table(data, mapping, list_initial_sources): + table_template = "\t" + R2RML_LOGICAL_TABLE + " [\n\t\ta " + R2RML_LOGICAL_TABLE_CLASS + \ + ";\n\t\t" + + final_list = [] + sources = get_sources(data, mapping) + for source in sources: + sql_version = False + if source in list_initial_sources: + source = data.get(YARRRML_SOURCES).get(source) + + if YARRRML_ACCESS in source and YARRRML_QUERY in source: + r2rml_access = database_source(mapping, source) + sql_version = True + elif YARRRML_QUERY in source: + r2rml_access = R2RML_SQL_QUERY + " \"" + source.get(YARRRML_QUERY).replace("\n", " ").replace("\"", "\\\"") + "\"" + elif YARRRML_TABLE in source: + r2rml_access = R2RML_TABLE_NAME + " \"" + source.get(YARRRML_TABLE) + "\"" + else: + raise Exception("ERROR: table or query is not provided in " + source + " of mapping " + mapping) + if not sql_version: + if YARRRML_QUERY_FORMULATION in source: + r2rml_access += ";\n\t\t" + R2RML_SQL_VERSION + " rr:" + source.get(YARRRML_QUERY_FORMULATION).upper() + else: + r2rml_access += ";\n\t\t" + R2RML_SQL_VERSION + " rr:SQL2008" + r2rml_access += "\n\t];\n" + final_list.append(table_template + r2rml_access) + return final_list + + def add_source_simplified(mapping, source): source_rdf = "" file_path = re.sub("~.*", "", source[0]) reference_formulation = source[0].split('~')[1] source_extension = file_path.split('.')[1] - ref_formulation_rml = reference_formulation.replace("json", "JSON").replace("csv", "CSV").replace("xpath", "XPath") + ref_formulation_rml = reference_formulation.replace("jsonpath", "JSONPath").replace("csv", "CSV").replace("xpath", "XPath") if switch_in_reference_formulation(reference_formulation) != source_extension: raise Exception( "ERROR: mismatch extension and referenceFormulation in source " + source + " in mapping " + mapping) @@ -77,7 +108,7 @@ def add_source_full(mapping, source): if YARRRML_REFERENCE_FORMULATION in source: reference_formulation = str(source.get(YARRRML_REFERENCE_FORMULATION)) format_from_reference = switch_in_reference_formulation(reference_formulation.lower()) - ref_formulation_rml = reference_formulation.replace("json", "JSON").replace("csv", "CSV").replace("xpath", + ref_formulation_rml = reference_formulation.replace("jsonpath", "JSONPath").replace("csv", "CSV").replace("xpath", "XPath") if extension != format_from_reference or format_from_reference == "ERROR": raise Exception("ERROR: not referenceFormulation found or mismatch between the format and " @@ -114,15 +145,19 @@ def database_source(mapping, source): username = source.get(YARRRML_CREDENTIALS).get(YARRRML_USERNAME) password = source.get(YARRRML_CREDENTIALS).get(YARRRML_PASSWORD) hash_datasource = abs(hash(access + type + username + password)) - source_rdf += "<#DataSource_" + str(hash_datasource) + ">;\n\t\t" \ - + R2RML_SQL_VERSION + " rr:SQL2008;\n\t\t" \ - + R2RML_SQL_QUERY + " \"" + source.get(YARRRML_QUERY) + "\";\n\t\t" + source_rdf += "<#DataSource_" + str(hash_datasource) + ">;\n\t\t" + if YARRRML_QUERY in source: + source_rdf += R2RML_SQL_QUERY + " \"" + source.get(YARRRML_QUERY) + "\"" + elif YARRRML_TABLE in source: + source_rdf += R2RML_TABLE_NAME + " \"" + source.get(YARRRML_TABLE) + "\"" if YARRRML_REFERENCE_FORMULATION in source: - source_rdf += "\t\t" + RML_REFERENCE_FORMULATION + " ql:" \ - + switch_in_reference_formulation( - source.get(YARRRML_REFERENCE_FORMULATION)) + "\n\t];\n" + source_rdf += ";\n\t\t" + RML_REFERENCE_FORMULATION + " ql:" \ + + switch_in_reference_formulation(source.get(YARRRML_REFERENCE_FORMULATION)) + if YARRRML_QUERY_FORMULATION in source: + source_rdf += ";\n\t\t" + R2RML_SQL_VERSION + " rr:" + source.get(YARRRML_QUERY_FORMULATION).upper() else: - source_rdf += "\n\t];\n" + source_rdf += ";\n\t\t" + R2RML_SQL_VERSION + " rr:SQL2008" + source_rdf += "\n\t];\n" else: raise Exception("ERROR: no credentials to get access to source in mapping " + mapping) else: @@ -154,7 +189,7 @@ def generate_database_connections(data): for mapping in data.get(YARRRML_MAPPINGS): sources = get_sources(data, mapping) for source in sources: - if YARRRML_ACCESS and YARRRML_QUERY in source: + if YARRRML_QUERY in source and YARRRML_ACCESS in source: type = source.get(YARRRML_TYPE) if type == "mysql": driver = "com.mysql.jdbc.Driver" diff --git a/pretty_yarrrml2rml/pretty_yarrrml2rml/subject.py b/yarrrml-translator/yarrrml_translator/subject.py similarity index 100% rename from pretty_yarrrml2rml/pretty_yarrrml2rml/subject.py rename to yarrrml-translator/yarrrml_translator/subject.py diff --git a/pretty_yarrrml2rml/pretty_yarrrml2rml/termmap.py b/yarrrml-translator/yarrrml_translator/termmap.py similarity index 67% rename from pretty_yarrrml2rml/pretty_yarrrml2rml/termmap.py rename to yarrrml-translator/yarrrml_translator/termmap.py index b188276..2f2b1ad 100644 --- a/pretty_yarrrml2rml/pretty_yarrrml2rml/termmap.py +++ b/yarrrml-translator/yarrrml_translator/termmap.py @@ -2,10 +2,13 @@ ## return the type of TermMap based on the input text -def get_termmap_type(text): +def get_termmap_type(text, mapping_format): if "$(" in text and ")" in text: if text[0] == "$" and text[len(text) - 1] == ")" and text.count("$(") == 1: - return RML_REFERENCE + if mapping_format == R2RML_URI: + return R2RML_COLUMN + else: + return RML_REFERENCE else: return R2RML_TEMPLATE else: @@ -13,13 +16,13 @@ def get_termmap_type(text): ## Generates a TermMap (subject, predicate, object) based on the property, class and the text -def generate_rml_termmap(rml_property, rml_class, text, identation): +def generate_rml_termmap(rml_property, rml_class, text, identation, mapping_format=RML_URI): template = identation[0:-1] + rml_property + " [\n"+identation+"a " + rml_class + ";\n" + identation - term_map = get_termmap_type(text) + term_map = get_termmap_type(text, mapping_format) if term_map == R2RML_TEMPLATE: text = text.replace("$(", "{") text = text.replace(")", "}") - elif term_map == RML_REFERENCE: + elif term_map == RML_REFERENCE or term_map == R2RML_COLUMN: text = text.replace("$(", "") text = text.replace(")", "") elif term_map == R2RML_CONSTANT and text == "a": @@ -27,7 +30,10 @@ def generate_rml_termmap(rml_property, rml_class, text, identation): if term_map != "rr:constant": template += term_map + " \"" + text + "\";\n"+identation[0:-1]+"];\n" else: - template += term_map + " " + text + ";\n"+identation[0:-1]+"];\n" + if text.startswith("http"): + template += term_map + " <" + text + ">;\n" + identation[0:-1] + "];\n" + else: + template += term_map + " " + text + ";\n"+identation[0:-1]+"];\n" return template	Fixed a typo in read me Typo in rml:referenceFormulation ql:JSONPath; Hi, When we convert to RML file, after the source at rml:referenceFormulation we have `rml:referenceFormulation ql:JSONpath;` where `p` is small in `ql:JSONpath` Please fix it to `ql"JSONPath` because `P` has to be capital. Thanks		2022-10-27T07:57:07	[]	[]
oeg-upm/yatter	oeg-upm__yatter-22	074c3153142d14e02247d31cbd44fae5b9d1f01d	diff --git a/.gitignore b/.gitignore new file mode 100644 index 0000000..d63758b --- /dev/null +++ b/.gitignore @@ -0,0 +1,3 @@ +.DS_Store +lib/__pychache__ +.idea \ No newline at end of file diff --git a/lib/__pycache__/mapping.cpython-38.pyc b/lib/__pycache__/mapping.cpython-38.pyc deleted file mode 100644 index 81ec0f4..0000000 Binary files a/lib/__pycache__/mapping.cpython-38.pyc and /dev/null differ diff --git a/lib/__pycache__/predicateobject.cpython-38.pyc b/lib/__pycache__/predicateobject.cpython-38.pyc deleted file mode 100644 index c8fe2f8..0000000 Binary files a/lib/__pycache__/predicateobject.cpython-38.pyc and /dev/null differ diff --git a/lib/__pycache__/source.cpython-38.pyc b/lib/__pycache__/source.cpython-38.pyc deleted file mode 100644 index 2a6d436..0000000 Binary files a/lib/__pycache__/source.cpython-38.pyc and /dev/null differ diff --git a/lib/__pycache__/subject.cpython-38.pyc b/lib/__pycache__/subject.cpython-38.pyc deleted file mode 100644 index 7f0afd7..0000000 Binary files a/lib/__pycache__/subject.cpython-38.pyc and /dev/null differ diff --git a/lib/constants.py b/lib/constants.py new file mode 100644 index 0000000..77d6aa7 --- /dev/null +++ b/lib/constants.py @@ -0,0 +1,105 @@ + + + +############################################################################## +############################# RML CONSTANTS ############################### +############################################################################## + +RML_URI = 'http://semweb.mmlab.be/ns/rml#' +R2RML_URI = 'http://www.w3.org/ns/r2rml#' +RDF_URI = 'http://www.w3.org/1999/02/22-rdf-syntax-ns#' +RDF_TYPE = 'rdf:type' + +RML_PREFIX = '@prefix' +RML_TRIPLES_MAP = 'rr:TriplesMap' + + +RML_LOGICAL_SOURCE = 'rml:LogicalSource' +RML_LOGICAL_SOURCE_PROPERTY = 'rml:logicalSource' +RML_SOURCE = 'rml:source' +RML_REFERENCE_FORMULATION = 'rml:referenceFormulation' +RML_ITERATOR = 'rml:iterator' +RML_REFERENCE = 'rml:reference' + + + + +R2RML_TEMPLATE = 'rr:template' +R2RML_CONSTANT = 'rr:constant' +R2RML_SUBJECT_PROPERTY = 'rr:subjectMap' +R2RML_SUBJECT_CLASS = 'rr:SubjectMap' +R2RML_SQL_VERSION = 'rr:sqlVersion' +R2RML_SQL_QUERY = 'rr:sqlQuery' +R2RML_PREDICATE_OBJECT_MAP = 'rr:predicateObjectMap' +R2RML_PREDICATE_OBJECT_MAP_CLASS = 'rr:PredicateObjectMap' +R2RML_SHORTCUT_PREDICATE = 'rr:predicate' +R2RML_PREDICATE = 'rr:predicateMap' +R2RML_PREDICATE_CLASS = 'rr:PredicateMap' +R2RML_SHORTCUT_OBJECT = 'rr:object' +R2RML_OBJECT = 'rr:objectMap' +R2RML_OBJECT_CLASS = 'rr:ObjectMap' +R2RML_DATATYPE = 'rr:datatype' +R2RML_TERMTYPE = 'rr:termType' +R2RML_LANGUAGE = 'rr:language' +R2RML_IRI = 'rr:IRI' +R2RML_REFOBJECT_CLASS = 'rr:RefObjectMap' +R2RML_PARENT_TRIPLESMAP = 'rr:parentTriplesMap' +R2RML_JOIN_CONITION = 'rr:joinCondition' +R2RML_CHILD = 'rr:child' +R2RML_PARENT = 'rr:parent' + +############################################################################## +############################# D2RQ CONSTANTS ########################### +############################################################################## +D2RQ_DATABASE_CLASS = 'd2rq:Database' +D2RQ_DSN = 'd2rq:jdbcDSN' +D2RQ_DRIVER = 'd2rq:jdbcDriver' +D2RQ_USER = 'd2rq:username' +D2RQ_PASS = 'd2rq:password' + + +############################################################################## +############################# YARRRML CONSTANTS ########################### +############################################################################## + +YARRRML_PREFIXES = 'prefixes' +YARRRML_MAPPINGS = 'mappings' + +YARRRML_SOURCES = 'sources' +YARRRML_SOURCE = 'source' +YARRRML_ACCESS = 'access' +YARRRML_QUERY = 'query' +YARRRML_REFERENCE_FORMULATION = 'referenceFormulation' +YARRRML_ITERATOR = 'iterator' +YARRRML_CREDENTIALS = 'credentials' +YARRRML_TYPE = 'type' +YARRRML_USERNAME = 'username' +YARRRML_PASSWORD = 'password' + +YARRRML_SHORTCUT_SUBJECTS = 's' +YARRRML_SUBJECT = 'subject' +YARRRML_SUBJECTS = 'subjects' + + +YARRRML_PREDICATEOBJECT = 'predicateobjects' +YARRRML_SHORTCUT_PREDICATEOBJECT = 'po' + +YARRRML_PREDICATES = 'predicates' +YARRRML_SHORTCUT_PREDICATES = 'p' +YARRRML_OBJECT = 'objects' +YARRRML_SHORTCUT_OBJECT = 'o' +YARRRML_VALUE = 'value' +YARRRML_MAPPING = 'mapping' +YARRRML_DATATYPE = 'datatype' +YARRRML_LANGUAGE = 'language' + +YARRRML_CONDITION = 'condition' +YARRRML_PARAMETERS = 'parameters' + +YARRRML_IRI = '~iri' +YARRRML_LANG = '~lang' + + + + + diff --git a/lib/main.py b/lib/main.py index 5a54367..7f47606 100644 --- a/lib/main.py +++ b/lib/main.py @@ -1,80 +1,90 @@ import yaml import sys -import os -import mapping as mappingmod -import source as sourcemod -import subject as subjectmod -import predicateobject as predicateobjectmod +from rdflib import Graph +import mapping as mapping_mod +import source as source_mod +import subject as subject_mod +import predicateobject as predicate_object_mod -if __name__ =="__main__": - if len(sys.argv)==1: - print("Introduce the .yml file you want to convert to RML") - file = input() - #file="../test/gtfs-csv.yml" - print("\n") - #../test/example1/rml/mapping.yml - with open(file) as f: - print("Name for your file:") - end=input() - #end="prueba" - if(end[:-4]!=".rml"): - nuevo_fich=open(end+".rml","a") - else: - nuevo_fich=open(end,"a") - data = yaml.safe_load(f) - yaml_stream = True - - elif len(sys.argv)==5 and sys.argv[1]=="-m" and sys.argv[3]=="-o": - file=sys.argv[2] + +def run_parsing_system_inputs(): + if len(sys.argv) == 1: + file = input("Introduce the .yml file you want to convert to RML") with open(file) as f: - end=sys.argv[4] - if(end[:-4]!=".rml"): - nuevo_fich=open(end+".rml","a") - else: - nuevo_fich=open(end,"a") - data = yaml.safe_load(f) - yaml_stream = True + yaml_data = yaml.safe_load(f) + + elif len(sys.argv) == 5 and sys.argv[1] == "-m" and sys.argv[3] == "-o": + with open(sys.argv[2]) as f: + yaml_data = yaml.safe_load(f) + else: + sys.tracebacklimit = 0 + raise Exception( + "\n####################################\nERROR: Wrong argument input. You can:" + "\n-Use no arguments\n-Use arguments (in this order): -m yarrrml.yml -o mapping.rml.ttl" + "\n####################################\n") + return yaml_data + + +def open_inputs(yarrml_path): + with open(yarrml_path) as f: + yaml_data = yaml.safe_load(f) + return yaml_data + + +def write_results(rml_content): + if len(sys.argv) == 1: + rml_output_path = input("Name the path for the output file:") else: - sys.tracebacklimit=0 - raise Exception("\n####################################\nERROR: Wrong argument input. You can:\n-Use no arguments\n-Use arguments: -m YARRRMLfile -o RMLFile\n####################################\n") + rml_output_path = sys.argv[4] + rml_output_file = open(rml_output_path, "a") + rml_output_file.write(rml_content) + rml_output_file.close() - print("------------------------START RML-------------------------------") - mapping_list=[] - list_initial_sources= sourcemod.getInitialSources(data) - final="" - final = mappingmod.addPrefix(data) + print("Validating the generated RDF file with RDFLib") try: - for mapp in data.get("mappings"): - subject_list=[] - subject_list = subjectmod.addSubject(data, mapp) - source_list=[] - source_list=sourcemod.addSource(data,mapp,list_initial_sources) - pred =predicateobjectmod.addPredicateObject(data,mapp) + graph = Graph() + graph.parse(rml_output_path, format="turtle") + except Exception as e: + print("------------------------ERROR-------------------------------") + print("File not created: " + str(e)) + sys.exit() + +def translate(yarrrml_data): + print("------------------------START TRANSLATING YARRRML TO RML-------------------------------") + list_initial_sources = source_mod.get_initial_sources(yarrrml_data) + final = mapping_mod.add_prefix(yarrrml_data) + try: + for map in yarrrml_data.get("mappings"): + subject_list = subject_mod.add_subject(yarrrml_data, map) + source_list = source_mod.add_source(yarrrml_data, map, list_initial_sources) + pred = predicate_object_mod.add_predicate_object_maps(yarrrml_data, map) it = 0 for source in source_list: for subject in subject_list: - map=mappingmod.addMapping(data,mapp,it) + map_aux = mapping_mod.add_mapping(map, it) if type(source) is list: - final += map + source[0] +subject + pred + source[1] + final += map_aux + source[0] + subject + pred + source[1] else: - final += map + source +subject + pred - final = final[:-3] - final+= ".\n\n\n" - it=it+1 - nuevo_fich.write(final) + final += map_aux + source + subject + pred + final = final[:-2] + final += ".\n\n\n" + it = it + 1 + + print("RML content successfully created!") except Exception as e: - print(str(e)) - print("------------------------END RML-------------------------------") - print("FILE NOT SUCCESSFULY CREATED") - if(".rml" in end): - os.remove(end) - else: - os.remove(end+".rml") - sys.exit() - - - print("------------------------END RML-------------------------------") - print("FILE SUCCESSFULY CREATED!") + print("------------------------ERROR-------------------------------") + print("RML content not generated: "+str(e)) + sys.exit() + + return final + + +if __name__ == "__main__": + yarrrml_data = run_parsing_system_inputs() + rml_content = translate(yarrrml_data) + write_results(rml_content) + + print("------------------------END TRANSLATION-------------------------------") diff --git a/lib/mapping.py b/lib/mapping.py index 415fe4f..6b211fd 100644 --- a/lib/mapping.py +++ b/lib/mapping.py @@ -1,21 +1,38 @@ -import yaml -def addMapping(data, mapping,it): - map_template="<#"+mapping+"_"+str(it)+"> a rr:TriplesMap;\n\n" +import constants + + +def add_mapping(mapping, it): + map_template = "<#" + mapping + "_" + str(it) + "> a " + constants.RML_TRIPLES_MAP + ";\n\n" return map_template -def addPrefix(data): - template="" - if "prefixes" in data: - prefixes=data.get("prefixes") + +def add_prefix(data): + template = "" + common_prefixes = [] + if constants.YARRRML_PREFIXES in data: + prefixes = data.get(constants.YARRRML_PREFIXES) for prefix in prefixes: - template+="@prefix "+ prefix +": <"+data.get("prefixes").get(prefix)+">.\n" - if "rr" not in data.get("prefixes"): - template+="@prefix rr: <http://www.w3.org/ns/r2rml#>.\n" - if "rml" not in data.get("prefixes"): - template+="@prefix rml: <http://semweb.mmlab.be/ns/rml#>.\n" - if "rdf" not in data.get("prefixes"): - template+="@prefix rdf: <http://www.w3.org/1999/02/22-rdf-syntax-ns#>.\n" - template+="\n\n" - return template - else: - return template + prefix_uri = data.get(constants.YARRRML_PREFIXES).get(prefix) + check_common_prefixes(prefix_uri, common_prefixes) + template += constants.RML_PREFIX + " " + prefix + ": <" + prefix_uri + ">.\n" + + if "r2rml" not in common_prefixes: + template += constants.RML_PREFIX + " rr: <" + constants.R2RML_URI + ">.\n" + if "rml" not in common_prefixes: + template += constants.RML_PREFIX + " rml: <" + constants.RML_URI + ">.\n" + if "rdf" not in common_prefixes: + template += constants.RML_PREFIX + " rdf: <" + constants.RDF_URI + ">.\n" + + template += "\n\n" + + return template + + +def check_common_prefixes(prefix_uri, common_prefixes): + + if prefix_uri == constants.R2RML_URI: + common_prefixes.append("r2rml") + elif prefix_uri == constants.RML_URI: + common_prefixes.append("rml") + elif prefix_uri == constants.RDF_URI: + common_prefixes.append("rdf") diff --git a/lib/predicateobject.py b/lib/predicateobject.py index 3fc7a6d..6e91142 100644 --- a/lib/predicateobject.py +++ b/lib/predicateobject.py @@ -1,475 +1,180 @@ -import yaml -import source as sourcemod -import subject as subjectmod +import constants +import source +import subject +import termmap -def addPredicateObject(data,mapping): - po_template = "" - if ("predicateobjects" in data.get("mappings").get(mapping)): - for predob in data.get("mappings").get(mapping).get("predicateobjects"): - if "predicates" in predob: - #full con acceso a predicates - if "objects" in predob: - po_template+=addPredicateObjectFull(data,mapping,predob,"predicates","objects") - elif "o" in predob: - po_template+= addPredicateObjectFull(data,mapping,predob,"predicates","o") - - elif "p" in predob: - #full con acceso a p - if "objects" in predob: - po_template+=addPredicateObjectFull(data,mapping,predob,"p","objects") - elif "o" in predob: - po_template+= addPredicateObjectFull(data,mapping,predob,"p","o") - else: - if type(predob) is list: - po_template+=addPredicateObjectSimplified(data,mapping,predob) - else: - #ERROR - raise Exception("Error: Incorrect predicateObject Map in mapping " + mapping) - elif ("po" in data.get("mappings").get(mapping)): - for predob in data.get("mappings").get(mapping).get("po"): - if "predicates" in predob: - #full - if "objects" in predob: - po_template+=addPredicateObjectFull(data,mapping,predob,"predicates","objects") - elif "o" in predob: - po_template+= addPredicateObjectFull(data,mapping,predob,"predicates","o") - elif "p" in predob: - #full con acceso a p - if "objects" in predob: - po_template+=addPredicateObjectFull(data,mapping,predob,"p","objects") - elif "o" in predob: - po_template+= addPredicateObjectFull(data,mapping,predob,"p","o") - else: - if type(predob) is list: - po_template+=addPredicateObjectSimplified(data,mapping,predob) - else: - #ERROR - raise Exception("Error: Incorrect predicateObject Map in mapping " + mapping) - return po_template +def get_object_access(predicate_object_map): + if constants.YARRRML_OBJECT in predicate_object_map: + object_access = constants.YARRRML_OBJECT + elif constants.YARRRML_SHORTCUT_OBJECT in predicate_object_map: + object_access = constants.YARRRML_SHORTCUT_OBJECT + else: + raise Exception("There isn't an object key correctly specify in " + predicate_object_map) + return object_access -def addPredicateObjectFull(data,mapping,predob,access,accessObject): - template="" - #COMPROBAR SI ES LISTA PRIMERO - if(type(predob.get(access))!= list): - template="\trr:predicateObjectMap [\n" + "\t\ta rr:PredicateObjectMap;\n\t\trr:predicate " + predob.get(access) +";\n" - if accessObject in predob: - if "value" in predob.get(accessObject): - template+="\t\trr:objectMap [ \n\t\t\ta rr:ObjectMap;\n" - object=predob.get(accessObject).get("value") - termMap=getTermMap(object) - if(termMap=='rr:template "'): - object=object.replace("$",'') - object=object.replace("(","{") - object=object.replace(")",'}') - else: - object=object.replace("$(",'') - object=object.replace(")",'') - if(termMap!='rr:constant '): - template+="\t\t\t"+ termMap + object + '"'+ ';\n' - else: - template+="\t\t\t"+ termMap + object + ';\n' - #"\t\t]\n\t];\n\n" - if "datatype" in predob.get(accessObject): - template+= "\t\t\trr:datatype " + predob.get(accessObject).get("datatype") + "\n\t\t]\n\t];" - elif "type" in predob.get(accessObject): - template+= "\t\t\trr:TermType " + predob.get(accessObject).get("type") + "\n\t\t]\n\t];" - elif "language" in predob.get(accessObject): - template+= '\t\t\trr:language "' + predob.get(accessObject).get("language") + '"\n\t\t]\n\t];' - elif "mapping" in predob.get(accessObject): - template+= joinMapping(data,mapping,predob.get(accessObject)) - else: - if(type(predob.get(accessObject))!=list): - template+="\t\trr:objectMap [ \n\t\t\ta rr:ObjectMap;\n" - object=predob.get(accessObject) - termMap=getTermMap(object) - if(termMap=='rr:template "'): - object=object.replace("$",'') - object=object.replace("(","{") - object=object.replace(")",'}') - else: - object=object.replace("$(",'') - object=object.replace(")",'') - if(termMap!='rr:constant '): - template+="\t\t\t"+ termMap + object + '"'+ ';\n\t\t]\n\t];\n\n' - else: - template+="\t\t\t"+ termMap + object + ';\n\t\t]\n\t];\n\n' - else: - for objec in predob.get(accessObject): - if "mapping" in objec: - template+= joinMapping(data,mapping,objec) - continue - else: - template+="\t\trr:objectMap [ \n\t\t\ta rr:ObjectMap;\n" - if(type(objec) is list): - object = objec[0] - termMap=getTermMap(object) - if(termMap=='rr:template "'): - object=object.replace("$",'') - object=object.replace("(","{") - object=object.replace(")",'}') - else: - object=object.replace("$(",'') - object=object.replace(")",'') - else: - termMap=getTermMap(object) - if(termMap=='rr:template "'): - object=object.replace("$",'') - object=object.replace("(","{") - object=object.replace(")",'}') - else: - object=object.replace("$(",'') - object=object.replace(")",'') - if(termMap!='rr:constant '): - template+="\t\t\t"+ termMap + object + '"'+ ';\n' - else: - template+="\t\t\t"+ termMap + object + ';\n' - if type(objec)== list and len(objec)==2: - types = check_type(objec,2) - if(types!="error"): - if(types=="iri"): - template+="\t\t\trr:TermType rr:IRI"+"\n\t\t];\n" - elif(types=="language"): - lenguage=objec[1].replace("~lang","") - template+='\t\t\trr:language "'+ lenguage +'"\n\t\t];\n' - elif(types=="datatype"): - template+="\t\t\trr:datatype "+ objec[1] +"\n\t\t];\n" - else: - raise Exception("Error: incorrect format of predicateObjectMap in mapping " + mapping) - else: - template+="\t\t];\n" - template=template[:-2] - template+= "\n\t];\n\n" - -#CASO LISTA +def get_predicate_access(predicate_object_map): + if constants.YARRRML_PREDICATES in predicate_object_map: + predicate_access = constants.YARRRML_PREDICATES + elif constants.YARRRML_SHORTCUT_PREDICATES in predicate_object_map: + predicate_access = constants.YARRRML_SHORTCUT_PREDICATES else: - if(accessObject in predob): - if (type(predob.get(accessObject))==list): - template="\trr:predicateObjectMap [\n" + "\t\ta rr:PredicateObjectMap;\n" - for predic in predob.get(access): - template+= "\t\trr:predicate " + str(predic) + ";\n" - if(len(predob.get(accessObject))==2 and type(predob.get(accessObject)[1])!=list and type(predob.get(accessObject)[1])!=list): #solo un objeto con tipo - object=predob.get(accessObject)[0] - termMap=getTermMap(object) - if(termMap=='rr:template "'): - object=object.replace("$",'') - object=object.replace("(","{") - object=object.replace(")",'}') - else: - object=object.replace("$(",'') - object=object.replace(")",'') + raise Exception("There isn't a predicate key correctly specify in " + predicate_object_map) + return predicate_access - if(termMap!='rr:constant '): - template+="\t\t\t"+ termMap + object + '"'+ ';\n' - else: - template+="\t\t\t"+ termMap + object + ';\n' - types = check_type_simple(predob.get(accessObject)[1]) - if(types!="error"): - if(types=="iri"): - template+="\t\t\trr:TermType rr:IRI"+"\n\t\t];\n" - elif(types=="language"): - lenguage=predob.get(accessObject)[1].replace("~lang","") - template+='\t\t\trr:language "'+ lenguage +'"\n\t\t];\n' - elif(types=="datatype"): - template+="\t\t\trr:datatype "+ predob.get(accessObject)[1] +"\n\t\t];\n" - else: - raise Exception("Error: incorrect format of predicateObjectMap in mapping " + mapping) - template=template[:-2] - template+= "\n\t];\n\n" - else: - for objec in predob.get(accessObject): - template+="\t\trr:objectMap [ \n\t\t\ta rr:ObjectMap;\n" - if(type(objec) is list): - object = objec[0] - termMap=getTermMap(object) - if(termMap=='rr:template "'): - object=object.replace("$",'') - object=object.replace("(","{") - object=object.replace(")",'}') - else: - object=object.replace("$(",'') - object=object.replace(")",'') - else: - object=objec - termMap=getTermMap(object) - if(termMap=='rr:template "'): - object=object.replace("$",'') - object=object.replace("(","{") - object=object.replace(")",'}') - else: - object=object.replace("$(",'') - object=object.replace(")",'') - if(termMap!='rr:constant '): - template+="\t\t\t"+ termMap + object + '"'+ ';\n' - else: - template+="\t\t\t"+ termMap + object + ';\n' - if type(objec)== list and len(objec)==2: - types = check_type(objec,2) - if(types!="error"): - if(types=="iri"): - template+="\t\t\trr:TermType rr:IRI"+"\n\t\t];\n" - elif(types=="language"): - lenguage=objec[1].replace("~lang","") - template+='\t\t\trr:language "'+ lenguage +'"\n\t\t];\n' - elif(types=="datatype"): - template+="\t\t\trr:datatype "+ objec[1] +"\n\t\t];\n" - else: - raise Exception("Error: incorrect format of predicateObjectMap in mapping " + mapping) - else: - template+="\t\t];\n" - template=template[:-2] - template+= "\n\t];\n\n" - else: - if "value" in predob.get(accessObject): - template+="\t\trr:objectMap [ \n\t\t\ta rr:ObjectMap;\n" - object=predob.get(accessObject).get("value") - termMap=getTermMap(object) - if(termMap=='rr:template "'): - object=object.replace("$",'') - object=object.replace("(","{") - object=object.replace(")",'}') - else: - object=object.replace("$(",'') - object=object.replace(")",'') - if(termMap!='rr:constant '): - template+="\t\t\t"+ termMap + object + '"'+ ';\n' - else: - template+="\t\t\t"+ termMap + object + ';\n' - #"\t\t]\n\t];\n\n" - if "datatype" in predob.get(accessObject): - template+= "\t\t\trr:datatype " + predob.get(accessObject).get("datatype") + "\n\t\t]\n\t];\n\n" - elif "type" in predob.get(accessObject): - template+= "\t\t\trr:TermType " + predob.get(accessObject).get("type") + "\n\t\t]\n\t];\n\n" - elif "language" in predob.get(accessObject): - template+= '\t\t\trr:language "' + predob.get(accessObject).get("language") + '"\n\t\t]\n\t];\n\n' - elif "mapping" in predob.get(accessObject): - template+= joinMapping(data,mapping,predob.get(accessObject)) - else:#creo que sobra - template+="\t\trr:objectMap [ \n\t\t\ta rr:ObjectMap;\n" - object=predob.get(accessObject) - termMap=getTermMap(object) - if(termMap=='rr:template "'): - object=object.replace("$",'') - object=object.replace("(","{") - object=object.replace(")",'}') - else: - object=object.replace("$(",'') - object=object.replace(")",'') - if(termMap!='rr:constant '): - template+="\t\t\t"+ termMap + object + '"'+ ';\n\t\t]\n\t];\n\n' - else: - template+="\t\t\t"+ termMap + object + ';\n\t\t]\n\t];\n\n' - - return template -def addPredicateObjectSimplified(data,mapping,predob): - template= "\trr:predicateObjectMap [\n\t\ta rr:PredicateObjectMap;\n" - if((type(predob[0]) == list or type(predob[1])==list) and len(predob)==2): - if(type(predob[0])==list): #varios predicados - for predicate in predob[0]: - template+="\t\trr:predicate "+str(predicate)+";\n" - else: #un predicado - template+="\t\trr:predicate "+str(predob[0])+";\n" - if(type(predob[1])==list): #varios objetos - for object in predob[1]: - template+="\t\trr:objectMap [ \n\t\t\ta rr:ObjectMap;\n" - object=str(object) - termMap=getTermMap(object) - if(termMap=='rr:template "'): - object=object.replace("$",'') - object=object.replace("(","{") - object=object.replace(")",'}') - else: - object=object.replace("$(",'') - object=object.replace(")",'') - types=check_type_simple(object) - if(types=="error"): - if(termMap!='rr:constant '): - template+="\t\t\t"+ termMap + object + '"'+ ';\n' - else: - template+="\t\t\t"+ termMap + object + ';\n' - else: - if(types=="iri"): - object=object.replace("~iri","") - template+="\t\t\t"+ termMap + object + '";\n' - template+="\t\t\trr:TermType rr:IRI"+"\n\t\t]\n\t];\n\n" - elif(types=="language"): - lenguage=object.replace("~lang","") - template+='\t\t\trr:language "'+ lenguage +'"\n\t\t]\n\t];\n\n' - elif(types=="datatype"): - template+="\t\t\trr:datatype "+ object +"\n\t\t]\n\t];\n\n" - else: #un solo objeto - template+="\t\trr:objectMap [ \n\t\t\ta rr:ObjectMap;\n" - object=str(predob[1]) - termMap=getTermMap(object) - if(termMap=='rr:template "'): - object=object.replace("$",'') - object=object.replace("(","{") - object=object.replace(")",'}') - else: - object=object.replace("$(",'') - object=object.replace(")",'') - types=check_type_simple(object) - if(termMap!='rr:constant '): - template+="\t\t\t"+ termMap + object + '"'+ ';\n' - else: - template+="\t\t\t"+ termMap + object + ';\n' - if(types!="error"): - if(types=="iri"): - template+="\t\t\trr:TermType rr:IRI"+"\n\t\t]\n\t];\n\n" - elif(types=="language"): - lenguage=object.replace("~lang","") - template+='\t\t\trr:language "'+ lenguage +'"\n\t\t]\n\t];\n\n' - elif(types=="datatype"): - template+="\t\t\trr:datatype "+ object +"\n\t\t]\n\t];\n\n" +def get_predicate_object_access(mapping): + if constants.YARRRML_PREDICATEOBJECT in mapping: + predicate_object_access = constants.YARRRML_PREDICATEOBJECT + elif constants.YARRRML_SHORTCUT_PREDICATEOBJECT in mapping: + predicate_object_access = constants.YARRRML_SHORTCUT_PREDICATEOBJECT + else: + raise Exception("There isn't a predicate_object_map key correctly specify in " + mapping) + return predicate_object_access - template = template[:-2] - template+="\n\t];\n" +def add_predicate_object_maps(data, mapping): + po_template = "" + pom_text = "\t" + constants.R2RML_PREDICATE_OBJECT_MAP + " [\n" + mapping_data = data.get(constants.YARRRML_MAPPINGS).get(mapping) + for predicate_object_map in mapping_data.get(get_predicate_object_access(mapping_data)): + if type(predicate_object_map) is list: + po_template += pom_text + add_predicate_object(data, mapping, predicate_object_map) + "\n" + else: + po_template += pom_text + add_predicate_object(data, mapping, predicate_object_map, + get_predicate_access(predicate_object_map), + get_object_access(predicate_object_map)) + "\n" + return po_template - elif(len(predob)==2): - ##solo predicado y objeto singulares - template+="\t\trr:predicate "+str(predob[0])+";\n"+"\t\trr:objectMap [ \n\t\t\ta rr:ObjectMap;\n" - object=str(predob[1]) - if "~" in object: - termMap=getTermMap(object) - if(termMap=='rr:template "'): - object=object.replace("$",'') - object=object.replace("(","{") - object=object.replace(")",'}') - else: - object=object.replace("$(",'') - object=object.replace(")",'') - types=check_type_simple(object) - object1=object.split("~") - if(termMap!='rr:constant '): - template+="\t\t\t"+ termMap + object1[0] + '"'+ ';\n' - else: - template+="\t\t\t"+ termMap + object1[0] + ';\n' - if(types!="error"): - if(types=="iri"): - template+="\t\t\trr:TermType rr:IRI"+"\n\t\t]\n\t];\n\n" - elif(types=="language"): - lenguage=objec[1].replace("~lang","") - template+='\t\t\trr:language "'+ lenguage +'"\n\t\t]\n\t];\n\n' - elif(types=="datatype"): - template+="\t\t\trr:datatype "+ objec[1] +"\n\t\t]\n\t];\n\n" - else: - termMap=getTermMap(object) - if(termMap=='rr:template "'): - object=object.replace("$",'') - object=object.replace("(","{") - object=object.replace(")",'}') - else: - object=object.replace("$(",'') - object=object.replace(")",'') - if(termMap!='rr:constant '): - template+="\t\t\t"+ termMap + object + '"'+ '\n\t\t]\n\t];\n\n' - else: - template+="\t\t\t"+ termMap + object + '\n\t\t]\n\t];\n\n' +def get_object_list(predicate_object, object_access): + object_list = [] + if object_access is not None: + object_maps = predicate_object.get(object_access) + else: + object_maps = predicate_object[1] - elif(len(predob)==3): - #1 pred, 2 obj, 3 datatype, leng - template+="\t\trr:predicate "+str(predob[0])+";\n"+"\t\trr:objectMap [ \n\t\t\ta rr:ObjectMap;\n" - object=str(predob[1]) - termMap=getTermMap(object) - if(termMap=='rr:template "'): - object=object.replace("$",'') - object=object.replace("(","{") - object=object.replace(")",'}') - else: - object=object.replace("$(",'') - object=object.replace(")",'') - if(termMap!='rr:constant '): - template+="\t\t\t"+ termMap + object + '"'+ ';\n' + if type(object_maps) == list: + if len(predicate_object) == 3: + for i in range(len(object_maps)): + object_list.append([object_maps[i], predicate_object[2]]) else: - template+="\t\t\t"+ termMap + object + ';\n' - types = check_type(predob,3) - if(types!="error"): - template+="\t\t\trr:" + types +" "+ predob[2] +"\n\t\t]\n\t];\n\n" - else: - raise Exception("Error: incorrect format of predicateObjectMap in mapping " + mapping) + object_list.extend(object_maps) else: - #ERROR - raise Exception("Error: incorrect format of predicateObjectMap in mapping " + mapping) - - return template - - - + if len(predicate_object) == 3: + object_list.append([object_maps, predicate_object[2]]) + else: + object_list.append(object_maps) + return object_list -def check_type(predob,pos): - type=str(predob[pos-1]) - if "~iri" in type: - return "iri" - elif "~lang" in type: - return "language" - elif ("xsd:" in type) or ("ex:" in type): - return "datatype" +def get_predicate_list(predicate_object, predicate_access): + predicate_list = [] + if predicate_access is not None: + predicate_maps = predicate_object.get(predicate_access) else: - return "error" - -def check_type_simple(object): - - if "~iri" in object: - return "iri" - elif "~lang" in object: - return "language" - elif ("xsd:" in object) or ("ex:" in object): - return "datatype" + predicate_maps = predicate_object[0] + if type(predicate_maps) == list: + predicate_list.extend(predicate_maps) else: - return "error" - - -def joinMapping(data,mapping,objec): - list_mappings=[] - template="" - for mappings in data.get("mappings"): + predicate_list.append(predicate_maps) + return predicate_list + + +def add_predicate_object(data, mapping, predicate_object, predicate_access=None, object_access=None): + template = "" + predicate_list = get_predicate_list(predicate_object, predicate_access) + object_list = get_object_list(predicate_object, object_access) + + for pm in predicate_list: + template += termmap.generate_rml_termmap(constants.R2RML_PREDICATE, constants.R2RML_PREDICATE_CLASS, pm, + "\t\t\t") + for om in object_list: + iri = False + if type(om) == list: + object_value = om[0] + if constants.YARRRML_IRI in om[0]: + object_value = om[0].split(constants.YARRRML_IRI)[0] + iri = True + template += termmap.generate_rml_termmap(constants.R2RML_OBJECT, constants.R2RML_OBJECT_CLASS, + object_value, "\t\t\t") + if len(om) == 2: + types = termmap.check_type(om[1]) + if types != "error": + if types == constants.YARRRML_LANGUAGE: + template = template[0:len(template) - 5] + "\t\t\t" + constants.R2RML_LANGUAGE + " " \ + + om[1].replace(constants.YARRRML_LANG, "") + "\"\n\t\t];\n" + elif types == constants.YARRRML_DATATYPE: + template = template[0:len(template) - 5] + "\t\t\t" + constants.R2RML_DATATYPE + " " \ + + om[1] + "\n\t\t];\n" + if iri: + template = template[0:len(template) - 5] + "\t\t\t" + constants.R2RML_TERMTYPE + " " \ + + constants.R2RML_IRI + "\n\t\t];\n" + elif constants.YARRRML_MAPPING in om: + template += join_mapping(data, mapping, om) + else: + if constants.YARRRML_VALUE in om: + object_value = om.get(constants.YARRRML_VALUE) + else: + object_value = om + if constants.YARRRML_IRI in om: + object_value = om.split(constants.YARRRML_IRI)[0] + iri = True + template += termmap.generate_rml_termmap(constants.R2RML_OBJECT, constants.R2RML_OBJECT_CLASS, + object_value, "\t\t\t") + if type(om) == list and constants.YARRRML_DATATYPE in om: + template = template[0:len(template) - 5] + "\t\t\t" + constants.R2RML_DATATYPE + " " \ + + om.get(constants.YARRRML_DATATYPE) + "\n\t\t];\n" + if (type(om) == list and constants.YARRRML_TYPE in om and om.get(constants.YARRRML_TYPE) == "iri") or iri: + template = template[0:len(template) - 5] + "\t\t\t" + constants.R2RML_TERMTYPE + " " \ + + constants.R2RML_IRI + "\n\t\t];\n" + if type(om) == list and constants.YARRRML_LANGUAGE in om: + template = template[0:len(template) - 5] + "\t\t\t" + constants.R2RML_LANGUAGE + " \"" \ + + om.get(constants.YARRRML_LANGUAGE) + "\"\n\t\t];\n" + + return template + "\t];" + + +def join_mapping(data, mapping, om): + list_mappings = [] + template = "" + for mappings in data.get(constants.YARRRML_MAPPINGS): list_mappings.append(mappings) - if objec.get("mapping") in list_mappings: - subject_list = subjectmod.addSubject(data, objec.get("mapping")) - list_initial_sources= sourcemod.getInitialSources(data) - source_list=sourcemod.addSource(data,objec.get("mapping"),list_initial_sources) - - sumSourSub = len(subject_list) * len(source_list) - for i in range(sumSourSub): - template+="\t\trr:objectMap [ \n\t\t\ta rr:RefObjectMap;\n\t\t\trr:parentTriplesMap <#"+ objec.get("mapping")+"_"+str(i)+">;\n" - if "condition" in objec: - if "parameters" in objec.get("condition"): - list_parameters = objec.get("condition").get("parameters") - if len(list_parameters)==2: - child=list_parameters[0][1] - parent=list_parameters[1][1] - child=child.replace("$(",'"') - child=child.replace(")",'"') - parent=parent.replace("$(",'"') - parent=parent.replace(")",'"') - - template+='\t\t\trr:joinCondition [\n\t\t\t\trr:child ' + child + ';\n\t\t\t\trr:parent ' + parent + ';\n\t\t\t]\n\t\t];\n' + mapping_join = om.get(constants.YARRRML_MAPPING) + + if mapping_join in list_mappings: + subject_list = subject.add_subject(data, mapping_join) + list_initial_sources = source.get_initial_sources(data) + source_list = source.add_source(data, mapping_join, list_initial_sources) + + number_joins_rml = len(subject_list) * len(source_list) + for i in range(number_joins_rml): + template += "\t\t" + constants.R2RML_OBJECT + \ + " [ \n\t\t\ta " + constants.R2RML_REFOBJECT_CLASS + \ + ";\n\t\t\t" + constants.R2RML_PARENT_TRIPLESMAP + " <#" + mapping_join + "_" + str(i) + ">;\n" + if constants.YARRRML_CONDITION in om: + if constants.YARRRML_PARAMETERS in om.get(constants.YARRRML_CONDITION): + list_parameters = om.get(constants.YARRRML_CONDITION).get(constants.YARRRML_PARAMETERS) + if len(list_parameters) == 2: + child = list_parameters[0][1] + parent = list_parameters[1][1] + child = child.replace("$(", '"') + child = child.replace(")", '"') + parent = parent.replace("$(", '"') + parent = parent.replace(")", '"') + + template += "\t\t\t" + constants.R2RML_JOIN_CONITION + \ + " [\n\t\t\t\t" + constants.R2RML_CHILD + " " + child + \ + ";\n\t\t\t\t" + constants.R2RML_PARENT + " " + parent + ";\n\t\t\t]; \n\t\t];\n" else: - raise Exception("Error: more than two parameters in join condition in mapping1 " + mapping) + raise Exception("Error: more than two parameters in join condition in mapping1 " + mapping) else: - template+="\n\t\t]\n\t];" - return template - + template += "\n\t\t]\n" else: raise Exception("Error in reference mapping another mapping in mapping " + mapping) return template - - - -def getTermMap(text): - if("$(" in text and ")" in text): - if text[0]=="$": - return 'rml:reference "' - else: - return 'rr:template "' - else: - return 'rr:constant ' diff --git a/lib/source.py b/lib/source.py index 8f05355..19201ca 100644 --- a/lib/source.py +++ b/lib/source.py @@ -1,181 +1,158 @@ -import yaml +import re +import constants -def getInitialSources(data): - list_initial_sources =[] - if "sources" in data: - for y in data.get("sources"): +def get_initial_sources(data): + list_initial_sources = [] + if constants.YARRRML_SOURCES in data: + for y in data.get(constants.YARRRML_SOURCES): list_initial_sources.append(y) return list_initial_sources -def addSource(data, mapping,list_initial_sources): +def add_source(data, mapping, list_initial_sources): + source_template = "\t" + constants.RML_LOGICAL_SOURCE + " [\n\t\ta " + constants.RML_LOGICAL_SOURCE_PROPERTY + \ + ";\n\t\t" + constants.RML_SOURCE + " " + final_list = [] - source_template="" - list_sources=[] - final_list=[] - #primero se comprueba que este en la lista inicial de sources, sino luego se comprueba si es forma completa o simplificada - - #CASO INITIAL SOURCE: forma de string - if "sources" in data.get("mappings").get(mapping): - if(type(data.get("mappings").get(mapping).get("sources")) is str ): #una unica source - final_list.append(addInitialSource(data,data.get("mappings").get(mapping).get("sources"),mapping)) - else: - for sources in data.get("mappings").get(mapping).get("sources"): #varias sources - list_sources.append(sources) - for sources in list_sources: - if sources in list_initial_sources: - final_list.append(addInitialSource(data,sources,mapping)) - elif "source" in data.get("mappings").get(mapping): - if(type(data.get("mappings").get(mapping).get("source")) is str ): #una unica source - final_list.append(addInitialSource(data,data.get("mappings").get(mapping).get("source"),mapping)) - else: - for sources in data.get("mappings").get(mapping).get("source"): #varias sources - list_sources.append(sources) - for sources in list_sources: - if sources in list_initial_sources: - final_list.append(addInitialSource(data,sources,mapping)) + if constants.YARRRML_SOURCES in data.get(constants.YARRRML_MAPPINGS).get(mapping): + sources = data.get(constants.YARRRML_MAPPINGS).get(mapping).get(constants.YARRRML_SOURCES) + elif constants.YARRRML_SOURCE in data.get(constants.YARRRML_MAPPINGS).get(mapping): # don't find this exist + sources = data.get(constants.YARRRML_MAPPINGS).get(mapping).get(constants.YARRRML_SOURCE) else: raise Exception("ERROR: sources not defined in mapping " + mapping) - #CASO SIMPLIFIED SOURCE Y FULL SOURCE: no tiene forma de string - if "sources" in data.get("mappings").get(mapping): - if (type(data.get("mappings").get(mapping).get("sources")) is not str): - for sources in data.get("mappings").get(mapping).get("sources"): - if "access" in sources: - final_list.append(addSourceFull(data,mapping,sources)) - elif type(sources) is list: - final_list.append(addSourceSimplified(data,mapping,sources)) - elif sources not in list_initial_sources: - raise Exception("ERROR: source " + sources + " in mapping "+ mapping + " not valid") - else: - if (type(data.get("mappings").get(mapping).get("source")) is not str): - for sources in data.get("mappings").get(mapping).get("source"): - if "access" in sources: - final_list.append(addSourceFull(data,mapping,sources)) - elif type(sources) is list: - final_list.append(addSourceSimplified(data,mapping,sources)) - elif sources not in list_initial_sources: - raise Exception("ERROR: source " + sources + " in mapping "+ mapping + " not valid") - return final_list -def addInitialSource(data,sources,mapping): - if type(data.get("sources").get(sources)) is list: - return addSourceSimplified(data,mapping,data.get("sources").get(sources)) + for source in sources: + if source in list_initial_sources: + source = data.get(constants.YARRRML_SOURCES).get(source) - else: - return addSourceFull(data,mapping,data.get("sources").get(sources)) - - -def addSourceSimplified(data,mapping,sources): - source_template = "\trml:logicalSource [\n" +"\t\ta rml:logicalSource;\n" + "\t\trml:source " - source= sources[0] - source_aux = source.split('~') - source_aux1= source_aux[0].split('.') - source_name = source_aux1[0] - source_extension = source_aux1[1] - source_referenceFormulation = source_aux[1] - if(checkExtension(source_extension,source_referenceFormulation)==1): - raise Exception("ERROR: mismatch extension and referenceFormulation in source "+ sources + " in mapping "+mapping) - else: - if len(sources)== 1: - if (source_extension=="csv" or source_extension=="SQL2008"): - source_template+= '"'+source_aux[0]+'"' + ";\n" + "\t\trml:referenceFormulation ql:" + source_referenceFormulation + "\n" + "\t];\n" + if constants.YARRRML_ACCESS in source: + if constants.YARRRML_QUERY in source: + final_list.extend(source_template + database_source(mapping, source)) else: - raise Exception("ERROR: source " + sources + " in mapping " + mapping +" has no iterator") + final_list.append(source_template + add_source_full(mapping, source)) + elif type(source) is list: + final_list.append(source_template + add_source_simplified(mapping, source)) else: - source_delim = sources[1] - source_template+= '"'+source_aux[0]+'"' + ";\n" + "\t\trml:referenceFormulation ql:" + source_referenceFormulation + ";\n" + '\t\trml:iterator "' +source_delim + '"'+ ";\n\t];\n" - return source_template - - -def addSourceFull(data,mapping,sour): - source_final="" - if "query" in sour: - list_source=databaseSource(data,mapping,sour) - return list_source - list_sources=[] - list_sources.append("sources") - list_sources.append(data.get("mappings").get(mapping).get("sources")) - source_template = "\trml:logicalSource [\n" +"\t\ta rml:logicalSource;\n" + "\t\trml:source " - if "access" in sour: - source = str(sour.get("access")) - extension = source.split(".") - else: - raise Exception('ERROR: source' +sour.get("access") + 'in mapping ' + mapping+' has no "access"') - - if "referenceFormulation" in sour: - source_referenceFormulation = str(sour.get("referenceFormulation")) - if checkExtension(extension[1],source_referenceFormulation)==1: - raise Exception("ERROR: mismatch extension and referenceFormulation in source " + sour.get("access") + "in mapping " + mapping) - if getReferenceFOrmulation(source_referenceFormulation)!= "ERROR": - source_referenceFormulation=getReferenceFOrmulation(source_referenceFormulation) + raise Exception("ERROR: source " + source + " in mapping " + mapping + " not valid") + return final_list + + +def add_source_simplified(mapping, source): + source_rdf = "" + file_path = re.sub("~.*", "", source[0]) + reference_formulation = source[0].split('~')[1] + source_extension = file_path.split('.')[1] + ref_formulation_rml = reference_formulation.replace("json", "JSON").replace("csv", "CSV").replace("xpath", "XPath") + if switch_in_reference_formulation(reference_formulation) != source_extension: + raise Exception( + "ERROR: mismatch extension and referenceFormulation in source " + source + " in mapping " + mapping) else: - if(extension[1]=="csv"): - if "iterator" in sour: - source_iterator = str(sour.get("iterator")) - source_template += '"'+source+'"' + ";\n" + '\t\trml:iterator "' +source_iterator + '"\n\t];\n' - return source_template + if len(source) == 1: # si no tiene iterador + if source_extension == "csv" or source_extension == "SQL2008": + source_rdf += '"' + file_path + '"' + ";\n" + "\t\t" + constants.RML_REFERENCE_FORMULATION + " ql:" \ + + ref_formulation_rml + "\n" + "\t];\n" else: - source_template += '"'+source+'"' + ";\n" + '"\n\t];\n' - return source_template + raise Exception("ERROR: source " + source + " in mapping " + mapping + " has no iterator") + else: # source[1] es el iterador en json y xml + source_rdf += "\"" + file_path + "\";\n\t\t" + constants.RML_REFERENCE_FORMULATION + " ql:" \ + + ref_formulation_rml + ";\n\t\t" + constants.RML_ITERATOR + " \"" \ + + source[1] + "\";\n\t];\n" + return source_rdf + + +def add_source_full(mapping, source): + source_rdf = "" + + access = str(source.get(constants.YARRRML_ACCESS)) + extension = source.split(".")[1] + + if constants.YARRRML_REFERENCE_FORMULATION in source: + reference_formulation = str(source.get(constants.YARRRML_REFERENCE_FORMULATION)) + format_from_reference = switch_in_reference_formulation(reference_formulation.lower()) + ref_formulation_rml = reference_formulation.replace("json", "JSON").replace("csv", "CSV").replace("xpath","XPath") + if extension == format_from_reference or format_from_reference == "ERROR": + raise Exception("ERROR: not referenceFormulation found or mismatch between the format and " + "referenceFormulation in source " + access + "in mapping " + mapping) + if constants.YARRRML_ITERATOR in source: + source_iterator = str(source.get(constants.YARRRML_ITERATOR)) + + source_rdf += "\"" + access + "\";\n\t\t" + constants.RML_REFERENCE_FORMULATION + " ql:" \ + + ref_formulation_rml + ";\n\t\t" + constants.RML_ITERATOR + " \"" \ + + source_iterator + "\"\n\t];\n" else: - raise Exception("ERROR: source "+sour.get("access") + "in mapping " + mapping + " has no referenceFormulation") - if "iterator" in sour: - source_iterator = str(sour.get("iterator")) - source_template += '"'+source+'"' + ";\n" + "\t\trml:referenceFormulation ql:"+ source_referenceFormulation +";\n" + '\t\trml:iterator "' +source_iterator + '"\n\t];\n' + if extension == "csv" or extension == "SQL2008": + source_rdf += "\"" + access + "\";\n\t\t" + constants.RML_REFERENCE_FORMULATION + " ql:" \ + + ref_formulation_rml + ";\n\n\t];\n" + else: + raise Exception("ERROR: source " + access + "in mapping " + mapping + " has no referenceFormulation") + else: - if (extension[1]=="csv" or extension[1]=="SQL2008"): - source_template += '"'+source+'"' + ";\n" + "\t\trml:referenceFormulation ql:"+ source_referenceFormulation +";\n" + '"\n\t];\n' + if extension == "csv": + source_rdf += "\"" + access + "\";\n\n\t];\n" else: - raise Exception("ERROR: source "+sour.get("access") + " in mapping " + mapping + " has no iterator") - source_final += source_template - return source_final - -def checkExtension (extension,referenceFormulation): - switcher={ - "json":"jsonpath", - "csv":"csv", - "xml":"xpath" - #para querys - } - reference = switcher.get(extension,"ERROR") - if reference == referenceFormulation: - return 0 - else: - return 1 - -def getReferenceFOrmulation (referenceFormulation): - switcher={ - "jsonpath":"JSONPath", - "csv":"CSV", - "xpath":"XPath" - } - return switcher.get(referenceFormulation,"ERROR") - - -def databaseSource(data,mapping,source): - list=[] - templatelog="\trml:logicalSource [\n\t\ta rml:LogicalSource;\n" - if "access" in source: - if "credentials" in source: - if "type" in source and source.get("type")=="mysql": - templatelog+='\t\trml:source <#DataSource_'+mapping+'>;\n\t\trr:sqlVersion rr:SQL2008;\n\t\trml:query "'+source.get("query") +'";\n\t\t' - if "referenceFormulation" in source: - formu=getReferenceFOrmulation(source.get("referenceFormulation")) - templatelog+="\t\trml:referenceFormulation ql:"+ formu +"\n\t];\n" + raise Exception("ERROR: source " + access + "in mapping " + mapping + " has no referenceFormulation") + + return source_rdf + + +def database_source(mapping, source): + source_database = [] + source_rdf = "" + if constants.YARRRML_ACCESS in source: + if constants.YARRRML_CREDENTIALS in source: + if constants.YARRRML_TYPE in source: + source_rdf += "<#DataSource_" + mapping + ">;\n\t\t" \ + + constants.R2RML_SQL_VERSION + " rr:SQL2008;\n\t\t" \ + + constants.R2RML_SQL_QUERY + " \"" + source.get(constants.YARRRML_QUERY) + "\";\n\t\t" + if constants.YARRRML_REFERENCE_FORMULATION in source: + source_rdf += "\t\t" + constants.RML_REFERENCE_FORMULATION + " ql:" \ + + switch_in_reference_formulation( + source.get(constants.YARRRML_REFERENCE_FORMULATION)) + "\n\t];\n" else: - templatelog+="\n\t];\n" + source_rdf += "\n\t];\n" + source_database.append(source_rdf) + source_database.append(generate_database_connection(mapping, source)) else: - raise Exception("ERROR: no credentials to get access to source in mapping "+ mapping) + raise Exception("ERROR: no credentials to get access to source in mapping " + mapping) else: raise Exception("ERROR: no access to the source in mapping " + mapping) + return source_database - list.append(templatelog) - - templateDat = '<#DataSource_'+mapping+'> a d2rq:Database;\n\tdrr1:jdbcDSN "'+ source.get("access")+'";\n\td2rq:jdbcDriver "com.mysql.jdbc.Driver";\n\td2rq:username "'+source.get("credentials").get("username")+'";\n\td2rq:password "'+source.get("credentials").get("password")+'".\n\n' - list.append(templateDat) - - return list +def switch_in_reference_formulation(value): + value = value.lower() + if value == "csv": + switcher = value + elif "json" in value: + if "path" in value: + switcher = "json" + else: + switcher = "jsonpath" + elif "x" in value: + if "path" in value: + switcher = "xml" + else: + switcher = "xpath" + return switcher + + +def generate_database_connection(mapping, source): + type = source.get(constants.YARRRML_TYPE) + if type == "mysql": + driver = "com.mysql.jdbc.Driver" + elif type == "postgresql": + driver = "org.postgresql.Driver" + elif type == "sqlserver": + driver = "com.microsoft.sqlserver.jdbc.SQLServerDriver" + else: + driver = "" + + database = "<#DataSource_" + mapping + "> a " + constants.D2RQ_DATABASE_CLASS + ";\n\t" \ + + constants.D2RQ_DSN + " \"" + source.get(constants.YARRRML_ACCESS) + "\";\n\t" \ + + constants.D2RQ_DRIVER + " \"" + driver + "\";\n\t" \ + + constants.D2RQ_USER + " \"" + source.get(constants.YARRRML_CREDENTIALS).get(constants.YARRRML_USERNAME) + "\";\n\t" \ + + constants.D2RQ_PASS + " \"" + source.get(constants.YARRRML_CREDENTIALS).get(constants.YARRRML_PASSWORD) + "\".\n\n" + return database diff --git a/lib/subject.py b/lib/subject.py index 9ce2375..3df859c 100644 --- a/lib/subject.py +++ b/lib/subject.py @@ -1,121 +1,40 @@ -import yaml +import sys -def addSubject(data, mapping): - list_subject=[] - final_list=[] +import constants +import termmap - if ("s" in data.get("mappings").get(mapping)): - list_subject=data.get("mappings").get(mapping).get("s") - if (type(list_subject)== list): #varios sujetos - for subject in list_subject: - subject_template = "\trr:subjectMap [\n" + "\t\ta rr:SubjectMap;\n"+ "\t\t" - termMap=getTermMap(subject) - if(termMap=='rr:template "'): - subject=subject.replace("$",'') - subject=subject.replace("(","{") - subject=subject.replace(")",'}') - else: - subject=subject.replace("$(",'') - subject=subject.replace(")",'') - if(termMap!='rr:constant '): - subject_template+= termMap + ' "'+ subject + '"'+ "\n\t];\n\n" - else: - subject_template+= termMap + subject + "\n\t];\n\n" - final_list.append(subject_template) +## Adds a subject or set of subjects to the given TriplesMap +def add_subject(data, mapping): + subjects = [] + rml_subjects = [] + only_one = False - else: #un solo sujeto - subject_template = "\trr:subjectMap [\n" + "\t\ta rr:SubjectMap;\n"+ "\t\t" - subject=list_subject - termMap=getTermMap(subject) - if(termMap=='rr:template "'): - subject=subject.replace("$",'') - subject=subject.replace("(","{") - subject=subject.replace(")",'}') - else: - subject=subject.replace("$(",'') - subject=subject.replace(")",'') - if(termMap!='rr:constant '): - subject_template+= termMap + ' "'+ subject + '"'+ "\n\t];\n\n" - else: - subject_template+= termMap + subject + "\n\t];\n\n" - final_list.append(subject_template) - - - elif ("subjects" in data.get("mappings").get(mapping)): - list_subject= data.get("mappings").get(mapping).get("subjects") - if (type(list_subject)== list): #varios sujetos - for subject in list_subject: - subject_template = "\trr:subjectMap [\n" + "\t\ta rr:SubjectMap;\n"+ "\t\t" - termMap=getTermMap(subject) - if(termMap=='rr:template "'): - subject=subject.replace("$",'') - subject=subject.replace("(","{") - subject=subject.replace(")",'}') - else: - subject=subject.replace("$(",'') - subject=subject.replace(")",'') - if(termMap!='rr:constant '): - subject_template+= termMap + ' "'+ subject + '"'+ "\n\t];\n\n" - else: - subject_template+= termMap + subject + "\n\t];\n\n" - final_list.append(subject_template) - - - else: #un solo sujeto - subject_template = "\trr:subjectMap [\n" + "\t\ta rr:SubjectMap;\n"+ "\t\t" - subject=list_subject - termMap=getTermMap(subject) - if(termMap=='rr:template "'): - subject=subject.replace("$",'') - subject=subject.replace("(","{") - subject=subject.replace(")",'}') - else: - subject=subject.replace("$(",'') - subject=subject.replace(")",'') - if(termMap!='rr:constant '): - subject_template+= termMap + ' "'+ subject + '"'+ "\n\t];\n\n" - else: - subject_template+= termMap + subject + "\n\t];\n\n" - final_list.append(subject_template) - + if constants.YARRRML_SHORTCUT_SUBJECTS in data.get(constants.YARRRML_MAPPINGS).get(mapping): + subject = data.get(constants.YARRRML_MAPPINGS).get(mapping).get(constants.YARRRML_SHORTCUT_SUBJECTS) + elif constants.YARRRML_SUBJECTS in data.get(constants.YARRRML_MAPPINGS).get(mapping): + subject = data.get(constants.YARRRML_MAPPINGS).get(mapping).get(constants.YARRRML_SUBJECTS) + elif constants.YARRRML_SUBJECT in data.get(constants.YARRRML_MAPPINGS).get(mapping): + subject = data.get(constants.YARRRML_MAPPINGS).get(mapping).get(constants.YARRRML_SUBJECT) + only_one = True + else: + raise Exception("ERROR: no subjects in mapping " + mapping) - elif ("subject" in data.get("mappings").get(mapping)): - list_subject=data.get("mappings").get(mapping).get("subject") - if (type(list_subject)== list): #varios sujetos ERROR - raise Exception("ERROR: Only one subject should be in mapping: " + mapping) + if type(subject) != list: + subjects.append(subject) + elif type(subject) == list and only_one: + raise Exception("ERROR: Only one subject can be define using keyword 'subject'") + else: + rml_subjects = subject - else: #un solo sujeto - subject_template = "\trr:subjectMap [\n" + "\t\ta rr:SubjectMap;\n"+ "\t\t" - subject=list_subject - termMap=getTermMap(subject) - if(termMap=='rr:template "'): - subject=subject.replace("$",'') - subject=subject.replace("(","{") - subject=subject.replace(")",'}') - else: - subject=subject.replace("$(",'') - subject=subject.replace(")",'') - if(termMap!='rr:constant '): - subject_template+= termMap + ' "'+ subject + '"'+ "\n\t];\n\n" - else: - subject_template+= termMap + subject + "\n\t];\n\n" - final_list.append(subject_template) + for subject in subjects: + subject_termmap = termmap.generate_rml_termmap(constants.R2RML_SUBJECT_PROPERTY, constants.R2RML_SUBJECT_CLASS, subject, "\t\t") + rml_subjects.append(subject_termmap) + return rml_subjects - else: - raise Exception("ERROR: no subjects in mapping " + mapping) - return final_list -def getTermMap(text): - if("$(" in text and ")" in text): - if text[0]=="$": - return "rml:reference " - else: - return "rr:template " - else: - return "rr:constant " diff --git a/lib/termmap.py b/lib/termmap.py new file mode 100644 index 0000000..666c744 --- /dev/null +++ b/lib/termmap.py @@ -0,0 +1,41 @@ +import constants + + +## return the type of TermMap based on the input text +def get_termmap_type(text): + if "$(" in text and ")" in text: + if text[0] == "$" and text[len(text) - 1] == ")" and text.count("$(") == 1: + return constants.RML_REFERENCE + else: + return constants.R2RML_TEMPLATE + else: + return constants.R2RML_CONSTANT + + +## Generates a TermMap (subject, predicate, object) based on the property, class and the text +def generate_rml_termmap(rml_property, rml_class, text, identation): + template = identation[0:-1] + rml_property + " [\n"+identation+"a " + rml_class + ";\n" + identation + term_map = get_termmap_type(text) + if term_map == constants.R2RML_TEMPLATE: + text = text.replace("$(", "{") + text = text.replace(")", "}") + elif term_map == constants.RML_REFERENCE: + text = text.replace("$(", "") + text = text.replace(")", "") + elif term_map == constants.R2RML_CONSTANT and text == "a": + text = constants.RDF_TYPE + if term_map != "rr:constant": + template += term_map + " \"" + text + "\";\n"+identation[0:-1]+"];\n" + else: + template += term_map + " " + text + ";\n"+identation[0:-1]+"];\n" + + return template + + +def check_type(om): + if "~lang" in om: + return "language" + elif ":" in om: + return "datatype" + else: + return "error" \ No newline at end of file diff --git a/requirements.txt b/requirements.txt new file mode 100644 index 0000000..af95972 --- /dev/null +++ b/requirements.txt @@ -0,0 +1,2 @@ +pyyaml == 5.4.1 +rdflib == 5.0.0 \ No newline at end of file	Does not support json as a source Does not support JSON format as a source ```yaml prefixes: ex: "http://example.com/" mappings: person: sources: - ['data.json~jsonpath', '$.persons[*]'] s: http://example.com/$(firstname) po: - [a, foaf:Person] - [ex:name, $(firstname)] ``` Output: ```yaml @prefix ex: <http://example.com/>. @prefix rr: <http://www.w3.org/ns/r2rml#>. @prefix rml: <http://semweb.mmlab.be/ns/rml#>. @prefix rdf: <http://www.w3.org/1999/02/22-rdf-syntax-ns#>. ```		2021-07-12T07:00:00	[]	[]
CouncilDataProject/cdp-backend	CouncilDataProject__cdp-backend-236	0982830358df2c62ab00a05bf8f862c4f3ee168c	diff --git a/cdp_backend/utils/file_utils.py b/cdp_backend/utils/file_utils.py index c8c27073..7ea300e4 100644 --- a/cdp_backend/utils/file_utils.py +++ b/cdp_backend/utils/file_utils.py @@ -6,6 +6,7 @@ import math import random import re +import shutil from hashlib import sha256 from pathlib import Path from uuid import uuid4 @@ -225,21 +226,27 @@ def resource_copy( # noqa: C901 ) return str(dst) - # Set custom timeout for http resources + # Common case: http(s) URI if uri.startswith("http"): # The verify=False is passed to any http URIs # It was added because it's very common for SSL certs to be bad # See: https://github.com/CouncilDataProject/cdp-scrapers/pull/85 # And: https://github.com/CouncilDataProject/seattle/runs/5957646032 - with open(dst, "wb") as open_dst: - open_dst.write( - requests.get( - uri, - verify=False, - timeout=1800, - ).content - ) + # Use stream=True to avoid downloading the entire file into memory + # See: https://github.com/CouncilDataProject/cdp-backend/issues/235 + try: + # This response must be closed after the copy is done. But using + # `with requests.get() as response` fails mypy type checking. + # See: https://requests.readthedocs.io/en/latest/user/advanced/#body-content-workflow + response = requests.get(uri, stream=True, verify=False, timeout=1800) + response.raise_for_status() + with open(dst, "wb") as open_dst: + shutil.copyfileobj( + response.raw, open_dst, length=64 * 1024 * 1024 # 64MB chunks + ) + finally: + response.close() else: # TODO: Add explicit use of GCS credentials until public read is fixed diff --git a/pyproject.toml b/pyproject.toml index bf6585a3..0e2bcd29 100644 --- a/pyproject.toml +++ b/pyproject.toml @@ -103,6 +103,7 @@ test = [ # Extras "networkx>=2.5", "pydot>=1.4", + "requests-mock>=1.10.0" ] docs = [ # Sphinx + Doc Gen + Styling	Inefficient usage of requests.get for very large videos causes event gather to fail ### Describe the Bug Usage of [requests](https://github.com/CouncilDataProject/cdp-backend/blob/0982830358df2c62ab00a05bf8f862c4f3ee168c/cdp_backend/utils/file_utils.py#L237) to write the content to the open destination file as a BufferedWriter causes unconstrained memory usage based on the size of the video file being copied from the remote origin. For example, consider the mp4 at https://sg001-media.sliq.net/00309/Media2/2023/04/27/House%20Floor%20Session-House%20Chamber-080505_46177_1113.mp4... ```sh curl -Ss -I https://sg001-media.sliq.net/00309/Media2/2023/04/27/House%20Floor%20Session-House%20Chamber-080505_46177_1113.mp4 \| grep ^content-length content-length: 8068641879 ``` The video is ~8GB in size and uses too much memory for n1-standard-4 machine type we use by default in Google Cloud especially when whisper model is already loaded. ### Expected Behavior Downloading video files from remote origin should use a pattern that constrains memory usage in the python process. ### Reproduction ```python import requests with open("movie.mp4", "wb") as f: f.write( requests.get("https://sg001-media.sliq.net/00309/Media2/2023/04/27/House%20Floor%20Session-House%20Chamber-080505_46177_1113.mp4").content ) ``` Running this will download the 8G file and copy to output file. The memory usage will be unconstrained since requests to will load the entire body into memory before copying the file to disk. ### Environment - OS Version: N/A - cdp-backend Version: latest		2023-05-13T18:50:38	[]	[]
CouncilDataProject/cdp-backend	CouncilDataProject__cdp-backend-214	ab3a65dee89f1c4a0c9ad9ad3abe0881f7c84d99	diff --git a/cdp_backend/pipeline/event_gather_pipeline.py b/cdp_backend/pipeline/event_gather_pipeline.py index 12dca809..e8cdc49f 100644 --- a/cdp_backend/pipeline/event_gather_pipeline.py +++ b/cdp_backend/pipeline/event_gather_pipeline.py @@ -160,7 +160,11 @@ def create_event_gather_flow( if session.caption_uri is not None: # If the caption doesn't exist, remove the property # This will result in Speech-to-Text being used instead - if not resource_exists(session.caption_uri): + if not resource_exists( + session.caption_uri + ) or not file_utils.caption_is_valid( + tmp_video_filepath, session.caption_uri + ): log.warning( f"File not found using provided caption URI: " f"'{session.caption_uri}'. " diff --git a/cdp_backend/utils/file_utils.py b/cdp_backend/utils/file_utils.py index a23bd446..e9fa0ae5 100644 --- a/cdp_backend/utils/file_utils.py +++ b/cdp_backend/utils/file_utils.py @@ -7,13 +7,21 @@ import re from hashlib import sha256 from pathlib import Path +from tempfile import TemporaryDirectory from typing import Optional, Tuple, Union from uuid import uuid4 +import ffmpeg import fireo import fsspec import requests +import webvtt from fsspec.core import url_to_fs +from webvtt.exceptions import ( + InvalidCaptionsError, + MalformedCaptionError, + MalformedFileError, +) from ..database import models as db_models @@ -626,3 +634,50 @@ def clip_and_reformat_video( log.error(ffmpeg_stderr) return output_path + + +def caption_is_valid(video_uri: str, caption_uri: str) -> bool: + """ + Validate the caption file at the URI provided. + + Parameters + ---------- + video_uri: str + The URI for the the target video. + uri: str + The URI to validate caption file for. + + Returns + ------- + status: bool + The validation status. + + Notes + ----- + Duration of the video at video_uri + and the duration of the caption file are compared. + The caption file is accepted if the durations differ by no more than 20%. + """ + try: + ffprobe = ffmpeg.probe(video_uri) + except ffmpeg.Error as e: + log.warning(f"ffprobe({video_uri}): {e.stderr}") + return False + + # Making sure temp copy of the caption file is deleted when finished + with TemporaryDirectory() as dir_path: + local_caption_path = resource_copy(caption_uri, dst=dir_path) + try: + caption_length = webvtt.read(local_caption_path).total_length + except (InvalidCaptionsError, MalformedCaptionError, MalformedFileError) as e: + log.warning(f"webvtt.read({caption_uri}): {str(e)}") + return False + + similar_audio_streams = filter( + lambda s: s.get("codec_type", "") == "audio" + and math.isclose( + float(s.get("duration", "0.0")), caption_length, rel_tol=0.2 + ), + ffprobe.get("streams", []), + ) + return any(similar_audio_streams)	Filter out bad caption files ### Feature Description ```Json {"generator": "CDP WebVTT Conversion -- CDP v3.2.3", "confidence": 0.97, "session_datetime": "2022-09-14T10:00:00-04:00", "created_datetime": "2022-09-17T15:51:32.537666", "sentences": [{"index": 0, "confidence": 0.97, "start_time": 0.0, "end_time": 0.0, "words": [], "text": "", "speaker_index": 0, "speaker_name": null, "annotations": null}], "annotations": null} ``` Note `end_time` in the above output from a PIttsburgh, PA captions processing. Turns out that event does have a caption file in the Legistar data structure for the event _but_ it is in fact empty. We want to filter these out just attempt speech-to-text. ### Use Case Allow for proper transcript generation in `generate_transcript()` by appropriately filtering out empty caption files. ### Solution Compare the lengths of the video and the caption file. If they differ by more than some threshold, e.g. 20%, throw away the caption file. This means the scraper can no longer just hand off the caption file URL as-is. ### Alternatives - Throw away caption files with file size less than some threshold, e.g. 100 bytes. - Throw away caption files less than ~1 minute.		2022-09-23T06:14:34	[]	[]
CouncilDataProject/cdp-backend	CouncilDataProject__cdp-backend-193	00f98763aab4c9a7e69e69c06015d9ed17d64c78	diff --git a/cdp_backend/sr_models/webvtt_sr_model.py b/cdp_backend/sr_models/webvtt_sr_model.py index e0abb019..5d986a5a 100644 --- a/cdp_backend/sr_models/webvtt_sr_model.py +++ b/cdp_backend/sr_models/webvtt_sr_model.py @@ -164,7 +164,15 @@ def _get_sentences( # List of text, representing a sentence lines: List[str] = [] start_time: Optional[float] = None + for caption in speaker_turn_captions: + + # Clean text of line breaks + caption.text = caption.text.replace("\n", " ") + + # Remove any double spaces as result of line break removal + caption.text = caption.text.replace(" ", " ") + if start_time is None: start_time = caption.start_in_seconds lines.append(caption.text)	WebVTT parsing fails for Boston <!-- ⚠️⚠️ Please do the following before submitting: ⚠️⚠️ 📖 Please read our Code of Conduct. 🔎 Please search existing issues to avoid creating duplicates. --> ### Describe the Bug _A clear and concise description of the bug._ WebVTT parsing isn't the greatest for Boston: https://councildataproject.org/boston/#/events/3c7f0df0ecd4 ### Expected Behavior _What did you expect to happen instead?_ WebVTT parsing should be the same level or better than Seattle	Boston event gather log for where closed caption is pulled from: https://github.com/CouncilDataProject/boston/runs/6759035563?check_suite_focus=true#step:9:7200 Seems like you should be able to pull that same URL: http://boston.granicus.com//videos/87/captions.vtt and test out changes locally @isaacna or @Shak2000 do either of you have the time or interest to take this one? > @isaacna or @Shak2000 do either of you have the time or interest to take this one? I should have some time later this week or next to take a look	2022-06-13T23:06:45	[]	[]
CouncilDataProject/cdp-backend	CouncilDataProject__cdp-backend-185	f77ca6d02099dcea71da8378ddf742951a4665a7	diff --git a/.github/workflows/build-docs.yml b/.github/workflows/build-docs.yml index 7ff4d233..bdbfda29 100644 --- a/.github/workflows/build-docs.yml +++ b/.github/workflows/build-docs.yml @@ -15,7 +15,7 @@ jobs: - name: Set up Python uses: actions/setup-python@v2 with: - python-version: 3.9 + python-version: '3.10' cache: 'pip' - name: Install Packages run: \| diff --git a/.github/workflows/build-main.yml b/.github/workflows/build-main.yml index 0e922800..f469d390 100644 --- a/.github/workflows/build-main.yml +++ b/.github/workflows/build-main.yml @@ -19,7 +19,7 @@ jobs: strategy: fail-fast: false matrix: - python-version: [3.7, 3.8, 3.9] + python-version: [3.8, 3.9, '3.10'] os: [ubuntu-latest, macOS-latest, windows-latest] steps: @@ -62,7 +62,7 @@ jobs: - name: Set up Python uses: actions/setup-python@v2 with: - python-version: 3.9 + python-version: '3.10' cache: 'pip' - name: Install graphviz run: \| @@ -85,7 +85,7 @@ jobs: - name: Set up Python uses: actions/setup-python@v2 with: - python-version: 3.9 + python-version: '3.10' cache: 'pip' - name: Install Dependencies run: \| diff --git a/.github/workflows/check-pr.yml b/.github/workflows/check-pr.yml index 8221da74..97842300 100644 --- a/.github/workflows/check-pr.yml +++ b/.github/workflows/check-pr.yml @@ -8,7 +8,7 @@ jobs: strategy: fail-fast: false matrix: - python-version: [3.7, 3.8, 3.9] + python-version: [3.8, 3.9, '3.10'] os: [ubuntu-latest, macOS-latest, windows-latest] steps: @@ -51,7 +51,7 @@ jobs: - name: Set up Python uses: actions/setup-python@v2 with: - python-version: 3.9 + python-version: '3.10' cache: 'pip' - name: Install graphviz run: \| @@ -71,7 +71,7 @@ jobs: - name: Set up Python uses: actions/setup-python@v2 with: - python-version: 3.9 + python-version: '3.10' cache: 'pip' - name: Install graphviz run: \| diff --git a/.pre-commit-config.yaml b/.pre-commit-config.yaml index bf891b19..c189180f 100644 --- a/.pre-commit-config.yaml +++ b/.pre-commit-config.yaml @@ -5,7 +5,7 @@ repos: hooks: - id: flake8 additional_dependencies: [flake8-typing-imports, flake8-debugger] - args: [--count, --show-source, --statistics, --min-python-version=3.7.0] + args: [--count, --show-source, --statistics, --min-python-version=3.8.0] - repo: https://github.com/myint/autoflake rev: v1.4 hooks: diff --git a/LICENSE b/LICENSE index c65d50b0..c9f5e4f2 100644 --- a/LICENSE +++ b/LICENSE @@ -1,6 +1,6 @@ MIT License -Copyright (c) 2020, Council Data Project Contributors +Copyright (c) 2022, Council Data Project Contributors Permission is hereby granted, free of charge, to any person obtaining a copy of this software and associated documentation files (the "Software"), to deal diff --git a/Makefile b/Makefile index 24b77a20..08a0550e 100644 --- a/Makefile +++ b/Makefile @@ -56,8 +56,8 @@ gen-docs: ## Generate Sphinx HTML documentation, including API docs create_cdp_database_uml \ -o docs/_static/cdp_database_diagram.dot dot \ - -T png \ - -o docs/_static/cdp_database_diagram.png docs/_static/cdp_database_diagram.dot + -T jpg \ + -o docs/_static/cdp_database_diagram.jpg docs/_static/cdp_database_diagram.dot create_cdp_ingestion_models_doc \ -t docs/ingestion_models.template \ -o docs/ingestion_models.md diff --git a/cdp_backend/__init__.py b/cdp_backend/__init__.py index 097173c9..f392cd0d 100644 --- a/cdp_backend/__init__.py +++ b/cdp_backend/__init__.py @@ -2,7 +2,7 @@ """Top-level package for cdp-backend.""" -__author__ = "Jackson Maxfield Brown" +__author__ = "Council Data Project Contributors" __email__ = "[email protected]" from .version import __version__ # noqa: F401 diff --git a/dev-infrastructure/README.md b/dev-infrastructure/README.md index cef61168..0a1de564 100644 --- a/dev-infrastructure/README.md +++ b/dev-infrastructure/README.md @@ -57,9 +57,6 @@ restart your terminal after installation._ After `pulumi` and `gcloud` have both been installed and terminal restarted, run the following commands to setup your local machine with credentials to both services. -Note: Pulumi only supports Python 3.7, when creating dev infrastructures you -need to run these scripts in a py37 environment. - ```bash cd cdp-backend/dev-infrastructure make login diff --git a/docs/conf.py b/docs/conf.py index 283f5135..477b6947 100755 --- a/docs/conf.py +++ b/docs/conf.py @@ -21,8 +21,6 @@ import os import sys -import sphinx_rtd_theme # noqa: F401 - import cdp_backend sys.path.insert(0, os.path.abspath("..")) @@ -70,8 +68,8 @@ # General information about the project. project = u"cdp-backend" -copyright = u"2020, Jackson Maxfield Brown" -author = u"Jackson Maxfield Brown" +copyright = u"2022, Council Data Project Contributors" +author = u"Council Data Project Contributors" # The version info for the project you"re documenting, acts as replacement # for \|version\| and \|release\|, also used in various other places throughout @@ -106,16 +104,13 @@ # The theme to use for HTML and HTML Help pages. See the documentation for # a list of builtin themes. # -html_theme = "sphinx_rtd_theme" +html_theme = "furo" # Theme options are theme-specific and customize the look and feel of a # theme further. For a list of options available for each theme, see the # documentation. # -html_theme_options = { - "collapse_navigation": False, - "prev_next_buttons_location": "top", -} +html_theme_options = {} # Add any paths that contain custom static files (such as style sheets) here, # relative to this directory. They are copied after the builtin static files, @@ -154,7 +149,7 @@ master_doc, "cdp_backend.tex", u"cdp-backend Documentation", - u"Jackson Maxfield Brown", + u"Council Data Project Contributors", "manual", ), ] diff --git a/docs/database_schema.md b/docs/database_schema.md index 50227c6e..df5b1130 100644 --- a/docs/database_schema.md +++ b/docs/database_schema.md @@ -5,7 +5,7 @@ While CDP utilizes a called [Cloud Firestore](https://firebase.google.com/docs/firestore), it is still useful to view the database schema as a UML diagram. -![cdp-database-schema](./_static/cdp_database_diagram.png) +![cdp-database-schema](./_static/cdp_database_diagram.jpg) ### Notes * `` denotes that the field is required diff --git a/setup.py b/setup.py index ec288ea4..d291b5bb 100644 --- a/setup.py +++ b/setup.py @@ -13,47 +13,46 @@ ] infrastructure_requirements = [ - "pulumi~=3.3", - "pulumi-google-native~=0.7.0", - "pulumi-gcp~=5.7", + "pulumi~=3.31", + "pulumi-google-native~=0.18", + "pulumi-gcp~=6.0", ] pipeline_requirements = [ "dask[distributed]>=2021.7.0", - "ffmpeg-python~=0.2.0", - "google-cloud-speech~=2.11.0", + "ffmpeg-python==0.2.0", + "google-cloud-speech~=2.13", "graphviz~=0.16", - "imageio~=2.9", - "imageio-ffmpeg~=0.4.4", + "imageio~=2.18", + "imageio-ffmpeg~=0.4", "nltk~=3.6", - "pandas~=1.2", - "prefect~=0.14.0", - "rapidfuzz~=1.4", + "pandas~=1.0", + "prefect~=1.2", + "rapidfuzz~=2.0", "spacy~=3.0", - "truecase~=0.0.12", - "webvtt-py~=0.4.6", - "yt-dlp~=2022.2.4", + "truecase==0.0.14", + "webvtt-py==0.4.6", + "yt-dlp>=2022.2.4", ] test_requirements = [ infrastructure_requirements, pipeline_requirements, "black>=22.3.0", - "codecov==2.1.12", - "fastparquet~=0.7", + "codecov>=2.1.12", "flake8>=3.8.3", "flake8-debugger>=3.2.1", "isort>=5.7.0", "mypy>=0.790", "networkx>=2.5", - "pyarrow~=5.0", + "pyarrow>=5.0", "pydot>=1.4", "pytest>=5.4.3", "pytest-cov>=2.9.0", "pytest-raises>=0.11", "tox>=3.15.2", "types-pytz>=2021.1.2", - "types-requests~=0.1.11", + "types-requests", ] dev_requirements = [ @@ -61,25 +60,24 @@ test_requirements, "bokeh>=2.3.2", "bump2version>=1.0.1", - "coverage==5.4", + "coverage>=5.4", "ipython>=7.15.0", "jinja2>=2.11.2", "m2r2>=0.2.7", - "prefect[viz]~=0.14.0", - "pytest-runner>=5.2", + "prefect[viz]", "Sphinx>=3.4.3", - "sphinx_rtd_theme>=0.5.1", + "furo>=2022.4.7", "twine>=3.1.1", "wheel>=0.34.2", ] requirements = [ - "aiohttp~=3.7.4.post0", - "dataclasses-json~=0.5", - "fireo~=1.4", + "aiohttp>=3.7.4.post0", + "dataclasses-json>=0.5", + "fireo>=1.4", "fsspec", # Version pin set by gcsfs - "gcsfs~=2021.7.0", - "requests~=2.26.0", + "gcsfs>=2021.7.0", + "requests>=2.26.0", ] extra_requirements = { @@ -95,16 +93,18 @@ } setup( - author="Jackson Maxfield Brown, To Huynh, Isaac Na", + author=( + "Jackson Maxfield Brown, To Huynh, Isaac Na, Council Data Project Contributors" + ), author_email="[email protected]", classifiers=[ - "Development Status :: 2 - Pre-Alpha", + "Development Status :: 5 - Production/Stable", "Intended Audience :: Developers", "License :: OSI Approved :: MIT License", "Natural Language :: English", - "Programming Language :: Python :: 3.7", "Programming Language :: Python :: 3.8", "Programming Language :: Python :: 3.9", + "Programming Language :: Python :: 3.10", ], description=( "Data storage utilities and processing pipelines to run on CDP server " @@ -145,7 +145,7 @@ keywords="civic technology, open government", name="cdp-backend", packages=find_packages(exclude=["tests", ".tests", ".tests.*"]), - python_requires=">=3.7", + python_requires=">=3.8", setup_requires=setup_requirements, test_suite="cdp_backend/tests", tests_require=test_requirements, diff --git a/tox.ini b/tox.ini index b3c54930..68f11aaa 100644 --- a/tox.ini +++ b/tox.ini @@ -1,5 +1,5 @@ [tox] -envlist = py37, py38, py39, lint +envlist = py38, py39, py310, lint skip_missing_interpreters = true toxworkdir={env:TOX_WORK_DIR:.tox}	Switch docs theme over to Furo Copy docs theme from cdp-data: http://councildataproject.org/cdp-data/index.html		2022-05-01T16:29:25	[]	[]
CouncilDataProject/cdp-backend	CouncilDataProject__cdp-backend-165	ede007f7be5c42cc7913f1872024dceadcf96f3f	diff --git a/cdp_backend/pipeline/event_index_pipeline.py b/cdp_backend/pipeline/event_index_pipeline.py index c3818d79..831a893b 100644 --- a/cdp_backend/pipeline/event_index_pipeline.py +++ b/cdp_backend/pipeline/event_index_pipeline.py @@ -207,6 +207,7 @@ def read_transcripts_and_generate_grams( cleaned_text=string_utils.clean_text( sentence.text, clean_stop_words=True, + clean_emojis=True, ), n_grams=[], ) diff --git a/cdp_backend/utils/string_utils.py b/cdp_backend/utils/string_utils.py index 457b0d7a..58f81313 100644 --- a/cdp_backend/utils/string_utils.py +++ b/cdp_backend/utils/string_utils.py @@ -12,7 +12,38 @@ ############################################################################### -def clean_text(text: str, clean_stop_words: bool = False) -> str: +def remove_emojis(text: str) -> str: + """ + Minor changes made from this answer on stackoverflow: + https://stackoverflow.com/a/58356570 + """ + emoj_patterns = re.compile( + "[" + "\U0001F600-\U0001F64F" # emoticons + "\U0001F300-\U0001F5FF" # symbols & pictographs + "\U0001F680-\U0001F6FF" # transport & map symbols + "\U0001F1E0-\U0001F1FF" # flags (iOS) + "\U00002702-\U000027B0" + "\U000024C2-\U0001F251" + "\U0001f926-\U0001f937" + "\u2600-\u2B55" + "\u200d" + "\u23cf" + "\u23e9" + "\u231a" + "\ufe0f" # dingbats + "\u3030" + "]+", + re.UNICODE, + ) + return re.sub(emoj_patterns, "", text) + + +def clean_text( + text: str, + clean_stop_words: bool = False, + clean_emojis: bool = False, +) -> str: """ Clean text of common characters and extra formatting. @@ -23,6 +54,9 @@ def clean_text(text: str, clean_stop_words: bool = False) -> str: clean_stop_words: bool Should English stop words be removed from the raw text or not. Default: False (do not remove stop words) + clean_emojis: bool + Should emojis, emoticons, pictograms, and other characters be removed. + Default: False (do not remove pictograms) Returns ------- @@ -57,22 +91,23 @@ def clean_text(text: str, clean_stop_words: bool = False) -> str: STOPWORDS = stopwords.words("english") joined_stopwords = "\|".join(STOPWORDS) - cleaned_stopwords = re.sub( + cleaned_text = re.sub( r"\b(" + joined_stopwords + r")\b", "", cleaned_punctuation, ) else: # Update for mypy typing - cleaned_stopwords = cleaned_punctuation + cleaned_text = cleaned_punctuation + + # Remove pictograms + if clean_emojis: + cleaned_text = remove_emojis(cleaned_text) # Remove gaps in string try: - cleaned_doc = re.sub(r" {2,}", " ", cleaned_stopwords) - if cleaned_doc[0] == " ": - cleaned_doc = cleaned_doc[1:] - if cleaned_doc[-1] == " ": - cleaned_doc = cleaned_doc[:-1] + cleaned_doc = re.sub(r" {2,}", " ", cleaned_text) + cleaned_doc = cleaned_doc.strip() # IndexError occurs when the string was cleaned and it contained entirely stop # words or punctuation for some reason	Context span selection fails during index creation <!-- ⚠️⚠️ Please do the following before submitting: ⚠️⚠️ 📖 Please read our Code of Conduct. 🔎 Please search existing issues to avoid creating duplicates. --> ### Describe the Bug _A clear and concise description of the bug._ https://github.com/CouncilDataProject/seattle-staging/runs/4862736150?check_suite_focus=true Looks like the `closest_term` was `""` and an empty string wasn't found in the list. Regardless of how valuable that is. This was from seattle-staging and we can run and store the index locally so hopefully this will allow us to check out both what term is being processed and what is going on in general.	https://github.com/CouncilDataProject/seattle-staging/runs/4916834417?check_suite_focus=true Another pipeline fail	2022-02-17T07:03:04	[]	[]
CouncilDataProject/cdp-backend	CouncilDataProject__cdp-backend-132	706b5fedff2ddf9c80dcf8f90a6e33f59beb4ff0	diff --git a/cdp_backend/pipeline/event_gather_pipeline.py b/cdp_backend/pipeline/event_gather_pipeline.py index 7389317f..4359f3cf 100644 --- a/cdp_backend/pipeline/event_gather_pipeline.py +++ b/cdp_backend/pipeline/event_gather_pipeline.py @@ -126,7 +126,15 @@ def create_event_gather_flow( session_processing_results: List[SessionProcessingResult] = [] for session in event.sessions: # Download video to local copy - tmp_video_filepath = resource_copy_task(uri=session.video_uri) + resource_copy_filepath = resource_copy_task(uri=session.video_uri) + + # Convert to mp4 if necessary + tmp_video_filepath = convert_video_to_mp4_and_upload( + video_filepath=resource_copy_filepath, + session=session, + credentials_file=config.google_credentials_file, + bucket=config.validated_gcs_bucket_name, + ) # Get unique session identifier session_content_hash = get_session_content_hash( @@ -274,6 +282,61 @@ def get_session_content_hash( return file_utils.hash_file_contents(uri=tmp_video_filepath) +@task +def convert_video_to_mp4_and_upload( + video_filepath: str, + session: Session, + credentials_file: str, + bucket: str, +) -> str: + """ + Convert a video to MP4 (if necessary), upload it to the file store, and remove + the original non-MP4 file that was resource copied. + + Parameters + ---------- + video_filepath: Union[str, Path] + The local path for video file to convert. + session: Session + The session to append the new MP4 video uri to. + credentials_file: str + Path to Google Service Account Credentials JSON file. + bucket: str + The GCS bucket to store the MP4 file to. + + Returns + ------- + mp4_filepath: str + The local filepath of the converted MP4 file. + """ + + # Get file extension + ext = Path(video_filepath).suffix + + # Convert to mp4 if necessary + if ext != ".mp4": + # Convert video to mp4 + mp4_filepath = file_utils.convert_video_to_mp4(video_filepath) + + # Remove old mkv file + fs_functions.remove_local_file(video_filepath) + + # Upload to gcsfs + mp4_uri = fs_functions.upload_file( + credentials_file=credentials_file, + bucket=bucket, + filepath=mp4_filepath, + ) + + # Set session video_uri to uri in remote file store + session.video_uri = mp4_uri + + # Return converted local mp4 filepath + return mp4_filepath + + return video_filepath + + @task def split_audio( session_content_hash: str, diff --git a/cdp_backend/utils/file_utils.py b/cdp_backend/utils/file_utils.py index a104228a..791a6285 100644 --- a/cdp_backend/utils/file_utils.py +++ b/cdp_backend/utils/file_utils.py @@ -8,6 +8,7 @@ from typing import Optional, Tuple, Union import aiohttp +import ffmpeg import fsspec from fsspec.core import url_to_fs @@ -363,6 +364,27 @@ def hash_file_contents(uri: str, buffer_size: int = 2 ** 16) -> str: return hasher.hexdigest() +def convert_video_to_mp4(video_filepath: str) -> str: + """ + Converts a video to an equivalent MP4 file. + + Parameters + ---------- + video_filepath: str + The filepath of the video to convert. + + Returns + ------- + mp4_filepath: str + The filepath of the converted MP4 video. + """ + mp4_filepath = str(Path(video_filepath).with_suffix(".mp4")) + ffmpeg.input(video_filepath).output(mp4_filepath).overwrite_output().run() + log.info("Finished converting {} to mp4".format(video_filepath)) + + return mp4_filepath + + def generate_file_storage_name(file_uri: str, suffix: str) -> str: """ Generate a filename using the hash of the file contents and some provided suffix.	Ensure or Convert video to MP4 or WebM before processing <!-- ⚠️⚠️ Please do the following before submitting: ⚠️⚠️ 📖 Please read our Code of Conduct. 🔎 Please search existing issues to avoid creating duplicates. --> ### Feature Description _A clear and concise description of the feature you're requesting._ If the video isn't MP4 / WebM, we won't be able to render the video on the website. Ensure or convert to a "web ready" video format. ### Use Case _Please provide a use case to help us understand your request in context._ Allow more video formats through the pipeline / catch common video recording format errors. (OBS records videos as MKV by default) ### Solution _Please describe your ideal solution._ If the video file isn't MP4 or WEBM, download it, convert, and upload our conversion to file storage. Use our converted video as the basis which would mean hosting it ourselves but :shrug: ### Alternatives _Please describe any alternatives you've considered, even if you've dismissed them._		2021-11-08T05:06:50	[]	[]
CouncilDataProject/cdp-backend	CouncilDataProject__cdp-backend-128	5b9a478316fe390171c09b59c01ee77714f8db64	diff --git a/cdp_backend/sr_models/google_cloud_sr_model.py b/cdp_backend/sr_models/google_cloud_sr_model.py index c573e8d1..c0b3c5ae 100644 --- a/cdp_backend/sr_models/google_cloud_sr_model.py +++ b/cdp_backend/sr_models/google_cloud_sr_model.py @@ -69,22 +69,22 @@ def transcribe( The transcript model for the supplied media file. """ # Create client - client = speech.SpeechClient.from_service_account_json(self.credentials_file) + client = speech.SpeechClient.from_service_account_file( + filename=str(self.credentials_file) + ) # Create basic metadata - metadata = speech.types.RecognitionMetadata() + metadata = speech.RecognitionMetadata() metadata.interaction_type = ( - speech.enums.RecognitionMetadata.InteractionType.DISCUSSION + speech.RecognitionMetadata.InteractionType.DISCUSSION ) # Add phrases - speech_context = speech.types.SpeechContext( - phrases=self._clean_phrases(phrases) - ) + speech_context = speech.SpeechContext(phrases=self._clean_phrases(phrases)) # Prepare for transcription - config = speech.types.RecognitionConfig( - encoding=speech.enums.RecognitionConfig.AudioEncoding.LINEAR16, + config = speech.RecognitionConfig( + encoding=speech.RecognitionConfig.AudioEncoding.LINEAR16, sample_rate_hertz=16000, language_code="en-US", enable_automatic_punctuation=True, @@ -92,11 +92,13 @@ def transcribe( speech_contexts=[speech_context], metadata=metadata, ) - audio = speech.types.RecognitionAudio(uri=file_uri) + audio = speech.RecognitionAudio(uri=file_uri) # Begin transcription log.debug(f"Beginning transcription for: {file_uri}") - operation = client.long_running_recognize(config, audio) + operation = client.long_running_recognize( + request={"config": config, "audio": audio} + ) # Wait for complete response = operation.result(timeout=10800) @@ -145,10 +147,16 @@ def transcribe( # Extract word from response word = result.alternatives[0].words[w_ind] + # Nanos no longer supported, use microseconds instead + # https://github.com/googleapis/python-speech/issues/71 start_time = ( - word.start_time.seconds + word.start_time.nanos * 1e-9 + word.start_time.seconds + + word.start_time.microseconds * 1e-6 + ) + + end_time = ( + word.end_time.seconds + word.end_time.microseconds * 1e-6 ) - end_time = word.end_time.seconds + word.end_time.nanos * 1e-9 # Add start_time to Sentence if first word if w_ind - w_marker == 0: diff --git a/setup.py b/setup.py index bce2d9c2..079d108e 100644 --- a/setup.py +++ b/setup.py @@ -21,7 +21,7 @@ pipeline_requirements = [ "dask[distributed]>=2021.7.0", "ffmpeg-python~=0.2.0", - "google-cloud-speech~=1.3", + "google-cloud-speech~=2.11.0", "graphviz~=0.16", "imageio~=2.9", "imageio-ffmpeg~=0.4.4",	Upgrade to GCP Speech-to-Text v2 To stay up-to-date we should upgrade to v2 when we get a chance, fortunately they released an upgrading guide here: https://github.com/googleapis/python-speech/blob/release-v2.0.0/UPGRADING.md		2021-11-02T00:19:58	[]	[]
CouncilDataProject/cdp-backend	CouncilDataProject__cdp-backend-112	8c8dbf4e4d2fdc7e4278206c75d550f345c9cbe6	diff --git a/cdp_backend/pipeline/event_gather_pipeline.py b/cdp_backend/pipeline/event_gather_pipeline.py index 6c609112..ed2c6fe1 100644 --- a/cdp_backend/pipeline/event_gather_pipeline.py +++ b/cdp_backend/pipeline/event_gather_pipeline.py @@ -126,8 +126,10 @@ def create_event_gather_flow( session_processing_results: List[SessionProcessingResult] = [] for session in event.sessions: # Get or create audio + tmp_video_filepath = resource_copy_task(uri=session.video_uri) + session_content_hash, audio_uri = get_video_and_split_audio( - video_uri=session.video_uri, + tmp_video_filepath=tmp_video_filepath, bucket=config.validated_gcs_bucket_name, credentials_file=config.google_credentials_file, ) @@ -167,12 +169,17 @@ def create_event_gather_flow( hover_thumbnail_uri, ) = get_video_and_generate_thumbnails( session_content_hash=session_content_hash, - video_uri=session.video_uri, + tmp_video_path=tmp_video_filepath, event=event, bucket=config.validated_gcs_bucket_name, credentials_file=config.google_credentials_file, ) + # Add audio uri and static thumbnail uri + resource_delete_task( + tmp_video_filepath, upstream_tasks=[audio_uri, static_thumbnail_uri] + ) + # Store all processed and provided data session_processing_results.append( compile_session_processing_result( # type: ignore @@ -197,9 +204,36 @@ def create_event_gather_flow( return flow +@task +def resource_copy_task(uri: str) -> str: + # Copy video file from the internet and store it locally + dirpath = tempfile.mkdtemp() + + file_utils.resource_copy( + uri=uri, + dst=dirpath, + overwrite=True, + ) + + return str(Path(dirpath) / uri.split("/")[-1]) + + +@task +def resource_delete_task(uri: str) -> None: + """ + Remove local file + + Parameters + ---------- + uri: str + The local video file. + """ + fs_functions.remove_local_file(uri) + + @task(nout=2, max_retries=3, retry_delay=timedelta(seconds=60)) def get_video_and_split_audio( - video_uri: str, bucket: str, credentials_file: str + tmp_video_filepath: str, bucket: str, credentials_file: str ) -> Tuple[str, str]: """ Download (or copy) a video file to temp storage, split's the audio, and uploads @@ -207,7 +241,7 @@ def get_video_and_split_audio( Parameters ---------- - video_uri: str + tmp_video_filepath: str The URI to the video file to split audio from. bucket: str The name of the GCS bucket to upload the produced audio to. @@ -226,16 +260,6 @@ def get_video_and_split_audio( We sometimes get file downloading failures when running in parallel so this has two retries attached to it that will run after a failure on a 3 minute delay. """ - # Get just the video filename from the full uri - # TODO: Handle windows file splitting??? - # Generally this is a URI but we do have a goal of a local file processing too... - video_filename = video_uri.split("/")[-1] - tmp_video_filepath = file_utils.resource_copy( - uri=video_uri, - dst=f"audio-splitting-{video_filename}", - overwrite=True, - ) - # Hash the video contents session_content_hash = file_utils.hash_file_contents(uri=tmp_video_filepath) @@ -285,9 +309,6 @@ def get_video_and_split_audio( ]: fs_functions.remove_local_file(local_path) - # Always remove tmp video file - fs_functions.remove_local_file(tmp_video_filepath) - return session_content_hash, audio_uri @@ -715,7 +736,7 @@ def generate_transcript( @task(nout=2) def get_video_and_generate_thumbnails( session_content_hash: str, - video_uri: str, + tmp_video_path: str, event: EventIngestionModel, bucket: str, credentials_file: str, @@ -727,7 +748,7 @@ def get_video_and_generate_thumbnails( ---------- session_content_hash: str The unique key (SHA256 hash of video content) for this session processing. - video_uri: str + tmp_video_path: str The URI to the video file to generate thumbnails from. event: EventIngestionModel The parent event of the session. If no captions are available, @@ -744,12 +765,6 @@ def get_video_and_generate_thumbnails( hover_thumbnail_url: str The URL of the hover thumbnail, stored on GCS. """ - # Create tmp directory to save file in - dirpath = tempfile.mkdtemp() - dst = Path(dirpath) - - tmp_video_path = file_utils.resource_copy(uri=video_uri, dst=dst) - if event.static_thumbnail_uri is None: # Generate new static_thumbnail_file = file_utils.get_static_thumbnail( @@ -784,8 +799,6 @@ def get_video_and_generate_thumbnails( ) fs_functions.remove_local_file(hover_thumbnail_file) - fs_functions.remove_local_file(tmp_video_path) - return ( static_thumbnail_url, hover_thumbnail_url,	Combine all the video processing tasks into a single task <!-- ⚠️⚠️ Please do the following before submitting: ⚠️⚠️ 📖 Please read our Code of Conduct. 🔎 Please search existing issues to avoid creating duplicates. --> ### Feature Description _A clear and concise description of the feature you're requesting._ Reduce the pipeline computation time by combining all the tasks that require the video on the worker together. ### Use Case _Please provide a use case to help us understand your request in context._ Drastically reducing the pipeline computation time. ### Solution _Please describe your ideal solution._ Move the function calls to generate thumbnails to the start of the pipeline with the strip audio and generate hash. ### Alternatives _Please describe any alternatives you've considered, even if you've dismissed them._		2021-09-27T04:21:23	[]	[]
CouncilDataProject/cdp-backend	CouncilDataProject__cdp-backend-59	9d0332cdf85747faa2c3c04b8ae991e43d7df4c4	diff --git a/cdp_backend/sr_models/google_cloud_sr_model.py b/cdp_backend/sr_models/google_cloud_sr_model.py index 4a213aeb..c6a1dc0e 100644 --- a/cdp_backend/sr_models/google_cloud_sr_model.py +++ b/cdp_backend/sr_models/google_cloud_sr_model.py @@ -8,6 +8,7 @@ from typing import Any, List, Optional, Union from google.cloud import speech_v1p1beta1 as speech +from spacy.lang.en import English from ..pipeline.transcript_model import Sentence, Transcript, Word from .sr_model import SRModel @@ -107,28 +108,65 @@ def transcribe( # Create timestamped sentences timestamped_sentences: List[Sentence] = [] - current_sentence = None - sentence_index = 0 - word_index = 0 + transcript_sentence_index = 0 + + # Create sentence boundary pipeline + nlp = English() + nlp.add_pipe("sentencizer") for result in response.results: # Some portions of audio may not have text if len(result.alternatives) > 0: - # Check length of transcript result - word_list = result.alternatives[0].words - if len(word_list) > 0: - for word in word_list: - # create Word + # Split transcript into sentences + doc = nlp(result.alternatives[0].transcript) + + # Convert generator to list + sentences = [str(sent) for sent in doc.sents] + + # Index holder for word results of response + w_marker = 0 + + for s_ind in range(0, len(sentences)): + # Sentence text + s_text = sentences[s_ind] + + num_words = len(s_text.split()) + + # Initialize sentence model + timestamped_sentence = Sentence( + index=transcript_sentence_index, + confidence=result.alternatives[0].confidence, + # Start and end time are placeholder values + start_time=0.0, + end_time=0.0, + words=[], + text=s_text, + ) + + for w_ind in range(w_marker, w_marker + num_words): + # Extract word from response + word = result.alternatives[0].words[w_ind] + start_time = ( word.start_time.seconds + word.start_time.nanos * 1e-9 ) end_time = word.end_time.seconds + word.end_time.nanos * 1e-9 + # Add start_time to Sentence if first word + if w_ind - w_marker == 0: + timestamped_sentence.start_time = start_time + + # Add end_time to Sentence if last word + if (w_ind - w_marker) == (num_words - 1): + timestamped_sentence.end_time = end_time + # Clean everything but non-delimiting characters make lowercase regex = re.compile(r"[^a-zA-Z0-9'\-]") cleaned_word = regex.sub("", word.word).lower() + + # Create Word model timestamped_word = Word( - index=word_index, + index=w_ind - w_marker, start_time=start_time, end_time=end_time, text=cleaned_word, @@ -136,43 +174,20 @@ def transcribe( annotations=None, ) - if current_sentence is None: - current_sentence = Sentence( - index=sentence_index, - confidence=result.alternatives[0].confidence, - start_time=word.start_time, - end_time=word.end_time, - # TODO: Add speaker and annotations? - words=[timestamped_word], - text=word.word, - ) - word_index += 1 - - # End current sentence and reset - # TODO: Account for non-sentence ending periods, such as - # prefixes like "Mr." or "Dr." - elif bool(re.match(r"\.\|\?", word.word[-1])): - # Finish sentence and append - current_sentence.end_time = word.end_time - current_sentence.text += " {}".format(word.word) - current_sentence.words.append(timestamped_word) - - timestamped_sentences.append(current_sentence) - - # Adjust indices - current_sentence = None - sentence_index += 1 - word_index = 0 - - # Update current sentence - else: - current_sentence.text += " {}".format(word.word) - current_sentence.words.append(timestamped_word) - word_index += 1 - - # Update confidence stats - confidence_sum += result.alternatives[0].confidence - segments += 1 + timestamped_sentence.words.append(timestamped_word) + + # Increment word marker + w_marker += num_words + + # Add Sentence to sentence list + timestamped_sentences.append(timestamped_sentence) + + # Increment transcript sentence index + transcript_sentence_index += 1 + + # Update confidence stats + confidence_sum += result.alternatives[0].confidence + segments += 1 # Compute mean confidence if segments > 0: diff --git a/setup.py b/setup.py index 12666cd3..bbcaed92 100644 --- a/setup.py +++ b/setup.py @@ -58,6 +58,7 @@ "pulumi~=3.0.0", "pulumi-google-native~=0.1.0", "pulumi-gcp~=5.0.0", + "spacy~=3.0.6", "truecase>=0.0.9", "webvtt-py>=0.4.5", ]	Use spacy to separate sentences in SR model transcribe function ## Use Case To account for edge cases in sentence detection (Prefixes like Mr.), we can use the [spacy](https://stackoverflow.com/questions/46290313/how-to-break-up-document-by-sentences-with-with-spacy) library. ## Solution Use the spacy library to separate sentences. Haven't looked into the library too deeply, but would probably require us to change the iteration logic since we create `Word` (and increment index) and potentially end the `Sentence` in the same iteration.		2021-05-31T07:13:54	[]	[]
CouncilDataProject/cdp-backend	CouncilDataProject__cdp-backend-38	86b9acc5019dd68989e6934805dc7592999b5a0f	diff --git a/.gitignore b/.gitignore index b81f174a..0bb34bc8 100644 --- a/.gitignore +++ b/.gitignore @@ -114,4 +114,5 @@ ENV/ # Project specific cdp_database_diagram.* +cdp_event_gather_flow_* event_data_ingestion_model.md diff --git a/Makefile b/Makefile index 2bf92b8f..5a605fcd 100644 --- a/Makefile +++ b/Makefile @@ -56,6 +56,7 @@ gen-docs: ## generate Sphinx HTML documentation, including API docs create_cdp_database_uml -o docs/_static/cdp_database_diagram.dot dot -T png -o docs/_static/cdp_database_diagram.png docs/_static/cdp_database_diagram.dot create_cdp_ingestion_model_doc -t docs/event_data_ingestion_model.template -o docs/event_data_ingestion_model.md + create_cdp_event_gather_flow_viz -o docs/_static/cdp_event_gather_flow_{ftype}.png sphinx-apidoc -o docs/ cdp_backend */tests/ $(MAKE) -C docs html diff --git a/cdp_backend/bin/cdp_event_gather.py b/cdp_backend/bin/cdp_event_gather.py index 81f9c46f..51f83b65 100644 --- a/cdp_backend/bin/cdp_event_gather.py +++ b/cdp_backend/bin/cdp_event_gather.py @@ -9,7 +9,7 @@ from pathlib import Path from typing import Callable -from ..pipeline import cdp_event_gather_pipeline as pipeline +from cdp_backend.pipeline import cdp_event_gather_pipeline as pipeline ############################################################################### @@ -34,7 +34,7 @@ def __parse(self) -> None: p.add_argument( "-g", "--google-credentials-file", - default=(Path(__file__).parent.parent.parent / "cdp-dev-infra-creds.json"), + default=(Path(__file__).parent.parent.parent / "cdp-creds.json"), type=Path, dest="google_credentials_file", help="Path to the Google Service Account Credentials JSON file.", diff --git a/cdp_backend/bin/create_cdp_event_gather_flow_viz.py b/cdp_backend/bin/create_cdp_event_gather_flow_viz.py new file mode 100644 index 00000000..7c35f0e4 --- /dev/null +++ b/cdp_backend/bin/create_cdp_event_gather_flow_viz.py @@ -0,0 +1,100 @@ +#!/usr/bin/env python +# -- coding: utf-8 -- + +import argparse +import logging +import sys +import traceback +from pathlib import Path +from typing import List + +from cdp_backend.pipeline import cdp_event_gather_pipeline as pipeline +from cdp_backend.pipeline.ingestion_models import ( + EXAMPLE_FILLED_EVENT, + EXAMPLE_MINIMAL_EVENT, + EventIngestionModel, +) + +############################################################################### + +logging.basicConfig( + level=logging.INFO, + format="[%(levelname)4s: %(module)s:%(lineno)4s %(asctime)s] %(message)s", +) +log = logging.getLogger(__name__) + +############################################################################### + + +class Args(argparse.Namespace): + def __init__(self) -> None: + self.__parse() + + def __parse(self) -> None: + p = argparse.ArgumentParser( + prog="cdp_event_gather", + description="Gather, process, and store event data to CDP infrastructure.", + ) + p.add_argument( + "-o", + "--output-file-spec", + type=Path, + default=Path("cdp_event_gather_flow_{ftype}.png"), + dest="output_file", + help=( + "Path spec to where to store the created PNG files. " + "Use `{ftype}` as the formatted spec parameter." + ), + ) + p.parse_args(namespace=self) + + +def fake_get_events_minimal() -> List[EventIngestionModel]: + return [EXAMPLE_MINIMAL_EVENT] + + +def fake_get_events_filled() -> List[EventIngestionModel]: + return [EXAMPLE_FILLED_EVENT] + + +def fake_get_events_many() -> List[EventIngestionModel]: + return [EXAMPLE_MINIMAL_EVENT] 4 + + +def main() -> None: + try: + args = Args() + minimal_flow = pipeline.create_cdp_event_gather_flow( + fake_get_events_minimal, "" + ) + minimal_flow.visualize( + filename=str(args.output_file.with_suffix("")).format(ftype="minimal"), + format="png", + ) + + filled_flow = pipeline.create_cdp_event_gather_flow(fake_get_events_filled, "") + filled_flow.visualize( + filename=str(args.output_file.with_suffix("")).format(ftype="filled"), + format="png", + ) + + many_flow = pipeline.create_cdp_event_gather_flow(fake_get_events_many, "") + many_flow.visualize( + filename=str(args.output_file.with_suffix("")).format(ftype="many"), + format="png", + ) + + except Exception as e: + log.error("=============================================") + log.error("\n\n" + traceback.format_exc()) + log.error("=============================================") + log.error("\n\n" + str(e) + "\n") + log.error("=============================================") + sys.exit(1) + + +############################################################################### +# Allow caller to directly run this module (usually in development scenarios) + +if __name__ == "__main__": + main() diff --git a/cdp_backend/pipeline/cdp_event_gather_pipeline.py b/cdp_backend/pipeline/cdp_event_gather_pipeline.py index 72b3766d..ed85d218 100644 --- a/cdp_backend/pipeline/cdp_event_gather_pipeline.py +++ b/cdp_backend/pipeline/cdp_event_gather_pipeline.py @@ -29,9 +29,6 @@ def create_cdp_event_gather_flow( get_events_func: Callable, credentials_file: str, ) -> Flow: - # Initialize fireo connection - fireo.connection(from_file=credentials_file) - # Create flow with Flow("CDP Event Gather Pipeline") as flow: events: List[EventIngestionModel] = get_events_func() @@ -42,18 +39,24 @@ def create_cdp_event_gather_flow( # Upload calls for minimal event body_ref = upload_db_model( - create_body_from_ingestion_model(event.body), event.body + create_body_from_ingestion_model(event.body), + event.body, + creds_file=credentials_file, ) # TODO add upload calls for non-minimal event event_ref = upload_db_model( - create_event_from_ingestion_model(event, body_ref), event + create_event_from_ingestion_model(event, body_ref), + event, + creds_file=credentials_file, ) for session in event.sessions: upload_db_model( - create_session_from_ingestion_model(session, event_ref), session + create_session_from_ingestion_model(session, event_ref), + session, + creds_file=credentials_file, ) return flow @@ -72,7 +75,6 @@ def update_db_model(db_model: Model, ingestion_model: Any, db_key: str) -> Model ] needs_update = False - for field in non_primary_db_fields: if hasattr(ingestion_model, field): db_val = getattr(db_model, field) @@ -93,9 +95,11 @@ def update_db_model(db_model: Model, ingestion_model: Any, db_key: str) -> Model @task -def upload_db_model(db_model: Model, ingestion_model: Any) -> Model: - uniqueness_validation = get_model_uniqueness(db_model) +def upload_db_model(db_model: Model, ingestion_model: Any, creds_file: str) -> Model: + # Initialize fireo connection + fireo.connection(from_file=creds_file) + uniqueness_validation = get_model_uniqueness(db_model) if uniqueness_validation.is_unique: db_model.save() log.info( diff --git a/docs/event_gather_pipeline.md b/docs/event_gather_pipeline.md new file mode 100644 index 00000000..23d78432 --- /dev/null +++ b/docs/event_gather_pipeline.md @@ -0,0 +1,64 @@ +# Event Gather Pipeline + +Council Data Project documentation generally shortens the steps that the +event gather pipeline takes into three short and easy to remember key points. + +1. Gather Events<br> + This function is created by each instance maintainer and + generally scrapes or requests data from an API and + returns a list of events to process and store. + [Learn more about the event ingestion model](./cdp-backend/event_data_ingestion_model.html) +2. Transcribe<br> + If the event was provided with a closed caption file we clean the file up + into the CDP format, if not, CDP tools generate a transcription. +3. Index<br> + An entirely different pipeline from the event gather pipeline but in + the context of the short and sweet version, this makes sense to include. + This is the pipeline that runs to generate and index using all the event + transcripts. + +These key points make sense from the thousand-foot-view, but is not +satifactory for the developers and others who are simply interested in +how the event gather pipeline works step-by-step. + +This document gives visualizations of each and every step the event gather +pipeline takes, starting immediately after the maintainer provided function +returns the `List[EventIngestionModel]`. + +### Minimal Event Processing + +Similar to the [event ingestion model](./cdp-backend/event_data_ingestion_model.html) +documentation, if the event is provided back with only a URI to a +video and nothing else, the processing is incredibly simple. + +![minimal cdp event gather flow viz](./_static/cdp_event_gather_flow_minimal.png) + +### Filled Event Processing + +The major difference from the minimal event processing workflow is that the +pipeline will process, prepare, and upload all of the attachments. + +![filled cdp event gather flow viz](./_static/cdp_event_gather_flow_filled.png) + +### Workflow Management and Parallel Processing + +We utilize [Prefect Core](https://github.com/prefecthq/prefect) to generate and +run the workflow. In general, and by default in our cookiecutter-cdp-deployment, +we do not run any part of the workflow in parallel. But, this is simply because we +generally don't need to. As all event gather pipelines are automated to run every day, +or multiple times a day, it is rare for a pipeline to pick up more than one event +to process. + +However, in the case where a maintainer is attempting to backfill large collections of +events (i.e. processing historical events), we recommend attaching a +[Dask Executor](https://docs.prefect.io/api/latest/executors.html#daskexecutor) +to the flow for parallel event processing. +[More details about Dask](https://dask.org/) + +I.E. +If the result of the `get_events` function was a list of four +`EventIngestionModel` objects, the resulting flow would look like the following: +![parallel four wide equal minimal event flows](./_static/cdp_event_gather_flow_many.png) + +And, if a `DaskExecutor` is also provided, all single event processing workflows +would process in parallel (provided enough compute resources). diff --git a/docs/event_gather_pipeline.rst b/docs/event_gather_pipeline.rst new file mode 100644 index 00000000..f1a2ebf2 --- /dev/null +++ b/docs/event_gather_pipeline.rst @@ -0,0 +1,1 @@ +.. mdinclude:: ./event_gather_pipeline.md diff --git a/docs/index.rst b/docs/index.rst index bef157d1..935488c5 100644 --- a/docs/index.rst +++ b/docs/index.rst @@ -10,6 +10,7 @@ Welcome to cdp-backend's documentation! installation Package Modules <modules> Event Data Ingestion Model <event_data_ingestion_model> + Event Gather Pipeline <event_gather_pipeline> Database Schema <database_schema> contributing diff --git a/setup.py b/setup.py index e78e72ce..7b180b66 100644 --- a/setup.py +++ b/setup.py @@ -34,6 +34,7 @@ "ipython>=7.15.0", "jinja2>=2.11.2", "m2r>=0.2.1", + "prefect[viz]~=0.14.0", "pytest-runner>=5.2", "Sphinx>=2.0.0b1,<3", "sphinx_rtd_theme>=0.4.3", @@ -43,15 +44,15 @@ ] requirements = [ - "dask[bag]~=2.30.0", + "dask[bag]~=2020.12.0", "fireo~=1.3.7", "fsspec~=0.8.3", "gcsfs~=0.7.1", "graphviz~=0.14", "pandas~=1.1.3", + "prefect~=0.14.0", "pulumi~=2.16.2", "pulumi-gcp~=4.6.0", - "prefect~=0.14.0" ] extra_requirements = { @@ -87,6 +88,10 @@ "create_cdp_ingestion_model_doc=" "cdp_backend.bin.create_ingestion_model_doc:main" ), + ( + "create_cdp_event_gather_flow_viz=" + "cdp_backend.bin.create_cdp_event_gather_flow_viz:main" + ), "cdp_event_gather=cdp_backend.bin.cdp_event_gather:main", ], },	Create a bin script to generate the flow visualization Create a bin script to generate event gather flow visualization to attach to documentation after #16 is completed. _Originally posted by @JacksonMaxfield in https://github.com/CouncilDataProject/cdp-backend/pull/33#discussion_r547063942_		2021-01-03T03:46:08	[]	[]

securefederatedai/openfl

securefederatedai__openfl-516

3adc29d3bdc0fcc71f6bd3fc578a1d0fad88a41f

diff --git a/openfl-tutorials/Federated_PyTorch_UNET_Tutorial.ipynb b/openfl-tutorials/Federated_PyTorch_UNET_Tutorial.ipynb index 72f75aef54..7ee6c2e692 100644 --- a/openfl-tutorials/Federated_PyTorch_UNET_Tutorial.ipynb +++ b/openfl-tutorials/Federated_PyTorch_UNET_Tutorial.ipynb @@ -76,8 +76,8 @@ "outputs": [], "source": [ "!wget 'https://datasets.simula.no/downloads/hyper-kvasir/hyper-kvasir-segmented-images.zip' -O kvasir.zip\n", - "ZIP_SHA384 = 'e30d18a772c6520476e55b610a4db457237f151e'\\\n", - " '19182849d54b49ae24699881c1e18e0961f77642be900450ef8b22e7'\n", + "ZIP_SHA384 = ('66cd659d0e8afd8c83408174'\n", + " '1ade2b75dada8d4648b816f2533c8748b1658efa3d49e205415d4116faade2c5810e241e')\n", "validate_file_hash('./kvasir.zip', ZIP_SHA384)\n", "!unzip -n kvasir.zip -d ./data" ] @@ -525,8 +525,19 @@ } ], "metadata": { + "kernelspec": { + "display_name": "Python 3.8.10 64-bit", + "language": "python", + "name": "python3" + }, "language_info": { - "name": "python" + "name": "python", + "version": "3.8.10" + }, + "vscode": { + "interpreter": { + "hash": "916dbcbb3f70747c44a77c7bcd40155683ae19c65e1c03b4aa3499c5328201f1" + } } }, "nbformat": 4, diff --git a/openfl-tutorials/interactive_api/PyTorch_Kvasir_UNet/envoy/kvasir_shard_descriptor.py b/openfl-tutorials/interactive_api/PyTorch_Kvasir_UNet/envoy/kvasir_shard_descriptor.py index 1107d2decf..2a83543250 100644 --- a/openfl-tutorials/interactive_api/PyTorch_Kvasir_UNet/envoy/kvasir_shard_descriptor.py +++ b/openfl-tutorials/interactive_api/PyTorch_Kvasir_UNet/envoy/kvasir_shard_descriptor.py @@ -100,8 +100,8 @@ def download_data(data_folder): "hyper-kvasir/hyper-kvasir-segmented-images.zip'" f' -O {zip_file_path.relative_to(Path.cwd())}' ) - zip_sha384 = ('e30d18a772c6520476e55b610a4db457237f151e' - '19182849d54b49ae24699881c1e18e0961f77642be900450ef8b22e7') + zip_sha384 = ('66cd659d0e8afd8c83408174' + '1ade2b75dada8d4648b816f2533c8748b1658efa3d49e205415d4116faade2c5810e241e') validate_file_hash(zip_file_path, zip_sha384) os.system(f'unzip -n {zip_file_path.relative_to(Path.cwd())}' f' -d {data_folder.relative_to(Path.cwd())}') diff --git a/openfl-tutorials/interactive_api/PyTorch_Kvasir_UNet/envoy/kvasir_shard_descriptor_with_data_splitter.py b/openfl-tutorials/interactive_api/PyTorch_Kvasir_UNet/envoy/kvasir_shard_descriptor_with_data_splitter.py index 687b1ceb87..09bcde87c0 100644 --- a/openfl-tutorials/interactive_api/PyTorch_Kvasir_UNet/envoy/kvasir_shard_descriptor_with_data_splitter.py +++ b/openfl-tutorials/interactive_api/PyTorch_Kvasir_UNet/envoy/kvasir_shard_descriptor_with_data_splitter.py @@ -97,8 +97,8 @@ def download_data(data_folder): " 'https://datasets.simula.no/downloads/hyper-kvasir/" "hyper-kvasir-segmented-images.zip'" f' -O {zip_file_path.relative_to(Path.cwd())}') - zip_sha384 = ('e30d18a772c6520476e55b610a4db457237f151e' - '19182849d54b49ae24699881c1e18e0961f77642be900450ef8b22e7') + zip_sha384 = ('66cd659d0e8afd8c83408174' + '1ade2b75dada8d4648b816f2533c8748b1658efa3d49e205415d4116faade2c5810e241e') validate_file_hash(zip_file_path, zip_sha384) os.system(f'unzip -n {zip_file_path.relative_to(Path.cwd())}' f' -d {data_folder.relative_to(Path.cwd())}') diff --git a/openfl-workspace/torch_unet_kvasir/src/data_loader.py b/openfl-workspace/torch_unet_kvasir/src/data_loader.py index b038a884ef..15db0d63c1 100644 --- a/openfl-workspace/torch_unet_kvasir/src/data_loader.py +++ b/openfl-workspace/torch_unet_kvasir/src/data_loader.py @@ -32,7 +32,7 @@ def read_data(image_path, mask_path): """ img = io.imread(image_path) - assert(img.shape[2] == 3) + assert (img.shape[2] == 3) mask = io.imread(mask_path) return (img, mask[:, :, 0].astype(np.uint8)) @@ -59,7 +59,7 @@ def __init__(self, is_validation, shard_num, collaborator_count, **kwargs): self.images_names = self.images_names[shard_num:: collaborator_count] self.is_validation = is_validation - assert(len(self.images_names) > 8) + assert (len(self.images_names) > 8) validation_size = len(self.images_names) // 8 if is_validation: @@ -92,8 +92,8 @@ def __len__(self): def load_kvasir_dataset(): """Load and unzip kvasir dataset.""" - zip_sha384 = ('e30d18a772c6520476e55b610a4db457237f151e' - '19182849d54b49ae24699881c1e18e0961f77642be900450ef8b22e7') + zip_sha384 = ('66cd659d0e8afd8c83408174' + '1ade2b75dada8d4648b816f2533c8748b1658efa3d49e205415d4116faade2c5810e241e') data_url = ('https://datasets.simula.no/downloads/' 'hyper-kvasir/hyper-kvasir-segmented-images.zip') filename = 'kvasir.zip' diff --git a/openfl-workspace/torch_unet_kvasir_gramine_ready/src/data_loader.py b/openfl-workspace/torch_unet_kvasir_gramine_ready/src/data_loader.py index b038a884ef..15db0d63c1 100644 --- a/openfl-workspace/torch_unet_kvasir_gramine_ready/src/data_loader.py +++ b/openfl-workspace/torch_unet_kvasir_gramine_ready/src/data_loader.py @@ -32,7 +32,7 @@ def read_data(image_path, mask_path): """ img = io.imread(image_path) - assert(img.shape[2] == 3) + assert (img.shape[2] == 3) mask = io.imread(mask_path) return (img, mask[:, :, 0].astype(np.uint8)) @@ -59,7 +59,7 @@ def __init__(self, is_validation, shard_num, collaborator_count, **kwargs): self.images_names = self.images_names[shard_num:: collaborator_count] self.is_validation = is_validation - assert(len(self.images_names) > 8) + assert (len(self.images_names) > 8) validation_size = len(self.images_names) // 8 if is_validation: @@ -92,8 +92,8 @@ def __len__(self): def load_kvasir_dataset(): """Load and unzip kvasir dataset.""" - zip_sha384 = ('e30d18a772c6520476e55b610a4db457237f151e' - '19182849d54b49ae24699881c1e18e0961f77642be900450ef8b22e7') + zip_sha384 = ('66cd659d0e8afd8c83408174' + '1ade2b75dada8d4648b816f2533c8748b1658efa3d49e205415d4116faade2c5810e241e') data_url = ('https://datasets.simula.no/downloads/' 'hyper-kvasir/hyper-kvasir-segmented-images.zip') filename = 'kvasir.zip'

Example fails: Dataset hash needs correction I faced errors running `torch_unet_kvasir` and `torch_unet_kvasir_gramine_ready` examples. The message is attached [error.log](https://github.com/intel/openfl/files/9597992/error.log)

2022-09-28T15:01:54

0.0

[]

securefederatedai/openfl

securefederatedai__openfl-369

b133a439dd44b55f66076aeaaef26f0e60e3119a

diff --git a/openfl/component/aggregator/aggregator.py b/openfl/component/aggregator/aggregator.py index 590cf9f243..c4f092058f 100644 --- a/openfl/component/aggregator/aggregator.py +++ b/openfl/component/aggregator/aggregator.py @@ -11,6 +11,7 @@ from openfl.pipelines import TensorCodec from openfl.protocols import base_pb2 from openfl.protocols import utils +from openfl.utilities import change_tags from openfl.utilities import TaskResultKey from openfl.utilities import TensorKey from openfl.utilities.logs import write_metric @@ -328,7 +329,7 @@ def get_aggregated_tensor(self, collaborator_name, tensor_name, require_lossless: bool round_number: int report: bool - tags: list[str] + tags: tuple[str, ...] Returns: named_tensor : protobuf NamedTensor the tensor requested by the collaborator @@ -341,18 +342,16 @@ def get_aggregated_tensor(self, collaborator_name, tensor_name, else: compress_lossless = False - tags = list(tags) - # TODO the TensorDB doesn't support compressed data yet. # The returned tensor will # be recompressed anyway. if 'compressed' in tags: - tags.remove('compressed') + tags = change_tags(tags, remove_field='compressed') if 'lossy_compressed' in tags: - tags.remove('lossy_compressed') + tags = change_tags(tags, remove_field='lossy_compressed') tensor_key = TensorKey( - tensor_name, self.uuid, round_number, report, tuple(tags) + tensor_name, self.uuid, round_number, report, tags ) tensor_name, origin, round_number, report, tags = tensor_key @@ -583,10 +582,8 @@ def _process_named_tensor(self, named_tensor, collaborator_name): ) dec_name, dec_origin, dec_round_num, dec_report, dec_tags = dec_tk # Need to add the collaborator tag to the resulting tensor - if type(dec_tags) == str: - new_tags = tuple([dec_tags] + [collaborator_name]) - else: - new_tags = tuple(list(dec_tags) + [collaborator_name]) + new_tags = change_tags(dec_tags, add_field=collaborator_name) + # layer.agg.n.trained.delta.col_i decompressed_tensor_key = TensorKey( dec_name, dec_origin, dec_round_num, dec_report, new_tags @@ -599,10 +596,7 @@ def _process_named_tensor(self, named_tensor, collaborator_name): require_lossless=False ) dec_name, dec_origin, dec_round_num, dec_report, dec_tags = dec_tk - if type(dec_tags) == str: - new_tags = tuple([dec_tags] + [collaborator_name]) - else: - new_tags = tuple(list(dec_tags) + [collaborator_name]) + new_tags = change_tags(dec_tags, add_field=collaborator_name) # layer.agg.n.trained.delta.lossy_decompressed.col_i decompressed_tensor_key = TensorKey( dec_name, dec_origin, dec_round_num, dec_report, new_tags @@ -778,11 +772,11 @@ def _compute_validation_related_task_metrics(self, task_name): task_key = TaskResultKey(task_name, collaborators_for_task[0], self.round_number) for tensor_key in self.collaborator_tasks_results[task_key]: tensor_name, origin, round_number, report, tags = tensor_key - assert (tags[-1] == collaborators_for_task[0]), ( + assert (collaborators_for_task[0] in tags), ( f'Tensor {tensor_key} in task {task_name} has not been processed correctly' ) # Strip the collaborator label, and lookup aggregated tensor - new_tags = tuple(tags[:-1]) + new_tags = change_tags(tags, remove_field=collaborators_for_task[0]) agg_tensor_key = TensorKey(tensor_name, origin, round_number, report, new_tags) agg_tensor_name, agg_origin, agg_round_number, agg_report, agg_tags = agg_tensor_key agg_function = WeightedAverage() if 'metric' in tags else task_agg_function diff --git a/openfl/databases/tensor_db.py b/openfl/databases/tensor_db.py index 077e1d2fe7..b412212a19 100644 --- a/openfl/databases/tensor_db.py +++ b/openfl/databases/tensor_db.py @@ -12,6 +12,7 @@ import pandas as pd from openfl.component.aggregation_functions import AggregationFunction +from openfl.utilities import change_tags from openfl.utilities import LocalTensor from openfl.utilities import TensorKey @@ -145,10 +146,7 @@ def get_aggregated_tensor(self, tensor_key: TensorKey, collaborator_weight_dict: return np.array(raw_df.iloc[0]), {} for col in collaborator_names: - if type(tags) == str: - new_tags = tuple([tags] + [col]) - else: - new_tags = tuple(list(tags) + [col]) + new_tags = change_tags(tags, add_field=col) raw_df = self.tensor_db[ (self.tensor_db['tensor_name'] == tensor_name) & (self.tensor_db['origin'] == origin) diff --git a/openfl/federated/task/runner_fe.py b/openfl/federated/task/runner_fe.py index acace8c61b..f76419cb14 100644 --- a/openfl/federated/task/runner_fe.py +++ b/openfl/federated/task/runner_fe.py @@ -4,6 +4,7 @@ import numpy as np +from openfl.utilities import change_tags from openfl.utilities import split_tensor_dict_for_holdouts from openfl.utilities import TensorKey from .runner import TaskRunner @@ -181,7 +182,8 @@ def validate(self, col_name, round_num, input_tensor_dict, **kwargs): suffix += '_local' else: suffix += '_agg' - tags = ('metric', suffix) + tags = ('metric',) + tags = change_tags(tags, add_field=suffix) output_tensor_dict = { TensorKey( metric, origin, round_num, True, tags diff --git a/openfl/federated/task/runner_keras.py b/openfl/federated/task/runner_keras.py index 93ac40b1d0..331e01414c 100644 --- a/openfl/federated/task/runner_keras.py +++ b/openfl/federated/task/runner_keras.py @@ -11,6 +11,7 @@ import numpy as np +from openfl.utilities import change_tags from openfl.utilities import Metric from openfl.utilities import split_tensor_dict_for_holdouts from openfl.utilities import TensorKey @@ -230,7 +231,8 @@ def validate(self, col_name, round_num, input_tensor_dict, **kwargs): suffix += '_local' else: suffix += '_agg' - tags = ('metric', suffix) + tags = ('metric',) + tags = change_tags(tags, add_field=suffix) output_tensor_dict = { TensorKey(metric, origin, round_num, True, tags): np.array(ret_dict[metric]) diff --git a/openfl/federated/task/runner_pt.py b/openfl/federated/task/runner_pt.py index 2905f09798..0746a48f76 100644 --- a/openfl/federated/task/runner_pt.py +++ b/openfl/federated/task/runner_pt.py @@ -12,6 +12,7 @@ import torch.nn as nn import tqdm +from openfl.utilities import change_tags from openfl.utilities import Metric from openfl.utilities import split_tensor_dict_for_holdouts from openfl.utilities import TensorKey @@ -117,7 +118,8 @@ def validate(self, col_name, round_num, input_tensor_dict, suffix += '_local' else: suffix += '_agg' - tags = ('metric', suffix) + tags = ('metric',) + tags = change_tags(tags, add_field=suffix) # TODO figure out a better way to pass in metric for this pytorch # validate function output_tensor_dict = { diff --git a/openfl/federated/task/task_runner.py b/openfl/federated/task/task_runner.py index 05023e4743..05d8ec0e83 100644 --- a/openfl/federated/task/task_runner.py +++ b/openfl/federated/task/task_runner.py @@ -6,6 +6,7 @@ import numpy as np +from openfl.utilities import change_tags from openfl.utilities import split_tensor_dict_for_holdouts from openfl.utilities import TensorKey @@ -73,7 +74,7 @@ def _prepare_tensorkeys_for_agggregation(self, metric_dict, validation_flag, else: suffix = 'validate' + validation_flag tags = (suffix,) - tags = ('metric', *tags) + tags = change_tags(tags, add_field='metric') metric_dict = { TensorKey(metric, origin, round_num, True, tags): np.array(value) for metric, value in metric_dict.items() diff --git a/openfl/pipelines/tensor_codec.py b/openfl/pipelines/tensor_codec.py index 5312a8c0f1..85dfa8c552 100644 --- a/openfl/pipelines/tensor_codec.py +++ b/openfl/pipelines/tensor_codec.py @@ -6,6 +6,7 @@ import numpy as np from openfl.pipelines import NoCompressionPipeline +from openfl.utilities import change_tags from openfl.utilities import TensorKey @@ -67,9 +68,9 @@ def compress(self, tensor_key, data, require_lossless=False, **kwargs): # returned ('trained.delta'->'trained.delta.lossy_compressed') tensor_name, origin, round_number, report, tags = tensor_key if not self.compression_pipeline.is_lossy() or require_lossless: - new_tags = tuple(list(tags) + ['compressed']) + new_tags = change_tags(tags, add_field='compressed') else: - new_tags = tuple(list(tags) + ['lossy_compressed']) + new_tags = change_tags(tags, add_field='lossy_compressed') compressed_tensor_key = TensorKey( tensor_name, origin, round_number, report, new_tags) return compressed_tensor_key, compressed_nparray, metadata @@ -121,18 +122,16 @@ def decompress(self, tensor_key, data, transformer_metadata, data, transformer_metadata, **kwargs) # Define the decompressed tensorkey that should be returned if 'lossy_compressed' in tags: - lc_idx = tags.index('lossy_compressed') - new_tags = list(tags) - new_tags[lc_idx] = 'lossy_decompressed' + new_tags = change_tags( + tags, add_field='lossy_decompressed', remove_field='lossy_compressed') decompressed_tensor_key = TensorKey( - tensor_name, origin, round_number, report, tuple(new_tags)) + tensor_name, origin, round_number, report, new_tags) elif 'compressed' in tags: # 'compressed' == lossless compression; no need for # compression related tag after decompression - new_tags = list(tags) - new_tags.remove('compressed') + new_tags = change_tags(tags, remove_field='compressed') decompressed_tensor_key = TensorKey( - tensor_name, origin, round_number, report, tuple(new_tags)) + tensor_name, origin, round_number, report, new_tags) else: raise NotImplementedError( 'Decompression is only supported on compressed data') @@ -170,10 +169,7 @@ def generate_delta(tensor_key, nparray, base_model_nparray): assert 'model' not in tags, ( 'The tensorkey should be provided ' 'from the layer with new weights, not the base model') - if type(tags) == str: - new_tags = tuple([tensor_key[3]] + ['delta']) - else: - new_tags = tuple(list(tags) + ['delta']) + new_tags = change_tags(tags, add_field='delta') delta_tensor_key = TensorKey( tensor_name, origin, round_number, report, new_tags) return delta_tensor_key, nparray - base_model_nparray @@ -209,9 +205,7 @@ def apply_delta(tensor_key, delta, base_model_nparray, creates_model=False): # from the base model' # Aggregator UUID has the prefix 'aggregator' if 'aggregator' in origin and not creates_model: - tags = list(tags) - tags.remove('delta') - new_tags = tuple(tags) + new_tags = change_tags(tags, remove_field='delta') new_model_tensor_key = TensorKey( tensor_name, origin, round_number, report, new_tags) else: diff --git a/openfl/transport/grpc/aggregator_server.py b/openfl/transport/grpc/aggregator_server.py index a62f18ccc1..cb19a56d77 100644 --- a/openfl/transport/grpc/aggregator_server.py +++ b/openfl/transport/grpc/aggregator_server.py @@ -185,7 +185,7 @@ def GetAggregatedTensor(self, request, context): # NOQA:N802 require_lossless = request.require_lossless round_number = request.round_number report = request.report - tags = request.tags + tags = tuple(request.tags) named_tensor = self.aggregator.get_aggregated_tensor( collaborator_name, tensor_name, round_number, report, tags, require_lossless) diff --git a/openfl/utilities/utils.py b/openfl/utilities/utils.py index 8bd0feb313..bc04aec990 100644 --- a/openfl/utilities/utils.py +++ b/openfl/utilities/utils.py @@ -236,3 +236,25 @@ def merge_configs( settings.set(key, operation(value)) settings.validators.validate() return settings + + +def change_tags(tags, *, add_field=None, remove_field=None) -> Tuple[str, ...]: + """Change tensor tags to add or remove fields. + + Args: + tags(tuple): tensor tags. + add_field(str): add a new tensor tag field. + remove_field(str): remove a tensor tag field. + """ + tags = list(set(tags)) + + if add_field is not None and add_field not in tags: + tags.append(add_field) + if remove_field is not None: + if remove_field in tags: + tags.remove(remove_field) + else: + raise Exception(f'{remove_field} not in tags {tuple(tags)}') + + tags = tuple(sorted(tags)) + return tags

Get organised with tensor tags type. Tensor tags, that are used by the aggregator, can be different types inside the aggregator in different methods. It can be list, tuple, str or <class 'google.protobuf.pyext._message.RepeatedScalarContainer'>. One methods cast it to list, others to tuple. We should unify it.

2022-03-07T20:33:28

0.0

[]

securefederatedai/openfl

securefederatedai__openfl-149

8dfe170b5d45e02daa34a2bff63d715ce996caf3

diff --git a/openfl-tutorials/interactive_api/Director_Pytorch_Kvasir_UNET/director_folder/director_config.yaml b/openfl-tutorials/interactive_api/Director_Pytorch_Kvasir_UNET/director_folder/director_config.yaml index 9b45b60869..bc52829c27 100644 --- a/openfl-tutorials/interactive_api/Director_Pytorch_Kvasir_UNET/director_folder/director_config.yaml +++ b/openfl-tutorials/interactive_api/Director_Pytorch_Kvasir_UNET/director_folder/director_config.yaml @@ -1,4 +1,5 @@ settings: - listen_ip: localhost + listen_addr: localhost + listen_port: 50051 sample_shape: ['300', '400', '3'] target_shape: ['300', '400'] \ No newline at end of file diff --git a/openfl-tutorials/interactive_api/Director_Pytorch_Kvasir_UNET/envoy_folder/kvasir_shard_descriptor.py b/openfl-tutorials/interactive_api/Director_Pytorch_Kvasir_UNET/envoy_folder/kvasir_shard_descriptor.py index bb6dc92681..ef7e5f5406 100644 --- a/openfl-tutorials/interactive_api/Director_Pytorch_Kvasir_UNET/envoy_folder/kvasir_shard_descriptor.py +++ b/openfl-tutorials/interactive_api/Director_Pytorch_Kvasir_UNET/envoy_folder/kvasir_shard_descriptor.py @@ -119,11 +119,13 @@ def dataset_description(self) -> str: from openfl.component.envoy.envoy import Envoy shard_name = 'one' - director_uri = 'localhost:50051' + director_host = 'localhost' + director_port = 50051 keeper = Envoy( shard_name=shard_name, - director_uri=director_uri, + director_host=director_host, + director_port=director_port, shard_descriptor=kvasir_sd, tls=True, root_certificate='./cert/root_ca.crt', diff --git a/openfl-tutorials/interactive_api/Director_Pytorch_Kvasir_UNET/envoy_folder/start_envoy.sh b/openfl-tutorials/interactive_api/Director_Pytorch_Kvasir_UNET/envoy_folder/start_envoy.sh index c546d40138..222d3988e0 100644 --- a/openfl-tutorials/interactive_api/Director_Pytorch_Kvasir_UNET/envoy_folder/start_envoy.sh +++ b/openfl-tutorials/interactive_api/Director_Pytorch_Kvasir_UNET/envoy_folder/start_envoy.sh @@ -1,4 +1,4 @@ #!/bin/bash set -e -fx envoy start -n env_one --disable-tls --shard-config-path shard_config.yaml -d localhost:50051 \ No newline at end of file +fx envoy start -n env_one --disable-tls --shard-config-path shard_config.yaml -dh localhost -dp 50051 \ No newline at end of file diff --git a/openfl-tutorials/interactive_api/Director_Pytorch_Kvasir_UNET/envoy_folder/start_envoy_with_tls.sh b/openfl-tutorials/interactive_api/Director_Pytorch_Kvasir_UNET/envoy_folder/start_envoy_with_tls.sh index 295f61b101..b4339cc0cb 100644 --- a/openfl-tutorials/interactive_api/Director_Pytorch_Kvasir_UNET/envoy_folder/start_envoy_with_tls.sh +++ b/openfl-tutorials/interactive_api/Director_Pytorch_Kvasir_UNET/envoy_folder/start_envoy_with_tls.sh @@ -3,4 +3,4 @@ set -e ENVOY_NAME=$1 DIRECTOR_FQDN=$2 -fx envoy start -n "$ENVOY_NAME" --shard-config-path shard_config.yaml -d "$DIRECTOR_FQDN":50051 -rc cert/root_ca.crt -pk cert/"$ENVOY_NAME".key -oc cert/"$ENVOY_NAME".crt \ No newline at end of file +fx envoy start -n "$ENVOY_NAME" --shard-config-path shard_config.yaml -dh"$DIRECTOR_FQDN" -dp 50051 -rc cert/root_ca.crt -pk cert/"$ENVOY_NAME".key -oc cert/"$ENVOY_NAME".crt \ No newline at end of file diff --git a/openfl/component/envoy/envoy.py b/openfl/component/envoy/envoy.py index de3b036779..ab02615b4b 100644 --- a/openfl/component/envoy/envoy.py +++ b/openfl/component/envoy/envoy.py @@ -24,7 +24,7 @@ class Envoy: """Envoy class.""" - def __init__(self, shard_name, director_uri, shard_descriptor, + def __init__(self, *, shard_name, director_host, director_port, shard_descriptor, root_certificate: str = None, private_key: str = None, certificate: str = None, tls: bool = True) -> None: """Initialize a envoy object.""" @@ -33,11 +33,15 @@ def __init__(self, shard_name, director_uri, shard_descriptor, root_certificate).absolute() if root_certificate is not None else None self.private_key = Path(private_key).absolute() if root_certificate is not None else None self.certificate = Path(certificate).absolute() if root_certificate is not None else None - self.director_client = ShardDirectorClient(director_uri, shard_name=shard_name, - tls=tls, - root_certificate=root_certificate, - private_key=private_key, - certificate=certificate) + self.director_client = ShardDirectorClient( + director_host=director_host, + director_port=director_port, + shard_name=shard_name, + tls=tls, + root_certificate=root_certificate, + private_key=private_key, + certificate=certificate + ) self.shard_descriptor = shard_descriptor self.executor = ThreadPoolExecutor() self.running_experiments = {} diff --git a/openfl/interface/director.py b/openfl/interface/director.py index 2c182aa42c..bd06dbad6d 100644 --- a/openfl/interface/director.py +++ b/openfl/interface/director.py @@ -48,14 +48,14 @@ def start(director_config_path, tls, root_certificate, private_key, certificate) sample_shape = settings.get('sample_shape', '') target_shape = settings.get('target_shape', '') logger.info(f'Sample shape: {sample_shape}, target shape: {target_shape}') - listen_ip = settings.get('listen_ip') + listen_addr = settings.get('listen_addr') listen_port = settings.get('listen_port') root_certificate = root_certificate or settings.get('root_certificate') private_key = private_key or settings.get('private_key') certificate = certificate or settings.get('certificate') kwargs = {} - if listen_ip: - kwargs['listen_ip'] = listen_ip + if listen_addr: + kwargs['listen_addr'] = listen_addr if listen_port: kwargs['listen_port'] = listen_port director_server = DirectorGRPCServer( diff --git a/openfl/interface/envoy.py b/openfl/interface/envoy.py index ffd0a4fdb1..bde3fc4cd1 100644 --- a/openfl/interface/envoy.py +++ b/openfl/interface/envoy.py @@ -31,8 +31,10 @@ def envoy(context): @envoy.command(name='start') @option('-n', '--shard-name', required=True, help='Current shard name') -@option('-d', '--director-uri', required=True, +@option('-dh', '--director-host', required=True, help='The FQDN of the federation director') +@option('-dp', '--director-port', required=True, + help='The federation director port') @option('--tls/--disable-tls', default=True, is_flag=True, help='Use TLS or not (By default TLS is enabled)') @option('-sc', '--shard-config-path', default='shard_config.yaml', @@ -43,7 +45,7 @@ def envoy(context): help='Path to a private key') @option('-oc', '--public-cert-path', 'certificate', default=None, help='Path to a signed certificate') -def start_(shard_name, director_uri, tls, shard_config_path, +def start_(shard_name, director_host, director_port, tls, shard_config_path, root_certificate, private_key, certificate): """Start the Envoy.""" logger.info('🧿 Starting the Envoy.') @@ -51,7 +53,8 @@ def start_(shard_name, director_uri, tls, shard_config_path, shard_descriptor = shard_descriptor_from_config(shard_config_path) envoy = Envoy( shard_name=shard_name, - director_uri=director_uri, + director_host=director_host, + director_port=director_port, shard_descriptor=shard_descriptor, tls=tls, root_certificate=root_certificate, diff --git a/openfl/interface/interactive_api/federation.py b/openfl/interface/interactive_api/federation.py index 420e696496..0026300412 100644 --- a/openfl/interface/interactive_api/federation.py +++ b/openfl/interface/interactive_api/federation.py @@ -44,8 +44,15 @@ def __init__(self, client_id, director_node_fqdn=None, director_port=None, tls=T self.api_private_key = api_private_key # Create Director client - self.dir_client = DirectorClient(client_id, f'{director_node_fqdn}:{director_port}', - tls, cert_chain, api_private_key, api_cert) + self.dir_client = DirectorClient( + client_id=client_id, + director_host=director_node_fqdn, + director_port=director_port, + tls=tls, + root_certificate=cert_chain, + private_key=api_private_key, + certificate=api_cert + ) # Request sample and target shapes from Director. # This is an internal method for finding out dataset properties in a Federation. diff --git a/openfl/transport/grpc/director_client.py b/openfl/transport/grpc/director_client.py index a21485d29b..be0ccfebcb 100644 --- a/openfl/transport/grpc/director_client.py +++ b/openfl/transport/grpc/director_client.py @@ -25,13 +25,14 @@ class ShardDirectorClient: """The internal director client class.""" - def __init__(self, director_uri, shard_name, tls=True, + def __init__(self, *, director_host, director_port, shard_name, tls=True, root_certificate=None, private_key=None, certificate=None) -> None: """Initialize a shard director client object.""" self.shard_name = shard_name + director_addr = f'{director_host}:{director_port}' options = [('grpc.max_message_length', 100 * 1024 * 1024)] if not tls: - channel = grpc.insecure_channel(director_uri, options=options) + channel = grpc.insecure_channel(director_addr, options=options) else: if not (root_certificate and private_key and certificate): raise Exception('No certificates provided') @@ -50,7 +51,7 @@ def __init__(self, director_uri, shard_name, tls=True, private_key=private_key_b, certificate_chain=certificate_b ) - channel = grpc.secure_channel(director_uri, credentials, options=options) + channel = grpc.secure_channel(director_addr, credentials, options=options) self.stub = director_pb2_grpc.FederationDirectorStub(channel) def report_shard_info(self, shard_descriptor) -> bool: @@ -154,13 +155,14 @@ def send_health_check(self, collaborator_name, is_experiment_running, valid_dura class DirectorClient: """Director client class for users.""" - def __init__(self, client_id, director_uri, tls=True, + def __init__(self, *, client_id, director_host, director_port, tls=True, root_certificate=None, private_key=None, certificate=None) -> None: """Initialize director client object.""" + director_addr = f'{director_host}:{director_port}' channel_opt = [('grpc.max_send_message_length', 512 * 1024 * 1024), ('grpc.max_receive_message_length', 512 * 1024 * 1024)] if not tls: - channel = grpc.insecure_channel(director_uri, options=channel_opt) + channel = grpc.insecure_channel(director_addr, options=channel_opt) else: if not (root_certificate and private_key and certificate): raise Exception('No certificates provided') @@ -180,7 +182,7 @@ def __init__(self, client_id, director_uri, tls=True, certificate_chain=certificate_b ) - channel = grpc.secure_channel(director_uri, credentials, options=channel_opt) + channel = grpc.secure_channel(director_addr, credentials, options=channel_opt) self.stub = director_pb2_grpc.FederationDirectorStub(channel) self.client_id = client_id diff --git a/openfl/transport/grpc/director_server.py b/openfl/transport/grpc/director_server.py index 8e96d22aad..3fc955d312 100644 --- a/openfl/transport/grpc/director_server.py +++ b/openfl/transport/grpc/director_server.py @@ -25,12 +25,12 @@ class DirectorGRPCServer(director_pb2_grpc.FederationDirectorServicer): def __init__(self, *, director_cls, tls: bool = True, root_certificate: str = None, private_key: str = None, certificate: str = None, - listen_ip='[::]', listen_port=50051, **kwargs) -> None: + listen_addr='[::]', listen_port=50051, **kwargs) -> None: """Initialize a director object.""" # TODO: add working directory super().__init__() - self.listen_addr = f'{listen_ip}:{listen_port}' + self.listen_uri = f'{listen_addr}:{listen_port}' self.tls = tls self.root_certificate = None self.private_key = None @@ -86,7 +86,7 @@ async def _run(self): director_pb2_grpc.add_FederationDirectorServicer_to_server(self, self.server) if not self.tls: - self.server.add_insecure_port(self.listen_addr) + self.server.add_insecure_port(self.listen_uri) else: with open(self.private_key, 'rb') as f: private_key_b = f.read() @@ -99,8 +99,8 @@ async def _run(self): root_certificates=root_certificate_b, require_client_auth=True ) - self.server.add_secure_port(self.listen_addr, server_credentials) - logger.info(f'Starting server on {self.listen_addr}') + self.server.add_secure_port(self.listen_uri, server_credentials) + logger.info(f'Starting server on {self.listen_uri}') await self.server.start() await self.server.wait_for_termination()

Need to have a consistent command for passing director's uri As of now, the envoy command asks for director's URI whereas the director for IP and port. It would be great if we could pass in the director's URI at both nodes.

2021-08-13T16:02:34

0.0

[]

securefederatedai/openfl

securefederatedai__openfl-104

bd73b749a9ea1b92dbcdd07e639752101d769fc0

diff --git a/.gitignore b/.gitignore index 4d19e080ce..7364deced3 100644 --- a/.gitignore +++ b/.gitignore @@ -1,4 +1,5 @@ *.egg-info +*.pkl __pycache__ /build /dist @@ -6,3 +7,7 @@ __pycache__ .ipynb_checkpoints venv/* .idea + +*.jpg +*.crt +*.key \ No newline at end of file diff --git a/docs/advanced_topics.rst b/docs/advanced_topics.rst index bb8cd1e953..dafa8b5bc0 100644 --- a/docs/advanced_topics.rst +++ b/docs/advanced_topics.rst @@ -14,5 +14,6 @@ Advanced Topics compression_settings overriding_agg_fn bash_autocomplete_activation + log_metric_callback diff --git a/docs/log_metric_callback.rst b/docs/log_metric_callback.rst new file mode 100644 index 0000000000..46255f6387 --- /dev/null +++ b/docs/log_metric_callback.rst @@ -0,0 +1,69 @@ +.. # Copyright (C) 2020-2021 Intel Corporation +.. # SPDX-License-Identifier: Apache-2.0 + +.. _log_metric_callback: +=============================== +Metric logging callback +=============================== + +------------------------------- +Usage +------------------------------- +|productName| allows developers to use custom metric logging functions. This function will call on aggregator node. +In order to define such function, you should: + +Python API +========== +Define function with follow signature: + +.. code-block:: python + + def callback_name(node_name, task_name, metric_name, metric, round_number): + """ + Write metric callback + + Args: + node_name (str): Name of node, which generate metric + task_name (str): Name of task + metric_name (str): Name of metric + metric (np.ndarray): Metric value + round_number (int): Round number + """ + your code +CLI +==== + +Define callback function similar way like in python api in ``src`` folder of your workspace. And provide a way to your function in ``aggregator`` part of ``plan/plan.yaml`` file in your workspace, use ``log_metric_callback`` key: + +.. code-block:: yaml + + aggregator : + defaults : plan/defaults/aggregator.yaml + template : openfl.component.Aggregator + settings : + init_state_path : save/torch_cnn_mnist_init.pbuf + best_state_path : save/torch_cnn_mnist_best.pbuf + last_state_path : save/torch_cnn_mnist_last.pbuf + rounds_to_train : 10 + log_metric_callback : + template : src.mnist_utils.callback_name + + + +Example +======================= + +Below is an example of a log metric callback, which writes metric values to tensorboard + +.. code-block:: python + + from torch.utils.tensorboard import SummaryWriter + + writer = SummaryWriter('./logs/cnn_mnist', flush_secs=5) + + + def write_metric(node_name, task_name, metric_name, metric, round_number): + writer.add_scalar("{}/{}/{}".format(node_name, task_name, metric_name), + metric, round_number) + +Full implementation can be found in ``openfl-tutorials/Federated_Pytorch_MNIST_Tutorial.ipynb`` and in ``torch_cnn_mnist`` workspace \ No newline at end of file diff --git a/docs/mermaid/pki_scheme.mmd b/docs/mermaid/pki_scheme.mmd new file mode 100644 index 0000000000..9e5b69d6ae --- /dev/null +++ b/docs/mermaid/pki_scheme.mmd @@ -0,0 +1,20 @@ +sequenceDiagram +Title: Collaborator Certificate Signing Flow + participant A as Aggregator + participant CA as CA + participant C as Collaborator + CA->>CA: 1. Create CA: `step ca init --password-file pass_file` + CA->>CA: 2. Up HTTPS CA server: `step_ca ca_config.json` + CA->>CA: 3. Generate JWK pair: `step crypto jwk create pub.json priv.json --password-file pass_file` + CA->>CA: 4. Get JWT for aggregator: `step ca token localhost --key priv.json --password-file pass_file --ca-url ca_url` + CA->>A: 5. Copy JWT to aggregator. + A->>CA: 6. Certify node: `step ca certificate localhost agg.crt agg.key --token AbC1d2E..` + Note over A,CA: Get agg.crt + CA->>CA: 7. Get JWT for collaborator: `step ca token col_name --key priv.json --password-file pass_file --ca-url ca_url` + CA->>C: 8. Copy JWT to collaborator. + C->>CA: 9. Certify node: `step ca certificate col_name col_name.crt col_name.key --token AbC1d2E..` + Note over C,CA: Get col_name.crt + CA->>A: 10. Copy root_ca.crt to aggregator + Note over A,CA: This could be done at step 5 with token + CA->>C: 11. Copy root_ca.crt to collaborator + Note over C,CA: This could be done at step 8 with token \ No newline at end of file diff --git a/docs/running_the_federation.certificates.rst b/docs/running_the_federation.certificates.rst index 3443454cae..4ea49d207b 100644 --- a/docs/running_the_federation.certificates.rst +++ b/docs/running_the_federation.certificates.rst @@ -11,7 +11,8 @@ Therefore, security keys and certificates will need to be created for the aggregator and collaborators to negotiate the connection securely. For the :ref:`Hello Federation <running_the_federation>` demo we will run the aggregator and collaborators on the same localhost server -so these configuration steps just need to be done once on that machine. +so these configuration steps just need to be done once on that machine. We have two pki +workflows: manual and semi-automatic (with step-ca). .. note:: @@ -20,10 +21,16 @@ so these configuration steps just need to be done once on that machine. .. _install_certs: .. kroki:: mermaid/CSR_signing.mmd - :caption: Certificate generation and signing + :caption: Manual certificate generation and signing :align: center :type: mermaid +.. _install_certs: + +.. kroki:: mermaid/pki_scheme.mmd + :caption: Step-ca certificate generation and signing + :align: center + :type: mermaid .. _install_certs_agg: diff --git a/docs/running_the_federation.interactive_api.rst b/docs/running_the_federation.interactive_api.rst index 824caa3f12..6acb9e5eb2 100644 --- a/docs/running_the_federation.interactive_api.rst +++ b/docs/running_the_federation.interactive_api.rst @@ -23,6 +23,58 @@ Python Environment =================== Create a virtual Python environment. Please, install only packages that are required for conducting the experiment, since Python environment will be replicated on collaborator nodes. +****************************************** +Certification +****************************************** +If you have trusted workspace and connection should not be encrypted you can use :code:`disable_tls` option while starting expirement. +Otherwise it is necessary to certify each node participating in the federation. Certificates allow to use mutual tls connection between nodes. +You can certify nodes by your own pki system or use pki provided by OpenFL. It is based on `step-ca <https://github.com/smallstep/certificates>`_ +as a server and `step <https://github.com/smallstep/cli>`_ as a client utilities. They are downloaded from github during workspace setup. Regardless of the certification method, +paths to certificates on each node are provided at start of experiment. Pki workflow from OpenFL will be discussed below. + +OpenFL PKI workflow +=================== +Openfl PKI pipeline asumes creating local CA with https server which listen signing requests. +Certificates from each node can be signed by requesting to CA server with special token. +Token must be copied to each node by some secure way. Each step is considered in detail below. + +1. Create CA, i.e create root key pair, CA server config and other. + .. code-block:: console + + $ fx pki install -p </path/to/ca/dir> --password <123> --ca-url <host:port> + | :code:`-p` - path to folder, which will contain ca files. + | :code:`--password` - password that will encrypts some ca files. + | :code:`--ca-url` - host and port which ca server will listen + This command will also download `step-ca <https://github.com/smallstep/certificates>`_ and `step <https://github.com/smallstep/cli>`_ binaries from github. + +2. Run CA https server. + .. code-block:: console + + $ fx pki run -p </path/to/ca/dir> + | :code:`-p` - path to folder, which will contain ca files. + +3. Get token for some node. + + .. code-block:: console + + $ fx pki get-token -n <subject> + | :code:`-n` - subject name, fqdn for director, collaborator name for envoy or api name for api-layer node + + Run this command on ca side, from ca folder. Output is a token which contains JWT (json web token) from CA server and CA + root certificate concatenated together. This JWT have twenty-four hours time-to-live. + +4. Copy token to node side (director or envoy) by some secure channel and run certify command. + .. code-block:: console + + $ fx pki certify -n <subject> -t <token> + | :code:`-n` - subject name, fqdn for director, collaborator name for envoy or api name for api-layer node + | :code:`-t` - output token from previous command + This command call step client, to connect to CA server over https. + Https is provided by root certificate which was copy with JWT. + Server authenticates client by JWT and client authenticates server by root certificate. + +Now signed certificate and private key are stored on current node. Signed certificate has one year time-to-live. You should certify all node that will participate in federation: director, all envoys and api-layer node. + ****************************************** Defining a Federated Learning Experiment ****************************************** diff --git a/openfl-tutorials/Federated_FedProx_PyTorch_MNIST_Tutorial.ipynb b/openfl-tutorials/Federated_FedProx_PyTorch_MNIST_Tutorial.ipynb index 9419056c12..12936d98e0 100644 --- a/openfl-tutorials/Federated_FedProx_PyTorch_MNIST_Tutorial.ipynb +++ b/openfl-tutorials/Federated_FedProx_PyTorch_MNIST_Tutorial.ipynb @@ -40,7 +40,7 @@ "cell_type": "markdown", "metadata": {}, "source": [ - "After importing the required packages, the next step is setting up our openfl workspace. To do this, simply run the `fx.init()` command as follows:" + "After importing the required packages, the next step is setting Up our openfl workspace. To do this, simply run the `fx.init()` command as follows:" ] }, { diff --git a/openfl-tutorials/Federated_PyTorch_UNET_Tutorial.ipynb b/openfl-tutorials/Federated_PyTorch_UNET_Tutorial.ipynb index a7f116f9ae..b6cc5f81ac 100644 --- a/openfl-tutorials/Federated_PyTorch_UNET_Tutorial.ipynb +++ b/openfl-tutorials/Federated_PyTorch_UNET_Tutorial.ipynb @@ -135,8 +135,11 @@ " def __init__(self, data_path, collaborator_count, collaborator_num, is_validation):\n", " self.images_path = './data/segmented-images/images/'\n", " self.masks_path = './data/segmented-images/masks/'\n", - " self.images_names = [img_name for img_name in sorted(listdir(\n", - " self.images_path)) if len(img_name) > 3 and img_name[-3:] == 'jpg']\n", + " self.images_names = [\n", + " img_name\n", + " for img_name in sorted(listdir(self.images_path))\n", + " if len(img_name) > 3 and img_name[-3:] == 'jpg'\n", + " ]\n", "\n", " self.images_names = self.images_names[collaborator_num:: collaborator_count]\n", " self.is_validation = is_validation\n", @@ -267,9 +270,9 @@ " return score.sum()\n", "\n", "\n", - "class double_conv(nn.Module):\n", + "class DoubleConv(nn.Module):\n", " def __init__(self, in_ch, out_ch):\n", - " super(double_conv, self).__init__()\n", + " super(DoubleConv, self).__init__()\n", " self.in_ch = in_ch\n", " self.out_ch = out_ch\n", " self.conv = nn.Sequential(\n", @@ -286,33 +289,36 @@ " return x\n", "\n", "\n", - "class down(nn.Module):\n", + "class Down(nn.Module):\n", " def __init__(self, in_ch, out_ch):\n", - " super(down, self).__init__()\n", + " super(Down, self).__init__()\n", " self.mpconv = nn.Sequential(\n", - " nn.MaxPool2d(2), double_conv(in_ch, out_ch))\n", + " nn.MaxPool2d(2),\n", + " DoubleConv(in_ch, out_ch)\n", + " )\n", "\n", " def forward(self, x):\n", " x = self.mpconv(x)\n", " return x\n", "\n", "\n", - "class up(nn.Module):\n", + "class Up(nn.Module):\n", " def __init__(self, in_ch, out_ch, bilinear=False):\n", - " super(up, self).__init__()\n", + " super(Up, self).__init__()\n", " self.in_ch = in_ch\n", " self.out_ch = out_ch\n", " if bilinear:\n", - " self.up = nn.Upsample(\n", - " scale_factor=2, mode=\"bilinear\", align_corners=True)\n", - " else:\n", - " self.up = nn.ConvTranspose2d(\n", - " in_ch, in_ch // 2, 2, stride=2\n", + " self.Up = nn.Upsample(\n", + " scale_factor=2,\n", + " mode=\"bilinear\",\n", + " align_corners=True\n", " )\n", - " self.conv = double_conv(in_ch, out_ch)\n", + " else:\n", + " self.Up = nn.ConvTranspose2d(in_ch, in_ch // 2, 2, stride=2)\n", + " self.conv = DoubleConv(in_ch, out_ch)\n", "\n", " def forward(self, x1, x2):\n", - " x1 = self.up(x1)\n", + " x1 = self.Up(x1)\n", " diffY = x2.size()[2] - x1.size()[2]\n", " diffX = x2.size()[3] - x1.size()[3]\n", "\n", @@ -327,15 +333,15 @@ "class UNet(nn.Module):\n", " def __init__(self, n_channels=3, n_classes=1):\n", " super().__init__()\n", - " self.inc = double_conv(n_channels, 64)\n", - " self.down1 = down(64, 128)\n", - " self.down2 = down(128, 256)\n", - " self.down3 = down(256, 512)\n", - " self.down4 = down(512, 1024)\n", - " self.up1 = up(1024, 512)\n", - " self.up2 = up(512, 256)\n", - " self.up3 = up(256, 128)\n", - " self.up4 = up(128, 64)\n", + " self.inc = DoubleConv(n_channels, 64)\n", + " self.down1 = Down(64, 128)\n", + " self.down2 = Down(128, 256)\n", + " self.down3 = Down(256, 512)\n", + " self.down4 = Down(512, 1024)\n", + " self.up1 = Up(1024, 512)\n", + " self.up2 = Up(512, 256)\n", + " self.up3 = Up(256, 128)\n", + " self.up4 = Up(128, 64)\n", " self.outc = nn.Conv2d(64, n_classes, 1)\n", "\n", " def forward(self, x):\n", @@ -558,4 +564,4 @@ }, "nbformat": 4, "nbformat_minor": 4 -} +} \ No newline at end of file diff --git a/openfl-tutorials/Federated_Pytorch_MNIST_Tutorial.ipynb b/openfl-tutorials/Federated_Pytorch_MNIST_Tutorial.ipynb index 52beb21ffb..c889b996f6 100644 --- a/openfl-tutorials/Federated_Pytorch_MNIST_Tutorial.ipynb +++ b/openfl-tutorials/Federated_Pytorch_MNIST_Tutorial.ipynb @@ -49,7 +49,7 @@ "outputs": [], "source": [ "#Setup default workspace, logging, etc.\n", - "fx.init('torch_cnn_mnist')" + "fx.init('torch_cnn_mnist', log_level='METRIC', log_file='./spam_metric.log')" ] }, { @@ -124,6 +124,29 @@ " return F.cross_entropy(input=output,target=target)" ] }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "Here we can define metric logging function. It should has the following signature described below. You can use it to write metrics to tensorboard or some another specific logging." + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "from torch.utils.tensorboard import SummaryWriter\n", + "\n", + "writer = SummaryWriter('./logs/cnn_mnist', flush_secs=5)\n", + "\n", + "\n", + "def write_metric(node_name, task_name, metric_name, metric, round_number):\n", + " writer.add_scalar(\"{}/{}/{}\".format(node_name, task_name, metric_name),\n", + " metric, round_number)" + ] + }, { "cell_type": "code", "execution_count": null, @@ -205,8 +228,12 @@ "metadata": {}, "outputs": [], "source": [ - "#Run experiment, return trained FederatedModel\n", - "final_fl_model = fx.run_experiment(collaborators,{'aggregator.settings.rounds_to_train':5})" + "# Run experiment, return trained FederatedModel\n", + "\n", + "final_fl_model = fx.run_experiment(collaborators, override_config={\n", + " 'aggregator.settings.rounds_to_train': 5,\n", + " 'aggregator.settings.log_metric_callback': write_metric,\n", + "})" ] }, { diff --git a/openfl-tutorials/interactive_api/Director_Pytorch_Kvasir_UNET/director_folder/director_config.yaml b/openfl-tutorials/interactive_api/Director_Pytorch_Kvasir_UNET/director_folder/director_config.yaml new file mode 100644 index 0000000000..9b45b60869 --- /dev/null +++ b/openfl-tutorials/interactive_api/Director_Pytorch_Kvasir_UNET/director_folder/director_config.yaml @@ -0,0 +1,4 @@ +settings: + listen_ip: localhost + sample_shape: ['300', '400', '3'] + target_shape: ['300', '400'] \ No newline at end of file diff --git a/openfl-tutorials/interactive_api/Director_Pytorch_Kvasir_UNET/director_folder/start_director.sh b/openfl-tutorials/interactive_api/Director_Pytorch_Kvasir_UNET/director_folder/start_director.sh new file mode 100644 index 0000000000..5806a6cc0a --- /dev/null +++ b/openfl-tutorials/interactive_api/Director_Pytorch_Kvasir_UNET/director_folder/start_director.sh @@ -0,0 +1,4 @@ +#!/bin/bash +set -e + +fx director start --disable-tls -c director_config.yaml \ No newline at end of file diff --git a/openfl-tutorials/interactive_api/Director_Pytorch_Kvasir_UNET/director_folder/start_director_with_tls.sh b/openfl-tutorials/interactive_api/Director_Pytorch_Kvasir_UNET/director_folder/start_director_with_tls.sh new file mode 100644 index 0000000000..5d6d46a792 --- /dev/null +++ b/openfl-tutorials/interactive_api/Director_Pytorch_Kvasir_UNET/director_folder/start_director_with_tls.sh @@ -0,0 +1,4 @@ +#!/bin/bash +set -e +FQDN=$1 +fx director start -c director_config.yaml -rc cert/root_ca.crt -pk cert/"${FQDN}".key -oc cert/"${FQDN}".crt \ No newline at end of file diff --git a/openfl-tutorials/interactive_api/Director_Pytorch_Kvasir_UNET/envoy_folder/kvasir_shard_descriptor.py b/openfl-tutorials/interactive_api/Director_Pytorch_Kvasir_UNET/envoy_folder/kvasir_shard_descriptor.py new file mode 100644 index 0000000000..69290c1956 --- /dev/null +++ b/openfl-tutorials/interactive_api/Director_Pytorch_Kvasir_UNET/envoy_folder/kvasir_shard_descriptor.py @@ -0,0 +1,135 @@ +# Copyright (C) 2020-2021 Intel Corporation +# SPDX-License-Identifier: Apache-2.0 +"""Kvasir shard descriptor.""" + + +import os +from hashlib import sha384 +from pathlib import Path + +import numpy as np +from PIL import Image + +from openfl.interface.interactive_api.shard_descriptor import ShardDescriptor + + +class KvasirShardDescriptor(ShardDescriptor): + """Shard descriptor class.""" + + def __init__(self, data_folder: str = 'kvasir_data', + rank_worldsize: str = '1,1', + enforce_image_hw: str = None) -> None: + """Initialize KvasirShardDescriptor.""" + super().__init__() + + self.data_folder = Path.cwd() / data_folder + self.download_data(self.data_folder) + + # Settings for resizing data + self.enforce_image_hw = None + if enforce_image_hw is not None: + self.enforce_image_hw = tuple(int(size) for size in enforce_image_hw.split(',')) + # Settings for sharding the dataset + self.rank_worldsize = tuple(int(num) for num in rank_worldsize.split(',')) + + self.images_path = self.data_folder / 'segmented-images' / 'images' + self.masks_path = self.data_folder / 'segmented-images' / 'masks' + + self.images_names = [ + img_name + for img_name in sorted(os.listdir(self.images_path)) + if len(img_name) > 3 and img_name[-3:] == 'jpg' + ] + # Sharding + self.images_names = self.images_names[self.rank_worldsize[0] - 1::self.rank_worldsize[1]] + + # Calculating data and target shapes + sample, target = self[0] + self._sample_shape = [str(dim) for dim in sample.shape] + self._target_shape = [str(dim) for dim in target.shape] + + @staticmethod + def download_data(data_folder): + """Download data.""" + zip_file_path = data_folder / 'kvasir.zip' + os.makedirs(data_folder, exist_ok=True) + os.system('wget -nc' + " 'https://datasets.simula.no/hyper-kvasir/hyper-kvasir-segmented-images.zip'" + f' -O {zip_file_path.relative_to(Path.cwd())}') + zip_sha384 = 'e30d18a772c6520476e55b610a4db457237f151e' \ + '19182849d54b49ae24699881c1e18e0961f77642be900450ef8b22e7' + assert sha384(open(zip_file_path, 'rb').read( + os.path.getsize(zip_file_path))).hexdigest() == zip_sha384 + os.system(f'unzip -n {zip_file_path.relative_to(Path.cwd())}' + f' -d {data_folder.relative_to(Path.cwd())}') + + def __getitem__(self, index): + """Return a item by the index.""" + name = self.images_names[index] + # Reading data + img = Image.open(self.images_path / name) + mask = Image.open(self.masks_path / name) + if self.enforce_image_hw is not None: + # If we need to resize data + # PIL accepts (w,h) tuple, not (h,w) + img = img.resize(self.enforce_image_hw[::-1]) + mask = mask.resize(self.enforce_image_hw[::-1]) + img = np.asarray(img) + mask = np.asarray(mask) + assert img.shape[2] == 3 + + return img, mask[:, :, 0].astype(np.uint8) + + def __len__(self): + """Return the len of the dataset.""" + return len(self.images_names) + + @property + def sample_shape(self): + """Return the sample shape info.""" + return self._sample_shape + + @property + def target_shape(self): + """Return the target shape info.""" + return self._target_shape + + @property + def dataset_description(self) -> str: + """Return the dataset description.""" + return f'Kvasir dataset, shard number {self.rank_worldsize[0]}' \ + f' out of {self.rank_worldsize[1]}' + + +if __name__ == '__main__': + from openfl.interface.cli import setup_logging + setup_logging() + + data_folder = 'data' + rank_worldsize = '1,100' + enforce_image_hw = '529,622' + + kvasir_sd = KvasirShardDescriptor( + data_folder, + rank_worldsize=rank_worldsize, + enforce_image_hw=enforce_image_hw) + + print(kvasir_sd.dataset_description) + print(kvasir_sd.sample_shape, kvasir_sd.target_shape) + + from openfl.component.envoy.envoy import Envoy + + shard_name = 'one' + director_uri = 'localhost:50051' + + keeper = Envoy( + shard_name=shard_name, + director_uri=director_uri, + shard_descriptor=kvasir_sd, + disable_tls=False, + root_ca='./cert/root_ca.crt', + key='./cert/one.key', + cert='./cert/one.crt', + ) + + keeper.start() diff --git a/openfl-tutorials/interactive_api/Director_Pytorch_Kvasir_UNET/envoy_folder/sd_requirements.txt b/openfl-tutorials/interactive_api/Director_Pytorch_Kvasir_UNET/envoy_folder/sd_requirements.txt new file mode 100644 index 0000000000..92606011b1 --- /dev/null +++ b/openfl-tutorials/interactive_api/Director_Pytorch_Kvasir_UNET/envoy_folder/sd_requirements.txt @@ -0,0 +1,2 @@ +numpy +pillow \ No newline at end of file diff --git a/openfl-tutorials/interactive_api/Director_Pytorch_Kvasir_UNET/envoy_folder/shard_config.yaml b/openfl-tutorials/interactive_api/Director_Pytorch_Kvasir_UNET/envoy_folder/shard_config.yaml new file mode 100644 index 0000000000..aba8be3627 --- /dev/null +++ b/openfl-tutorials/interactive_api/Director_Pytorch_Kvasir_UNET/envoy_folder/shard_config.yaml @@ -0,0 +1,5 @@ +template: kvasir_shard_descriptor.KvasirShardDescriptor +params: + data_folder: kvasir_data + rank_worldsize: 1,90 + enforce_image_hw: '300,400' \ No newline at end of file diff --git a/openfl-tutorials/interactive_api/Director_Pytorch_Kvasir_UNET/envoy_folder/start_envoy.sh b/openfl-tutorials/interactive_api/Director_Pytorch_Kvasir_UNET/envoy_folder/start_envoy.sh new file mode 100644 index 0000000000..c546d40138 --- /dev/null +++ b/openfl-tutorials/interactive_api/Director_Pytorch_Kvasir_UNET/envoy_folder/start_envoy.sh @@ -0,0 +1,4 @@ +#!/bin/bash +set -e + +fx envoy start -n env_one --disable-tls --shard-config-path shard_config.yaml -d localhost:50051 \ No newline at end of file diff --git a/openfl-tutorials/interactive_api/Director_Pytorch_Kvasir_UNET/envoy_folder/start_envoy_with_tls.sh b/openfl-tutorials/interactive_api/Director_Pytorch_Kvasir_UNET/envoy_folder/start_envoy_with_tls.sh new file mode 100644 index 0000000000..295f61b101 --- /dev/null +++ b/openfl-tutorials/interactive_api/Director_Pytorch_Kvasir_UNET/envoy_folder/start_envoy_with_tls.sh @@ -0,0 +1,6 @@ +#!/bin/bash +set -e +ENVOY_NAME=$1 +DIRECTOR_FQDN=$2 + +fx envoy start -n "$ENVOY_NAME" --shard-config-path shard_config.yaml -d "$DIRECTOR_FQDN":50051 -rc cert/root_ca.crt -pk cert/"$ENVOY_NAME".key -oc cert/"$ENVOY_NAME".crt \ No newline at end of file diff --git a/openfl-tutorials/interactive_api_tutorials_(experimental)/Pytorch_Kvasir_UNET_workspace/new_python_api_UNET.ipynb b/openfl-tutorials/interactive_api/Director_Pytorch_Kvasir_UNET/workspace/Updated_Kvasir_with_Director.ipynb similarity index 57% rename from openfl-tutorials/interactive_api_tutorials_(experimental)/Pytorch_Kvasir_UNET_workspace/new_python_api_UNET.ipynb rename to openfl-tutorials/interactive_api/Director_Pytorch_Kvasir_UNET/workspace/Updated_Kvasir_with_Director.ipynb index fc4ffe3734..f9081b07b8 100644 --- a/openfl-tutorials/interactive_api_tutorials_(experimental)/Pytorch_Kvasir_UNET_workspace/new_python_api_UNET.ipynb +++ b/openfl-tutorials/interactive_api/Director_Pytorch_Kvasir_UNET/workspace/Updated_Kvasir_with_Director.ipynb @@ -5,10 +5,37 @@ "id": "liquid-jacket", "metadata": {}, "source": [ - "# Federated PyTorch UNET Tutorial\n", + "# Federated Kvasir with Director example\n", "## Using low-level Python API" ] }, + { + "cell_type": "markdown", + "id": "af0579f8", + "metadata": {}, + "source": [ + "# Long-Living entities update\n", + "\n", + "* We now may have director running on another machine.\n", + "* We use Federation API to communicate with Director.\n", + "* Federation object should hold a Director's client (for user service)\n", + "* Keeping in mind that several API instances may be connacted to one Director.\n", + "\n", + "\n", + "* We do not think for now how we start a Director.\n", + "* But it knows the data shape and target shape for the DataScience problem in the Federation.\n", + "* Director holds the list of connected envoys, we do not need to specify it anymore.\n", + "* Director and Envoys are responsible for encrypting connections, we do not need to worry about certs.\n", + "\n", + "\n", + "* Yet we MUST have a cert to communicate to the Director.\n", + "* We MUST know the FQDN of a Director.\n", + "* Director communicates data and target shape to the Federation interface object.\n", + "\n", + "\n", + "* Experiment API may use this info to construct a dummy dataset and a `shard descriptor` stub." + ] + }, { "cell_type": "code", "execution_count": null, @@ -19,180 +46,141 @@ "outputs": [], "source": [ "# Install dependencies if not already installed\n", - "!pip install torchvision==0.8.1\n", - "!pip install scikit-image" + "!pip install torchvision==0.8.1" ] }, { "cell_type": "markdown", - "id": "caring-distinction", + "id": "16986f22", "metadata": {}, "source": [ - "### Describe the model and optimizer" + "# Connect to the Federation" ] }, { "cell_type": "code", "execution_count": null, - "id": "visible-victor", + "id": "4485ac79", "metadata": {}, "outputs": [], "source": [ - "import torch\n", - "import torch.nn as nn\n", - "import torch.optim as optim" + "# Create a federation\n", + "from openfl.interface.interactive_api.federation import Federation\n", + "\n", + "# please use the same identificator that was used in signed certificate\n", + "cliend_id = 'frontend'\n", + "\n", + "# 1) Run with API layer - Director mTLS \n", + "# If the user wants to enable mTLS their must provide CA root chain, and signed key pair to the federation interface\n", + "# cert_chain = 'cert/root_ca.crt'\n", + "# API_certificate = 'cert/frontend.crt'\n", + "# API_private_key = 'cert/frontend.key'\n", + "\n", + "# federation = Federation(client_id='frontend', director_node_fqdn='localhost', director_port='50051', disable_tls=False,\n", + "# cert_chain=cert_chain, api_cert=API_certificate, api_private_key=API_private_key)\n", + "\n", + "# --------------------------------------------------------------------------------------------------------------------\n", + "\n", + "# 2) Run with TLS disabled (trusted environment)\n", + "# Federation can also determine local fqdn automatically\n", + "federation = Federation(client_id='frontend', director_node_fqdn='localhost', director_port='50051', disable_tls=True)\n" ] }, { "cell_type": "code", "execution_count": null, - "id": "foreign-gospel", - "metadata": {}, + "id": "e35802d5", + "metadata": { + "scrolled": true + }, "outputs": [], "source": [ - "\"\"\"\n", - "UNet model definition\n", - "\"\"\"\n", - "from layers import soft_dice_coef, soft_dice_loss, DoubleConv, Down, Up\n", - "\n", - "class UNet(nn.Module):\n", - " def __init__(self, n_channels=3, n_classes=1):\n", - " super().__init__()\n", - " self.inc = DoubleConv(n_channels, 64)\n", - " self.down1 = Down(64, 128)\n", - " self.down2 = Down(128, 256)\n", - " self.down3 = Down(256, 512)\n", - " self.down4 = Down(512, 1024)\n", - " self.up1 = Up(1024, 512)\n", - " self.up2 = Up(512, 256)\n", - " self.up3 = Up(256, 128)\n", - " self.up4 = Up(128, 64)\n", - " self.outc = nn.Conv2d(64, n_classes, 1)\n", - "\n", - " def forward(self, x):\n", - " x1 = self.inc(x)\n", - " x2 = self.down1(x1)\n", - " x3 = self.down2(x2)\n", - " x4 = self.down3(x3)\n", - " x5 = self.down4(x4)\n", - " x = self.up1(x5, x4)\n", - " x = self.up2(x, x3)\n", - " x = self.up3(x, x2)\n", - " x = self.up4(x, x1)\n", - " x = self.outc(x)\n", - " x = torch.sigmoid(x)\n", - " return x\n", - " \n", - " \n", - "model_unet = UNet()" + "shard_registry = federation.get_shard_registry()\n", + "shard_registry" ] }, { "cell_type": "code", "execution_count": null, - "id": "greater-activation", + "id": "67ae50de", "metadata": {}, "outputs": [], "source": [ - "optimizer_adam = optim.Adam(model_unet.parameters(), lr=1e-4)" + "federation.target_shape" ] }, { - "cell_type": "markdown", - "id": "insured-metallic", - "metadata": {}, + "cell_type": "code", + "execution_count": null, + "id": "920216d3", + "metadata": { + "scrolled": true + }, + "outputs": [], "source": [ - "### Prepare data" + "# First, request a dummy_shard_desc that holds information about the federated dataset \n", + "dummy_shard_desc = federation.get_dummy_shard_descriptor(size=10)\n", + "sample, target = dummy_shard_desc[0]" ] }, { "cell_type": "markdown", - "id": "essential-remainder", + "id": "obvious-tyler", "metadata": {}, "source": [ - "We ask user to keep all the test data in `data/` folder under the workspace as it will not be sent to collaborators" + "## Creating a FL experiment using Interactive API" ] }, { "cell_type": "code", "execution_count": null, - "id": "blind-circuit", + "id": "rubber-address", "metadata": {}, "outputs": [], "source": [ - "import os\n", - "from hashlib import sha384\n", - "import PIL\n", - "from torch.utils.data import Dataset, DataLoader\n", - "from torchvision import transforms as tsf\n", - "from skimage import io" + "from openfl.interface.interactive_api.experiment import TaskInterface, DataInterface, ModelInterface, FLExperiment" ] }, { - "cell_type": "code", - "execution_count": null, - "id": "recorded-input", - "metadata": { - "scrolled": true - }, - "outputs": [], + "cell_type": "markdown", + "id": "sustainable-public", + "metadata": {}, "source": [ - "os.makedirs('data', exist_ok=True)\n", - "!wget -nc 'https://datasets.simula.no/hyper-kvasir/hyper-kvasir-segmented-images.zip' -O ./data/kvasir.zip\n", - "ZIP_SHA384 = 'e30d18a772c6520476e55b610a4db457237f151e'\\\n", - " '19182849d54b49ae24699881c1e18e0961f77642be900450ef8b22e7'\n", - "assert sha384(open('./data/kvasir.zip', 'rb').read(\n", - " os.path.getsize('./data/kvasir.zip'))).hexdigest() == ZIP_SHA384\n", - "!unzip -n ./data/kvasir.zip -d ./data" + "### Register dataset" + ] + }, + { + "cell_type": "markdown", + "id": "unlike-texas", + "metadata": {}, + "source": [ + "We extract User dataset class implementation.\n", + "Is it convinient?\n", + "What if the dataset is not a class?" ] }, { "cell_type": "code", "execution_count": null, - "id": "prepared-israel", - "metadata": {}, + "id": "64f37dcf", + "metadata": { + "scrolled": true + }, "outputs": [], "source": [ - "DATA_PATH = './data/segmented-images/'\n", + "import os\n", + "import PIL\n", "import numpy as np\n", + "from torch.utils.data import Dataset, DataLoader, SubsetRandomSampler\n", + "from torchvision import transforms as tsf\n", "\n", - "def read_data(image_path, mask_path):\n", - " \"\"\"\n", - " Read image and mask from disk.\n", - " \"\"\"\n", - " img = io.imread(image_path)\n", - " assert(img.shape[2] == 3)\n", - " mask = io.imread(mask_path)\n", - " return (img, mask[:, :, 0].astype(np.uint8))\n", - "\n", - "\n", - "class KvasirDataset(Dataset):\n", - " \"\"\"\n", - " Kvasir dataset contains 1000 images for all collaborators.\n", - " Args:\n", - " data_path: path to dataset on disk\n", - " collaborator_count: total number of collaborators\n", - " collaborator_num: number of current collaborator\n", - " is_validation: validation option\n", - " \"\"\"\n", - "\n", - " def __init__(self, images_path = './data/segmented-images/images/', \\\n", - " masks_path = './data/segmented-images/masks/',\n", - " validation_fraction=1/8, is_validation=False):\n", - "\n", - " self.images_path = images_path\n", - " self.masks_path = masks_path\n", - " self.images_names = [img_name for img_name in sorted(os.listdir(\n", - " self.images_path)) if len(img_name) > 3 and img_name[-3:] == 'jpg']\n", - "\n", - " assert(len(self.images_names) > 2), \"Too few images\"\n", - " \n", - " validation_size = max(1, int(len(self.images_names) * validation_fraction))\n", + "# Now you can implement you data loaders using dummy_shard_desc\n", + "class KvasirSD(DataInterface, Dataset):\n", + "\n", + " def __init__(self, validation_fraction=1/8, **kwargs):\n", + " super().__init__(**kwargs)\n", " \n", - " if is_validation:\n", - " self.images_names = self.images_names[-validation_size :]\n", - " else:\n", - " self.images_names = self.images_names[: -validation_size]\n", + " self.validation_fraction = validation_fraction\n", " \n", " # Prepare transforms\n", " self.img_trans = tsf.Compose([\n", @@ -205,239 +193,206 @@ " tsf.Resize((332, 332), interpolation=PIL.Image.NEAREST),\n", " tsf.ToTensor()])\n", " \n", + " @property\n", + " def shard_descriptor(self):\n", + " return self._shard_descriptor\n", + " \n", + " @shard_descriptor.setter\n", + " def shard_descriptor(self, shard_descriptor):\n", + " \"\"\"\n", + " Describe per-collaborator procedures or sharding.\n", + "\n", + " This method will be called during a collaborator initialization.\n", + " Local shard_descriptor will be set by Envoy.\n", + " \"\"\"\n", + " self._shard_descriptor = shard_descriptor\n", + " \n", + " validation_size = max(1, int(len(self.shard_descriptor) * self.validation_fraction))\n", + " \n", + " self.train_indeces = np.arange(len(self.shard_descriptor) - validation_size)\n", + " self.val_indeces = np.arange(len(self.shard_descriptor) - validation_size, len(self.shard_descriptor))\n", + " \n", "\n", " def __getitem__(self, index):\n", - " name = self.images_names[index]\n", - " img, mask = read_data(self.images_path + name, self.masks_path + name)\n", + " img, mask = self.shard_descriptor[index]\n", " img = self.img_trans(img).numpy()\n", " mask = self.mask_trans(mask).numpy()\n", " return img, mask\n", "\n", " def __len__(self):\n", - " return len(self.images_names)" - ] - }, - { - "cell_type": "markdown", - "id": "portuguese-groove", - "metadata": {}, - "source": [ - "### Define Federated Learning tasks" - ] - }, - { - "cell_type": "code", - "execution_count": null, - "id": "enclosed-abuse", - "metadata": {}, - "outputs": [], - "source": [ - "def train(unet_model, train_loader, optimizer, device, loss_fn=soft_dice_loss):\n", + " return len(self.shard_descriptor)\n", " \n", - " function_defined_in_notebook()\n", " \n", - " unet_model.train()\n", - " unet_model.to(device)\n", - "\n", - " losses = []\n", - "\n", - " for data, target in train_loader:\n", - " data, target = torch.tensor(data).to(device), torch.tensor(\n", - " target).to(device, dtype=torch.float32)\n", - " optimizer.zero_grad()\n", - " output = unet_model(data)\n", - " loss = loss_fn(output=output, target=target)\n", - " loss.backward()\n", - " optimizer.step()\n", - " losses.append(loss.detach().cpu().numpy())\n", - " \n", - " return {'train_loss': np.mean(losses),}\n", - "\n", + " def get_train_loader(self, **kwargs):\n", + " \"\"\"\n", + " Output of this method will be provided to tasks with optimizer in contract\n", + " \"\"\"\n", + " train_sampler = SubsetRandomSampler(self.train_indeces)\n", + " return DataLoader(\n", + " self, num_workers=8, batch_size=self.kwargs['train_bs'], sampler=train_sampler\n", + " )\n", "\n", - "def validate(unet_model, val_loader, device):\n", - " unet_model.eval()\n", - " unet_model.to(device)\n", + " def get_valid_loader(self, **kwargs):\n", + " \"\"\"\n", + " Output of this method will be provided to tasks without optimizer in contract\n", + " \"\"\"\n", + " val_sampler = SubsetRandomSampler(self.val_indeces)\n", + " return DataLoader(self, num_workers=8, batch_size=self.kwargs['valid_bs'], sampler=val_sampler)\n", "\n", - " val_score = 0\n", - " total_samples = 0\n", + " def get_train_data_size(self):\n", + " \"\"\"\n", + " Information for aggregation\n", + " \"\"\"\n", + " return len(self.train_indeces)\n", "\n", - " with torch.no_grad():\n", - " for data, target in val_loader:\n", - " samples = target.shape[0]\n", - " total_samples += samples\n", - " data, target = torch.tensor(data).to(device), \\\n", - " torch.tensor(target).to(device, dtype=torch.int64)\n", - " output = unet_model(data)\n", - " val = soft_dice_coef(output, target)\n", - " val_score += val.sum().cpu().numpy()\n", - " \n", - " return {'dice_coef': val_score / total_samples,}" + " def get_valid_data_size(self):\n", + " \"\"\"\n", + " Information for aggregation\n", + " \"\"\"\n", + " return len(self.val_indeces)" ] }, { "cell_type": "code", "execution_count": null, - "id": "secure-protest", + "id": "d8df35f5", "metadata": {}, "outputs": [], - "source": [] + "source": [ + "fed_dataset = KvasirSD(train_bs=4, valid_bs=8)\n", + "fed_dataset.shard_descriptor = dummy_shard_desc\n", + "for i, (sample, target) in enumerate(fed_dataset.get_train_loader()):\n", + " print(sample.shape)" + ] }, { "cell_type": "markdown", - "id": "obvious-tyler", + "id": "caring-distinction", "metadata": {}, "source": [ - "## Describing FL experiment" + "### Describe a model and optimizer" ] }, { "cell_type": "code", "execution_count": null, - "id": "rubber-address", + "id": "visible-victor", "metadata": {}, "outputs": [], "source": [ - "from openfl.interface.interactive_api.experiment import TaskInterface, DataInterface, ModelInterface, FLExperiment" - ] - }, - { - "cell_type": "markdown", - "id": "caroline-passion", - "metadata": {}, - "source": [ - "### Register model" + "import torch\n", + "import torch.nn as nn\n", + "import torch.optim as optim" ] }, { "cell_type": "code", "execution_count": null, - "id": "handled-teens", + "id": "foreign-gospel", "metadata": {}, "outputs": [], "source": [ - "from copy import deepcopy\n", + "\"\"\"\n", + "UNet model definition\n", + "\"\"\"\n", + "from layers import soft_dice_coef, soft_dice_loss, DoubleConv, Down, Up\n", "\n", - "framework_adapter = 'openfl.plugins.frameworks_adapters.pytorch_adapter.FrameworkAdapterPlugin'\n", - "MI = ModelInterface(model=model_unet, optimizer=optimizer_adam, framework_plugin=framework_adapter)\n", + "class UNet(nn.Module):\n", + " def __init__(self, n_channels=3, n_classes=1):\n", + " super().__init__()\n", + " self.inc = DoubleConv(n_channels, 64)\n", + " self.down1 = Down(64, 128)\n", + " self.down2 = Down(128, 256)\n", + " self.down3 = Down(256, 512)\n", + " self.down4 = Down(512, 1024)\n", + " self.up1 = Up(1024, 512)\n", + " self.up2 = Up(512, 256)\n", + " self.up3 = Up(256, 128)\n", + " self.up4 = Up(128, 64)\n", + " self.outc = nn.Conv2d(64, n_classes, 1)\n", "\n", - "# Save the initial model state\n", - "initial_model = deepcopy(model_unet)" + " def forward(self, x):\n", + " x1 = self.inc(x)\n", + " x2 = self.down1(x1)\n", + " x3 = self.down2(x2)\n", + " x4 = self.down3(x3)\n", + " x5 = self.down4(x4)\n", + " x = self.up1(x5, x4)\n", + " x = self.up2(x, x3)\n", + " x = self.up3(x, x2)\n", + " x = self.up4(x, x1)\n", + " x = self.outc(x)\n", + " x = torch.sigmoid(x)\n", + " return x\n", + " \n", + " \n", + "class UNet(nn.Module):\n", + " def __init__(self, n_channels=3, n_classes=1):\n", + " super().__init__()\n", + " self.inc = DoubleConv(n_channels, 64)\n", + " self.down1 = Down(64, 128)\n", + " self.down2 = Down(128, 256)\n", + " self.down3 = Down(256, 512)\n", + " self.up2 = Up(512, 256)\n", + " self.up3 = Up(256, 128)\n", + " self.up4 = Up(128, 64)\n", + " self.outc = nn.Conv2d(64, n_classes, 1)\n", + "\n", + " def forward(self, x):\n", + " x1 = self.inc(x)\n", + " x2 = self.down1(x1)\n", + " x3 = self.down2(x2)\n", + " x4 = self.down3(x3)\n", + " x = self.up2(x4, x3)\n", + " x = self.up3(x, x2)\n", + " x = self.up4(x, x1)\n", + " x = self.outc(x)\n", + " x = torch.sigmoid(x)\n", + " return x\n", + " \n", + "model_unet = UNet()" ] }, { - "cell_type": "markdown", - "id": "sustainable-public", + "cell_type": "code", + "execution_count": null, + "id": "greater-activation", "metadata": {}, + "outputs": [], "source": [ - "### Register dataset" + "optimizer_adam = optim.Adam(model_unet.parameters(), lr=1e-4)" ] }, { "cell_type": "markdown", - "id": "unlike-texas", + "id": "caroline-passion", "metadata": {}, "source": [ - "We extract User dataset class implementation.\n", - "Is it convinient?\n", - "What if the dataset is not a class?" + "#### Register model" ] }, { "cell_type": "code", "execution_count": null, - "id": "anticipated-coverage", - "metadata": {}, - "outputs": [], - "source": [ - "class UserDataset:\n", - " def __init__(self, path_to_local_data):\n", - " print(f'User Dataset initialized with {path_to_local_data}')\n", - " \n", - " \n", - "class OpenflMixin: \n", - " def _delayed_init(self):\n", - " raise NotImplementedError\n", - " \n", - " \n", - "class FedDataset(OpenflMixin):\n", - " def __init__(self, UserDataset):\n", - " self.user_dataset_class = UserDataset\n", - " print('We implement all abstract methods from mixin in this class')\n", - " \n", - " def _delayed_init(self, data_path):\n", - " print('This method is called on the collaborator node')\n", - " dataset_obj = self.user_dataset_class(data_path)\n", - " \n", - " \n", - "fed_dataset = FedDataset(UserDataset)\n", - "fed_dataset._delayed_init('data path on the collaborator node')" - ] - }, - { - "cell_type": "code", - "execution_count": null, - "id": "verbal-newton", + "id": "handled-teens", "metadata": {}, "outputs": [], "source": [ - "class FedDataset(DataInterface):\n", - " def __init__(self, UserDatasetClass, **kwargs):\n", - " self.UserDatasetClass = UserDatasetClass\n", - " self.kwargs = kwargs\n", - " \n", - " def _delayed_init(self, data_path='1,1'):\n", - " # With the next command the local dataset will be loaded on the collaborator node\n", - " # For this example we have the same dataset on the same path, and we will shard it\n", - " # So we use `data_path` information for this purpose.\n", - " self.rank, self.world_size = [int(part) for part in data_path.split(',')]\n", - " \n", - " validation_fraction=1/8\n", - " self.train_set = self.UserDatasetClass(validation_fraction=validation_fraction, is_validation=False)\n", - " self.valid_set = self.UserDatasetClass(validation_fraction=validation_fraction, is_validation=True)\n", - " \n", - " # Do the actual sharding\n", - " self._do_sharding( self.rank, self.world_size)\n", - " \n", - " def _do_sharding(self, rank, world_size):\n", - " # This method relies on the dataset's implementation\n", - " # i.e. coupled in a bad way\n", - " self.train_set.images_names = self.train_set.images_names[ rank-1 :: world_size ]\n", - "\n", - " def get_train_loader(self, **kwargs):\n", - " \"\"\"\n", - " Output of this method will be provided to tasks with optimizer in contract\n", - " \"\"\"\n", - " return DataLoader(\n", - " self.train_set, num_workers=8, batch_size=self.kwargs['train_bs'], shuffle=True\n", - " )\n", - "\n", - " def get_valid_loader(self, **kwargs):\n", - " \"\"\"\n", - " Output of this method will be provided to tasks without optimizer in contract\n", - " \"\"\"\n", - " return DataLoader(self.valid_set, num_workers=8, batch_size=self.kwargs['valid_bs'])\n", + "from copy import deepcopy\n", "\n", - " def get_train_data_size(self):\n", - " \"\"\"\n", - " Information for aggregation\n", - " \"\"\"\n", - " return len(self.train_set)\n", + "framework_adapter = 'openfl.plugins.frameworks_adapters.pytorch_adapter.FrameworkAdapterPlugin'\n", + "MI = ModelInterface(model=model_unet, optimizer=optimizer_adam, framework_plugin=framework_adapter)\n", "\n", - " def get_valid_data_size(self):\n", - " \"\"\"\n", - " Information for aggregation\n", - " \"\"\"\n", - " return len(self.valid_set)\n", - " \n", - "fed_dataset = FedDataset(KvasirDataset, train_bs=8, valid_bs=8)" + "# Save the initial model state\n", + "initial_model = deepcopy(model_unet)" ] }, { "cell_type": "markdown", - "id": "marine-kansas", + "id": "portuguese-groove", "metadata": {}, "source": [ - "### Register tasks" + "### Define and register FL tasks" ] }, { @@ -463,6 +418,8 @@ "def train(unet_model, train_loader, optimizer, device, loss_fn=soft_dice_loss, some_parameter=None):\n", " if not torch.cuda.is_available():\n", " device = 'cpu'\n", + " else:\n", + " device = 'cuda'\n", " \n", " function_defined_in_notebook(some_parameter)\n", " \n", @@ -488,6 +445,11 @@ "\n", "@TI.register_fl_task(model='unet_model', data_loader='val_loader', device='device') \n", "def validate(unet_model, val_loader, device):\n", + " if not torch.cuda.is_available():\n", + " device = 'cpu'\n", + " else:\n", + " device = 'cuda'\n", + " \n", " unet_model.eval()\n", " unet_model.to(device)\n", " \n", @@ -517,58 +479,6 @@ "## Time to start a federated learning experiment" ] }, - { - "cell_type": "code", - "execution_count": null, - "id": "silver-baltimore", - "metadata": {}, - "outputs": [], - "source": [ - "# Create a federation\n", - "from openfl.interface.interactive_api.federation import Federation\n", - "\n", - "# 1) Run with aggregator-collaborator mTLS \n", - "# If the user wants to enable mTLS their must provide CA root chain, and signed key pair to the federation interface\n", - "cert_chain = 'cert/cert_chain.crt'\n", - "agg_certificate = 'cert/agg_certificate.crt'\n", - "agg_private_key = 'cert/agg_private.key'\n", - "\n", - "federation = Federation(central_node_fqdn='some.fqdn', disable_tls=False,\n", - " cert_chain=cert_chain, agg_certificate=agg_certificate, agg_private_key=agg_private_key)\n", - "col_data_paths = {'one': '1,1',}\n", - "federation.register_collaborators(col_data_paths=col_data_paths)\n", - "\n", - "# --------------------------------------------------------------------------------------------------------------------\n", - "\n", - "# 2) Run with TLS disabled (trusted environment)\n", - "# Federation can also determine local fqdn automatically\n", - "federation = Federation(central_node_fqdn='localhost', disable_tls=True)\n", - "# First number which is a collaborators rank may also be passed as a cuda device identifier\n", - "col_data_paths = {'one': '1,2',\n", - " 'two': '2,2'}\n", - "federation.register_collaborators(col_data_paths=col_data_paths)" - ] - }, - { - "cell_type": "markdown", - "id": "adad7744", - "metadata": {}, - "source": [ - "#### Certification of an aggregator\n", - "* fx workspace certify: creates cert folder and CA as well as cert_chain\n", - "* fx aggregator generate-cert-request --fqdn `FQDN`: you can pass a specific aggregator FQDN if you want\n", - "* fx aggregator certify --fqdn `FQDN` --silent: signes aggregators cert\n", - " After that just pass the paths to required certs to the Federation API\n", - "\n", - "#### Certification of a collaborator\n", - "just follow the usual procedure: \n", - "fx collaborator generate-cert-request -d {DATA_PATH} -n {COL} \n", - "\n", - "fx collaborator certify --request-pkg {COL_DIRECTORY}/{FED_WORKSPACE}/col_{COL}_to_agg_cert_request.zip\n", - "\n", - "fx collaborator certify --import {FED_DIRECTORY}/agg_to_col_{COL}_signed_cert.zip" - ] - }, { "cell_type": "code", "execution_count": null, @@ -577,7 +487,8 @@ "outputs": [], "source": [ "# create an experimnet in federation\n", - "fl_experiment = FLExperiment(federation=federation,)" + "experiment_name = 'kvasir_test_experiment'\n", + "fl_experiment = FLExperiment(federation=federation, experiment_name=experiment_name)" ] }, { @@ -585,23 +496,31 @@ "execution_count": null, "id": "lightweight-causing", "metadata": { - "scrolled": false + "scrolled": true }, "outputs": [], "source": [ "# If I use autoreload I got a pickling error\n", "\n", "# The following command zips the workspace and python requirements to be transfered to collaborator nodes\n", - "fl_experiment.prepare_workspace_distribution(model_provider=MI, task_keeper=TI, data_loader=fed_dataset, rounds_to_train=7, \\\n", - " opt_treatment='CONTINUE_GLOBAL')\n", - "# This command starts the aggregator server\n", - "fl_experiment.start_experiment(model_provider=MI)\n", + "fl_experiment.start(model_provider=MI, \n", + " task_keeper=TI,\n", + " data_loader=fed_dataset,\n", + " rounds_to_train=2,\n", + " opt_treatment='CONTINUE_GLOBAL')\n" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "id": "f1543a36", + "metadata": {}, + "outputs": [], + "source": [ + "# If user want to stop IPython session, then reconnect and check how experiment is going \n", + "# fl_experiment.restore_experiment_state(MI)\n", "\n", - "# When the aggregator server blocks the notebook one can start collaborators\n", - "# For the test run just type console command from the workspace directory:\n", - "# `fx collaborator start -d data.yaml -n {col_name}` for all collaborators\n", - "# For the distributed experiment transfer zipped workspace to the collaborator nodes and run\n", - "# `fx workspace import --archive {workspace_name}.zip` cd to the workspace and start collaborators" + "fl_experiment.stream_metrics()" ] }, { @@ -616,7 +535,9 @@ "cell_type": "code", "execution_count": null, "id": "55acff59", - "metadata": {}, + "metadata": { + "scrolled": true + }, "outputs": [], "source": [ "best_model = fl_experiment.get_best_model()" @@ -625,11 +546,23 @@ { "cell_type": "code", "execution_count": null, - "id": "9fb5718d", + "id": "9479fb7f", "metadata": {}, "outputs": [], "source": [ - "fed_dataset._delayed_init()" + "# We remove exremove_experiment_datamove_experiment_datamove_experiment_datariment data from director\n", + "fl_experiment.remove_experiment_data()" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "id": "75c8aeab", + "metadata": {}, + "outputs": [], + "source": [ + "best_model.inc.conv[0].weight\n", + "# model_unet.inc.conv[0].weight" ] }, { @@ -685,6 +618,70 @@ "# Validating trained model\n", "validate(best_model, fed_dataset.get_valid_loader(), 'cpu')" ] + }, + { + "cell_type": "code", + "execution_count": null, + "id": "2dcdbfeb", + "metadata": {}, + "outputs": [], + "source": [ + "a = (np.zeros((2,4)), np.ones(2,), 2*np.ones(2,))\n", + "a = [elem for elem in a if elem.shape==2 else elem.newaxis()]\n", + "np.concatenate(a)" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "id": "70863215", + "metadata": {}, + "outputs": [], + "source": [ + "class A:\n", + " def __init__(self, **kwargs):\n", + " print(\"Class A\", kwargs)\n", + " \n", + "class B:\n", + " def __init__(self, **kwargs):\n", + " super().__init__()\n", + " print(\"Class B\", kwargs)\n", + " \n", + "class C(B, A):\n", + " def __init__(self, **kwargs):\n", + " super().__init__()\n", + " print(\"Class C\", kwargs)\n", + " \n", + "# class A:\n", + "# def __init__(self, **kwargs):\n", + "# print(\"Class A\", kwargs)" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "id": "4ba44919", + "metadata": {}, + "outputs": [], + "source": [ + "c = C(x=1, z=5)" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "id": "0637f26f", + "metadata": {}, + "outputs": [], + "source": [] + }, + { + "cell_type": "code", + "execution_count": null, + "id": "e00ff26c", + "metadata": {}, + "outputs": [], + "source": [] } ], "metadata": { @@ -708,4 +705,4 @@ }, "nbformat": 4, "nbformat_minor": 5 -} +} \ No newline at end of file diff --git a/openfl-tutorials/interactive_api/Director_Pytorch_Kvasir_UNET/workspace/layers.py b/openfl-tutorials/interactive_api/Director_Pytorch_Kvasir_UNET/workspace/layers.py new file mode 100644 index 0000000000..5165dcc97e --- /dev/null +++ b/openfl-tutorials/interactive_api/Director_Pytorch_Kvasir_UNET/workspace/layers.py @@ -0,0 +1,100 @@ +"""Layers for Unet model.""" + +import torch +import torch.nn as nn +import torch.nn.functional as F + + +def soft_dice_loss(output, target): + """Calculate loss.""" + num = target.size(0) + m1 = output.view(num, -1) + m2 = target.view(num, -1) + intersection = m1 * m2 + score = 2.0 * (intersection.sum(1) + 1) / (m1.sum(1) + m2.sum(1) + 1) + score = 1 - score.sum() / num + return score + + +def soft_dice_coef(output, target): + """Calculate soft DICE coefficient.""" + num = target.size(0) + m1 = output.view(num, -1) + m2 = target.view(num, -1) + intersection = m1 * m2 + score = 2.0 * (intersection.sum(1) + 1) / (m1.sum(1) + m2.sum(1) + 1) + return score.sum() + + +class DoubleConv(nn.Module): + """Pytorch double conv class.""" + + def __init__(self, in_ch, out_ch): + """Initialize layer.""" + super(DoubleConv, self).__init__() + self.in_ch = in_ch + self.out_ch = out_ch + self.conv = nn.Sequential( + nn.Conv2d(in_ch, out_ch, 3, padding=1), + nn.BatchNorm2d(out_ch), + nn.ReLU(inplace=True), + nn.Conv2d(out_ch, out_ch, 3, padding=1), + nn.BatchNorm2d(out_ch), + nn.ReLU(inplace=True), + ) + + def forward(self, x): + """Do forward pass.""" + x = self.conv(x) + return x + + +class Down(nn.Module): + """Pytorch nn module subclass.""" + + def __init__(self, in_ch, out_ch): + """Initialize layer.""" + super(Down, self).__init__() + self.mpconv = nn.Sequential( + nn.MaxPool2d(2), + DoubleConv(in_ch, out_ch) + ) + + def forward(self, x): + """Do forward pass.""" + x = self.mpconv(x) + return x + + +class Up(nn.Module): + """Pytorch nn module subclass.""" + + def __init__(self, in_ch, out_ch, bilinear=False): + """Initialize layer.""" + super(Up, self).__init__() + self.in_ch = in_ch + self.out_ch = out_ch + if bilinear: + self.up = nn.Upsample( + scale_factor=2, + mode='bilinear', + align_corners=True + ) + else: + self.up = nn.ConvTranspose2d(in_ch, in_ch // 2, 2, stride=2) + self.conv = DoubleConv(in_ch, out_ch) + + def forward(self, x1, x2): + """Do forward pass.""" + x1 = self.up(x1) + diff_y = x2.size()[2] - x1.size()[2] + diff_x = x2.size()[3] - x1.size()[3] + + x1 = F.pad( + x1, + (diff_x // 2, diff_x - diff_x // 2, diff_y // 2, diff_y - diff_y // 2) + ) + + x = torch.cat([x2, x1], dim=1) + x = self.conv(x) + return x diff --git a/openfl-tutorials/interactive_api/Director_Pytorch_Kvasir_UNET/workspace/new_python_api_UNET.ipynb b/openfl-tutorials/interactive_api/Director_Pytorch_Kvasir_UNET/workspace/new_python_api_UNET.ipynb new file mode 100644 index 0000000000..b72244349f --- /dev/null +++ b/openfl-tutorials/interactive_api/Director_Pytorch_Kvasir_UNET/workspace/new_python_api_UNET.ipynb @@ -0,0 +1,960 @@ +{ + "cells": [ + { + "cell_type": "markdown", + "id": "liquid-jacket", + "metadata": {}, + "source": [ + "# Federated PyTorch UNET Tutorial\n", + "## Using low-level Python API" + ] + }, + { + "cell_type": "code", + "execution_count": 1, + "id": "alike-sharing", + "metadata": { + "scrolled": true + }, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "Requirement already satisfied: torchvision in /home/davidyuk/.virtualenvs/openfl/lib/python3.8/site-packages (0.9.1)\n", + "Requirement already satisfied: torch==1.8.1 in /home/davidyuk/.virtualenvs/openfl/lib/python3.8/site-packages (from torchvision) (1.8.1)\n", + "Requirement already satisfied: numpy in /home/davidyuk/.virtualenvs/openfl/lib/python3.8/site-packages (from torchvision) (1.20.3)\n", + "Requirement already satisfied: pillow>=4.1.1 in /home/davidyuk/.virtualenvs/openfl/lib/python3.8/site-packages (from torchvision) (8.2.0)\n", + "Requirement already satisfied: typing-extensions in /home/davidyuk/.virtualenvs/openfl/lib/python3.8/site-packages (from torch==1.8.1->torchvision) (3.10.0.0)\n", + "\u001B[33mWARNING: You are using pip version 21.1; however, version 21.1.2 is available.\n", + "You should consider upgrading via the '/home/davidyuk/.virtualenvs/openfl/bin/python -m pip install --upgrade pip' command.\u001B[0m\n" + ] + } + ], + "source": [ + "# Install dependencies if not already installed\n", + "!pip install torchvision" + ] + }, + { + "cell_type": "markdown", + "id": "caring-distinction", + "metadata": {}, + "source": [ + "### Describe the model and optimizer" + ] + }, + { + "cell_type": "code", + "execution_count": 2, + "id": "visible-victor", + "metadata": {}, + "outputs": [], + "source": [ + "import torch\n", + "import torch.nn as nn\n", + "import torch.optim as optim" + ] + }, + { + "cell_type": "code", + "execution_count": 3, + "id": "foreign-gospel", + "metadata": {}, + "outputs": [], + "source": [ + "\"\"\"\n", + "UNet model definition\n", + "\"\"\"\n", + "from layers import soft_dice_coef, soft_dice_loss, DoubleConv, Down, Up\n", + "\n", + "class UNet(nn.Module):\n", + " def __init__(self, n_channels=3, n_classes=1):\n", + " super().__init__()\n", + " self.inc = DoubleConv(n_channels, 64)\n", + " self.down1 = Down(64, 128)\n", + " self.down2 = Down(128, 256)\n", + " self.down3 = Down(256, 512)\n", + " self.down4 = Down(512, 1024)\n", + " self.up1 = Up(1024, 512)\n", + " self.up2 = Up(512, 256)\n", + " self.up3 = Up(256, 128)\n", + " self.up4 = Up(128, 64)\n", + " self.outc = nn.Conv2d(64, n_classes, 1)\n", + "\n", + " def forward(self, x):\n", + " x1 = self.inc(x)\n", + " x2 = self.down1(x1)\n", + " x3 = self.down2(x2)\n", + " x4 = self.down3(x3)\n", + " x5 = self.down4(x4)\n", + " x = self.up1(x5, x4)\n", + " x = self.up2(x, x3)\n", + " x = self.up3(x, x2)\n", + " x = self.up4(x, x1)\n", + " x = self.outc(x)\n", + " x = torch.sigmoid(x)\n", + " return x\n", + " \n", + " \n", + "model_unet = UNet()" + ] + }, + { + "cell_type": "code", + "execution_count": 4, + "id": "24e5cf06", + "metadata": { + "scrolled": true + }, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "class UNet(nn.Module):\n", + " def __init__(self, n_channels=3, n_classes=1):\n", + " super().__init__()\n", + " self.inc = DoubleConv(n_channels, 64)\n", + " self.down1 = Down(64, 128)\n", + " self.down2 = Down(128, 256)\n", + " self.down3 = Down(256, 512)\n", + " self.down4 = Down(512, 1024)\n", + " self.up1 = Up(1024, 512)\n", + " self.up2 = Up(512, 256)\n", + " self.up3 = Up(256, 128)\n", + " self.up4 = Up(128, 64)\n", + " self.outc = nn.Conv2d(64, n_classes, 1)\n", + "\n", + " def forward(self, x):\n", + " x1 = self.inc(x)\n", + " x2 = self.down1(x1)\n", + " x3 = self.down2(x2)\n", + " x4 = self.down3(x3)\n", + " x5 = self.down4(x4)\n", + " x = self.up1(x5, x4)\n", + " x = self.up2(x, x3)\n", + " x = self.up3(x, x2)\n", + " x = self.up4(x, x1)\n", + " x = self.outc(x)\n", + " x = torch.sigmoid(x)\n", + " return x\n", + "\n" + ] + } + ], + "source": [ + "import inspect\n", + "from IPython.core.magics.code import extract_symbols\n", + "\n", + "import sys\n", + "\n", + "def new_getfile(object, _old_getfile=inspect.getfile):\n", + " if not inspect.isclass(object):\n", + " return _old_getfile(object)\n", + " \n", + " # Lookup by parent module (as in current inspect)\n", + " if hasattr(object, '__module__'):\n", + " object_ = sys.modules.get(object.__module__)\n", + " if hasattr(object_, '__file__'):\n", + " return object_.__file__\n", + " \n", + " # If parent module is __main__, lookup by methods (NEW)\n", + " for name, member in inspect.getmembers(object):\n", + " if inspect.isfunction(member) and object.__qualname__ + '.' + member.__name__ == member.__qualname__:\n", + " return inspect.getfile(member)\n", + " else:\n", + " raise TypeError('Source for {!r} not found'.format(object))\n", + "inspect.getfile = new_getfile\n", + "\n", + "obj = UNet\n", + "cell_code = \"\".join(inspect.linecache.getlines(new_getfile(obj)))\n", + "class_code = extract_symbols(cell_code, obj.__name__)[0][0]\n", + "print(class_code)" + ] + }, + { + "cell_type": "code", + "execution_count": 5, + "id": "greater-activation", + "metadata": {}, + "outputs": [], + "source": [ + "optimizer_adam = optim.Adam(model_unet.parameters(), lr=1e-4)" + ] + }, + { + "cell_type": "markdown", + "id": "obvious-tyler", + "metadata": {}, + "source": [ + "## Describing FL experiment" + ] + }, + { + "cell_type": "code", + "execution_count": 6, + "id": "rubber-address", + "metadata": {}, + "outputs": [], + "source": [ + "from openfl.interface.interactive_api.experiment import TaskInterface, DataInterface, ModelInterface, FLExperiment" + ] + }, + { + "cell_type": "markdown", + "id": "caroline-passion", + "metadata": {}, + "source": [ + "### Register model" + ] + }, + { + "cell_type": "code", + "execution_count": 7, + "id": "handled-teens", + "metadata": {}, + "outputs": [], + "source": [ + "from copy import deepcopy\n", + "\n", + "framework_adapter = 'openfl.plugins.frameworks_adapters.pytorch_adapter.FrameworkAdapterPlugin'\n", + "MI = ModelInterface(model=model_unet, optimizer=optimizer_adam, framework_plugin=framework_adapter)\n", + "\n", + "# Save the initial model state\n", + "initial_model = deepcopy(model_unet)" + ] + }, + { + "cell_type": "markdown", + "id": "portuguese-groove", + "metadata": {}, + "source": [ + "### Define Federated Learning tasks" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "id": "secure-protest", + "metadata": {}, + "outputs": [], + "source": [] + }, + { + "cell_type": "markdown", + "id": "sustainable-public", + "metadata": {}, + "source": [ + "### Register dataset" + ] + }, + { + "cell_type": "markdown", + "id": "unlike-texas", + "metadata": {}, + "source": [ + "We extract User dataset class implementation.\n", + "Is it convinient?\n", + "What if the dataset is not a class?" + ] + }, + { + "cell_type": "code", + "execution_count": 8, + "id": "verbal-newton", + "metadata": {}, + "outputs": [], + "source": [ + "# class FedDataset(DataInterface):\n", + "# def __init__(self, UserDatasetClass, **kwargs):\n", + "# self.UserDatasetClass = UserDatasetClass\n", + "# self.kwargs = kwargs\n", + " \n", + "# def _delayed_init(self, data_path='1,1'):\n", + "# # With the next command the local dataset will be loaded on the collaborator node\n", + "# # For this example we have the same dataset on the same path, and we will shard it\n", + "# # So we use `data_path` information for this purpose.\n", + "# self.rank, self.world_size = [int(part) for part in data_path.split(',')]\n", + " \n", + "# validation_fraction=1/8\n", + "# self.train_set = self.UserDatasetClass(validation_fraction=validation_fraction, is_validation=False)\n", + "# self.valid_set = self.UserDatasetClass(validation_fraction=validation_fraction, is_validation=True)\n", + " \n", + "# # Do the actual sharding\n", + "# self._do_sharding( self.rank, self.world_size)\n", + " \n", + "# def _do_sharding(self, rank, world_size):\n", + "# # This method relies on the dataset's implementation\n", + "# # i.e. coupled in a bad way\n", + "# self.train_set.images_names = self.train_set.images_names[ rank-1 :: world_size ]\n", + "\n", + "# def get_train_loader(self, **kwargs):\n", + "# \"\"\"\n", + "# Output of this method will be provided to tasks with optimizer in contract\n", + "# \"\"\"\n", + "# return DataLoader(\n", + "# self.train_set, num_workers=8, batch_size=self.kwargs['train_bs'], shuffle=True\n", + "# )\n", + "\n", + "# def get_valid_loader(self, **kwargs):\n", + "# \"\"\"\n", + "# Output of this method will be provided to tasks without optimizer in contract\n", + "# \"\"\"\n", + "# return DataLoader(self.valid_set, num_workers=8, batch_size=self.kwargs['valid_bs'])\n", + "\n", + "# def get_train_data_size(self):\n", + "# \"\"\"\n", + "# Information for aggregation\n", + "# \"\"\"\n", + "# return len(self.train_set)\n", + "\n", + "# def get_valid_data_size(self):\n", + "# \"\"\"\n", + "# Information for aggregation\n", + "# \"\"\"\n", + "# return len(self.valid_set)\n", + " \n", + "# fed_dataset = FedDataset(KvasirDataset, train_bs=8, valid_bs=8)" + ] + }, + { + "cell_type": "markdown", + "id": "marine-kansas", + "metadata": {}, + "source": [ + "### Register tasks" + ] + }, + { + "cell_type": "code", + "execution_count": 9, + "id": "increasing-builder", + "metadata": {}, + "outputs": [], + "source": [ + "TI = TaskInterface()\n", + "import torch\n", + "\n", + "import tqdm\n", + "\n", + "# The Interactive API supports registering functions definied in main module or imported.\n", + "def function_defined_in_notebook(some_parameter):\n", + " print(f'Also I accept a parameter and it is {some_parameter}')\n", + "\n", + "# Task interface currently supports only standalone functions.\n", + "@TI.add_kwargs(**{'some_parameter': 42})\n", + "@TI.register_fl_task(model='unet_model', data_loader='train_loader', \\\n", + " device='device', optimizer='optimizer') \n", + "def train(unet_model, train_loader, optimizer, device, loss_fn=soft_dice_loss, some_parameter=None):\n", + " if not torch.cuda.is_available():\n", + " device = 'cpu'\n", + " \n", + " function_defined_in_notebook(some_parameter)\n", + " \n", + " train_loader = tqdm.tqdm(train_loader, desc=\"train\")\n", + " \n", + " unet_model.train()\n", + " unet_model.to(device)\n", + "\n", + " losses = []\n", + "\n", + " for data, target in train_loader:\n", + " data, target = torch.tensor(data).to(device), torch.tensor(\n", + " target).to(device, dtype=torch.float32)\n", + " optimizer.zero_grad()\n", + " output = unet_model(data)\n", + " loss = loss_fn(output=output, target=target)\n", + " loss.backward()\n", + " optimizer.step()\n", + " losses.append(loss.detach().cpu().numpy())\n", + " \n", + " return {'train_loss': np.mean(losses),}\n", + "\n", + "\n", + "@TI.register_fl_task(model='unet_model', data_loader='val_loader', device='device') \n", + "def validate(unet_model, val_loader, device):\n", + " unet_model.eval()\n", + " unet_model.to(device)\n", + " \n", + " val_loader = tqdm.tqdm(val_loader, desc=\"validate\")\n", + "\n", + " val_score = 0\n", + " total_samples = 0\n", + "\n", + " with torch.no_grad():\n", + " for data, target in val_loader:\n", + " samples = target.shape[0]\n", + " total_samples += samples\n", + " data, target = torch.tensor(data).to(device), \\\n", + " torch.tensor(target).to(device, dtype=torch.int64)\n", + " output = unet_model(data)\n", + " val = soft_dice_coef(output, target)\n", + " val_score += val.sum().cpu().numpy()\n", + " \n", + " return {'dice_coef': val_score / total_samples,}" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "id": "b50a39c7", + "metadata": {}, + "outputs": [], + "source": [] + }, + { + "cell_type": "markdown", + "id": "derived-bride", + "metadata": {}, + "source": [ + "## Time to start a federated learning experiment" + ] + }, + { + "cell_type": "code", + "execution_count": 10, + "id": "4485ac79", + "metadata": {}, + "outputs": [], + "source": [ + "# Create a federation\n", + "from openfl.interface.interactive_api.federation import Federation\n", + "\n", + "# 1) Run with API layer - Director mTLS \n", + "# If the user wants to enable mTLS their must provide CA root chain, and signed key pair to the federation interface\n", + "cert_chain = 'cert/cert_chain.crt'\n", + "API_certificate = 'cert/API_certificate.crt'\n", + "API_private_key = 'cert/API_private.key'\n", + "\n", + "# federation = Federation(director_node_fqdn='some.fqdn', disable_tls=False,\n", + "# cert_chain=cert_chain, API_cert=API_certificate, API_private_key=API_private_key)\n", + "\n", + "# --------------------------------------------------------------------------------------------------------------------\n", + "\n", + "# 2) Run with TLS disabled (trusted environment)\n", + "# Federation can also determine local fqdn automatically\n", + "federation = Federation(director_node_fqdn='localhost:50051', disable_tls=True)" + ] + }, + { + "cell_type": "code", + "execution_count": 11, + "id": "e35802d5", + "metadata": {}, + "outputs": [ + { + "data": { + "text/plain": [ + "[node_info {\n", + " name: \"1,2\"\n", + "}\n", + "shard_description: \"Kvasir dataset, shard number 1 out of 2\"\n", + "n_samples: 500\n", + "sample_shape: \"529\"\n", + "sample_shape: \"622\"\n", + "sample_shape: \"3\"\n", + "target_shape: \"529\"\n", + "target_shape: \"622\"\n", + "]" + ] + }, + "execution_count": 11, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "shard_registry = federation.get_shard_registry()\n", + "shard_registry" + ] + }, + { + "cell_type": "code", + "execution_count": 12, + "id": "67ae50de", + "metadata": {}, + "outputs": [ + { + "data": { + "text/plain": [ + "['529', '622']" + ] + }, + "execution_count": 12, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "federation.target_shape" + ] + }, + { + "cell_type": "code", + "execution_count": 13, + "id": "920216d3", + "metadata": { + "scrolled": true + }, + "outputs": [], + "source": [ + "# First, request a dummy_shard_desc that holds information about the federated dataset \n", + "dummy_shard_desc = federation.get_dummy_shard_descriptor(size=10)\n", + "sample, target = dummy_shard_desc[0]" + ] + }, + { + "cell_type": "markdown", + "id": "insured-metallic", + "metadata": {}, + "source": [ + "### Prepare data" + ] + }, + { + "cell_type": "markdown", + "id": "essential-remainder", + "metadata": {}, + "source": [ + "We ask user to keep all the test data in `data/` folder under the workspace as it will not be sent to collaborators" + ] + }, + { + "cell_type": "code", + "execution_count": 14, + "id": "blind-circuit", + "metadata": {}, + "outputs": [], + "source": [ + "import os\n", + "import PIL\n", + "import numpy as np\n", + "from torch.utils.data import Dataset, DataLoader, SubsetRandomSampler\n", + "from torchvision import transforms as tsf" + ] + }, + { + "cell_type": "code", + "execution_count": 15, + "id": "64f37dcf", + "metadata": { + "scrolled": true + }, + "outputs": [], + "source": [ + "# Now you can implement you data loaders using dummy_shard_desc\n", + "class KvasirSD(DataInterface, Dataset):\n", + "\n", + " def __init__(self, validation_fraction=1/8, **kwargs):\n", + " super().__init__(**kwargs)\n", + " \n", + " self.validation_fraction = validation_fraction\n", + " \n", + " # Prepare transforms\n", + " self.img_trans = tsf.Compose([\n", + " tsf.ToPILImage(),\n", + " tsf.Resize((332, 332)),\n", + " tsf.ToTensor(),\n", + " tsf.Normalize(mean=[0.5, 0.5, 0.5], std=[0.5, 0.5, 0.5])])\n", + " self.mask_trans = tsf.Compose([\n", + " tsf.ToPILImage(),\n", + " tsf.Resize((332, 332), interpolation=PIL.Image.NEAREST),\n", + " tsf.ToTensor()])\n", + " \n", + " @property\n", + " def shard_descriptor(self):\n", + " return self._shard_descriptor\n", + " \n", + " @shard_descriptor.setter\n", + " def shard_descriptor(self, shard_descriptor):\n", + " \"\"\"\n", + " Describe per-collaborator procedures or sharding.\n", + "\n", + " This method will be called during a collaborator initialization.\n", + " Local shard_descriptor will be set by Envoy.\n", + " \"\"\"\n", + " self._shard_descriptor = shard_descriptor\n", + " \n", + " validation_size = max(1, int(len(self.shard_descriptor) * self.validation_fraction))\n", + " \n", + " self.train_indeces = np.arange(len(self.shard_descriptor) - validation_size)\n", + " self.val_indeces = np.arange(len(self.shard_descriptor) - validation_size, len(self.shard_descriptor))\n", + " \n", + "\n", + " def __getitem__(self, index):\n", + " img, mask = self.shard_descriptor[index]\n", + " img = self.img_trans(img).numpy()\n", + " mask = self.mask_trans(mask).numpy()\n", + " return img, mask\n", + "\n", + " def __len__(self):\n", + " return len(self.shard_descriptor)\n", + " \n", + " \n", + " def get_train_loader(self, **kwargs):\n", + " \"\"\"\n", + " Output of this method will be provided to tasks with optimizer in contract\n", + " \"\"\"\n", + " train_sampler = SubsetRandomSampler(self.train_indeces)\n", + " return DataLoader(\n", + " self, num_workers=8, batch_size=self.kwargs['train_bs'], sampler=train_sampler\n", + " )\n", + "\n", + " def get_valid_loader(self, **kwargs):\n", + " \"\"\"\n", + " Output of this method will be provided to tasks without optimizer in contract\n", + " \"\"\"\n", + " val_sampler = SubsetRandomSampler(self.val_indeces)\n", + " return DataLoader(self, num_workers=8, batch_size=self.kwargs['valid_bs'], sampler=val_sampler)\n", + "\n", + " def get_train_data_size(self):\n", + " \"\"\"\n", + " Information for aggregation\n", + " \"\"\"\n", + " return len(self.train_indeces)\n", + "\n", + " def get_valid_data_size(self):\n", + " \"\"\"\n", + " Information for aggregation\n", + " \"\"\"\n", + " return len(self.val_indeces)" + ] + }, + { + "cell_type": "code", + "execution_count": 16, + "id": "d8df35f5", + "metadata": {}, + "outputs": [ + { + "name": "stderr", + "output_type": "stream", + "text": [ + "/home/davidyuk/.virtualenvs/openfl/lib/python3.8/site-packages/torchvision/transforms/transforms.py:257: UserWarning: Argument interpolation should be of type InterpolationMode instead of int. Please, use InterpolationMode enum.\n", + " warnings.warn(\n" + ] + }, + { + "name": "stdout", + "output_type": "stream", + "text": [ + "torch.Size([8, 3, 332, 332])\n", + "torch.Size([1, 3, 332, 332])\n" + ] + } + ], + "source": [ + "fed_dataset = KvasirSD(train_bs=8, valid_bs=8)\n", + "fed_dataset.shard_descriptor = dummy_shard_desc\n", + "for i, (sample, target) in enumerate(fed_dataset.get_train_loader()):\n", + " print(sample.shape)" + ] + }, + { + "cell_type": "markdown", + "id": "af0579f8", + "metadata": {}, + "source": [ + "# Long-Living entities update\n", + "\n", + "* We now may have director running on another machine.\n", + "* We use Federation API to communicate with Director.\n", + "* Federation object should hold a Director's client (for user service)\n", + "* Keeping in mind that several API instances may be connacted to one Director.\n", + "\n", + "\n", + "* We do not think for now how we start a Director.\n", + "* But it knows the data shape and target shape for the DataScience problem in the Federation.\n", + "* Director holds the list of connected envoys, we do not need to specify it anymore.\n", + "* Director and Envoys are responsible for encrypting connections, we do not need to worry about certs.\n", + "\n", + "\n", + "* Yet we MUST have a cert to communicate to the Director.\n", + "* We MUST know the FQDN of a Director.\n", + "* Director communicates data and target shape to the Federation interface object.\n", + "\n", + "\n", + "* Experiment API may use this info to construct a dummy dataset and a `shard descriptor` stub." + ] + }, + { + "cell_type": "markdown", + "id": "adad7744", + "metadata": {}, + "source": [ + "#### Certification of an aggregator\n", + "* fx workspace certify: creates cert folder and CA as well as cert_chain\n", + "* fx aggregator generate-cert-request --fqdn `FQDN`: you can pass a specific aggregator FQDN if you want\n", + "* fx aggregator certify --fqdn `FQDN` --silent: signes aggregators cert\n", + " After that just pass the paths to required certs to the Federation API\n", + "\n", + "#### Certification of a collaborator\n", + "just follow the usual procedure: \n", + "fx collaborator generate-cert-request -d {DATA_PATH} -n {COL} \n", + "\n", + "fx collaborator certify --request-pkg {COL_DIRECTORY}/{FED_WORKSPACE}/col_{COL}_to_agg_cert_request.zip\n", + "\n", + "fx collaborator certify --import {FED_DIRECTORY}/agg_to_col_{COL}_signed_cert.zip" + ] + }, + { + "cell_type": "code", + "execution_count": 17, + "id": "12f5663f", + "metadata": {}, + "outputs": [ + { + "data": { + "text/plain": [ + "2" + ] + }, + "execution_count": 17, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "# Interface generator\n", + "class IFO:\n", + " def __init__(self):\n", + " self.a = 1\n", + " \n", + "class EXP:\n", + " def __init__(self):\n", + " self.ifo = IFO()\n", + " \n", + " def get_ifo(self):\n", + " return self.ifo\n", + " \n", + "exp = EXP()\n", + "ifo = exp.get_ifo()\n", + "ifo.a = 2\n", + "\n", + "# print(exp.ifo.a)\n", + "exp2 = exp.__class__()\n", + "exp.ifo.a" + ] + }, + { + "cell_type": "code", + "execution_count": 18, + "id": "mature-renewal", + "metadata": {}, + "outputs": [], + "source": [ + "# create an experimnet in federation\n", + "fl_experiment = FLExperiment(federation=federation,)" + ] + }, + { + "cell_type": "code", + "execution_count": 19, + "id": "lightweight-causing", + "metadata": { + "scrolled": true + }, + "outputs": [ + { + "name": "stderr", + "output_type": "stream", + "text": [ + "tried to remove tensor: __opt_state_needed not present in the tensor dict\n", + "tried to remove tensor: __opt_state_needed not present in the tensor dict\n", + "INFO:openfl.transport.grpc.director_client:SetNewExperiment\n", + "WARNING:openfl.transport.grpc.server:gRPC is running on insecure channel with TLS disabled.\n", + "INFO:openfl.transport.grpc.server:Starting Aggregator gRPC Server\n", + "INFO:openfl.component.aggregator.aggregator:Sending tasks to collaborator 1,2 for round 0\n" + ] + } + ], + "source": [ + "# If I use autoreload I got a pickling error\n", + "\n", + "# The following command zips the workspace and python requirements to be transfered to collaborator nodes\n", + "fl_experiment.prepare_workspace_distribution(model_provider=MI, \n", + " task_keeper=TI, data_loader=fed_dataset, rounds_to_train=7, opt_treatment='CONTINUE_GLOBAL')\n", + "\n", + "\n", + "# This command starts the aggregator server\n", + "fl_experiment.start_experiment(model_provider=MI)\n" + ] + }, + { + "cell_type": "markdown", + "id": "8c30b301", + "metadata": {}, + "source": [ + "## Now we validate the best model!" + ] + }, + { + "cell_type": "code", + "execution_count": 20, + "id": "55acff59", + "metadata": {}, + "outputs": [ + { + "ename": "KeyError", + "evalue": "'inc.conv.0.weight'", + "output_type": "error", + "traceback": [ + "\u001B[0;31m---------------------------------------------------------------------------\u001B[0m", + "\u001B[0;31mKeyError\u001B[0m Traceback (most recent call last)", + "\u001B[0;32m<ipython-input-20-a539666cced0>\u001B[0m in \u001B[0;36m<module>\u001B[0;34m\u001B[0m\n\u001B[0;32m----> 1\u001B[0;31m \u001B[0mbest_model\u001B[0m \u001B[0;34m=\u001B[0m \u001B[0mfl_experiment\u001B[0m\u001B[0;34m.\u001B[0m\u001B[0mget_best_model\u001B[0m\u001B[0;34m(\u001B[0m\u001B[0;34m)\u001B[0m\u001B[0;34m\u001B[0m\u001B[0;34m\u001B[0m\u001B[0m\n\u001B[0m", + "\u001B[0;32m~/openfl/openfl/interface/interactive_api/experiment.py\u001B[0m in \u001B[0;36mget_best_model\u001B[0;34m(self)\u001B[0m\n\u001B[1;32m 41\u001B[0m \u001B[0;31m# Next line relies on aggregator inner field where model dicts are stored\u001B[0m\u001B[0;34m\u001B[0m\u001B[0;34m\u001B[0m\u001B[0;34m\u001B[0m\u001B[0m\n\u001B[1;32m 42\u001B[0m \u001B[0mbest_tensor_dict\u001B[0m \u001B[0;34m=\u001B[0m \u001B[0mself\u001B[0m\u001B[0;34m.\u001B[0m\u001B[0mserver\u001B[0m\u001B[0;34m.\u001B[0m\u001B[0maggregator\u001B[0m\u001B[0;34m.\u001B[0m\u001B[0mbest_tensor_dict\u001B[0m\u001B[0;34m\u001B[0m\u001B[0;34m\u001B[0m\u001B[0m\n\u001B[0;32m---> 43\u001B[0;31m \u001B[0mself\u001B[0m\u001B[0;34m.\u001B[0m\u001B[0mtask_runner_stub\u001B[0m\u001B[0;34m.\u001B[0m\u001B[0mrebuild_model\u001B[0m\u001B[0;34m(\u001B[0m\u001B[0mbest_tensor_dict\u001B[0m\u001B[0;34m,\u001B[0m \u001B[0mvalidation\u001B[0m\u001B[0;34m=\u001B[0m\u001B[0;32mTrue\u001B[0m\u001B[0;34m,\u001B[0m \u001B[0mdevice\u001B[0m\u001B[0;34m=\u001B[0m\u001B[0;34m'cpu'\u001B[0m\u001B[0;34m)\u001B[0m\u001B[0;34m\u001B[0m\u001B[0;34m\u001B[0m\u001B[0m\n\u001B[0m\u001B[1;32m 44\u001B[0m \u001B[0;32mreturn\u001B[0m \u001B[0mself\u001B[0m\u001B[0;34m.\u001B[0m\u001B[0mtask_runner_stub\u001B[0m\u001B[0;34m.\u001B[0m\u001B[0mmodel\u001B[0m\u001B[0;34m\u001B[0m\u001B[0;34m\u001B[0m\u001B[0m\n\u001B[1;32m 45\u001B[0m \u001B[0;34m\u001B[0m\u001B[0m\n", + "\u001B[0;32m~/openfl/openfl/federated/task/task_runner.py\u001B[0m in \u001B[0;36mrebuild_model\u001B[0;34m(self, input_tensor_dict, validation, device)\u001B[0m\n\u001B[1;32m 220\u001B[0m \u001B[0;32mif\u001B[0m \u001B[0mself\u001B[0m\u001B[0;34m.\u001B[0m\u001B[0mopt_treatment\u001B[0m \u001B[0;34m==\u001B[0m \u001B[0;34m'RESET'\u001B[0m\u001B[0;34m:\u001B[0m\u001B[0;34m\u001B[0m\u001B[0;34m\u001B[0m\u001B[0m\n\u001B[1;32m 221\u001B[0m \u001B[0mself\u001B[0m\u001B[0;34m.\u001B[0m\u001B[0mreset_opt_vars\u001B[0m\u001B[0;34m(\u001B[0m\u001B[0;34m)\u001B[0m\u001B[0;34m\u001B[0m\u001B[0;34m\u001B[0m\u001B[0m\n\u001B[0;32m--> 222\u001B[0;31m \u001B[0mself\u001B[0m\u001B[0;34m.\u001B[0m\u001B[0mset_tensor_dict\u001B[0m\u001B[0;34m(\u001B[0m\u001B[0minput_tensor_dict\u001B[0m\u001B[0;34m,\u001B[0m \u001B[0mwith_opt_vars\u001B[0m\u001B[0;34m=\u001B[0m\u001B[0;32mFalse\u001B[0m\u001B[0;34m,\u001B[0m \u001B[0mdevice\u001B[0m\u001B[0;34m=\u001B[0m\u001B[0mdevice\u001B[0m\u001B[0;34m)\u001B[0m\u001B[0;34m\u001B[0m\u001B[0;34m\u001B[0m\u001B[0m\n\u001B[0m\u001B[1;32m 223\u001B[0m elif (self.training_round_completed\n\u001B[1;32m 224\u001B[0m and self.opt_treatment == 'CONTINUE_GLOBAL' and not validation):\n", + "\u001B[0;32m~/openfl/openfl/federated/task/task_runner.py\u001B[0m in \u001B[0;36mset_tensor_dict\u001B[0;34m(self, tensor_dict, with_opt_vars, device)\u001B[0m\n\u001B[1;32m 373\u001B[0m \u001B[0mkwargs\u001B[0m \u001B[0;34m=\u001B[0m \u001B[0;34m{\u001B[0m\u001B[0;34m'device'\u001B[0m\u001B[0;34m:\u001B[0m \u001B[0mdevice\u001B[0m\u001B[0;34m,\u001B[0m \u001B[0;34m}\u001B[0m\u001B[0;34m\u001B[0m\u001B[0;34m\u001B[0m\u001B[0m\n\u001B[1;32m 374\u001B[0m \u001B[0;34m\u001B[0m\u001B[0m\n\u001B[0;32m--> 375\u001B[0;31m \u001B[0;32mreturn\u001B[0m \u001B[0mself\u001B[0m\u001B[0;34m.\u001B[0m\u001B[0mframework_adapter\u001B[0m\u001B[0;34m.\u001B[0m\u001B[0mset_tensor_dict\u001B[0m\u001B[0;34m(\u001B[0m\u001B[0;34m*\u001B[0m\u001B[0margs\u001B[0m\u001B[0;34m,\u001B[0m \u001B[0;34m**\u001B[0m\u001B[0mkwargs\u001B[0m\u001B[0;34m)\u001B[0m\u001B[0;34m\u001B[0m\u001B[0;34m\u001B[0m\u001B[0m\n\u001B[0m", + "\u001B[0;32m~/openfl/openfl/plugins/frameworks_adapters/pytorch_adapter.py\u001B[0m in \u001B[0;36mset_tensor_dict\u001B[0;34m(model, tensor_dict, optimizer, device)\u001B[0m\n\u001B[1;32m 43\u001B[0m \u001B[0;31m# everything\u001B[0m\u001B[0;34m\u001B[0m\u001B[0;34m\u001B[0m\u001B[0;34m\u001B[0m\u001B[0m\n\u001B[1;32m 44\u001B[0m \u001B[0;32mfor\u001B[0m \u001B[0mk\u001B[0m \u001B[0;32min\u001B[0m \u001B[0mmodel\u001B[0m\u001B[0;34m.\u001B[0m\u001B[0mstate_dict\u001B[0m\u001B[0;34m(\u001B[0m\u001B[0;34m)\u001B[0m\u001B[0;34m:\u001B[0m\u001B[0;34m\u001B[0m\u001B[0;34m\u001B[0m\u001B[0m\n\u001B[0;32m---> 45\u001B[0;31m \u001B[0mnew_state\u001B[0m\u001B[0;34m[\u001B[0m\u001B[0mk\u001B[0m\u001B[0;34m]\u001B[0m \u001B[0;34m=\u001B[0m \u001B[0mpt\u001B[0m\u001B[0;34m.\u001B[0m\u001B[0mfrom_numpy\u001B[0m\u001B[0;34m(\u001B[0m\u001B[0mtensor_dict\u001B[0m\u001B[0;34m.\u001B[0m\u001B[0mpop\u001B[0m\u001B[0;34m(\u001B[0m\u001B[0mk\u001B[0m\u001B[0;34m)\u001B[0m\u001B[0;34m)\u001B[0m\u001B[0;34m.\u001B[0m\u001B[0mto\u001B[0m\u001B[0;34m(\u001B[0m\u001B[0mdevice\u001B[0m\u001B[0;34m)\u001B[0m\u001B[0;34m\u001B[0m\u001B[0;34m\u001B[0m\u001B[0m\n\u001B[0m\u001B[1;32m 46\u001B[0m \u001B[0;34m\u001B[0m\u001B[0m\n\u001B[1;32m 47\u001B[0m \u001B[0;31m# set model state\u001B[0m\u001B[0;34m\u001B[0m\u001B[0;34m\u001B[0m\u001B[0;34m\u001B[0m\u001B[0m\n", + "\u001B[0;31mKeyError\u001B[0m: 'inc.conv.0.weight'" + ] + } + ], + "source": [ + "best_model = fl_experiment.get_best_model()" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "id": "9fb5718d", + "metadata": {}, + "outputs": [], + "source": [ + "fed_dataset._delayed_init()" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "id": "a2acb7e6", + "metadata": {}, + "outputs": [], + "source": [ + "# Validating initial model\n", + "validate(initial_model, fed_dataset.get_valid_loader(), 'cpu')" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "id": "c12ca93f", + "metadata": {}, + "outputs": [], + "source": [ + "# Validating trained model\n", + "validate(best_model, fed_dataset.get_valid_loader(), 'cpu')" + ] + }, + { + "cell_type": "markdown", + "id": "1e6734f6", + "metadata": {}, + "source": [ + "## We can tune model further!" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "id": "3940e75e", + "metadata": {}, + "outputs": [], + "source": [ + "MI = ModelInterface(model=best_model, optimizer=optimizer_adam, framework_plugin=framework_adapter)\n", + "fl_experiment.start_experiment(model_provider=MI, task_keeper=TI, data_loader=fed_dataset, rounds_to_train=4, \\\n", + " opt_treatment='CONTINUE_GLOBAL')" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "id": "1bd786d2", + "metadata": {}, + "outputs": [], + "source": [ + "best_model = fl_experiment.get_best_model()\n", + "# Validating trained model\n", + "validate(best_model, fed_dataset.get_valid_loader(), 'cpu')" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "id": "2dcdbfeb", + "metadata": {}, + "outputs": [], + "source": [ + "a = (np.zeros((2,4)), np.ones(2,), 2*np.ones(2,))\n", + "a = [elem for elem in a if elem.shape==2 else elem.newaxis()]\n", + "np.concatenate(a)" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "id": "70863215", + "metadata": {}, + "outputs": [], + "source": [ + "class A:\n", + " def __init__(self, **kwargs):\n", + " print(\"Class A\", kwargs)\n", + " \n", + "class B:\n", + " def __init__(self, **kwargs):\n", + " super().__init__()\n", + " print(\"Class B\", kwargs)\n", + " \n", + "class C(B, A):\n", + " def __init__(self, **kwargs):\n", + " super().__init__()\n", + " print(\"Class C\", kwargs)\n", + " \n", + "# class A:\n", + "# def __init__(self, **kwargs):\n", + "# print(\"Class A\", kwargs)" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "id": "4ba44919", + "metadata": {}, + "outputs": [], + "source": [ + "c = C(x=1, z=5)" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "id": "0637f26f", + "metadata": {}, + "outputs": [], + "source": [] + } + ], + "metadata": { + "kernelspec": { + "display_name": "Python 3", + "language": "python", + "name": "python3" + }, + "language_info": { + "codemirror_mode": { + "name": "ipython", + "version": 3 + }, + "file_extension": ".py", + "mimetype": "text/x-python", + "name": "python", + "nbconvert_exporter": "python", + "pygments_lexer": "ipython3", + "version": "3.8.5" + } + }, + "nbformat": 4, + "nbformat_minor": 5 +} \ No newline at end of file diff --git a/openfl-tutorials/interactive_api_tutorials_(experimental)/Pytorch_Kvasir_UNET_workspace/layers.py b/openfl-tutorials/interactive_api/Pytorch_Kvasir_UNET/layers.py similarity index 90% rename from openfl-tutorials/interactive_api_tutorials_(experimental)/Pytorch_Kvasir_UNET_workspace/layers.py rename to openfl-tutorials/interactive_api/Pytorch_Kvasir_UNET/layers.py index 75e1e6902e..bf876c1cb3 100644 --- a/openfl-tutorials/interactive_api_tutorials_(experimental)/Pytorch_Kvasir_UNET_workspace/layers.py +++ b/openfl-tutorials/interactive_api/Pytorch_Kvasir_UNET/layers.py @@ -56,7 +56,9 @@ def __init__(self, in_ch, out_ch): """Initialize layer.""" super(Down, self).__init__() self.mpconv = nn.Sequential( - nn.MaxPool2d(2), DoubleConv(in_ch, out_ch)) + nn.MaxPool2d(2), + DoubleConv(in_ch, out_ch) + ) def forward(self, x): """Do forward pass.""" @@ -74,11 +76,12 @@ def __init__(self, in_ch, out_ch, bilinear=False): self.out_ch = out_ch if bilinear: self.up = nn.Upsample( - scale_factor=2, mode='bilinear', align_corners=True) - else: - self.up = nn.ConvTranspose2d( - in_ch, in_ch // 2, 2, stride=2 + scale_factor=2, + mode='bilinear', + align_corners=True ) + else: + self.up = nn.ConvTranspose2d(in_ch, in_ch // 2, 2, stride=2) self.conv = DoubleConv(in_ch, out_ch) def forward(self, x1, x2): diff --git a/openfl-tutorials/interactive_api/Pytorch_Kvasir_UNET/new_python_api_UNET.ipynb b/openfl-tutorials/interactive_api/Pytorch_Kvasir_UNET/new_python_api_UNET.ipynb new file mode 100644 index 0000000000..e236cbea56 --- /dev/null +++ b/openfl-tutorials/interactive_api/Pytorch_Kvasir_UNET/new_python_api_UNET.ipynb @@ -0,0 +1,1159 @@ +{ + "cells": [ + { + "cell_type": "markdown", + "id": "liquid-jacket", + "metadata": {}, + "source": [ + "# Federated PyTorch UNET Tutorial\n", + "## Using low-level Python API" + ] + }, + { + "cell_type": "code", + "execution_count": 1, + "id": "alike-sharing", + "metadata": { + "scrolled": true + }, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "Requirement already satisfied: torchvision in /home/davidyuk/.virtualenvs/openfl/lib/python3.8/site-packages (0.9.1)\n", + "Requirement already satisfied: numpy in /home/davidyuk/.virtualenvs/openfl/lib/python3.8/site-packages (from torchvision) (1.20.3)\n", + "Requirement already satisfied: torch==1.8.1 in /home/davidyuk/.virtualenvs/openfl/lib/python3.8/site-packages (from torchvision) (1.8.1)\n", + "Requirement already satisfied: pillow>=4.1.1 in /home/davidyuk/.virtualenvs/openfl/lib/python3.8/site-packages (from torchvision) (8.2.0)\n", + "Requirement already satisfied: typing-extensions in /home/davidyuk/.virtualenvs/openfl/lib/python3.8/site-packages (from torch==1.8.1->torchvision) (3.10.0.0)\n", + "\u001B[33mWARNING: You are using pip version 21.1; however, version 21.1.2 is available.\n", + "You should consider upgrading via the '/home/davidyuk/.virtualenvs/openfl/bin/python -m pip install --upgrade pip' command.\u001B[0m\n", + "Requirement already satisfied: scikit-image in /home/davidyuk/.virtualenvs/openfl/lib/python3.8/site-packages (0.18.1)\n", + "Requirement already satisfied: matplotlib!=3.0.0,>=2.0.0 in /home/davidyuk/.virtualenvs/openfl/lib/python3.8/site-packages (from scikit-image) (3.4.2)\n", + "Requirement already satisfied: numpy>=1.16.5 in /home/davidyuk/.virtualenvs/openfl/lib/python3.8/site-packages (from scikit-image) (1.20.3)\n", + "Requirement already satisfied: scipy>=1.0.1 in /home/davidyuk/.virtualenvs/openfl/lib/python3.8/site-packages (from scikit-image) (1.6.3)\n", + "Requirement already satisfied: pillow!=7.1.0,!=7.1.1,>=4.3.0 in /home/davidyuk/.virtualenvs/openfl/lib/python3.8/site-packages (from scikit-image) (8.2.0)\n", + "Requirement already satisfied: networkx>=2.0 in /home/davidyuk/.virtualenvs/openfl/lib/python3.8/site-packages (from scikit-image) (2.5.1)\n", + "Requirement already satisfied: PyWavelets>=1.1.1 in /home/davidyuk/.virtualenvs/openfl/lib/python3.8/site-packages (from scikit-image) (1.1.1)\n", + "Requirement already satisfied: tifffile>=2019.7.26 in /home/davidyuk/.virtualenvs/openfl/lib/python3.8/site-packages (from scikit-image) (2021.4.8)\n", + "Requirement already satisfied: imageio>=2.3.0 in /home/davidyuk/.virtualenvs/openfl/lib/python3.8/site-packages (from scikit-image) (2.9.0)\n", + "Requirement already satisfied: cycler>=0.10 in /home/davidyuk/.virtualenvs/openfl/lib/python3.8/site-packages (from matplotlib!=3.0.0,>=2.0.0->scikit-image) (0.10.0)\n", + "Requirement already satisfied: pyparsing>=2.2.1 in /home/davidyuk/.virtualenvs/openfl/lib/python3.8/site-packages (from matplotlib!=3.0.0,>=2.0.0->scikit-image) (2.4.7)\n", + "Requirement already satisfied: python-dateutil>=2.7 in /home/davidyuk/.virtualenvs/openfl/lib/python3.8/site-packages (from matplotlib!=3.0.0,>=2.0.0->scikit-image) (2.8.1)\n", + "Requirement already satisfied: kiwisolver>=1.0.1 in /home/davidyuk/.virtualenvs/openfl/lib/python3.8/site-packages (from matplotlib!=3.0.0,>=2.0.0->scikit-image) (1.3.1)\n", + "Requirement already satisfied: six in /home/davidyuk/.virtualenvs/openfl/lib/python3.8/site-packages (from cycler>=0.10->matplotlib!=3.0.0,>=2.0.0->scikit-image) (1.16.0)\n", + "Requirement already satisfied: decorator<5,>=4.3 in /home/davidyuk/.virtualenvs/openfl/lib/python3.8/site-packages (from networkx>=2.0->scikit-image) (4.4.2)\n", + "\u001B[33mWARNING: You are using pip version 21.1; however, version 21.1.2 is available.\n", + "You should consider upgrading via the '/home/davidyuk/.virtualenvs/openfl/bin/python -m pip install --upgrade pip' command.\u001B[0m\n" + ] + } + ], + "source": [ + "# Install dependencies if not already installed\n", + "!pip install torchvision==0.8.1\n", + "!pip install scikit-image" + ] + }, + { + "cell_type": "markdown", + "id": "caring-distinction", + "metadata": {}, + "source": [ + "### Describe the model and optimizer" + ] + }, + { + "cell_type": "code", + "execution_count": 11, + "id": "visible-victor", + "metadata": {}, + "outputs": [], + "source": [ + "import torch\n", + "import torch.nn as nn\n", + "import torch.optim as optim" + ] + }, + { + "cell_type": "code", + "execution_count": 12, + "id": "foreign-gospel", + "metadata": {}, + "outputs": [], + "source": [ + "\"\"\"\n", + "UNet model definition\n", + "\"\"\"\n", + "from layers import soft_dice_coef, soft_dice_loss, DoubleConv, Down, Up\n", + "\n", + "class UNet(nn.Module):\n", + " def __init__(self, n_channels=3, n_classes=1):\n", + " super().__init__()\n", + " self.inc = DoubleConv(n_channels, 64)\n", + " self.down1 = Down(64, 128)\n", + " self.down2 = Down(128, 256)\n", + " self.down3 = Down(256, 512)\n", + " self.down4 = Down(512, 1024)\n", + " self.up1 = Up(1024, 512)\n", + " self.up2 = Up(512, 256)\n", + " self.up3 = Up(256, 128)\n", + " self.up4 = Up(128, 64)\n", + " self.outc = nn.Conv2d(64, n_classes, 1)\n", + "\n", + " def forward(self, x):\n", + " x1 = self.inc(x)\n", + " x2 = self.down1(x1)\n", + " x3 = self.down2(x2)\n", + " x4 = self.down3(x3)\n", + " x5 = self.down4(x4)\n", + " x = self.up1(x5, x4)\n", + " x = self.up2(x, x3)\n", + " x = self.up3(x, x2)\n", + " x = self.up4(x, x1)\n", + " x = self.outc(x)\n", + " x = torch.sigmoid(x)\n", + " return x\n", + " \n", + " # Lookup by parent module (as in current inspect)\n", + " if hasattr(object, '__module__'):\n", + " object_ = sys.modules.get(object.__module__)\n", + " if hasattr(object_, '__file__'):\n", + " return object_.__file__\n", + " \n", + " # If parent module is __main__, lookup by methods (NEW)\n", + " for name, member in inspect.getmembers(object):\n", + " if inspect.isfunction(member) and object.__qualname__ + '.' + member.__name__ == member.__qualname__:\n", + " return inspect.getfile(member)\n", + " else:\n", + " raise TypeError('Source for {!r} not found'.format(object))\n", + "inspect.getfile = new_getfile\n", + "\n", + "obj = UNet\n", + "cell_code = \"\".join(inspect.linecache.getlines(new_getfile(obj)))\n", + "class_code = extract_symbols(cell_code, obj.__name__)[0][0]\n", + "print(class_code)" + ] + }, + { + "cell_type": "code", + "execution_count": 14, + "id": "greater-activation", + "metadata": {}, + "outputs": [], + "source": [ + "optimizer_adam = optim.Adam(model_unet.parameters(), lr=1e-4)" + ] + }, + { + "cell_type": "markdown", + "id": "insured-metallic", + "metadata": {}, + "source": [ + "### Prepare data" + ] + }, + { + "cell_type": "markdown", + "id": "essential-remainder", + "metadata": {}, + "source": [ + "We ask user to keep all the test data in `data/` folder under the workspace as it will not be sent to collaborators" + ] + }, + { + "cell_type": "code", + "execution_count": 15, + "id": "blind-circuit", + "metadata": {}, + "outputs": [], + "source": [ + "import os\n", + "from hashlib import sha384\n", + "import PIL\n", + "from PIL import Image\n", + "from torch.utils.data import Dataset, DataLoader\n", + "from torchvision import transforms as tsf" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "id": "recorded-input", + "metadata": { + "scrolled": true + }, + "outputs": [], + "source": [ + "os.makedirs('data', exist_ok=True)\n", + "!wget -nc 'https://datasets.simula.no/hyper-kvasir/hyper-kvasir-segmented-images.zip' -O ./data/kvasir.zip\n", + "ZIP_SHA384 = 'e30d18a772c6520476e55b610a4db457237f151e'\\\n", + " '19182849d54b49ae24699881c1e18e0961f77642be900450ef8b22e7'\n", + "assert sha384(open('./data/kvasir.zip', 'rb').read(\n", + " os.path.getsize('./data/kvasir.zip'))).hexdigest() == ZIP_SHA384\n", + "!unzip -n ./data/kvasir.zip -d ./data" + ] + }, + { + "cell_type": "code", + "execution_count": 10, + "id": "5fa108f5", + "metadata": {}, + "outputs": [ + { + "data": { + "text/plain": [ + "<generator object <genexpr> at 0x7f616dce3f90>" + ] + }, + "execution_count": 10, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "rank_worldsize='3,4'\n", + "(int(num) for num in rank_worldsize.split(','))" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "id": "fec3c6d4", + "metadata": {}, + "outputs": [], + "source": [] + }, + { + "cell_type": "code", + "execution_count": 40, + "id": "prepared-israel", + "metadata": {}, + "outputs": [], + "source": [ + "DATA_PATH = './data/segmented-images/'\n", + "import numpy as np\n", + "\n", + "def read_data(image_path, mask_path):\n", + " \"\"\"\n", + " Read image and mask from disk.\n", + " \"\"\"\n", + " img = Image.open(image_path)\n", + " img = np.asarray(img)\n", + " assert(img.shape[2] == 3)\n", + " mask = Image.open(mask_path)\n", + " mask = np.asarray(mask)\n", + " return (img, mask[:, :, 0].astype(np.uint8))\n", + "\n", + "\n", + "class KvasirDataset(Dataset):\n", + " \"\"\"\n", + " Kvasir dataset contains 1000 images for all collaborators.\n", + " Args:\n", + " data_path: path to dataset on disk\n", + " collaborator_count: total number of collaborators\n", + " collaborator_num: number of current collaborator\n", + " is_validation: validation option\n", + " \"\"\"\n", + "\n", + " def __init__(self, images_path = './kvasir_data/segmented-images/images/', \\\n", + " masks_path = './kvasir_data/segmented-images/masks/',\n", + " validation_fraction=1/8, is_validation=False):\n", + "\n", + " self.images_path = images_path\n", + " self.masks_path = masks_path\n", + " self.images_names = [\n", + " img_name\n", + " for img_name in sorted(os.listdir(self.images_path))\n", + " if len(img_name) > 3 and img_name[-3:] == 'jpg'\n", + " ]\n", + "\n", + " assert(len(self.images_names) > 2), \"Too few images\"\n", + " \n", + " validation_size = max(1, int(len(self.images_names) * validation_fraction))\n", + " \n", + " if is_validation:\n", + " self.images_names = self.images_names[-validation_size :]\n", + " else:\n", + " self.images_names = self.images_names[: -validation_size]\n", + "\n", + " # Prepare transforms\n", + " self.img_trans = tsf.Compose([\n", + " tsf.ToPILImage(),\n", + " tsf.Resize((332, 332)),\n", + " tsf.ToTensor(),\n", + " tsf.Normalize(mean=[0.5, 0.5, 0.5], std=[0.5, 0.5, 0.5])])\n", + " self.mask_trans = tsf.Compose([\n", + " tsf.ToPILImage(),\n", + " tsf.Resize((332, 332), interpolation=PIL.Image.NEAREST),\n", + " tsf.ToTensor()])\n", + " \n", + "\n", + " def __getitem__(self, index):\n", + " name = self.images_names[index]\n", + " img, mask = read_data(self.images_path + name, self.masks_path + name)\n", + " img = self.img_trans(img).numpy()\n", + " mask = self.mask_trans(mask).numpy()\n", + " return img, mask\n", + "\n", + " def __len__(self):\n", + " return len(self.images_names)" + ] + }, + { + "cell_type": "code", + "execution_count": 41, + "id": "c17c0944", + "metadata": {}, + "outputs": [], + "source": [ + "img = Image.open('./kvasir_data/segmented-images/images/cju0qkwl35piu0993l0dewei2.jpg')\n", + "img.resize()" + ] + }, + { + "cell_type": "code", + "execution_count": 42, + "id": "9c889655", + "metadata": {}, + "outputs": [ + { + "name": "stderr", + "output_type": "stream", + "text": [ + "/home/davidyuk/.virtualenvs/openfl/lib/python3.8/site-packages/torchvision/transforms/transforms.py:257: UserWarning: Argument interpolation should be of type InterpolationMode instead of int. Please, use InterpolationMode enum.\n", + " warnings.warn(\n" + ] + }, + { + "data": { + "text/plain": [ + "array([[[-1., -1., -1., ..., -1., -1., -1.],\n", + " [-1., -1., -1., ..., -1., -1., -1.],\n", + " [-1., -1., -1., ..., -1., -1., -1.],\n", + " ...,\n", + " [-1., -1., -1., ..., -1., -1., -1.],\n", + " [-1., -1., -1., ..., -1., -1., -1.],\n", + " [-1., -1., -1., ..., -1., -1., -1.]],\n", + "\n", + " [[-1., -1., -1., ..., -1., -1., -1.],\n", + " [-1., -1., -1., ..., -1., -1., -1.],\n", + " [-1., -1., -1., ..., -1., -1., -1.],\n", + " ...,\n", + " [-1., -1., -1., ..., -1., -1., -1.],\n", + " [-1., -1., -1., ..., -1., -1., -1.],\n", + " [-1., -1., -1., ..., -1., -1., -1.]],\n", + "\n", + " [[-1., -1., -1., ..., -1., -1., -1.],\n", + " [-1., -1., -1., ..., -1., -1., -1.],\n", + " [-1., -1., -1., ..., -1., -1., -1.],\n", + " ...,\n", + " [-1., -1., -1., ..., -1., -1., -1.],\n", + " [-1., -1., -1., ..., -1., -1., -1.],\n", + " [-1., -1., -1., ..., -1., -1., -1.]]], dtype=float32)" + ] + }, + "execution_count": 42, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "dset = KvasirDataset()\n", + "dset[0][0]" + ] + }, + { + "cell_type": "markdown", + "id": "portuguese-groove", + "metadata": {}, + "source": [ + "### Define Federated Learning tasks" + ] + }, + { + "cell_type": "code", + "execution_count": 7, + "id": "enclosed-abuse", + "metadata": {}, + "outputs": [], + "source": [ + "def train(unet_model, train_loader, optimizer, device, loss_fn=soft_dice_loss):\n", + " \n", + " function_defined_in_notebook()\n", + " \n", + " unet_model.train()\n", + " unet_model.to(device)\n", + "\n", + " losses = []\n", + "\n", + " for data, target in train_loader:\n", + " data, target = torch.tensor(data).to(device), torch.tensor(\n", + " target).to(device, dtype=torch.float32)\n", + " optimizer.zero_grad()\n", + " output = unet_model(data)\n", + " loss = loss_fn(output=output, target=target)\n", + " loss.backward()\n", + " optimizer.step()\n", + " losses.append(loss.detach().cpu().numpy())\n", + " \n", + " return {'train_loss': np.mean(losses),}\n", + "\n", + "\n", + "def validate(unet_model, val_loader, device):\n", + " unet_model.eval()\n", + " unet_model.to(device)\n", + "\n", + " val_score = 0\n", + " total_samples = 0\n", + "\n", + " with torch.no_grad():\n", + " for data, target in val_loader:\n", + " samples = target.shape[0]\n", + " total_samples += samples\n", + " data, target = torch.tensor(data).to(device), \\\n", + " torch.tensor(target).to(device, dtype=torch.int64)\n", + " output = unet_model(data)\n", + " val = soft_dice_coef(output, target)\n", + " val_score += val.sum().cpu().numpy()\n", + " \n", + " return {'dice_coef': val_score / total_samples,}" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "id": "secure-protest", + "metadata": {}, + "outputs": [], + "source": [] + }, + { + "cell_type": "markdown", + "id": "obvious-tyler", + "metadata": {}, + "source": [ + "## Describing FL experiment" + ] + }, + { + "cell_type": "code", + "execution_count": 1, + "id": "rubber-address", + "metadata": {}, + "outputs": [], + "source": [ + "from openfl.interface.interactive_api.experiment import TaskInterface, DataInterface, ModelInterface, FLExperiment" + ] + }, + { + "cell_type": "code", + "execution_count": 2, + "id": "fd0a70ed", + "metadata": {}, + "outputs": [], + "source": [ + "from torch.utils.data import Dataset, DataLoader\n", + "class A(Dataset, DataInterface):\n", + " pass\n", + "a = A()\n", + "# dir(a)" + ] + }, + { + "cell_type": "markdown", + "id": "caroline-passion", + "metadata": {}, + "source": [ + "### Register model" + ] + }, + { + "cell_type": "code", + "execution_count": 9, + "id": "handled-teens", + "metadata": {}, + "outputs": [], + "source": [ + "from copy import deepcopy\n", + "\n", + "framework_adapter = 'openfl.plugins.frameworks_adapters.pytorch_adapter.FrameworkAdapterPlugin'\n", + "MI = ModelInterface(model=model_unet, optimizer=optimizer_adam, framework_plugin=framework_adapter)\n", + "\n", + "# Save the initial model state\n", + "initial_model = deepcopy(model_unet)" + ] + }, + { + "cell_type": "markdown", + "id": "sustainable-public", + "metadata": {}, + "source": [ + "### Register dataset" + ] + }, + { + "cell_type": "markdown", + "id": "unlike-texas", + "metadata": {}, + "source": [ + "We extract User dataset class implementation.\n", + "Is it convinient?\n", + "What if the dataset is not a class?" + ] + }, + { + "cell_type": "code", + "execution_count": 10, + "id": "anticipated-coverage", + "metadata": {}, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "We implement all abstract methods from mixin in this class\n", + "This method is called on the collaborator node\n", + "User Dataset initialized with data path on the collaborator node\n" + ] + } + ], + "source": [ + "class UserDataset:\n", + " def __init__(self, path_to_local_data):\n", + " print(f'User Dataset initialized with {path_to_local_data}')\n", + " \n", + " \n", + "class OpenflMixin: \n", + " def _delayed_init(self):\n", + " raise NotImplementedError\n", + " \n", + " \n", + "class FedDataset(OpenflMixin):\n", + " def __init__(self, UserDataset):\n", + " self.user_dataset_class = UserDataset\n", + " print('We implement all abstract methods from mixin in this class')\n", + " \n", + " def _delayed_init(self, data_path):\n", + " print('This method is called on the collaborator node')\n", + " dataset_obj = self.user_dataset_class(data_path)\n", + " \n", + " \n", + "fed_dataset = FedDataset(UserDataset)\n", + "fed_dataset._delayed_init('data path on the collaborator node')" + ] + }, + { + "cell_type": "code", + "execution_count": 11, + "id": "verbal-newton", + "metadata": {}, + "outputs": [], + "source": [ + "# class FedDataset(DataInterface):\n", + "# def __init__(self, UserDatasetClass, **kwargs):\n", + "# self.UserDatasetClass = UserDatasetClass\n", + "# self.kwargs = kwargs\n", + " \n", + "# def _delayed_init(self, data_path='1,1'):\n", + "# # With the next command the local dataset will be loaded on the collaborator node\n", + "# # For this example we have the same dataset on the same path, and we will shard it\n", + "# # So we use `data_path` information for this purpose.\n", + "# self.rank, self.world_size = [int(part) for part in data_path.split(',')]\n", + " \n", + "# validation_fraction=1/8\n", + "# self.train_set = self.UserDatasetClass(validation_fraction=validation_fraction, is_validation=False)\n", + "# self.valid_set = self.UserDatasetClass(validation_fraction=validation_fraction, is_validation=True)\n", + " \n", + "# # Do the actual sharding\n", + "# self._do_sharding( self.rank, self.world_size)\n", + " \n", + "# def _do_sharding(self, rank, world_size):\n", + "# # This method relies on the dataset's implementation\n", + "# # i.e. coupled in a bad way\n", + "# self.train_set.images_names = self.train_set.images_names[ rank-1 :: world_size ]\n", + "\n", + "# def get_train_loader(self, **kwargs):\n", + "# \"\"\"\n", + "# Output of this method will be provided to tasks with optimizer in contract\n", + "# \"\"\"\n", + "# return DataLoader(\n", + "# self.train_set, num_workers=8, batch_size=self.kwargs['train_bs'], shuffle=True\n", + "# )\n", + "\n", + "# def get_valid_loader(self, **kwargs):\n", + "# \"\"\"\n", + "# Output of this method will be provided to tasks without optimizer in contract\n", + "# \"\"\"\n", + "# return DataLoader(self.valid_set, num_workers=8, batch_size=self.kwargs['valid_bs'])\n", + "\n", + "# def get_train_data_size(self):\n", + "# \"\"\"\n", + "# Information for aggregation\n", + "# \"\"\"\n", + "# return len(self.train_set)\n", + "\n", + "# def get_valid_data_size(self):\n", + "# \"\"\"\n", + "# Information for aggregation\n", + "# \"\"\"\n", + "# return len(self.valid_set)\n", + " \n", + "# fed_dataset = FedDataset(KvasirDataset, train_bs=8, valid_bs=8)" + ] + }, + { + "cell_type": "markdown", + "id": "marine-kansas", + "metadata": {}, + "source": [ + "### Register tasks" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "id": "increasing-builder", + "metadata": {}, + "outputs": [], + "source": [ + "TI = TaskInterface()\n", + "import torch\n", + "\n", + "import tqdm\n", + "\n", + "# The Interactive API supports registering functions definied in main module or imported.\n", + "def function_defined_in_notebook(some_parameter):\n", + " print(f'Also I accept a parameter and it is {some_parameter}')\n", + "\n", + "# Task interface currently supports only standalone functions.\n", + "@TI.add_kwargs(**{'some_parameter': 42})\n", + "@TI.register_fl_task(model='unet_model', data_loader='train_loader', \\\n", + " device='device', optimizer='optimizer') \n", + "def train(unet_model, train_loader, optimizer, device, loss_fn=soft_dice_loss, some_parameter=None):\n", + " if not torch.cuda.is_available():\n", + " device = 'cpu'\n", + " \n", + " function_defined_in_notebook(some_parameter)\n", + " \n", + " train_loader = tqdm.tqdm(train_loader, desc=\"train\")\n", + " \n", + " unet_model.train()\n", + " unet_model.to(device)\n", + "\n", + " losses = []\n", + "\n", + " for data, target in train_loader:\n", + " data, target = torch.tensor(data).to(device), torch.tensor(\n", + " target).to(device, dtype=torch.float32)\n", + " optimizer.zero_grad()\n", + " output = unet_model(data)\n", + " loss = loss_fn(output=output, target=target)\n", + " loss.backward()\n", + " optimizer.step()\n", + " losses.append(loss.detach().cpu().numpy())\n", + " \n", + " return {'train_loss': np.mean(losses),}\n", + "\n", + "\n", + "@TI.register_fl_task(model='unet_model', data_loader='val_loader', device='device') \n", + "def validate(unet_model, val_loader, device):\n", + " unet_model.eval()\n", + " unet_model.to(device)\n", + " \n", + " val_loader = tqdm.tqdm(val_loader, desc=\"validate\")\n", + "\n", + " val_score = 0\n", + " total_samples = 0\n", + "\n", + " with torch.no_grad():\n", + " for data, target in val_loader:\n", + " samples = target.shape[0]\n", + " total_samples += samples\n", + " data, target = torch.tensor(data).to(device), \\\n", + " torch.tensor(target).to(device, dtype=torch.int64)\n", + " output = unet_model(data)\n", + " val = soft_dice_coef(output, target)\n", + " val_score += val.sum().cpu().numpy()\n", + " \n", + " return {'dice_coef': val_score / total_samples,}" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "id": "b50a39c7", + "metadata": {}, + "outputs": [], + "source": [] + }, + { + "cell_type": "markdown", + "id": "derived-bride", + "metadata": {}, + "source": [ + "## Time to start a federated learning experiment" + ] + }, + { + "cell_type": "code", + "execution_count": 1, + "id": "4485ac79", + "metadata": {}, + "outputs": [], + "source": [ + "# Create a federation\n", + "from openfl.interface.interactive_api.federation import Federation\n", + "\n", + "# 1) Run with API layer - Director mTLS \n", + "# If the user wants to enable mTLS their must provide CA root chain, and signed key pair to the federation interface\n", + "cert_chain = 'cert/cert_chain.crt'\n", + "API_certificate = 'cert/API_certificate.crt'\n", + "API_private_key = 'cert/API_private.key'\n", + "\n", + "federation = Federation(director_node_fqdn='some.fqdn', disable_tls=False,\n", + " cert_chain=cert_chain, API_cert=API_certificate, API_private_key=API_private_key)\n", + "\n", + "# --------------------------------------------------------------------------------------------------------------------\n", + "\n", + "# 2) Run with TLS disabled (trusted environment)\n", + "# Federation can also determine local fqdn automatically\n", + "federation = Federation(director_node_fqdn='localhost', disable_tls=True)" + ] + }, + { + "cell_type": "code", + "execution_count": 2, + "id": "920216d3", + "metadata": { + "scrolled": true + }, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "(300, 300, 3) (6,) 10\n" + ] + } + ], + "source": [ + "# First, request a dummy_shard_desc that holds information about the federated dataset \n", + "dummy_shard_desc = federation.get_dummy_shard_descriptor(size=10)" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "id": "996f6460", + "metadata": {}, + "outputs": [], + "source": [ + "\"1,6\"" + ] + }, + { + "cell_type": "code", + "execution_count": 17, + "id": "64f37dcf", + "metadata": { + "scrolled": true + }, + "outputs": [ + { + "ename": "NameError", + "evalue": "name 'DataInterface' is not defined", + "output_type": "error", + "traceback": [ + "\u001B[0;31m---------------------------------------------------------------------------\u001B[0m", + "\u001B[0;31mNameError\u001B[0m Traceback (most recent call last)", + "\u001B[0;32m<ipython-input-17-858ae44ef56e>\u001B[0m in \u001B[0;36m<module>\u001B[0;34m\u001B[0m\n\u001B[1;32m 1\u001B[0m \u001B[0;31m# Now you can implement you data loaders using dummy_shard_desc\u001B[0m\u001B[0;34m\u001B[0m\u001B[0;34m\u001B[0m\u001B[0;34m\u001B[0m\u001B[0m\n\u001B[0;32m----> 2\u001B[0;31m \u001B[0;32mclass\u001B[0m \u001B[0mKvasirSD\u001B[0m\u001B[0;34m(\u001B[0m\u001B[0mDataInterface\u001B[0m\u001B[0;34m,\u001B[0m \u001B[0mDataset\u001B[0m\u001B[0;34m)\u001B[0m\u001B[0;34m:\u001B[0m\u001B[0;34m\u001B[0m\u001B[0;34m\u001B[0m\u001B[0m\n\u001B[0m\u001B[1;32m 3\u001B[0m \u001B[0;34m\u001B[0m\u001B[0m\n\u001B[1;32m 4\u001B[0m \u001B[0;32mdef\u001B[0m \u001B[0m__init__\u001B[0m\u001B[0;34m(\u001B[0m\u001B[0mself\u001B[0m\u001B[0;34m,\u001B[0m \u001B[0mshard_descriptor\u001B[0m\u001B[0;34m,\u001B[0m \u001B[0mvalidation_fraction\u001B[0m\u001B[0;34m=\u001B[0m\u001B[0;36m1\u001B[0m\u001B[0;34m/\u001B[0m\u001B[0;36m8\u001B[0m\u001B[0;34m,\u001B[0m \u001B[0mis_validation\u001B[0m\u001B[0;34m=\u001B[0m\u001B[0;32mFalse\u001B[0m\u001B[0;34m,\u001B[0m \u001B[0;34m**\u001B[0m\u001B[0mkwargs\u001B[0m\u001B[0;34m)\u001B[0m\u001B[0;34m:\u001B[0m\u001B[0;34m\u001B[0m\u001B[0;34m\u001B[0m\u001B[0m\n\u001B[1;32m 5\u001B[0m \u001B[0msuper\u001B[0m\u001B[0;34m(\u001B[0m\u001B[0;34m)\u001B[0m\u001B[0;34m.\u001B[0m\u001B[0m__init__\u001B[0m\u001B[0;34m(\u001B[0m\u001B[0;34m**\u001B[0m\u001B[0mkwargs\u001B[0m\u001B[0;34m)\u001B[0m\u001B[0;34m\u001B[0m\u001B[0;34m\u001B[0m\u001B[0m\n", + "\u001B[0;31mNameError\u001B[0m: name 'DataInterface' is not defined" + ] + } + ], + "source": [ + "# Now you can implement you data loaders using dummy_shard_desc\n", + "class KvasirSD(DataInterface, Dataset):\n", + "\n", + " def __init__(self, shard_descriptor, validation_fraction=1/8, is_validation=False, **kwargs):\n", + " super().__init__(**kwargs)\n", + "\n", + " self.shard_descriptor = shard_descriptor\n", + " \n", + " self.validation_fraction = validation_fraction\n", + " \n", + " validation_size = max(1, int(len(self.shard_descriptor) * validation_fraction))\n", + " self.train_indeces = list(range())\n", + " \n", + "# if is_validation:\n", + "# self.images_names = self.images_names[-validation_size :]\n", + "# else:\n", + "# self.images_names = self.images_names[: -validation_size]\n", + " \n", + " # Prepare transforms\n", + " self.img_trans = tsf.Compose([\n", + " tsf.ToPILImage(),\n", + " tsf.Resize((332, 332)),\n", + " tsf.ToTensor(),\n", + " tsf.Normalize(mean=[0.5, 0.5, 0.5], std=[0.5, 0.5, 0.5])])\n", + " self.mask_trans = tsf.Compose([\n", + " tsf.ToPILImage(),\n", + " tsf.Resize((332, 332), interpolation=PIL.Image.NEAREST),\n", + " tsf.ToTensor()])\n", + " \n", + "\n", + " def __getitem__(self, index):\n", + " img, mask = self.shard_descriptor[index]\n", + " img = self.img_trans(img).numpy()\n", + " mask = self.mask_trans(mask).numpy()\n", + " return img, mask\n", + "\n", + " def __len__(self):\n", + " return len(self.shard_descriptor)\n", + " \n", + " \n", + " def get_train_loader(self, **kwargs):\n", + " \"\"\"\n", + " Output of this method will be provided to tasks with optimizer in contract\n", + " \"\"\"\n", + " return DataLoader(\n", + " self.train_set, num_workers=8, batch_size=self.kwargs['train_bs'], shuffle=True\n", + " )\n", + "\n", + " def get_valid_loader(self, **kwargs):\n", + " \"\"\"\n", + " Output of this method will be provided to tasks without optimizer in contract\n", + " \"\"\"\n", + " return DataLoader(self.valid_set, num_workers=8, batch_size=self.kwargs['valid_bs'])\n", + "\n", + " def get_train_data_size(self):\n", + " \"\"\"\n", + " Information for aggregation\n", + " \"\"\"\n", + " return len(self.train_set)\n", + "\n", + " def get_valid_data_size(self):\n", + " \"\"\"\n", + " Information for aggregation\n", + " \"\"\"\n", + " return len(self.valid_set)" + ] + }, + { + "cell_type": "code", + "execution_count": 22, + "id": "d8df35f5", + "metadata": {}, + "outputs": [ + { + "ename": "SyntaxError", + "evalue": "invalid syntax (<ipython-input-22-9778031892f4>, line 1)", + "output_type": "error", + "traceback": [ + "\u001B[0;36m File \u001B[0;32m\"<ipython-input-22-9778031892f4>\"\u001B[0;36m, line \u001B[0;32m1\u001B[0m\n\u001B[0;31m [1:10]\u001B[0m\n\u001B[0m ^\u001B[0m\n\u001B[0;31mSyntaxError\u001B[0m\u001B[0;31m:\u001B[0m invalid syntax\n" + ] + } + ], + "source": [] + }, + { + "cell_type": "markdown", + "id": "af0579f8", + "metadata": {}, + "source": [ + "# Long-Living entities update\n", + "\n", + "* We now may have director running on another machine.\n", + "* We use Federation API to communicate with Director.\n", + "* Federation object should hold a Director's client (for user service)\n", + "* Keeping in mind that several API instances may be connacted to one Director.\n", + "\n", + "\n", + "* We do not think for now how we start a Director.\n", + "* But it knows the data shape and target shape for the DataScience problem in the Federation.\n", + "* Director holds the list of connected envoys, we do not need to specify it anymore.\n", + "* Director and Envoys are responsible for encrypting connections, we do not need to worry about certs.\n", + "\n", + "\n", + "* Yet we MUST have a cert to communicate to the Director.\n", + "* We MUST know the FQDN of a Director.\n", + "* Director communicates data and target shape to the Federation interface object.\n", + "\n", + "\n", + "* Experiment API may use this info to construct a dummy dataset and a `shard descriptor` stub." + ] + }, + { + "cell_type": "markdown", + "id": "adad7744", + "metadata": {}, + "source": [ + "#### Certification of an aggregator\n", + "* fx workspace certify: creates cert folder and CA as well as cert_chain\n", + "* fx aggregator generate-cert-request --fqdn `FQDN`: you can pass a specific aggregator FQDN if you want\n", + "* fx aggregator certify --fqdn `FQDN` --silent: signes aggregators cert\n", + " After that just pass the paths to required certs to the Federation API\n", + "\n", + "#### Certification of a collaborator\n", + "just follow the usual procedure: \n", + "fx collaborator generate-cert-request -d {DATA_PATH} -n {COL} \n", + "\n", + "fx collaborator certify --request-pkg {COL_DIRECTORY}/{FED_WORKSPACE}/col_{COL}_to_agg_cert_request.zip\n", + "\n", + "fx collaborator certify --import {FED_DIRECTORY}/agg_to_col_{COL}_signed_cert.zip" + ] + }, + { + "cell_type": "code", + "execution_count": 2, + "id": "12f5663f", + "metadata": {}, + "outputs": [ + { + "data": { + "text/plain": [ + "2" + ] + }, + "execution_count": 2, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "# Interface generator\n", + "class IFO:\n", + " def __init__(self):\n", + " self.a = 1\n", + " \n", + "class EXP:\n", + " def __init__(self):\n", + " self.ifo = IFO()\n", + " \n", + " def get_ifo(self):\n", + " return self.ifo\n", + " \n", + "exp = EXP()\n", + "ifo = exp.get_ifo()\n", + "ifo.a = 2\n", + "\n", + "# print(exp.ifo.a)\n", + "exp2 = exp.__class__()\n", + "exp.ifo.a" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "id": "mature-renewal", + "metadata": {}, + "outputs": [], + "source": [ + "# create an experimnet in federation\n", + "fl_experiment = FLExperiment(federation=federation,)" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "id": "lightweight-causing", + "metadata": { + "scrolled": false + }, + "outputs": [], + "source": [ + "# If I use autoreload I got a pickling error\n", + "\n", + "# The following command zips the workspace and python requirements to be transfered to collaborator nodes\n", + "fl_experiment.prepare_workspace_distribution(model_provider=MI, task_keeper=TI, data_loader=fed_dataset, rounds_to_train=7, \\\n", + " opt_treatment='CONTINUE_GLOBAL')\n", + "# This command starts the aggregator server. You can pass 'metric' log_level\n", + "fl_experiment.start_experiment(model_provider=MI, log_level='INFO', log_file='federation.log')\n", + "\n", + "# When the aggregator server blocks the notebook one can start collaborators\n", + "# For the test run just type console command from the workspace directory:\n", + "# `fx collaborator start -d data.yaml -n {col_name}` for all collaborators\n", + "# For the distributed experiment transfer zipped workspace to the collaborator nodes and run\n", + "# `fx workspace import --archive {workspace_name}.zip` cd to the workspace and start collaborators" + ] + }, + { + "cell_type": "markdown", + "id": "8c30b301", + "metadata": {}, + "source": [ + "## Now we validate the best model!" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "id": "55acff59", + "metadata": {}, + "outputs": [], + "source": [ + "best_model = fl_experiment.get_best_model()" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "id": "9fb5718d", + "metadata": {}, + "outputs": [], + "source": [ + "fed_dataset._delayed_init()" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "id": "a2acb7e6", + "metadata": {}, + "outputs": [], + "source": [ + "# Validating initial model\n", + "validate(initial_model, fed_dataset.get_valid_loader(), 'cpu')" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "id": "c12ca93f", + "metadata": {}, + "outputs": [], + "source": [ + "# Validating trained model\n", + "validate(best_model, fed_dataset.get_valid_loader(), 'cpu')" + ] + }, + { + "cell_type": "markdown", + "id": "1e6734f6", + "metadata": {}, + "source": [ + "## We can tune model further!" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "id": "3940e75e", + "metadata": {}, + "outputs": [], + "source": [ + "MI = ModelInterface(model=best_model, optimizer=optimizer_adam, framework_plugin=framework_adapter)\n", + "fl_experiment.start_experiment(model_provider=MI, task_keeper=TI, data_loader=fed_dataset, rounds_to_train=4, \\\n", + " opt_treatment='CONTINUE_GLOBAL')" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "id": "1bd786d2", + "metadata": {}, + "outputs": [], + "source": [ + "best_model = fl_experiment.get_best_model()\n", + "# Validating trained model\n", + "validate(best_model, fed_dataset.get_valid_loader(), 'cpu')" + ] + }, + { + "cell_type": "code", + "execution_count": 40, + "id": "2dcdbfeb", + "metadata": {}, + "outputs": [ + { + "ename": "SyntaxError", + "evalue": "invalid syntax (<ipython-input-40-b0f7ed81de85>, line 2)", + "output_type": "error", + "traceback": [ + "\u001B[0;36m File \u001B[0;32m\"<ipython-input-40-b0f7ed81de85>\"\u001B[0;36m, line \u001B[0;32m2\u001B[0m\n\u001B[0;31m a = [elem for elem in a if elem.shape==2 else elem.newaxis()]\u001B[0m\n\u001B[0m ^\u001B[0m\n\u001B[0;31mSyntaxError\u001B[0m\u001B[0;31m:\u001B[0m invalid syntax\n" + ] + } + ], + "source": [ + "a = (np.zeros((2,4)), np.ones(2,), 2*np.ones(2,))\n", + "a = [elem for elem in a if elem.shape==2 else elem.newaxis()]\n", + "np.concatenate(a)" + ] + }, + { + "cell_type": "code", + "execution_count": 15, + "id": "70863215", + "metadata": {}, + "outputs": [], + "source": [ + "class A:\n", + " def __init__(self, **kwargs):\n", + " print(\"Class A\", kwargs)\n", + " \n", + "class B:\n", + " def __init__(self, **kwargs):\n", + " super().__init__()\n", + " print(\"Class B\", kwargs)\n", + " \n", + "class C(B, A):\n", + " def __init__(self, **kwargs):\n", + " super().__init__()\n", + " print(\"Class C\", kwargs)\n", + " \n", + "# class A:\n", + "# def __init__(self, **kwargs):\n", + "# print(\"Class A\", kwargs)" + ] + }, + { + "cell_type": "code", + "execution_count": 16, + "id": "4ba44919", + "metadata": {}, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "Class A {}\n", + "Class B {}\n", + "Class C {'x': 1, 'z': 5}\n" + ] + } + ], + "source": [ + "c = C(x=1, z=5)" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "id": "0637f26f", + "metadata": {}, + "outputs": [], + "source": [] + } + ], + "metadata": { + "kernelspec": { + "display_name": "Python 3", + "language": "python", + "name": "python3" + }, + "language_info": { + "codemirror_mode": { + "name": "ipython", + "version": 3 + }, + "file_extension": ".py", + "mimetype": "text/x-python", + "name": "python", + "nbconvert_exporter": "python", + "pygments_lexer": "ipython3", + "version": "3.6.9" + } + }, + "nbformat": 4, + "nbformat_minor": 5 +} \ No newline at end of file diff --git a/openfl-tutorials/interactive_api_tutorials_(experimental)/Pytorch_TinyImageNet_workspace/new_python_api_tinyimagenet.ipynb b/openfl-tutorials/interactive_api/Pytorch_TinyImageNet/new_python_api_tinyimagenet.ipynb similarity index 100% rename from openfl-tutorials/interactive_api_tutorials_(experimental)/Pytorch_TinyImageNet_workspace/new_python_api_tinyimagenet.ipynb rename to openfl-tutorials/interactive_api/Pytorch_TinyImageNet/new_python_api_tinyimagenet.ipynb diff --git a/openfl-tutorials/interactive_api_tutorials_(experimental)/Tensorflow_MNIST_workspace/new_python_api_Tensorflow_MNIST.ipynb b/openfl-tutorials/interactive_api/Tensorflow_MNIST/new_python_api_Tensorflow_MNIST.ipynb similarity index 100% rename from openfl-tutorials/interactive_api_tutorials_(experimental)/Tensorflow_MNIST_workspace/new_python_api_Tensorflow_MNIST.ipynb rename to openfl-tutorials/interactive_api/Tensorflow_MNIST/new_python_api_Tensorflow_MNIST.ipynb diff --git a/openfl-workspace/default/director.yaml b/openfl-workspace/default/director.yaml new file mode 100644 index 0000000000..32fd4caa30 --- /dev/null +++ b/openfl-workspace/default/director.yaml @@ -0,0 +1,11 @@ +# Copyright (C) 2020-2021 Intel Corporation +# SPDX-License-Identifier: Apache-2.0 + +# Director's config. +# Parameters: +# 1. sample_shape - sample shape interface unified across the Federation +# 1. target_shape - target shape interface unified across the Federation + +settings: + sample_shape: [] + target_shape: [] \ No newline at end of file diff --git a/openfl-workspace/default/shard_config.yaml b/openfl-workspace/default/shard_config.yaml new file mode 100644 index 0000000000..cb08ac22d2 --- /dev/null +++ b/openfl-workspace/default/shard_config.yaml @@ -0,0 +1,17 @@ +# Copyright (C) 2020-2021 Intel Corporation +# SPDX-License-Identifier: Apache-2.0 + +# To start envoy implement LocalShardDescriptor class in shard_descriptor module. +# Alternatively, point to an implemented Shard Descriptor in 'template' field. + +# Put in 'params' group any argument needed to initialize choosen Shard Descriptor. +# Parameters are optional. For clarity, this example config contains three parameters: +# 1. data_path - string with data location on disk +# 2. sample_shape - shape of sample's numpy representaion that will by return from __getitem__ +# 3. target_shape - shape of target's numpy representaion that will by return from __getitem__ + +template: shard_descriptor.LocalShardDescriptor +params: + data_path: data + sample_shape: [] + target_shape: [] \ No newline at end of file diff --git a/openfl-workspace/default/shard_descriptor.py b/openfl-workspace/default/shard_descriptor.py new file mode 100644 index 0000000000..2f41db0f40 --- /dev/null +++ b/openfl-workspace/default/shard_descriptor.py @@ -0,0 +1,21 @@ +# Copyright (C) 2020-2021 Intel Corporation +# SPDX-License-Identifier: Apache-2.0 +""" +Shard Descriptor template. + +It is recommended to perform tensor manipulations using numpy. +""" + +from openfl.interface.interactive_api.shard_descriptor import ShardDescriptor + + +class LocalShardDescriptor(ShardDescriptor): + """Shard descriptor subclass.""" + + def __init__(self, data_path: str, sample_shape: tuple, target_shape: tuple) -> None: + """ + Initialize local Shard Descriptor. + + Parameters are arbitrary, set up a shard_config.yaml as you need. + """ + super().__init__() diff --git a/openfl-workspace/torch_cnn_mnist/plan/plan.yaml b/openfl-workspace/torch_cnn_mnist/plan/plan.yaml index ffef4de9ac..903d94e27a 100644 --- a/openfl-workspace/torch_cnn_mnist/plan/plan.yaml +++ b/openfl-workspace/torch_cnn_mnist/plan/plan.yaml @@ -5,10 +5,13 @@ aggregator : defaults : plan/defaults/aggregator.yaml template : openfl.component.Aggregator settings : - init_state_path : save/torch_cnn_mnist_init.pbuf - best_state_path : save/torch_cnn_mnist_best.pbuf - last_state_path : save/torch_cnn_mnist_last.pbuf - rounds_to_train : 10 + init_state_path : save/torch_cnn_mnist_init.pbuf + best_state_path : save/torch_cnn_mnist_best.pbuf + last_state_path : save/torch_cnn_mnist_last.pbuf + rounds_to_train : 10 + log_metric_callback : + template : src.mnist_utils.write_metric + collaborator : defaults : plan/defaults/collaborator.yaml diff --git a/openfl-workspace/torch_cnn_mnist/requirements.txt b/openfl-workspace/torch_cnn_mnist/requirements.txt index 9ef102a829..29815a9214 100644 --- a/openfl-workspace/torch_cnn_mnist/requirements.txt +++ b/openfl-workspace/torch_cnn_mnist/requirements.txt @@ -1,2 +1,3 @@ torch >= 1.2.0 torchvision >= 0.9.1 +tensorboard diff --git a/openfl-workspace/torch_cnn_mnist/src/mnist_utils.py b/openfl-workspace/torch_cnn_mnist/src/mnist_utils.py index a03e1e6da2..16ee801b4d 100644 --- a/openfl-workspace/torch_cnn_mnist/src/mnist_utils.py +++ b/openfl-workspace/torch_cnn_mnist/src/mnist_utils.py @@ -6,11 +6,27 @@ from logging import getLogger import numpy as np +from torch.utils.tensorboard import SummaryWriter from torchvision import datasets from torchvision import transforms logger = getLogger(__name__) +writer = None + + +def get_writer(): + """Create global writer object.""" + global writer + if not writer: + writer = SummaryWriter('./logs/cnn_mnist', flush_secs=5) + + +def write_metric(node_name, task_name, metric_name, metric, round_number): + """Write metric callback.""" + get_writer() + writer.add_scalar(f'{node_name}/{task_name}/{metric_name}', metric, round_number) + def one_hot(labels, classes): """ diff --git a/openfl-workspace/torch_unet_kvasir/src/data_loader.py b/openfl-workspace/torch_unet_kvasir/src/data_loader.py index 22b555593a..3ae0b8e65e 100644 --- a/openfl-workspace/torch_unet_kvasir/src/data_loader.py +++ b/openfl-workspace/torch_unet_kvasir/src/data_loader.py @@ -51,8 +51,11 @@ def __init__(self, is_validation, shard_num, collaborator_count, **kwargs): """ self.images_path = './data/segmented-images/images/' self.masks_path = './data/segmented-images/masks/' - self.images_names = [img_name for img_name in sorted(listdir( - self.images_path)) if len(img_name) > 3 and img_name[-3:] == 'jpg'] + self.images_names = [ + img_name + for img_name in sorted(listdir(self.images_path)) + if len(img_name) > 3 and img_name[-3:] == 'jpg' + ] self.images_names = self.images_names[shard_num:: collaborator_count] self.is_validation = is_validation diff --git a/openfl-workspace/torch_unet_kvasir/src/pt_unet_parts.py b/openfl-workspace/torch_unet_kvasir/src/pt_unet_parts.py index 1a7318a64f..5883c3ec6a 100644 --- a/openfl-workspace/torch_unet_kvasir/src/pt_unet_parts.py +++ b/openfl-workspace/torch_unet_kvasir/src/pt_unet_parts.py @@ -71,7 +71,8 @@ def __init__(self, in_ch, out_ch): super(Down, self).__init__() self.mpconv = nn.Sequential( nn.MaxPool2d(2), - DoubleConv(in_ch, out_ch)) + DoubleConv(in_ch, out_ch) + ) def forward(self, x): """Run forward.""" diff --git a/openfl/component/aggregator/aggregator.py b/openfl/component/aggregator/aggregator.py index 2a02cd03bd..8e5a35b8b4 100644 --- a/openfl/component/aggregator/aggregator.py +++ b/openfl/component/aggregator/aggregator.py @@ -2,6 +2,7 @@ # SPDX-License-Identifier: Apache-2.0 """Aggregator module.""" +import queue from logging import getLogger from openfl.databases import TensorDB @@ -11,6 +12,7 @@ from openfl.protocols import utils from openfl.utilities import TaskResultKey from openfl.utilities import TensorKey +from openfl.utilities.logs import write_metric class Aggregator: @@ -42,6 +44,7 @@ def __init__(self, assigner, rounds_to_train=256, + log_metric_callback=None, single_col_cert_common_name=None, compression_pipeline=None, db_store_rounds=1, @@ -81,6 +84,7 @@ def __init__(self, self.best_tensor_dict: dict = {} self.last_tensor_dict: dict = {} + self.metric_queue = queue.Queue() self.best_model_score = None if kwargs.get('initial_tensor_dict', None) is not None: @@ -103,6 +107,8 @@ def __init__(self, self.collaborator_task_weight = {} # {TaskResultKey: data_size} + self.log_metric = write_metric + def _load_initial_tensors(self): """ Load all of the tensors required to begin federated learning. @@ -482,7 +488,13 @@ def send_local_task_results(self, collaborator_name, round_number, task_name, tensor_key, nparray = self._process_named_tensor( named_tensor, collaborator_name ) - + if 'metric' in tensor_key.tags: + self.log_metric(tensor_key.tags[-1], task_name, + tensor_key.tensor_name, nparray, round_number) + self.logger.metric(f'Round {round_number}, collaborator {tensor_key.tags[-1]} ' + f'{task_name} result {tensor_key.tensor_name}:\t{nparray}') + self.metric_queue.put((tensor_key.tags[-1], task_name, + tensor_key.tensor_name, nparray, round_number)) task_results.append(tensor_key) # By giving task_key it's own weight, we can support different # training/validation weights @@ -708,7 +720,6 @@ def _compute_validation_related_task_metrics(self, task_name): task_name : str The task name to compute """ - self.logger.info(f'{task_name} task metrics...') # By default, print out all of the metrics that the validation # task sent # This handles getting the subset of collaborators that may be @@ -748,19 +759,25 @@ def _compute_validation_related_task_metrics(self, task_name): if agg_results is None: self.logger.warning( f'Aggregated metric {agg_tensor_name} could not be collected ' - f'for round {self.round_number}. Skipping reporting for this round' - ) + f'for round {self.round_number}. Skipping reporting for this round') if agg_function: - self.logger.info(f'{agg_function} {agg_tensor_name}:\t{agg_results:.4f}') + self.logger.metric(f'Round {round_number}, aggregator: {task_name} ' + f'{agg_function} {agg_tensor_name}:\t{agg_results:.4f}') else: - self.logger.info(f'{agg_tensor_name}:\t{agg_results:.4f}') + self.logger.metric(f'Round {round_number}, aggregator: {task_name} ' + f'{agg_tensor_name}:\t{agg_results:.4f}') + self.log_metric('Aggregator', task_name, tensor_key.tensor_name, + agg_results, round_number) + self.metric_queue.put(('Aggregator', task_name, tensor_key.tensor_name, + agg_results, round_number)) # TODO Add all of the logic for saving the model based # on best accuracy, lowest loss, etc. if 'validate_agg' in tags: # Compare the accuracy of the model, and # potentially save it if self.best_model_score is None or self.best_model_score < agg_results: - self.logger.info(f'Saved the best model with score {agg_results:f}') + self.logger.metric(f'Round {round_number}: saved the best ' + f'model with score {agg_results:f}') self.best_model_score = agg_results self._save_model(round_number, self.best_state_path) if 'trained' in tags: @@ -904,5 +921,5 @@ def _log_big_warning(self): # #*#@##, .++:.,# `* @# +. @@@ - #`@ + # `@ , ''' diff --git a/openfl/component/ca/__init__.py b/openfl/component/ca/__init__.py new file mode 100644 index 0000000000..1d77e7fdee --- /dev/null +++ b/openfl/component/ca/__init__.py @@ -0,0 +1,4 @@ +# Copyright (C) 2020-2021 Intel Corporation +# SPDX-License-Identifier: Apache-2.0 + +"""CA package.""" diff --git a/openfl/component/ca/ca.py b/openfl/component/ca/ca.py new file mode 100644 index 0000000000..3bf6aa8458 --- /dev/null +++ b/openfl/component/ca/ca.py @@ -0,0 +1,244 @@ +# Copyright (C) 2020-2021 Intel Corporation +# SPDX-License-Identifier: Apache-2.0 + +"""Aggregator module.""" + + +import base64 +import json +import os +import shutil +import signal +import subprocess +import time +import urllib.request +from logging import getLogger +from pathlib import Path +from subprocess import call + +import requests +from click import confirm + + +logger = getLogger(__name__) + +TOKEN_DELIMITER = '.' + + +def download_step_bin(url, grep_name, architecture, prefix='.', confirmation=True): + """ + Donwload step binaries from github. + + Args: + url: address of latest release + grep_name: name to grep over github assets + architecture: architecture type to grep + prefix: folder path to download + confirmation: request user confirmation or not + """ + if confirmation: + confirm('CA binaries from github will be downloaded now', default=True, abort=True) + result = requests.get(url) + if result.status_code != 200: + logger.warning('Can\'t download binaries from github. Please try lately.') + return + + assets = result.json().get('assets', []) + urls = [ + a['browser_download_url'] + for a in assets + if grep_name in a['name'] and architecture in a['name'] + ] + url = urls[-1] + url = url.replace('https', 'http') + name = url.split('/')[-1] + logger.info(f'Downloading {name}') + urllib.request.urlretrieve(url, f'{prefix}/{name}') + shutil.unpack_archive(f'{prefix}/{name}', f'{prefix}/step') + + +def get_token(name, ca_url, ca_path='.'): + """ + Create authentication token. + + Args: + name: common name for following certificate + (aggregator fqdn or collaborator name) + ca_url: full url of CA server + ca_path: path to ca folder + """ + ca_path = Path(ca_path) + step_config_dir = ca_path / 'step_config' + pki_dir = ca_path / 'cert' + step_path, _ = get_ca_bin_paths(ca_path) + if not step_path: + raise Exception('Step-CA is not installed!\nRun `fx pki install` first') + + priv_json = step_config_dir / 'secrets' / 'priv.json' + pass_file = pki_dir / 'pass_file' + root_crt = step_config_dir / 'certs' / 'root_ca.crt' + try: + token = subprocess.check_output( + f'{step_path} ca token {name} ' + f'--key {priv_json} --root {root_crt} ' + f'--password-file {pass_file} 'f'--ca-url {ca_url}', shell=True) + except subprocess.CalledProcessError as exc: + logger.error(f'Error code {exc.returncode}: {exc.output}') + return + + token = token.strip() + token_b64 = base64.b64encode(token) + + with open(step_config_dir / 'certs' / 'root_ca.crt', mode='rb') as file: + root_ca = file.read() + root_ca_b64 = base64.b64encode(root_ca) + + return TOKEN_DELIMITER.join([ + token_b64.decode('utf-8'), + root_ca_b64.decode('utf-8'), + ]) + + +def get_ca_bin_paths(ca_path): + """Get paths of step binaries.""" + ca_path = Path(ca_path) + step = None + step_ca = None + if (ca_path / 'step').exists(): + dirs = os.listdir(ca_path / 'step') + for dir_ in dirs: + if 'step_' in dir_: + step = ca_path / 'step' / dir_ / 'bin' / 'step' + if 'step-ca' in dir_: + step_ca = ca_path / 'step' / dir_ / 'bin' / 'step-ca' + return step, step_ca + + +def certify(name, cert_path: Path, token_with_cert, ca_path: Path): + """Create an envoy workspace.""" + os.makedirs(cert_path, exist_ok=True) + + token, root_ca = token_with_cert.split(TOKEN_DELIMITER) + token = base64.b64decode(token).decode('utf-8') + root_ca = base64.b64decode(root_ca) + + step_path, _ = get_ca_bin_paths(ca_path) + if not step_path: + url = 'http://api.github.com/repos/smallstep/cli/releases/latest' + download_step_bin(url, 'step_linux', 'amd', prefix=ca_path) + step_path, _ = get_ca_bin_paths(ca_path) + if not step_path: + raise Exception('Step-CA is not installed!\nRun `fx pki install` first') + + with open(f'{cert_path}/root_ca.crt', mode='wb') as file: + file.write(root_ca) + call(f'./{step_path} ca certificate {name} {cert_path}/{name}.crt ' + f'{cert_path}/{name}.key -f --token {token}', shell=True) + + +def remove_ca(ca_path): + """Kill step-ca process and rm ca directory.""" + _check_kill_process('step-ca') + shutil.rmtree(ca_path, ignore_errors=True) + + +def install(ca_path, ca_url, password): + """ + Create certificate authority for federation. + + Args: + ca_path: path to ca directory + ca_url: url for ca server like: 'host:port' + password: Simple password for encrypting root private keys + + """ + logger.info('Creating CA') + + ca_path = Path(ca_path) + ca_path.mkdir(parents=True, exist_ok=True) + step_config_dir = ca_path / 'step_config' + os.environ['STEPPATH'] = str(step_config_dir) + step_path, step_ca_path = get_ca_bin_paths(ca_path) + + if not (step_path and step_ca_path and step_path.exists() and step_ca_path.exists()): + confirm('CA binaries from github will be downloaded now', default=True, abort=True) + url = 'http://api.github.com/repos/smallstep/certificates/releases/latest' + download_step_bin(url, 'step-ca_linux', 'amd', prefix=ca_path, confirmation=False) + url = 'http://api.github.com/repos/smallstep/cli/releases/latest' + download_step_bin(url, 'step_linux', 'amd', prefix=ca_path, confirmation=False) + step_config_dir = ca_path / 'step_config' + if (not step_config_dir.exists() + or confirm('CA exists, do you want to recreate it?', default=True)): + _create_ca(ca_path, ca_url, password) + _configure(step_config_dir) + + +def run_ca(step_ca, pass_file, ca_json): + """Run CA server.""" + if _check_kill_process('step-ca', confirmation=True): + logger.info('Up CA server') + call(f'{step_ca} --password-file {pass_file} {ca_json}', shell=True) + + +def _check_kill_process(pstring, confirmation=False): + """Kill process by name.""" + pids = [] + proc = subprocess.Popen(f'ps ax | grep {pstring} | grep -v grep', + shell=True, stdout=subprocess.PIPE) + text = proc.communicate()[0].decode('utf-8') + + for line in text.splitlines(): + fields = line.split() + pids.append(fields[0]) + + if len(pids): + if confirmation and not confirm('CA server is already running. Stop him?', default=True): + return False + for pid in pids: + os.kill(int(pid), signal.SIGKILL) + time.sleep(2) + return True + + +def _create_ca(ca_path: Path, ca_url: str, password: str): + """Create a ca workspace.""" + pki_dir = ca_path / 'cert' + step_config_dir = ca_path / 'step_config' + + pki_dir.mkdir(parents=True, exist_ok=True) + step_config_dir.mkdir(parents=True, exist_ok=True) + + with open(f'{pki_dir}/pass_file', 'w') as f: + f.write(password) + step_path, step_ca_path = get_ca_bin_paths(ca_path) + assert(step_path and step_ca_path and step_path.exists() and step_ca_path.exists()) + + logger.info('Create CA Config') + os.environ['STEPPATH'] = str(step_config_dir) + shutil.rmtree(step_config_dir, ignore_errors=True) + name = ca_url.split(':')[0] + call(f'{step_path} ca init --name name --dns {name} ' + f'--address {ca_url} --provisioner prov ' + f'--password-file {pki_dir}/pass_file', shell=True) + + call(f'{step_path} ca provisioner remove prov --all', shell=True) + call(f'{step_path} crypto jwk create {step_config_dir}/certs/pub.json ' + f'{step_config_dir}/secrets/priv.json --password-file={pki_dir}/pass_file', shell=True) + call( + f'{step_path} ca provisioner add provisioner {step_config_dir}/certs/pub.json', + shell=True + ) + + +def _configure(step_config_dir): + conf_file = step_config_dir / 'config' / 'ca.json' + with open(conf_file, 'r+') as f: + data = json.load(f) + data.setdefault('authority', {}).setdefault('claims', {}) + data['authority']['claims']['maxTLSCertDuration'] = f'{365 * 24}h' + data['authority']['claims']['defaultTLSCertDuration'] = f'{365 * 24}h' + data['authority']['claims']['maxUserSSHCertDuration'] = '24h' + data['authority']['claims']['defaultUserSSHCertDuration'] = '24h' + f.seek(0) + json.dump(data, f, indent=4) + f.truncate() diff --git a/openfl/component/collaborator/collaborator.py b/openfl/component/collaborator/collaborator.py index 840ae54331..b39ce9d09f 100644 --- a/openfl/component/collaborator/collaborator.py +++ b/openfl/component/collaborator/collaborator.py @@ -363,9 +363,9 @@ def send_task_results(self, tensor_dict, round_number, task_name): tensor_name, origin, fl_round, report, tags = tensor if report: - self.logger.info( - f'Sending metric for task {task_name},' - f' round number {round_number}:' + self.logger.metric( + f'Round {round_number}, collaborator {self.collaborator_name} ' + f'is sending metric for task {task_name}:' f' {tensor_name}\t{tensor_dict[tensor]}') self.client.send_local_task_results( diff --git a/openfl/component/director/__init__.py b/openfl/component/director/__init__.py new file mode 100644 index 0000000000..580285bb0e --- /dev/null +++ b/openfl/component/director/__init__.py @@ -0,0 +1,11 @@ +# Copyright (C) 2020-2021 Intel Corporation +# SPDX-License-Identifier: Apache-2.0 + +"""Director package.""" + +from .director import Director + + +__all__ = [ + 'Director', +] diff --git a/openfl/component/director/director.py b/openfl/component/director/director.py new file mode 100644 index 0000000000..81506cf04a --- /dev/null +++ b/openfl/component/director/director.py @@ -0,0 +1,235 @@ +# Copyright (C) 2020-2021 Intel Corporation +# SPDX-License-Identifier: Apache-2.0 + +"""Director module.""" + +import asyncio +import logging +import os +import shutil +import time +import typing +from collections import defaultdict +from pathlib import Path + +from openfl.federated import Plan +from openfl.protocols import director_pb2 + +logger = logging.getLogger(__name__) + + +class Director: + """Director class.""" + + def __init__(self, *, disable_tls: bool = False, + root_ca: Path = None, key: Path = None, cert: Path = None, + sample_shape: list = None, target_shape: list = None) -> None: + """Initialize a director object.""" + # TODO: add working directory + super().__init__() + self.sample_shape, self.target_shape = sample_shape, target_shape + self._shard_registry = {} + self.col_exp_queues = defaultdict(asyncio.Queue) + + self.experiment_data = {} # {Experiment name : archive bytes} + # What if two experiments come with the same name from different users? + self.experiments_queue = asyncio.Queue() # experiments waiting to be executed + self.experiment_stash = defaultdict(dict) # Running of finished experiments + # {API name : {experiment name : aggregator}} + + self.disable_tls = disable_tls + self.root_ca = root_ca + self.key = key + self.cert = cert + + def acknowledge_shard(self, shard_info: director_pb2.ShardInfo) -> bool: + """Save shard info to shard registry if it's acceptable.""" + is_accepted = False + if (self.sample_shape != shard_info.sample_shape + or self.target_shape != shard_info.target_shape): + logger.info('Request was not accepted') + return is_accepted + logger.info('Request was accepted') + self._shard_registry[shard_info.node_info.name] = { + 'shard_info': shard_info, + 'is_online': True, + 'is_experiment_running': False + } + is_accepted = True + return is_accepted + + async def set_new_experiment(self, *, experiment_name: str, sender_name: str, + tensor_dict: dict, + collaborator_names: typing.Iterable[str], data: bytes) -> bool: + """Set new experiment.""" + # TODO: save to file + self.experiment_data[experiment_name] = data + + self.create_workspace(experiment_name, data) + asyncio.create_task(self._run_aggregator( + sender_name, + tensor_dict, + experiment_name, + collaborator_names + )) + + logger.info(f'New experiment {experiment_name} for ' + f'collaborators {collaborator_names}') + for col_name in collaborator_names: + queue = self.col_exp_queues[col_name] + await queue.put(experiment_name) + + return True + + async def get_trained_model(self, experiment_name: str, caller: str): + """Get trained models.""" + if experiment_name not in self.experiment_stash[caller]: + logger.error('Aggregator has not started yet') + return None, None + + aggregator = self.experiment_stash[caller][experiment_name].aggregator + + if aggregator.last_tensor_dict is None: + logger.error('Aggregator have no aggregated model to return') + return None, None + + return aggregator.best_tensor_dict, aggregator.last_tensor_dict + + def get_experiment_data(self, experiment_name: str) -> bytes: + """Get experiment data.""" + return self.experiment_data.get(experiment_name, b'') + + async def wait_experiment(self, collaborator_name: str) -> str: + """Wait an experiment.""" + queue = self.col_exp_queues[collaborator_name] + experiment_name = await queue.get() + + return experiment_name + + def get_dataset_info(self): + """Get dataset info.""" + return self.sample_shape, self.target_shape + + def get_registered_shards(self) -> list: + """Get registered shard infos.""" + return [shard_status['shard_info'] for shard_status in self._shard_registry.values()] + + async def stream_metrics(self, experiment_name: str, caller: str): + """Stream metrics from the aggregator.""" + aggregator = self.experiment_stash[caller][experiment_name].aggregator + + while not aggregator.all_quit_jobs_sent() or not aggregator.metric_queue.empty(): + # If the aggregator has not fineished the experimnet + # or it finished but metric_queue is not empty we send metrics + + # But here we may have a problem if the new experiment starts too quickly + + while not aggregator.metric_queue.empty(): + metric_origin, task_name, metric_name, metric_value, round_ = \ + aggregator.metric_queue.get() + yield director_pb2.StreamMetricsResponse( + metric_origin=metric_origin, + task_name=task_name, + metric_name=metric_name, + metric_value=float(metric_value), + round=round_) + + # Awaiting quit job sent to collaborators + await asyncio.sleep(5) + + def remove_experiment_data(self, name: str, caller: str): + """Remove experiment data RPC.""" + if name in self.experiment_stash.get(caller, {}): + del self.experiment_stash[caller][name] + + def collaborator_health_check(self, *, collaborator_name: str, + is_experiment_running: bool, + valid_duration: int) -> bool: + """Accept health check from envoy.""" + is_accepted = False + shard_info = self._shard_registry.get(collaborator_name) + if not shard_info: + logger.error(f'Unknown shard {collaborator_name}') + return is_accepted + is_accepted = True + + shard_info['is_online']: True + shard_info['is_experiment_running'] = is_experiment_running + shard_info['valid_duration'] = valid_duration + shard_info['last_updated'] = time.time() + + return is_accepted + + def get_envoys(self) -> list: + """Get a status information about envoys.""" + logger.info(f'Shard registry: {self._shard_registry}') + envoy_infos = [] + for envoy in self._shard_registry.values(): + envoy_info = director_pb2.EnvoyInfo( + shard_info=envoy['shard_info'], + is_online=time.time() < envoy['last_updated'] + envoy['valid_duration'], + is_experiment_running=envoy['is_experiment_running'] + ) + envoy_info.valid_duration.seconds = envoy['valid_duration'] + envoy_info.last_updated.seconds = int(envoy['last_updated']) + + envoy_infos.append(envoy_info) + + return envoy_infos + + @staticmethod + def create_workspace(experiment_name, npbytes): + """Create the aggregator workspace.""" + if os.path.exists(experiment_name): + shutil.rmtree(experiment_name) + os.makedirs(experiment_name) + + arch_name = f'{experiment_name}/{experiment_name}' + '.zip' + logger.info(f'arch_name: {arch_name}') + with open(arch_name, 'wb') as content_file: + content_file.write(npbytes) + + shutil.unpack_archive(arch_name, experiment_name) + + async def _run_aggregator( + self, + experiment_sender, + initial_tensor_dict, + experiment_name, + collaborator_names, + plan='plan/plan.yaml', + ): # TODO: path params, change naming + """Run aggregator.""" + cwd = os.getcwd() + os.chdir(f'{cwd}/{experiment_name}') + plan = Plan.parse(plan_config_path=Path(plan)) + + plan.authorized_cols = list(collaborator_names) + + logger.info('🧿 Starting the Aggregator Service.') + aggregator_server = plan.interactive_api_get_server( + tensor_dict=initial_tensor_dict, + chain=self.root_ca, + certificate=self.cert, + private_key=self.key, + disable_tls=self.disable_tls, + ) + self.experiment_stash[experiment_sender][experiment_name] = \ + aggregator_server + + grpc_server = aggregator_server.get_server() + grpc_server.start() + logger.info('Starting Aggregator gRPC Server') + + try: + while not aggregator_server.aggregator.all_quit_jobs_sent(): + # Awaiting quit job sent to collaborators + await asyncio.sleep(10) + except KeyboardInterrupt: + pass + finally: + os.chdir(cwd) + shutil.rmtree(experiment_name) + grpc_server.stop(0) + # Temporary solution to free RAM used by TensorDB + aggregator_server.aggregator.tensor_db.clean_up(0) diff --git a/openfl/component/envoy/__init__.py b/openfl/component/envoy/__init__.py new file mode 100644 index 0000000000..17fe14353f --- /dev/null +++ b/openfl/component/envoy/__init__.py @@ -0,0 +1,4 @@ +# Copyright (C) 2020-2021 Intel Corporation +# SPDX-License-Identifier: Apache-2.0 + +"""Envoy package.""" diff --git a/openfl/component/envoy/envoy.py b/openfl/component/envoy/envoy.py new file mode 100644 index 0000000000..4ae1d8e85c --- /dev/null +++ b/openfl/component/envoy/envoy.py @@ -0,0 +1,112 @@ +# Copyright (C) 2020-2021 Intel Corporation +# SPDX-License-Identifier: Apache-2.0 + +"""Envoy module.""" + +import logging +import os +import sys +import time +from concurrent.futures import ThreadPoolExecutor +from pathlib import Path + +from click import echo + +from openfl.federated import Plan +from openfl.transport.grpc.director_client import ShardDirectorClient + +logger = logging.getLogger(__name__) + + +DEFAULT_TIMEOUT_IN_SECONDS = 60 # TODO: make configurable +DEFAULT_RETRY_TIMEOUT_IN_SECONDS = 5 + + +class Envoy: + """Envoy class.""" + + def __init__(self, shard_name, director_uri, shard_descriptor, + root_ca: str = None, key: str = None, cert: str = None, + disable_tls: bool = False) -> None: + """Initialize a envoy object.""" + self.name = shard_name + self.root_ca = Path(root_ca).absolute() if root_ca is not None else None + self.key = Path(key).absolute() if root_ca is not None else None + self.cert = Path(cert).absolute() if root_ca is not None else None + self.director_client = ShardDirectorClient(director_uri, shard_name=shard_name, + disable_tls=disable_tls, + root_ca=root_ca, key=key, cert=cert) + self.shard_descriptor = shard_descriptor + self.executor = ThreadPoolExecutor() + self.running_experiments = {} + self.is_experiment_running = False + self._health_check_future = None + + def run(self): + """Run of the envoy working cycle.""" + while True: + try: + # Workspace import should not be done by gRPC client! + experiment_name = self.director_client.get_experiment_data() + except Exception as exc: + logger.error(f'Failed to get experiment: {exc}') + time.sleep(DEFAULT_RETRY_TIMEOUT_IN_SECONDS) + continue + self.is_experiment_running = True + try: + self._run_collaborator(experiment_name) + except Exception as exc: + logger.error(f'Collaborator failed: {exc}') + finally: + # Workspace cleaning should not be done by gRPC client! + self.director_client.remove_workspace(experiment_name) + self.is_experiment_running = False + + def send_health_check(self): + """Send health check to the director.""" + logger.info('The health check sender is started.') + while True: + self.director_client.send_health_check( + self.name, + self.is_experiment_running, + DEFAULT_TIMEOUT_IN_SECONDS + ) + time.sleep(DEFAULT_TIMEOUT_IN_SECONDS / 2) + + def _run_collaborator(self, experiment_name, plan='plan/plan.yaml',): + """Run the collaborator for the experiment running.""" + cwd = os.getcwd() + os.chdir(f'{cwd}/{experiment_name}') # TODO: probably it should be another way + + # This is needed for python module finder + sys.path.append(os.getcwd()) + + plan = Plan.parse(plan_config_path=Path(plan)) + + # TODO: Need to restructure data loader config file loader + echo(f'Data = {plan.cols_data_paths}') + logger.info('🧿 Starting a Collaborator Service.') + + col = plan.get_collaborator(self.name, self.root_ca, self.key, + self.cert, shard_descriptor=self.shard_descriptor) + try: + col.run() + finally: + os.chdir(cwd) + + def start(self): + """Start the envoy.""" + try: + is_accepted = self.director_client.report_shard_info(self.shard_descriptor) + except Exception as exc: + logger.exception(str(exc)) + logger.exception('Failed to report shard info') + else: + if is_accepted: + # Shard accepted for participation in the federation + logger.info('Shard accepted') + self._health_check_future = self.executor.submit(self.send_health_check) + self.run() + else: + # Shut down + logger.error('Report shard info was not accepted') diff --git a/openfl/federated/plan/plan.py b/openfl/federated/plan/plan.py index e78585ad4e..e2c1b0a454 100644 --- a/openfl/federated/plan/plan.py +++ b/openfl/federated/plan/plan.py @@ -182,6 +182,27 @@ def build(template, settings, **override): return instance + @staticmethod + def import_(template): + """ + Import an instance of a openfl Component or Federated DataLoader/TaskRunner. + + Args: + template: Fully qualified object path + + Returns: + A Python object + """ + class_name = splitext(template)[1].strip('.') + module_path = splitext(template)[0] + Plan.logger.info(f'Importing [red]🡆[/] Object [red]{class_name}[/] ' + f'from [red]{module_path}[/] Module.', + extra={'markup': True}) + module = import_module(module_path) + instance = getattr(module, class_name) + + return instance + def __init__(self): """Initialize.""" self.config = {} # dictionary containing patched plan definition @@ -204,7 +225,7 @@ def __init__(self): self.name_ = None @property - def hash(self): # NOQA + def hash(self): # NOQA """Generate hash for this instance.""" self.hash_ = sha384(dump(self.config).encode('utf-8')) Plan.logger.info(f'FL-Plan hash is [blue]{self.hash_.hexdigest()}[/]', @@ -258,7 +279,16 @@ def get_aggregator(self, tensor_dict=None): defaults[SETTINGS]['authorized_cols'] = self.authorized_cols defaults[SETTINGS]['assigner'] = self.get_assigner() defaults[SETTINGS]['compression_pipeline'] = self.get_tensor_pipe() + log_metric_callback = defaults[SETTINGS].get('log_metric_callback') + + if log_metric_callback: + if isinstance(log_metric_callback, dict): + log_metric_callback = Plan.import_(**log_metric_callback) + elif not callable(log_metric_callback): + raise TypeError(f'log_metric_callback should be callable object ' + f'or be import from code part, get {log_metric_callback}') + defaults[SETTINGS]['log_metric_callback'] = log_metric_callback if self.aggregator_ is None: self.aggregator_ = Plan.build(**defaults, initial_tensor_dict=tensor_dict) @@ -318,12 +348,9 @@ def get_data_loader(self, collaborator_name): return self.loader_ # Python interactive api - def initialize_data_loader(self, data_loader, collaborator_name): + def initialize_data_loader(self, data_loader, shard_descriptor): """Get data loader.""" - data_path = self.cols_data_paths[ - collaborator_name - ] - data_loader._delayed_init(data_path=data_path) + data_loader.shard_descriptor = shard_descriptor return data_loader # legacy api (TaskRunner subclassing) @@ -372,8 +399,8 @@ def get_core_task_runner(self, data_loader=None, return self.runner_ - def get_collaborator(self, collaborator_name, - task_runner=None, client=None): + def get_collaborator(self, collaborator_name, root_ca=None, key=None, cert=None, + task_runner=None, client=None, shard_descriptor=None): """Get collaborator.""" defaults = self.config.get( 'collaborator', @@ -398,7 +425,7 @@ def get_collaborator(self, collaborator_name, if 'openfl.federated.task.task_runner' in self.config['task_runner']['template']: # Interactive API model_provider, task_keeper, data_loader = self.deserialize_interface_objects() - data_loader = self.initialize_data_loader(data_loader, collaborator_name) + data_loader = self.initialize_data_loader(data_loader, shard_descriptor) defaults[SETTINGS]['task_runner'] = self.get_core_task_runner( data_loader=data_loader, model_provider=model_provider, @@ -416,7 +443,10 @@ def get_collaborator(self, collaborator_name, defaults[SETTINGS]['client'] = self.get_client( collaborator_name, self.aggregator_uuid, - self.federation_uuid + self.federation_uuid, + root_ca, + key, + cert ) if self.collaborator_ is None: @@ -424,13 +454,18 @@ def get_collaborator(self, collaborator_name, return self.collaborator_ - def get_client(self, collaborator_name, aggregator_uuid, federation_uuid): + def get_client(self, collaborator_name, aggregator_uuid, federation_uuid, + root_ca=None, key=None, cert=None): """Get gRPC client for the specified collaborator.""" common_name = collaborator_name - - chain = 'cert/cert_chain.crt' - certificate = f'cert/client/col_{common_name}.crt' - private_key = f'cert/client/col_{common_name}.key' + if root_ca and key and cert: + chain = root_ca + certificate = cert + private_key = key + else: + chain = 'cert/cert_chain.crt' + certificate = f'cert/client/col_{common_name}.crt' + private_key = f'cert/client/col_{common_name}.key' client_args = self.config['network'][SETTINGS] @@ -448,13 +483,18 @@ def get_client(self, collaborator_name, aggregator_uuid, federation_uuid): return self.client_ - def get_server(self): + def get_server(self, root_ca=None, key=None, cert=None, **kwargs): """Get gRPC server of the aggregator instance.""" common_name = self.config['network'][SETTINGS]['agg_addr'].lower() - chain = 'cert/cert_chain.crt' - certificate = f'cert/server/agg_{common_name}.crt' - private_key = f'cert/server/agg_{common_name}.key' + if root_ca and key and cert: + chain = '../cert/root_ca.crt' + certificate = f'agg_{common_name}.crt' + private_key = f'agg_{common_name}.key' + else: + chain = 'cert/cert_chain.crt' + certificate = f'cert/server/agg_{common_name}.crt' + private_key = f'cert/server/agg_{common_name}.key' server_args = self.config['network'][SETTINGS] @@ -464,6 +504,8 @@ def get_server(self): server_args['certificate'] = certificate server_args['private_key'] = private_key + server_args.update(kwargs) + server_args['aggregator'] = self.get_aggregator() if self.server_ is None: @@ -471,7 +513,8 @@ def get_server(self): return self.server_ - def interactive_api_get_server(self, tensor_dict, chain, certificate, private_key): + def interactive_api_get_server(self, *, tensor_dict, chain, certificate, + private_key, disable_tls): """Get gRPC server of the aggregator instance.""" server_args = self.config['network'][SETTINGS] @@ -479,6 +522,7 @@ def interactive_api_get_server(self, tensor_dict, chain, certificate, private_ke server_args['ca'] = chain server_args['certificate'] = certificate server_args['private_key'] = private_key + server_args['disable_tls'] = disable_tls server_args['aggregator'] = self.get_aggregator(tensor_dict) @@ -489,13 +533,16 @@ def interactive_api_get_server(self, tensor_dict, chain, certificate, private_ke def deserialize_interface_objects(self): """Deserialize objects for TaskRunner.""" - interface_objects = [] serializer = Plan.build( self.config['api_layer']['required_plugin_components']['serializer_plugin'], {}) - for filename in ['model_interface_file', - 'tasks_interface_file', 'dataloader_interface_file']: - interface_objects.append( - serializer.restore_object(self.config['api_layer']['settings'][filename]) - ) + filenames = [ + 'model_interface_file', + 'tasks_interface_file', + 'dataloader_interface_file' + ] + interface_objects = [ + serializer.restore_object(self.config['api_layer']['settings'][filename]) + for filename in filenames + ] model_provider, task_keeper, data_loader = interface_objects return model_provider, task_keeper, data_loader diff --git a/openfl/federated/task/task_runner.py b/openfl/federated/task/task_runner.py index a261fed5ac..755426c6e6 100644 --- a/openfl/federated/task/task_runner.py +++ b/openfl/federated/task/task_runner.py @@ -72,8 +72,8 @@ def _prepare_tensorkeys_for_agggregation(self, metric_dict, validation_flag, else: suffix = 'validate' + validation_flag - tags = ('metric', suffix) - + tags = (suffix,) + tags = ('metric', *tags) metric_dict = { TensorKey(metric, origin, round_num, True, tags): np.array(value) for metric, value in metric_dict.items() diff --git a/openfl/interface/cli.py b/openfl/interface/cli.py index e094f7689e..434c2fde49 100755 --- a/openfl/interface/cli.py +++ b/openfl/interface/cli.py @@ -12,39 +12,36 @@ from click import pass_context from click import style +from openfl.utilities import add_log_level -def setup_logging(level='info'): + +def setup_logging(level='info', log_file=None): """Initialize logging settings.""" + import logging from logging import basicConfig - from logging import CRITICAL - from logging import DEBUG - from logging import ERROR - from logging import INFO - from logging import NOTSET - from logging import WARNING + from rich.console import Console from rich.logging import RichHandler - import os - - console = Console(width=160) - levels = \ - { - 'notset': NOTSET, - 'debug': DEBUG, - 'info': INFO, - 'warning': WARNING, - 'error': ERROR, - 'critical': CRITICAL - } + metric = 25 + add_log_level('METRIC', metric) - if level.lower() in ['debug', 'error']: - os.environ['GRPC_VERBOSITY'] = level.upper() + if isinstance(level, str): + level = level.upper() - level = levels.get(level.lower(), levels['notset']) + handlers = [] + if log_file: + fh = logging.FileHandler(log_file) + formatter = logging.Formatter( + '%(asctime)s %(levelname)s %(message)s %(filename)s:%(lineno)d' + ) + fh.setFormatter(formatter) + handlers.append(fh) + console = Console(width=160) + handlers.append(RichHandler(console=console)) basicConfig(level=level, format='%(message)s', - datefmt='[%X]', handlers=[RichHandler(console=console)]) + datefmt='[%X]', handlers=handlers) def disable_warnings(): @@ -129,6 +126,7 @@ def format_help(self, ctx, formatter): @pass_context def cli(context, log_level): """Command-line Interface.""" + import os from sys import argv context.ensure_object(dict) @@ -137,10 +135,11 @@ def cli(context, log_level): context.obj['script'] = argv[0] context.obj['arguments'] = argv[1:] - setup_logging(log_level) + log_file = os.getenv('LOG_FILE') + setup_logging(log_level, log_file) [email protected]() [email protected]_callback() @pass_context def end(context, result, **kwargs): """Print the result of the operation.""" @@ -190,7 +189,6 @@ def entry(): file = Path(__file__).resolve() root = file.parent.resolve() # interface root, containing command modules work = Path.cwd().resolve() - path.append(str(root)) path.insert(0, str(work)) diff --git a/openfl/interface/director.py b/openfl/interface/director.py new file mode 100644 index 0000000000..fb294ad4f4 --- /dev/null +++ b/openfl/interface/director.py @@ -0,0 +1,86 @@ +# Copyright (C) 2020-2021 Intel Corporation +# SPDX-License-Identifier: Apache-2.0 +"""Director CLI.""" + +import logging +import shutil +import sys +from pathlib import Path + +import click +from click import group +from click import option +from click import pass_context +from click import Path as ClickPath +from yaml import safe_load + +from openfl.component.director import Director +from openfl.interface.cli_helper import WORKSPACE +from openfl.transport import DirectorGRPCServer + +logger = logging.getLogger(__name__) + + +@group() +@pass_context +def director(context): + """Manage Federated Learning Director.""" + context.obj['group'] = 'director' + + [email protected](name='start') +@option('-c', '--director-config-path', default='director.yaml', + help='The director config file path', type=ClickPath(exists=True)) +@option('--disable-tls', default=False, + is_flag=True) +@option('-rc', '--root-cert-path', 'root_ca', default=None, + help='Path to a root CA cert') +@option('-pk', '--private-key-path', 'key', default=None, + help='Path to a private key') +@option('-oc', '--public-cert-path', 'cert', default=None, + help='Path to a signed certificate') +def start(director_config_path, disable_tls, root_ca, key, cert): + """Start the director service.""" + logger.info('🧿 Starting the Director Service.') + with open(director_config_path) as stream: + director_config = safe_load(stream) + settings = director_config.get('settings', {}) + sample_shape = settings.get('sample_shape', '') + target_shape = settings.get('target_shape', '') + logger.info(f'Sample shape: {sample_shape}, target shape: {target_shape}') + listen_ip = settings.get('listen_ip') + listen_port = settings.get('listen_port') + root_ca = root_ca or settings.get('root_ca') + key = key or settings.get('key') + cert = cert or settings.get('cert') + kwargs = {} + if listen_ip: + kwargs['listen_ip'] = listen_ip + if listen_port: + kwargs['listen_port'] = listen_port + director_server = DirectorGRPCServer( + director_cls=Director, + disable_tls=disable_tls, + sample_shape=sample_shape, + target_shape=target_shape, + root_ca=root_ca, + key=key, + cert=cert, + **kwargs + ) + director_server.start() + + [email protected](name='create-workspace') +@option('-p', '--director-path', required=True, + help='The director path', type=ClickPath()) +def create(director_path): + """Create a director workspace.""" + director_path = Path(director_path) + if director_path.exists(): + if not click.confirm('Director workspace already exists. Recreate?', default=True): + sys.exit(1) + shutil.rmtree(director_path) + (director_path / 'cert').mkdir(parents=True, exist_ok=True) + (director_path / 'logs').mkdir(parents=True, exist_ok=True) + shutil.copyfile(WORKSPACE / 'default/director.yaml', director_path / 'director.yaml') diff --git a/openfl/interface/envoy.py b/openfl/interface/envoy.py new file mode 100644 index 0000000000..abff25356c --- /dev/null +++ b/openfl/interface/envoy.py @@ -0,0 +1,102 @@ +# Copyright (C) 2020-2021 Intel Corporation +# SPDX-License-Identifier: Apache-2.0 +"""Envoy CLI.""" + +import logging +import shutil +import sys +from importlib import import_module +from pathlib import Path + +import click +from click import group +from click import option +from click import pass_context +from click import Path as ClickPath +from yaml import safe_load + +from openfl.component.envoy.envoy import Envoy +from openfl.interface.cli_helper import WORKSPACE + +logger = logging.getLogger(__name__) + + +@group() +@pass_context +def envoy(context): + """Manage Federated Learning Envoy.""" + context.obj['group'] = 'envoy' + + [email protected](name='start') +@option('-n', '--shard-name', required=True, + help='Current shard name') +@option('-d', '--director-uri', required=True, + help='The FQDN of the federation director') +@option('--disable-tls', default=False, + is_flag=True) +@option('-sc', '--shard-config-path', default='shard_config.yaml', + help='The shard config path', type=ClickPath(exists=True)) +@option('-rc', '--root-cert-path', 'root_ca', default=None, + help='Path to a root CA cert') +@option('-pk', '--private-key-path', 'key', default=None, + help='Path to a private key') +@option('-oc', '--public-cert-path', 'cert', default=None, + help='Path to a signed certificate') +def start_(shard_name, director_uri, disable_tls, shard_config_path, + root_ca, key, cert): + """Start the Envoy.""" + logger.info('🧿 Starting the Envoy.') + + shard_descriptor = shard_descriptor_from_config(shard_config_path) + envoy = Envoy( + shard_name=shard_name, + director_uri=director_uri, + shard_descriptor=shard_descriptor, + disable_tls=disable_tls, + root_ca=root_ca, + key=key, + cert=cert + ) + + envoy.start() + + [email protected](name='create-workspace') +@option('-p', '--envoy-path', required=True, + help='The Envoy path', type=ClickPath()) +def create(envoy_path): + """Create an envoy workspace.""" + envoy_path = Path(envoy_path) + if envoy_path.exists(): + if not click.confirm('Envoy workspace already exists. Recreate?', + default=True): + sys.exit(1) + shutil.rmtree(envoy_path) + (envoy_path / 'cert').mkdir(parents=True, exist_ok=True) + (envoy_path / 'logs').mkdir(parents=True, exist_ok=True) + (envoy_path / 'data').mkdir(parents=True, exist_ok=True) + shutil.copyfile(WORKSPACE / 'default/shard_config.yaml', + envoy_path / 'shard_config.yaml') + shutil.copyfile(WORKSPACE / 'default/shard_descriptor.py', + envoy_path / 'shard_descriptor.py') + shutil.copyfile(WORKSPACE / 'default/requirements.txt', + envoy_path / 'requirements.txt') + + +def shard_descriptor_from_config(shard_config_path: str): + """Build a shard descriptor from config.""" + with open(shard_config_path) as stream: + shard_config = safe_load(stream) + template = shard_config.get('template') + if not template: + raise Exception(f'You should define a shard ' + f'descriptor template in {shard_config_path}') + class_name = template.split('.')[-1] + module_path = '.'.join(template.split('.')[:-1]) + params = shard_config.get('params', {}) + + module = import_module(module_path) + instance = getattr(module, class_name)(**params) + + return instance diff --git a/openfl/interface/interactive_api/experiment.py b/openfl/interface/interactive_api/experiment.py index 0899ac7af7..3baabcd2db 100644 --- a/openfl/interface/interactive_api/experiment.py +++ b/openfl/interface/interactive_api/experiment.py @@ -3,14 +3,17 @@ """Python low-level API module.""" import functools -import logging import os +import time from collections import defaultdict from copy import deepcopy from logging import getLogger from pathlib import Path +from tensorboardX import SummaryWriter + from openfl.federated import Plan +from openfl.interface.cli import setup_logging from openfl.interface.cli_helper import WORKSPACE from openfl.utilities import split_tensor_dict_for_holdouts @@ -18,7 +21,13 @@ class FLExperiment: """Central class for FL experiment orchestration.""" - def __init__(self, federation, serializer_plugin=None) -> None: + def __init__( + self, + federation, + experiment_name: str = None, + serializer_plugin: str = 'openfl.plugins.interface_serializer.' + 'cloudpickle_serializer.CloudpickleSerializer' + ) -> None: """ Initialize an experiment inside a federation. @@ -26,22 +35,90 @@ def __init__(self, federation, serializer_plugin=None) -> None: Information about the data on collaborators is contained on the federation level. """ self.federation = federation + self.experiment_name = experiment_name or 'test-' + time.strftime('%Y%m%d-%H%M%S') + self.summary_writer = None + self.serializer_plugin = serializer_plugin - if serializer_plugin is None: - self.serializer_plugin = \ - 'openfl.plugins.interface_serializer.cloudpickle_serializer.CloudpickleSerializer' - else: - self.serializer_plugin = serializer_plugin + self.experiment_accepted = False self.logger = getLogger(__name__) + setup_logging() + + def _assert_experiment_accepted(self): + """Assure experiment is sent to director.""" + if not self.experiment_accepted: + self.logger.error('The experimnet has not been accepted by director') + self.logger.error( + 'Report the experiment first: ' + 'use the Experiment.start() method.') + raise Exception def get_best_model(self): """Retrieve the model with the best score.""" - # Next line relies on aggregator inner field where model dicts are stored - best_tensor_dict = self.server.aggregator.best_tensor_dict - self.task_runner_stub.rebuild_model(best_tensor_dict, validation=True, device='cpu') + self._assert_experiment_accepted() + tensor_dict = self.federation.dir_client.get_best_model( + experiment_name=self.experiment_name) + + return self._rebuild_model(tensor_dict) + + def get_last_model(self): + """Retrieve the aggregated model after the last round.""" + self._assert_experiment_accepted() + tensor_dict = self.federation.dir_client.get_last_model( + experiment_name=self.experiment_name) + + return self._rebuild_model(tensor_dict) + + def _rebuild_model(self, tensor_dict): + """Use tensor dict to update model weights.""" + if len(tensor_dict) == 0: + self.logger.error('No tensors received from director') + self.logger.error( + 'Possible reasons:\n' + '1. Aggregated model is not ready \n' + '2. Experiment data removed from director' + ) + else: + self.task_runner_stub.rebuild_model(tensor_dict, validation=True, device='cpu') + return self.task_runner_stub.model + def stream_metrics(self, tensorboard_logs: bool = True) -> None: + """Stream metrics.""" + self._assert_experiment_accepted() + for metric_message_dict in self.federation.dir_client.stream_metrics(self.experiment_name): + self.logger.metric( + f'Round {metric_message_dict["round"]}, ' + f'collaborator {metric_message_dict["metric_origin"]} ' + f'{metric_message_dict["task_name"]} result ' + f'{metric_message_dict["metric_name"]}:\t{metric_message_dict["metric_value"]}') + + if tensorboard_logs: + self.write_tensorboard_metric(metric_message_dict) + + def write_tensorboard_metric(self, metric: dict) -> None: + """Write metric callback.""" + if not self.summary_writer: + self.summary_writer = SummaryWriter(f'./logs/{self.experiment_name}', flush_secs=5) + + self.summary_writer.add_scalar( + f'{metric["metric_origin"]}/{metric["task_name"]}/{metric["metric_name"]}', + metric['metric_value'], metric['round']) + + def remove_experiment_data(self): + """Remove experiment data.""" + self._assert_experiment_accepted() + log_message = 'Removing experiment data ' + if self.federation.dir_client.remove_experiment_data( + name=self.experiment_name + ): + log_message += 'succeed.' + self.experiment_accepted = False + else: + log_message += 'failed.' + + self.logger.info(log_message) + def prepare_workspace_distribution( self, model_provider, task_keeper, data_loader, rounds_to_train, @@ -64,41 +141,55 @@ def prepare_workspace_distribution( self._export_python_env() # Compress te workspace to restore it on collaborator - self._pack_the_workspace() + self.arch_path = self._pack_the_workspace() # DO CERTIFICATES exchange - def start_experiment(self, model_provider): - """ - Start the aggregator. - - This method requires model_provider to start an experiment with another - model initialization without workspace redistribution. - """ - # Start the aggregator + def start(self, *, model_provider, task_keeper, data_loader, + rounds_to_train, delta_updates=False, opt_treatment='RESET'): + """Prepare experiment and run.""" + self.prepare_workspace_distribution( + model_provider, task_keeper, data_loader, rounds_to_train, + delta_updates=delta_updates, opt_treatment=opt_treatment + ) + self.logger.info('Starting experiment!') self.plan.resolve() - initial_tensor_dict = self._get_initial_tensor_dict(model_provider) - self.server = self.plan.interactive_api_get_server( - initial_tensor_dict, - chain=self.federation.cert_chain, - certificate=self.federation.agg_certificate, - private_key=self.federation.agg_private_key) + try: + response = self.federation.dir_client.set_new_experiment( + name=self.experiment_name, + col_names=self.plan.authorized_cols, + arch_path=self.arch_path, + initial_tensor_dict=initial_tensor_dict + ) + finally: + self.remove_workspace_archive() + + if response.accepted: + self.logger.info('Experiment was accepted and launched.') + self.experiment_accepted = True + else: + self.logger.info('Experiment was not accepted or failed.') - logging.basicConfig(level=logging.INFO) - self.server.serve() + def restore_experiment_state(self, model_provider): + """Restore accepted experimnet object.""" + self.task_runner_stub = self.plan.get_core_task_runner(model_provider=model_provider) + self.experiment_accepted = True @staticmethod def _export_python_env(): """Prepare requirements.txt.""" from pip._internal.operations import freeze requirements_generator = freeze.freeze() + + def is_package_has_version(package: str) -> bool: + return '==' in package and package != 'pkg-resources==0.0.0' + with open('./requirements.txt', 'w') as f: - for package in requirements_generator: - if '==' not in package: - # We do not export dependencies without version + for pack in requirements_generator: + if is_package_has_version(pack): continue - f.write(package + '\n') + f.write(pack + '\n') @staticmethod def _pack_the_workspace(): @@ -115,7 +206,7 @@ def _pack_the_workspace(): archive_name = basename(getcwd()) tmp_dir = 'temp_' + archive_name - makedirs(tmp_dir) + makedirs(tmp_dir, exist_ok=True) ignore = ignore_patterns( '__pycache__', 'data', 'cert', tmp_dir, '*.crt', '*.key', @@ -123,10 +214,17 @@ def _pack_the_workspace(): copytree('./', tmp_dir + '/workspace', ignore=ignore) - make_archive(archive_name, archive_type, tmp_dir + '/workspace') + arch_path = make_archive(archive_name, archive_type, tmp_dir + '/workspace') rmtree(tmp_dir) + return arch_path + + def remove_workspace_archive(self): + """Remove the workspace archive.""" + os.remove(self.arch_path) + del self.arch_path + def _get_initial_tensor_dict(self, model_provider): """Extract initial weights from the model.""" self.task_runner_stub = self.plan.get_core_task_runner(model_provider=model_provider) @@ -152,9 +250,20 @@ def _prepare_plan(self, model_provider, task_keeper, data_loader, # Change plan name to default one plan.name = 'plan.yaml' - plan.authorized_cols = list(self.federation.col_data_paths.keys()) + # Seems like we still need to fill authorized_cols list + # So aggregator know when to start sending tasks + # We also could change the aggregator logic so it will send tasks to aggregator + # as soon as it connects. This change should be a part of a bigger PR + # brining in fault tolerance changes + shard_registry = self.federation.get_shard_registry() + plan.authorized_cols = [ + name for name, info in shard_registry.items() if info['is_online'] + ] # Network part of the plan - plan.config['network']['settings']['agg_addr'] = self.federation.fqdn + # We keep in mind that an aggregator FQND will be the same as the directors FQDN + # We just choose a port randomly from plan hash + plan.config['network']['settings']['agg_addr'] = \ + self.federation.director_node_fqdn.split(':')[0] # We drop the port plan.config['network']['settings']['disable_tls'] = self.federation.disable_tls # Aggregator part of the plan @@ -356,18 +465,24 @@ class DataInterface: at initialization time for dataloader customization """ - def __init__(self, user_dataset_class, **kwargs): + def __init__(self, **kwargs): """Initialize DataLoader.""" - self.UserDatasetClass = user_dataset_class self.kwargs = kwargs - def _delayed_init(self, data_path): + @property + def shard_descriptor(self): + """Return shard descriptor.""" + return self._shard_descriptor + + @shard_descriptor.setter + def shard_descriptor(self, shard_descriptor): """ Describe per-collaborator procedures or sharding. This method will be called during a collaborator initialization. - data_path variable will be set according to data.yaml. + Local shard_descriptor will be set by Envoy. """ + self._shard_descriptor = shard_descriptor raise NotImplementedError def get_train_loader(self, **kwargs): diff --git a/openfl/interface/interactive_api/federation.py b/openfl/interface/interactive_api/federation.py index 863ab41de0..910c882cf1 100644 --- a/openfl/interface/interactive_api/federation.py +++ b/openfl/interface/interactive_api/federation.py @@ -2,7 +2,9 @@ # SPDX-License-Identifier: Apache-2.0 """Federation API module.""" +from openfl.transport.grpc.director_client import DirectorClient from openfl.utilities.utils import getfqdn_env +from .shard_descriptor import DummyShardDescriptor class Federation: @@ -13,37 +15,46 @@ class Federation: their local data and network setting to enable communication in federation. """ - def __init__(self, central_node_fqdn=None, disable_tls=False, - cert_chain=None, agg_certificate=None, agg_private_key=None) -> None: + def __init__(self, client_id, director_node_fqdn=None, director_port=None, disable_tls=False, + cert_chain=None, api_cert=None, api_private_key=None) -> None: """ Initialize federation. - Federation API class should be initialized with the aggregator node FQDN + Federation API class should be initialized with the Director node FQDN and encryption settings. One may disable mTLS in trusted environments or - provide paths to a certificate chain to CA, aggregator certificate and + provide paths to a certificate chain to CA, API certificate and pricate key to enable mTLS. + + Args: + - client_id: name of created Frontend API instance. + The same name user certify. + - director_node_fqdn: Address and port a director's service is running on. + User passes here an address with a port. """ - if central_node_fqdn is None: - self.fqdn = getfqdn_env() + self.client_id = client_id + if director_node_fqdn is None: + self.director_node_fqdn = getfqdn_env() else: - self.fqdn = central_node_fqdn + self.director_node_fqdn = director_node_fqdn self.disable_tls = disable_tls self.cert_chain = cert_chain - self.agg_certificate = agg_certificate - self.agg_private_key = agg_private_key + self.api_cert = api_cert + self.api_private_key = api_private_key - def register_collaborators(self, col_data_paths: dict) -> None: - """ - Provide data to be stored in data.yaml. + # Create Director client + self.dir_client = DirectorClient(client_id, f'{director_node_fqdn}:{director_port}', + disable_tls, cert_chain, api_private_key, api_cert) - This method should be used to provide information about collaborators - participating federation. - Arguments: - col_data_paths: dict(collaborator name : local data path) - """ - self.col_data_paths = col_data_paths - with open('./data.yaml', 'w') as f: - for col_name, data_path in self.col_data_paths.items(): - f.write(f'{col_name},{data_path}\n') + # Request sample and target shapes from Director. + # This is an internal method for finding out dataset properties in a Federation. + self.sample_shape, self.target_shape = self.dir_client.get_dataset_info() + + def get_dummy_shard_descriptor(self, size): + """Return a dummy shard descriptor.""" + return DummyShardDescriptor(self.sample_shape, self.target_shape, size) + + def get_shard_registry(self): + """Return a shard registry.""" + return self.dir_client.get_envoys() diff --git a/openfl/interface/interactive_api/shard_descriptor.py b/openfl/interface/interactive_api/shard_descriptor.py new file mode 100644 index 0000000000..a7413be10e --- /dev/null +++ b/openfl/interface/interactive_api/shard_descriptor.py @@ -0,0 +1,68 @@ +# Copyright (C) 2020-2021 Intel Corporation +# SPDX-License-Identifier: Apache-2.0 +"""Shard descriptor.""" + + +import numpy as np + + +class ShardDescriptor: + """Shard descriptor class.""" + + def __len__(self): + """Return the len of the dataset.""" + raise NotImplementedError + + def __getitem__(self, index: int): + """Return a item by the index.""" + raise NotImplementedError + + @property + def sample_shape(self): + """Return the sample shape info.""" + raise NotImplementedError + + @property + def target_shape(self): + """Return the target shape info.""" + raise NotImplementedError + + @property + def dataset_description(self) -> str: + """Return the dataset description.""" + return '' + + +class DummyShardDescriptor(ShardDescriptor): + """Dummy shard descriptor class.""" + + def __init__(self, sample_shape, target_shape, size) -> None: + """Initialize DummyShardDescriptor.""" + self._sample_shape = [int(dim) for dim in sample_shape] + self._target_shape = [int(dim) for dim in target_shape] + self.size = size + self.samples = np.random.randint(0, 255, (self.size, *self.sample_shape), np.uint8) + self.targets = np.random.randint(0, 255, (self.size, *self.target_shape), np.uint8) + + def __len__(self): + """Return the len of the dataset.""" + return self.size + + def __getitem__(self, index: int): + """Return a item by the index.""" + return self.samples[index], self.targets[index] + + @property + def sample_shape(self): + """Return the sample shape info.""" + return self._sample_shape + + @property + def target_shape(self): + """Return the target shape info.""" + return self._target_shape + + @property + def dataset_description(self) -> str: + """Return the dataset description.""" + return 'Dummy shard descriptor' diff --git a/openfl/interface/pki.py b/openfl/interface/pki.py new file mode 100644 index 0000000000..2f35f83b11 --- /dev/null +++ b/openfl/interface/pki.py @@ -0,0 +1,103 @@ +# Copyright (C) 2020-2021 Intel Corporation +# SPDX-License-Identifier: Apache-2.0 +"""PKI CLI.""" + +import logging +import os +from pathlib import Path + +from click import group +from click import option +from click import pass_context +from click import Path as ClickPath + +from openfl.component.ca.ca import certify +from openfl.component.ca.ca import get_ca_bin_paths +from openfl.component.ca.ca import get_token +from openfl.component.ca.ca import install +from openfl.component.ca.ca import remove_ca +from openfl.component.ca.ca import run_ca + + +logger = logging.getLogger(__name__) + +CA_URL = 'localhost:9123' + + +@group() +@pass_context +def pki(context): + """Manage Step-ca PKI.""" + context.obj['group'] = 'pki' + + [email protected](name='run') +@option('-p', '--ca-path', required=True, + help='The ca path', type=ClickPath()) +def run(ca_path): + """Run CA server.""" + ca_path = Path(ca_path) + step_config_dir = ca_path / 'step_config' + pki_dir = ca_path / 'cert' + pass_file = pki_dir / 'pass_file' + ca_json = step_config_dir / 'config' / 'ca.json' + _, step_ca_path = get_ca_bin_paths(ca_path) + if (not os.path.exists(step_config_dir) or not os.path.exists(pki_dir) + or not os.path.exists(pass_file) or not os.path.exists(ca_json) + or not os.path.exists(step_ca_path)): + logger.warning('CA is not installed or corrupted, please install it first') + return + run_ca(step_ca_path, pass_file, ca_json) + + [email protected](name='install') +@option('-p', '--ca-path', required=True, + help='The ca path', type=ClickPath()) +@option('--password', required=True) +@option('--ca-url', required=False, default=CA_URL) +def install_(ca_path, password, ca_url): + """Create a ca workspace.""" + install(ca_path, ca_url, password) + + [email protected](name='uninstall') +@option('-p', '--ca-path', required=True, + help='The CA path', type=ClickPath()) +def uninstall(ca_path): + """Remove step-CA.""" + remove_ca(ca_path) + + [email protected](name='get-token') +@option('-n', '--name', required=True) +@option('--ca-url', required=False, default=CA_URL) +@option('-p', '--ca-path', default='.', + help='The CA path', type=ClickPath()) +def get_token_(name, ca_url, ca_path): + """ + Create authentication token. + + Args: + name: common name for following certificate + (aggregator fqdn or collaborator name) + ca_url: full url of CA server + ca_path: the path to CA binaries + """ + token = get_token(name, ca_url, ca_path) + print('Token:') + print(token) + + [email protected](name='certify') +@option('-n', '--name', required=True) +@option('-t', '--token', 'token_with_cert', required=True) +@option('-c', '--certs-path', required=False, default=Path('.') / 'cert', + help='The path where certificates will be stored', type=ClickPath()) +@option('-p', '--ca-path', default='.', help='The path to CA client', + type=ClickPath(), required=False) +def certify_(name, token_with_cert, certs_path, ca_path): + """Create an envoy workspace.""" + certs_path = Path(certs_path) + ca_path = Path(ca_path) + certs_path.mkdir(parents=True, exist_ok=True) + certify(name, certs_path, token_with_cert, ca_path) diff --git a/openfl/native/native.py b/openfl/native/native.py index e9c810f935..885cb4bafc 100644 --- a/openfl/native/native.py +++ b/openfl/native/native.py @@ -5,6 +5,7 @@ This file defines openfl entrypoints to be used directly through python (not CLI) """ +import logging import os from copy import copy from logging import getLogger @@ -17,6 +18,7 @@ import openfl.interface.workspace as workspace from openfl.federated import Plan from openfl.protocols import utils +from openfl.utilities import add_log_level from openfl.utilities import split_tensor_dict_for_holdouts logger = getLogger(__name__) @@ -39,12 +41,13 @@ def setup_plan(log_level='CRITICAL'): cols_config = 'plan/cols.yaml' data_config = 'plan/data.yaml' + current_level = logging.root.level getLogger().setLevel(log_level) plan = Plan.parse(plan_config_path=Path(plan_config), cols_config_path=Path(cols_config), data_config_path=Path(data_config), resolve=False) - getLogger().setLevel('INFO') + getLogger().setLevel(current_level) return plan @@ -107,7 +110,7 @@ def unflatten(config, separator='.'): return config -def setup_logging(): +def setup_logging(level='INFO', log_file=None): """Initialize logging settings.""" # Setup logging from logging import basicConfig @@ -117,16 +120,29 @@ def setup_logging(): if True if pkgutil.find_loader('tensorflow') else False: import tensorflow as tf tf.compat.v1.logging.set_verbosity(tf.compat.v1.logging.ERROR) + metric = 25 + add_log_level('METRIC', metric) + + if isinstance(level, str): + level = level.upper() + + handlers = [] + if log_file: + fh = logging.FileHandler(log_file) + formatter = logging.Formatter( + '%(asctime)s %(levelname)s %(message)s %(filename)s:%(lineno)d' + ) + fh.setFormatter(formatter) + handlers.append(fh) + console = Console(width=160) - basicConfig( - level='INFO', - format='%(message)s', - datefmt='[%X]', - handlers=[RichHandler(console=console)] - ) + handlers.append(RichHandler(console=console)) + basicConfig(level=level, format='%(message)s', + datefmt='[%X]', handlers=handlers) -def init(workspace_template: str = 'default', agg_fqdn: str = None, col_names=None): +def init(workspace_template: str = 'default', log_level: str = 'INFO', + log_file: str = None, agg_fqdn: str = None, col_names=None): """ Initialize the openfl package. @@ -149,6 +165,10 @@ def init(workspace_template: str = 'default', agg_fqdn: str = None, col_names=No Other options include are any of the template names [ keras_cnn_mnist, tf_2dunet, tf_cnn_histology, mtorch_cnn_histology, torch_cnn_mnist] + log_level : str + Log level for logging. METRIC level is available + log_file : str + Name of the file in which the log will be duplicated agg_fqdn : str The local node's fully qualified domain name (if it can't be resolved automatically) @@ -174,7 +194,7 @@ def init(workspace_template: str = 'default', agg_fqdn: str = None, col_names=No collaborator.certify(col_name, silent=True) data_path += 1 - setup_logging() + setup_logging(level=log_level, log_file=log_file) def create_collaborator(plan, name, model, aggregator): diff --git a/openfl/protocols/director.proto b/openfl/protocols/director.proto new file mode 100644 index 0000000000..9d9156b873 --- /dev/null +++ b/openfl/protocols/director.proto @@ -0,0 +1,148 @@ +// Copyright (C) 2020 Intel Corporation +// Licensed subject to the terms of the separately executed evaluation license agreement between Intel Corporation and you. +syntax = "proto3"; + +import "google/protobuf/timestamp.proto"; +import "google/protobuf/duration.proto"; +import "federation.proto"; + + +message RequestHeader { + string sender = 1; +} + +// Envoy Messages + +message NodeInfo { + string name = 1; + string adress = 2; + bool cuda_available = 3; + uint32 memory_size = 4; +} + +message ShardInfo { + NodeInfo node_info = 1; + string shard_description = 2; + uint64 n_samples = 3; + // We just pass numpy shapes + repeated string sample_shape = 4; + repeated string target_shape = 5; +} + +message ShardAcknowledgement { + bool accepted = 1; +} + +message WaitExperimentRequest { + string collaborator_name = 1; +} + +message WaitExperimentResponse { + string experiment_name = 1; +} + +message GetExperimentDataRequest { + string experiment_name = 1; + string collaborator_name = 2; +} + +message ExperimentData { + uint32 size = 1; // size, in bytes, of the data sent in npbytes + bytes npbytes = 2; //actual data +} + +// API Messages + +message ExperimentInfo { + RequestHeader header = 1; + string name = 2; + repeated string collaborator_names = 3; + ExperimentData experiment_data = 4; + ModelProto model_proto = 5; +} + +message SetNewExperimentResponse{ + bool accepted = 1; + string tensorboard_address = 2; +} + +message GetTrainedModelRequest { + enum ModelType { + BEST_MODEL = 0; + LAST_MODEL = 1; + } + RequestHeader header = 1; + string experiment_name = 2; + ModelType model_type = 3; +} + +message TrainedModelResponse { + ModelProto model_proto = 1; +} + +message GetDatasetInfoRequest { + RequestHeader header = 1; +} + +message StreamMetricsRequest { + RequestHeader header = 1; + string experiment_name = 2; +} + +message StreamMetricsResponse { + string metric_origin = 1; + string task_name = 2; + string metric_name = 3; + float metric_value = 4; + uint32 round = 5; +} + +message RemoveExperimentRequest { + RequestHeader header = 1; + string experiment_name = 2; +} + +message RemoveExperimentResponse { + bool acknowledgement = 1; +} + +message CollaboratorStatus { + string name = 1; + bool is_experiment_running = 2; + google.protobuf.Duration valid_duration = 3; +} + +message CollaboratorHealthCheckResponse { + bool accepted = 1; +} + +message EnvoyInfo { + ShardInfo shard_info = 1; + bool is_online = 2; + bool is_experiment_running = 3; + google.protobuf.Timestamp last_updated = 4; + google.protobuf.Duration valid_duration = 5; +} + +message GetEnvoysRequest {} + +message GetEnvoysResponse { + repeated EnvoyInfo envoy_infos = 1; +} + +service FederationDirector { + // Envoy RPCs + rpc AcknowledgeShard (ShardInfo) returns (ShardAcknowledgement) {} + // Shard owner could also provide some public data for tests + rpc WaitExperiment (stream WaitExperimentRequest) returns (stream WaitExperimentResponse) {} + rpc GetExperimentData (GetExperimentDataRequest) returns (stream ExperimentData) {} + + // API RPCs + rpc SetNewExperiment (stream ExperimentInfo) returns (SetNewExperimentResponse) {} + rpc GetDatasetInfo (GetDatasetInfoRequest) returns (ShardInfo) {} + rpc GetTrainedModel (GetTrainedModelRequest) returns (TrainedModelResponse) {} + rpc StreamMetrics (StreamMetricsRequest) returns (stream StreamMetricsResponse) {} + rpc RemoveExperimentData (RemoveExperimentRequest) returns (RemoveExperimentResponse) {} + rpc CollaboratorHealthCheck (CollaboratorStatus) returns (CollaboratorHealthCheckResponse) {} + rpc GetEnvoys (GetEnvoysRequest) returns (GetEnvoysResponse) {} +} diff --git a/openfl/protocols/director_pb2.py b/openfl/protocols/director_pb2.py new file mode 100644 index 0000000000..adbaaee0f0 --- /dev/null +++ b/openfl/protocols/director_pb2.py @@ -0,0 +1,1261 @@ +# -*- coding: utf-8 -*- +# Generated by the protocol buffer compiler. DO NOT EDIT! +# source: director.proto +"""Generated protocol buffer code.""" +from google.protobuf import descriptor as _descriptor +from google.protobuf import message as _message +from google.protobuf import reflection as _reflection +from google.protobuf import symbol_database as _symbol_database +# @@protoc_insertion_point(imports) + +_sym_db = _symbol_database.Default() + + +from google.protobuf import timestamp_pb2 as google_dot_protobuf_dot_timestamp__pb2 +from google.protobuf import duration_pb2 as google_dot_protobuf_dot_duration__pb2 +import openfl.protocols.federation_pb2 as federation__pb2 + + +DESCRIPTOR = _descriptor.FileDescriptor( + name='director.proto', + package='', + syntax='proto3', + serialized_options=None, + create_key=_descriptor._internal_create_key, + serialized_pb=b'\n\x0e\x64irector.proto\x1a\x1fgoogle/protobuf/timestamp.proto\x1a\x1egoogle/protobuf/duration.proto\x1a\x10\x66\x65\x64\x65ration.proto\"\x1f\n\rRequestHeader\x12\x0e\n\x06sender\x18\x01 \x01(\t\"U\n\x08NodeInfo\x12\x0c\n\x04name\x18\x01 \x01(\t\x12\x0e\n\x06\x61\x64ress\x18\x02 \x01(\t\x12\x16\n\x0e\x63uda_available\x18\x03 \x01(\x08\x12\x13\n\x0bmemory_size\x18\x04 \x01(\r\"\x83\x01\n\tShardInfo\x12\x1c\n\tnode_info\x18\x01 \x01(\x0b\x32\t.NodeInfo\x12\x19\n\x11shard_description\x18\x02 \x01(\t\x12\x11\n\tn_samples\x18\x03 \x01(\x04\x12\x14\n\x0csample_shape\x18\x04 \x03(\t\x12\x14\n\x0ctarget_shape\x18\x05 \x03(\t\"(\n\x14ShardAcknowledgement\x12\x10\n\x08\x61\x63\x63\x65pted\x18\x01 \x01(\x08\"2\n\x15WaitExperimentRequest\x12\x19\n\x11\x63ollaborator_name\x18\x01 \x01(\t\"1\n\x16WaitExperimentResponse\x12\x17\n\x0f\x65xperiment_name\x18\x01 \x01(\t\"N\n\x18GetExperimentDataRequest\x12\x17\n\x0f\x65xperiment_name\x18\x01 \x01(\t\x12\x19\n\x11\x63ollaborator_name\x18\x02 \x01(\t\"/\n\x0e\x45xperimentData\x12\x0c\n\x04size\x18\x01 \x01(\r\x12\x0f\n\x07npbytes\x18\x02 \x01(\x0c\"\xa6\x01\n\x0e\x45xperimentInfo\x12\x1e\n\x06header\x18\x01 \x01(\x0b\x32\x0e.RequestHeader\x12\x0c\n\x04name\x18\x02 \x01(\t\x12\x1a\n\x12\x63ollaborator_names\x18\x03 \x03(\t\x12(\n\x0f\x65xperiment_data\x18\x04 \x01(\x0b\x32\x0f.ExperimentData\x12 \n\x0bmodel_proto\x18\x05 \x01(\x0b\x32\x0b.ModelProto\"I\n\x18SetNewExperimentResponse\x12\x10\n\x08\x61\x63\x63\x65pted\x18\x01 \x01(\x08\x12\x1b\n\x13tensorboard_address\x18\x02 \x01(\t\"\xb5\x01\n\x16GetTrainedModelRequest\x12\x1e\n\x06header\x18\x01 \x01(\x0b\x32\x0e.RequestHeader\x12\x17\n\x0f\x65xperiment_name\x18\x02 \x01(\t\x12\x35\n\nmodel_type\x18\x03 \x01(\x0e\x32!.GetTrainedModelRequest.ModelType\"+\n\tModelType\x12\x0e\n\nBEST_MODEL\x10\x00\x12\x0e\n\nLAST_MODEL\x10\x01\"8\n\x14TrainedModelResponse\x12 \n\x0bmodel_proto\x18\x01 \x01(\x0b\x32\x0b.ModelProto\"7\n\x15GetDatasetInfoRequest\x12\x1e\n\x06header\x18\x01 \x01(\x0b\x32\x0e.RequestHeader\"O\n\x14StreamMetricsRequest\x12\x1e\n\x06header\x18\x01 \x01(\x0b\x32\x0e.RequestHeader\x12\x17\n\x0f\x65xperiment_name\x18\x02 \x01(\t\"{\n\x15StreamMetricsResponse\x12\x15\n\rmetric_origin\x18\x01 \x01(\t\x12\x11\n\ttask_name\x18\x02 \x01(\t\x12\x13\n\x0bmetric_name\x18\x03 \x01(\t\x12\x14\n\x0cmetric_value\x18\x04 \x01(\x02\x12\r\n\x05round\x18\x05 \x01(\r\"R\n\x17RemoveExperimentRequest\x12\x1e\n\x06header\x18\x01 \x01(\x0b\x32\x0e.RequestHeader\x12\x17\n\x0f\x65xperiment_name\x18\x02 \x01(\t\"3\n\x18RemoveExperimentResponse\x12\x17\n\x0f\x61\x63knowledgement\x18\x01 \x01(\x08\"t\n\x12\x43ollaboratorStatus\x12\x0c\n\x04name\x18\x01 \x01(\t\x12\x1d\n\x15is_experiment_running\x18\x02 \x01(\x08\x12\x31\n\x0evalid_duration\x18\x03 \x01(\x0b\x32\x19.google.protobuf.Duration\"3\n\x1f\x43ollaboratorHealthCheckResponse\x12\x10\n\x08\x61\x63\x63\x65pted\x18\x01 \x01(\x08\"\xc2\x01\n\tEnvoyInfo\x12\x1e\n\nshard_info\x18\x01 \x01(\x0b\x32\n.ShardInfo\x12\x11\n\tis_online\x18\x02 \x01(\x08\x12\x1d\n\x15is_experiment_running\x18\x03 \x01(\x08\x12\x30\n\x0clast_updated\x18\x04 \x01(\x0b\x32\x1a.google.protobuf.Timestamp\x12\x31\n\x0evalid_duration\x18\x05 \x01(\x0b\x32\x19.google.protobuf.Duration\"\x12\n\x10GetEnvoysRequest\"4\n\x11GetEnvoysResponse\x12\x1f\n\x0b\x65nvoy_infos\x18\x01 \x03(\x0b\x32\n.EnvoyInfo2\xb9\x05\n\x12\x46\x65\x64\x65rationDirector\x12\x37\n\x10\x41\x63knowledgeShard\x12\n.ShardInfo\x1a\x15.ShardAcknowledgement\"\x00\x12G\n\x0eWaitExperiment\x12\x16.WaitExperimentRequest\x1a\x17.WaitExperimentResponse\"\x00(\x01\x30\x01\x12\x43\n\x11GetExperimentData\x12\x19.GetExperimentDataRequest\x1a\x0f.ExperimentData\"\x00\x30\x01\x12\x42\n\x10SetNewExperiment\x12\x0f.ExperimentInfo\x1a\x19.SetNewExperimentResponse\"\x00(\x01\x12\x36\n\x0eGetDatasetInfo\x12\x16.GetDatasetInfoRequest\x1a\n.ShardInfo\"\x00\x12\x43\n\x0fGetTrainedModel\x12\x17.GetTrainedModelRequest\x1a\x15.TrainedModelResponse\"\x00\x12\x42\n\rStreamMetrics\x12\x15.StreamMetricsRequest\x1a\x16.StreamMetricsResponse\"\x00\x30\x01\x12M\n\x14RemoveExperimentData\x12\x18.RemoveExperimentRequest\x1a\x19.RemoveExperimentResponse\"\x00\x12R\n\x17\x43ollaboratorHealthCheck\x12\x13.CollaboratorStatus\x1a .CollaboratorHealthCheckResponse\"\x00\x12\x34\n\tGetEnvoys\x12\x11.GetEnvoysRequest\x1a\x12.GetEnvoysResponse\"\x00\x62\x06proto3' + , + dependencies=[google_dot_protobuf_dot_timestamp__pb2.DESCRIPTOR,google_dot_protobuf_dot_duration__pb2.DESCRIPTOR,federation__pb2.DESCRIPTOR,]) + + + +_GETTRAINEDMODELREQUEST_MODELTYPE = _descriptor.EnumDescriptor( + name='ModelType', + full_name='GetTrainedModelRequest.ModelType', + filename=None, + file=DESCRIPTOR, + create_key=_descriptor._internal_create_key, + values=[ + _descriptor.EnumValueDescriptor( + name='BEST_MODEL', index=0, number=0, + serialized_options=None, + type=None, + create_key=_descriptor._internal_create_key), + _descriptor.EnumValueDescriptor( + name='LAST_MODEL', index=1, number=1, + serialized_options=None, + type=None, + create_key=_descriptor._internal_create_key), + ], + containing_type=None, + serialized_options=None, + serialized_start=1012, + serialized_end=1055, +) +_sym_db.RegisterEnumDescriptor(_GETTRAINEDMODELREQUEST_MODELTYPE) + + +_REQUESTHEADER = _descriptor.Descriptor( + name='RequestHeader', + full_name='RequestHeader', + filename=None, + file=DESCRIPTOR, + containing_type=None, + create_key=_descriptor._internal_create_key, + fields=[ + _descriptor.FieldDescriptor( + name='sender', full_name='RequestHeader.sender', index=0, + number=1, type=9, cpp_type=9, label=1, + has_default_value=False, default_value=b"".decode('utf-8'), + message_type=None, enum_type=None, containing_type=None, + is_extension=False, extension_scope=None, + serialized_options=None, file=DESCRIPTOR, create_key=_descriptor._internal_create_key), + ], + extensions=[ + ], + nested_types=[], + enum_types=[ + ], + serialized_options=None, + is_extendable=False, + syntax='proto3', + extension_ranges=[], + oneofs=[ + ], + serialized_start=101, + serialized_end=132, +) + + +_NODEINFO = _descriptor.Descriptor( + name='NodeInfo', + full_name='NodeInfo', + filename=None, + file=DESCRIPTOR, + containing_type=None, + create_key=_descriptor._internal_create_key, + fields=[ + _descriptor.FieldDescriptor( + name='name', full_name='NodeInfo.name', index=0, + number=1, type=9, cpp_type=9, label=1, + has_default_value=False, default_value=b"".decode('utf-8'), + message_type=None, enum_type=None, containing_type=None, + is_extension=False, extension_scope=None, + serialized_options=None, file=DESCRIPTOR, create_key=_descriptor._internal_create_key), + _descriptor.FieldDescriptor( + name='adress', full_name='NodeInfo.adress', index=1, + number=2, type=9, cpp_type=9, label=1, + has_default_value=False, default_value=b"".decode('utf-8'), + message_type=None, enum_type=None, containing_type=None, + is_extension=False, extension_scope=None, + serialized_options=None, file=DESCRIPTOR, create_key=_descriptor._internal_create_key), + _descriptor.FieldDescriptor( + name='cuda_available', full_name='NodeInfo.cuda_available', index=2, + number=3, type=8, cpp_type=7, label=1, + has_default_value=False, default_value=False, + message_type=None, enum_type=None, containing_type=None, + is_extension=False, extension_scope=None, + serialized_options=None, file=DESCRIPTOR, create_key=_descriptor._internal_create_key), + _descriptor.FieldDescriptor( + name='memory_size', full_name='NodeInfo.memory_size', index=3, + number=4, type=13, cpp_type=3, label=1, + has_default_value=False, default_value=0, + message_type=None, enum_type=None, containing_type=None, + is_extension=False, extension_scope=None, + serialized_options=None, file=DESCRIPTOR, create_key=_descriptor._internal_create_key), + ], + extensions=[ + ], + nested_types=[], + enum_types=[ + ], + serialized_options=None, + is_extendable=False, + syntax='proto3', + extension_ranges=[], + oneofs=[ + ], + serialized_start=134, + serialized_end=219, +) + + +_SHARDINFO = _descriptor.Descriptor( + name='ShardInfo', + full_name='ShardInfo', + filename=None, + file=DESCRIPTOR, + containing_type=None, + create_key=_descriptor._internal_create_key, + fields=[ + _descriptor.FieldDescriptor( + name='node_info', full_name='ShardInfo.node_info', index=0, + number=1, type=11, cpp_type=10, label=1, + has_default_value=False, default_value=None, + message_type=None, enum_type=None, containing_type=None, + is_extension=False, extension_scope=None, + serialized_options=None, file=DESCRIPTOR, create_key=_descriptor._internal_create_key), + _descriptor.FieldDescriptor( + name='shard_description', full_name='ShardInfo.shard_description', index=1, + number=2, type=9, cpp_type=9, label=1, + has_default_value=False, default_value=b"".decode('utf-8'), + message_type=None, enum_type=None, containing_type=None, + is_extension=False, extension_scope=None, + serialized_options=None, file=DESCRIPTOR, create_key=_descriptor._internal_create_key), + _descriptor.FieldDescriptor( + name='n_samples', full_name='ShardInfo.n_samples', index=2, + number=3, type=4, cpp_type=4, label=1, + has_default_value=False, default_value=0, + message_type=None, enum_type=None, containing_type=None, + is_extension=False, extension_scope=None, + serialized_options=None, file=DESCRIPTOR, create_key=_descriptor._internal_create_key), + _descriptor.FieldDescriptor( + name='sample_shape', full_name='ShardInfo.sample_shape', index=3, + number=4, type=9, cpp_type=9, label=3, + has_default_value=False, default_value=[], + message_type=None, enum_type=None, containing_type=None, + is_extension=False, extension_scope=None, + serialized_options=None, file=DESCRIPTOR, create_key=_descriptor._internal_create_key), + _descriptor.FieldDescriptor( + name='target_shape', full_name='ShardInfo.target_shape', index=4, + number=5, type=9, cpp_type=9, label=3, + has_default_value=False, default_value=[], + message_type=None, enum_type=None, containing_type=None, + is_extension=False, extension_scope=None, + serialized_options=None, file=DESCRIPTOR, create_key=_descriptor._internal_create_key), + ], + extensions=[ + ], + nested_types=[], + enum_types=[ + ], + serialized_options=None, + is_extendable=False, + syntax='proto3', + extension_ranges=[], + oneofs=[ + ], + serialized_start=222, + serialized_end=353, +) + + +_SHARDACKNOWLEDGEMENT = _descriptor.Descriptor( + name='ShardAcknowledgement', + full_name='ShardAcknowledgement', + filename=None, + file=DESCRIPTOR, + containing_type=None, + create_key=_descriptor._internal_create_key, + fields=[ + _descriptor.FieldDescriptor( + name='accepted', full_name='ShardAcknowledgement.accepted', index=0, + number=1, type=8, cpp_type=7, label=1, + has_default_value=False, default_value=False, + message_type=None, enum_type=None, containing_type=None, + is_extension=False, extension_scope=None, + serialized_options=None, file=DESCRIPTOR, create_key=_descriptor._internal_create_key), + ], + extensions=[ + ], + nested_types=[], + enum_types=[ + ], + serialized_options=None, + is_extendable=False, + syntax='proto3', + extension_ranges=[], + oneofs=[ + ], + serialized_start=355, + serialized_end=395, +) + + +_WAITEXPERIMENTREQUEST = _descriptor.Descriptor( + name='WaitExperimentRequest', + full_name='WaitExperimentRequest', + filename=None, + file=DESCRIPTOR, + containing_type=None, + create_key=_descriptor._internal_create_key, + fields=[ + _descriptor.FieldDescriptor( + name='collaborator_name', full_name='WaitExperimentRequest.collaborator_name', index=0, + number=1, type=9, cpp_type=9, label=1, + has_default_value=False, default_value=b"".decode('utf-8'), + message_type=None, enum_type=None, containing_type=None, + is_extension=False, extension_scope=None, + serialized_options=None, file=DESCRIPTOR, create_key=_descriptor._internal_create_key), + ], + extensions=[ + ], + nested_types=[], + enum_types=[ + ], + serialized_options=None, + is_extendable=False, + syntax='proto3', + extension_ranges=[], + oneofs=[ + ], + serialized_start=397, + serialized_end=447, +) + + +_WAITEXPERIMENTRESPONSE = _descriptor.Descriptor( + name='WaitExperimentResponse', + full_name='WaitExperimentResponse', + filename=None, + file=DESCRIPTOR, + containing_type=None, + create_key=_descriptor._internal_create_key, + fields=[ + _descriptor.FieldDescriptor( + name='experiment_name', full_name='WaitExperimentResponse.experiment_name', index=0, + number=1, type=9, cpp_type=9, label=1, + has_default_value=False, default_value=b"".decode('utf-8'), + message_type=None, enum_type=None, containing_type=None, + is_extension=False, extension_scope=None, + serialized_options=None, file=DESCRIPTOR, create_key=_descriptor._internal_create_key), + ], + extensions=[ + ], + nested_types=[], + enum_types=[ + ], + serialized_options=None, + is_extendable=False, + syntax='proto3', + extension_ranges=[], + oneofs=[ + ], + serialized_start=449, + serialized_end=498, +) + + +_GETEXPERIMENTDATAREQUEST = _descriptor.Descriptor( + name='GetExperimentDataRequest', + full_name='GetExperimentDataRequest', + filename=None, + file=DESCRIPTOR, + containing_type=None, + create_key=_descriptor._internal_create_key, + fields=[ + _descriptor.FieldDescriptor( + name='experiment_name', full_name='GetExperimentDataRequest.experiment_name', index=0, + number=1, type=9, cpp_type=9, label=1, + has_default_value=False, default_value=b"".decode('utf-8'), + message_type=None, enum_type=None, containing_type=None, + is_extension=False, extension_scope=None, + serialized_options=None, file=DESCRIPTOR, create_key=_descriptor._internal_create_key), + _descriptor.FieldDescriptor( + name='collaborator_name', full_name='GetExperimentDataRequest.collaborator_name', index=1, + number=2, type=9, cpp_type=9, label=1, + has_default_value=False, default_value=b"".decode('utf-8'), + message_type=None, enum_type=None, containing_type=None, + is_extension=False, extension_scope=None, + serialized_options=None, file=DESCRIPTOR, create_key=_descriptor._internal_create_key), + ], + extensions=[ + ], + nested_types=[], + enum_types=[ + ], + serialized_options=None, + is_extendable=False, + syntax='proto3', + extension_ranges=[], + oneofs=[ + ], + serialized_start=500, + serialized_end=578, +) + + +_EXPERIMENTDATA = _descriptor.Descriptor( + name='ExperimentData', + full_name='ExperimentData', + filename=None, + file=DESCRIPTOR, + containing_type=None, + create_key=_descriptor._internal_create_key, + fields=[ + _descriptor.FieldDescriptor( + name='size', full_name='ExperimentData.size', index=0, + number=1, type=13, cpp_type=3, label=1, + has_default_value=False, default_value=0, + message_type=None, enum_type=None, containing_type=None, + is_extension=False, extension_scope=None, + serialized_options=None, file=DESCRIPTOR, create_key=_descriptor._internal_create_key), + _descriptor.FieldDescriptor( + name='npbytes', full_name='ExperimentData.npbytes', index=1, + number=2, type=12, cpp_type=9, label=1, + has_default_value=False, default_value=b"", + message_type=None, enum_type=None, containing_type=None, + is_extension=False, extension_scope=None, + serialized_options=None, file=DESCRIPTOR, create_key=_descriptor._internal_create_key), + ], + extensions=[ + ], + nested_types=[], + enum_types=[ + ], + serialized_options=None, + is_extendable=False, + syntax='proto3', + extension_ranges=[], + oneofs=[ + ], + serialized_start=580, + serialized_end=627, +) + + +_EXPERIMENTINFO = _descriptor.Descriptor( + name='ExperimentInfo', + full_name='ExperimentInfo', + filename=None, + file=DESCRIPTOR, + containing_type=None, + create_key=_descriptor._internal_create_key, + fields=[ + _descriptor.FieldDescriptor( + name='header', full_name='ExperimentInfo.header', index=0, + number=1, type=11, cpp_type=10, label=1, + has_default_value=False, default_value=None, + message_type=None, enum_type=None, containing_type=None, + is_extension=False, extension_scope=None, + serialized_options=None, file=DESCRIPTOR, create_key=_descriptor._internal_create_key), + _descriptor.FieldDescriptor( + name='name', full_name='ExperimentInfo.name', index=1, + number=2, type=9, cpp_type=9, label=1, + has_default_value=False, default_value=b"".decode('utf-8'), + message_type=None, enum_type=None, containing_type=None, + is_extension=False, extension_scope=None, + serialized_options=None, file=DESCRIPTOR, create_key=_descriptor._internal_create_key), + _descriptor.FieldDescriptor( + name='collaborator_names', full_name='ExperimentInfo.collaborator_names', index=2, + number=3, type=9, cpp_type=9, label=3, + has_default_value=False, default_value=[], + message_type=None, enum_type=None, containing_type=None, + is_extension=False, extension_scope=None, + serialized_options=None, file=DESCRIPTOR, create_key=_descriptor._internal_create_key), + _descriptor.FieldDescriptor( + name='experiment_data', full_name='ExperimentInfo.experiment_data', index=3, + number=4, type=11, cpp_type=10, label=1, + has_default_value=False, default_value=None, + message_type=None, enum_type=None, containing_type=None, + is_extension=False, extension_scope=None, + serialized_options=None, file=DESCRIPTOR, create_key=_descriptor._internal_create_key), + _descriptor.FieldDescriptor( + name='model_proto', full_name='ExperimentInfo.model_proto', index=4, + number=5, type=11, cpp_type=10, label=1, + has_default_value=False, default_value=None, + message_type=None, enum_type=None, containing_type=None, + is_extension=False, extension_scope=None, + serialized_options=None, file=DESCRIPTOR, create_key=_descriptor._internal_create_key), + ], + extensions=[ + ], + nested_types=[], + enum_types=[ + ], + serialized_options=None, + is_extendable=False, + syntax='proto3', + extension_ranges=[], + oneofs=[ + ], + serialized_start=630, + serialized_end=796, +) + + +_SETNEWEXPERIMENTRESPONSE = _descriptor.Descriptor( + name='SetNewExperimentResponse', + full_name='SetNewExperimentResponse', + filename=None, + file=DESCRIPTOR, + containing_type=None, + create_key=_descriptor._internal_create_key, + fields=[ + _descriptor.FieldDescriptor( + name='accepted', full_name='SetNewExperimentResponse.accepted', index=0, + number=1, type=8, cpp_type=7, label=1, + has_default_value=False, default_value=False, + message_type=None, enum_type=None, containing_type=None, + is_extension=False, extension_scope=None, + serialized_options=None, file=DESCRIPTOR, create_key=_descriptor._internal_create_key), + _descriptor.FieldDescriptor( + name='tensorboard_address', full_name='SetNewExperimentResponse.tensorboard_address', index=1, + number=2, type=9, cpp_type=9, label=1, + has_default_value=False, default_value=b"".decode('utf-8'), + message_type=None, enum_type=None, containing_type=None, + is_extension=False, extension_scope=None, + serialized_options=None, file=DESCRIPTOR, create_key=_descriptor._internal_create_key), + ], + extensions=[ + ], + nested_types=[], + enum_types=[ + ], + serialized_options=None, + is_extendable=False, + syntax='proto3', + extension_ranges=[], + oneofs=[ + ], + serialized_start=798, + serialized_end=871, +) + + +_GETTRAINEDMODELREQUEST = _descriptor.Descriptor( + name='GetTrainedModelRequest', + full_name='GetTrainedModelRequest', + filename=None, + file=DESCRIPTOR, + containing_type=None, + create_key=_descriptor._internal_create_key, + fields=[ + _descriptor.FieldDescriptor( + name='header', full_name='GetTrainedModelRequest.header', index=0, + number=1, type=11, cpp_type=10, label=1, + has_default_value=False, default_value=None, + message_type=None, enum_type=None, containing_type=None, + is_extension=False, extension_scope=None, + serialized_options=None, file=DESCRIPTOR, create_key=_descriptor._internal_create_key), + _descriptor.FieldDescriptor( + name='experiment_name', full_name='GetTrainedModelRequest.experiment_name', index=1, + number=2, type=9, cpp_type=9, label=1, + has_default_value=False, default_value=b"".decode('utf-8'), + message_type=None, enum_type=None, containing_type=None, + is_extension=False, extension_scope=None, + serialized_options=None, file=DESCRIPTOR, create_key=_descriptor._internal_create_key), + _descriptor.FieldDescriptor( + name='model_type', full_name='GetTrainedModelRequest.model_type', index=2, + number=3, type=14, cpp_type=8, label=1, + has_default_value=False, default_value=0, + message_type=None, enum_type=None, containing_type=None, + is_extension=False, extension_scope=None, + serialized_options=None, file=DESCRIPTOR, create_key=_descriptor._internal_create_key), + ], + extensions=[ + ], + nested_types=[], + enum_types=[ + _GETTRAINEDMODELREQUEST_MODELTYPE, + ], + serialized_options=None, + is_extendable=False, + syntax='proto3', + extension_ranges=[], + oneofs=[ + ], + serialized_start=874, + serialized_end=1055, +) + + +_TRAINEDMODELRESPONSE = _descriptor.Descriptor( + name='TrainedModelResponse', + full_name='TrainedModelResponse', + filename=None, + file=DESCRIPTOR, + containing_type=None, + create_key=_descriptor._internal_create_key, + fields=[ + _descriptor.FieldDescriptor( + name='model_proto', full_name='TrainedModelResponse.model_proto', index=0, + number=1, type=11, cpp_type=10, label=1, + has_default_value=False, default_value=None, + message_type=None, enum_type=None, containing_type=None, + is_extension=False, extension_scope=None, + serialized_options=None, file=DESCRIPTOR, create_key=_descriptor._internal_create_key), + ], + extensions=[ + ], + nested_types=[], + enum_types=[ + ], + serialized_options=None, + is_extendable=False, + syntax='proto3', + extension_ranges=[], + oneofs=[ + ], + serialized_start=1057, + serialized_end=1113, +) + + +_GETDATASETINFOREQUEST = _descriptor.Descriptor( + name='GetDatasetInfoRequest', + full_name='GetDatasetInfoRequest', + filename=None, + file=DESCRIPTOR, + containing_type=None, + create_key=_descriptor._internal_create_key, + fields=[ + _descriptor.FieldDescriptor( + name='header', full_name='GetDatasetInfoRequest.header', index=0, + number=1, type=11, cpp_type=10, label=1, + has_default_value=False, default_value=None, + message_type=None, enum_type=None, containing_type=None, + is_extension=False, extension_scope=None, + serialized_options=None, file=DESCRIPTOR, create_key=_descriptor._internal_create_key), + ], + extensions=[ + ], + nested_types=[], + enum_types=[ + ], + serialized_options=None, + is_extendable=False, + syntax='proto3', + extension_ranges=[], + oneofs=[ + ], + serialized_start=1115, + serialized_end=1170, +) + + +_STREAMMETRICSREQUEST = _descriptor.Descriptor( + name='StreamMetricsRequest', + full_name='StreamMetricsRequest', + filename=None, + file=DESCRIPTOR, + containing_type=None, + create_key=_descriptor._internal_create_key, + fields=[ + _descriptor.FieldDescriptor( + name='header', full_name='StreamMetricsRequest.header', index=0, + number=1, type=11, cpp_type=10, label=1, + has_default_value=False, default_value=None, + message_type=None, enum_type=None, containing_type=None, + is_extension=False, extension_scope=None, + serialized_options=None, file=DESCRIPTOR, create_key=_descriptor._internal_create_key), + _descriptor.FieldDescriptor( + name='experiment_name', full_name='StreamMetricsRequest.experiment_name', index=1, + number=2, type=9, cpp_type=9, label=1, + has_default_value=False, default_value=b"".decode('utf-8'), + message_type=None, enum_type=None, containing_type=None, + is_extension=False, extension_scope=None, + serialized_options=None, file=DESCRIPTOR, create_key=_descriptor._internal_create_key), + ], + extensions=[ + ], + nested_types=[], + enum_types=[ + ], + serialized_options=None, + is_extendable=False, + syntax='proto3', + extension_ranges=[], + oneofs=[ + ], + serialized_start=1172, + serialized_end=1251, +) + + +_STREAMMETRICSRESPONSE = _descriptor.Descriptor( + name='StreamMetricsResponse', + full_name='StreamMetricsResponse', + filename=None, + file=DESCRIPTOR, + containing_type=None, + create_key=_descriptor._internal_create_key, + fields=[ + _descriptor.FieldDescriptor( + name='metric_origin', full_name='StreamMetricsResponse.metric_origin', index=0, + number=1, type=9, cpp_type=9, label=1, + has_default_value=False, default_value=b"".decode('utf-8'), + message_type=None, enum_type=None, containing_type=None, + is_extension=False, extension_scope=None, + serialized_options=None, file=DESCRIPTOR, create_key=_descriptor._internal_create_key), + _descriptor.FieldDescriptor( + name='task_name', full_name='StreamMetricsResponse.task_name', index=1, + number=2, type=9, cpp_type=9, label=1, + has_default_value=False, default_value=b"".decode('utf-8'), + message_type=None, enum_type=None, containing_type=None, + is_extension=False, extension_scope=None, + serialized_options=None, file=DESCRIPTOR, create_key=_descriptor._internal_create_key), + _descriptor.FieldDescriptor( + name='metric_name', full_name='StreamMetricsResponse.metric_name', index=2, + number=3, type=9, cpp_type=9, label=1, + has_default_value=False, default_value=b"".decode('utf-8'), + message_type=None, enum_type=None, containing_type=None, + is_extension=False, extension_scope=None, + serialized_options=None, file=DESCRIPTOR, create_key=_descriptor._internal_create_key), + _descriptor.FieldDescriptor( + name='metric_value', full_name='StreamMetricsResponse.metric_value', index=3, + number=4, type=2, cpp_type=6, label=1, + has_default_value=False, default_value=float(0), + message_type=None, enum_type=None, containing_type=None, + is_extension=False, extension_scope=None, + serialized_options=None, file=DESCRIPTOR, create_key=_descriptor._internal_create_key), + _descriptor.FieldDescriptor( + name='round', full_name='StreamMetricsResponse.round', index=4, + number=5, type=13, cpp_type=3, label=1, + has_default_value=False, default_value=0, + message_type=None, enum_type=None, containing_type=None, + is_extension=False, extension_scope=None, + serialized_options=None, file=DESCRIPTOR, create_key=_descriptor._internal_create_key), + ], + extensions=[ + ], + nested_types=[], + enum_types=[ + ], + serialized_options=None, + is_extendable=False, + syntax='proto3', + extension_ranges=[], + oneofs=[ + ], + serialized_start=1253, + serialized_end=1376, +) + + +_REMOVEEXPERIMENTREQUEST = _descriptor.Descriptor( + name='RemoveExperimentRequest', + full_name='RemoveExperimentRequest', + filename=None, + file=DESCRIPTOR, + containing_type=None, + create_key=_descriptor._internal_create_key, + fields=[ + _descriptor.FieldDescriptor( + name='header', full_name='RemoveExperimentRequest.header', index=0, + number=1, type=11, cpp_type=10, label=1, + has_default_value=False, default_value=None, + message_type=None, enum_type=None, containing_type=None, + is_extension=False, extension_scope=None, + serialized_options=None, file=DESCRIPTOR, create_key=_descriptor._internal_create_key), + _descriptor.FieldDescriptor( + name='experiment_name', full_name='RemoveExperimentRequest.experiment_name', index=1, + number=2, type=9, cpp_type=9, label=1, + has_default_value=False, default_value=b"".decode('utf-8'), + message_type=None, enum_type=None, containing_type=None, + is_extension=False, extension_scope=None, + serialized_options=None, file=DESCRIPTOR, create_key=_descriptor._internal_create_key), + ], + extensions=[ + ], + nested_types=[], + enum_types=[ + ], + serialized_options=None, + is_extendable=False, + syntax='proto3', + extension_ranges=[], + oneofs=[ + ], + serialized_start=1378, + serialized_end=1460, +) + + +_REMOVEEXPERIMENTRESPONSE = _descriptor.Descriptor( + name='RemoveExperimentResponse', + full_name='RemoveExperimentResponse', + filename=None, + file=DESCRIPTOR, + containing_type=None, + create_key=_descriptor._internal_create_key, + fields=[ + _descriptor.FieldDescriptor( + name='acknowledgement', full_name='RemoveExperimentResponse.acknowledgement', index=0, + number=1, type=8, cpp_type=7, label=1, + has_default_value=False, default_value=False, + message_type=None, enum_type=None, containing_type=None, + is_extension=False, extension_scope=None, + serialized_options=None, file=DESCRIPTOR, create_key=_descriptor._internal_create_key), + ], + extensions=[ + ], + nested_types=[], + enum_types=[ + ], + serialized_options=None, + is_extendable=False, + syntax='proto3', + extension_ranges=[], + oneofs=[ + ], + serialized_start=1462, + serialized_end=1513, +) + + +_COLLABORATORSTATUS = _descriptor.Descriptor( + name='CollaboratorStatus', + full_name='CollaboratorStatus', + filename=None, + file=DESCRIPTOR, + containing_type=None, + create_key=_descriptor._internal_create_key, + fields=[ + _descriptor.FieldDescriptor( + name='name', full_name='CollaboratorStatus.name', index=0, + number=1, type=9, cpp_type=9, label=1, + has_default_value=False, default_value=b"".decode('utf-8'), + message_type=None, enum_type=None, containing_type=None, + is_extension=False, extension_scope=None, + serialized_options=None, file=DESCRIPTOR, create_key=_descriptor._internal_create_key), + _descriptor.FieldDescriptor( + name='is_experiment_running', full_name='CollaboratorStatus.is_experiment_running', index=1, + number=2, type=8, cpp_type=7, label=1, + has_default_value=False, default_value=False, + message_type=None, enum_type=None, containing_type=None, + is_extension=False, extension_scope=None, + serialized_options=None, file=DESCRIPTOR, create_key=_descriptor._internal_create_key), + _descriptor.FieldDescriptor( + name='valid_duration', full_name='CollaboratorStatus.valid_duration', index=2, + number=3, type=11, cpp_type=10, label=1, + has_default_value=False, default_value=None, + message_type=None, enum_type=None, containing_type=None, + is_extension=False, extension_scope=None, + serialized_options=None, file=DESCRIPTOR, create_key=_descriptor._internal_create_key), + ], + extensions=[ + ], + nested_types=[], + enum_types=[ + ], + serialized_options=None, + is_extendable=False, + syntax='proto3', + extension_ranges=[], + oneofs=[ + ], + serialized_start=1515, + serialized_end=1631, +) + + +_COLLABORATORHEALTHCHECKRESPONSE = _descriptor.Descriptor( + name='CollaboratorHealthCheckResponse', + full_name='CollaboratorHealthCheckResponse', + filename=None, + file=DESCRIPTOR, + containing_type=None, + create_key=_descriptor._internal_create_key, + fields=[ + _descriptor.FieldDescriptor( + name='accepted', full_name='CollaboratorHealthCheckResponse.accepted', index=0, + number=1, type=8, cpp_type=7, label=1, + has_default_value=False, default_value=False, + message_type=None, enum_type=None, containing_type=None, + is_extension=False, extension_scope=None, + serialized_options=None, file=DESCRIPTOR, create_key=_descriptor._internal_create_key), + ], + extensions=[ + ], + nested_types=[], + enum_types=[ + ], + serialized_options=None, + is_extendable=False, + syntax='proto3', + extension_ranges=[], + oneofs=[ + ], + serialized_start=1633, + serialized_end=1684, +) + + +_ENVOYINFO = _descriptor.Descriptor( + name='EnvoyInfo', + full_name='EnvoyInfo', + filename=None, + file=DESCRIPTOR, + containing_type=None, + create_key=_descriptor._internal_create_key, + fields=[ + _descriptor.FieldDescriptor( + name='shard_info', full_name='EnvoyInfo.shard_info', index=0, + number=1, type=11, cpp_type=10, label=1, + has_default_value=False, default_value=None, + message_type=None, enum_type=None, containing_type=None, + is_extension=False, extension_scope=None, + serialized_options=None, file=DESCRIPTOR, create_key=_descriptor._internal_create_key), + _descriptor.FieldDescriptor( + name='is_online', full_name='EnvoyInfo.is_online', index=1, + number=2, type=8, cpp_type=7, label=1, + has_default_value=False, default_value=False, + message_type=None, enum_type=None, containing_type=None, + is_extension=False, extension_scope=None, + serialized_options=None, file=DESCRIPTOR, create_key=_descriptor._internal_create_key), + _descriptor.FieldDescriptor( + name='is_experiment_running', full_name='EnvoyInfo.is_experiment_running', index=2, + number=3, type=8, cpp_type=7, label=1, + has_default_value=False, default_value=False, + message_type=None, enum_type=None, containing_type=None, + is_extension=False, extension_scope=None, + serialized_options=None, file=DESCRIPTOR, create_key=_descriptor._internal_create_key), + _descriptor.FieldDescriptor( + name='last_updated', full_name='EnvoyInfo.last_updated', index=3, + number=4, type=11, cpp_type=10, label=1, + has_default_value=False, default_value=None, + message_type=None, enum_type=None, containing_type=None, + is_extension=False, extension_scope=None, + serialized_options=None, file=DESCRIPTOR, create_key=_descriptor._internal_create_key), + _descriptor.FieldDescriptor( + name='valid_duration', full_name='EnvoyInfo.valid_duration', index=4, + number=5, type=11, cpp_type=10, label=1, + has_default_value=False, default_value=None, + message_type=None, enum_type=None, containing_type=None, + is_extension=False, extension_scope=None, + serialized_options=None, file=DESCRIPTOR, create_key=_descriptor._internal_create_key), + ], + extensions=[ + ], + nested_types=[], + enum_types=[ + ], + serialized_options=None, + is_extendable=False, + syntax='proto3', + extension_ranges=[], + oneofs=[ + ], + serialized_start=1687, + serialized_end=1881, +) + + +_GETENVOYSREQUEST = _descriptor.Descriptor( + name='GetEnvoysRequest', + full_name='GetEnvoysRequest', + filename=None, + file=DESCRIPTOR, + containing_type=None, + create_key=_descriptor._internal_create_key, + fields=[ + ], + extensions=[ + ], + nested_types=[], + enum_types=[ + ], + serialized_options=None, + is_extendable=False, + syntax='proto3', + extension_ranges=[], + oneofs=[ + ], + serialized_start=1883, + serialized_end=1901, +) + + +_GETENVOYSRESPONSE = _descriptor.Descriptor( + name='GetEnvoysResponse', + full_name='GetEnvoysResponse', + filename=None, + file=DESCRIPTOR, + containing_type=None, + create_key=_descriptor._internal_create_key, + fields=[ + _descriptor.FieldDescriptor( + name='envoy_infos', full_name='GetEnvoysResponse.envoy_infos', index=0, + number=1, type=11, cpp_type=10, label=3, + has_default_value=False, default_value=[], + message_type=None, enum_type=None, containing_type=None, + is_extension=False, extension_scope=None, + serialized_options=None, file=DESCRIPTOR, create_key=_descriptor._internal_create_key), + ], + extensions=[ + ], + nested_types=[], + enum_types=[ + ], + serialized_options=None, + is_extendable=False, + syntax='proto3', + extension_ranges=[], + oneofs=[ + ], + serialized_start=1903, + serialized_end=1955, +) + +_SHARDINFO.fields_by_name['node_info'].message_type = _NODEINFO +_EXPERIMENTINFO.fields_by_name['header'].message_type = _REQUESTHEADER +_EXPERIMENTINFO.fields_by_name['experiment_data'].message_type = _EXPERIMENTDATA +_EXPERIMENTINFO.fields_by_name['model_proto'].message_type = federation__pb2._MODELPROTO +_GETTRAINEDMODELREQUEST.fields_by_name['header'].message_type = _REQUESTHEADER +_GETTRAINEDMODELREQUEST.fields_by_name['model_type'].enum_type = _GETTRAINEDMODELREQUEST_MODELTYPE +_GETTRAINEDMODELREQUEST_MODELTYPE.containing_type = _GETTRAINEDMODELREQUEST +_TRAINEDMODELRESPONSE.fields_by_name['model_proto'].message_type = federation__pb2._MODELPROTO +_GETDATASETINFOREQUEST.fields_by_name['header'].message_type = _REQUESTHEADER +_STREAMMETRICSREQUEST.fields_by_name['header'].message_type = _REQUESTHEADER +_REMOVEEXPERIMENTREQUEST.fields_by_name['header'].message_type = _REQUESTHEADER +_COLLABORATORSTATUS.fields_by_name['valid_duration'].message_type = google_dot_protobuf_dot_duration__pb2._DURATION +_ENVOYINFO.fields_by_name['shard_info'].message_type = _SHARDINFO +_ENVOYINFO.fields_by_name['last_updated'].message_type = google_dot_protobuf_dot_timestamp__pb2._TIMESTAMP +_ENVOYINFO.fields_by_name['valid_duration'].message_type = google_dot_protobuf_dot_duration__pb2._DURATION +_GETENVOYSRESPONSE.fields_by_name['envoy_infos'].message_type = _ENVOYINFO +DESCRIPTOR.message_types_by_name['RequestHeader'] = _REQUESTHEADER +DESCRIPTOR.message_types_by_name['NodeInfo'] = _NODEINFO +DESCRIPTOR.message_types_by_name['ShardInfo'] = _SHARDINFO +DESCRIPTOR.message_types_by_name['ShardAcknowledgement'] = _SHARDACKNOWLEDGEMENT +DESCRIPTOR.message_types_by_name['WaitExperimentRequest'] = _WAITEXPERIMENTREQUEST +DESCRIPTOR.message_types_by_name['WaitExperimentResponse'] = _WAITEXPERIMENTRESPONSE +DESCRIPTOR.message_types_by_name['GetExperimentDataRequest'] = _GETEXPERIMENTDATAREQUEST +DESCRIPTOR.message_types_by_name['ExperimentData'] = _EXPERIMENTDATA +DESCRIPTOR.message_types_by_name['ExperimentInfo'] = _EXPERIMENTINFO +DESCRIPTOR.message_types_by_name['SetNewExperimentResponse'] = _SETNEWEXPERIMENTRESPONSE +DESCRIPTOR.message_types_by_name['GetTrainedModelRequest'] = _GETTRAINEDMODELREQUEST +DESCRIPTOR.message_types_by_name['TrainedModelResponse'] = _TRAINEDMODELRESPONSE +DESCRIPTOR.message_types_by_name['GetDatasetInfoRequest'] = _GETDATASETINFOREQUEST +DESCRIPTOR.message_types_by_name['StreamMetricsRequest'] = _STREAMMETRICSREQUEST +DESCRIPTOR.message_types_by_name['StreamMetricsResponse'] = _STREAMMETRICSRESPONSE +DESCRIPTOR.message_types_by_name['RemoveExperimentRequest'] = _REMOVEEXPERIMENTREQUEST +DESCRIPTOR.message_types_by_name['RemoveExperimentResponse'] = _REMOVEEXPERIMENTRESPONSE +DESCRIPTOR.message_types_by_name['CollaboratorStatus'] = _COLLABORATORSTATUS +DESCRIPTOR.message_types_by_name['CollaboratorHealthCheckResponse'] = _COLLABORATORHEALTHCHECKRESPONSE +DESCRIPTOR.message_types_by_name['EnvoyInfo'] = _ENVOYINFO +DESCRIPTOR.message_types_by_name['GetEnvoysRequest'] = _GETENVOYSREQUEST +DESCRIPTOR.message_types_by_name['GetEnvoysResponse'] = _GETENVOYSRESPONSE +_sym_db.RegisterFileDescriptor(DESCRIPTOR) + +RequestHeader = _reflection.GeneratedProtocolMessageType('RequestHeader', (_message.Message,), { + 'DESCRIPTOR' : _REQUESTHEADER, + '__module__' : 'director_pb2' + # @@protoc_insertion_point(class_scope:RequestHeader) + }) +_sym_db.RegisterMessage(RequestHeader) + +NodeInfo = _reflection.GeneratedProtocolMessageType('NodeInfo', (_message.Message,), { + 'DESCRIPTOR' : _NODEINFO, + '__module__' : 'director_pb2' + # @@protoc_insertion_point(class_scope:NodeInfo) + }) +_sym_db.RegisterMessage(NodeInfo) + +ShardInfo = _reflection.GeneratedProtocolMessageType('ShardInfo', (_message.Message,), { + 'DESCRIPTOR' : _SHARDINFO, + '__module__' : 'director_pb2' + # @@protoc_insertion_point(class_scope:ShardInfo) + }) +_sym_db.RegisterMessage(ShardInfo) + +ShardAcknowledgement = _reflection.GeneratedProtocolMessageType('ShardAcknowledgement', (_message.Message,), { + 'DESCRIPTOR' : _SHARDACKNOWLEDGEMENT, + '__module__' : 'director_pb2' + # @@protoc_insertion_point(class_scope:ShardAcknowledgement) + }) +_sym_db.RegisterMessage(ShardAcknowledgement) + +WaitExperimentRequest = _reflection.GeneratedProtocolMessageType('WaitExperimentRequest', (_message.Message,), { + 'DESCRIPTOR' : _WAITEXPERIMENTREQUEST, + '__module__' : 'director_pb2' + # @@protoc_insertion_point(class_scope:WaitExperimentRequest) + }) +_sym_db.RegisterMessage(WaitExperimentRequest) + +WaitExperimentResponse = _reflection.GeneratedProtocolMessageType('WaitExperimentResponse', (_message.Message,), { + 'DESCRIPTOR' : _WAITEXPERIMENTRESPONSE, + '__module__' : 'director_pb2' + # @@protoc_insertion_point(class_scope:WaitExperimentResponse) + }) +_sym_db.RegisterMessage(WaitExperimentResponse) + +GetExperimentDataRequest = _reflection.GeneratedProtocolMessageType('GetExperimentDataRequest', (_message.Message,), { + 'DESCRIPTOR' : _GETEXPERIMENTDATAREQUEST, + '__module__' : 'director_pb2' + # @@protoc_insertion_point(class_scope:GetExperimentDataRequest) + }) +_sym_db.RegisterMessage(GetExperimentDataRequest) + +ExperimentData = _reflection.GeneratedProtocolMessageType('ExperimentData', (_message.Message,), { + 'DESCRIPTOR' : _EXPERIMENTDATA, + '__module__' : 'director_pb2' + # @@protoc_insertion_point(class_scope:ExperimentData) + }) +_sym_db.RegisterMessage(ExperimentData) + +ExperimentInfo = _reflection.GeneratedProtocolMessageType('ExperimentInfo', (_message.Message,), { + 'DESCRIPTOR' : _EXPERIMENTINFO, + '__module__' : 'director_pb2' + # @@protoc_insertion_point(class_scope:ExperimentInfo) + }) +_sym_db.RegisterMessage(ExperimentInfo) + +SetNewExperimentResponse = _reflection.GeneratedProtocolMessageType('SetNewExperimentResponse', (_message.Message,), { + 'DESCRIPTOR' : _SETNEWEXPERIMENTRESPONSE, + '__module__' : 'director_pb2' + # @@protoc_insertion_point(class_scope:SetNewExperimentResponse) + }) +_sym_db.RegisterMessage(SetNewExperimentResponse) + +GetTrainedModelRequest = _reflection.GeneratedProtocolMessageType('GetTrainedModelRequest', (_message.Message,), { + 'DESCRIPTOR' : _GETTRAINEDMODELREQUEST, + '__module__' : 'director_pb2' + # @@protoc_insertion_point(class_scope:GetTrainedModelRequest) + }) +_sym_db.RegisterMessage(GetTrainedModelRequest) + +TrainedModelResponse = _reflection.GeneratedProtocolMessageType('TrainedModelResponse', (_message.Message,), { + 'DESCRIPTOR' : _TRAINEDMODELRESPONSE, + '__module__' : 'director_pb2' + # @@protoc_insertion_point(class_scope:TrainedModelResponse) + }) +_sym_db.RegisterMessage(TrainedModelResponse) + +GetDatasetInfoRequest = _reflection.GeneratedProtocolMessageType('GetDatasetInfoRequest', (_message.Message,), { + 'DESCRIPTOR' : _GETDATASETINFOREQUEST, + '__module__' : 'director_pb2' + # @@protoc_insertion_point(class_scope:GetDatasetInfoRequest) + }) +_sym_db.RegisterMessage(GetDatasetInfoRequest) + +StreamMetricsRequest = _reflection.GeneratedProtocolMessageType('StreamMetricsRequest', (_message.Message,), { + 'DESCRIPTOR' : _STREAMMETRICSREQUEST, + '__module__' : 'director_pb2' + # @@protoc_insertion_point(class_scope:StreamMetricsRequest) + }) +_sym_db.RegisterMessage(StreamMetricsRequest) + +StreamMetricsResponse = _reflection.GeneratedProtocolMessageType('StreamMetricsResponse', (_message.Message,), { + 'DESCRIPTOR' : _STREAMMETRICSRESPONSE, + '__module__' : 'director_pb2' + # @@protoc_insertion_point(class_scope:StreamMetricsResponse) + }) +_sym_db.RegisterMessage(StreamMetricsResponse) + +RemoveExperimentRequest = _reflection.GeneratedProtocolMessageType('RemoveExperimentRequest', (_message.Message,), { + 'DESCRIPTOR' : _REMOVEEXPERIMENTREQUEST, + '__module__' : 'director_pb2' + # @@protoc_insertion_point(class_scope:RemoveExperimentRequest) + }) +_sym_db.RegisterMessage(RemoveExperimentRequest) + +RemoveExperimentResponse = _reflection.GeneratedProtocolMessageType('RemoveExperimentResponse', (_message.Message,), { + 'DESCRIPTOR' : _REMOVEEXPERIMENTRESPONSE, + '__module__' : 'director_pb2' + # @@protoc_insertion_point(class_scope:RemoveExperimentResponse) + }) +_sym_db.RegisterMessage(RemoveExperimentResponse) + +CollaboratorStatus = _reflection.GeneratedProtocolMessageType('CollaboratorStatus', (_message.Message,), { + 'DESCRIPTOR' : _COLLABORATORSTATUS, + '__module__' : 'director_pb2' + # @@protoc_insertion_point(class_scope:CollaboratorStatus) + }) +_sym_db.RegisterMessage(CollaboratorStatus) + +CollaboratorHealthCheckResponse = _reflection.GeneratedProtocolMessageType('CollaboratorHealthCheckResponse', (_message.Message,), { + 'DESCRIPTOR' : _COLLABORATORHEALTHCHECKRESPONSE, + '__module__' : 'director_pb2' + # @@protoc_insertion_point(class_scope:CollaboratorHealthCheckResponse) + }) +_sym_db.RegisterMessage(CollaboratorHealthCheckResponse) + +EnvoyInfo = _reflection.GeneratedProtocolMessageType('EnvoyInfo', (_message.Message,), { + 'DESCRIPTOR' : _ENVOYINFO, + '__module__' : 'director_pb2' + # @@protoc_insertion_point(class_scope:EnvoyInfo) + }) +_sym_db.RegisterMessage(EnvoyInfo) + +GetEnvoysRequest = _reflection.GeneratedProtocolMessageType('GetEnvoysRequest', (_message.Message,), { + 'DESCRIPTOR' : _GETENVOYSREQUEST, + '__module__' : 'director_pb2' + # @@protoc_insertion_point(class_scope:GetEnvoysRequest) + }) +_sym_db.RegisterMessage(GetEnvoysRequest) + +GetEnvoysResponse = _reflection.GeneratedProtocolMessageType('GetEnvoysResponse', (_message.Message,), { + 'DESCRIPTOR' : _GETENVOYSRESPONSE, + '__module__' : 'director_pb2' + # @@protoc_insertion_point(class_scope:GetEnvoysResponse) + }) +_sym_db.RegisterMessage(GetEnvoysResponse) + + + +_FEDERATIONDIRECTOR = _descriptor.ServiceDescriptor( + name='FederationDirector', + full_name='FederationDirector', + file=DESCRIPTOR, + index=0, + serialized_options=None, + create_key=_descriptor._internal_create_key, + serialized_start=1958, + serialized_end=2655, + methods=[ + _descriptor.MethodDescriptor( + name='AcknowledgeShard', + full_name='FederationDirector.AcknowledgeShard', + index=0, + containing_service=None, + input_type=_SHARDINFO, + output_type=_SHARDACKNOWLEDGEMENT, + serialized_options=None, + create_key=_descriptor._internal_create_key, + ), + _descriptor.MethodDescriptor( + name='WaitExperiment', + full_name='FederationDirector.WaitExperiment', + index=1, + containing_service=None, + input_type=_WAITEXPERIMENTREQUEST, + output_type=_WAITEXPERIMENTRESPONSE, + serialized_options=None, + create_key=_descriptor._internal_create_key, + ), + _descriptor.MethodDescriptor( + name='GetExperimentData', + full_name='FederationDirector.GetExperimentData', + index=2, + containing_service=None, + input_type=_GETEXPERIMENTDATAREQUEST, + output_type=_EXPERIMENTDATA, + serialized_options=None, + create_key=_descriptor._internal_create_key, + ), + _descriptor.MethodDescriptor( + name='SetNewExperiment', + full_name='FederationDirector.SetNewExperiment', + index=3, + containing_service=None, + input_type=_EXPERIMENTINFO, + output_type=_SETNEWEXPERIMENTRESPONSE, + serialized_options=None, + create_key=_descriptor._internal_create_key, + ), + _descriptor.MethodDescriptor( + name='GetDatasetInfo', + full_name='FederationDirector.GetDatasetInfo', + index=4, + containing_service=None, + input_type=_GETDATASETINFOREQUEST, + output_type=_SHARDINFO, + serialized_options=None, + create_key=_descriptor._internal_create_key, + ), + _descriptor.MethodDescriptor( + name='GetTrainedModel', + full_name='FederationDirector.GetTrainedModel', + index=5, + containing_service=None, + input_type=_GETTRAINEDMODELREQUEST, + output_type=_TRAINEDMODELRESPONSE, + serialized_options=None, + create_key=_descriptor._internal_create_key, + ), + _descriptor.MethodDescriptor( + name='StreamMetrics', + full_name='FederationDirector.StreamMetrics', + index=6, + containing_service=None, + input_type=_STREAMMETRICSREQUEST, + output_type=_STREAMMETRICSRESPONSE, + serialized_options=None, + create_key=_descriptor._internal_create_key, + ), + _descriptor.MethodDescriptor( + name='RemoveExperimentData', + full_name='FederationDirector.RemoveExperimentData', + index=7, + containing_service=None, + input_type=_REMOVEEXPERIMENTREQUEST, + output_type=_REMOVEEXPERIMENTRESPONSE, + serialized_options=None, + create_key=_descriptor._internal_create_key, + ), + _descriptor.MethodDescriptor( + name='CollaboratorHealthCheck', + full_name='FederationDirector.CollaboratorHealthCheck', + index=8, + containing_service=None, + input_type=_COLLABORATORSTATUS, + output_type=_COLLABORATORHEALTHCHECKRESPONSE, + serialized_options=None, + create_key=_descriptor._internal_create_key, + ), + _descriptor.MethodDescriptor( + name='GetEnvoys', + full_name='FederationDirector.GetEnvoys', + index=9, + containing_service=None, + input_type=_GETENVOYSREQUEST, + output_type=_GETENVOYSRESPONSE, + serialized_options=None, + create_key=_descriptor._internal_create_key, + ), +]) +_sym_db.RegisterServiceDescriptor(_FEDERATIONDIRECTOR) + +DESCRIPTOR.services_by_name['FederationDirector'] = _FEDERATIONDIRECTOR + +# @@protoc_insertion_point(module_scope) diff --git a/openfl/protocols/director_pb2_grpc.py b/openfl/protocols/director_pb2_grpc.py new file mode 100644 index 0000000000..18064a3c81 --- /dev/null +++ b/openfl/protocols/director_pb2_grpc.py @@ -0,0 +1,366 @@ +# Generated by the gRPC Python protocol compiler plugin. DO NOT EDIT! +"""Client and server classes corresponding to protobuf-defined services.""" +import grpc + +from . import director_pb2 as director__pb2 + + +class FederationDirectorStub(object): + """Missing associated documentation comment in .proto file.""" + + def __init__(self, channel): + """Constructor. + + Args: + channel: A grpc.Channel. + """ + self.AcknowledgeShard = channel.unary_unary( + '/FederationDirector/AcknowledgeShard', + request_serializer=director__pb2.ShardInfo.SerializeToString, + response_deserializer=director__pb2.ShardAcknowledgement.FromString, + ) + self.WaitExperiment = channel.stream_stream( + '/FederationDirector/WaitExperiment', + request_serializer=director__pb2.WaitExperimentRequest.SerializeToString, + response_deserializer=director__pb2.WaitExperimentResponse.FromString, + ) + self.GetExperimentData = channel.unary_stream( + '/FederationDirector/GetExperimentData', + request_serializer=director__pb2.GetExperimentDataRequest.SerializeToString, + response_deserializer=director__pb2.ExperimentData.FromString, + ) + self.SetNewExperiment = channel.stream_unary( + '/FederationDirector/SetNewExperiment', + request_serializer=director__pb2.ExperimentInfo.SerializeToString, + response_deserializer=director__pb2.SetNewExperimentResponse.FromString, + ) + self.GetDatasetInfo = channel.unary_unary( + '/FederationDirector/GetDatasetInfo', + request_serializer=director__pb2.GetDatasetInfoRequest.SerializeToString, + response_deserializer=director__pb2.ShardInfo.FromString, + ) + self.GetTrainedModel = channel.unary_unary( + '/FederationDirector/GetTrainedModel', + request_serializer=director__pb2.GetTrainedModelRequest.SerializeToString, + response_deserializer=director__pb2.TrainedModelResponse.FromString, + ) + self.StreamMetrics = channel.unary_stream( + '/FederationDirector/StreamMetrics', + request_serializer=director__pb2.StreamMetricsRequest.SerializeToString, + response_deserializer=director__pb2.StreamMetricsResponse.FromString, + ) + self.RemoveExperimentData = channel.unary_unary( + '/FederationDirector/RemoveExperimentData', + request_serializer=director__pb2.RemoveExperimentRequest.SerializeToString, + response_deserializer=director__pb2.RemoveExperimentResponse.FromString, + ) + self.CollaboratorHealthCheck = channel.unary_unary( + '/FederationDirector/CollaboratorHealthCheck', + request_serializer=director__pb2.CollaboratorStatus.SerializeToString, + response_deserializer=director__pb2.CollaboratorHealthCheckResponse.FromString, + ) + self.GetEnvoys = channel.unary_unary( + '/FederationDirector/GetEnvoys', + request_serializer=director__pb2.GetEnvoysRequest.SerializeToString, + response_deserializer=director__pb2.GetEnvoysResponse.FromString, + ) + + +class FederationDirectorServicer(object): + """Missing associated documentation comment in .proto file.""" + + def AcknowledgeShard(self, request, context): + """Envoy RPCs + """ + context.set_code(grpc.StatusCode.UNIMPLEMENTED) + context.set_details('Method not implemented!') + raise NotImplementedError('Method not implemented!') + + def WaitExperiment(self, request_iterator, context): + """Shard owner could also provide some public data for tests + """ + context.set_code(grpc.StatusCode.UNIMPLEMENTED) + context.set_details('Method not implemented!') + raise NotImplementedError('Method not implemented!') + + def GetExperimentData(self, request, context): + """Missing associated documentation comment in .proto file.""" + context.set_code(grpc.StatusCode.UNIMPLEMENTED) + context.set_details('Method not implemented!') + raise NotImplementedError('Method not implemented!') + + def SetNewExperiment(self, request_iterator, context): + """API RPCs + """ + context.set_code(grpc.StatusCode.UNIMPLEMENTED) + context.set_details('Method not implemented!') + raise NotImplementedError('Method not implemented!') + + def GetDatasetInfo(self, request, context): + """Missing associated documentation comment in .proto file.""" + context.set_code(grpc.StatusCode.UNIMPLEMENTED) + context.set_details('Method not implemented!') + raise NotImplementedError('Method not implemented!') + + def GetTrainedModel(self, request, context): + """Missing associated documentation comment in .proto file.""" + context.set_code(grpc.StatusCode.UNIMPLEMENTED) + context.set_details('Method not implemented!') + raise NotImplementedError('Method not implemented!') + + def StreamMetrics(self, request, context): + """Missing associated documentation comment in .proto file.""" + context.set_code(grpc.StatusCode.UNIMPLEMENTED) + context.set_details('Method not implemented!') + raise NotImplementedError('Method not implemented!') + + def RemoveExperimentData(self, request, context): + """Missing associated documentation comment in .proto file.""" + context.set_code(grpc.StatusCode.UNIMPLEMENTED) + context.set_details('Method not implemented!') + raise NotImplementedError('Method not implemented!') + + def CollaboratorHealthCheck(self, request, context): + """Missing associated documentation comment in .proto file.""" + context.set_code(grpc.StatusCode.UNIMPLEMENTED) + context.set_details('Method not implemented!') + raise NotImplementedError('Method not implemented!') + + def GetEnvoys(self, request, context): + """Missing associated documentation comment in .proto file.""" + context.set_code(grpc.StatusCode.UNIMPLEMENTED) + context.set_details('Method not implemented!') + raise NotImplementedError('Method not implemented!') + + +def add_FederationDirectorServicer_to_server(servicer, server): + rpc_method_handlers = { + 'AcknowledgeShard': grpc.unary_unary_rpc_method_handler( + servicer.AcknowledgeShard, + request_deserializer=director__pb2.ShardInfo.FromString, + response_serializer=director__pb2.ShardAcknowledgement.SerializeToString, + ), + 'WaitExperiment': grpc.stream_stream_rpc_method_handler( + servicer.WaitExperiment, + request_deserializer=director__pb2.WaitExperimentRequest.FromString, + response_serializer=director__pb2.WaitExperimentResponse.SerializeToString, + ), + 'GetExperimentData': grpc.unary_stream_rpc_method_handler( + servicer.GetExperimentData, + request_deserializer=director__pb2.GetExperimentDataRequest.FromString, + response_serializer=director__pb2.ExperimentData.SerializeToString, + ), + 'SetNewExperiment': grpc.stream_unary_rpc_method_handler( + servicer.SetNewExperiment, + request_deserializer=director__pb2.ExperimentInfo.FromString, + response_serializer=director__pb2.SetNewExperimentResponse.SerializeToString, + ), + 'GetDatasetInfo': grpc.unary_unary_rpc_method_handler( + servicer.GetDatasetInfo, + request_deserializer=director__pb2.GetDatasetInfoRequest.FromString, + response_serializer=director__pb2.ShardInfo.SerializeToString, + ), + 'GetTrainedModel': grpc.unary_unary_rpc_method_handler( + servicer.GetTrainedModel, + request_deserializer=director__pb2.GetTrainedModelRequest.FromString, + response_serializer=director__pb2.TrainedModelResponse.SerializeToString, + ), + 'StreamMetrics': grpc.unary_stream_rpc_method_handler( + servicer.StreamMetrics, + request_deserializer=director__pb2.StreamMetricsRequest.FromString, + response_serializer=director__pb2.StreamMetricsResponse.SerializeToString, + ), + 'RemoveExperimentData': grpc.unary_unary_rpc_method_handler( + servicer.RemoveExperimentData, + request_deserializer=director__pb2.RemoveExperimentRequest.FromString, + response_serializer=director__pb2.RemoveExperimentResponse.SerializeToString, + ), + 'CollaboratorHealthCheck': grpc.unary_unary_rpc_method_handler( + servicer.CollaboratorHealthCheck, + request_deserializer=director__pb2.CollaboratorStatus.FromString, + response_serializer=director__pb2.CollaboratorHealthCheckResponse.SerializeToString, + ), + 'GetEnvoys': grpc.unary_unary_rpc_method_handler( + servicer.GetEnvoys, + request_deserializer=director__pb2.GetEnvoysRequest.FromString, + response_serializer=director__pb2.GetEnvoysResponse.SerializeToString, + ), + } + generic_handler = grpc.method_handlers_generic_handler( + 'FederationDirector', rpc_method_handlers) + server.add_generic_rpc_handlers((generic_handler,)) + + + # This class is part of an EXPERIMENTAL API. +class FederationDirector(object): + """Missing associated documentation comment in .proto file.""" + + @staticmethod + def AcknowledgeShard(request, + target, + options=(), + channel_credentials=None, + call_credentials=None, + insecure=False, + compression=None, + wait_for_ready=None, + timeout=None, + metadata=None): + return grpc.experimental.unary_unary(request, target, '/FederationDirector/AcknowledgeShard', + director__pb2.ShardInfo.SerializeToString, + director__pb2.ShardAcknowledgement.FromString, + options, channel_credentials, + insecure, call_credentials, compression, wait_for_ready, timeout, metadata) + + @staticmethod + def WaitExperiment(request_iterator, + target, + options=(), + channel_credentials=None, + call_credentials=None, + insecure=False, + compression=None, + wait_for_ready=None, + timeout=None, + metadata=None): + return grpc.experimental.stream_stream(request_iterator, target, '/FederationDirector/WaitExperiment', + director__pb2.WaitExperimentRequest.SerializeToString, + director__pb2.WaitExperimentResponse.FromString, + options, channel_credentials, + insecure, call_credentials, compression, wait_for_ready, timeout, metadata) + + @staticmethod + def GetExperimentData(request, + target, + options=(), + channel_credentials=None, + call_credentials=None, + insecure=False, + compression=None, + wait_for_ready=None, + timeout=None, + metadata=None): + return grpc.experimental.unary_stream(request, target, '/FederationDirector/GetExperimentData', + director__pb2.GetExperimentDataRequest.SerializeToString, + director__pb2.ExperimentData.FromString, + options, channel_credentials, + insecure, call_credentials, compression, wait_for_ready, timeout, metadata) + + @staticmethod + def SetNewExperiment(request_iterator, + target, + options=(), + channel_credentials=None, + call_credentials=None, + insecure=False, + compression=None, + wait_for_ready=None, + timeout=None, + metadata=None): + return grpc.experimental.stream_unary(request_iterator, target, '/FederationDirector/SetNewExperiment', + director__pb2.ExperimentInfo.SerializeToString, + director__pb2.SetNewExperimentResponse.FromString, + options, channel_credentials, + insecure, call_credentials, compression, wait_for_ready, timeout, metadata) + + @staticmethod + def GetDatasetInfo(request, + target, + options=(), + channel_credentials=None, + call_credentials=None, + insecure=False, + compression=None, + wait_for_ready=None, + timeout=None, + metadata=None): + return grpc.experimental.unary_unary(request, target, '/FederationDirector/GetDatasetInfo', + director__pb2.GetDatasetInfoRequest.SerializeToString, + director__pb2.ShardInfo.FromString, + options, channel_credentials, + insecure, call_credentials, compression, wait_for_ready, timeout, metadata) + + @staticmethod + def GetTrainedModel(request, + target, + options=(), + channel_credentials=None, + call_credentials=None, + insecure=False, + compression=None, + wait_for_ready=None, + timeout=None, + metadata=None): + return grpc.experimental.unary_unary(request, target, '/FederationDirector/GetTrainedModel', + director__pb2.GetTrainedModelRequest.SerializeToString, + director__pb2.TrainedModelResponse.FromString, + options, channel_credentials, + insecure, call_credentials, compression, wait_for_ready, timeout, metadata) + + @staticmethod + def StreamMetrics(request, + target, + options=(), + channel_credentials=None, + call_credentials=None, + insecure=False, + compression=None, + wait_for_ready=None, + timeout=None, + metadata=None): + return grpc.experimental.unary_stream(request, target, '/FederationDirector/StreamMetrics', + director__pb2.StreamMetricsRequest.SerializeToString, + director__pb2.StreamMetricsResponse.FromString, + options, channel_credentials, + insecure, call_credentials, compression, wait_for_ready, timeout, metadata) + + @staticmethod + def RemoveExperimentData(request, + target, + options=(), + channel_credentials=None, + call_credentials=None, + insecure=False, + compression=None, + wait_for_ready=None, + timeout=None, + metadata=None): + return grpc.experimental.unary_unary(request, target, '/FederationDirector/RemoveExperimentData', + director__pb2.RemoveExperimentRequest.SerializeToString, + director__pb2.RemoveExperimentResponse.FromString, + options, channel_credentials, + insecure, call_credentials, compression, wait_for_ready, timeout, metadata) + + @staticmethod + def CollaboratorHealthCheck(request, + target, + options=(), + channel_credentials=None, + call_credentials=None, + insecure=False, + compression=None, + wait_for_ready=None, + timeout=None, + metadata=None): + return grpc.experimental.unary_unary(request, target, '/FederationDirector/CollaboratorHealthCheck', + director__pb2.CollaboratorStatus.SerializeToString, + director__pb2.CollaboratorHealthCheckResponse.FromString, + options, channel_credentials, + insecure, call_credentials, compression, wait_for_ready, timeout, metadata) + + @staticmethod + def GetEnvoys(request, + target, + options=(), + channel_credentials=None, + call_credentials=None, + insecure=False, + compression=None, + wait_for_ready=None, + timeout=None, + metadata=None): + return grpc.experimental.unary_unary(request, target, '/FederationDirector/GetEnvoys', + director__pb2.GetEnvoysRequest.SerializeToString, + director__pb2.GetEnvoysResponse.FromString, + options, channel_credentials, + insecure, call_credentials, compression, wait_for_ready, timeout, metadata) diff --git a/openfl/protocols/federation.proto b/openfl/protocols/federation.proto index 893204990c..83c01dab97 100644 --- a/openfl/protocols/federation.proto +++ b/openfl/protocols/federation.proto @@ -1,5 +1,6 @@ // Copyright (C) 2020 Intel Corporation // Licensed subject to the terms of the separately executed evaluation license agreement between Intel Corporation and you. +syntax = "proto2"; service Aggregator { rpc GetTasks(TasksRequest) returns (TasksResponse) {} diff --git a/openfl/transport/__init__.py b/openfl/transport/__init__.py index 4953716521..e9307d443c 100644 --- a/openfl/transport/__init__.py +++ b/openfl/transport/__init__.py @@ -5,8 +5,10 @@ from .grpc import AggregatorGRPCServer from .grpc import CollaboratorGRPCClient +from .grpc import DirectorGRPCServer __all__ = [ 'AggregatorGRPCServer', - 'CollaboratorGRPCClient' + 'CollaboratorGRPCClient', + 'DirectorGRPCServer', ] diff --git a/openfl/transport/grpc/__init__.py b/openfl/transport/grpc/__init__.py index bfe365a265..2258bb30aa 100644 --- a/openfl/transport/grpc/__init__.py +++ b/openfl/transport/grpc/__init__.py @@ -4,9 +4,11 @@ """openfl.transport.grpc package.""" from .client import CollaboratorGRPCClient +from .director_server import DirectorGRPCServer from .server import AggregatorGRPCServer __all__ = [ 'AggregatorGRPCServer', 'CollaboratorGRPCClient', + 'DirectorGRPCServer', ] diff --git a/openfl/transport/grpc/client.py b/openfl/transport/grpc/client.py index 70f4d8ee9f..a769b9c781 100644 --- a/openfl/transport/grpc/client.py +++ b/openfl/transport/grpc/client.py @@ -56,6 +56,7 @@ def _intercept_call(self, continuation, client_call_details, request_or_iterator if isinstance(response, grpc.RpcError): # If status code is not in retryable status codes + self.sleeping_policy.logger.info(f'Response code: {response.code()}') if ( self.status_for_retry and response.code() not in self.status_for_retry diff --git a/openfl/transport/grpc/director_client.py b/openfl/transport/grpc/director_client.py new file mode 100644 index 0000000000..ac6c587f93 --- /dev/null +++ b/openfl/transport/grpc/director_client.py @@ -0,0 +1,286 @@ +# Copyright (C) 2020-2021 Intel Corporation +# SPDX-License-Identifier: Apache-2.0 + +"""Director clients module.""" + +import logging +import os +import shutil +import time +from datetime import datetime +from subprocess import check_call +from sys import executable + +import grpc + +from openfl.pipelines import NoCompressionPipeline +from openfl.protocols import director_pb2 +from openfl.protocols import director_pb2_grpc +from openfl.protocols.utils import construct_model_proto +from openfl.protocols.utils import deconstruct_model_proto + + +logger = logging.getLogger(__name__) + + +class ShardDirectorClient: + """The internal director client class.""" + + def __init__(self, director_uri, shard_name, disable_tls=False, + root_ca=None, key=None, cert=None) -> None: + """Initialize a shard director client object.""" + self.shard_name = shard_name + options = [('grpc.max_message_length', 100 * 1024 * 1024)] + if disable_tls: + channel = grpc.insecure_channel(director_uri, options=options) + else: + if not (root_ca and key and cert): + raise Exception('No certificates provided') + try: + with open(root_ca, 'rb') as f: + root_ca_b = f.read() + with open(key, 'rb') as f: + key_b = f.read() + with open(cert, 'rb') as f: + cert_b = f.read() + except FileNotFoundError as exc: + raise Exception(f'Provided certificate file is not exist: {exc.filename}') + + credentials = grpc.ssl_channel_credentials( + root_certificates=root_ca_b, + private_key=key_b, + certificate_chain=cert_b + ) + channel = grpc.secure_channel(director_uri, credentials, options=options) + self.stub = director_pb2_grpc.FederationDirectorStub(channel) + + def report_shard_info(self, shard_descriptor) -> bool: + """Report shard info to the director.""" + logger.info('Send report AcknowledgeShard') + # True considered as successful registration + shard_info = director_pb2.ShardInfo( + shard_description=shard_descriptor.dataset_description, + n_samples=len(shard_descriptor), + sample_shape=shard_descriptor.sample_shape, + target_shape=shard_descriptor.target_shape + ) + + shard_info.node_info.CopyFrom(self._get_node_info()) + + acknowledgement = self.stub.AcknowledgeShard(shard_info) + return acknowledgement.accepted + + def get_experiment_data(self): + """Get an experiment data from the director.""" + logger.info('Send WaitExperiment request') + response_iter = self.stub.WaitExperiment(self._get_experiment_data()) + logger.info('WaitExperiment response has received') + response = next(response_iter) + experiment_name = response.experiment_name + if not experiment_name: + raise Exception('No experiment') + logger.info(f'Request experiment {experiment_name}') + request = director_pb2.GetExperimentDataRequest( + experiment_name=experiment_name, + collaborator_name=self.shard_name + ) + response_iter = self.stub.GetExperimentData(request) + + self.create_workspace(experiment_name, response_iter) + + return experiment_name + + @staticmethod + def remove_workspace(experiment_name): + """Remove the workspace.""" + shutil.rmtree(experiment_name, ignore_errors=True) + + @staticmethod + def create_workspace(experiment_name, response_iter): + """Create a collaborator workspace for the experiment.""" + if os.path.exists(experiment_name): + shutil.rmtree(experiment_name) + os.makedirs(experiment_name) + + arch_name = f'{experiment_name}/{experiment_name}' + '.zip' + logger.info(f'arch_name: {arch_name}') + with open(arch_name, 'wb') as content_file: + for response in response_iter: + logger.info(f'Size: {response.size}') + if response.size == len(response.npbytes): + content_file.write(response.npbytes) + else: + raise Exception('Broken archive') + + shutil.unpack_archive(arch_name, experiment_name) + os.remove(arch_name) + + requirements_filename = f'./{experiment_name}/requirements.txt' + + if os.path.isfile(requirements_filename): + attempts = 3 + for _ in range(attempts): + try: + check_call([ + executable, '-m', 'pip', 'install', '-r', requirements_filename], + shell=False) + except Exception as exc: + logger.error(f'Failed to install requirements: {exc}') + time.sleep(3) + else: + break + else: + logger.error('No ' + requirements_filename + ' file found.') + + def _get_experiment_data(self): + """Generate the experiment data request.""" + yield director_pb2.WaitExperimentRequest(collaborator_name=self.shard_name) + + def _get_node_info(self): + """Generate a node info message.""" + return director_pb2.NodeInfo(name=self.shard_name) + + def send_health_check(self, collaborator_name, is_experiment_running, valid_duration): + """Send envoy health check.""" + status = director_pb2.CollaboratorStatus( + name=collaborator_name, + is_experiment_running=is_experiment_running, + ) + status.valid_duration.seconds = valid_duration + logger.debug(f'Sending health check status: {status}') + + return self.stub.CollaboratorHealthCheck(status) + + +class DirectorClient: + """Director client class for users.""" + + def __init__(self, client_id, director_uri, disable_tls=False, + root_ca=None, key=None, cert=None) -> None: + """Initialize director client object.""" + channel_opt = [('grpc.max_send_message_length', 512 * 1024 * 1024), + ('grpc.max_receive_message_length', 512 * 1024 * 1024)] + if disable_tls: + channel = grpc.insecure_channel(director_uri, options=channel_opt) + else: + if not (root_ca and key and cert): + raise Exception('No certificates provided') + try: + with open(root_ca, 'rb') as f: + root_ca_b = f.read() + with open(key, 'rb') as f: + key_b = f.read() + with open(cert, 'rb') as f: + cert_b = f.read() + except FileNotFoundError as exc: + raise Exception(f'Provided certificate file is not exist: {exc.filename}') + + credentials = grpc.ssl_channel_credentials( + root_certificates=root_ca_b, + private_key=key_b, + certificate_chain=cert_b + ) + + channel = grpc.secure_channel(director_uri, credentials, options=channel_opt) + self.stub = director_pb2_grpc.FederationDirectorStub(channel) + + self.client_id = client_id + self.header = director_pb2.RequestHeader(sender=self.client_id) + + def set_new_experiment(self, name, col_names, arch_path, + initial_tensor_dict=None): + """Send the new experiment to director to launch.""" + logger.info('SetNewExperiment') + model_proto = None + if initial_tensor_dict: + model_proto = construct_model_proto(initial_tensor_dict, 0, NoCompressionPipeline()) + + with open(arch_path, 'rb') as arch: + def st(): + max_buffer_size = (2 * 1024 * 1024) + chunk = arch.read(max_buffer_size) + while chunk != b'': + if not chunk: + raise StopIteration + # TODO: add hash or/and size to check + experiment_info = director_pb2.ExperimentInfo( + header=self.header, + name=name, + collaborator_names=col_names, + model_proto=model_proto + ) + experiment_info.experiment_data.size = len(chunk) + experiment_info.experiment_data.npbytes = chunk + yield experiment_info + chunk = arch.read(max_buffer_size) + + resp = self.stub.SetNewExperiment(st()) + return resp + + def get_dataset_info(self): + """Request the dataset info from the director.""" + resp = self.stub.GetDatasetInfo(director_pb2.GetDatasetInfoRequest(header=self.header)) + return resp.sample_shape, resp.target_shape + + def _get_trained_model(self, experiment_name, model_type): + """Get trained model RPC.""" + get_model_request = director_pb2.GetTrainedModelRequest( + header=self.header, + experiment_name=experiment_name, + model_type=model_type) + model_proto_response = self.stub.GetTrainedModel(get_model_request) + tensor_dict, _ = deconstruct_model_proto( + model_proto_response.model_proto, + NoCompressionPipeline() + ) + return tensor_dict + + def get_best_model(self, experiment_name): + """Get best model method.""" + model_type = director_pb2.GetTrainedModelRequest.BEST_MODEL + return self._get_trained_model(experiment_name, model_type) + + def get_last_model(self, experiment_name): + """Get last model method.""" + model_type = director_pb2.GetTrainedModelRequest.LAST_MODEL + return self._get_trained_model(experiment_name, model_type) + + def stream_metrics(self, experiment_name): + """Stream metrics RPC.""" + request = director_pb2.StreamMetricsRequest( + header=self.header, + experiment_name=experiment_name) + for metric_message in self.stub.StreamMetrics(request): + yield { + 'metric_origin': metric_message.metric_origin, + 'task_name': metric_message.task_name, + 'metric_name': metric_message.metric_name, + 'metric_value': metric_message.metric_value, + 'round': metric_message.round} + + def remove_experiment_data(self, name): + """Remove experiment data RPC.""" + request = director_pb2.RemoveExperimentRequest( + header=self.header, + experiment_name=name) + response = self.stub.RemoveExperimentData(request) + return response.acknowledgement + + def get_envoys(self, raw_result=False): + """Get envoys info.""" + envoys = self.stub.GetEnvoys(director_pb2.GetEnvoysRequest()) + if raw_result: + return envoys + now = datetime.now().strftime('%Y-%m-%d %H:%M:%S') + result = {} + for envoy in envoys.envoy_infos: + result[envoy.shard_info.node_info.name] = { + 'shard_info': envoy.shard_info, + 'is_online': envoy.is_online or False, + 'is_experiment_running': envoy.is_experiment_running or False, + 'last_updated': datetime.fromtimestamp( + envoy.last_updated.seconds).strftime('%Y-%m-%d %H:%M:%S'), + 'current_time': now, + 'valid_duration': envoy.valid_duration + } + return result diff --git a/openfl/transport/grpc/director_server.py b/openfl/transport/grpc/director_server.py new file mode 100644 index 0000000000..a802b35f44 --- /dev/null +++ b/openfl/transport/grpc/director_server.py @@ -0,0 +1,244 @@ +# Copyright (C) 2020-2021 Intel Corporation +# SPDX-License-Identifier: Apache-2.0 + +"""Director server.""" + +import asyncio +import logging +from pathlib import Path + +from grpc import aio +from grpc import ssl_server_credentials + +from openfl.pipelines import NoCompressionPipeline +from openfl.protocols import director_pb2 +from openfl.protocols import director_pb2_grpc +from openfl.protocols.utils import construct_model_proto +from openfl.protocols.utils import deconstruct_model_proto + +logger = logging.getLogger(__name__) + + +class DirectorGRPCServer(director_pb2_grpc.FederationDirectorServicer): + """Director transport class.""" + + def __init__(self, *, director_cls, disable_tls: bool = False, + root_ca: str = None, key: str = None, cert: str = None, + listen_ip='[::]', listen_port=50051, **kwargs) -> None: + """Initialize a director object.""" + # TODO: add working directory + super().__init__() + + self.listen_addr = f'{listen_ip}:{listen_port}' + self.disable_tls = disable_tls + self.root_ca = None + self.key = None + self.cert = None + self._fill_certs(root_ca, key, cert) + self.server = None + self.director = director_cls( + disable_tls=self.disable_tls, + root_ca=self.root_ca, + key=self.key, + cert=self.cert, + **kwargs + ) + + def _fill_certs(self, root_ca, key, cert): + """Fill certificates.""" + if not self.disable_tls: + if not (root_ca and key and cert): + raise Exception('No certificates provided') + self.root_ca = Path(root_ca).absolute() + self.key = Path(key).absolute() + self.cert = Path(cert).absolute() + + def validate_caller(self, request, context): + """ + Validate the caller. + + Args: + request: The gRPC message request + context: The gRPC context + + Returns: + True - if the caller has valid cert + False - if the callers cert name is invalid + """ + # Can we close the request right here? + if not self.disable_tls: + caller_cert_name = context.auth_context()['x509_common_name'][0].decode('utf-8') + caller_common_name = request.header.sender + if caller_cert_name != caller_common_name: + return False + + return True + + def start(self): + """Launch the director GRPC server.""" + asyncio.run(self._run()) + + async def _run(self): + channel_opt = [('grpc.max_send_message_length', 512 * 1024 * 1024), + ('grpc.max_receive_message_length', 512 * 1024 * 1024)] + self.server = aio.server(options=channel_opt) + director_pb2_grpc.add_FederationDirectorServicer_to_server(self, self.server) + + if self.disable_tls: + self.server.add_insecure_port(self.listen_addr) + else: + with open(self.key, 'rb') as f: + key_b = f.read() + with open(self.cert, 'rb') as f: + cert_b = f.read() + with open(self.root_ca, 'rb') as f: + root_ca_b = f.read() + server_credentials = ssl_server_credentials( + ((key_b, cert_b),), + root_certificates=root_ca_b, + require_client_auth=True + ) + self.server.add_secure_port(self.listen_addr, server_credentials) + logger.info(f'Starting server on {self.listen_addr}') + await self.server.start() + await self.server.wait_for_termination() + + async def AcknowledgeShard(self, shard_info, context): # NOQA:N802 + """Receive acknowledge shard info.""" + logger.info(f'AcknowledgeShard request has got: {shard_info}') + is_accepted = self.director.acknowledge_shard(shard_info) + reply = director_pb2.ShardAcknowledgement(accepted=is_accepted) + + return reply + + async def SetNewExperiment(self, stream, context): # NOQA:N802 + """Request to set new experiment.""" + logger.info(f'SetNewExperiment request has got {stream}') + # TODO: add streaming reader + npbytes = b'' + async for request in stream: + if request.experiment_data.size == len(request.experiment_data.npbytes): + npbytes += request.experiment_data.npbytes + else: + raise Exception('Bad request') + + if not self.validate_caller(request, context): + # Can we send reject before reading the stream? + return director_pb2.SetNewExperimentResponse(accepted=False) + tensor_dict = None + if request.model_proto: + tensor_dict, _ = deconstruct_model_proto(request.model_proto, NoCompressionPipeline()) + + is_accepted = await self.director.set_new_experiment( + experiment_name=request.name, + sender_name=request.header.sender, + tensor_dict=tensor_dict, + collaborator_names=request.collaborator_names, + data=npbytes + ) + + logger.info('Send response') + return director_pb2.SetNewExperimentResponse(accepted=is_accepted) + + async def GetTrainedModel(self, request, context): # NOQA:N802 + """RPC for retrieving trained models.""" + if not self.validate_caller(request, context): + return director_pb2.TrainedModelResponse() + logger.info('Request GetTrainedModel has got!') + + best_tensor_dict, last_tensor_dict = await self.director.get_trained_model( + request.experiment_name, + request.header.sender + ) + + if request.model_type == director_pb2.GetTrainedModelRequest.BEST_MODEL: + tensor_dict = best_tensor_dict + elif request.model_type == director_pb2.GetTrainedModelRequest.LAST_MODEL: + tensor_dict = last_tensor_dict + else: + logger.error('Incorrect model type') + return director_pb2.TrainedModelResponse() + if not tensor_dict: + return director_pb2.TrainedModelResponse() + + model_proto = construct_model_proto(tensor_dict, 0, NoCompressionPipeline()) + + return director_pb2.TrainedModelResponse(model_proto=model_proto) + + async def GetExperimentData(self, request, context): # NOQA:N802 + """Receive experiment data.""" + # TODO: add size filling + # TODO: add experiment name field + # TODO: rename npbytes to data + content = self.director.get_experiment_data(request.experiment_name) + logger.info(f'Content length: {len(content)}') + max_buffer_size = (2 * 1024 * 1024) + + for i in range(0, len(content), max_buffer_size): + chunk = content[i:i + max_buffer_size] + logger.info(f'Send {len(chunk)} bytes') + yield director_pb2.ExperimentData(size=len(chunk), npbytes=chunk) + + async def WaitExperiment(self, request_iterator, context): # NOQA:N802 + """Request for wait an experiment.""" + logger.info('Request WaitExperiment has got!') + async for msg in request_iterator: + logger.info(msg) + experiment_name = await self.director.wait_experiment(msg.collaborator_name) + logger.info(f'Experiment {experiment_name} was prepared') + + yield director_pb2.WaitExperimentResponse(experiment_name=experiment_name) + + async def GetDatasetInfo(self, request, context): # NOQA:N802 + """Request the info about target and sample shapes in the dataset.""" + logger.info('Request GetDatasetInfo has got!') + if not self.validate_caller(request, context): + return director_pb2.ShardInfo() + + sample_shape, target_shape = self.director.get_dataset_info() + resp = director_pb2.ShardInfo( + sample_shape=sample_shape, + target_shape=target_shape + ) + return resp + + async def StreamMetrics(self, request, context): # NOQA:N802 + """Request to stream metrics from the aggregator to frontend.""" + if not self.validate_caller(request, context): + return + # We should probably set a name to the aggregator and verify it here. + # Moreover, we may save the experiment name in plan.yaml and retrieve it + # during the aggregator initialization + logger.info(f'Request StreamMetrics for {request.experiment_name} experiment has got!') + + async for message in \ + self.director.stream_metrics(request.experiment_name, request.header.sender): + yield message + + async def RemoveExperimentData(self, request, context): # NOQA:N802 + """Remove experiment data RPC.""" + response = director_pb2.RemoveExperimentResponse(acknowledgement=False) + if not self.validate_caller(request, context): + return response + + self.director.remove_experiment_data(request.experiment_name, request.header.sender) + + response.acknowledgement = True + return response + + async def CollaboratorHealthCheck(self, request, context): # NOQA:N802 + """Accept health check from envoy.""" + logger.debug(f'Request CollaboratorHealthCheck has got: {request}') + is_accepted = self.director.collaborator_health_check( + collaborator_name=request.name, + is_experiment_running=request.is_experiment_running, + valid_duration=request.valid_duration.seconds, + ) + + return director_pb2.CollaboratorHealthCheckResponse(accepted=is_accepted) + + async def GetEnvoys(self, request, context): # NOQA:N802 + """Get a status information about envoys.""" + envoy_infos = self.director.get_envoys() + + return director_pb2.GetEnvoysResponse(envoy_infos=envoy_infos) diff --git a/openfl/transport/grpc/server.py b/openfl/transport/grpc/server.py index 801a11f25f..ce58ee1270 100644 --- a/openfl/transport/grpc/server.py +++ b/openfl/transport/grpc/server.py @@ -205,8 +205,8 @@ def SendLocalTaskResults(self, request, context): # NOQA:N802 # turn data stream into local model update return Acknowledgement(header=self.get_header(collaborator_name)) - def serve(self): - """Start an aggregator gRPC service.""" + def get_server(self): + """Return gRPC server.""" self.server = server(ThreadPoolExecutor(max_workers=cpu_count()), options=self.channel_options) @@ -216,8 +216,8 @@ def serve(self): self.logger.warn( 'gRPC is running on insecure channel with TLS disabled.') - - self.server.add_insecure_port(self.uri) + port = self.server.add_insecure_port(self.uri) + self.logger.info(f'Insecure port: {port}') else: @@ -239,8 +239,13 @@ def serve(self): self.server.add_secure_port(self.uri, self.server_credentials) - self.logger.info('Starting Aggregator gRPC Server') + return self.server + def serve(self): + """Start an aggregator gRPC service.""" + self.get_server() + + self.logger.info('Starting Aggregator gRPC Server') self.server.start() try: diff --git a/openfl/utilities/ca.py b/openfl/utilities/ca.py new file mode 100644 index 0000000000..5371635291 --- /dev/null +++ b/openfl/utilities/ca.py @@ -0,0 +1,18 @@ +# Copyright (C) 2020-2021 Intel Corporation +# SPDX-License-Identifier: Apache-2.0 +"""Generic check functions.""" +import os + + +def get_credentials(folder_path): + """Get credentials from folder by template.""" + root_ca, key, cert = None, None, None + if os.path.exists(folder_path): + for f in os.listdir(folder_path): + if '.key' in f: + key = folder_path + '/' + f + if '.crt' in f and 'root_ca' not in f: + cert = folder_path + '/' + f + if 'root_ca' in f: + root_ca = folder_path + '/' + f + return root_ca, key, cert diff --git a/openfl/utilities/logs.py b/openfl/utilities/logs.py new file mode 100644 index 0000000000..3507042058 --- /dev/null +++ b/openfl/utilities/logs.py @@ -0,0 +1,38 @@ +# Copyright (C) 2020-2021 Intel Corporation +# SPDX-License-Identifier: Apache-2.0 + +"""Logs utilities.""" + +import logging + +from rich.console import Console +from rich.logging import RichHandler +from tensorboardX import SummaryWriter + +writer = None + + +def get_writer(): + """Create global writer object.""" + global writer + if not writer: + writer = SummaryWriter('./logs/cnn_mnist', flush_secs=5) + + +def write_metric(node_name, task_name, metric_name, metric, round_number): + """Write metric callback.""" + get_writer() + writer.add_scalar(f'{node_name}/{task_name}/{metric_name}', metric, round_number) + + +def setup_loggers(log_level=logging.INFO): + """Configure loggers.""" + root = logging.getLogger() + root.setLevel(log_level) + console = Console(width=160) + handler = RichHandler(console=console) + formatter = logging.Formatter( + '[%(asctime)s][%(name)s][%(levelname)s] - %(message)s' + ) + handler.setFormatter(formatter) + root.addHandler(handler) diff --git a/openfl/utilities/utils.py b/openfl/utilities/utils.py index 05d55cb181..40cab8fb40 100644 --- a/openfl/utilities/utils.py +++ b/openfl/utilities/utils.py @@ -1,7 +1,10 @@ # Copyright (C) 2020-2021 Intel Corporation # SPDX-License-Identifier: Apache-2.0 """Utilities module.""" + +import logging import os +import shutil from socket import getfqdn import numpy as np @@ -23,6 +26,62 @@ def getfqdn_env(name: str = '') -> str: return getfqdn(name) +def add_log_level(level_name, level_num, method_name=None): + """ + Add a new logging level to the logging module. + + Args: + level_name: name of log level. + level_num: log level value. + method_name: log method wich will use new log level (default = level_name.lower()) + + """ + if not method_name: + method_name = level_name.lower() + + def log_for_level(self, message, *args, **kwargs): + if self.isEnabledFor(level_num): + self._log(level_num, message, args, **kwargs) + + def log_to_root(message, *args, **kwargs): + logging.log(level_num, message, *args, **kwargs) + + logging.addLevelName(level_num, level_name) + setattr(logging, level_name, level_num) + setattr(logging.getLoggerClass(), method_name, log_for_level) + setattr(logging, method_name, log_to_root) + + +class UnpackWorkspace: + """Workspace context manager. + + It unpacks the archive at the begining of the experiment and removes the archive. + Then in moving to the workspace's folder. + Exiting it removes the workspace + """ + + def __init__(self, archive_path) -> None: + """Initialize workspace context manager.""" + self.cwd = os.getcwd() + + self.workspace_name = os.path.basename(archive_path).split('.')[0] + if os.path.exists(self.workspace_name): + shutil.rmtree(self.workspace_name) + os.makedirs(self.workspace_name) + shutil.unpack_archive(archive_path, self.workspace_name) + + os.remove(archive_path) + + def __enter__(self): + """Enter workspace context manager.""" + os.chdir(f'{self.cwd}/{self.workspace_name}') + + def __exit__(self, exc_type, exc_value, traceback): + """Exit workspace context manager.""" + os.chdir(self.cwd) + shutil.rmtree(self.workspace_name) + + def split_tensor_dict_into_floats_and_non_floats(tensor_dict): """ Split the tensor dictionary into float and non-floating point values. diff --git a/setup.cfg b/setup.cfg index 7d2b39a206..ae06205e93 100644 --- a/setup.cfg +++ b/setup.cfg @@ -7,7 +7,7 @@ ignore = W503, N812 inline-quotes = ' multiline-quotes = ' docstring-quotes = """ -exclude = *_pb2* +exclude = *_pb2*,tests/github/interactive_api,tests/github/interactive_api_director max-line-length = 99 avoid-escape = False import-order-style = smarkets diff --git a/setup.py b/setup.py index 349cc12b6b..20531b565c 100644 --- a/setup.py +++ b/setup.py @@ -26,6 +26,9 @@ 'openfl.component.assigner', 'openfl.component.aggregator', 'openfl.component.collaborator', + 'openfl.component.envoy', + 'openfl.component.director', + 'openfl.component.ca', 'openfl.utilities', 'openfl.utilities.optimizers.torch', 'openfl.utilities.optimizers.keras', @@ -53,8 +56,8 @@ 'numpy', 'pandas', 'protobuf', - 'grpcio==1.34.0', - 'grpcio-tools==1.34.0', + 'grpcio~=1.34.0', + 'grpcio-tools~=1.34.0', 'rich==9.1.0', 'tqdm', 'scikit-learn', @@ -64,6 +67,9 @@ 'flatten_json', 'cryptography>=3.4.6', 'cloudpickle', + 'tensorboardX', + 'tensorboard', + 'requests' ], python_requires='>=3.6, <3.9', project_urls={

Scalable PKIs In avoid manual copying for current PKIs/certificate exchange between Aggregator & Collaborators we need an automatic system for that.

2021-06-04T07:41:00

0.0

[]

securefederatedai/openfl

securefederatedai__openfl-62

6a1ea6979dad2f181a000874f1cbf2004e18652f

diff --git a/docs/advanced_topics.rst b/docs/advanced_topics.rst index 7506d2d0a2..2c16c5e7c0 100644 --- a/docs/advanced_topics.rst +++ b/docs/advanced_topics.rst @@ -12,5 +12,6 @@ Advanced Topics multiple_plans compression_settings + overriding_agg_fn diff --git a/docs/overriding_agg_fn.rst b/docs/overriding_agg_fn.rst new file mode 100644 index 0000000000..a4c7d0de82 --- /dev/null +++ b/docs/overriding_agg_fn.rst @@ -0,0 +1,76 @@ +.. # Copyright (C) 2020-2021 Intel Corporation +.. # SPDX-License-Identifier: Apache-2.0 + +.. _overriding_agg_fn: + +Overriding aggregation function +####################### + +Usage +===================== +|productName| allows developers to use custom aggregation functions per task. +In order to do this, you should create an implementation of :class:`openfl.component.aggregation_functions.AggregationFunctionInterface` +and pass it as ``tasks.{task_name}.aggregation_type`` parameter of ``override_config`` keyword argument of :func:`openfl.native.run_experiment` native function. + +.. warning:: + + Currently custom aggregation functions supported in native API (``fx.run_experiment(...)``) only. + Command Line Interface always uses FedAvg and this behavior cannot be overriden due to disability of serialization and deserialization of complex Python objects in Plan files. + +``AggregationFunctionInterface`` requires a single ``call`` function. +This function receives tensors for a single parameter from multiple collaborators with additional metadata (see definition of :meth:`openfl.component.aggregation_functions.AggregationFunctionInterface.call`) and returns a single tensor that represents the result of aggregation. + +.. note:: + By default, we use weighted averaging with weights equal to data parts (FedAvg). + +Example +======================= + +Below is an example of custom tensor clipping aggregation function that multiplies all local tensors by 0.3 and averages them according to weights equal to data parts to produce the resulting global tensor. + +.. code-block:: python + + from openfl.component.aggregation_functions import AggregationFunctionInterface + import numpy as np + + class ClippedAveraging(AggregationFunctionInterface): + def __init__(self, ratio): + self.ratio = ratio + + def call(self, + local_tensors, + db_iterator, + tensor_name, + fl_round, + *__): + """Aggregate tensors. + + Args: + local_tensors(list[openfl.utilities.LocalTensor]): List of local tensors to aggregate. + db_iterator: iterator over history of all tensors. + Columns: ['tensor_name', 'round', 'tags', 'nparray'] + tensor_name: name of the tensor + fl_round: round number + tags: tuple of tags for this tensor + """ + clipped_tensors = [] + previous_tensor_value = None + for record in db_iterator: + if ( + record['round'] == (fl_round - 1) + and record['tensor_name'] == tensor_name + and 'aggregated' in record['tags'] + and 'delta' not in record['tags'] + ): + previous_tensor_value = record['nparray'] + weights = [] + for local_tensor in local_tensors: + prev_tensor = previous_tensor_value if previous_tensor_value is not None else local_tensor.tensor + delta = local_tensor.tensor - prev_tensor + new_tensor = prev_tensor + delta * self.ratio + clipped_tensors.append(new_tensor) + weights.append(local_tensor.weight) + + return np.average(clipped_tensors, weights=weights, axis=0) + +Full implementation can be found at ``openfl-tutorials/Federated_Pytorch_MNIST_custom_aggregation_Tutorial.ipynb`` \ No newline at end of file diff --git a/docs/running_the_federation.notebook.rst b/docs/running_the_federation.notebook.rst index 22f2cb4738..eecbe0c376 100644 --- a/docs/running_the_federation.notebook.rst +++ b/docs/running_the_federation.notebook.rst @@ -47,7 +47,7 @@ At this point you may be wondering what goes into a FL.Plan, and how you can cus .. code-block:: python - print(json.dumps(fx.get_plan(), indent=4, sort_keys=True)) + print(fx.get_plan()) { "aggregator.settings.best_state_path": "save/keras_cnn_mnist_best.pbuf", diff --git a/openfl-tutorials/Federated_Keras_MNIST_Tutorial.ipynb b/openfl-tutorials/Federated_Keras_MNIST_Tutorial.ipynb index bec5ee5ad5..cb570c4f4a 100644 --- a/openfl-tutorials/Federated_Keras_MNIST_Tutorial.ipynb +++ b/openfl-tutorials/Federated_Keras_MNIST_Tutorial.ipynb @@ -205,8 +205,7 @@ "outputs": [], "source": [ "#Get the current values of the plan. Each of these can be overridden\n", - "import json\n", - "print(json.dumps(fx.get_plan(), indent=4, sort_keys=True))" + "print(fx.get_plan())" ] }, { diff --git a/openfl-tutorials/Federated_PyTorch_UNET_Tutorial.ipynb b/openfl-tutorials/Federated_PyTorch_UNET_Tutorial.ipynb index 378e624470..2a7c13a923 100644 --- a/openfl-tutorials/Federated_PyTorch_UNET_Tutorial.ipynb +++ b/openfl-tutorials/Federated_PyTorch_UNET_Tutorial.ipynb @@ -9,9 +9,21 @@ }, { "cell_type": "code", - "execution_count": null, + "execution_count": 1, "metadata": {}, - "outputs": [], + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "Requirement already satisfied: torch in /home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages (1.8.1)\n", + "Requirement already satisfied: typing-extensions in /home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages (from torch) (3.7.4.3)\n", + "Requirement already satisfied: numpy in /home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages (from torch) (1.18.5)\n", + "\u001b[33mWARNING: You are using pip version 21.0.1; however, version 21.1 is available.\n", + "You should consider upgrading via the '/home/itrushkin/.virtualenvs/openfl_research/bin/python -m pip install --upgrade pip' command.\u001b[0m\n" + ] + } + ], "source": [ "# Install dependencies if not already installed\n", "!pip install torch" @@ -26,9 +38,111 @@ }, { "cell_type": "code", - "execution_count": null, + "execution_count": 2, "metadata": {}, - "outputs": [], + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "Creating Workspace Directories\n", + "Creating Workspace Templates\n", + "Successfully installed packages from /home/itrushkin/.local/workspace/requirements.txt.\n", + "\n", + "New workspace directory structure:\n", + "workspace\n", + "├── plan\n", + "│ ├── defaults\n", + "│ │ ├── data_loader.yaml\n", + "│ │ ├── network.yaml\n", + "│ │ ├── tasks_keras.yaml\n", + "│ │ ├── collaborator.yaml\n", + "│ │ ├── assigner.yaml\n", + "│ │ ├── task_runner.yaml\n", + "│ │ ├── defaults\n", + "│ │ ├── tasks_tensorflow.yaml\n", + "│ │ ├── aggregator.yaml\n", + "│ │ ├── tasks_torch.yaml\n", + "│ │ └── tasks_fast_estimator.yaml\n", + "│ ├── plan.yaml\n", + "│ ├── data.yaml\n", + "│ └── cols.yaml\n", + "├── logs\n", + "├── agg_to_col_two_signed_cert.zip\n", + "├── data\n", + "│ └── MNIST\n", + "│ ├── raw\n", + "│ └── processed\n", + "├── cert\n", + "│ ├── ca\n", + "│ │ ├── signing-ca.crt\n", + "│ │ ├── root-ca.crt\n", + "│ │ ├── signing-ca\n", + "│ │ ├── signing-ca.csr\n", + "│ │ └── root-ca\n", + "│ ├── server\n", + "│ │ ├── agg_nnlicv611.inn.intel.com.csr\n", + "│ │ ├── agg_nnlicv611.inn.intel.com.key\n", + "│ │ └── agg_nnlicv611.inn.intel.com.crt\n", + "│ ├── client\n", + "│ │ ├── col_two.crt\n", + "│ │ ├── col_one.crt\n", + "│ │ ├── col_one.key\n", + "│ │ └── col_two.key\n", + "│ ├── cert_chain.crt\n", + "│ ├── col_one.csr\n", + "│ └── col_two.csr\n", + "├── code\n", + "│ ├── __init__.py\n", + "│ ├── data_loader.py\n", + "│ ├── mnist_utils.py\n", + "│ ├── fed_unet_runner.py\n", + "│ ├── ptmnist_inmemory.py\n", + "│ ├── pt_cnn.py\n", + "│ └── pt_unet_parts.py\n", + "├── final_pytorch_model\n", + "├── .workspace\n", + "├── requirements.txt\n", + "├── agg_to_col_one_signed_cert.zip\n", + "└── save\n", + " ├── torch_cnn_mnist_init.pbuf\n", + " ├── torch_cnn_mnist_best.pbuf\n", + " └── torch_cnn_mnist_last.pbuf\n", + "\n", + "15 directories, 42 files\n", + "Setting Up Certificate Authority...\n", + "\n", + "1. Create Root CA\n", + "1.1 Create Directories\n", + "1.2 Create Database\n", + "1.3 Create CA Request and Certificate\n", + "2. Create Signing Certificate\n", + "2.1 Create Directories\n", + "2.2 Create Database\n", + "2.3 Create Signing Certificate CSR\n", + "2.4 Sign Signing Certificate CSR\n", + "3 Create Certificate Chain\n", + "\n", + "Done.\n", + "Creating AGGREGATOR certificate key pair with following settings: CN=\u001b[31mnnlicv611.inn.intel.com\u001b[0m, SAN=\u001b[31mDNS:nnlicv611.inn.intel.com\u001b[0m\n", + " Writing AGGREGATOR certificate key pair to: \u001b[32mcert/server\u001b[0m\n", + "The CSR Hash for file \u001b[32mserver/agg_nnlicv611.inn.intel.com.csr\u001b[0m = \u001b[31mcabebd7a61c27b859c152be3573d77664096762d25e616294f56273fe65bd21d9afe883f2e745c4784a1fa10620d79d9\u001b[0m\n", + " Signing AGGREGATOR certificate\n", + "Creating COLLABORATOR certificate key pair with following settings: CN=\u001b[31mone\u001b[0m, SAN=\u001b[31mDNS:one\u001b[0m\n", + " Moving COLLABORATOR certificate to: \u001b[32mcert/col_one\u001b[0m\n", + "The CSR Hash for file \u001b[32mcol_one.csr\u001b[0m = \u001b[31m5e39b9007a7cedcaab3330907cbf1f471ae221483b2810de42483aec960171e8a95a3e67dd7985477592c873bf8876e2\u001b[0m\n", + " Signing COLLABORATOR certificate\n", + "\n", + "Registering \u001b[32mone\u001b[0m in \u001b[32mplan/cols.yaml\u001b[0m\n", + "Creating COLLABORATOR certificate key pair with following settings: CN=\u001b[31mtwo\u001b[0m, SAN=\u001b[31mDNS:two\u001b[0m\n", + " Moving COLLABORATOR certificate to: \u001b[32mcert/col_two\u001b[0m\n", + "The CSR Hash for file \u001b[32mcol_two.csr\u001b[0m = \u001b[31m54f92e51abdfa46fc18339b0c59f44af5bd55ce3c562daf0254208b49545db231c8ea963e0b2383272ea74a5aa2faa3f\u001b[0m\n", + " Signing COLLABORATOR certificate\n", + "\n", + "Registering \u001b[32mtwo\u001b[0m in \u001b[32mplan/cols.yaml\u001b[0m\n" + ] + } + ], "source": [ "import openfl.native as fx\n", "\n", @@ -38,7 +152,7 @@ }, { "cell_type": "code", - "execution_count": null, + "execution_count": 3, "metadata": { "code_folding": [] }, @@ -73,9 +187,2069 @@ }, { "cell_type": "code", - "execution_count": null, + "execution_count": 4, "metadata": {}, - "outputs": [], + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "--2021-04-29 17:42:32-- https://datasets.simula.no/hyper-kvasir/hyper-kvasir-segmented-images.zip\n", + "Resolving proxy-dmz.intel.com (proxy-dmz.intel.com)... 10.102.248.16\n", + "Connecting to proxy-dmz.intel.com (proxy-dmz.intel.com)|10.102.248.16|:912... connected.\n", + "Proxy request sent, awaiting response... 200 OK\n", + "Length: 46251583 (44M) [application/zip]\n", + "Saving to: ‘kvasir.zip’\n", + "\n", + "kvasir.zip 100%[===================>] 44,11M 514KB/s in 98s \n", + "\n", + "2021-04-29 17:44:11 (460 KB/s) - ‘kvasir.zip’ saved [46251583/46251583]\n", + "\n", + "Archive: kvasir.zip\n", + " creating: ./data/segmented-images/\n", + " inflating: ./data/segmented-images/bounding-boxes.json \n", + " creating: ./data/segmented-images/masks/\n", + " inflating: ./data/segmented-images/masks/cju88oh0po9gq0801nge4tgr1.jpg \n", + " inflating: ./data/segmented-images/masks/cju34uhepd3dd0799hs8782ad.jpg \n", + " inflating: ./data/segmented-images/masks/cju8bgdmqrksy0801tozdmraa.jpg \n", + " inflating: ./data/segmented-images/masks/cju5ddda9bkkt0850enzwatb1.jpg \n", + " inflating: ./data/segmented-images/masks/cju0s690hkp960855tjuaqvv0.jpg \n", + " inflating: ./data/segmented-images/masks/cju5nyu31gv8e0871zpk74a2n.jpg \n", + " inflating: ./data/segmented-images/masks/cju358pwtdby20878cg7nm0np.jpg \n", + " inflating: ./data/segmented-images/masks/cju5cu8qkb84x08186jwo8yin.jpg \n", + " inflating: ./data/segmented-images/masks/cju77j66ez52p08019xygi0co.jpg \n", + " inflating: ./data/segmented-images/masks/cju42xpi8lw4w0871ve317a1p.jpg \n", + " inflating: ./data/segmented-images/masks/cju884985nlmx0817vzpax3y4.jpg \n", + " inflating: ./data/segmented-images/masks/cju5fu081d8gc0818l3yylujk.jpg \n", + " inflating: ./data/segmented-images/masks/cju8a2itsq4dv0755ntlovpxe.jpg \n", + " inflating: ./data/segmented-images/masks/cju5bj926aiec07559rshy4wa.jpg \n", + " inflating: ./data/segmented-images/masks/cju1hirfi7ekp0855q0vgm9qq.jpg \n", + " inflating: ./data/segmented-images/masks/cju85bf1algsq0871y9gtlq97.jpg \n", + " inflating: ./data/segmented-images/masks/cju7fd6yt2p740987wkr8exo1.jpg \n", + " inflating: ./data/segmented-images/masks/cju1cqc7n4gpy0855jt246k68.jpg \n", + " inflating: ./data/segmented-images/masks/cju41z76wlgbz0801qdetlvby.jpg \n", + " inflating: ./data/segmented-images/masks/cjyzuio1qgh040763k56deohv.jpg \n", + " inflating: ./data/segmented-images/masks/cju43jcqim2cp08172dvjvyui.jpg \n", + " inflating: ./data/segmented-images/masks/cju8b7aqtr4a00987coba14b7.jpg \n", + " inflating: ./data/segmented-images/masks/cju2zkpdl9h7t0799ix60teqg.jpg \n", + " inflating: ./data/segmented-images/masks/cju5g163vd6mt0817uccuga6u.jpg \n", + " inflating: ./data/segmented-images/masks/cju2z2x3nvd3c099350zgty7w.jpg \n", + " inflating: ./data/segmented-images/masks/cju30u1hbakn808019g15nb8b.jpg \n", + " inflating: ./data/segmented-images/masks/cju2pag1f0s4r0878h52uq83s.jpg \n", + " inflating: ./data/segmented-images/masks/cju5uget8krjy0818kvywd0zu.jpg \n", + " inflating: ./data/segmented-images/masks/cju5yimthmlv80850zhoc90c2.jpg \n", + " inflating: ./data/segmented-images/masks/cju2txjfzv60w098839dcimys.jpg \n", + " inflating: ./data/segmented-images/masks/cju0sxqiclckk08551ycbwhno.jpg \n", + " inflating: ./data/segmented-images/masks/cju1brhsj3rls0855a1vgdlen.jpg \n", + " inflating: ./data/segmented-images/masks/cju8bff9nrfi10850fmfzbf8v.jpg \n", + " inflating: ./data/segmented-images/masks/cju8at3s1qqqx0850hcq8nmnq.jpg \n", + " inflating: ./data/segmented-images/masks/cju1cvkfwqrec0993wbp1jlzm.jpg \n", + " inflating: ./data/segmented-images/masks/cju2zy1e49pqk0878t6ncqn12.jpg \n", + " inflating: ./data/segmented-images/masks/cju3v72v5h1qz0818fggilwtq.jpg \n", + " inflating: ./data/segmented-images/masks/cju8b5p40r2c60987ofa0mu03.jpg \n", + " inflating: ./data/segmented-images/masks/cju1hhj6mxfp90835n3wofrap.jpg \n", + " inflating: ./data/segmented-images/masks/cju6z600qwh4z081700qimgl9.jpg \n", + " inflating: ./data/segmented-images/masks/cju3ul8dogf1z09872y2ecowp.jpg \n", + " inflating: ./data/segmented-images/masks/cju1cu1u2474n0878tt7v4tdr.jpg \n", + " inflating: ./data/segmented-images/masks/cju2t3ibkuecr0835o7si16zv.jpg \n", + " inflating: ./data/segmented-images/masks/cju7b4mtw1n9n080186209f3d.jpg \n", + " inflating: ./data/segmented-images/masks/cju8ayeq7r1fb0818z1junacy.jpg \n", + " inflating: ./data/segmented-images/masks/cju5vcmrqla7i0817x4sp4pqw.jpg \n", + " inflating: ./data/segmented-images/masks/cju83ipu3jwpx0801z5pvguf8.jpg \n", + " inflating: ./data/segmented-images/masks/cju7cp6dw244p0818gncdol4m.jpg \n", + " inflating: ./data/segmented-images/masks/cju2m56cryvqd0801gtn2yp8t.jpg \n", + " inflating: ./data/segmented-images/masks/cju3381d8bz3h07991xtl7ra0.jpg \n", + " inflating: ./data/segmented-images/masks/cju5wqonpm0e60801z88ewmy1.jpg \n", + " inflating: ./data/segmented-images/masks/cju2zr3c3vwb00993jn06bbaz.jpg \n", + " inflating: ./data/segmented-images/masks/cju2zxja9w1eh09933609ho9z.jpg \n", + " inflating: ./data/segmented-images/masks/cju5v8pgplg6k0755rvi2t63h.jpg \n", + " inflating: ./data/segmented-images/masks/cju34fojcctcf0799ebolbvkn.jpg \n", + " inflating: ./data/segmented-images/masks/cju8c5zcbsdfz0801o5t6jag1.jpg \n", + " inflating: ./data/segmented-images/masks/cju5y84q3mdv50817eyp82xf3.jpg \n", + " inflating: ./data/segmented-images/masks/cju88t4fvokxf07558ymyh281.jpg \n", + " inflating: ./data/segmented-images/masks/cju1c8ffau5770835g0g343o8.jpg \n", + " inflating: ./data/segmented-images/masks/cju2hfqnmhisa0993gpleeldd.jpg \n", + " inflating: ./data/segmented-images/masks/cju83h9ysjwe808716nt35oah.jpg \n", + " inflating: ./data/segmented-images/masks/cju1efbr0rqxz09931z0lf4vf.jpg \n", + " inflating: ./data/segmented-images/masks/cju2upu4evw7g08358guwozxv.jpg \n", + " inflating: ./data/segmented-images/masks/cju5wmvsdlx1j0871npgj8j4b.jpg \n", + " inflating: ./data/segmented-images/masks/cju8b3ka8r64u0801fh18hk7l.jpg \n", + " inflating: ./data/segmented-images/masks/cju8a3nhbpwnb0850d37fo2na.jpg \n", + " inflating: ./data/segmented-images/masks/cju2ntxtdzlvu0799xl3j9pan.jpg \n", + " inflating: ./data/segmented-images/masks/cju5vzjoslpj708186z2fusmz.jpg \n", + " inflating: ./data/segmented-images/masks/cju1fuoa4wmc50835qfd11sp9.jpg \n", + " inflating: ./data/segmented-images/masks/cju5vwbr4lhqn0987a1pji0ux.jpg \n", + " inflating: ./data/segmented-images/masks/cju2lyynuymli0855g7fxgbhe.jpg \n", + " inflating: ./data/segmented-images/masks/cju41kd7yl4nm0850gil5qqwh.jpg \n", + " inflating: ./data/segmented-images/masks/cju33jon3ygbj0993pu22a4k6.jpg \n", + " inflating: ./data/segmented-images/masks/cju1gi7jlwyld0835cdf6g6qz.jpg \n", + " inflating: ./data/segmented-images/masks/cju7dqcwi2dz00850gcmr2ert.jpg \n", + " inflating: ./data/segmented-images/masks/cju83kxitjv340987z09m0ezy.jpg \n", + " inflating: ./data/segmented-images/masks/cju2mh8t6p07008350e01tx2a.jpg \n", + " inflating: ./data/segmented-images/masks/cju40taxlkrho0987smigg0x0.jpg \n", + " inflating: ./data/segmented-images/masks/cju31rb7vb6110801p9rhacuw.jpg \n", + " inflating: ./data/segmented-images/masks/cju34zivp3fq80988opxbaqyn.jpg \n", + " inflating: ./data/segmented-images/masks/cju87mrypnb1e0818scv1mxxg.jpg \n", + " inflating: ./data/segmented-images/masks/cju1fjsb4sipq09931lvd8e41.jpg \n", + " inflating: ./data/segmented-images/masks/cju2zrojo9kcd0878ld2epejq.jpg \n", + " inflating: ./data/segmented-images/masks/cju1cbokpuiw70988j4lq1fpi.jpg \n", + " inflating: ./data/segmented-images/masks/cju7b5afm1nfw0801xqm8bf8q.jpg \n", + " inflating: ./data/segmented-images/masks/cju5waeduln160817w0agirve.jpg \n", + " inflating: ./data/segmented-images/masks/cju7do8c72dbo0801vxfzxdc4.jpg \n", + " inflating: ./data/segmented-images/masks/cju15l5ubz9yh0855b3ivdpse.jpg \n", + " inflating: ./data/segmented-images/masks/cju1cdxvz48hw0801i0fjwcnk.jpg \n", + " inflating: ./data/segmented-images/masks/cju2yw4s7z7p20988lmf2gdgd.jpg \n", + " inflating: ./data/segmented-images/masks/cju3v664kh0px0818y4y7wolf.jpg \n", + " inflating: ./data/segmented-images/masks/cju5xneamme2p0801qdf7fdwv.jpg \n", + " inflating: ./data/segmented-images/masks/cju2lcyfgkf5809932fn9gucn.jpg \n", + " inflating: ./data/segmented-images/masks/cju85a8h8llwm07559wxg4t5w.jpg \n", + " inflating: ./data/segmented-images/masks/cju426tomlhll0818fc0i7nvh.jpg \n", + " inflating: ./data/segmented-images/masks/cju0vtox5ain6099360pu62rp.jpg \n", + " inflating: ./data/segmented-images/masks/cju1b75x63ddl0799sdp0i2j3.jpg \n", + " inflating: ./data/segmented-images/masks/cju2phaksnahz0993yxogjcpv.jpg \n", + " inflating: ./data/segmented-images/masks/cju33yemn2qb20988wfjxximx.jpg \n", + " inflating: ./data/segmented-images/masks/cju3128yi0rpu0988o4oo5n8n.jpg \n", + " inflating: ./data/segmented-images/masks/cju7afqon1ip40850ue2308b6.jpg \n", + " inflating: ./data/segmented-images/masks/cju5hl8nee8a40755fm8qjj0o.jpg \n", + " inflating: ./data/segmented-images/masks/cju8abobpqbir08189u01huru.jpg \n", + " inflating: ./data/segmented-images/masks/cju7awzmu1ncs0871hziy65zx.jpg \n", + " inflating: ./data/segmented-images/masks/cju7ejm2l2ncl0801wq6y84nw.jpg \n", + " inflating: ./data/segmented-images/masks/cju8bi8q7rlmn0871abc5ch8k.jpg \n", + " inflating: ./data/segmented-images/masks/cju5xq3tdm9fn0987pbedxdg5.jpg \n", + " inflating: ./data/segmented-images/masks/cju6vvxsev9y30987kespucdg.jpg \n", + " inflating: ./data/segmented-images/masks/cju5fi0yxd3ei0801v7u0yudn.jpg \n", + " inflating: ./data/segmented-images/masks/cju7d7aut2a2p0818z4uxc6cd.jpg \n", + " inflating: ./data/segmented-images/masks/cju8auylgqx0z0871u4o4db7o.jpg \n", + " inflating: ./data/segmented-images/masks/cju1dg44i4z3w0801nyz4p6zf.jpg \n", + " inflating: ./data/segmented-images/masks/cju1dia8hvc6v098827mgffnm.jpg \n", + " inflating: ./data/segmented-images/masks/cju2trtjf4qjd0878a2zle9v9.jpg \n", + " inflating: ./data/segmented-images/masks/cju2t2ivz43i10878zeg8r1br.jpg \n", + " inflating: ./data/segmented-images/masks/cju31t8xd17bk0835rnb893jk.jpg \n", + " inflating: ./data/segmented-images/masks/cju8axq24r4an0755yhv9d4ly.jpg \n", + " inflating: ./data/segmented-images/masks/cju5gucasds9d0801019axylx.jpg \n", + " inflating: ./data/segmented-images/masks/cju6us80mv1b50871ebyq2wxa.jpg \n", + " inflating: ./data/segmented-images/masks/cju2s2527pfyr0993l3h1149a.jpg \n", + " inflating: ./data/segmented-images/masks/cju5klveuff6w0871wbibgh3m.jpg \n", + " inflating: ./data/segmented-images/masks/cju88fpm4o0tl0871w1i6a4ds.jpg \n", + " inflating: ./data/segmented-images/masks/cju1fm3id6gl50801r3fok20c.jpg \n", + " inflating: ./data/segmented-images/masks/cju16d65tzw9d0799ouslsw25.jpg \n", + " inflating: ./data/segmented-images/masks/cju6w733bveoz0817e600tw72.jpg \n", + " inflating: ./data/segmented-images/masks/cju2qs32r1vys07999conmbvx.jpg \n", + " inflating: ./data/segmented-images/masks/cju2zdvjn9h7r08553cp4eed5.jpg \n", + " inflating: ./data/segmented-images/masks/cju2hgsptlfam0835o3b59h1o.jpg \n", + " inflating: ./data/segmented-images/masks/cju5yhgznmkzb0801cji2vi8j.jpg \n", + " inflating: ./data/segmented-images/masks/cju7dvl5m2n4t0755hlnnjjet.jpg \n", + " inflating: ./data/segmented-images/masks/cju6ut4l8va6y0755tyw3vfqq.jpg \n", + " inflating: ./data/segmented-images/masks/cju6ur9l9v9jq0755paud9uka.jpg \n", + " inflating: ./data/segmented-images/masks/cju18ibp219ub08783i6o98g7.jpg \n", + " inflating: ./data/segmented-images/masks/cju2hewssldzx0835ep795xu0.jpg \n", + " inflating: ./data/segmented-images/masks/cju3xvoo2iqlc0817eku2r3wl.jpg \n", + " inflating: ./data/segmented-images/masks/cju2r7h21sj9608354gzks3ae.jpg \n", + " inflating: ./data/segmented-images/masks/cju5k7r0yf98c09878csbxb4d.jpg \n", + " inflating: ./data/segmented-images/masks/cju43h43am1dy08176gwfhmnt.jpg \n", + " inflating: ./data/segmented-images/masks/cju2rnkt22xep0801as160g9t.jpg \n", + " inflating: ./data/segmented-images/masks/cju32l161bi1v07990vm376in.jpg \n", + " inflating: ./data/segmented-images/masks/cju5k3j3uf6de0817hszzfr7n.jpg \n", + " inflating: ./data/segmented-images/masks/cju2hugv9vget0799hhk7ksvg.jpg \n", + " inflating: ./data/segmented-images/masks/cju7aez2x1jtj0871ztezs3oi.jpg \n", + " inflating: ./data/segmented-images/masks/cju1euuc65wm00799m4sjdnnn.jpg \n", + " inflating: ./data/segmented-images/masks/cju330ofbc2l30801th5g3hw6.jpg \n", + " inflating: ./data/segmented-images/masks/cju43lcnum9y10755bjs7z87f.jpg \n", + " inflating: ./data/segmented-images/masks/cju14g8o4xui30878gkgbrvqj.jpg \n", + " inflating: ./data/segmented-images/masks/cju2z45kuzf6d0988nz2c819m.jpg \n", + " inflating: ./data/segmented-images/masks/cju30j1rgadut0801vuyrsnt8.jpg \n", + " inflating: ./data/segmented-images/masks/cju8cgi2kspp308011nxdtjp6.jpg \n", + " inflating: ./data/segmented-images/masks/cju1f79yhsb5w0993txub59ol.jpg \n", + " inflating: ./data/segmented-images/masks/cju33x0f22peh0988g0ln7w5v.jpg \n", + " inflating: ./data/segmented-images/masks/cju85c2d4ln1b0755zz1z3onx.jpg \n", + " inflating: ./data/segmented-images/masks/cju2syxa93yw40799x2iuwabz.jpg \n", + " inflating: ./data/segmented-images/masks/cju3xuj20ivgp0818mij8bjrd.jpg \n", + " inflating: ./data/segmented-images/masks/cju33o12x2jm50988944mxq0v.jpg \n", + " inflating: ./data/segmented-images/masks/cju5thdbrjp1108715xdfx356.jpg \n", + " inflating: ./data/segmented-images/masks/cju17r8il13910799dr2wme2e.jpg \n", + " inflating: ./data/segmented-images/masks/cju3y0pjrj1c30755nxekxccj.jpg \n", + " inflating: ./data/segmented-images/masks/cju7f0ec32txj08184asb8w5f.jpg \n", + " inflating: ./data/segmented-images/masks/cju87zv8lni0o0850hbbecbq6.jpg \n", + " inflating: ./data/segmented-images/masks/cju88trl3ogi208716qvti51b.jpg \n", + " inflating: ./data/segmented-images/masks/cju8bj2ssrmlm0871gc2ug2rs.jpg \n", + " inflating: ./data/segmented-images/masks/cju1egh885m1l0855ci1lt37c.jpg \n", + " inflating: ./data/segmented-images/masks/cju41p90plcsx08018cnzpndc.jpg \n", + " inflating: ./data/segmented-images/masks/cju0tl3uz8blh0993wxvn7ly3.jpg \n", + " inflating: ./data/segmented-images/masks/cju2hx006vidl0799igm81vmh.jpg \n", + " inflating: ./data/segmented-images/masks/cju2spdagu1l50835da1f46fr.jpg \n", + " inflating: ./data/segmented-images/masks/cju34o6dbd2lo0855aqlcy1hs.jpg \n", + " inflating: ./data/segmented-images/masks/cju5ufn3skquf0818dhapnhba.jpg \n", + " inflating: ./data/segmented-images/masks/cju30mm25a53s0799qa5wiqe8.jpg \n", + " inflating: ./data/segmented-images/masks/cju1ddr6p4k5z08780uuuzit2.jpg \n", + " inflating: ./data/segmented-images/masks/cju2zp89k9q1g0855k1x0f1xa.jpg \n", + " inflating: ./data/segmented-images/masks/cju88nroho44508500129f1nh.jpg \n", + " inflating: ./data/segmented-images/masks/cju76o55nymqd0871h31sph9w.jpg \n", + " inflating: ./data/segmented-images/masks/cju160wshltz10993i1gmqxbe.jpg \n", + " inflating: ./data/segmented-images/masks/cju2ouil2mssu0993hvxsed6d.jpg \n", + " inflating: ./data/segmented-images/masks/cju1amqw6p8pw0993d9gc5crl.jpg \n", + " inflating: ./data/segmented-images/masks/cju2rqo702wpx0855fn7d5cxh.jpg \n", + " inflating: ./data/segmented-images/masks/cju843yjskhq30818qre4rwm2.jpg \n", + " inflating: ./data/segmented-images/masks/cju2hjrqcvi2j0801bx1i6gxg.jpg \n", + " inflating: ./data/segmented-images/masks/cju6v5ilsv8hk0850rb5sgh6o.jpg \n", + " inflating: ./data/segmented-images/masks/cju2rmd2rsw9g09888hh1efu0.jpg \n", + " inflating: ./data/segmented-images/masks/cju43kj2pm34f0850l28ahpni.jpg \n", + " inflating: ./data/segmented-images/masks/cju2rxm8rpbaf0993o3qr2oph.jpg \n", + " inflating: ./data/segmented-images/masks/cju5u4pywk81x0817vn9pe14z.jpg \n", + " inflating: ./data/segmented-images/masks/cju88vx2uoocy075531lc63n3.jpg \n", + " inflating: ./data/segmented-images/masks/cju30ov1oah920801mi8thuyg.jpg \n", + " inflating: ./data/segmented-images/masks/cju7ae7bq1f820987toc8si1d.jpg \n", + " inflating: ./data/segmented-images/masks/cju3xiic0ilzp0850lusrb42j.jpg \n", + " inflating: ./data/segmented-images/masks/cju2urqpwvxw70835rvndvtsi.jpg \n", + " inflating: ./data/segmented-images/masks/cju35740hzm0g0993zl5ic246.jpg \n", + " inflating: ./data/segmented-images/masks/cju2rn0hasxri0835nfy3buay.jpg \n", + " inflating: ./data/segmented-images/masks/cju357rxxdaz30878y2esjpjt.jpg \n", + " inflating: ./data/segmented-images/masks/cju7fnfv02tt90801djnix9m8.jpg \n", + " inflating: ./data/segmented-images/masks/cju89z6pqpqfx0817mfv8ixjc.jpg \n", + " inflating: ./data/segmented-images/masks/cju7dz5yy2i7z0801ausi7rna.jpg \n", + " inflating: ./data/segmented-images/masks/cju8ceacrsqkr0755hdz145es.jpg \n", + " inflating: ./data/segmented-images/masks/cju17hw9hr9c5098800fu4u8e.jpg \n", + " inflating: ./data/segmented-images/masks/cju8b542nr81x0871uxnkm9ih.jpg \n", + " inflating: ./data/segmented-images/masks/cju5uhrdwkmsu0817ervv91l8.jpg \n", + " inflating: ./data/segmented-images/masks/cju85dx63lic408017f0l0400.jpg \n", + " inflating: ./data/segmented-images/masks/cju1egx9pvz2n0988eoy8jp23.jpg \n", + " inflating: ./data/segmented-images/masks/cju88v2f9oi8w0871hx9auh01.jpg \n", + " inflating: ./data/segmented-images/masks/cju7ekbo32pft0871fv7kzwb9.jpg \n", + " inflating: ./data/segmented-images/masks/cju5uzmaol56l0817flxh4w9p.jpg \n", + " inflating: ./data/segmented-images/masks/cju6vvb8svhed0801jjcquh5e.jpg \n", + " inflating: ./data/segmented-images/masks/cju8brv16rx7f0818uf5n89pv.jpg \n", + " inflating: ./data/segmented-images/masks/cju2rzpsmtb0f0835jabkbao1.jpg \n", + " inflating: ./data/segmented-images/masks/cju16b6ynq8e40988m8vx0xnj.jpg \n", + " inflating: ./data/segmented-images/masks/cju8914beokbf0850isxpocrk.jpg \n", + " inflating: ./data/segmented-images/masks/cju8chxndsre008015uisl4si.jpg \n", + " inflating: ./data/segmented-images/masks/cju5bwhapakm90987c1v4z46a.jpg \n", + " inflating: ./data/segmented-images/masks/cju7da88w2eod0755wejzynvt.jpg \n", + " inflating: ./data/segmented-images/masks/cju8c6hnxsdvr0801wn0vrsa6.jpg \n", + " inflating: ./data/segmented-images/masks/cju2qu37qobl50993aw7ghcfq.jpg \n", + " inflating: ./data/segmented-images/masks/cju2wxv0hxs2f09884w48v8fi.jpg \n", + " inflating: ./data/segmented-images/masks/cju5ymyd8mmdc0801ry3by1xr.jpg \n", + " inflating: ./data/segmented-images/masks/cju5wrrs0m2af0818vmnajbtw.jpg \n", + " inflating: ./data/segmented-images/masks/cju7ff97z2ow40817u2r83my5.jpg \n", + " inflating: ./data/segmented-images/masks/cju2zpw4q9vzr0801p0lysjdl.jpg \n", + " inflating: ./data/segmented-images/masks/cju77b3wyz4160755qis4ljsb.jpg \n", + " inflating: ./data/segmented-images/masks/cju85citjlnfm0755i4rk5tqj.jpg \n", + " inflating: ./data/segmented-images/masks/cju887ftknop008177nnjt46y.jpg \n", + " inflating: ./data/segmented-images/masks/cju77re6fz5bb0817vp9redjg.jpg \n", + " inflating: ./data/segmented-images/masks/cju33qpdvc9g0087825jhf3s9.jpg \n", + " inflating: ./data/segmented-images/masks/cju8b0jr0r2oi0801jiquetd5.jpg \n", + " inflating: ./data/segmented-images/masks/cju32fhnhbds40799broyoptc.jpg \n", + " inflating: ./data/segmented-images/masks/cju6xmqd9w0250817l5kxfnsk.jpg \n", + " inflating: ./data/segmented-images/masks/cju2wve9v7esz0878mxsdcy04.jpg \n", + " inflating: ./data/segmented-images/masks/cju3xtufwiv9c0818djsc4cqd.jpg \n", + " inflating: ./data/segmented-images/masks/cju8dk7eztzup08182yxko5zh.jpg \n", + " inflating: ./data/segmented-images/masks/cju5hi52odyf90817prvcwg45.jpg \n", + " inflating: ./data/segmented-images/masks/cju7ezs7g2mxm098787atbran.jpg \n", + " inflating: ./data/segmented-images/masks/cju1b0y2e396p08558ois175d.jpg \n", + " inflating: ./data/segmented-images/masks/cju8aeei7q8k308173n9y4klv.jpg \n", + " inflating: ./data/segmented-images/masks/cju1h89h6xbnx08352k2790o9.jpg \n", + " inflating: ./data/segmented-images/masks/cju2xwm1s84l50799i60mq0pu.jpg \n", + " inflating: ./data/segmented-images/masks/cju88cddensj00987788yotmg.jpg \n", + " inflating: ./data/segmented-images/masks/cju3v3ac9gyz30755hfqwyp1i.jpg \n", + " inflating: ./data/segmented-images/masks/cjyzkpsbjdsjq07211dfi4sru.jpg \n", + " inflating: ./data/segmented-images/masks/cju30lncba3ny0878jwnous8n.jpg \n", + " inflating: ./data/segmented-images/masks/cju2lz8vqktne0993fuym6drw.jpg \n", + " inflating: ./data/segmented-images/masks/cju1fmsyf6gxb0801cimx2gle.jpg \n", + " inflating: ./data/segmented-images/masks/cju14pxbaoksp0835qzorx6g6.jpg \n", + " inflating: ./data/segmented-images/masks/cju1djtprvd7b0988thwwrg09.jpg \n", + " inflating: ./data/segmented-images/masks/cju7b2l561oas0871decgslaf.jpg \n", + " inflating: ./data/segmented-images/masks/cju1c4fcu40hl07992b8gj0c8.jpg \n", + " inflating: ./data/segmented-images/masks/cju7eueum2oqn0850rodmx8zo.jpg \n", + " inflating: ./data/segmented-images/masks/cju42py9mlqyd0818u3d1d7ga.jpg \n", + " inflating: ./data/segmented-images/masks/cju84hibuktj80871u519o71q.jpg \n", + " inflating: ./data/segmented-images/masks/cju42tauqlo5p08171l3cuo4b.jpg \n", + " inflating: ./data/segmented-images/masks/cju8bzzy2s66m08016z6mouqt.jpg \n", + " inflating: ./data/segmented-images/masks/cju41lojblbs307555jdci937.jpg \n", + " inflating: ./data/segmented-images/masks/cju85ia1slh220987y7c20sm2.jpg \n", + " inflating: ./data/segmented-images/masks/cju83vvmik9wa08710yeh7cuk.jpg \n", + " inflating: ./data/segmented-images/masks/cju1hp9i2xu8e0988u2dazk7m.jpg \n", + " inflating: ./data/segmented-images/masks/cju5b9oyda4yr0850g9viziyv.jpg \n", + " inflating: ./data/segmented-images/masks/cju2p4ddkmzxj0993p94o62av.jpg \n", + " inflating: ./data/segmented-images/masks/cju8adb60qbiu080188mxpf8d.jpg \n", + " inflating: ./data/segmented-images/masks/cju5es375cnzy0801nkq35ffs.jpg \n", + " inflating: ./data/segmented-images/masks/cju5wphwwlu3m0987hh3ltg88.jpg \n", + " inflating: ./data/segmented-images/masks/cju353d1eda8c07992afde611.jpg \n", + " inflating: ./data/segmented-images/masks/cju34ymm8d6700799uop0cw33.jpg \n", + " inflating: ./data/segmented-images/masks/cju785htizjzo08017tvlhtg4.jpg \n", + " inflating: ./data/segmented-images/masks/cju2qtee81yd708787bsjr75d.jpg \n", + " inflating: ./data/segmented-images/masks/cju34eqjpcpm508788b3lhp97.jpg \n", + " inflating: ./data/segmented-images/masks/cju6z2616wqbk07555bvnuyr1.jpg \n", + " inflating: ./data/segmented-images/masks/cju7aqkue1i2k09879uzcpt8r.jpg \n", + " inflating: ./data/segmented-images/masks/cju2r2obh2bjm08553kng0rh7.jpg \n", + " inflating: ./data/segmented-images/masks/cju17z0qongpa0993de4boim4.jpg \n", + " inflating: ./data/segmented-images/masks/cju774fmayxif0818u2g79usw.jpg \n", + " inflating: ./data/segmented-images/masks/cju1euant5l960878iqj5vvto.jpg \n", + " inflating: ./data/segmented-images/masks/cju2t16vuucaq0835xcpsivn2.jpg \n", + " inflating: ./data/segmented-images/masks/cju1g4nsb6ngy0799l4ezm8ab.jpg \n", + " inflating: ./data/segmented-images/masks/cju2wx0gh7fpz0878wwyd9ep8.jpg \n", + " inflating: ./data/segmented-images/masks/cju5cky5xb0ay0801oxet697t.jpg \n", + " inflating: ./data/segmented-images/masks/cju7ap09p1kz10850ldccjebj.jpg \n", + " inflating: ./data/segmented-images/masks/cju42qet0lsq90871e50xbnuv.jpg \n", + " inflating: ./data/segmented-images/masks/cju32pzh9bpw10855q4vaxfhe.jpg \n", + " inflating: ./data/segmented-images/masks/cju2rga4psq9n09881z519xx0.jpg \n", + " inflating: ./data/segmented-images/masks/cju76lsehyia10987u54vn8rb.jpg \n", + " inflating: ./data/segmented-images/masks/cju7ey10f2rvf0871bwbi9x82.jpg \n", + " inflating: ./data/segmented-images/masks/cju41s6nbleqy0755e2mslg0b.jpg \n", + " inflating: ./data/segmented-images/masks/cju40w3hbkwpn08015rbs3wko.jpg \n", + " inflating: ./data/segmented-images/masks/cju2m71z2ywwv080131bcrsd3.jpg \n", + " inflating: ./data/segmented-images/masks/cju7bd1qu1mx409877xjxibox.jpg \n", + " inflating: ./data/segmented-images/masks/cju7aifyo1p3n07552nxjx51f.jpg \n", + " inflating: ./data/segmented-images/masks/cju35ldepdtlm0801yv79y8vu.jpg \n", + " inflating: ./data/segmented-images/masks/cju2ij9uiic2l09933ljiv6gm.jpg \n", + " inflating: ./data/segmented-images/masks/cju7agj961l2r0818z29iq8yn.jpg \n", + " inflating: ./data/segmented-images/masks/cju6vrs1ov8cr098788h8gs6j.jpg \n", + " inflating: ./data/segmented-images/masks/cju7cue9b232j0801qdzk1ykj.jpg \n", + " inflating: ./data/segmented-images/masks/cju5vbo6jldrt0871jf6f1700.jpg \n", + " inflating: ./data/segmented-images/masks/cju3y9difj6th0801kd1rqm3w.jpg \n", + " inflating: ./data/segmented-images/masks/cju1871y11d6r0799k6cw4yze.jpg \n", + " inflating: ./data/segmented-images/masks/cju2nsmwjlzyl0993jl80chvz.jpg \n", + " inflating: ./data/segmented-images/masks/cju3521y5d5mq0878t3ezsu4p.jpg \n", + " inflating: ./data/segmented-images/masks/cju83rcnzkbsj0755x5anfrcg.jpg \n", + " inflating: ./data/segmented-images/masks/cju8bop5jrsid08716i24fqda.jpg \n", + " inflating: ./data/segmented-images/masks/cju8ando2qqdo0818ck7i1be1.jpg \n", + " inflating: ./data/segmented-images/masks/cju85nr8elly209872w9n5m0s.jpg \n", + " inflating: ./data/segmented-images/masks/cju35oyvd3y850988km12hdz1.jpg \n", + " inflating: ./data/segmented-images/masks/cju8c9akjsdjj0850s67uzlxq.jpg \n", + " inflating: ./data/segmented-images/masks/cju35eg0tdmjt085525sb4bua.jpg \n", + " inflating: ./data/segmented-images/masks/cju5bdwa3aatx0818b79i18zf.jpg \n", + " inflating: ./data/segmented-images/masks/cju5von04litr08718j8po40a.jpg \n", + " inflating: ./data/segmented-images/masks/cju32upim1z7u0988l883nqp6.jpg \n", + " inflating: ./data/segmented-images/masks/cju2zm0axztpe0988r8s9twjr.jpg \n", + " inflating: ./data/segmented-images/masks/cju7epwj82koz098713apjnzo.jpg \n", + " inflating: ./data/segmented-images/masks/cju84ffdzkrjn08183jh1fxmb.jpg \n", + " inflating: ./data/segmented-images/masks/cju2u73dj53oz0878486k8k4b.jpg \n", + " inflating: ./data/segmented-images/masks/cju2nyc5f02m40801ojqbtiea.jpg \n", + " inflating: ./data/segmented-images/masks/cju76l27oyrw907551ri2a7fl.jpg \n", + " inflating: ./data/segmented-images/masks/cju772304yw5t0818vbw8kkjf.jpg \n", + " inflating: ./data/segmented-images/masks/cju6xa0qmvzun0818xjukgncj.jpg \n", + " inflating: ./data/segmented-images/masks/cju7787c5yy3l080159mwqsnj.jpg \n", + " inflating: ./data/segmented-images/masks/cju2y26c588bo07993ksd8eoz.jpg \n", + " inflating: ./data/segmented-images/masks/cju5kre09fhka0850h7b1898j.jpg \n", + " inflating: ./data/segmented-images/masks/cju8cj10qsrau0871o2dr6ai1.jpg \n", + " inflating: ./data/segmented-images/masks/cju2pkwt3r8b90988v2ywq1px.jpg \n", + " inflating: ./data/segmented-images/masks/cju42u5bjlvi10801dc13sskp.jpg \n", + " inflating: ./data/segmented-images/masks/cju7dizi82h2i0755doucgnt3.jpg \n", + " inflating: ./data/segmented-images/masks/cju3x9lttikfb0818a0g104zn.jpg \n", + " inflating: ./data/segmented-images/masks/cju31ugmfb3dz0855xtqshki6.jpg \n", + " inflating: ./data/segmented-images/masks/cju424hy5lckr085073fva1ok.jpg \n", + " inflating: ./data/segmented-images/masks/cju7azuu31mia0801pf9ib5ed.jpg \n", + " inflating: ./data/segmented-images/masks/cju40u30gkuzc0871rq7t666d.jpg \n", + " inflating: ./data/segmented-images/masks/cju8apjewqrk00801k5d71gky.jpg \n", + " inflating: ./data/segmented-images/masks/cju1haab178i70799tk9z8y8x.jpg \n", + " inflating: ./data/segmented-images/masks/cju5wuhm1lwm40987vugqn3vv.jpg \n", + " inflating: ./data/segmented-images/masks/cju7etr3y2p4t0801cdzjj8ab.jpg \n", + " inflating: ./data/segmented-images/masks/cju1dhfok4mhe0878jlgrag0h.jpg \n", + " inflating: ./data/segmented-images/masks/cju2otvvv0l7z0855x7we8cb0.jpg \n", + " inflating: ./data/segmented-images/masks/cju5wj0faly5008187n6530af.jpg \n", + " inflating: ./data/segmented-images/masks/cju7ea4om2l910801bohqjccy.jpg \n", + " inflating: ./data/segmented-images/masks/cju1d50a94qf50855wsowacrc.jpg \n", + " inflating: ./data/segmented-images/masks/cju2oo0wh0bqy0878biujeyhe.jpg \n", + " inflating: ./data/segmented-images/masks/cju18kevfrojc0835bn90f1in.jpg \n", + " inflating: ./data/segmented-images/masks/cju5c7oijaqmq09878qwgqv8n.jpg \n", + " inflating: ./data/segmented-images/masks/cju7bduyq1rjf08719giru9ho.jpg \n", + " inflating: ./data/segmented-images/masks/cju5fw37edaae0801vkwvocn7.jpg \n", + " inflating: ./data/segmented-images/masks/cju83wwn1k55e0850kw6i2d81.jpg \n", + " inflating: ./data/segmented-images/masks/cju0ue769mxii08019zqgdbxn.jpg \n", + " inflating: ./data/segmented-images/masks/cju5hwonqedw10801vsd3w6kk.jpg \n", + " inflating: ./data/segmented-images/masks/cju2uy8ox62jo0801g88hh42z.jpg \n", + " inflating: ./data/segmented-images/masks/cju3tx1qyg0c907552fglumhc.jpg \n", + " inflating: ./data/segmented-images/masks/cju3v11mrgwwb0755u242ygye.jpg \n", + " inflating: ./data/segmented-images/masks/cju2suk42469908015ngmq6f2.jpg \n", + " inflating: ./data/segmented-images/masks/cju7dsrtb2f8i085064kwugfk.jpg \n", + " inflating: ./data/segmented-images/masks/cju8dm2cau2km0818jsv9eeq2.jpg \n", + " inflating: ./data/segmented-images/masks/cju784jpdzeae0987q5ypq883.jpg \n", + " inflating: ./data/segmented-images/masks/cju885ikhnmkn09878s2lqtuh.jpg \n", + " inflating: ./data/segmented-images/masks/cju45ty6zn9oz0850qy4qnck1.jpg \n", + " inflating: ./data/segmented-images/masks/ck2bxlujamu330725szlc2jdu.jpg \n", + " inflating: ./data/segmented-images/masks/cju5yclrymlgj0818k426ud6z.jpg \n", + " inflating: ./data/segmented-images/masks/cju5clr68b48r0755cmuvponm.jpg \n", + " inflating: ./data/segmented-images/masks/cju32srle1xfq083575i3fl75.jpg \n", + " inflating: ./data/segmented-images/masks/cju3ya7goj6at0818v2l5ay7f.jpg \n", + " inflating: ./data/segmented-images/masks/cju1d31sp4d4k0878r3fr02ul.jpg \n", + " inflating: ./data/segmented-images/masks/cju422cm8lfxn0818ojicxejb.jpg \n", + " inflating: ./data/segmented-images/masks/cju783tmkzkqu081803g7q5vk.jpg \n", + " inflating: ./data/segmented-images/masks/cju1h5w4wxajx0835mc954kxy.jpg \n", + " inflating: ./data/segmented-images/masks/cju30k2z40ds308353kdew70n.jpg \n", + " inflating: ./data/segmented-images/masks/cju7et17a2vjk0755e743npl1.jpg \n", + " inflating: ./data/segmented-images/masks/cju85omszllp30850b6rm9mi3.jpg \n", + " inflating: ./data/segmented-images/masks/cju32qr9tbvsj08013pkpjenq.jpg \n", + " inflating: ./data/segmented-images/masks/cju7dp3dw2k4n0755zhe003ad.jpg \n", + " inflating: ./data/segmented-images/masks/cju2ti1du4idn0878giuozonw.jpg \n", + " inflating: ./data/segmented-images/masks/ck2da7fwcjfis07218r1rvm95.jpg \n", + " inflating: ./data/segmented-images/masks/cju2rpa30t07b0835im0erql0.jpg \n", + " inflating: ./data/segmented-images/masks/cju5o1vu9gz8a0818eyy92bns.jpg \n", + " inflating: ./data/segmented-images/masks/cju2xbk0080y80801eghyddi2.jpg \n", + " inflating: ./data/segmented-images/masks/cju3xl264ingx0850rcf0rshj.jpg \n", + " inflating: ./data/segmented-images/masks/cju2z9vlp9j0w0801oag91sy9.jpg \n", + " inflating: ./data/segmented-images/masks/cju77bvg0yv4r0987yh60xmjo.jpg \n", + " inflating: ./data/segmented-images/masks/cju3xwpgviwlx0871rwm15q7v.jpg \n", + " inflating: ./data/segmented-images/masks/cju43o6n7m9nk087191ijwqq9.jpg \n", + " inflating: ./data/segmented-images/masks/cju6v1m1xv07w09870ah3njy1.jpg \n", + " inflating: ./data/segmented-images/masks/cju15ptjtppz40988odsm9azx.jpg \n", + " inflating: ./data/segmented-images/masks/cju6yxyt0wh080871sqpepu47.jpg \n", + " inflating: ./data/segmented-images/masks/cju45lbgznahl08180xz1h7u6.jpg \n", + " inflating: ./data/segmented-images/masks/cju5xopi0md7q0871r1sjc1av.jpg \n", + " inflating: ./data/segmented-images/masks/cju5wtdu4m0im0871mix0yvc0.jpg \n", + " inflating: ./data/segmented-images/masks/cju5f26ebcuai0818xlwh6116.jpg \n", + " inflating: ./data/segmented-images/masks/cju6yywx1whbb0871ksgfgf9f.jpg \n", + " inflating: ./data/segmented-images/masks/cju6vucxvvlda0755j7msqnya.jpg \n", + " inflating: ./data/segmented-images/masks/cju30gxjq0djk0988jytm49rs.jpg \n", + " inflating: ./data/segmented-images/masks/cju303j5r062k098835zxfds5.jpg \n", + " inflating: ./data/segmented-images/masks/cju35c4wzdhow0799h6eq4sgs.jpg \n", + " inflating: ./data/segmented-images/masks/cju5bbtwsa8cl0987wgfsqpao.jpg \n", + " inflating: ./data/segmented-images/masks/cju35atpxdjot0855q46aqrd0.jpg \n", + " inflating: ./data/segmented-images/masks/cju5buy2bal250818ipl6fqwv.jpg \n", + " inflating: ./data/segmented-images/masks/cju3uwz6ogsp10801h2r3bj5l.jpg \n", + " inflating: ./data/segmented-images/masks/cju5wrapcm2290818jsh26ppb.jpg \n", + " inflating: ./data/segmented-images/masks/cju45qbf3n9sa0987oonbkly9.jpg \n", + " inflating: ./data/segmented-images/masks/cju43c92lm5cj0755lorsorfg.jpg \n", + " inflating: ./data/segmented-images/masks/cju2zi4l09f5807991s8do2b2.jpg \n", + " inflating: ./data/segmented-images/masks/cju77vvcwzcm50850lzoykuva.jpg \n", + " inflating: ./data/segmented-images/masks/cju77u1sjz77b0817ft44r3fk.jpg \n", + " inflating: ./data/segmented-images/masks/cju88gx09o2vk0818610zody3.jpg \n", + " inflating: ./data/segmented-images/masks/cju2i03ptvkiu0799xbbd4det.jpg \n", + " inflating: ./data/segmented-images/masks/cju7alcgr1lsr0871riqk84z7.jpg \n", + " inflating: ./data/segmented-images/masks/cju7bb3ss1uoo0755pmhyco7t.jpg \n", + " inflating: ./data/segmented-images/masks/cju5wi6bqlxy90755bu227nvb.jpg \n", + " inflating: ./data/segmented-images/masks/cju7aklv31h4309871m29l4e7.jpg \n", + " inflating: ./data/segmented-images/masks/cju5bf6hxa6m50817rbwettgu.jpg \n", + " inflating: ./data/segmented-images/masks/cju1bhnfitmge0835ynls0l6b.jpg \n", + " inflating: ./data/segmented-images/masks/cju8djdqztu6408506pzhlo18.jpg \n", + " inflating: ./data/segmented-images/masks/cju18849rrsgr0988p90hkygb.jpg \n", + " inflating: ./data/segmented-images/masks/cju8clorgsuwn08714toqb7v6.jpg \n", + " inflating: ./data/segmented-images/masks/cju3ttznuftyf09875t11850w.jpg \n", + " inflating: ./data/segmented-images/masks/cju35mdz73x890835eynq1h9v.jpg \n", + " inflating: ./data/segmented-images/masks/cju5bmhdcafs909878qfzrqzi.jpg \n", + " inflating: ./data/segmented-images/masks/cju8bssulrrcy0987h1vq5060.jpg \n", + " inflating: ./data/segmented-images/masks/cju5huurrecm70801y680y13m.jpg \n", + " inflating: ./data/segmented-images/masks/cju87ox0kncom0801b98hqnd2.jpg \n", + " inflating: ./data/segmented-images/masks/cju30525w04r10835ygp257sb.jpg \n", + " inflating: ./data/segmented-images/masks/cju8chdlqsu620755azjty1tj.jpg \n", + " inflating: ./data/segmented-images/masks/cju30bmab08bi0835mvlr6e0r.jpg \n", + " inflating: ./data/segmented-images/masks/cju5uxjnol2r509871qv2yeia.jpg \n", + " inflating: ./data/segmented-images/masks/cju2saez63gxl08559ucjq3kt.jpg \n", + " inflating: ./data/segmented-images/masks/cju5w7xn0lrkq0801f9k0htgx.jpg \n", + " inflating: ./data/segmented-images/masks/cju7frtqu2xa20818wq8r9fzf.jpg \n", + " inflating: ./data/segmented-images/masks/cju83nwu1jxte0987h1krpfmv.jpg \n", + " inflating: ./data/segmented-images/masks/cju7fbndk2sl608015ravktum.jpg \n", + " inflating: ./data/segmented-images/masks/cju3x7xsaijq80818f0psavav.jpg \n", + " inflating: ./data/segmented-images/masks/cju6x97w4vwua0850x0997r0a.jpg \n", + " inflating: ./data/segmented-images/masks/cju8bpctzrqkr0850zeldv9kt.jpg \n", + " inflating: ./data/segmented-images/masks/cju2yv4imv6cz099314jveiib.jpg \n", + " inflating: ./data/segmented-images/masks/cju42m60jlpcm08186kqppzqv.jpg \n", + " inflating: ./data/segmented-images/masks/cju5jz5fff8c50871hbe6108f.jpg \n", + " inflating: ./data/segmented-images/masks/cju2y8s56ymqr083541ggdsml.jpg \n", + " inflating: ./data/segmented-images/masks/cju773hsyyosz0817pk1e7sjq.jpg \n", + " inflating: ./data/segmented-images/masks/cju35k2fr3vc50988c85qkrwg.jpg \n", + " inflating: ./data/segmented-images/masks/cju40sdwukv3k0755y99ug1k8.jpg \n", + " inflating: ./data/segmented-images/masks/cju2mfjndoz700988b9lc3zeq.jpg \n", + " inflating: ./data/segmented-images/masks/cju7deifq2fzn0755lc8idyh8.jpg \n", + " inflating: ./data/segmented-images/masks/cju7d7tly27h408016fyp5nr7.jpg \n", + " inflating: ./data/segmented-images/masks/cju87kbcen2av0987usezo8kn.jpg \n", + " inflating: ./data/segmented-images/masks/cju1aqqv02qwz0878a5cyhr67.jpg \n", + " inflating: ./data/segmented-images/masks/cju2pjb9v0ywn0878j5g5n69j.jpg \n", + " inflating: ./data/segmented-images/masks/cju1gkndf6yi10801o1qnje19.jpg \n", + " inflating: ./data/segmented-images/masks/cju7apr0c1qqm0755s7msqot4.jpg \n", + " inflating: ./data/segmented-images/masks/cju2qz06823a40878ojcz9ccx.jpg \n", + " inflating: ./data/segmented-images/masks/cju7bc95p1mdm0817yqj5jc6j.jpg \n", + " inflating: ./data/segmented-images/masks/cju1819curo000988pd5xcqme.jpg \n", + " inflating: ./data/segmented-images/masks/cju1fyb1d69et0878muzdak9u.jpg \n", + " inflating: ./data/segmented-images/masks/cju42romflni20817etb9a0fl.jpg \n", + " inflating: ./data/segmented-images/masks/cju13hp5rnbjx0835bf0jowgx.jpg \n", + " inflating: ./data/segmented-images/masks/cju849c23kgnk0817cgv2hw1e.jpg \n", + " inflating: ./data/segmented-images/masks/cju5x15djm7ae0755h8czf6nt.jpg \n", + " inflating: ./data/segmented-images/masks/cju0rx1idathl0835detmsp84.jpg \n", + " inflating: ./data/segmented-images/masks/cju8bqxxurs6i0850mu7mtef9.jpg \n", + " inflating: ./data/segmented-images/masks/cju2tqfgw4oat0799rn0g5b2z.jpg \n", + " inflating: ./data/segmented-images/masks/cju1dnz61vfp40988e78bkjga.jpg \n", + " inflating: ./data/segmented-images/masks/cju33za6l2qy70988jhrlp2ev.jpg \n", + " inflating: ./data/segmented-images/masks/cju42nm68lpyo0818xvvqmupq.jpg \n", + " inflating: ./data/segmented-images/masks/cju8bm24yrrdp081829mbo8ic.jpg \n", + " inflating: ./data/segmented-images/masks/cju6vqarjv7yo0987q4b1btk1.jpg \n", + " inflating: ./data/segmented-images/masks/cju7amjna1ly40871ugiokehb.jpg \n", + " inflating: ./data/segmented-images/masks/cju2pmhtr17a00855cvpelzb0.jpg \n", + " inflating: ./data/segmented-images/masks/cju2zblxw9848087853csbrx1.jpg \n", + " inflating: ./data/segmented-images/masks/cju7fcgbe2z3p07550vaflqdb.jpg \n", + " inflating: ./data/segmented-images/masks/cju7fob3x301u0755x985pmmq.jpg \n", + " inflating: ./data/segmented-images/masks/cju84ih17kp5l09876bkooocl.jpg \n", + " inflating: ./data/segmented-images/masks/cju1hs0za7jha0855vj0mdrjt.jpg \n", + " inflating: ./data/segmented-images/masks/cju5x7iskmad90818frchyfwd.jpg \n", + " inflating: ./data/segmented-images/masks/cju5ccpvqash50850kb4bs22k.jpg \n", + " inflating: ./data/segmented-images/masks/cju5fb86jd1jp0755b1ukbhq5.jpg \n", + " inflating: ./data/segmented-images/masks/cju846ec0kj7z08012o10klrb.jpg \n", + " inflating: ./data/segmented-images/masks/cju33w4sdcivk0855x879zht7.jpg \n", + " inflating: ./data/segmented-images/masks/cju87nkyrnb970801q84m47yt.jpg \n", + " inflating: ./data/segmented-images/masks/cju8cbsyssiqj0871gr4jedjp.jpg \n", + " inflating: ./data/segmented-images/masks/cju77k828z46w0871r0avuoo9.jpg \n", + " inflating: ./data/segmented-images/masks/cju16jgnyzp970878melv7r25.jpg \n", + " inflating: ./data/segmented-images/masks/cju2yyhsp933j0855hp32e012.jpg \n", + " inflating: ./data/segmented-images/masks/cju87z6o6nh73085045bzsx6o.jpg \n", + " inflating: ./data/segmented-images/masks/cju5cjh3xattc0817j2vbulzi.jpg \n", + " inflating: ./data/segmented-images/masks/cju43mkj9m8wb0871qiadahub.jpg \n", + " inflating: ./data/segmented-images/masks/cju8bysfgrzkl081786jwac09.jpg \n", + " inflating: ./data/segmented-images/masks/cju1ftaji6isw0855108yqcse.jpg \n", + " inflating: ./data/segmented-images/masks/cju8cdeazsm8h0801jxifmzur.jpg \n", + " inflating: ./data/segmented-images/masks/cju33belnbyhm0878yxl42233.jpg \n", + " inflating: ./data/segmented-images/masks/cju888fr7nveu0818r9uwtiit.jpg \n", + " inflating: ./data/segmented-images/masks/cju890guyoiti098753yg6cdu.jpg \n", + " inflating: ./data/segmented-images/masks/cju34sh43d8zm08019xbwhc0o.jpg \n", + " inflating: ./data/segmented-images/masks/cju3x2s11ibzi0817kk284k0j.jpg \n", + " inflating: ./data/segmented-images/masks/cju2wzu8wxtgu09880ku9x1pg.jpg \n", + " inflating: ./data/segmented-images/masks/cju77q10sz9ug0801449wu1nu.jpg \n", + " inflating: ./data/segmented-images/masks/cju3280wv1ir009882jze27tc.jpg \n", + " inflating: ./data/segmented-images/masks/cju45jpvfn6c809873pv1i34s.jpg \n", + " inflating: ./data/segmented-images/masks/cju30df2j09dd08351ayx2t6w.jpg \n", + " inflating: ./data/segmented-images/masks/cju17bz250pgd0799u1hqkj5u.jpg \n", + " inflating: ./data/segmented-images/masks/cju2t9tdwuk700835kv0ljmtl.jpg \n", + " inflating: ./data/segmented-images/masks/ck2bxskgxxzfv08386xkqtqdy.jpg \n", + " inflating: ./data/segmented-images/masks/cju7crgxa28550755wbsgqkel.jpg \n", + " inflating: ./data/segmented-images/masks/cju2xa3i4y0160988i679zsqd.jpg \n", + " inflating: ./data/segmented-images/masks/cju45n0oxn5vu08500yfrt9jn.jpg \n", + " inflating: ./data/segmented-images/masks/cju3yb47cj1xq0817zfotbni4.jpg \n", + " inflating: ./data/segmented-images/masks/cju88z8bson4h0871nnd7fdxo.jpg \n", + " inflating: ./data/segmented-images/masks/cju5chrxxawka0871qcj171yz.jpg \n", + " inflating: ./data/segmented-images/masks/cju1dq3x1vgx109889c7wyirg.jpg \n", + " inflating: ./data/segmented-images/masks/cju40r6jrksyk0871wg98zgho.jpg \n", + " inflating: ./data/segmented-images/masks/cju7bmi1v1pnj0987pa52jjok.jpg \n", + " inflating: ./data/segmented-images/masks/cju2zwg05a0oy0801yr73ig7g.jpg \n", + " inflating: ./data/segmented-images/masks/cju3umoh1geet0817cmpef5am.jpg \n", + " inflating: ./data/segmented-images/masks/cju1alwgo30z60855fm3y23sm.jpg \n", + " inflating: ./data/segmented-images/masks/cju2tpfa5uyx408359datxqqj.jpg \n", + " inflating: ./data/segmented-images/masks/cju83syhdk6gs0801rf1rekdl.jpg \n", + " inflating: ./data/segmented-images/masks/cju2htabevq9108015qjei0x7.jpg \n", + " inflating: ./data/segmented-images/masks/cju8bafgqrf4x0818twisk3ea.jpg \n", + " inflating: ./data/segmented-images/masks/cju7b9vcs1luz0987ta60j1dy.jpg \n", + " inflating: ./data/segmented-images/masks/cju2hlm19vjjf0801o69qnber.jpg \n", + " inflating: ./data/segmented-images/masks/cju8ca4geseia0850i2ru11hw.jpg \n", + " inflating: ./data/segmented-images/masks/cju5y7buemcw80987p0r30g9f.jpg \n", + " inflating: ./data/segmented-images/masks/cjyzjzssvd8pq0838f4nolj5l.jpg \n", + " inflating: ./data/segmented-images/masks/cju2oq5570avm079959o20op1.jpg \n", + " inflating: ./data/segmented-images/masks/cju3xhpvvimda0987ygrpzni2.jpg \n", + " inflating: ./data/segmented-images/masks/cju324q101fhe08350wae9cif.jpg \n", + " inflating: ./data/segmented-images/masks/cju8cwy02t9eq08185qn12c02.jpg \n", + " inflating: ./data/segmented-images/masks/cju1ewnoh5z030855vpex9uzt.jpg \n", + " inflating: ./data/segmented-images/masks/cju32csyfblyh080170aa3x5p.jpg \n", + " inflating: ./data/segmented-images/masks/cju5vutu7ll8w0871dfp92n9p.jpg \n", + " inflating: ./data/segmented-images/masks/cju18gzrq18zw0878wbf4ftw6.jpg \n", + " inflating: ./data/segmented-images/masks/cju7cufm7298k0755j09uf3of.jpg \n", + " inflating: ./data/segmented-images/masks/cju2hqt33lmra0988fr5ijv8j.jpg \n", + " inflating: ./data/segmented-images/masks/cju3u1c8tfyqx08503iedc3mx.jpg \n", + " inflating: ./data/segmented-images/masks/cju1f8w0t65en0799m9oacq0q.jpg \n", + " inflating: ./data/segmented-images/masks/cju34m7h536wq0988xz7gx79v.jpg \n", + " inflating: ./data/segmented-images/masks/cju6v6g6kvdw007552x6mb0po.jpg \n", + " inflating: ./data/segmented-images/masks/cju418rckl3ur08012psrx1r1.jpg \n", + " inflating: ./data/segmented-images/masks/cju5ktjwofed70817eg58ef7u.jpg \n", + " inflating: ./data/segmented-images/masks/cju1c6yfz42md08550zgoz3pw.jpg \n", + " inflating: ./data/segmented-images/masks/cju2qfie4rvz508357kad9z5o.jpg \n", + " inflating: ./data/segmented-images/masks/cju5yjq1pmlgc0801z0t24bly.jpg \n", + " inflating: ./data/segmented-images/masks/cju2xjz2ju8pe0993ysv9wg17.jpg \n", + " inflating: ./data/segmented-images/masks/cju5ukkg6kv7u08011x2b6zl5.jpg \n", + " inflating: ./data/segmented-images/masks/cju5xkwzxmf0z0818gk4xabdm.jpg \n", + " inflating: ./data/segmented-images/masks/cju8azmhcr66e0755t61atz72.jpg \n", + " inflating: ./data/segmented-images/masks/cju83u9ftk3ni0987qnhlcinv.jpg \n", + " inflating: ./data/segmented-images/masks/cju3y79ofj3va0871uqfb1mzo.jpg \n", + " inflating: ./data/segmented-images/masks/cju7d4jk723eu0817bqz2n39m.jpg \n", + " inflating: ./data/segmented-images/masks/cju2i6acqvo6l0799u20fift8.jpg \n", + " inflating: ./data/segmented-images/masks/cju7cq6su27qv075574dir0r3.jpg \n", + " inflating: ./data/segmented-images/masks/cju2nfnvxzdkd0878399axlco.jpg \n", + " inflating: ./data/segmented-images/masks/cju306x7w05nb0835cunv799x.jpg \n", + " inflating: ./data/segmented-images/masks/cju5x28nzm7t907558ocq4bt7.jpg \n", + " inflating: ./data/segmented-images/masks/cju7ahtkb1jr90801jck4kbds.jpg \n", + " inflating: ./data/segmented-images/masks/cju83k8fyjsxr0817d6nxs6r4.jpg \n", + " inflating: ./data/segmented-images/masks/cju17x0j4nfc10993y31pvlgs.jpg \n", + " inflating: ./data/segmented-images/masks/cju3x5u2tiihx0818914gzxy1.jpg \n", + " inflating: ./data/segmented-images/masks/cju5f0dezct4q08183ydw11dx.jpg \n", + " inflating: ./data/segmented-images/masks/cju7b3f5h1sm40755i572jden.jpg \n", + " inflating: ./data/segmented-images/masks/cju7ddtz729960801uazp1knc.jpg \n", + " inflating: ./data/segmented-images/masks/cju87xn2snfmv0987sc3d9xnq.jpg \n", + " inflating: ./data/segmented-images/masks/cju13cgqmnhwn0988yrainhcp.jpg \n", + " inflating: ./data/segmented-images/masks/cju2hdr06v2bq0799mbm3bks1.jpg \n", + " inflating: ./data/segmented-images/masks/cju15mhjczc8z0801kit5c6di.jpg \n", + " inflating: ./data/segmented-images/masks/cju88k75inzyb0850ccv5x3vk.jpg \n", + " inflating: ./data/segmented-images/masks/cju8c5223s8j80850b4kealt4.jpg \n", + " inflating: ./data/segmented-images/masks/cju5ft6mcd5q40987rhjgbrr6.jpg \n", + " inflating: ./data/segmented-images/masks/ck2bxknhjvs1x0794iogrq49k.jpg \n", + " inflating: ./data/segmented-images/masks/cju5tgbzhjllu08174ca41eus.jpg \n", + " inflating: ./data/segmented-images/masks/cju5vxuc5loxw0818u8xgf45p.jpg \n", + " inflating: ./data/segmented-images/masks/cju1ats0y372e08011yazcsxm.jpg \n", + " inflating: ./data/segmented-images/masks/cju2ycp1u8g2r0799jslnp7cz.jpg \n", + " inflating: ./data/segmented-images/masks/cju5cetivauok0987ok3e5bre.jpg \n", + " inflating: ./data/segmented-images/masks/cju886ryxnsl50801r93jai7q.jpg \n", + " inflating: ./data/segmented-images/masks/cju2qozsk20cq0855ugrg3cri.jpg \n", + " inflating: ./data/segmented-images/masks/cju2tvrvm53ws0801a0jfjdxg.jpg \n", + " inflating: ./data/segmented-images/masks/cju7f4sc62xqj075597xpmuoy.jpg \n", + " inflating: ./data/segmented-images/masks/cju6v4szov55u0871qmqz3v8n.jpg \n", + " inflating: ./data/segmented-images/masks/cju857ad0l88m0817qx4cwxnf.jpg \n", + " inflating: ./data/segmented-images/masks/cju5ygh1zmmdi0755uod5e17i.jpg \n", + " inflating: ./data/segmented-images/masks/cju8d4jgatgpj0871q2ophhkm.jpg \n", + " inflating: ./data/segmented-images/masks/cju3u4lxmg59o0755rz42b9en.jpg \n", + " inflating: ./data/segmented-images/masks/cju33231uy4gi0993qc7b1jch.jpg \n", + " inflating: ./data/segmented-images/masks/cju858eswlepn0871pzvdrhj1.jpg \n", + " inflating: ./data/segmented-images/masks/cju8c5mxls96t0850wvkvsity.jpg \n", + " inflating: ./data/segmented-images/masks/cju2r11x7sdgx0988o8ule0wl.jpg \n", + " inflating: ./data/segmented-images/masks/cju6wjm81vgsc0987enk9n3pr.jpg \n", + " inflating: ./data/segmented-images/masks/cju2np2k9zi3v079992ypxqkn.jpg \n", + " inflating: ./data/segmented-images/masks/cju3ua8u0g9rg0801uayhdxhu.jpg \n", + " inflating: ./data/segmented-images/masks/cju2p0eveqtdc0835gpi3p93i.jpg \n", + " inflating: ./data/segmented-images/masks/cju1f15k3w4ct0835cmde6ypo.jpg \n", + " inflating: ./data/segmented-images/masks/cju1f320ewfyu0988ndz6blh5.jpg \n", + " inflating: ./data/segmented-images/masks/ck2bxpfgxu2mk0748gsh7xelu.jpg \n", + " inflating: ./data/segmented-images/masks/cju45t5ddnbio0987qtqzx762.jpg \n", + " inflating: ./data/segmented-images/masks/cju2hos57llxm08359g92p6jj.jpg \n", + " inflating: ./data/segmented-images/masks/cju334jzo261t0835yqudnfs1.jpg \n", + " inflating: ./data/segmented-images/masks/cju32jcdabepz0878d0cznmfe.jpg \n", + " inflating: ./data/segmented-images/masks/cju31y80qbawn0801twwm2l5s.jpg \n", + " inflating: ./data/segmented-images/masks/cju8a56vxpy780850r45yu4wk.jpg \n", + " inflating: ./data/segmented-images/masks/cju8amfdtqi4x09871tygrgqe.jpg \n", + " inflating: ./data/segmented-images/masks/cju1f5x1164xv08555654c24r.jpg \n", + " inflating: ./data/segmented-images/masks/cju84gpefknwm098714oq8q61.jpg \n", + " inflating: ./data/segmented-images/masks/cju1csmlc4ht10799b8ymmghg.jpg \n", + " inflating: ./data/segmented-images/masks/cju85rkbnlo1c08503uxcpax1.jpg \n", + " inflating: ./data/segmented-images/masks/cju787jnjzjuj0871p94nck9g.jpg \n", + " inflating: ./data/segmented-images/masks/cju5wcc90lu020850mjrxppv6.jpg \n", + " inflating: ./data/segmented-images/masks/cju2zo0fwzv580988qlijd2xa.jpg \n", + " inflating: ./data/segmented-images/masks/cju5hyi9yegob0755ho3do8en.jpg \n", + " inflating: ./data/segmented-images/masks/cju3xjqtpikx50817tppy6g84.jpg \n", + " inflating: ./data/segmented-images/masks/cjyzu9th0qt4r0a46pyl4zik0.jpg \n", + " inflating: ./data/segmented-images/masks/cju1erep75us208553i4ofwwe.jpg \n", + " inflating: ./data/segmented-images/masks/cju2rkjfwoxys0993x768l1j2.jpg \n", + " inflating: ./data/segmented-images/masks/cju6ywm40wdbo0987pbftsvtg.jpg \n", + " inflating: ./data/segmented-images/masks/cju45v0pungu40871acnwtmu5.jpg \n", + " inflating: ./data/segmented-images/masks/cju8bh8surexp0987o5pzklk1.jpg \n", + " inflating: ./data/segmented-images/masks/cju1gv7106qd008784gk603mg.jpg \n", + " inflating: ./data/segmented-images/masks/cju5ekty5ckzf07550c9u3ckk.jpg \n", + " inflating: ./data/segmented-images/masks/cju0roawvklrq0799vmjorwfv.jpg \n", + " inflating: ./data/segmented-images/masks/cju7f9umg2olj0987fj5y285w.jpg \n", + " inflating: ./data/segmented-images/masks/cju3518w2d838079939fqztbc.jpg \n", + " inflating: ./data/segmented-images/masks/cju16fpvhzypl0799p9phnlx6.jpg \n", + " inflating: ./data/segmented-images/masks/cju5nxkujgscq0817l9gss626.jpg \n", + " inflating: ./data/segmented-images/masks/cju2yb31a8e8u0878wdashg7o.jpg \n", + " inflating: ./data/segmented-images/masks/cju5k503sfa5f0871lx0rpu5y.jpg \n", + " inflating: ./data/segmented-images/masks/cju5y4hgqmk0i08180rjhbwvp.jpg \n", + " inflating: ./data/segmented-images/masks/cju31w6goazci0799n014ly1q.jpg \n", + " inflating: ./data/segmented-images/masks/cju2tjrog4jy30878pawyazqc.jpg \n", + " inflating: ./data/segmented-images/masks/cju5udcufki0s09874ll1dbr5.jpg \n", + " inflating: ./data/segmented-images/masks/cju2osuru0ki00855txo0n3uu.jpg \n", + " inflating: ./data/segmented-images/masks/cju6x0yqbvxqt0755dhxislgb.jpg \n", + " inflating: ./data/segmented-images/masks/cju2trbpkv0c00988hxla5dzz.jpg \n", + " inflating: ./data/segmented-images/masks/cju8567gdlcbq0801dwwyo2jt.jpg \n", + " inflating: ./data/segmented-images/masks/cju7dtb1e2j0t0818deq51ib3.jpg \n", + " inflating: ./data/segmented-images/masks/cju5f8hxdcxxn08188obby0ea.jpg \n", + " inflating: ./data/segmented-images/masks/cju8b2rmgr52s0801p54eyflx.jpg \n", + " inflating: ./data/segmented-images/masks/cju2qh5le1ock0878oahaql7d.jpg \n", + " inflating: ./data/segmented-images/masks/cju8czvnztbf40871b4m7t78w.jpg \n", + " inflating: ./data/segmented-images/masks/cju7dymur2od30755eg8yv2ht.jpg \n", + " inflating: ./data/segmented-images/masks/cju2hw5gjlr5h0988so2qqres.jpg \n", + " inflating: ./data/segmented-images/masks/cju1c3218411b08014g9f6gig.jpg \n", + " inflating: ./data/segmented-images/masks/cju2uzabhs6er0993x3aaf87p.jpg \n", + " inflating: ./data/segmented-images/masks/cju8arof2qpf20850ifr1bnqj.jpg \n", + " inflating: ./data/segmented-images/masks/cju87r56lnkyp0755hz30leew.jpg \n", + " inflating: ./data/segmented-images/masks/cju5xjn5mm78b09871spyqhhr.jpg \n", + " inflating: ./data/segmented-images/masks/cju7f6cqy2ur20818t1saazbm.jpg \n", + " inflating: ./data/segmented-images/masks/cju2y40d8ulqo0993q0adtgtb.jpg \n", + " inflating: ./data/segmented-images/masks/cju7fpfzq2wyf0818xxd1oziv.jpg \n", + " inflating: ./data/segmented-images/masks/cju302fqq9spc0878rrygyzzz.jpg \n", + " inflating: ./data/segmented-images/masks/cju8828oxnool0801qno9luhr.jpg \n", + " inflating: ./data/segmented-images/masks/cju893jmdompz0817xn3g1w4h.jpg \n", + " inflating: ./data/segmented-images/masks/cju6z9a9kwsl007552s49rx6i.jpg \n", + " inflating: ./data/segmented-images/masks/cju15wdt3zla10801odjiw7sy.jpg \n", + " inflating: ./data/segmented-images/masks/cju14hjh2ob2o0835ouz3r5aa.jpg \n", + " inflating: ./data/segmented-images/masks/cju13fwthn9mq0835gacxgy01.jpg \n" + ] + }, + { + "name": "stdout", + "output_type": "stream", + "text": [ + " inflating: ./data/segmented-images/masks/cju85qefyln6v0850szeb9byi.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju6vgdmivcvb08018fra5lnv.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju8b8yair65w09878pyqtr96.jpg \r\n", + " inflating: ./data/segmented-images/masks/ck2395w2mb4vu07480otsu6tw.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju2nguelpmlj0835rojdn097.jpg \r\n", + " inflating: ./data/segmented-images/masks/cjyzkmjy8evns070165gf9dmq.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju0s2a9ekvms080138tjjpxr.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju2rlqdnoz9k0993cpjae3x0.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju7es23b2vcp0755gpbm9s7v.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju5tenjojp1j0755ms4949h2.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju7fq7mm2pw508176uk5ugtx.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju83mki1jv5w0817kubxm31r.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju5c5xc7algd0817pb1ej5yo.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju7d3oc82cho0755dajlwldz.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju83yddek68q0850d2x7zfkm.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju5wkonqlrl409877y8zvnub.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju2i8br1vqtd08784u6vmcjk.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju336l68y7if0993wf092166.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju7d1tvt25bu08019dvw3uff.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju7bgnvb1sf808717qa799ir.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju6x4t13vyw60755gtcf9ndu.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju7er4kc2opa0801anuxc0eb.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju77idwfz36d0871tzfzz51i.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju1cfhyg48bb0799cl5pr2jh.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju88aq6vo1ij0755c2ey7z7n.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju7dn24o296i09871qfxb8s2.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju7bfx651qr80801cs7epotb.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju45pm27n80u08174kyow1gj.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju7b1ygu1msd0801hywhy0mc.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju6z7e4bwgdd0987ogkzq9kt.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju8dpa89u6l80818dj6lldh9.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju5hjxaae3i40850h5z2laf5.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju1g20bdwq6u0835e16xugcd.jpg \r\n", + " inflating: ./data/segmented-images/masks/cjyzlw7f9faqr070129au64sq.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju41r6v2lcww0871ps8k8pf5.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju7db7lp2f400755tntd1ohf.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju43b8daly4408170e5ev06g.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju42dwedlmk60871jbgu4ehi.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju41nz76lcxu0755cya2qefx.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju8aj01yqeqm0850lhdz3xdw.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju6z1bzbwfq50817b2alatvr.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju3tvffffx5f0818t5ov22al.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju5ew4h9cqaf0818rrczkmqh.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju6uy20suzbl0987rzuhz7z9.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju43in5fm22c08175rxziqrk.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju5o4pk9h0720755lgp9jq8m.jpg \r\n", + " inflating: ./data/segmented-images/masks/ck2bxqz3evvg20794iiyv5v2m.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju8buos5rz9b08715lfr0f4f.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju2ma647l0nj0993ot4deq2q.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju35a77vdj4n08556jj2lgmc.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju8bk8oirjhw0817hgkua2w8.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju6wt9jvvn500871hjn3t3g0.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju1b3zgj3d8e0801kpolea6c.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju8d2q30tfhs0801n7lx77xl.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju35fxqyzt5p0993vusm54qz.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju7atnm31if40817pqclnjer.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju3v0fl3gwce0755qkjhzmd4.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju300m3s04fg0988uzupuf7z.jpg \r\n", + " inflating: ./data/segmented-images/masks/cjz14qsk2wci60794un9ozwmw.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju2u2b9o4zvp08788qb9nqxj.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju1fj6axwfp30835ukhuzhw9.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju5jx7jzf7c90871c2i9aiov.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju34ouumcznz07996gg1xq7v.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju2t62nq45jl0799odpufwx6.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju7arvfe1ldu0850erdmphgj.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju5hqz50e7o90850e0prlpa0.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju2lejzcy4pc0878c9rlonot.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju85hguellg50818kwu3s8d1.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju5bhv81abur0850ean02atv.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju428k5fldt108177s6g6f45.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju6wll7wvo3y08502pagos8m.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju35i2e63uxr0835h7zgkg9k.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju15jr8jz8sb0855ukmkswkz.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju1cnnziug1l0835yh4ropyg.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju34repocy5208780gswillm.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju2lberzkdzm09938cl40pog.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju2xlcqxy9c60988vjacdznb.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju2sxf3iqbpv09937iksn8ep.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju6vta3kvazg0817qbeppjtm.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju7evxt12m730987rxivne3x.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju2y0z6g87p10878fpk5d3rq.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju7fazv92ywx0755xov2erga.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju45rj7ln8980850a7821fov.jpg \r\n", + " inflating: ./data/segmented-images/masks/cjyzl833ndne80838pzuq6ila.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju2qn2fzs1vy0988l243cvzy.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju30qbm1ad3x0855znuhpz9u.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju2yg5ht8i4p087800js8hp4.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju84aoa3ktwn0755pfl4gfwd.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju7dlk532dsh0871zvr6qz0r.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju8402x1kcy70801t6kz6bdi.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju5enq1tcn1i0755hnkon787.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju33eqwbcch208012jikwdky.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju7ajnbo1gvm098749rdouk0.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju5i39mreass0817au8p22zy.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju1fr4etsmrr09933u4t4aql.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju5fs6j6d8350801vglraq4u.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju6wi3akvn8r0801px8eligc.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju2omjpeqj5a0988pjdlb8l1.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju42wamblrqn098798r2yyok.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju8c3xs7sauj0801ieyzezr5.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju2raxlosl630988jdbfy9b0.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju87li0zn3yb0817kbwgjiz8.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju2z1nxlzaj40835wj81s1iy.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju42g865lorv07552ytz6xxa.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju7f900s2o0k08175gl1giid.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju1ejj7dvqfa0835ra184v5m.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju76erapykj30871x5eaxh4q.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju2i3hzclw3o0988rrgh911i.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju892fesoq2g0801n0e0jyia.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju88evxanv9r08176zkeovec.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju43gfosm63n08714rpih8pe.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju88msmoo3470817m441j4sg.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju7b10ce1mnm08011c5bwyr4.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju1bm8063nmh07996rsjjemq.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju8dic9mtppa0987swn23wbc.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju2xf8e5y2wm08359vcgk09b.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju2yi9tz8vky0801yqip0xyl.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju7fmvpk2q170987v6i3ola8.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju2tzypl4wss0799ow05oxb9.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju7dmlgf2ebw0871ieqas5fh.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju85mpuglq8k0818d2it6hzb.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju7ctvqn25dy08186g442m1r.jpg \r\n", + " inflating: ./data/segmented-images/masks/cjyzurzvohqnr0794es1itzek.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju8b1v3br45u087189kku66u.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju1cj3f0qi5n0993ut8f49rj.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju7adqyj1jcx08712r1ro5gx.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju34c1xfyz920993itxkkfad.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju2top2ruxxy0988p1svx36g.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju5vi4nxlc530817uoqm2m7a.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju7d2q1k27nf08715zshsckt.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju1gghyjwxt80835vx0wgxw0.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju6uzxk0v83p0801rcwnexdu.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju2qxxko229x08786gvxxhur.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju2xyd9vyi7m098831qcucse.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju88itqbny720987hxizbj5y.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju7ev2b12owa08500bpfpwyw.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju8b4ja9r2s808509d45ma86.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju5boicjagt20871b1fotkh4.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju85fc11ljr40818edpb0inh.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju88l66no10s0850rsda7ej1.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju2nbdpmlmcj0993s1cht0dz.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju2nqapmzvk20801f9us40dx.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju6wuojavt740818b5qcv3iw.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju77g99iyxc00817zqi2ppor.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju3wstckialg0871xs0vevsj.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju85plp7lmkw0850rx42jdpf.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju7bf1lp1shi081835vs84lc.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju1ffnjn6ctm08015perkg37.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju7d8m3b2e210755l8fj1yph.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju7dxffn2eam0817qxosfwch.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju5h57xedz5h0755mjpc8694.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju5vgawslbe30987ndeepc1b.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju3v56bwgy8v0871w14pz8fx.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju6x35ervu2808015c7eoqe4.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju2xs6na81t20878pt6nkfip.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju85l4yjlops0801fvmnwptf.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju16ach3m1da0993r1dq3sn2.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju8bbznkrf5g0871jncffynk.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju33mirdc8mj0799k33wzoes.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju2u4pymvc720988wsxrmi84.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju6v3bb2v7xo085090blubyw.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju88q6h6obpd0871ckmiabbo.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju2zdhsczmn50988z64qwg2q.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju30xqmh0ni00835ix3batv1.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju34ds2531520988qjpqt6e3.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju5eyfe9cpk90987laa7tsl3.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju2s16zp317h0799gr67jqc2.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju8bljw9rqk20801kr54akrl.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju32gzs6xo8x0993r8tedbpb.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju8aw9n1qyg10801jkjlmors.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju5x00l6m5j608503k78ptee.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju5bycdkalkb09875f7bfrvx.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju7dubap2g0w0801fgl42mg9.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju6xifswvwbo0987nibtdr50.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju1expq45zst0855rjqwwj4m.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju2yo1j1v0qz09934o0e683p.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju3yht87j83m08507yk1u1fg.jpg \r\n", + " inflating: ./data/segmented-images/masks/cjyzu3reghjya0794w7pwoi50.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju5ht88gedbu0755xrcuddcx.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju7d9seq29zd0871nzl2uu5m.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju183od81ff608017ekzif89.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju0qkwl35piu0993l0dewei2.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju3xga12iixg0817dijbvjxw.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju8alhigqn2h0801zksudldd.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju7ez7r22qbc08015xfoz2wb.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju2qdj95ru8g09886gfi9rsz.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju40jl7skiuo0817p0smlgg8.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju7el6xv2k520817qxx9wdr5.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju7dhpsc2dnn0818025m6857.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju7dda8w2br20818zhsuz8s7.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju77196iyshb0850ycbto50a.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju40m0rjkpw80871z6n6yg1u.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju7csvlb22fr0850lvm45n3x.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju7dwe282dc309876rco45ts.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju7ecl9i2i060987xawjp4l0.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju88rl5eo94l0850kf5wtrm1.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju7cl8zm1xcu0817ado0jpas.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju8b6rp0r5st0850184f79xt.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju2okvco06xc0799kxe5n1qh.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju2iatlki5u309930zmgkv6h.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju7g7ba42z310987bqzbi2bq.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju0t4oil7vzk099370nun5h9.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju87tyddnnad0755bj0wxahe.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju326h4v1gxw08352px40p7r.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju8c1a0ws7o208181c6lbsom.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju5u8gz4kj5b07552e2wpkwp.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju17otoe119u0799nqcbl8n1.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju2uokeg5jm20799xwgsyz89.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju7dglf226g50987ohbthl19.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju1fb9236a110801yvg0fwju.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju2sggy13na70855tbeoqgha.jpg \r\n", + " inflating: ./data/segmented-images/masks/ck2bxw18mmz1k0725litqq2mc.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju2sevf53lkx08558h5bpaig.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju2qvuj1s9ok0835tp2k4ozh.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju30ftgja7170855xl9bkdm0.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju5eq8c8ck690850vix98hv3.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju2ulk385h170799rlklxob0.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju43eigtm6ev0801mv0m96t1.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju175facms5f0993a5tjikvt.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju1d96gsv62d09881b3wecw2.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju3xzvnzj0hd0755xprz39nj.jpg \r\n", + " inflating: ./data/segmented-images/masks/cjyztzaqtrv430848l8xgcerw.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju89y9h0puti0818i5yw29e6.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju2ysg748ru80878sp6j0gm0.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju2rz4k434s70855wwx3ddtx.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju5eftctcdbj08712gdp989f.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju8aqq8uqmoq0987hphto9gg.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju8bn7m2rmm70817hgxpb1uq.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju5ca9hcatkc0801jzwe7tfx.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju17g6ykn1cs0993dww6qdi8.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju15czxqp3lv0835jvhgzurz.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju323ypb1fbb0988gx5rzudb.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju2r6mt2om21099352pny5gw.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju88y1mwoln50871emyfny1g.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju17v6ih0u7808783zcbg1jy.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju1c0qb4tzi308355wtsnp0y.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju5i5oh2efg60987ez6cpf72.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju0u2g7pmnux0801vkk47ivj.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju2z6ez69g4u0801qwt088lw.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju7ehljc2or70871261br8ai.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju8432cmkgq90871cxe4iptl.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju2zgbj9zmrw0835nnlzxj4c.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju7fen322ou10817ziqkob4k.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju34xspwzenf0993cyzajv9n.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju3y54kwj3nr0801biidlb4e.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju32zhbnc1oy0801iyv1ix6p.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju0u82z3cuma0835wlxrnrjv.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju83qd0yjyht0817ktkfl268.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju412uwlkva50850d1ps1ww7.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju8doa16u5gh0818w1ywda3q.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju7dbppn28nx085097654msi.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju5u6wf0kh1t0755bg1ssixv.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju3uhb79gcgr0871orbrbi3x.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju1drnhbrb9409935wi7vkhg.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju6vifjlv55z0987un6y4zdo.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju87vqa0ndwg0850onjdz7ol.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju8418jhkf7d0818ga2v0xq0.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju2srvy5440s0801y1ba9akr.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju87q6yoneim0871dl4phvkd.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju7d5m0p23kn09871rk7pu3v.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju1cyjb5qtie0993njqne9m3.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju84kplnl1y30755ropua1b0.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju8c82iosagu0817l74s4m5g.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju77t0razbvm080106o56289.jpg \r\n", + " inflating: ./data/segmented-images/masks/cjyzufihqquiw0a46jatrbwln.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju0qx73cjw570799j4n5cjze.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju6xlygpw7bs0818n691jsq4.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju171py4qiha0835u8sl59ds.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju8cattbsivm0818p446wgel.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju3u815rg4ek0850vvhtcvcm.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju30nyxe0gfb0835p256yoju.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju34aozyyy830993bn16u32n.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju7d6ux323ze0987xos3srkx.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju7efffp2ivf0817etg3jehl.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju7dff9529h208503w60lbil.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju1hmff8tkp809931jps6fbr.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju3xeexgii1j0817zs68tb4g.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju34i3qvcyog0855qiejxx5w.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju77afzlz3kp07550x5nafzs.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju3y21quj0ir0818kgjagr15.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju2s9g11pnra0993gn4eh793.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju2ro5jqsy680988pi6qsujw.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju8c2rqzs5t80850d0zky5dy.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju5fydrud94708507vo6oy21.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju2wtwj87kys0855kx6mddzw.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju439oazm2fu0871ma0vvrft.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju5knbbqfipk080128cggukq.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju8a84g0q76m0818hwiggkod.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju84dsvaklpx098750hp83x4.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju2sszfq3uye0878sucelzk2.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju2y5zas8m7f0801d34g5owq.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju7emdni2py40871ivhxjtut.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju7flevb2wii08188otgs9p2.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju6wn57mvooj0850rp78hhy7.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju7ebe962hr409872ovibahw.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju30ia8da2bq0799klnehml2.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju847pxykriq0755268ktrk2.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju5yeqiwmkgl0801fzv2douc.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju320gyvbch60801v2amdi2g.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju45ofd9ne1j0801ri8dup7t.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju2nnqrqzp580855z8mhzgd6.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju2igw4gvxds0878808qj398.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju7druhp2gp308715i6km7be.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju310f6val1v0855xo8tc3gu.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju2ricdv2iys0878sv1adh0u.jpg \r\n", + " inflating: ./data/segmented-images/masks/cjyzk8qieoboa0848ogj51wwm.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju30ajhw09sx0988qyahx9s8.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju3tp94kfstl08181awh6z49.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju16whaj0e7n0855q7b6cjkm.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju3tsh4lfsok0987w6x3a0v1.jpg \r\n", + " inflating: ./data/segmented-images/masks/ck2bxiswtxuw80838qkisqjwz.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju8bw697rwg308177tg8huas.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju32a52lb9rc0799xi40qs00.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju2qqn5ys4uo0988ewrt2ip2.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju414lf2l1lt0801rl3hjllj.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju7avvi51iox0817ym55y6tt.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju430pm2lz0y0755jkhcc3d1.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju2zjcvj9qma0801dk71hhi0.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju7eotqi2qea0871y8yc7tqh.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju410dnfl0960755y8lu8d79.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju32phw2bv130801yj7bkouq.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju3ykamdj9u208503pygyuc8.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju3u39fog1bo0871lxjrabks.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju8ashhnquqr0801rwduzt7d.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju2r91dg2k090801bh0xzbxk.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju2xd75m82720801q4s4ik3n.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju2p91qir00k08350ddfif0w.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju40wto5kxwi0755it190f2k.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju3x4blzieu30850x10uuvbm.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju1dfeupuzlw0835gnxip369.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju1hyolc7aqu0878rrkfn1lr.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju2yljr0yzhw0988ecf271ly.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju8a1jtvpt9m081712iwkca7.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju7eea9b2m0z0801ynqv1fqu.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju2nd7l7z98o0799gfjvyfmw.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju2ueb6j5ado0878vf5md13o.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju30ywtc0oar0835bp2en7ec.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju2uwz9f5yf1085506cfamfx.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju5woy82m07m08505dmjg7g1.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju3uz4o6gr9z0850lhxyxvsj.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju0sr5ghl0nd08789uzf1raf.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju7f5ghb2r5s0801chwkxxh9.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju85je7vlht70817c9jcjwi4.jpg \r\n", + " inflating: ./data/segmented-images/masks/cjyzul1qggwwj07216mhiv5sy.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju84jdl9kv0i0871eog9b3i9.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju0qoxqj9q6s0835b43399p4.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju40poe4kt9s0755f9cnm3h5.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju8dn0c3u2v50801k8rvq02f.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju2x7vw87mu30878hye2ca0m.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju8dqkrqu83i0818ev74qpxq.jpg \r\n", + " inflating: ./data/segmented-images/masks/cju2oi8sq0i2y0801mektzvw8.jpg \r\n", + " creating: ./data/segmented-images/images/\r\n", + " inflating: ./data/segmented-images/images/cju2htabevq9108015qjei0x7.jpg \r\n", + " inflating: ./data/segmented-images/images/cju2r2obh2bjm08553kng0rh7.jpg \r\n", + " inflating: ./data/segmented-images/images/cju2zdvjn9h7r08553cp4eed5.jpg \r\n", + " inflating: ./data/segmented-images/images/cju6v4szov55u0871qmqz3v8n.jpg \r\n", + " inflating: ./data/segmented-images/images/cju16d65tzw9d0799ouslsw25.jpg \r\n", + " inflating: ./data/segmented-images/images/cju2txjfzv60w098839dcimys.jpg \r\n", + " inflating: ./data/segmented-images/images/cju8bff9nrfi10850fmfzbf8v.jpg \r\n", + " inflating: ./data/segmented-images/images/cju3xvoo2iqlc0817eku2r3wl.jpg \r\n", + " inflating: ./data/segmented-images/images/cju5buy2bal250818ipl6fqwv.jpg \r\n", + " inflating: ./data/segmented-images/images/cju7f0ec32txj08184asb8w5f.jpg \r\n", + " inflating: ./data/segmented-images/images/cju31y80qbawn0801twwm2l5s.jpg \r\n", + " inflating: ./data/segmented-images/images/cju3xiic0ilzp0850lusrb42j.jpg \r\n", + " inflating: ./data/segmented-images/images/cju6v6g6kvdw007552x6mb0po.jpg \r\n", + " inflating: ./data/segmented-images/images/cju5hi52odyf90817prvcwg45.jpg \r\n", + " inflating: ./data/segmented-images/images/cju87li0zn3yb0817kbwgjiz8.jpg \r\n", + " inflating: ./data/segmented-images/images/cju357rxxdaz30878y2esjpjt.jpg \r\n", + " inflating: ./data/segmented-images/images/cju35740hzm0g0993zl5ic246.jpg \r\n", + " inflating: ./data/segmented-images/images/cju7da88w2eod0755wejzynvt.jpg \r\n", + " inflating: ./data/segmented-images/images/cju6z2616wqbk07555bvnuyr1.jpg \r\n", + " inflating: ./data/segmented-images/images/cju35oyvd3y850988km12hdz1.jpg \r\n", + " inflating: ./data/segmented-images/images/cju892fesoq2g0801n0e0jyia.jpg \r\n", + " inflating: ./data/segmented-images/images/cju5x7iskmad90818frchyfwd.jpg \r\n", + " inflating: ./data/segmented-images/images/cju6w733bveoz0817e600tw72.jpg \r\n", + " inflating: ./data/segmented-images/images/cju41z76wlgbz0801qdetlvby.jpg \r\n", + " inflating: ./data/segmented-images/images/cju42py9mlqyd0818u3d1d7ga.jpg \r\n", + " inflating: ./data/segmented-images/images/cju8ceacrsqkr0755hdz145es.jpg \r\n", + " inflating: ./data/segmented-images/images/cju42xpi8lw4w0871ve317a1p.jpg \r\n", + " inflating: ./data/segmented-images/images/cju5wj0faly5008187n6530af.jpg \r\n", + " inflating: ./data/segmented-images/images/cju2saez63gxl08559ucjq3kt.jpg \r\n", + " inflating: ./data/segmented-images/images/cju88fpm4o0tl0871w1i6a4ds.jpg \r\n", + " inflating: ./data/segmented-images/images/cju2rnkt22xep0801as160g9t.jpg \r\n", + " inflating: ./data/segmented-images/images/cju87r56lnkyp0755hz30leew.jpg \r\n", + " inflating: ./data/segmented-images/images/cju1egh885m1l0855ci1lt37c.jpg \r\n", + " inflating: ./data/segmented-images/images/cju34eqjpcpm508788b3lhp97.jpg \r\n", + " inflating: ./data/segmented-images/images/cju2t9tdwuk700835kv0ljmtl.jpg \r\n", + " inflating: ./data/segmented-images/images/cju2qxxko229x08786gvxxhur.jpg \r\n", + " inflating: ./data/segmented-images/images/cju8apjewqrk00801k5d71gky.jpg \r\n", + " inflating: ./data/segmented-images/images/cju6ur9l9v9jq0755paud9uka.jpg \r\n", + " inflating: ./data/segmented-images/images/cju843yjskhq30818qre4rwm2.jpg \r\n", + " inflating: ./data/segmented-images/images/cju8bj2ssrmlm0871gc2ug2rs.jpg \r\n", + " inflating: ./data/segmented-images/images/cju2yyhsp933j0855hp32e012.jpg \r\n", + " inflating: ./data/segmented-images/images/cju2top2ruxxy0988p1svx36g.jpg \r\n", + " inflating: ./data/segmented-images/images/cju88trl3ogi208716qvti51b.jpg \r\n", + " inflating: ./data/segmented-images/images/cju5fw37edaae0801vkwvocn7.jpg \r\n", + " inflating: ./data/segmented-images/images/cju17x0j4nfc10993y31pvlgs.jpg \r\n", + " inflating: ./data/segmented-images/images/cju7bd1qu1mx409877xjxibox.jpg \r\n", + " inflating: ./data/segmented-images/images/cju7agj961l2r0818z29iq8yn.jpg \r\n", + " inflating: ./data/segmented-images/images/cju7cue9b232j0801qdzk1ykj.jpg \r\n", + " inflating: ./data/segmented-images/images/cju1cvkfwqrec0993wbp1jlzm.jpg \r\n", + " inflating: ./data/segmented-images/images/cju7b5afm1nfw0801xqm8bf8q.jpg \r\n", + " inflating: ./data/segmented-images/images/cju2qu37qobl50993aw7ghcfq.jpg \r\n", + " inflating: ./data/segmented-images/images/cju5wphwwlu3m0987hh3ltg88.jpg \r\n", + " inflating: ./data/segmented-images/images/cju2ti1du4idn0878giuozonw.jpg \r\n", + " inflating: ./data/segmented-images/images/cju83kxitjv340987z09m0ezy.jpg \r\n", + " inflating: ./data/segmented-images/images/cju2zp89k9q1g0855k1x0f1xa.jpg \r\n", + " inflating: ./data/segmented-images/images/cju6v5ilsv8hk0850rb5sgh6o.jpg \r\n", + " inflating: ./data/segmented-images/images/cju5xkwzxmf0z0818gk4xabdm.jpg \r\n", + " inflating: ./data/segmented-images/images/cju83u9ftk3ni0987qnhlcinv.jpg \r\n", + " inflating: ./data/segmented-images/images/cju8chxndsre008015uisl4si.jpg \r\n", + " inflating: ./data/segmented-images/images/cju1b75x63ddl0799sdp0i2j3.jpg \r\n", + " inflating: ./data/segmented-images/images/cju8arof2qpf20850ifr1bnqj.jpg \r\n", + " inflating: ./data/segmented-images/images/cju5fb86jd1jp0755b1ukbhq5.jpg \r\n", + " inflating: ./data/segmented-images/images/cju6z600qwh4z081700qimgl9.jpg \r\n", + " inflating: ./data/segmented-images/images/cju2yw4s7z7p20988lmf2gdgd.jpg \r\n", + " inflating: ./data/segmented-images/images/cju45qbf3n9sa0987oonbkly9.jpg \r\n", + " inflating: ./data/segmented-images/images/cju8c9akjsdjj0850s67uzlxq.jpg \r\n", + " inflating: ./data/segmented-images/images/cju45jpvfn6c809873pv1i34s.jpg \r\n", + " inflating: ./data/segmented-images/images/cju2lyynuymli0855g7fxgbhe.jpg \r\n", + " inflating: ./data/segmented-images/images/cju3v3ac9gyz30755hfqwyp1i.jpg \r\n", + " inflating: ./data/segmented-images/images/cju77k828z46w0871r0avuoo9.jpg \r\n", + " inflating: ./data/segmented-images/images/cju2z2x3nvd3c099350zgty7w.jpg \r\n", + " inflating: ./data/segmented-images/images/cju41kd7yl4nm0850gil5qqwh.jpg \r\n", + " inflating: ./data/segmented-images/images/cju7b2l561oas0871decgslaf.jpg \r\n", + " inflating: ./data/segmented-images/images/cju7cl8zm1xcu0817ado0jpas.jpg \r\n", + " inflating: ./data/segmented-images/images/cju45ty6zn9oz0850qy4qnck1.jpg \r\n", + " inflating: ./data/segmented-images/images/cju7ae7bq1f820987toc8si1d.jpg \r\n", + " inflating: ./data/segmented-images/images/cju76l27oyrw907551ri2a7fl.jpg \r\n", + " inflating: ./data/segmented-images/images/cju0vtox5ain6099360pu62rp.jpg \r\n", + " inflating: ./data/segmented-images/images/cju1hp9i2xu8e0988u2dazk7m.jpg \r\n", + " inflating: ./data/segmented-images/images/cju88k75inzyb0850ccv5x3vk.jpg \r\n", + " inflating: ./data/segmented-images/images/cju30bmab08bi0835mvlr6e0r.jpg \r\n", + " inflating: ./data/segmented-images/images/cju85bf1algsq0871y9gtlq97.jpg \r\n", + " inflating: ./data/segmented-images/images/cju2uokeg5jm20799xwgsyz89.jpg \r\n", + " inflating: ./data/segmented-images/images/cju1871y11d6r0799k6cw4yze.jpg \r\n", + " inflating: ./data/segmented-images/images/cju2omjpeqj5a0988pjdlb8l1.jpg \r\n", + " inflating: ./data/segmented-images/images/cju7crgxa28550755wbsgqkel.jpg \r\n", + " inflating: ./data/segmented-images/images/cju6xlygpw7bs0818n691jsq4.jpg \r\n", + " inflating: ./data/segmented-images/images/cju7787c5yy3l080159mwqsnj.jpg \r\n", + " inflating: ./data/segmented-images/images/cju1bhnfitmge0835ynls0l6b.jpg \r\n", + " inflating: ./data/segmented-images/images/cju5c7oijaqmq09878qwgqv8n.jpg \r\n", + " inflating: ./data/segmented-images/images/cju7ezs7g2mxm098787atbran.jpg \r\n", + " inflating: ./data/segmented-images/images/cju2qtee81yd708787bsjr75d.jpg \r\n", + " inflating: ./data/segmented-images/images/cju42g865lorv07552ytz6xxa.jpg \r\n" + ] + }, + { + "name": "stdout", + "output_type": "stream", + "text": [ + " inflating: ./data/segmented-images/images/cju6us80mv1b50871ebyq2wxa.jpg \r\n", + " inflating: ./data/segmented-images/images/cju87vqa0ndwg0850onjdz7ol.jpg \r\n", + " inflating: ./data/segmented-images/images/cju2sxf3iqbpv09937iksn8ep.jpg \r\n", + " inflating: ./data/segmented-images/images/cju5wtdu4m0im0871mix0yvc0.jpg \r\n", + " inflating: ./data/segmented-images/images/cju8ando2qqdo0818ck7i1be1.jpg \r\n", + " inflating: ./data/segmented-images/images/cju85plp7lmkw0850rx42jdpf.jpg \r\n", + " inflating: ./data/segmented-images/images/cju2u4pymvc720988wsxrmi84.jpg \r\n", + " inflating: ./data/segmented-images/images/cju35k2fr3vc50988c85qkrwg.jpg \r\n", + " inflating: ./data/segmented-images/images/cju2qz06823a40878ojcz9ccx.jpg \r\n", + " inflating: ./data/segmented-images/images/cju324q101fhe08350wae9cif.jpg \r\n", + " inflating: ./data/segmented-images/images/cju42u5bjlvi10801dc13sskp.jpg \r\n", + " inflating: ./data/segmented-images/images/cju7bb3ss1uoo0755pmhyco7t.jpg \r\n", + " inflating: ./data/segmented-images/images/cju2oo0wh0bqy0878biujeyhe.jpg \r\n", + " inflating: ./data/segmented-images/images/cju87z6o6nh73085045bzsx6o.jpg \r\n", + " inflating: ./data/segmented-images/images/cju1haab178i70799tk9z8y8x.jpg \r\n", + " inflating: ./data/segmented-images/images/cju7dda8w2br20818zhsuz8s7.jpg \r\n", + " inflating: ./data/segmented-images/images/cju2trbpkv0c00988hxla5dzz.jpg \r\n", + " inflating: ./data/segmented-images/images/cju1c6yfz42md08550zgoz3pw.jpg \r\n", + " inflating: ./data/segmented-images/images/cju7dsrtb2f8i085064kwugfk.jpg \r\n", + " inflating: ./data/segmented-images/images/cju2t16vuucaq0835xcpsivn2.jpg \r\n", + " inflating: ./data/segmented-images/images/cju3521y5d5mq0878t3ezsu4p.jpg \r\n", + " inflating: ./data/segmented-images/images/cju2hdr06v2bq0799mbm3bks1.jpg \r\n", + " inflating: ./data/segmented-images/images/cju41s6nbleqy0755e2mslg0b.jpg \r\n", + " inflating: ./data/segmented-images/images/cju13hp5rnbjx0835bf0jowgx.jpg \r\n", + " inflating: ./data/segmented-images/images/cju1fm3id6gl50801r3fok20c.jpg \r\n", + " inflating: ./data/segmented-images/images/cju2pmhtr17a00855cvpelzb0.jpg \r\n", + " inflating: ./data/segmented-images/images/cju2u2b9o4zvp08788qb9nqxj.jpg \r\n", + " inflating: ./data/segmented-images/images/cju8a56vxpy780850r45yu4wk.jpg \r\n", + " inflating: ./data/segmented-images/images/cju6vvb8svhed0801jjcquh5e.jpg \r\n", + " inflating: ./data/segmented-images/images/cju2sggy13na70855tbeoqgha.jpg \r\n", + " inflating: ./data/segmented-images/images/ck2bxknhjvs1x0794iogrq49k.jpg \r\n", + " inflating: ./data/segmented-images/images/cju785htizjzo08017tvlhtg4.jpg \r\n", + " inflating: ./data/segmented-images/images/cju5bdwa3aatx0818b79i18zf.jpg \r\n", + " inflating: ./data/segmented-images/images/cju5yjq1pmlgc0801z0t24bly.jpg \r\n", + " inflating: ./data/segmented-images/images/cju16b6ynq8e40988m8vx0xnj.jpg \r\n", + " inflating: ./data/segmented-images/images/cju41nz76lcxu0755cya2qefx.jpg \r\n", + " inflating: ./data/segmented-images/images/cju5xopi0md7q0871r1sjc1av.jpg \r\n", + " inflating: ./data/segmented-images/images/cju8bafgqrf4x0818twisk3ea.jpg \r\n", + " inflating: ./data/segmented-images/images/cju43b8daly4408170e5ev06g.jpg \r\n", + " inflating: ./data/segmented-images/images/cju0rx1idathl0835detmsp84.jpg \r\n", + " inflating: ./data/segmented-images/images/cju2qvuj1s9ok0835tp2k4ozh.jpg \r\n", + " inflating: ./data/segmented-images/images/cju33w4sdcivk0855x879zht7.jpg \r\n", + " inflating: ./data/segmented-images/images/cju2ycp1u8g2r0799jslnp7cz.jpg \r\n", + " inflating: ./data/segmented-images/images/cju2tjrog4jy30878pawyazqc.jpg \r\n", + " inflating: ./data/segmented-images/images/cju5jz5fff8c50871hbe6108f.jpg \r\n", + " inflating: ./data/segmented-images/images/cju6xa0qmvzun0818xjukgncj.jpg \r\n", + " inflating: ./data/segmented-images/images/cju8bysfgrzkl081786jwac09.jpg \r\n", + " inflating: ./data/segmented-images/images/cju5bhv81abur0850ean02atv.jpg \r\n", + " inflating: ./data/segmented-images/images/cju42nm68lpyo0818xvvqmupq.jpg \r\n", + " inflating: ./data/segmented-images/images/cju6wi3akvn8r0801px8eligc.jpg \r\n", + " inflating: ./data/segmented-images/images/cju1fuoa4wmc50835qfd11sp9.jpg \r\n", + " inflating: ./data/segmented-images/images/cju2tzypl4wss0799ow05oxb9.jpg \r\n", + " inflating: ./data/segmented-images/images/cju8djdqztu6408506pzhlo18.jpg \r\n", + " inflating: ./data/segmented-images/images/cju1fr4etsmrr09933u4t4aql.jpg \r\n", + " inflating: ./data/segmented-images/images/cju2pkwt3r8b90988v2ywq1px.jpg \r\n", + " inflating: ./data/segmented-images/images/cju336l68y7if0993wf092166.jpg \r\n", + " inflating: ./data/segmented-images/images/cju5k3j3uf6de0817hszzfr7n.jpg \r\n", + " inflating: ./data/segmented-images/images/cju5ft6mcd5q40987rhjgbrr6.jpg \r\n", + " inflating: ./data/segmented-images/images/cju7db7lp2f400755tntd1ohf.jpg \r\n", + " inflating: ./data/segmented-images/images/cju893jmdompz0817xn3g1w4h.jpg \r\n", + " inflating: ./data/segmented-images/images/cju1cqc7n4gpy0855jt246k68.jpg \r\n", + " inflating: ./data/segmented-images/images/cju5enq1tcn1i0755hnkon787.jpg \r\n", + " inflating: ./data/segmented-images/images/cju8cgi2kspp308011nxdtjp6.jpg \r\n", + " inflating: ./data/segmented-images/images/cju5hl8nee8a40755fm8qjj0o.jpg \r\n", + " inflating: ./data/segmented-images/images/cju8c5mxls96t0850wvkvsity.jpg \r\n", + " inflating: ./data/segmented-images/images/cju3u1c8tfyqx08503iedc3mx.jpg \r\n", + " inflating: ./data/segmented-images/images/cju89z6pqpqfx0817mfv8ixjc.jpg \r\n", + " inflating: ./data/segmented-images/images/cju6xmqd9w0250817l5kxfnsk.jpg \r\n", + " inflating: ./data/segmented-images/images/cju2m71z2ywwv080131bcrsd3.jpg \r\n", + " inflating: ./data/segmented-images/images/cju34sh43d8zm08019xbwhc0o.jpg \r\n", + " inflating: ./data/segmented-images/images/cju2nyc5f02m40801ojqbtiea.jpg \r\n", + " inflating: ./data/segmented-images/images/cju7fd6yt2p740987wkr8exo1.jpg \r\n", + " inflating: ./data/segmented-images/images/cju846ec0kj7z08012o10klrb.jpg \r\n", + " inflating: ./data/segmented-images/images/cju7ey10f2rvf0871bwbi9x82.jpg \r\n", + " inflating: ./data/segmented-images/images/cju2u73dj53oz0878486k8k4b.jpg \r\n", + " inflating: ./data/segmented-images/images/cju7apr0c1qqm0755s7msqot4.jpg \r\n", + " inflating: ./data/segmented-images/images/cju83h9ysjwe808716nt35oah.jpg \r\n", + " inflating: ./data/segmented-images/images/cju3xzvnzj0hd0755xprz39nj.jpg \r\n", + " inflating: ./data/segmented-images/images/cju7fbndk2sl608015ravktum.jpg \r\n", + " inflating: ./data/segmented-images/images/cju85mpuglq8k0818d2it6hzb.jpg \r\n", + " inflating: ./data/segmented-images/images/cju2zblxw9848087853csbrx1.jpg \r\n", + " inflating: ./data/segmented-images/images/cju6v1m1xv07w09870ah3njy1.jpg \r\n", + " inflating: ./data/segmented-images/images/cju8d4jgatgpj0871q2ophhkm.jpg \r\n", + " inflating: ./data/segmented-images/images/cju2zo0fwzv580988qlijd2xa.jpg \r\n", + " inflating: ./data/segmented-images/images/cju2hewssldzx0835ep795xu0.jpg \r\n", + " inflating: ./data/segmented-images/images/cju1b0y2e396p08558ois175d.jpg \r\n", + " inflating: ./data/segmented-images/images/cju1ftaji6isw0855108yqcse.jpg \r\n", + " inflating: ./data/segmented-images/images/cju7aifyo1p3n07552nxjx51f.jpg \r\n", + " inflating: ./data/segmented-images/images/cju418rckl3ur08012psrx1r1.jpg \r\n", + " inflating: ./data/segmented-images/images/cju6vqarjv7yo0987q4b1btk1.jpg \r\n", + " inflating: ./data/segmented-images/images/cju8adb60qbiu080188mxpf8d.jpg \r\n", + " inflating: ./data/segmented-images/images/cju8aj01yqeqm0850lhdz3xdw.jpg \r\n", + " inflating: ./data/segmented-images/images/cju5uxjnol2r509871qv2yeia.jpg \r\n", + " inflating: ./data/segmented-images/images/cju8aeei7q8k308173n9y4klv.jpg \r\n", + " inflating: ./data/segmented-images/images/cju2hlm19vjjf0801o69qnber.jpg \r\n", + " inflating: ./data/segmented-images/images/cju33o12x2jm50988944mxq0v.jpg \r\n", + " inflating: ./data/segmented-images/images/cju5vwbr4lhqn0987a1pji0ux.jpg \r\n", + " inflating: ./data/segmented-images/images/cju88vx2uoocy075531lc63n3.jpg \r\n", + " inflating: ./data/segmented-images/images/cju7et17a2vjk0755e743npl1.jpg \r\n", + " inflating: ./data/segmented-images/images/cju43mkj9m8wb0871qiadahub.jpg \r\n", + " inflating: ./data/segmented-images/images/cju5o1vu9gz8a0818eyy92bns.jpg \r\n", + " inflating: ./data/segmented-images/images/cju77re6fz5bb0817vp9redjg.jpg \r\n", + " inflating: ./data/segmented-images/images/cju160wshltz10993i1gmqxbe.jpg \r\n", + " inflating: ./data/segmented-images/images/cju34ymm8d6700799uop0cw33.jpg \r\n", + " inflating: ./data/segmented-images/images/cju2rkjfwoxys0993x768l1j2.jpg \r\n", + " inflating: ./data/segmented-images/images/cju783tmkzkqu081803g7q5vk.jpg \r\n", + " inflating: ./data/segmented-images/images/cju7amjna1ly40871ugiokehb.jpg \r\n", + " inflating: ./data/segmented-images/images/cju0sxqiclckk08551ycbwhno.jpg \r\n", + " inflating: ./data/segmented-images/images/cju7f4sc62xqj075597xpmuoy.jpg \r\n", + " inflating: ./data/segmented-images/images/cju45rj7ln8980850a7821fov.jpg \r\n", + " inflating: ./data/segmented-images/images/cju1fjsb4sipq09931lvd8e41.jpg \r\n", + " inflating: ./data/segmented-images/images/cju7fcgbe2z3p07550vaflqdb.jpg \r\n", + " inflating: ./data/segmented-images/images/cju2qdj95ru8g09886gfi9rsz.jpg \r\n", + " inflating: ./data/segmented-images/images/cju8buos5rz9b08715lfr0f4f.jpg \r\n", + " inflating: ./data/segmented-images/images/cju2zr3c3vwb00993jn06bbaz.jpg \r\n", + " inflating: ./data/segmented-images/images/cju1euuc65wm00799m4sjdnnn.jpg \r\n", + " inflating: ./data/segmented-images/images/cju8cj10qsrau0871o2dr6ai1.jpg \r\n", + " inflating: ./data/segmented-images/images/cju5klveuff6w0871wbibgh3m.jpg \r\n", + " inflating: ./data/segmented-images/images/cju5x00l6m5j608503k78ptee.jpg \r\n", + " inflating: ./data/segmented-images/images/cju42qet0lsq90871e50xbnuv.jpg \r\n", + " inflating: ./data/segmented-images/images/cju330ofbc2l30801th5g3hw6.jpg \r\n", + " inflating: ./data/segmented-images/images/cju320gyvbch60801v2amdi2g.jpg \r\n", + " inflating: ./data/segmented-images/images/cju5bf6hxa6m50817rbwettgu.jpg \r\n", + " inflating: ./data/segmented-images/images/cju5thdbrjp1108715xdfx356.jpg \r\n", + " inflating: ./data/segmented-images/images/cju1hhj6mxfp90835n3wofrap.jpg \r\n", + " inflating: ./data/segmented-images/images/ck2bxskgxxzfv08386xkqtqdy.jpg \r\n", + " inflating: ./data/segmented-images/images/cju5hwonqedw10801vsd3w6kk.jpg \r\n", + " inflating: ./data/segmented-images/images/cju888fr7nveu0818r9uwtiit.jpg \r\n", + " inflating: ./data/segmented-images/images/cju1c0qb4tzi308355wtsnp0y.jpg \r\n", + " inflating: ./data/segmented-images/images/cju6wll7wvo3y08502pagos8m.jpg \r\n", + " inflating: ./data/segmented-images/images/cju8b2rmgr52s0801p54eyflx.jpg \r\n", + " inflating: ./data/segmented-images/images/cju6xifswvwbo0987nibtdr50.jpg \r\n", + " inflating: ./data/segmented-images/images/cju45v0pungu40871acnwtmu5.jpg \r\n", + " inflating: ./data/segmented-images/images/cju7d2q1k27nf08715zshsckt.jpg \r\n", + " inflating: ./data/segmented-images/images/cju8azmhcr66e0755t61atz72.jpg \r\n", + " inflating: ./data/segmented-images/images/cju8dic9mtppa0987swn23wbc.jpg \r\n", + " inflating: ./data/segmented-images/images/cju424hy5lckr085073fva1ok.jpg \r\n", + " inflating: ./data/segmented-images/images/cju8bi8q7rlmn0871abc5ch8k.jpg \r\n", + " inflating: ./data/segmented-images/images/cju1d50a94qf50855wsowacrc.jpg \r\n", + " inflating: ./data/segmented-images/images/cju1amqw6p8pw0993d9gc5crl.jpg \r\n", + " inflating: ./data/segmented-images/images/cju5huurrecm70801y680y13m.jpg \r\n", + " inflating: ./data/segmented-images/images/cju1brhsj3rls0855a1vgdlen.jpg \r\n", + " inflating: ./data/segmented-images/images/cju85dx63lic408017f0l0400.jpg \r\n", + " inflating: ./data/segmented-images/images/cju88t4fvokxf07558ymyh281.jpg \r\n", + " inflating: ./data/segmented-images/images/cju31w6goazci0799n014ly1q.jpg \r\n", + " inflating: ./data/segmented-images/images/cju8bbznkrf5g0871jncffynk.jpg \r\n", + " inflating: ./data/segmented-images/images/cju85citjlnfm0755i4rk5tqj.jpg \r\n", + " inflating: ./data/segmented-images/images/cju3x5u2tiihx0818914gzxy1.jpg \r\n", + " inflating: ./data/segmented-images/images/cju1djtprvd7b0988thwwrg09.jpg \r\n", + " inflating: ./data/segmented-images/images/cju7fmvpk2q170987v6i3ola8.jpg \r\n", + " inflating: ./data/segmented-images/images/cju7fq7mm2pw508176uk5ugtx.jpg \r\n", + " inflating: ./data/segmented-images/images/cju774fmayxif0818u2g79usw.jpg \r\n", + " inflating: ./data/segmented-images/images/cju2zrojo9kcd0878ld2epejq.jpg \r\n", + " inflating: ./data/segmented-images/images/cju6z1bzbwfq50817b2alatvr.jpg \r\n", + " inflating: ./data/segmented-images/images/cju6ut4l8va6y0755tyw3vfqq.jpg \r\n", + " inflating: ./data/segmented-images/images/cju5yimthmlv80850zhoc90c2.jpg \r\n", + " inflating: ./data/segmented-images/images/cju5g163vd6mt0817uccuga6u.jpg \r\n", + " inflating: ./data/segmented-images/images/cju84ffdzkrjn08183jh1fxmb.jpg \r\n", + " inflating: ./data/segmented-images/images/cju5y7buemcw80987p0r30g9f.jpg \r\n", + " inflating: ./data/segmented-images/images/cju5bmhdcafs909878qfzrqzi.jpg \r\n", + " inflating: ./data/segmented-images/images/cju8a2itsq4dv0755ntlovpxe.jpg \r\n", + " inflating: ./data/segmented-images/images/cju1gghyjwxt80835vx0wgxw0.jpg \r\n", + " inflating: ./data/segmented-images/images/cju30ov1oah920801mi8thuyg.jpg \r\n", + " inflating: ./data/segmented-images/images/cju5v8pgplg6k0755rvi2t63h.jpg \r\n", + " inflating: ./data/segmented-images/images/cju85nr8elly209872w9n5m0s.jpg \r\n", + " inflating: ./data/segmented-images/images/cju2spdagu1l50835da1f46fr.jpg \r\n", + " inflating: ./data/segmented-images/images/cju7dff9529h208503w60lbil.jpg \r\n", + " inflating: ./data/segmented-images/images/cju7bf1lp1shi081835vs84lc.jpg \r\n", + " inflating: ./data/segmented-images/images/cju32upim1z7u0988l883nqp6.jpg \r\n", + " inflating: ./data/segmented-images/images/cju7evxt12m730987rxivne3x.jpg \r\n", + " inflating: ./data/segmented-images/images/cju2mfjndoz700988b9lc3zeq.jpg \r\n", + " inflating: ./data/segmented-images/images/cju88v2f9oi8w0871hx9auh01.jpg \r\n", + " inflating: ./data/segmented-images/images/cju35a77vdj4n08556jj2lgmc.jpg \r\n", + " inflating: ./data/segmented-images/images/cju1cdxvz48hw0801i0fjwcnk.jpg \r\n", + " inflating: ./data/segmented-images/images/cju8bpctzrqkr0850zeldv9kt.jpg \r\n", + " inflating: ./data/segmented-images/images/cju2nbdpmlmcj0993s1cht0dz.jpg \r\n", + " inflating: ./data/segmented-images/images/cju7ahtkb1jr90801jck4kbds.jpg \r\n", + " inflating: ./data/segmented-images/images/cju8b542nr81x0871uxnkm9ih.jpg \r\n", + " inflating: ./data/segmented-images/images/cju7fpfzq2wyf0818xxd1oziv.jpg \r\n", + " inflating: ./data/segmented-images/images/cju5ufn3skquf0818dhapnhba.jpg \r\n", + " inflating: ./data/segmented-images/images/cju45t5ddnbio0987qtqzx762.jpg \r\n", + " inflating: ./data/segmented-images/images/cju42dwedlmk60871jbgu4ehi.jpg \r\n", + " inflating: ./data/segmented-images/images/cju5f0dezct4q08183ydw11dx.jpg \r\n", + " inflating: ./data/segmented-images/images/cju3xl264ingx0850rcf0rshj.jpg \r\n", + " inflating: ./data/segmented-images/images/cju3tvffffx5f0818t5ov22al.jpg \r\n", + " inflating: ./data/segmented-images/images/cju2ij9uiic2l09933ljiv6gm.jpg \r\n", + " inflating: ./data/segmented-images/images/cju0roawvklrq0799vmjorwfv.jpg \r\n", + " inflating: ./data/segmented-images/images/cju33qpdvc9g0087825jhf3s9.jpg \r\n", + " inflating: ./data/segmented-images/images/cju849c23kgnk0817cgv2hw1e.jpg \r\n", + " inflating: ./data/segmented-images/images/cju2rxm8rpbaf0993o3qr2oph.jpg \r\n", + " inflating: ./data/segmented-images/images/cju40u30gkuzc0871rq7t666d.jpg \r\n", + " inflating: ./data/segmented-images/images/cju3v11mrgwwb0755u242ygye.jpg \r\n", + " inflating: ./data/segmented-images/images/cju7d5m0p23kn09871rk7pu3v.jpg \r\n", + " inflating: ./data/segmented-images/images/cju8bssulrrcy0987h1vq5060.jpg \r\n", + " inflating: ./data/segmented-images/images/cju5y84q3mdv50817eyp82xf3.jpg \r\n", + " inflating: ./data/segmented-images/images/cju43in5fm22c08175rxziqrk.jpg \r\n", + " inflating: ./data/segmented-images/images/cju5wrrs0m2af0818vmnajbtw.jpg \r\n", + " inflating: ./data/segmented-images/images/cju5wi6bqlxy90755bu227nvb.jpg \r\n", + " inflating: ./data/segmented-images/images/cju1cu1u2474n0878tt7v4tdr.jpg \r\n", + " inflating: ./data/segmented-images/images/cju8bn7m2rmm70817hgxpb1uq.jpg \r\n", + " inflating: ./data/segmented-images/images/cju1dhfok4mhe0878jlgrag0h.jpg \r\n", + " inflating: ./data/segmented-images/images/cju45n0oxn5vu08500yfrt9jn.jpg \r\n", + " inflating: ./data/segmented-images/images/cju3y0pjrj1c30755nxekxccj.jpg \r\n", + " inflating: ./data/segmented-images/images/cju858eswlepn0871pzvdrhj1.jpg \r\n", + " inflating: ./data/segmented-images/images/cju88evxanv9r08176zkeovec.jpg \r\n", + " inflating: ./data/segmented-images/images/cju5f8hxdcxxn08188obby0ea.jpg \r\n", + " inflating: ./data/segmented-images/images/cju5bbtwsa8cl0987wgfsqpao.jpg \r\n", + " inflating: ./data/segmented-images/images/cju35i2e63uxr0835h7zgkg9k.jpg \r\n", + " inflating: ./data/segmented-images/images/cju2r7h21sj9608354gzks3ae.jpg \r\n", + " inflating: ./data/segmented-images/images/cju2z9vlp9j0w0801oag91sy9.jpg \r\n", + " inflating: ./data/segmented-images/images/cju2okvco06xc0799kxe5n1qh.jpg \r\n", + " inflating: ./data/segmented-images/images/cju7alcgr1lsr0871riqk84z7.jpg \r\n", + " inflating: ./data/segmented-images/images/cju7azuu31mia0801pf9ib5ed.jpg \r\n", + " inflating: ./data/segmented-images/images/cju2nfnvxzdkd0878399axlco.jpg \r\n", + " inflating: ./data/segmented-images/images/cju34fojcctcf0799ebolbvkn.jpg \r\n", + " inflating: ./data/segmented-images/images/cju8bh8surexp0987o5pzklk1.jpg \r\n", + " inflating: ./data/segmented-images/images/cju8418jhkf7d0818ga2v0xq0.jpg \r\n", + " inflating: ./data/segmented-images/images/cju43o6n7m9nk087191ijwqq9.jpg \r\n", + " inflating: ./data/segmented-images/images/cju2trtjf4qjd0878a2zle9v9.jpg \r\n", + " inflating: ./data/segmented-images/images/cju5vbo6jldrt0871jf6f1700.jpg \r\n", + " inflating: ./data/segmented-images/images/cju3xga12iixg0817dijbvjxw.jpg \r\n", + " inflating: ./data/segmented-images/images/cju7dymur2od30755eg8yv2ht.jpg \r\n", + " inflating: ./data/segmented-images/images/cju5ccpvqash50850kb4bs22k.jpg \r\n", + " inflating: ./data/segmented-images/images/cju8c5zcbsdfz0801o5t6jag1.jpg \r\n", + " inflating: ./data/segmented-images/images/cju2y5zas8m7f0801d34g5owq.jpg \r\n", + " inflating: ./data/segmented-images/images/cju2nnqrqzp580855z8mhzgd6.jpg \r\n", + " inflating: ./data/segmented-images/images/cju3u815rg4ek0850vvhtcvcm.jpg \r\n", + " inflating: ./data/segmented-images/images/cju3xhpvvimda0987ygrpzni2.jpg \r\n", + " inflating: ./data/segmented-images/images/cju3y21quj0ir0818kgjagr15.jpg \r\n", + " inflating: ./data/segmented-images/images/cju30u1hbakn808019g15nb8b.jpg \r\n", + " inflating: ./data/segmented-images/images/cju8c6hnxsdvr0801wn0vrsa6.jpg \r\n", + " inflating: ./data/segmented-images/images/cju5fu081d8gc0818l3yylujk.jpg \r\n", + " inflating: ./data/segmented-images/images/cju77g99iyxc00817zqi2ppor.jpg \r\n", + " inflating: ./data/segmented-images/images/cju30lncba3ny0878jwnous8n.jpg \r\n", + " inflating: ./data/segmented-images/images/cju43h43am1dy08176gwfhmnt.jpg \r\n", + " inflating: ./data/segmented-images/images/cju8cattbsivm0818p446wgel.jpg \r\n", + " inflating: ./data/segmented-images/images/cju1csmlc4ht10799b8ymmghg.jpg \r\n", + " inflating: ./data/segmented-images/images/cju35ldepdtlm0801yv79y8vu.jpg \r\n", + " inflating: ./data/segmented-images/images/cju1euant5l960878iqj5vvto.jpg \r\n", + " inflating: ./data/segmented-images/images/cju87q6yoneim0871dl4phvkd.jpg \r\n", + " inflating: ./data/segmented-images/images/cju5vzjoslpj708186z2fusmz.jpg \r\n", + " inflating: ./data/segmented-images/images/cju5wmvsdlx1j0871npgj8j4b.jpg \r\n", + " inflating: ./data/segmented-images/images/cju83mki1jv5w0817kubxm31r.jpg \r\n", + " inflating: ./data/segmented-images/images/cju40taxlkrho0987smigg0x0.jpg \r\n", + " inflating: ./data/segmented-images/images/cju32pzh9bpw10855q4vaxfhe.jpg \r\n", + " inflating: ./data/segmented-images/images/cju3x9lttikfb0818a0g104zn.jpg \r\n", + " inflating: ./data/segmented-images/images/cju773hsyyosz0817pk1e7sjq.jpg \r\n", + " inflating: ./data/segmented-images/images/cju2zwg05a0oy0801yr73ig7g.jpg \r\n", + " inflating: ./data/segmented-images/images/cju2tvrvm53ws0801a0jfjdxg.jpg \r\n", + " inflating: ./data/segmented-images/images/cju3uwz6ogsp10801h2r3bj5l.jpg \r\n", + " inflating: ./data/segmented-images/images/cju2y0z6g87p10878fpk5d3rq.jpg \r\n", + " inflating: ./data/segmented-images/images/cju32jcdabepz0878d0cznmfe.jpg \r\n", + " inflating: ./data/segmented-images/images/cju15mhjczc8z0801kit5c6di.jpg \r\n", + " inflating: ./data/segmented-images/images/cju8432cmkgq90871cxe4iptl.jpg \r\n", + " inflating: ./data/segmented-images/images/cju5w7xn0lrkq0801f9k0htgx.jpg \r\n", + " inflating: ./data/segmented-images/images/cju7d7aut2a2p0818z4uxc6cd.jpg \r\n", + " inflating: ./data/segmented-images/images/cju2i8br1vqtd08784u6vmcjk.jpg \r\n", + " inflating: ./data/segmented-images/images/cju2qh5le1ock0878oahaql7d.jpg \r\n", + " inflating: ./data/segmented-images/images/ck2bxpfgxu2mk0748gsh7xelu.jpg \r\n", + " inflating: ./data/segmented-images/images/cju87kbcen2av0987usezo8kn.jpg \r\n", + " inflating: ./data/segmented-images/images/cju7bmi1v1pnj0987pa52jjok.jpg \r\n", + " inflating: ./data/segmented-images/images/cju5o4pk9h0720755lgp9jq8m.jpg \r\n", + " inflating: ./data/segmented-images/images/cju7f6cqy2ur20818t1saazbm.jpg \r\n", + " inflating: ./data/segmented-images/images/cju1cyjb5qtie0993njqne9m3.jpg \r\n", + " inflating: ./data/segmented-images/images/ck2bxiswtxuw80838qkisqjwz.jpg \r\n", + " inflating: ./data/segmented-images/images/cju7frtqu2xa20818wq8r9fzf.jpg \r\n", + " inflating: ./data/segmented-images/images/cju31ugmfb3dz0855xtqshki6.jpg \r\n", + " inflating: ./data/segmented-images/images/cju2xjz2ju8pe0993ysv9wg17.jpg \r\n", + " inflating: ./data/segmented-images/images/cju784jpdzeae0987q5ypq883.jpg \r\n" + ] + }, + { + "name": "stdout", + "output_type": "stream", + "text": [ + " inflating: ./data/segmented-images/images/cju8b8yair65w09878pyqtr96.jpg \r\n", + " inflating: ./data/segmented-images/images/cju7cp6dw244p0818gncdol4m.jpg \r\n", + " inflating: ./data/segmented-images/images/cju88aq6vo1ij0755c2ey7z7n.jpg \r\n", + " inflating: ./data/segmented-images/images/cju3v664kh0px0818y4y7wolf.jpg \r\n", + " inflating: ./data/segmented-images/images/cju5x15djm7ae0755h8czf6nt.jpg \r\n", + " inflating: ./data/segmented-images/images/cju88l66no10s0850rsda7ej1.jpg \r\n", + " inflating: ./data/segmented-images/images/cju6ywm40wdbo0987pbftsvtg.jpg \r\n", + " inflating: ./data/segmented-images/images/cju17r8il13910799dr2wme2e.jpg \r\n", + " inflating: ./data/segmented-images/images/cju7cufm7298k0755j09uf3of.jpg \r\n", + " inflating: ./data/segmented-images/images/cju2xf8e5y2wm08359vcgk09b.jpg \r\n", + " inflating: ./data/segmented-images/images/cju7g7ba42z310987bqzbi2bq.jpg \r\n", + " inflating: ./data/segmented-images/images/cju3ul8dogf1z09872y2ecowp.jpg \r\n", + " inflating: ./data/segmented-images/images/cju33eqwbcch208012jikwdky.jpg \r\n", + " inflating: ./data/segmented-images/images/cju2hfqnmhisa0993gpleeldd.jpg \r\n", + " inflating: ./data/segmented-images/images/cju34c1xfyz920993itxkkfad.jpg \r\n", + " inflating: ./data/segmented-images/images/cju43gfosm63n08714rpih8pe.jpg \r\n", + " inflating: ./data/segmented-images/images/cju7dhpsc2dnn0818025m6857.jpg \r\n", + " inflating: ./data/segmented-images/images/cju34ds2531520988qjpqt6e3.jpg \r\n", + " inflating: ./data/segmented-images/images/cju88gx09o2vk0818610zody3.jpg \r\n", + " inflating: ./data/segmented-images/images/cju6uy20suzbl0987rzuhz7z9.jpg \r\n", + " inflating: ./data/segmented-images/images/cju1g20bdwq6u0835e16xugcd.jpg \r\n", + " inflating: ./data/segmented-images/images/cju6x35ervu2808015c7eoqe4.jpg \r\n", + " inflating: ./data/segmented-images/images/cju5eftctcdbj08712gdp989f.jpg \r\n", + " inflating: ./data/segmented-images/images/cju30qbm1ad3x0855znuhpz9u.jpg \r\n", + " inflating: ./data/segmented-images/images/cju7awzmu1ncs0871hziy65zx.jpg \r\n", + " inflating: ./data/segmented-images/images/cju7csvlb22fr0850lvm45n3x.jpg \r\n", + " inflating: ./data/segmented-images/images/cju87xn2snfmv0987sc3d9xnq.jpg \r\n", + " inflating: ./data/segmented-images/images/cju7bgnvb1sf808717qa799ir.jpg \r\n", + " inflating: ./data/segmented-images/images/cju7d7tly27h408016fyp5nr7.jpg \r\n", + " inflating: ./data/segmented-images/images/cju3yb47cj1xq0817zfotbni4.jpg \r\n", + " inflating: ./data/segmented-images/images/cju412uwlkva50850d1ps1ww7.jpg \r\n", + " inflating: ./data/segmented-images/images/cju85rkbnlo1c08503uxcpax1.jpg \r\n", + " inflating: ./data/segmented-images/images/cju88cddensj00987788yotmg.jpg \r\n", + " inflating: ./data/segmented-images/images/cju8b5p40r2c60987ofa0mu03.jpg \r\n", + " inflating: ./data/segmented-images/images/cju13cgqmnhwn0988yrainhcp.jpg \r\n", + " inflating: ./data/segmented-images/images/cju8d2q30tfhs0801n7lx77xl.jpg \r\n", + " inflating: ./data/segmented-images/images/cju2ysg748ru80878sp6j0gm0.jpg \r\n", + " inflating: ./data/segmented-images/images/cju3x7xsaijq80818f0psavav.jpg \r\n", + " inflating: ./data/segmented-images/images/cju85qefyln6v0850szeb9byi.jpg \r\n", + " inflating: ./data/segmented-images/images/cju2zdhsczmn50988z64qwg2q.jpg \r\n", + " inflating: ./data/segmented-images/images/cju7b1ygu1msd0801hywhy0mc.jpg \r\n", + " inflating: ./data/segmented-images/images/cju41lojblbs307555jdci937.jpg \r\n", + " inflating: ./data/segmented-images/images/cju34ouumcznz07996gg1xq7v.jpg \r\n", + " inflating: ./data/segmented-images/images/cju34zivp3fq80988opxbaqyn.jpg \r\n", + " inflating: ./data/segmented-images/images/cju2lberzkdzm09938cl40pog.jpg \r\n", + " inflating: ./data/segmented-images/images/cju6vvxsev9y30987kespucdg.jpg \r\n", + " inflating: ./data/segmented-images/images/cju2raxlosl630988jdbfy9b0.jpg \r\n", + " inflating: ./data/segmented-images/images/cju7atnm31if40817pqclnjer.jpg \r\n", + " inflating: ./data/segmented-images/images/cju5xq3tdm9fn0987pbedxdg5.jpg \r\n", + " inflating: ./data/segmented-images/images/cju183od81ff608017ekzif89.jpg \r\n", + " inflating: ./data/segmented-images/images/cju7aqkue1i2k09879uzcpt8r.jpg \r\n", + " inflating: ./data/segmented-images/images/cju2rn0hasxri0835nfy3buay.jpg \r\n", + " inflating: ./data/segmented-images/images/cju2t3ibkuecr0835o7si16zv.jpg \r\n", + " inflating: ./data/segmented-images/images/cju5x28nzm7t907558ocq4bt7.jpg \r\n", + " inflating: ./data/segmented-images/images/cju787jnjzjuj0871p94nck9g.jpg \r\n", + " inflating: ./data/segmented-images/images/cju2pag1f0s4r0878h52uq83s.jpg \r\n", + " inflating: ./data/segmented-images/images/cju5boicjagt20871b1fotkh4.jpg \r\n", + " inflating: ./data/segmented-images/images/cju3v72v5h1qz0818fggilwtq.jpg \r\n", + " inflating: ./data/segmented-images/images/cju1fj6axwfp30835ukhuzhw9.jpg \r\n", + " inflating: ./data/segmented-images/images/cju2phaksnahz0993yxogjcpv.jpg \r\n", + " inflating: ./data/segmented-images/images/cju87tyddnnad0755bj0wxahe.jpg \r\n", + " inflating: ./data/segmented-images/images/cju77q10sz9ug0801449wu1nu.jpg \r\n", + " inflating: ./data/segmented-images/images/cju35c4wzdhow0799h6eq4sgs.jpg \r\n", + " inflating: ./data/segmented-images/images/cju33yemn2qb20988wfjxximx.jpg \r\n", + " inflating: ./data/segmented-images/images/cju43eigtm6ev0801mv0m96t1.jpg \r\n", + " inflating: ./data/segmented-images/images/cju17z0qongpa0993de4boim4.jpg \r\n", + " inflating: ./data/segmented-images/images/cju7d6ux323ze0987xos3srkx.jpg \r\n", + " inflating: ./data/segmented-images/images/cju32fhnhbds40799broyoptc.jpg \r\n", + " inflating: ./data/segmented-images/images/cju1f320ewfyu0988ndz6blh5.jpg \r\n", + " inflating: ./data/segmented-images/images/cju16jgnyzp970878melv7r25.jpg \r\n", + " inflating: ./data/segmented-images/images/cju77b3wyz4160755qis4ljsb.jpg \r\n", + " inflating: ./data/segmented-images/images/cju5bycdkalkb09875f7bfrvx.jpg \r\n", + " inflating: ./data/segmented-images/images/cju2zpw4q9vzr0801p0lysjdl.jpg \r\n", + " inflating: ./data/segmented-images/images/cju32qr9tbvsj08013pkpjenq.jpg \r\n", + " inflating: ./data/segmented-images/images/cju3uhb79gcgr0871orbrbi3x.jpg \r\n", + " inflating: ./data/segmented-images/images/cju6wt9jvvn500871hjn3t3g0.jpg \r\n", + " inflating: ./data/segmented-images/images/cju2z6ez69g4u0801qwt088lw.jpg \r\n", + " inflating: ./data/segmented-images/images/cju303j5r062k098835zxfds5.jpg \r\n", + " inflating: ./data/segmented-images/images/cju30gxjq0djk0988jytm49rs.jpg \r\n", + " inflating: ./data/segmented-images/images/cju7cq6su27qv075574dir0r3.jpg \r\n", + " inflating: ./data/segmented-images/images/cju3tp94kfstl08181awh6z49.jpg \r\n", + " inflating: ./data/segmented-images/images/cju85l4yjlops0801fvmnwptf.jpg \r\n", + " inflating: ./data/segmented-images/images/cju8cbsyssiqj0871gr4jedjp.jpg \r\n", + " inflating: ./data/segmented-images/images/cju6yywx1whbb0871ksgfgf9f.jpg \r\n", + " inflating: ./data/segmented-images/images/cju1dg44i4z3w0801nyz4p6zf.jpg \r\n", + " inflating: ./data/segmented-images/images/cju31rb7vb6110801p9rhacuw.jpg \r\n", + " inflating: ./data/segmented-images/images/cju2zi4l09f5807991s8do2b2.jpg \r\n", + " inflating: ./data/segmented-images/images/cju1cj3f0qi5n0993ut8f49rj.jpg \r\n", + " inflating: ./data/segmented-images/images/cju8c1a0ws7o208181c6lbsom.jpg \r\n", + " inflating: ./data/segmented-images/images/cju30525w04r10835ygp257sb.jpg \r\n", + " inflating: ./data/segmented-images/images/cju7dlk532dsh0871zvr6qz0r.jpg \r\n", + " inflating: ./data/segmented-images/images/cju5vgawslbe30987ndeepc1b.jpg \r\n", + " inflating: ./data/segmented-images/images/cju6x97w4vwua0850x0997r0a.jpg \r\n", + " inflating: ./data/segmented-images/images/cju35mdz73x890835eynq1h9v.jpg \r\n", + " inflating: ./data/segmented-images/images/cju8bqxxurs6i0850mu7mtef9.jpg \r\n", + " inflating: ./data/segmented-images/images/cju7ea4om2l910801bohqjccy.jpg \r\n", + " inflating: ./data/segmented-images/images/cju8at3s1qqqx0850hcq8nmnq.jpg \r\n", + " inflating: ./data/segmented-images/images/cju6v3bb2v7xo085090blubyw.jpg \r\n", + " inflating: ./data/segmented-images/images/cju885ikhnmkn09878s2lqtuh.jpg \r\n", + " inflating: ./data/segmented-images/images/cju7b10ce1mnm08011c5bwyr4.jpg \r\n", + " inflating: ./data/segmented-images/images/cju88rl5eo94l0850kf5wtrm1.jpg \r\n", + " inflating: ./data/segmented-images/images/cju32srle1xfq083575i3fl75.jpg \r\n", + " inflating: ./data/segmented-images/images/cju34o6dbd2lo0855aqlcy1hs.jpg \r\n", + " inflating: ./data/segmented-images/images/cju1gi7jlwyld0835cdf6g6qz.jpg \r\n", + " inflating: ./data/segmented-images/images/cju3xwpgviwlx0871rwm15q7v.jpg \r\n", + " inflating: ./data/segmented-images/images/cju5knbbqfipk080128cggukq.jpg \r\n", + " inflating: ./data/segmented-images/images/cju8czvnztbf40871b4m7t78w.jpg \r\n", + " inflating: ./data/segmented-images/images/cju5ht88gedbu0755xrcuddcx.jpg \r\n", + " inflating: ./data/segmented-images/images/cju1ffnjn6ctm08015perkg37.jpg \r\n", + " inflating: ./data/segmented-images/images/cju2hgsptlfam0835o3b59h1o.jpg \r\n", + " inflating: ./data/segmented-images/images/cju85c2d4ln1b0755zz1z3onx.jpg \r\n", + " inflating: ./data/segmented-images/images/cju1h89h6xbnx08352k2790o9.jpg \r\n", + " inflating: ./data/segmented-images/images/cju3128yi0rpu0988o4oo5n8n.jpg \r\n", + " inflating: ./data/segmented-images/images/cju426tomlhll0818fc0i7nvh.jpg \r\n", + " inflating: ./data/segmented-images/images/cju3xjqtpikx50817tppy6g84.jpg \r\n", + " inflating: ./data/segmented-images/images/cju8567gdlcbq0801dwwyo2jt.jpg \r\n", + " inflating: ./data/segmented-images/images/cju5von04litr08718j8po40a.jpg \r\n", + " inflating: ./data/segmented-images/images/cju2ma647l0nj0993ot4deq2q.jpg \r\n", + " inflating: ./data/segmented-images/images/cju7arvfe1ldu0850erdmphgj.jpg \r\n", + " inflating: ./data/segmented-images/images/cju2yb31a8e8u0878wdashg7o.jpg \r\n", + " inflating: ./data/segmented-images/images/cju2nsmwjlzyl0993jl80chvz.jpg \r\n", + " inflating: ./data/segmented-images/images/cju88msmoo3470817m441j4sg.jpg \r\n", + " inflating: ./data/segmented-images/images/cju1dq3x1vgx109889c7wyirg.jpg \r\n", + " inflating: ./data/segmented-images/images/cju83qd0yjyht0817ktkfl268.jpg \r\n", + " inflating: ./data/segmented-images/images/cju45ofd9ne1j0801ri8dup7t.jpg \r\n", + " inflating: ./data/segmented-images/images/cju35fxqyzt5p0993vusm54qz.jpg \r\n", + " inflating: ./data/segmented-images/images/cju8b6rp0r5st0850184f79xt.jpg \r\n", + " inflating: ./data/segmented-images/images/cju7d3oc82cho0755dajlwldz.jpg \r\n", + " inflating: ./data/segmented-images/images/cju5cky5xb0ay0801oxet697t.jpg \r\n", + " inflating: ./data/segmented-images/images/cju3y9difj6th0801kd1rqm3w.jpg \r\n", + " inflating: ./data/segmented-images/images/cju6wn57mvooj0850rp78hhy7.jpg \r\n", + " inflating: ./data/segmented-images/images/cju8aw9n1qyg10801jkjlmors.jpg \r\n", + " inflating: ./data/segmented-images/images/cju2hw5gjlr5h0988so2qqres.jpg \r\n", + " inflating: ./data/segmented-images/images/cju428k5fldt108177s6g6f45.jpg \r\n", + " inflating: ./data/segmented-images/images/cju45pm27n80u08174kyow1gj.jpg \r\n", + " inflating: ./data/segmented-images/images/cju5uget8krjy0818kvywd0zu.jpg \r\n", + " inflating: ./data/segmented-images/images/cju2iatlki5u309930zmgkv6h.jpg \r\n", + " inflating: ./data/segmented-images/images/cju2oq5570avm079959o20op1.jpg \r\n", + " inflating: ./data/segmented-images/images/cju171py4qiha0835u8sl59ds.jpg \r\n", + " inflating: ./data/segmented-images/images/cju30nyxe0gfb0835p256yoju.jpg \r\n", + " inflating: ./data/segmented-images/images/cju83rcnzkbsj0755x5anfrcg.jpg \r\n", + " inflating: ./data/segmented-images/images/cju3v0fl3gwce0755qkjhzmd4.jpg \r\n", + " inflating: ./data/segmented-images/images/cju2y8s56ymqr083541ggdsml.jpg \r\n", + " inflating: ./data/segmented-images/images/cju18gzrq18zw0878wbf4ftw6.jpg \r\n", + " inflating: ./data/segmented-images/images/cju8a84g0q76m0818hwiggkod.jpg \r\n", + " inflating: ./data/segmented-images/images/cju8dk7eztzup08182yxko5zh.jpg \r\n", + " inflating: ./data/segmented-images/images/cju2zjcvj9qma0801dk71hhi0.jpg \r\n", + " inflating: ./data/segmented-images/images/cju8b7aqtr4a00987coba14b7.jpg \r\n", + " inflating: ./data/segmented-images/images/cju2hjrqcvi2j0801bx1i6gxg.jpg \r\n", + " inflating: ./data/segmented-images/images/cju17hw9hr9c5098800fu4u8e.jpg \r\n", + " inflating: ./data/segmented-images/images/cju3x2s11ibzi0817kk284k0j.jpg \r\n", + " inflating: ./data/segmented-images/images/cju5vutu7ll8w0871dfp92n9p.jpg \r\n", + " inflating: ./data/segmented-images/images/cju2zxja9w1eh09933609ho9z.jpg \r\n", + " inflating: ./data/segmented-images/images/cju5xneamme2p0801qdf7fdwv.jpg \r\n", + " inflating: ./data/segmented-images/images/cju358pwtdby20878cg7nm0np.jpg \r\n", + " inflating: ./data/segmented-images/images/cju1hs0za7jha0855vj0mdrjt.jpg \r\n", + " inflating: ./data/segmented-images/images/cju2syxa93yw40799x2iuwabz.jpg \r\n", + " inflating: ./data/segmented-images/images/cju5ktjwofed70817eg58ef7u.jpg \r\n", + " inflating: ./data/segmented-images/images/cju3ttznuftyf09875t11850w.jpg \r\n", + " inflating: ./data/segmented-images/images/cju7ctvqn25dy08186g442m1r.jpg \r\n", + " inflating: ./data/segmented-images/images/cju5eyfe9cpk90987laa7tsl3.jpg \r\n", + " inflating: ./data/segmented-images/images/cju310f6val1v0855xo8tc3gu.jpg \r\n", + " inflating: ./data/segmented-images/images/cju306x7w05nb0835cunv799x.jpg \r\n", + " inflating: ./data/segmented-images/images/cju8bop5jrsid08716i24fqda.jpg \r\n", + " inflating: ./data/segmented-images/images/cju5wrapcm2290818jsh26ppb.jpg \r\n", + " inflating: ./data/segmented-images/images/cju7dqcwi2dz00850gcmr2ert.jpg \r\n", + " inflating: ./data/segmented-images/images/cju1g4nsb6ngy0799l4ezm8ab.jpg \r\n", + " inflating: ./data/segmented-images/images/cju7aklv31h4309871m29l4e7.jpg \r\n", + " inflating: ./data/segmented-images/images/cju2i3hzclw3o0988rrgh911i.jpg \r\n", + " inflating: ./data/segmented-images/images/cju7er4kc2opa0801anuxc0eb.jpg \r\n", + " inflating: ./data/segmented-images/images/cju2xlcqxy9c60988vjacdznb.jpg \r\n", + " inflating: ./data/segmented-images/images/cju8bljw9rqk20801kr54akrl.jpg \r\n", + " inflating: ./data/segmented-images/images/cju45lbgznahl08180xz1h7u6.jpg \r\n", + " inflating: ./data/segmented-images/images/cju7b4mtw1n9n080186209f3d.jpg \r\n", + " inflating: ./data/segmented-images/images/cju7f9umg2olj0987fj5y285w.jpg \r\n", + " inflating: ./data/segmented-images/images/cju887ftknop008177nnjt46y.jpg \r\n", + " inflating: ./data/segmented-images/images/cju7b9vcs1luz0987ta60j1dy.jpg \r\n", + " inflating: ./data/segmented-images/images/cju8auylgqx0z0871u4o4db7o.jpg \r\n", + " inflating: ./data/segmented-images/images/cju3ya7goj6at0818v2l5ay7f.jpg \r\n", + " inflating: ./data/segmented-images/images/cju7eueum2oqn0850rodmx8zo.jpg \r\n", + " inflating: ./data/segmented-images/images/cju0u82z3cuma0835wlxrnrjv.jpg \r\n", + " inflating: ./data/segmented-images/images/cju2i03ptvkiu0799xbbd4det.jpg \r\n", + " inflating: ./data/segmented-images/images/cju87mrypnb1e0818scv1mxxg.jpg \r\n", + " inflating: ./data/segmented-images/images/cju7dizi82h2i0755doucgnt3.jpg \r\n", + " inflating: ./data/segmented-images/images/cju5waeduln160817w0agirve.jpg \r\n", + " inflating: ./data/segmented-images/images/cju88nroho44508500129f1nh.jpg \r\n", + " inflating: ./data/segmented-images/images/cju3yht87j83m08507yk1u1fg.jpg \r\n", + " inflating: ./data/segmented-images/images/cju1cfhyg48bb0799cl5pr2jh.jpg \r\n", + " inflating: ./data/segmented-images/images/cju8clorgsuwn08714toqb7v6.jpg \r\n", + " inflating: ./data/segmented-images/images/cju16fpvhzypl0799p9phnlx6.jpg \r\n", + " inflating: ./data/segmented-images/images/cju5wuhm1lwm40987vugqn3vv.jpg \r\n", + " inflating: ./data/segmented-images/images/cju7dz5yy2i7z0801ausi7rna.jpg \r\n", + " inflating: ./data/segmented-images/images/cju2xwm1s84l50799i60mq0pu.jpg \r\n", + " inflating: ./data/segmented-images/images/cju88itqbny720987hxizbj5y.jpg \r\n", + " inflating: ./data/segmented-images/images/cju7ev2b12owa08500bpfpwyw.jpg \r\n", + " inflating: ./data/segmented-images/images/cju84ih17kp5l09876bkooocl.jpg \r\n", + " inflating: ./data/segmented-images/images/cju7afqon1ip40850ue2308b6.jpg \r\n", + " inflating: ./data/segmented-images/images/cju14g8o4xui30878gkgbrvqj.jpg \r\n", + " inflating: ./data/segmented-images/images/cju5cu8qkb84x08186jwo8yin.jpg \r\n", + " inflating: ./data/segmented-images/images/cju5tgbzhjllu08174ca41eus.jpg \r\n", + " inflating: ./data/segmented-images/images/cju1f5x1164xv08555654c24r.jpg \r\n", + " inflating: ./data/segmented-images/images/cju0s690hkp960855tjuaqvv0.jpg \r\n", + " inflating: ./data/segmented-images/images/cju2rzpsmtb0f0835jabkbao1.jpg \r\n", + " inflating: ./data/segmented-images/images/cju7dvl5m2n4t0755hlnnjjet.jpg \r\n", + " inflating: ./data/segmented-images/images/cju7fen322ou10817ziqkob4k.jpg \r\n", + " inflating: ./data/segmented-images/images/cju1cnnziug1l0835yh4ropyg.jpg \r\n", + " inflating: ./data/segmented-images/images/cju88oh0po9gq0801nge4tgr1.jpg \r\n", + " inflating: ./data/segmented-images/images/cju8b3ka8r64u0801fh18hk7l.jpg \r\n", + " inflating: ./data/segmented-images/images/cju18849rrsgr0988p90hkygb.jpg \r\n", + " inflating: ./data/segmented-images/images/cju323ypb1fbb0988gx5rzudb.jpg \r\n", + " inflating: ./data/segmented-images/images/cju6wuojavt740818b5qcv3iw.jpg \r\n", + " inflating: ./data/segmented-images/images/cju0tl3uz8blh0993wxvn7ly3.jpg \r\n", + " inflating: ./data/segmented-images/images/cju2zkpdl9h7t0799ix60teqg.jpg \r\n", + " inflating: ./data/segmented-images/images/cju2urqpwvxw70835rvndvtsi.jpg \r\n", + " inflating: ./data/segmented-images/images/cju5ddda9bkkt0850enzwatb1.jpg \r\n", + " inflating: ./data/segmented-images/images/cju2hx006vidl0799igm81vmh.jpg \r\n", + " inflating: ./data/segmented-images/images/cju5nxkujgscq0817l9gss626.jpg \r\n", + " inflating: ./data/segmented-images/images/cju1dnz61vfp40988e78bkjga.jpg \r\n", + " inflating: ./data/segmented-images/images/cju83syhdk6gs0801rf1rekdl.jpg \r\n", + " inflating: ./data/segmented-images/images/cju42tauqlo5p08171l3cuo4b.jpg \r\n", + " inflating: ./data/segmented-images/images/cju7d1tvt25bu08019dvw3uff.jpg \r\n", + " inflating: ./data/segmented-images/images/cju16whaj0e7n0855q7b6cjkm.jpg \r\n", + " inflating: ./data/segmented-images/images/cju2lcyfgkf5809932fn9gucn.jpg \r\n", + " inflating: ./data/segmented-images/images/cju8c2rqzs5t80850d0zky5dy.jpg \r\n", + " inflating: ./data/segmented-images/images/cju3umoh1geet0817cmpef5am.jpg \r\n", + " inflating: ./data/segmented-images/images/cju34xspwzenf0993cyzajv9n.jpg \r\n", + " inflating: ./data/segmented-images/images/cju2sszfq3uye0878sucelzk2.jpg \r\n", + " inflating: ./data/segmented-images/images/cju84dsvaklpx098750hp83x4.jpg \r\n", + " inflating: ./data/segmented-images/images/cju6wjm81vgsc0987enk9n3pr.jpg \r\n", + " inflating: ./data/segmented-images/images/cju8bk8oirjhw0817hgkua2w8.jpg \r\n", + " inflating: ./data/segmented-images/images/cju8cwy02t9eq08185qn12c02.jpg \r\n", + " inflating: ./data/segmented-images/images/cju2s9g11pnra0993gn4eh793.jpg \r\n", + " inflating: ./data/segmented-images/images/cju2lz8vqktne0993fuym6drw.jpg \r\n", + " inflating: ./data/segmented-images/images/cju8bm24yrrdp081829mbo8ic.jpg \r\n" + ] + }, + { + "name": "stdout", + "output_type": "stream", + "text": [ + " inflating: ./data/segmented-images/images/cju8abobpqbir08189u01huru.jpg \r\n", + " inflating: ./data/segmented-images/images/cju8axq24r4an0755yhv9d4ly.jpg \r\n", + " inflating: ./data/segmented-images/images/cju32csyfblyh080170aa3x5p.jpg \r\n", + " inflating: ./data/segmented-images/images/cju76erapykj30871x5eaxh4q.jpg \r\n", + " inflating: ./data/segmented-images/images/cju77vvcwzcm50850lzoykuva.jpg \r\n", + " inflating: ./data/segmented-images/images/cju30df2j09dd08351ayx2t6w.jpg \r\n", + " inflating: ./data/segmented-images/images/cju43c92lm5cj0755lorsorfg.jpg \r\n", + " inflating: ./data/segmented-images/images/cju5vxuc5loxw0818u8xgf45p.jpg \r\n", + " inflating: ./data/segmented-images/images/cju0t4oil7vzk099370nun5h9.jpg \r\n", + " inflating: ./data/segmented-images/images/cju886ryxnsl50801r93jai7q.jpg \r\n", + " inflating: ./data/segmented-images/images/cju5eq8c8ck690850vix98hv3.jpg \r\n", + " inflating: ./data/segmented-images/images/cju5cjh3xattc0817j2vbulzi.jpg \r\n", + " inflating: ./data/segmented-images/images/cju1fb9236a110801yvg0fwju.jpg \r\n", + " inflating: ./data/segmented-images/images/cju3tsh4lfsok0987w6x3a0v1.jpg \r\n", + " inflating: ./data/segmented-images/images/cju2ricdv2iys0878sv1adh0u.jpg \r\n", + " inflating: ./data/segmented-images/images/cju5uzmaol56l0817flxh4w9p.jpg \r\n", + " inflating: ./data/segmented-images/images/cju42m60jlpcm08186kqppzqv.jpg \r\n", + " inflating: ./data/segmented-images/images/cju7el6xv2k520817qxx9wdr5.jpg \r\n", + " inflating: ./data/segmented-images/images/cju2np2k9zi3v079992ypxqkn.jpg \r\n", + " inflating: ./data/segmented-images/images/cju2zy1e49pqk0878t6ncqn12.jpg \r\n", + " inflating: ./data/segmented-images/images/cju2rz4k434s70855wwx3ddtx.jpg \r\n", + " inflating: ./data/segmented-images/images/cju15l5ubz9yh0855b3ivdpse.jpg \r\n", + " inflating: ./data/segmented-images/images/cju5es375cnzy0801nkq35ffs.jpg \r\n", + " inflating: ./data/segmented-images/images/cju85hguellg50818kwu3s8d1.jpg \r\n", + " inflating: ./data/segmented-images/images/cju5ew4h9cqaf0818rrczkmqh.jpg \r\n", + " inflating: ./data/segmented-images/images/cju7dmlgf2ebw0871ieqas5fh.jpg \r\n", + " inflating: ./data/segmented-images/images/cju77j66ez52p08019xygi0co.jpg \r\n", + " inflating: ./data/segmented-images/images/cju13fwthn9mq0835gacxgy01.jpg \r\n", + " inflating: ./data/segmented-images/images/cju8amfdtqi4x09871tygrgqe.jpg \r\n", + " inflating: ./data/segmented-images/images/cju1hirfi7ekp0855q0vgm9qq.jpg \r\n", + " inflating: ./data/segmented-images/images/cju2yv4imv6cz099314jveiib.jpg \r\n", + " inflating: ./data/segmented-images/images/cju35eg0tdmjt085525sb4bua.jpg \r\n", + " inflating: ./data/segmented-images/images/cju7adqyj1jcx08712r1ro5gx.jpg \r\n", + " inflating: ./data/segmented-images/images/cju8chdlqsu620755azjty1tj.jpg \r\n", + " inflating: ./data/segmented-images/images/cju2qn2fzs1vy0988l243cvzy.jpg \r\n", + " inflating: ./data/segmented-images/images/cju7ebe962hr409872ovibahw.jpg \r\n", + " inflating: ./data/segmented-images/images/cju5k503sfa5f0871lx0rpu5y.jpg \r\n", + " inflating: ./data/segmented-images/images/cju5f26ebcuai0818xlwh6116.jpg \r\n", + " inflating: ./data/segmented-images/images/cju2tqfgw4oat0799rn0g5b2z.jpg \r\n", + " inflating: ./data/segmented-images/images/cju2nqapmzvk20801f9us40dx.jpg \r\n", + " inflating: ./data/segmented-images/images/cju7etr3y2p4t0801cdzjj8ab.jpg \r\n", + " inflating: ./data/segmented-images/images/cju77bvg0yv4r0987yh60xmjo.jpg \r\n", + " inflating: ./data/segmented-images/images/cju83vvmik9wa08710yeh7cuk.jpg \r\n", + " inflating: ./data/segmented-images/images/cju5y4hgqmk0i08180rjhbwvp.jpg \r\n", + " inflating: ./data/segmented-images/images/cju17bz250pgd0799u1hqkj5u.jpg \r\n", + " inflating: ./data/segmented-images/images/cju5i39mreass0817au8p22zy.jpg \r\n", + " inflating: ./data/segmented-images/images/cju41r6v2lcww0871ps8k8pf5.jpg \r\n", + " inflating: ./data/segmented-images/images/cju5vi4nxlc530817uoqm2m7a.jpg \r\n", + " inflating: ./data/segmented-images/images/cju5u4pywk81x0817vn9pe14z.jpg \r\n", + " inflating: ./data/segmented-images/images/cju3y79ofj3va0871uqfb1mzo.jpg \r\n", + " inflating: ./data/segmented-images/images/cju87nkyrnb970801q84m47yt.jpg \r\n", + " inflating: ./data/segmented-images/images/cju1drnhbrb9409935wi7vkhg.jpg \r\n", + " inflating: ./data/segmented-images/images/cju5b9oyda4yr0850g9viziyv.jpg \r\n", + " inflating: ./data/segmented-images/images/cju7fob3x301u0755x985pmmq.jpg \r\n", + " inflating: ./data/segmented-images/images/cju2i6acqvo6l0799u20fift8.jpg \r\n", + " inflating: ./data/segmented-images/images/cju7epwj82koz098713apjnzo.jpg \r\n", + " inflating: ./data/segmented-images/images/cju2m56cryvqd0801gtn2yp8t.jpg \r\n", + " inflating: ./data/segmented-images/images/cju32zhbnc1oy0801iyv1ix6p.jpg \r\n", + " inflating: ./data/segmented-images/images/cju3381d8bz3h07991xtl7ra0.jpg \r\n", + " inflating: ./data/segmented-images/images/cju15jr8jz8sb0855ukmkswkz.jpg \r\n", + " inflating: ./data/segmented-images/images/cju3ua8u0g9rg0801uayhdxhu.jpg \r\n", + " inflating: ./data/segmented-images/images/cju17otoe119u0799nqcbl8n1.jpg \r\n", + " inflating: ./data/segmented-images/images/cju84aoa3ktwn0755pfl4gfwd.jpg \r\n", + " inflating: ./data/segmented-images/images/cju5wkonqlrl409877y8zvnub.jpg \r\n", + " inflating: ./data/segmented-images/images/cju8bw697rwg308177tg8huas.jpg \r\n", + " inflating: ./data/segmented-images/images/cju77u1sjz77b0817ft44r3fk.jpg \r\n", + " inflating: ./data/segmented-images/images/cju7efffp2ivf0817etg3jehl.jpg \r\n", + " inflating: ./data/segmented-images/images/cju890guyoiti098753yg6cdu.jpg \r\n", + " inflating: ./data/segmented-images/images/cju18ibp219ub08783i6o98g7.jpg \r\n", + " inflating: ./data/segmented-images/images/cju32l161bi1v07990vm376in.jpg \r\n", + " inflating: ./data/segmented-images/images/cju30ftgja7170855xl9bkdm0.jpg \r\n", + " inflating: ./data/segmented-images/images/cju1egx9pvz2n0988eoy8jp23.jpg \r\n", + " inflating: ./data/segmented-images/images/cju5u6wf0kh1t0755bg1ssixv.jpg \r\n", + " inflating: ./data/segmented-images/images/cju2rlqdnoz9k0993cpjae3x0.jpg \r\n", + " inflating: ./data/segmented-images/images/cju1cbokpuiw70988j4lq1fpi.jpg \r\n", + " inflating: ./data/segmented-images/images/cju18kevfrojc0835bn90f1in.jpg \r\n", + " inflating: ./data/segmented-images/images/cju5yhgznmkzb0801cji2vi8j.jpg \r\n", + " inflating: ./data/segmented-images/images/cju3y54kwj3nr0801biidlb4e.jpg \r\n", + " inflating: ./data/segmented-images/images/cju5chrxxawka0871qcj171yz.jpg \r\n", + " inflating: ./data/segmented-images/images/cju76o55nymqd0871h31sph9w.jpg \r\n", + " inflating: ./data/segmented-images/images/cju3518w2d838079939fqztbc.jpg \r\n", + " inflating: ./data/segmented-images/images/cju7deifq2fzn0755lc8idyh8.jpg \r\n", + " inflating: ./data/segmented-images/images/cju15wdt3zla10801odjiw7sy.jpg \r\n", + " inflating: ./data/segmented-images/images/cju5hjxaae3i40850h5z2laf5.jpg \r\n", + " inflating: ./data/segmented-images/images/cju2wx0gh7fpz0878wwyd9ep8.jpg \r\n", + " inflating: ./data/segmented-images/images/cju7es23b2vcp0755gpbm9s7v.jpg \r\n", + " inflating: ./data/segmented-images/images/cju8dm2cau2km0818jsv9eeq2.jpg \r\n", + " inflating: ./data/segmented-images/images/cju2y26c588bo07993ksd8eoz.jpg \r\n", + " inflating: ./data/segmented-images/images/cju2suk42469908015ngmq6f2.jpg \r\n", + " inflating: ./data/segmented-images/images/cju1ddr6p4k5z08780uuuzit2.jpg \r\n", + " inflating: ./data/segmented-images/images/cju5vcmrqla7i0817x4sp4pqw.jpg \r\n", + " inflating: ./data/segmented-images/images/cju1gv7106qd008784gk603mg.jpg \r\n", + " inflating: ./data/segmented-images/images/cju6vucxvvlda0755j7msqnya.jpg \r\n", + " inflating: ./data/segmented-images/images/cju7ekbo32pft0871fv7kzwb9.jpg \r\n", + " inflating: ./data/segmented-images/images/cju30ia8da2bq0799klnehml2.jpg \r\n", + " inflating: ./data/segmented-images/images/cju2ntxtdzlvu0799xl3j9pan.jpg \r\n", + " inflating: ./data/segmented-images/images/cju2otvvv0l7z0855x7we8cb0.jpg \r\n", + " inflating: ./data/segmented-images/images/cju8bgdmqrksy0801tozdmraa.jpg \r\n", + " inflating: ./data/segmented-images/images/cju30ajhw09sx0988qyahx9s8.jpg \r\n", + " inflating: ./data/segmented-images/images/cju8c3xs7sauj0801ieyzezr5.jpg \r\n", + " inflating: ./data/segmented-images/images/cju84hibuktj80871u519o71q.jpg \r\n", + " inflating: ./data/segmented-images/images/cju33231uy4gi0993qc7b1jch.jpg \r\n", + " inflating: ./data/segmented-images/images/ck2bxw18mmz1k0725litqq2mc.jpg \r\n", + " inflating: ./data/segmented-images/images/cju88q6h6obpd0871ckmiabbo.jpg \r\n", + " inflating: ./data/segmented-images/images/cju2wtwj87kys0855kx6mddzw.jpg \r\n", + " inflating: ./data/segmented-images/images/cju7ajnbo1gvm098749rdouk0.jpg \r\n", + " inflating: ./data/segmented-images/images/cju8a3nhbpwnb0850d37fo2na.jpg \r\n", + " inflating: ./data/segmented-images/images/cju6vrs1ov8cr098788h8gs6j.jpg \r\n", + " inflating: ./data/segmented-images/images/cju7ez7r22qbc08015xfoz2wb.jpg \r\n", + " inflating: ./data/segmented-images/images/cju3u4lxmg59o0755rz42b9en.jpg \r\n", + " inflating: ./data/segmented-images/images/cju1fmsyf6gxb0801cimx2gle.jpg \r\n", + " inflating: ./data/segmented-images/images/cju7ap09p1kz10850ldccjebj.jpg \r\n", + " inflating: ./data/segmented-images/images/cju1ejj7dvqfa0835ra184v5m.jpg \r\n", + " inflating: ./data/segmented-images/images/cju2wve9v7esz0878mxsdcy04.jpg \r\n", + " inflating: ./data/segmented-images/images/cju2ulk385h170799rlklxob0.jpg \r\n", + " inflating: ./data/segmented-images/images/cju7avvi51iox0817ym55y6tt.jpg \r\n", + " inflating: ./data/segmented-images/images/cju7d4jk723eu0817bqz2n39m.jpg \r\n", + " inflating: ./data/segmented-images/images/cju8brv16rx7f0818uf5n89pv.jpg \r\n", + " inflating: ./data/segmented-images/images/cju2hqt33lmra0988fr5ijv8j.jpg \r\n", + " inflating: ./data/segmented-images/images/cju2upu4evw7g08358guwozxv.jpg \r\n", + " inflating: ./data/segmented-images/images/cju7b3f5h1sm40755i572jden.jpg \r\n", + " inflating: ./data/segmented-images/images/cju6x0yqbvxqt0755dhxislgb.jpg \r\n", + " inflating: ./data/segmented-images/images/cju0ue769mxii08019zqgdbxn.jpg \r\n", + " inflating: ./data/segmented-images/images/cju5clr68b48r0755cmuvponm.jpg \r\n", + " inflating: ./data/segmented-images/images/cju85omszllp30850b6rm9mi3.jpg \r\n", + " inflating: ./data/segmented-images/images/cju2sevf53lkx08558h5bpaig.jpg \r\n", + " inflating: ./data/segmented-images/images/cju2yg5ht8i4p087800js8hp4.jpg \r\n", + " inflating: ./data/segmented-images/images/cju5gucasds9d0801019axylx.jpg \r\n", + " inflating: ./data/segmented-images/images/cju87ox0kncom0801b98hqnd2.jpg \r\n", + " inflating: ./data/segmented-images/images/cju7ecl9i2i060987xawjp4l0.jpg \r\n", + " inflating: ./data/segmented-images/images/cju5nyu31gv8e0871zpk74a2n.jpg \r\n", + " inflating: ./data/segmented-images/images/cju2qs32r1vys07999conmbvx.jpg \r\n", + " inflating: ./data/segmented-images/images/cju7dp3dw2k4n0755zhe003ad.jpg \r\n", + " inflating: ./data/segmented-images/images/cju2zm0axztpe0988r8s9twjr.jpg \r\n", + " inflating: ./data/segmented-images/images/cju1c4fcu40hl07992b8gj0c8.jpg \r\n", + " inflating: ./data/segmented-images/images/cju33za6l2qy70988jhrlp2ev.jpg \r\n", + " inflating: ./data/segmented-images/images/cju2ouil2mssu0993hvxsed6d.jpg \r\n", + " inflating: ./data/segmented-images/images/cju7bc95p1mdm0817yqj5jc6j.jpg \r\n", + " inflating: ./data/segmented-images/images/cju1f15k3w4ct0835cmde6ypo.jpg \r\n", + " inflating: ./data/segmented-images/images/cju40m0rjkpw80871z6n6yg1u.jpg \r\n", + " inflating: ./data/segmented-images/images/cju6vgdmivcvb08018fra5lnv.jpg \r\n", + " inflating: ./data/segmented-images/images/cju1hmff8tkp809931jps6fbr.jpg \r\n", + " inflating: ./data/segmented-images/images/cju8c5223s8j80850b4kealt4.jpg \r\n", + " inflating: ./data/segmented-images/images/cju439oazm2fu0871ma0vvrft.jpg \r\n", + " inflating: ./data/segmented-images/images/cju1819curo000988pd5xcqme.jpg \r\n", + " inflating: ./data/segmented-images/images/cju5fydrud94708507vo6oy21.jpg \r\n", + " inflating: ./data/segmented-images/images/cju2r11x7sdgx0988o8ule0wl.jpg \r\n", + " inflating: ./data/segmented-images/images/cju5wqonpm0e60801z88ewmy1.jpg \r\n", + " inflating: ./data/segmented-images/images/cju76lsehyia10987u54vn8rb.jpg \r\n", + " inflating: ./data/segmented-images/images/cju85ia1slh220987y7c20sm2.jpg \r\n", + " inflating: ./data/segmented-images/images/cju847pxykriq0755268ktrk2.jpg \r\n", + " inflating: ./data/segmented-images/images/cju33jon3ygbj0993pu22a4k6.jpg \r\n", + " inflating: ./data/segmented-images/images/cju3wstckialg0871xs0vevsj.jpg \r\n", + " inflating: ./data/segmented-images/images/cju5fi0yxd3ei0801v7u0yudn.jpg \r\n", + " inflating: ./data/segmented-images/images/cju2s2527pfyr0993l3h1149a.jpg \r\n", + " inflating: ./data/segmented-images/images/cju2tpfa5uyx408359datxqqj.jpg \r\n", + " inflating: ./data/segmented-images/images/cju2xbk0080y80801eghyddi2.jpg \r\n", + " inflating: ./data/segmented-images/images/cju15czxqp3lv0835jvhgzurz.jpg \r\n", + " inflating: ./data/segmented-images/images/cju30j1rgadut0801vuyrsnt8.jpg \r\n", + " inflating: ./data/segmented-images/images/cju77idwfz36d0871tzfzz51i.jpg \r\n", + " inflating: ./data/segmented-images/images/cju14hjh2ob2o0835ouz3r5aa.jpg \r\n", + " inflating: ./data/segmented-images/images/cju5hyi9yegob0755ho3do8en.jpg \r\n", + " inflating: ./data/segmented-images/images/cju3280wv1ir009882jze27tc.jpg \r\n", + " inflating: ./data/segmented-images/images/cju83k8fyjsxr0817d6nxs6r4.jpg \r\n", + " inflating: ./data/segmented-images/images/cju5hqz50e7o90850e0prlpa0.jpg \r\n", + " inflating: ./data/segmented-images/images/cju6uzxk0v83p0801rcwnexdu.jpg \r\n", + " inflating: ./data/segmented-images/images/cju0s2a9ekvms080138tjjpxr.jpg \r\n", + " inflating: ./data/segmented-images/images/cju42wamblrqn098798r2yyok.jpg \r\n", + " inflating: ./data/segmented-images/images/cju34m7h536wq0988xz7gx79v.jpg \r\n", + " inflating: ./data/segmented-images/images/cju17g6ykn1cs0993dww6qdi8.jpg \r\n", + " inflating: ./data/segmented-images/images/cju2s16zp317h0799gr67jqc2.jpg \r\n", + " inflating: ./data/segmented-images/images/cju2wxv0hxs2f09884w48v8fi.jpg \r\n", + " inflating: ./data/segmented-images/images/cju7dubap2g0w0801fgl42mg9.jpg \r\n", + " inflating: ./data/segmented-images/images/cju7do8c72dbo0801vxfzxdc4.jpg \r\n", + " inflating: ./data/segmented-images/images/cju2t2ivz43i10878zeg8r1br.jpg \r\n", + " inflating: ./data/segmented-images/images/cju5u8gz4kj5b07552e2wpkwp.jpg \r\n", + " inflating: ./data/segmented-images/images/cju302fqq9spc0878rrygyzzz.jpg \r\n", + " inflating: ./data/segmented-images/images/cju30k2z40ds308353kdew70n.jpg \r\n", + " inflating: ./data/segmented-images/images/cju1expq45zst0855rjqwwj4m.jpg \r\n", + " inflating: ./data/segmented-images/images/cju43lcnum9y10755bjs7z87f.jpg \r\n", + " inflating: ./data/segmented-images/images/cju43jcqim2cp08172dvjvyui.jpg \r\n", + " inflating: ./data/segmented-images/images/cju3xtufwiv9c0818djsc4cqd.jpg \r\n", + " inflating: ./data/segmented-images/images/cju2qqn5ys4uo0988ewrt2ip2.jpg \r\n", + " inflating: ./data/segmented-images/images/cju422cm8lfxn0818ojicxejb.jpg \r\n", + " inflating: ./data/segmented-images/images/cju2ro5jqsy680988pi6qsujw.jpg \r\n", + " inflating: ./data/segmented-images/images/cju84gpefknwm098714oq8q61.jpg \r\n", + " inflating: ./data/segmented-images/images/cju1b3zgj3d8e0801kpolea6c.jpg \r\n", + " inflating: ./data/segmented-images/images/cju1erep75us208553i4ofwwe.jpg \r\n", + " inflating: ./data/segmented-images/images/cju77196iyshb0850ycbto50a.jpg \r\n", + " inflating: ./data/segmented-images/images/cju1bm8063nmh07996rsjjemq.jpg \r\n", + " inflating: ./data/segmented-images/images/cju772304yw5t0818vbw8kkjf.jpg \r\n", + " inflating: ./data/segmented-images/images/cju2rqo702wpx0855fn7d5cxh.jpg \r\n", + " inflating: ./data/segmented-images/images/cju3v56bwgy8v0871w14pz8fx.jpg \r\n", + " inflating: ./data/segmented-images/images/cju2xs6na81t20878pt6nkfip.jpg \r\n", + " inflating: ./data/segmented-images/images/cju2osuru0ki00855txo0n3uu.jpg \r\n", + " inflating: ./data/segmented-images/images/cju41p90plcsx08018cnzpndc.jpg \r\n", + " inflating: ./data/segmented-images/images/cju30xqmh0ni00835ix3batv1.jpg \r\n", + " inflating: ./data/segmented-images/images/cju2nguelpmlj0835rojdn097.jpg \r\n", + " inflating: ./data/segmented-images/images/cju2srvy5440s0801y1ba9akr.jpg \r\n", + " inflating: ./data/segmented-images/images/cju7dbppn28nx085097654msi.jpg \r\n", + " inflating: ./data/segmented-images/images/cju5c5xc7algd0817pb1ej5yo.jpg \r\n", + " inflating: ./data/segmented-images/images/cju1efbr0rqxz09931z0lf4vf.jpg \r\n", + " inflating: ./data/segmented-images/images/cju8b4ja9r2s808509d45ma86.jpg \r\n", + " inflating: ./data/segmented-images/images/cju33x0f22peh0988g0ln7w5v.jpg \r\n", + " inflating: ./data/segmented-images/images/cju2z1nxlzaj40835wj81s1iy.jpg \r\n", + " inflating: ./data/segmented-images/images/cju7fnfv02tt90801djnix9m8.jpg \r\n", + " inflating: ./data/segmented-images/images/cju2hugv9vget0799hhk7ksvg.jpg \r\n", + " inflating: ./data/segmented-images/images/cju414lf2l1lt0801rl3hjllj.jpg \r\n", + " inflating: ./data/segmented-images/images/cju8b1v3br45u087189kku66u.jpg \r\n", + " inflating: ./data/segmented-images/images/cju7ff97z2ow40817u2r83my5.jpg \r\n", + " inflating: ./data/segmented-images/images/cju8cdeazsm8h0801jxifmzur.jpg \r\n", + " inflating: ./data/segmented-images/images/cju2yo1j1v0qz09934o0e683p.jpg \r\n" + ] + }, + { + "name": "stdout", + "output_type": "stream", + "text": [ + " inflating: ./data/segmented-images/images/cju14pxbaoksp0835qzorx6g6.jpg \r\n", + " inflating: ./data/segmented-images/images/cju5bj926aiec07559rshy4wa.jpg \r\n", + " inflating: ./data/segmented-images/images/cju2yi9tz8vky0801yqip0xyl.jpg \r\n", + " inflating: ./data/segmented-images/images/cju7fazv92ywx0755xov2erga.jpg \r\n", + " inflating: ./data/segmented-images/images/cju8aqq8uqmoq0987hphto9gg.jpg \r\n", + " inflating: ./data/segmented-images/images/cju40jl7skiuo0817p0smlgg8.jpg \r\n", + " inflating: ./data/segmented-images/images/cju2igw4gvxds0878808qj398.jpg \r\n", + " inflating: ./data/segmented-images/images/cju5ekty5ckzf07550c9u3ckk.jpg \r\n", + " inflating: ./data/segmented-images/images/cju5wcc90lu020850mjrxppv6.jpg \r\n", + " inflating: ./data/segmented-images/images/cju7druhp2gp308715i6km7be.jpg \r\n", + " inflating: ./data/segmented-images/images/cju8914beokbf0850isxpocrk.jpg \r\n", + " inflating: ./data/segmented-images/images/cju7ehljc2or70871261br8ai.jpg \r\n", + " inflating: ./data/segmented-images/images/cju83yddek68q0850d2x7zfkm.jpg \r\n", + " inflating: ./data/segmented-images/images/cju2uy8ox62jo0801g88hh42z.jpg \r\n", + " inflating: ./data/segmented-images/images/cju5udcufki0s09874ll1dbr5.jpg \r\n", + " inflating: ./data/segmented-images/images/cju83wwn1k55e0850kw6i2d81.jpg \r\n", + " inflating: ./data/segmented-images/images/cju2uzabhs6er0993x3aaf87p.jpg \r\n", + " inflating: ./data/segmented-images/images/cju300m3s04fg0988uzupuf7z.jpg \r\n", + " inflating: ./data/segmented-images/images/cju1f8w0t65en0799m9oacq0q.jpg \r\n", + " inflating: ./data/segmented-images/images/cju1dia8hvc6v098827mgffnm.jpg \r\n", + " inflating: ./data/segmented-images/images/cju8c82iosagu0817l74s4m5g.jpg \r\n", + " inflating: ./data/segmented-images/images/cju2y40d8ulqo0993q0adtgtb.jpg \r\n", + " inflating: ./data/segmented-images/images/cju2pjb9v0ywn0878j5g5n69j.jpg \r\n", + " inflating: ./data/segmented-images/images/cju7dwe282dc309876rco45ts.jpg \r\n", + " inflating: ./data/segmented-images/images/cju2rga4psq9n09881z519xx0.jpg \r\n", + " inflating: ./data/segmented-images/images/cju2wzu8wxtgu09880ku9x1pg.jpg \r\n", + " inflating: ./data/segmented-images/images/cju3tx1qyg0c907552fglumhc.jpg \r\n", + " inflating: ./data/segmented-images/images/cju7d8m3b2e210755l8fj1yph.jpg \r\n", + " inflating: ./data/segmented-images/images/cju34uhepd3dd0799hs8782ad.jpg \r\n", + " inflating: ./data/segmented-images/images/cju7dxffn2eam0817qxosfwch.jpg \r\n", + " inflating: ./data/segmented-images/images/cju8b0jr0r2oi0801jiquetd5.jpg \r\n", + " inflating: ./data/segmented-images/images/cju34aozyyy830993bn16u32n.jpg \r\n", + " inflating: ./data/segmented-images/images/cju5bwhapakm90987c1v4z46a.jpg \r\n", + " inflating: ./data/segmented-images/images/cju40sdwukv3k0755y99ug1k8.jpg \r\n", + " inflating: ./data/segmented-images/images/cju7emdni2py40871ivhxjtut.jpg \r\n", + " inflating: ./data/segmented-images/images/cju1aqqv02qwz0878a5cyhr67.jpg \r\n", + " inflating: ./data/segmented-images/images/cju85a8h8llwm07559wxg4t5w.jpg \r\n", + " inflating: ./data/segmented-images/images/cju334jzo261t0835yqudnfs1.jpg \r\n", + " inflating: ./data/segmented-images/images/cju5cetivauok0987ok3e5bre.jpg \r\n", + " inflating: ./data/segmented-images/images/cju8dpa89u6l80818dj6lldh9.jpg \r\n", + " inflating: ./data/segmented-images/images/cju88y1mwoln50871emyfny1g.jpg \r\n", + " inflating: ./data/segmented-images/images/cju7bduyq1rjf08719giru9ho.jpg \r\n", + " inflating: ./data/segmented-images/images/cju2t62nq45jl0799odpufwx6.jpg \r\n", + " inflating: ./data/segmented-images/images/cju410dnfl0960755y8lu8d79.jpg \r\n", + " inflating: ./data/segmented-images/images/cju8alhigqn2h0801zksudldd.jpg \r\n", + " inflating: ./data/segmented-images/images/cju884985nlmx0817vzpax3y4.jpg \r\n", + " inflating: ./data/segmented-images/images/cju32gzs6xo8x0993r8tedbpb.jpg \r\n", + " inflating: ./data/segmented-images/images/cju6z7e4bwgdd0987ogkzq9kt.jpg \r\n", + " inflating: ./data/segmented-images/images/cju7d9seq29zd0871nzl2uu5m.jpg \r\n", + " inflating: ./data/segmented-images/images/cju2hos57llxm08359g92p6jj.jpg \r\n", + " inflating: ./data/segmented-images/images/cju5uhrdwkmsu0817ervv91l8.jpg \r\n", + " inflating: ./data/segmented-images/images/cju85fc11ljr40818edpb0inh.jpg \r\n", + " inflating: ./data/segmented-images/images/cju5xjn5mm78b09871spyqhhr.jpg \r\n", + " inflating: ./data/segmented-images/images/cju2rpa30t07b0835im0erql0.jpg \r\n", + " inflating: ./data/segmented-images/images/cju7dtb1e2j0t0818deq51ib3.jpg \r\n", + " inflating: ./data/segmented-images/images/cju8bzzy2s66m08016z6mouqt.jpg \r\n", + " inflating: ./data/segmented-images/images/cju7bfx651qr80801cs7epotb.jpg \r\n", + " inflating: ./data/segmented-images/images/cju0u2g7pmnux0801vkk47ivj.jpg \r\n", + " inflating: ./data/segmented-images/images/cju5h57xedz5h0755mjpc8694.jpg \r\n", + " inflating: ./data/segmented-images/images/cju1d31sp4d4k0878r3fr02ul.jpg \r\n", + " inflating: ./data/segmented-images/images/cju87zv8lni0o0850hbbecbq6.jpg \r\n", + " inflating: ./data/segmented-images/images/cju5ukkg6kv7u08011x2b6zl5.jpg \r\n", + " inflating: ./data/segmented-images/images/cju5k7r0yf98c09878csbxb4d.jpg \r\n", + " inflating: ./data/segmented-images/images/cju32phw2bv130801yj7bkouq.jpg \r\n", + " inflating: ./data/segmented-images/images/cju77t0razbvm080106o56289.jpg \r\n", + " inflating: ./data/segmented-images/images/cju83nwu1jxte0987h1krpfmv.jpg \r\n", + " inflating: ./data/segmented-images/images/cju7aez2x1jtj0871ztezs3oi.jpg \r\n", + " inflating: ./data/segmented-images/images/cju88z8bson4h0871nnd7fdxo.jpg \r\n", + " inflating: ./data/segmented-images/images/cju5i5oh2efg60987ez6cpf72.jpg \r\n", + " inflating: ./data/segmented-images/images/cju8doa16u5gh0818w1ywda3q.jpg \r\n", + " inflating: ./data/segmented-images/images/cju5fs6j6d8350801vglraq4u.jpg \r\n", + " inflating: ./data/segmented-images/images/cju40w3hbkwpn08015rbs3wko.jpg \r\n", + " inflating: ./data/segmented-images/images/cju83ipu3jwpx0801z5pvguf8.jpg \r\n", + " inflating: ./data/segmented-images/images/cju2qfie4rvz508357kad9z5o.jpg \r\n", + " inflating: ./data/segmented-images/images/cju1c3218411b08014g9f6gig.jpg \r\n", + " inflating: ./data/segmented-images/images/cju6vifjlv55z0987un6y4zdo.jpg \r\n", + " inflating: ./data/segmented-images/images/cju430pm2lz0y0755jkhcc3d1.jpg \r\n", + " inflating: ./data/segmented-images/images/cju5ymyd8mmdc0801ry3by1xr.jpg \r\n", + " inflating: ./data/segmented-images/images/cju5ygh1zmmdi0755uod5e17i.jpg \r\n", + " inflating: ./data/segmented-images/images/cju89y9h0puti0818i5yw29e6.jpg \r\n", + " inflating: ./data/segmented-images/images/cju5ca9hcatkc0801jzwe7tfx.jpg \r\n", + " inflating: ./data/segmented-images/images/cju7ejm2l2ncl0801wq6y84nw.jpg \r\n", + " inflating: ./data/segmented-images/images/cju34repocy5208780gswillm.jpg \r\n", + " inflating: ./data/segmented-images/images/cju8ayeq7r1fb0818z1junacy.jpg \r\n", + " inflating: ./data/segmented-images/images/cju1alwgo30z60855fm3y23sm.jpg \r\n", + " inflating: ./data/segmented-images/images/cju7eotqi2qea0871y8yc7tqh.jpg \r\n", + " inflating: ./data/segmented-images/images/cju2r6mt2om21099352pny5gw.jpg \r\n", + " inflating: ./data/segmented-images/images/cju857ad0l88m0817qx4cwxnf.jpg \r\n", + " inflating: ./data/segmented-images/images/cju5tenjojp1j0755ms4949h2.jpg \r\n", + " inflating: ./data/segmented-images/images/cju6z9a9kwsl007552s49rx6i.jpg \r\n", + " inflating: ./data/segmented-images/images/cju34i3qvcyog0855qiejxx5w.jpg \r\n", + " inflating: ./data/segmented-images/images/cju2p4ddkmzxj0993p94o62av.jpg \r\n", + " inflating: ./data/segmented-images/images/cju7f900s2o0k08175gl1giid.jpg \r\n", + " inflating: ./data/segmented-images/images/cju5yeqiwmkgl0801fzv2douc.jpg \r\n", + " inflating: ./data/segmented-images/images/cju175facms5f0993a5tjikvt.jpg \r\n", + " inflating: ./data/segmented-images/images/cju30mm25a53s0799qa5wiqe8.jpg \r\n", + " inflating: ./data/segmented-images/images/cju353d1eda8c07992afde611.jpg \r\n", + " inflating: ./data/segmented-images/images/cju2z45kuzf6d0988nz2c819m.jpg \r\n", + " inflating: ./data/segmented-images/images/cju33belnbyhm0878yxl42233.jpg \r\n", + " inflating: ./data/segmented-images/images/cju7ddtz729960801uazp1knc.jpg \r\n", + " inflating: ./data/segmented-images/images/cju3xeexgii1j0817zs68tb4g.jpg \r\n", + " inflating: ./data/segmented-images/images/cju1h5w4wxajx0835mc954kxy.jpg \r\n", + " inflating: ./data/segmented-images/images/cju1gkndf6yi10801o1qnje19.jpg \r\n", + " inflating: ./data/segmented-images/images/cju7dn24o296i09871qfxb8s2.jpg \r\n", + " inflating: ./data/segmented-images/images/cju6x4t13vyw60755gtcf9ndu.jpg \r\n", + " inflating: ./data/segmented-images/images/cju31t8xd17bk0835rnb893jk.jpg \r\n", + " inflating: ./data/segmented-images/images/cju3xuj20ivgp0818mij8bjrd.jpg \r\n", + " inflating: ./data/segmented-images/images/cju33mirdc8mj0799k33wzoes.jpg \r\n", + " inflating: ./data/segmented-images/images/cju42romflni20817etb9a0fl.jpg \r\n", + " inflating: ./data/segmented-images/images/cju5yclrymlgj0818k426ud6z.jpg \r\n", + " inflating: ./data/segmented-images/images/cju1fyb1d69et0878muzdak9u.jpg \r\n", + " inflating: ./data/segmented-images/images/cju8402x1kcy70801t6kz6bdi.jpg \r\n", + " inflating: ./data/segmented-images/images/cju84kplnl1y30755ropua1b0.jpg \r\n", + " inflating: ./data/segmented-images/images/cju326h4v1gxw08352px40p7r.jpg \r\n", + " inflating: ./data/segmented-images/images/cju40r6jrksyk0871wg98zgho.jpg \r\n", + " inflating: ./data/segmented-images/images/cju77afzlz3kp07550x5nafzs.jpg \r\n", + " inflating: ./data/segmented-images/images/cju0qx73cjw570799j4n5cjze.jpg \r\n", + " inflating: ./data/segmented-images/images/cju1ats0y372e08011yazcsxm.jpg \r\n", + " inflating: ./data/segmented-images/images/cju8ca4geseia0850i2ru11hw.jpg \r\n", + " inflating: ./data/segmented-images/images/cju7dglf226g50987ohbthl19.jpg \r\n", + " inflating: ./data/segmented-images/images/cju43kj2pm34f0850l28ahpni.jpg \r\n", + " inflating: ./data/segmented-images/images/cju6yxyt0wh080871sqpepu47.jpg \r\n", + " inflating: ./data/segmented-images/images/cju2p0eveqtdc0835gpi3p93i.jpg \r\n", + " inflating: ./data/segmented-images/images/cju1c8ffau5770835g0g343o8.jpg \r\n", + " inflating: ./data/segmented-images/images/cju2lejzcy4pc0878c9rlonot.jpg \r\n", + " inflating: ./data/segmented-images/images/cju2zgbj9zmrw0835nnlzxj4c.jpg \r\n", + " inflating: ./data/segmented-images/images/cju1d96gsv62d09881b3wecw2.jpg \r\n", + " inflating: ./data/segmented-images/images/cju5kre09fhka0850h7b1898j.jpg \r\n", + " inflating: ./data/segmented-images/images/cju2mh8t6p07008350e01tx2a.jpg \r\n", + " inflating: ./data/segmented-images/images/cju8828oxnool0801qno9luhr.jpg \r\n", + " inflating: ./data/segmented-images/images/cju5jx7jzf7c90871c2i9aiov.jpg \r\n", + " inflating: ./data/segmented-images/images/cju6vta3kvazg0817qbeppjtm.jpg \r\n", + " inflating: ./data/segmented-images/images/cju2xa3i4y0160988i679zsqd.jpg \r\n", + " inflating: ./data/segmented-images/images/cju1f79yhsb5w0993txub59ol.jpg \r\n", + " inflating: ./data/segmented-images/images/cju16ach3m1da0993r1dq3sn2.jpg \r\n", + " inflating: ./data/segmented-images/images/cju7flevb2wii08188otgs9p2.jpg \r\n", + " inflating: ./data/segmented-images/images/cju2rmd2rsw9g09888hh1efu0.jpg \r\n", + " inflating: ./data/segmented-images/images/cju32a52lb9rc0799xi40qs00.jpg \r\n", + " inflating: ./data/segmented-images/images/cju1ewnoh5z030855vpex9uzt.jpg \r\n", + " inflating: ./data/segmented-images/images/cju0qkwl35piu0993l0dewei2.jpg \r\n", + " inflating: ./data/segmented-images/images/cju15ptjtppz40988odsm9azx.jpg \r\n", + " inflating: ./data/segmented-images/images/cju35atpxdjot0855q46aqrd0.jpg \r\n", + " inflating: ./data/segmented-images/images/cju2xyd9vyi7m098831qcucse.jpg \r\n", + " inflating: ./data/segmented-images/images/cju17v6ih0u7808783zcbg1jy.jpg \r\n", + " inflating: ./data/segmented-images/images/cju2qozsk20cq0855ugrg3cri.jpg \r\n", + " inflating: ./data/segmented-images/images/cju3u39fog1bo0871lxjrabks.jpg \r\n", + " inflating: ./data/segmented-images/images/cju8ashhnquqr0801rwduzt7d.jpg \r\n", + " inflating: ./data/segmented-images/images/cju30ywtc0oar0835bp2en7ec.jpg \r\n", + " inflating: ./data/segmented-images/images/cju7f5ghb2r5s0801chwkxxh9.jpg \r\n", + " inflating: ./data/segmented-images/images/cju3ykamdj9u208503pygyuc8.jpg \r\n", + " inflating: ./data/segmented-images/images/cju2p91qir00k08350ddfif0w.jpg \r\n", + " inflating: ./data/segmented-images/images/cju8a1jtvpt9m081712iwkca7.jpg \r\n", + " inflating: ./data/segmented-images/images/cju1hyolc7aqu0878rrkfn1lr.jpg \r\n", + " inflating: ./data/segmented-images/images/cju3uz4o6gr9z0850lhxyxvsj.jpg \r\n", + " inflating: ./data/segmented-images/images/cju5woy82m07m08505dmjg7g1.jpg \r\n", + " inflating: ./data/segmented-images/images/cju2r91dg2k090801bh0xzbxk.jpg \r\n", + " inflating: ./data/segmented-images/images/cju2ueb6j5ado0878vf5md13o.jpg \r\n", + " inflating: ./data/segmented-images/images/cju2uwz9f5yf1085506cfamfx.jpg \r\n", + " inflating: ./data/segmented-images/images/cju1dfeupuzlw0835gnxip369.jpg \r\n", + " inflating: ./data/segmented-images/images/cju0qoxqj9q6s0835b43399p4.jpg \r\n", + " inflating: ./data/segmented-images/images/cju2x7vw87mu30878hye2ca0m.jpg \r\n", + " inflating: ./data/segmented-images/images/cju40poe4kt9s0755f9cnm3h5.jpg \r\n", + " inflating: ./data/segmented-images/images/cju2nd7l7z98o0799gfjvyfmw.jpg \r\n", + " inflating: ./data/segmented-images/images/cju84jdl9kv0i0871eog9b3i9.jpg \r\n" + ] + }, + { + "name": "stdout", + "output_type": "stream", + "text": [ + " inflating: ./data/segmented-images/images/cju8dn0c3u2v50801k8rvq02f.jpg \r\n", + " inflating: ./data/segmented-images/images/cju3x4blzieu30850x10uuvbm.jpg \r\n", + " inflating: ./data/segmented-images/images/cjyzkmjy8evns070165gf9dmq.jpg \r\n", + " inflating: ./data/segmented-images/images/cju40wto5kxwi0755it190f2k.jpg \r\n", + " inflating: ./data/segmented-images/images/cjyztzaqtrv430848l8xgcerw.jpg \r\n", + " inflating: ./data/segmented-images/images/cjyzurzvohqnr0794es1itzek.jpg \r\n", + " inflating: ./data/segmented-images/images/cju2oi8sq0i2y0801mektzvw8.jpg \r\n", + " inflating: ./data/segmented-images/images/cju7eea9b2m0z0801ynqv1fqu.jpg \r\n", + " inflating: ./data/segmented-images/images/cju8dqkrqu83i0818ev74qpxq.jpg \r\n", + " inflating: ./data/segmented-images/images/cjyzkpsbjdsjq07211dfi4sru.jpg \r\n", + " inflating: ./data/segmented-images/images/cjyzk8qieoboa0848ogj51wwm.jpg \r\n", + " inflating: ./data/segmented-images/images/cjyzu3reghjya0794w7pwoi50.jpg \r\n", + " inflating: ./data/segmented-images/images/cjyzufihqquiw0a46jatrbwln.jpg \r\n", + " inflating: ./data/segmented-images/images/cju85je7vlht70817c9jcjwi4.jpg \r\n", + " inflating: ./data/segmented-images/images/cjz14qsk2wci60794un9ozwmw.jpg \r\n", + " inflating: ./data/segmented-images/images/ck2bxlujamu330725szlc2jdu.jpg \r\n", + " inflating: ./data/segmented-images/images/cju0sr5ghl0nd08789uzf1raf.jpg \r\n", + " inflating: ./data/segmented-images/images/cju2yljr0yzhw0988ecf271ly.jpg \r\n", + " inflating: ./data/segmented-images/images/cju2xd75m82720801q4s4ik3n.jpg \r\n", + " inflating: ./data/segmented-images/images/ck2da7fwcjfis07218r1rvm95.jpg \r\n", + " inflating: ./data/segmented-images/images/ck2395w2mb4vu07480otsu6tw.jpg \r\n", + " inflating: ./data/segmented-images/images/cjyzuio1qgh040763k56deohv.jpg \r\n", + " inflating: ./data/segmented-images/images/cjyzlw7f9faqr070129au64sq.jpg \r\n", + " inflating: ./data/segmented-images/images/cjyzjzssvd8pq0838f4nolj5l.jpg \r\n", + " inflating: ./data/segmented-images/images/cjyzul1qggwwj07216mhiv5sy.jpg \r\n", + " inflating: ./data/segmented-images/images/ck2bxqz3evvg20794iiyv5v2m.jpg \r\n", + " inflating: ./data/segmented-images/images/cjyzl833ndne80838pzuq6ila.jpg \r\n", + " inflating: ./data/segmented-images/images/cjyzu9th0qt4r0a46pyl4zik0.jpg \r\n", + " inflating: ./data/segmented-images/license.txt \r\n" + ] + } + ], "source": [ "!wget 'https://datasets.simula.no/hyper-kvasir/hyper-kvasir-segmented-images.zip' -O kvasir.zip\n", "ZIP_SHA384 = 'e30d18a772c6520476e55b610a4db457237f151e'\\\n", @@ -94,7 +2268,7 @@ }, { "cell_type": "code", - "execution_count": null, + "execution_count": 5, "metadata": {}, "outputs": [], "source": [ @@ -103,7 +2277,7 @@ }, { "cell_type": "code", - "execution_count": null, + "execution_count": 6, "metadata": { "code_folding": [], "scrolled": false @@ -175,7 +2349,7 @@ }, { "cell_type": "code", - "execution_count": null, + "execution_count": 7, "metadata": { "code_folding": [] }, @@ -240,7 +2414,7 @@ }, { "cell_type": "code", - "execution_count": null, + "execution_count": 8, "metadata": { "code_folding": [] }, @@ -404,7 +2578,7 @@ }, { "cell_type": "code", - "execution_count": null, + "execution_count": 9, "metadata": {}, "outputs": [], "source": [ @@ -420,9 +2594,23 @@ }, { "cell_type": "code", - "execution_count": null, + "execution_count": 10, "metadata": {}, - "outputs": [], + "outputs": [ + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">[17:44:12] WARNING tried to remove tensor: __opt_state_needed not present in the tensor dict <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/utilities/utils.py\">utils.py</a>:85\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f752a79ed90>" + ] + }, + "metadata": {}, + "output_type": "display_data" + } + ], "source": [ "# Create a federated model using the pytorch class, optimizer function, and loss function\n", "fl_model = FederatedModel(build_model=UNet, optimizer=optimizer,\n", @@ -431,9 +2619,36 @@ }, { "cell_type": "code", - "execution_count": null, + "execution_count": 11, "metadata": {}, - "outputs": [], + "outputs": [ + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> WARNING tried to remove tensor: __opt_state_needed not present in the tensor dict <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/utilities/utils.py\">utils.py</a>:85\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f7529a69820>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">[17:44:13] WARNING tried to remove tensor: __opt_state_needed not present in the tensor dict <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/utilities/utils.py\">utils.py</a>:85\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f75299b8b20>" + ] + }, + "metadata": {}, + "output_type": "display_data" + } + ], "source": [ "collaborator_models = fl_model.setup(num_collaborators=2)\n", "collaborators = {'one': collaborator_models[0], 'two': collaborator_models[1]}" @@ -448,14 +2663,83 @@ }, { "cell_type": "code", - "execution_count": null, + "execution_count": 12, "metadata": { "code_folding": [] }, - "outputs": [], + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "{\n", + " \"aggregator.settings.best_state_path\": \"save/torch_unet_kvasir_best.pbuf\",\n", + " \"aggregator.settings.db_store_rounds\": 1,\n", + " \"aggregator.settings.init_state_path\": \"save/torch_unet_kvasir_init.pbuf\",\n", + " \"aggregator.settings.last_state_path\": \"save/torch_unet_kvasir_last.pbuf\",\n", + " \"aggregator.settings.rounds_to_train\": 40,\n", + " \"aggregator.template\": \"openfl.component.Aggregator\",\n", + " \"assigner.settings.task_groups\": [\n", + " {\n", + " \"name\": \"train_and_validate\",\n", + " \"percentage\": 1.0,\n", + " \"tasks\": [\n", + " \"aggregated_model_validation\",\n", + " \"train\",\n", + " \"locally_tuned_model_validation\"\n", + " ]\n", + " }\n", + " ],\n", + " \"assigner.template\": \"openfl.component.RandomGroupedAssigner\",\n", + " \"collaborator.settings.db_store_rounds\": 1,\n", + " \"collaborator.settings.delta_updates\": false,\n", + " \"collaborator.settings.epochs_per_round\": 1.0,\n", + " \"collaborator.settings.opt_treatment\": \"RESET\",\n", + " \"collaborator.settings.polling_interval\": 4,\n", + " \"collaborator.template\": \"openfl.component.Collaborator\",\n", + " \"data_loader.settings.batch_size\": 4,\n", + " \"data_loader.settings.collaborator_count\": 2,\n", + " \"data_loader.settings.data_group_name\": \"kvasir\",\n", + " \"data_loader.template\": \"code.data_loader.PyTorchKvasirDataLoader\",\n", + " \"network.settings.agg_addr\": \"nnlicv611.inn.intel.com\",\n", + " \"network.settings.agg_port\": 53966,\n", + " \"network.settings.cert_folder\": \"cert\",\n", + " \"network.settings.client_reconnect_interval\": 5,\n", + " \"network.settings.disable_client_auth\": false,\n", + " \"network.settings.disable_tls\": false,\n", + " \"network.settings.hash_salt\": \"auto\",\n", + " \"network.template\": \"openfl.federation.Network\",\n", + " \"task_runner.settings.n_channels\": 3,\n", + " \"task_runner.settings.n_classes\": 1,\n", + " \"task_runner.template\": \"code.fed_unet_runner.PyTorchFederatedUnet\",\n", + " \"tasks.aggregated_model_validation.function\": \"validate\",\n", + " \"tasks.aggregated_model_validation.kwargs\": {\n", + " \"apply\": \"global\",\n", + " \"metrics\": [\n", + " \"dice_coef\"\n", + " ]\n", + " },\n", + " \"tasks.locally_tuned_model_validation.function\": \"validate\",\n", + " \"tasks.locally_tuned_model_validation.kwargs\": {\n", + " \"apply\": \"local\",\n", + " \"metrics\": [\n", + " \"dice_coef\"\n", + " ]\n", + " },\n", + " \"tasks.settings\": {},\n", + " \"tasks.train.function\": \"train_batches\",\n", + " \"tasks.train.kwargs\": {\n", + " \"metrics\": [\n", + " \"loss\"\n", + " ]\n", + " }\n", + "}\n" + ] + } + ], "source": [ "# Get the current values of the FL plan. Each of these can be overridden\n", - "print(json.dumps(fx.get_plan(), indent=4, sort_keys=True))" + "print(fx.get_plan())" ] }, { @@ -467,11 +2751,4254 @@ }, { "cell_type": "code", - "execution_count": null, + "execution_count": 13, "metadata": { "scrolled": true }, - "outputs": [], + "outputs": [ + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO Updating aggregator.settings.rounds_to_train to 30... <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/native/native.py\">native.py</a>:82\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f75299ec340>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO Building 🡆 Object NoCompressionPipeline from openfl.pipelines Module. <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/federated/plan/plan.py\">plan.py</a>:162\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f75299ec7f0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO Settings 🡆 {} <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/federated/plan/plan.py\">plan.py</a>:165\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f75299eca60>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO Override 🡆 {} <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/federated/plan/plan.py\">plan.py</a>:167\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f75299ec5b0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO Creating Initial Weights File 🠆 save/torch_unet_kvasir_init.pbuf <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/native/native.py\">native.py</a>:240\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f7490d61c70>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO Starting Experiment... <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/native/native.py\">native.py</a>:244\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f7490d61ca0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO Building 🡆 Object RandomGroupedAssigner from openfl.component Module. <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/federated/plan/plan.py\">plan.py</a>:162\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f7490d61d90>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO Settings 🡆 {'task_groups': [{'name': 'train_and_validate', 'percentage': 1.0, 'tasks': ['aggregated_model_validation', 'train', <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/federated/plan/plan.py\">plan.py</a>:165\n", + " 'locally_tuned_model_validation']}], 'authorized_cols': ['one', 'two'], 'rounds_to_train': 40, 'tasks': {'defaults': \n", + " 'plan/defaults/tasks_torch.yaml', 'settings': {}, 'aggregated_model_validation': {'function': 'validate', 'kwargs': {'apply': \n", + " 'global', 'metrics': ['dice_coef']}}, 'train': {'function': 'train_batches', 'kwargs': {'metrics': ['loss']}}, \n", + " 'locally_tuned_model_validation': {'function': 'validate', 'kwargs': {'apply': 'local', 'metrics': ['dice_coef']}}}} \n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f7490d61dc0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO Override 🡆 {'defaults': 'plan/defaults/assigner.yaml'} <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/federated/plan/plan.py\">plan.py</a>:167\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f7490d61af0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO Building 🡆 Object Aggregator from openfl.component Module. <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/federated/plan/plan.py\">plan.py</a>:162\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f7490d2ef70>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO Settings 🡆 {'last_state_path': 'save/torch_unet_kvasir_last.pbuf', 'db_store_rounds': 1, 'init_state_path': <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/federated/plan/plan.py\">plan.py</a>:165\n", + " 'save/torch_unet_kvasir_init.pbuf', 'rounds_to_train': 30, 'best_state_path': 'save/torch_unet_kvasir_best.pbuf', \n", + " 'aggregator_uuid': 'aggregator_plan.yaml_851debec', 'federation_uuid': 'plan.yaml_851debec', 'authorized_cols': ['one', 'two'], \n", + " 'assigner': <openfl.component.assigner.random_grouped_assigner.RandomGroupedAssigner object at 0x7f75299ec520>} \n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f7490d2efa0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO Override 🡆 {'defaults': 'plan/defaults/aggregator.yaml'} <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/federated/plan/plan.py\">plan.py</a>:167\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f7490d2ee80>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO Building 🡆 Object Collaborator from openfl.component Module. <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/federated/plan/plan.py\">plan.py</a>:162\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f7490c9eac0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO Settings 🡆 {'db_store_rounds': 1, 'epochs_per_round': 1.0, 'delta_updates': False, 'polling_interval': 4, 'opt_treatment': <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/federated/plan/plan.py\">plan.py</a>:165\n", + " 'RESET', 'collaborator_name': 'one', 'aggregator_uuid': 'aggregator_plan.yaml_851debec', 'federation_uuid': \n", + " 'plan.yaml_851debec', 'task_runner': <openfl.federated.task.fl_model.FederatedModel object at 0x7f7529a742b0>, 'tensor_pipe': \n", + " <openfl.pipelines.no_compression_pipeline.NoCompressionPipeline object at 0x7f7529a746a0>, 'task_config': {'defaults': \n", + " 'plan/defaults/tasks_torch.yaml', 'settings': {}, 'aggregated_model_validation': {'function': 'validate', 'kwargs': {'apply': \n", + " 'global', 'metrics': ['dice_coef']}}, 'train': {'function': 'train_batches', 'kwargs': {'metrics': ['loss']}}, \n", + " 'locally_tuned_model_validation': {'function': 'validate', 'kwargs': {'apply': 'local', 'metrics': ['dice_coef']}}}, 'client': \n", + " <openfl.component.aggregator.aggregator.Aggregator object at 0x7f7490d61c40>} \n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f75299ecf10>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO Override 🡆 {'defaults': 'plan/defaults/collaborator.yaml'} <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/federated/plan/plan.py\">plan.py</a>:167\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f7490c9eac0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">[17:44:14] INFO Building 🡆 Object Collaborator from openfl.component Module. <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/federated/plan/plan.py\">plan.py</a>:162\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f7490c9e130>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO Settings 🡆 {'db_store_rounds': 1, 'epochs_per_round': 1.0, 'delta_updates': False, 'polling_interval': 4, 'opt_treatment': <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/federated/plan/plan.py\">plan.py</a>:165\n", + " 'RESET', 'collaborator_name': 'two', 'aggregator_uuid': 'aggregator_plan.yaml_851debec', 'federation_uuid': \n", + " 'plan.yaml_851debec', 'task_runner': <openfl.federated.task.fl_model.FederatedModel object at 0x7f7529a74490>, 'tensor_pipe': \n", + " <openfl.pipelines.no_compression_pipeline.NoCompressionPipeline object at 0x7f7529a746a0>, 'task_config': {'defaults': \n", + " 'plan/defaults/tasks_torch.yaml', 'settings': {}, 'aggregated_model_validation': {'function': 'validate', 'kwargs': {'apply': \n", + " 'global', 'metrics': ['dice_coef']}}, 'train': {'function': 'train_batches', 'kwargs': {'metrics': ['loss']}}, \n", + " 'locally_tuned_model_validation': {'function': 'validate', 'kwargs': {'apply': 'local', 'metrics': ['dice_coef']}}}, 'client': \n", + " <openfl.component.aggregator.aggregator.Aggregator object at 0x7f7490d61c40>, 'defaults': 'plan/defaults/collaborator.yaml'} \n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f7490c9edf0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO Override 🡆 {'defaults': 'plan/defaults/collaborator.yaml'} <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/federated/plan/plan.py\">plan.py</a>:167\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f7490c9efd0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO Waiting for tasks... <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\">collaborator.py</a>:156\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f7490c9eee0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO Sending tasks to collaborator one for round 0 <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\">aggregator.py</a>:276\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f7490c9ebb0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO Received the following tasks: ['aggregated_model_validation', 'train', 'locally_tuned_model_validation'] <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\">collaborator.py</a>:144\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f7490c9eca0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "name": "stderr", + "output_type": "stream", + "text": [ + "/home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/federated/task/runner_pt.py:287: UserWarning: The given NumPy array is not writeable, and PyTorch does not support non-writeable tensors. This means you can write to the underlying (supposedly non-writeable) NumPy array using the tensor. You may want to copy the array to protect its data or make it writeable before converting it to a tensor. This type of warning will be suppressed for the rest of this program. (Triggered internally at /pytorch/torch/csrc/utils/tensor_numpy.cpp:143.)\n", + " new_state[k] = pt.from_numpy(tensor_dict.pop(k)).to(device)\n", + "<ipython-input-8-5e98da9ec23c>:124: UserWarning: To copy construct from a tensor, it is recommended to use sourceTensor.clone().detach() or sourceTensor.clone().detach().requires_grad_(True), rather than torch.tensor(sourceTensor).\n", + " torch.tensor(data).to(self.device),\n", + "<ipython-input-8-5e98da9ec23c>:125: UserWarning: To copy construct from a tensor, it is recommended to use sourceTensor.clone().detach() or sourceTensor.clone().detach().requires_grad_(True), rather than torch.tensor(sourceTensor).\n", + " torch.tensor(target).to(self.device),\n" + ] + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">[17:44:17] INFO Sending metric for task aggregated_model_validation, round number 0: dice_coef 0.23560654732488817 <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\">collaborator.py</a>:365\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f74200291c0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO Collaborator one is sending task results for aggregated_model_validation, round 0 <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\">aggregator.py</a>:428\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f74200296a0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "name": "stderr", + "output_type": "stream", + "text": [ + "/home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/federated/task/runner_pt.py:154: UserWarning: To copy construct from a tensor, it is recommended to use sourceTensor.clone().detach() or sourceTensor.clone().detach().requires_grad_(True), rather than torch.tensor(sourceTensor).\n", + " data, target = pt.tensor(data).to(self.device), pt.tensor(\n" + ] + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">[17:44:37] INFO Sending metric for task train, round number 0: soft_dice_loss 0.6395286917686462 <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\">collaborator.py</a>:365\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f7420032f40>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO Collaborator one is sending task results for train, round 0 <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\">aggregator.py</a>:428\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f7420032d30>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">[17:44:39] INFO Sending metric for task locally_tuned_model_validation, round number 0: dice_coef 0.43012889739005794 <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\">collaborator.py</a>:365\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f7490cba880>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO Collaborator one is sending task results for locally_tuned_model_validation, round 0 <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\">aggregator.py</a>:428\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740bf1a7c0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO All tasks completed on one for round 0... <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\">collaborator.py</a>:148\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f7490cba760>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO Waiting for tasks... <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\">collaborator.py</a>:156\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740bf1a940>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO Sending tasks to collaborator two for round 0 <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\">aggregator.py</a>:276\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f7420029a90>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO Received the following tasks: ['aggregated_model_validation', 'train', 'locally_tuned_model_validation'] <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\">collaborator.py</a>:144\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f7490cc0a90>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">[17:44:42] INFO Sending metric for task aggregated_model_validation, round number 0: dice_coef 0.18667999943417887 <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\">collaborator.py</a>:365\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f73fb224130>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO Collaborator two is sending task results for aggregated_model_validation, round 0 <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\">aggregator.py</a>:428\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740be4c490>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">[17:45:02] INFO Sending metric for task train, round number 0: soft_dice_loss 0.6324753165245056 <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\">collaborator.py</a>:365\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f742002ceb0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO Collaborator two is sending task results for train, round 0 <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\">aggregator.py</a>:428\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f742002c970>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">[17:45:04] INFO Sending metric for task locally_tuned_model_validation, round number 0: dice_coef 0.3584949864495185 <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\">collaborator.py</a>:365\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f742002ca60>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO Collaborator two is sending task results for locally_tuned_model_validation, round 0 <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\">aggregator.py</a>:428\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740be83430>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO train task metrics... <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\">aggregator.py</a>:694\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740be83430>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO soft_dice_loss: 0.6360 <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\">aggregator.py</a>:742\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740be83430>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">[17:45:09] INFO locally_tuned_model_validation task metrics... <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\">aggregator.py</a>:694\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740bdacac0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO dice_coef: 0.3943 <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\">aggregator.py</a>:742\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740bdacbb0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO aggregated_model_validation task metrics... <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\">aggregator.py</a>:694\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740bdacc10>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO dice_coef: 0.2111 <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\">aggregator.py</a>:742\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740bdac850>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO Saved the best model with score 0.211143 <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\">aggregator.py</a>:749\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740bdac6d0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO Saving round 1 model... <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\">aggregator.py</a>:783\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740bdacac0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO Starting round 1... <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\">aggregator.py</a>:791\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740bdac850>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO All tasks completed on two for round 0... <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\">collaborator.py</a>:148\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740be4cd00>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">[17:45:10] INFO Waiting for tasks... <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\">collaborator.py</a>:156\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740be5bfa0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO Sending tasks to collaborator one for round 1 <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\">aggregator.py</a>:276\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740be87a30>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO Received the following tasks: ['aggregated_model_validation', 'train', 'locally_tuned_model_validation'] <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\">collaborator.py</a>:144\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740be87b80>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">[17:45:12] INFO Sending metric for task aggregated_model_validation, round number 1: dice_coef 0.4241240389885441 <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\">collaborator.py</a>:365\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f742000ea30>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO Collaborator one is sending task results for aggregated_model_validation, round 1 <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\">aggregator.py</a>:428\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f7420029a60>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">[17:45:32] INFO Sending metric for task train, round number 1: soft_dice_loss 0.6150628328323364 <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\">collaborator.py</a>:365\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f75299eca30>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO Collaborator one is sending task results for train, round 1 <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\">aggregator.py</a>:428\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f7490cc0be0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">[17:45:34] INFO Sending metric for task locally_tuned_model_validation, round number 1: dice_coef 0.44148768724933746 <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\">collaborator.py</a>:365\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f74200299d0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO Collaborator one is sending task results for locally_tuned_model_validation, round 1 <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\">aggregator.py</a>:428\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f74200296d0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO All tasks completed on one for round 1... <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\">collaborator.py</a>:148\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f7420029640>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO Waiting for tasks... <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\">collaborator.py</a>:156\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f7420029f40>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO Sending tasks to collaborator two for round 1 <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\">aggregator.py</a>:276\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f7420029100>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO Received the following tasks: ['aggregated_model_validation', 'train', 'locally_tuned_model_validation'] <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\">collaborator.py</a>:144\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f7420029190>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">[17:45:37] INFO Sending metric for task aggregated_model_validation, round number 1: dice_coef 0.34520076744018063 <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\">collaborator.py</a>:365\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740be839a0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO Collaborator two is sending task results for aggregated_model_validation, round 1 <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\">aggregator.py</a>:428\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740be4c880>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">[17:45:57] INFO Sending metric for task train, round number 1: soft_dice_loss 0.6058140993118286 <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\">collaborator.py</a>:365\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740be80b20>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO Collaborator two is sending task results for train, round 1 <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\">aggregator.py</a>:428\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740be80910>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">[17:45:59] INFO Sending metric for task locally_tuned_model_validation, round number 1: dice_coef 0.3571049436446159 <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\">collaborator.py</a>:365\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f7490cc05e0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO Collaborator two is sending task results for locally_tuned_model_validation, round 1 <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\">aggregator.py</a>:428\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f7490cc0550>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO train task metrics... <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\">aggregator.py</a>:694\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740be8e220>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO soft_dice_loss: 0.6104 <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\">aggregator.py</a>:742\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740be8efd0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">[17:46:05] INFO locally_tuned_model_validation task metrics... <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\">aggregator.py</a>:694\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f7490cbae80>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO dice_coef: 0.3993 <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\">aggregator.py</a>:742\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740be8ea30>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO aggregated_model_validation task metrics... <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\">aggregator.py</a>:694\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740bf159a0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO dice_coef: 0.3847 <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\">aggregator.py</a>:742\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740bf0e6d0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO Saved the best model with score 0.384662 <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\">aggregator.py</a>:749\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740bf15940>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">[17:46:06] INFO Saving round 2 model... <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\">aggregator.py</a>:783\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740bf0e6d0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO Starting round 2... <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\">aggregator.py</a>:791\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740bf2d160>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO All tasks completed on two for round 1... <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\">collaborator.py</a>:148\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f74200327f0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO Waiting for tasks... <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\">collaborator.py</a>:156\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f7420032130>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO Sending tasks to collaborator one for round 2 <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\">aggregator.py</a>:276\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f7490d3feb0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO Received the following tasks: ['aggregated_model_validation', 'train', 'locally_tuned_model_validation'] <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\">collaborator.py</a>:144\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f7420032130>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">[17:46:09] INFO Sending metric for task aggregated_model_validation, round number 2: dice_coef 0.43549662059353245 <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\">collaborator.py</a>:365\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f7490d3fac0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO Collaborator one is sending task results for aggregated_model_validation, round 2 <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\">aggregator.py</a>:428\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740be5be50>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">[17:46:29] INFO Sending metric for task train, round number 2: soft_dice_loss 0.5963471531867981 <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\">collaborator.py</a>:365\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740bf1aac0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO Collaborator one is sending task results for train, round 2 <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\">aggregator.py</a>:428\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740bf1a8e0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">[17:46:31] INFO Sending metric for task locally_tuned_model_validation, round number 2: dice_coef 0.4504963544107253 <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\">collaborator.py</a>:365\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740bffce50>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">[17:46:32] INFO Collaborator one is sending task results for locally_tuned_model_validation, round 2 <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\">aggregator.py</a>:428\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740bffc5e0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO All tasks completed on one for round 2... <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\">collaborator.py</a>:148\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740bffc670>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO Waiting for tasks... <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\">collaborator.py</a>:156\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740bffc6d0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO Sending tasks to collaborator two for round 2 <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\">aggregator.py</a>:276\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740bffccd0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO Received the following tasks: ['aggregated_model_validation', 'train', 'locally_tuned_model_validation'] <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\">collaborator.py</a>:144\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740bffc940>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">[17:46:34] INFO Sending metric for task aggregated_model_validation, round number 2: dice_coef 0.3556097505554076 <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\">collaborator.py</a>:365\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740bdacc40>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO Collaborator two is sending task results for aggregated_model_validation, round 2 <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\">aggregator.py</a>:428\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f7490cbaa90>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">[17:46:55] INFO Sending metric for task train, round number 2: soft_dice_loss 0.5920241475105286 <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\">collaborator.py</a>:365\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740bea9d90>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO Collaborator two is sending task results for train, round 2 <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\">aggregator.py</a>:428\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740bea90d0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">[17:46:57] INFO Sending metric for task locally_tuned_model_validation, round number 2: dice_coef 0.37100199874370327 <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\">collaborator.py</a>:365\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f7490d3feb0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO Collaborator two is sending task results for locally_tuned_model_validation, round 2 <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\">aggregator.py</a>:428\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740bf0e910>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO train task metrics... <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\">aggregator.py</a>:694\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740bf0e220>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO soft_dice_loss: 0.5942 <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\">aggregator.py</a>:742\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740bf0e400>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">[17:47:03] INFO locally_tuned_model_validation task metrics... <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\">aggregator.py</a>:694\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f742002c5e0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO dice_coef: 0.4107 <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\">aggregator.py</a>:742\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f7490d3f7c0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO aggregated_model_validation task metrics... <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\">aggregator.py</a>:694\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740bd1b9a0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO dice_coef: 0.3956 <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\">aggregator.py</a>:742\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740bf15ca0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO Saved the best model with score 0.395553 <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\">aggregator.py</a>:749\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740bd1b820>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">[17:47:04] INFO Saving round 3 model... <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\">aggregator.py</a>:783\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f742002cc70>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO Starting round 3... <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\">aggregator.py</a>:791\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740bfedca0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO All tasks completed on two for round 2... <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\">collaborator.py</a>:148\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740bfd9e50>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO Waiting for tasks... <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\">collaborator.py</a>:156\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740bf94eb0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO Sending tasks to collaborator one for round 3 <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\">aggregator.py</a>:276\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740bf94310>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO Received the following tasks: ['aggregated_model_validation', 'train', 'locally_tuned_model_validation'] <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\">collaborator.py</a>:144\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740bf94520>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">[17:47:06] INFO Sending metric for task aggregated_model_validation, round number 3: dice_coef 0.43500437275055914 <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\">collaborator.py</a>:365\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f7490cbaa60>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO Collaborator one is sending task results for aggregated_model_validation, round 3 <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\">aggregator.py</a>:428\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740bf2f8e0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">[17:47:27] INFO Sending metric for task train, round number 3: soft_dice_loss 0.5834751725196838 <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\">collaborator.py</a>:365\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f748dc86a30>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO Collaborator one is sending task results for train, round 3 <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\">aggregator.py</a>:428\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f742003f6d0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">[17:47:30] INFO Sending metric for task locally_tuned_model_validation, round number 3: dice_coef 0.45610655507733744 <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\">collaborator.py</a>:365\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740bf146d0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO Collaborator one is sending task results for locally_tuned_model_validation, round 3 <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\">aggregator.py</a>:428\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740bf14580>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO All tasks completed on one for round 3... <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\">collaborator.py</a>:148\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740bf14b80>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO Waiting for tasks... <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\">collaborator.py</a>:156\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740bf14940>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO Sending tasks to collaborator two for round 3 <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\">aggregator.py</a>:276\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740bf14e20>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO Received the following tasks: ['aggregated_model_validation', 'train', 'locally_tuned_model_validation'] <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\">collaborator.py</a>:144\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740bf14b20>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">[17:47:32] INFO Sending metric for task aggregated_model_validation, round number 3: dice_coef 0.3653699919100731 <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\">collaborator.py</a>:365\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740bf94670>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO Collaborator two is sending task results for aggregated_model_validation, round 3 <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\">aggregator.py</a>:428\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740be8edf0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">[17:47:53] INFO Sending metric for task train, round number 3: soft_dice_loss 0.5735182762145996 <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\">collaborator.py</a>:365\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740bf76e80>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO Collaborator two is sending task results for train, round 3 <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\">aggregator.py</a>:428\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740bf76820>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">[17:47:56] INFO Sending metric for task locally_tuned_model_validation, round number 3: dice_coef 0.3892689275164758 <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\">collaborator.py</a>:365\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740be80340>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO Collaborator two is sending task results for locally_tuned_model_validation, round 3 <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\">aggregator.py</a>:428\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740be80190>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO train task metrics... <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\">aggregator.py</a>:694\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740be800d0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO soft_dice_loss: 0.5785 <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\">aggregator.py</a>:742\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740be801c0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">[17:48:02] INFO locally_tuned_model_validation task metrics... <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\">aggregator.py</a>:694\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740bfff7f0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO dice_coef: 0.4227 <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\">aggregator.py</a>:742\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740bfff730>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO aggregated_model_validation task metrics... <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\">aggregator.py</a>:694\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740bf2d340>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO dice_coef: 0.4002 <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\">aggregator.py</a>:742\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740bf2d6d0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO Saved the best model with score 0.400187 <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\">aggregator.py</a>:749\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740bfff0a0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO Saving round 4 model... <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\">aggregator.py</a>:783\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f7420029640>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">[17:48:03] INFO Starting round 4... <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\">aggregator.py</a>:791\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740be80580>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO All tasks completed on two for round 3... <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\">collaborator.py</a>:148\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740bf060d0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO Waiting for tasks... <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\">collaborator.py</a>:156\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f7490c9e9a0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO Sending tasks to collaborator one for round 4 <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\">aggregator.py</a>:276\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740bf060d0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO Received the following tasks: ['aggregated_model_validation', 'train', 'locally_tuned_model_validation'] <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\">collaborator.py</a>:144\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f7490c9e9a0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">[17:48:05] INFO Sending metric for task aggregated_model_validation, round number 4: dice_coef 0.4355728107113992 <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\">collaborator.py</a>:365\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f742003fa00>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO Collaborator one is sending task results for aggregated_model_validation, round 4 <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\">aggregator.py</a>:428\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f742003f730>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">[17:48:26] INFO Sending metric for task train, round number 4: soft_dice_loss 0.5637962222099304 <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\">collaborator.py</a>:365\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740bf76a60>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO Collaborator one is sending task results for train, round 4 <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\">aggregator.py</a>:428\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740bdb93a0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">[17:48:29] INFO Sending metric for task locally_tuned_model_validation, round number 4: dice_coef 0.46870911505914503 <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\">collaborator.py</a>:365\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740bf06730>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO Collaborator one is sending task results for locally_tuned_model_validation, round 4 <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\">aggregator.py</a>:428\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f7420015730>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO All tasks completed on one for round 4... <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\">collaborator.py</a>:148\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740bf06730>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO Waiting for tasks... <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\">collaborator.py</a>:156\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740bf15f70>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO Sending tasks to collaborator two for round 4 <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\">aggregator.py</a>:276\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740bfffc70>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO Received the following tasks: ['aggregated_model_validation', 'train', 'locally_tuned_model_validation'] <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\">collaborator.py</a>:144\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740bf154f0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">[17:48:31] INFO Sending metric for task aggregated_model_validation, round number 4: dice_coef 0.37851148987970046 <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\">collaborator.py</a>:365\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740be2fb50>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO Collaborator two is sending task results for aggregated_model_validation, round 4 <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\">aggregator.py</a>:428\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740be2fbb0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">[17:48:52] INFO Sending metric for task train, round number 4: soft_dice_loss 0.5667974948883057 <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\">collaborator.py</a>:365\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f742002cd60>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO Collaborator two is sending task results for train, round 4 <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\">aggregator.py</a>:428\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740bdb93a0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">[17:48:55] INFO Sending metric for task locally_tuned_model_validation, round number 4: dice_coef 0.3846700369350372 <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\">collaborator.py</a>:365\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740be2f4f0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO Collaborator two is sending task results for locally_tuned_model_validation, round 4 <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\">aggregator.py</a>:428\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740bf624c0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO train task metrics... <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\">aggregator.py</a>:694\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740bfe3190>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO soft_dice_loss: 0.5653 <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\">aggregator.py</a>:742\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f742003f400>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">[17:49:01] INFO locally_tuned_model_validation task metrics... <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\">aggregator.py</a>:694\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740bd5c610>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO dice_coef: 0.4267 <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\">aggregator.py</a>:742\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740bf14970>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO aggregated_model_validation task metrics... <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\">aggregator.py</a>:694\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740bf144f0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO dice_coef: 0.4070 <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\">aggregator.py</a>:742\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740bd2e190>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO Saved the best model with score 0.407042 <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\">aggregator.py</a>:749\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740bf2ad00>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO Saving round 5 model... <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\">aggregator.py</a>:783\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740bd94970>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">[17:49:02] INFO Starting round 5... <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\">aggregator.py</a>:791\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740bf15640>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO All tasks completed on two for round 4... <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\">collaborator.py</a>:148\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740bd9e640>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO Waiting for tasks... <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\">collaborator.py</a>:156\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f7490d3fc70>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO Sending tasks to collaborator one for round 5 <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\">aggregator.py</a>:276\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740bd9e640>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO Received the following tasks: ['aggregated_model_validation', 'train', 'locally_tuned_model_validation'] <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\">collaborator.py</a>:144\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f742002c6d0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">[17:49:04] INFO Sending metric for task aggregated_model_validation, round number 5: dice_coef 0.4695041198884287 <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\">collaborator.py</a>:365\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740bf1ab20>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO Collaborator one is sending task results for aggregated_model_validation, round 5 <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\">aggregator.py</a>:428\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f748dc86040>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">[17:49:25] INFO Sending metric for task train, round number 5: soft_dice_loss 0.5579285025596619 <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\">collaborator.py</a>:365\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740be8ebb0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO Collaborator one is sending task results for train, round 5 <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\">aggregator.py</a>:428\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740be8ebb0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">[17:49:28] INFO Sending metric for task locally_tuned_model_validation, round number 5: dice_coef 0.477940563232668 <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\">collaborator.py</a>:365\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740be8e310>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO Collaborator one is sending task results for locally_tuned_model_validation, round 5 <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\">aggregator.py</a>:428\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740bf15910>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO All tasks completed on one for round 5... <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\">collaborator.py</a>:148\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740be8e280>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO Waiting for tasks... <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\">collaborator.py</a>:156\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740be2f8b0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO Sending tasks to collaborator two for round 5 <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\">aggregator.py</a>:276\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740bf154f0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO Received the following tasks: ['aggregated_model_validation', 'train', 'locally_tuned_model_validation'] <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\">collaborator.py</a>:144\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740be8e310>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">[17:49:30] INFO Sending metric for task aggregated_model_validation, round number 5: dice_coef 0.3998434495541357 <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\">collaborator.py</a>:365\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740be8e280>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO Collaborator two is sending task results for aggregated_model_validation, round 5 <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\">aggregator.py</a>:428\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f7420029be0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">[17:49:52] INFO Sending metric for task train, round number 5: soft_dice_loss 0.5507904291152954 <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\">collaborator.py</a>:365\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f742003f9a0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO Collaborator two is sending task results for train, round 5 <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\">aggregator.py</a>:428\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740be2fd00>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">[17:49:54] INFO Sending metric for task locally_tuned_model_validation, round number 5: dice_coef 0.4104092457602101 <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\">collaborator.py</a>:365\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740bd5ef40>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO Collaborator two is sending task results for locally_tuned_model_validation, round 5 <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\">aggregator.py</a>:428\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740be1ce20>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO train task metrics... <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\">aggregator.py</a>:694\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740bfe8ca0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO soft_dice_loss: 0.5544 <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\">aggregator.py</a>:742\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f7420015b20>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">[17:50:00] INFO locally_tuned_model_validation task metrics... <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\">aggregator.py</a>:694\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740be36b20>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO dice_coef: 0.4442 <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\">aggregator.py</a>:742\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740bcf59a0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO aggregated_model_validation task metrics... <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\">aggregator.py</a>:694\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740bcf59a0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO dice_coef: 0.4347 <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\">aggregator.py</a>:742\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740bcf5dc0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO Saved the best model with score 0.434674 <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\">aggregator.py</a>:749\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740be369a0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">[17:50:01] INFO Saving round 6 model... <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\">aggregator.py</a>:783\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740bcf59a0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO Starting round 6... <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\">aggregator.py</a>:791\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740bd94be0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO All tasks completed on two for round 5... <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\">collaborator.py</a>:148\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740bfff3a0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO Waiting for tasks... <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\">collaborator.py</a>:156\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f748dc865b0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO Sending tasks to collaborator one for round 6 <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\">aggregator.py</a>:276\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740bfff3a0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO Received the following tasks: ['aggregated_model_validation', 'train', 'locally_tuned_model_validation'] <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\">collaborator.py</a>:144\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740bfff3a0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">[17:50:04] INFO Sending metric for task aggregated_model_validation, round number 6: dice_coef 0.4847551199697679 <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\">collaborator.py</a>:365\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f742002cf40>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO Collaborator one is sending task results for aggregated_model_validation, round 6 <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\">aggregator.py</a>:428\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f742002c160>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">[17:50:25] INFO Sending metric for task train, round number 6: soft_dice_loss 0.549356997013092 <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\">collaborator.py</a>:365\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740bcf5d60>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO Collaborator one is sending task results for train, round 6 <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\">aggregator.py</a>:428\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740bcf5220>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">[17:50:28] INFO Sending metric for task locally_tuned_model_validation, round number 6: dice_coef 0.49405526922595117 <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\">collaborator.py</a>:365\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740bfe3250>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO Collaborator one is sending task results for locally_tuned_model_validation, round 6 <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\">aggregator.py</a>:428\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740bfe3670>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO All tasks completed on one for round 6... <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\">collaborator.py</a>:148\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740bfe3b50>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO Waiting for tasks... <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\">collaborator.py</a>:156\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740bfe3820>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO Sending tasks to collaborator two for round 6 <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\">aggregator.py</a>:276\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740bfe3310>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO Received the following tasks: ['aggregated_model_validation', 'train', 'locally_tuned_model_validation'] <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\">collaborator.py</a>:144\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740bfe3580>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">[17:50:30] INFO Sending metric for task aggregated_model_validation, round number 6: dice_coef 0.4081264974609498 <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\">collaborator.py</a>:365\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740bea51c0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO Collaborator two is sending task results for aggregated_model_validation, round 6 <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\">aggregator.py</a>:428\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740bea5730>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">[17:50:52] INFO Sending metric for task train, round number 6: soft_dice_loss 0.5499082803726196 <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\">collaborator.py</a>:365\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f73fb224880>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO Collaborator two is sending task results for train, round 6 <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\">aggregator.py</a>:428\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f73fb224790>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">[17:50:55] INFO Sending metric for task locally_tuned_model_validation, round number 6: dice_coef 0.4086666549405744 <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\">collaborator.py</a>:365\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740bf06ca0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO Collaborator two is sending task results for locally_tuned_model_validation, round 6 <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\">aggregator.py</a>:428\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740bf2f070>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO train task metrics... <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\">aggregator.py</a>:694\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f73fb224d90>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO soft_dice_loss: 0.5496 <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\">aggregator.py</a>:742\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740bf621c0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">[17:51:00] INFO locally_tuned_model_validation task metrics... <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\">aggregator.py</a>:694\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740be872b0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO dice_coef: 0.4514 <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\">aggregator.py</a>:742\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740be361c0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO aggregated_model_validation task metrics... <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\">aggregator.py</a>:694\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740bea5220>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO dice_coef: 0.4464 <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\">aggregator.py</a>:742\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740ac917c0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO Saved the best model with score 0.446441 <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\">aggregator.py</a>:749\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740bea5220>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">[17:51:01] INFO Saving round 7 model... <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\">aggregator.py</a>:783\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740be872b0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO Starting round 7... <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\">aggregator.py</a>:791\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740bf62970>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO All tasks completed on two for round 6... <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\">collaborator.py</a>:148\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740bea9f70>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">[17:51:02] INFO Waiting for tasks... <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\">collaborator.py</a>:156\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740bfe3520>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO Sending tasks to collaborator one for round 7 <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\">aggregator.py</a>:276\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740bffcc10>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO Received the following tasks: ['aggregated_model_validation', 'train', 'locally_tuned_model_validation'] <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\">collaborator.py</a>:144\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740bfe3d90>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">[17:51:04] INFO Sending metric for task aggregated_model_validation, round number 7: dice_coef 0.48679458902728173 <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\">collaborator.py</a>:365\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f7420032a00>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO Collaborator one is sending task results for aggregated_model_validation, round 7 <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\">aggregator.py</a>:428\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740be36b20>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">[17:51:25] INFO Sending metric for task train, round number 7: soft_dice_loss 0.5552520751953125 <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\">collaborator.py</a>:365\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f7490c9e5b0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO Collaborator one is sending task results for train, round 7 <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\">aggregator.py</a>:428\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f748dc86d30>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">[17:51:28] INFO Sending metric for task locally_tuned_model_validation, round number 7: dice_coef 0.4828214414658085 <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\">collaborator.py</a>:365\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740be83160>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO Collaborator one is sending task results for locally_tuned_model_validation, round 7 <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\">aggregator.py</a>:428\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f742000e160>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO All tasks completed on one for round 7... <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\">collaborator.py</a>:148\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f742000ee80>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO Waiting for tasks... <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\">collaborator.py</a>:156\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f742000eeb0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO Sending tasks to collaborator two for round 7 <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\">aggregator.py</a>:276\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f748dc854f0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO Received the following tasks: ['aggregated_model_validation', 'train', 'locally_tuned_model_validation'] <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\">collaborator.py</a>:144\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f742000e280>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">[17:51:31] INFO Sending metric for task aggregated_model_validation, round number 7: dice_coef 0.41096355165204695 <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\">collaborator.py</a>:365\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740be8e730>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO Collaborator two is sending task results for aggregated_model_validation, round 7 <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\">aggregator.py</a>:428\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740be8e730>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">[17:51:52] INFO Sending metric for task train, round number 7: soft_dice_loss 0.5383422374725342 <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\">collaborator.py</a>:365\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740bfffbb0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO Collaborator two is sending task results for train, round 7 <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\">aggregator.py</a>:428\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740bd941c0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">[17:51:55] INFO Sending metric for task locally_tuned_model_validation, round number 7: dice_coef 0.4175524327062791 <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\">collaborator.py</a>:365\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740be87dc0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO Collaborator two is sending task results for locally_tuned_model_validation, round 7 <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\">aggregator.py</a>:428\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740be87dc0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO train task metrics... <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\">aggregator.py</a>:694\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740bea5ca0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO soft_dice_loss: 0.5468 <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\">aggregator.py</a>:742\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740bf06d30>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">[17:52:01] INFO locally_tuned_model_validation task metrics... <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\">aggregator.py</a>:694\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f7490cc0be0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO dice_coef: 0.4502 <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\">aggregator.py</a>:742\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f7490cc0be0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO aggregated_model_validation task metrics... <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\">aggregator.py</a>:694\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740bcdcca0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO dice_coef: 0.4489 <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\">aggregator.py</a>:742\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740bcdcdc0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO Saved the best model with score 0.448879 <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\">aggregator.py</a>:749\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f748dc85d30>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">[17:52:02] INFO Saving round 8 model... <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\">aggregator.py</a>:783\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740bfd5580>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO Starting round 8... <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\">aggregator.py</a>:791\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740be8e730>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO All tasks completed on two for round 7... <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\">collaborator.py</a>:148\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f73fb224580>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO Waiting for tasks... <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\">collaborator.py</a>:156\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f7490d3f3d0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO Sending tasks to collaborator one for round 8 <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\">aggregator.py</a>:276\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740bd5c340>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO Received the following tasks: ['aggregated_model_validation', 'train', 'locally_tuned_model_validation'] <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\">collaborator.py</a>:144\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f73fb224580>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">[17:52:05] INFO Sending metric for task aggregated_model_validation, round number 8: dice_coef 0.49110506234630463 <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\">collaborator.py</a>:365\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740bd5c340>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO Collaborator one is sending task results for aggregated_model_validation, round 8 <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\">aggregator.py</a>:428\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740be36490>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">[17:52:26] INFO Sending metric for task train, round number 8: soft_dice_loss 0.5618945360183716 <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\">collaborator.py</a>:365\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740b4ba550>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO Collaborator one is sending task results for train, round 8 <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\">aggregator.py</a>:428\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740b4ba220>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">[17:52:29] INFO Sending metric for task locally_tuned_model_validation, round number 8: dice_coef 0.46602303943326395 <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\">collaborator.py</a>:365\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740bd9e130>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO Collaborator one is sending task results for locally_tuned_model_validation, round 8 <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\">aggregator.py</a>:428\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740bd9e070>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO All tasks completed on one for round 8... <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\">collaborator.py</a>:148\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740bd9ef70>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO Waiting for tasks... <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\">collaborator.py</a>:156\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740bd9eee0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO Sending tasks to collaborator two for round 8 <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\">aggregator.py</a>:276\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740bd9e6a0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO Received the following tasks: ['aggregated_model_validation', 'train', 'locally_tuned_model_validation'] <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\">collaborator.py</a>:144\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740bd9ef70>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">[17:52:32] INFO Sending metric for task aggregated_model_validation, round number 8: dice_coef 0.41813477200846516 <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\">collaborator.py</a>:365\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740bf2fee0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO Collaborator two is sending task results for aggregated_model_validation, round 8 <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\">aggregator.py</a>:428\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740bf2fd30>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">[17:52:54] INFO Sending metric for task train, round number 8: soft_dice_loss 0.534250020980835 <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\">collaborator.py</a>:365\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f7490c9ecd0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO Collaborator two is sending task results for train, round 8 <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\">aggregator.py</a>:428\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740bf06070>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">[17:52:56] INFO Sending metric for task locally_tuned_model_validation, round number 8: dice_coef 0.41231230381996403 <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\">collaborator.py</a>:365\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740bf62520>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO Collaborator two is sending task results for locally_tuned_model_validation, round 8 <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\">aggregator.py</a>:428\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740bf62a00>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO train task metrics... <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\">aggregator.py</a>:694\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740bf62070>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO soft_dice_loss: 0.5481 <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\">aggregator.py</a>:742\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740bf62520>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">[17:53:02] INFO locally_tuned_model_validation task metrics... <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\">aggregator.py</a>:694\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740bef3ee0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO dice_coef: 0.4392 <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\">aggregator.py</a>:742\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740bee4c10>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO aggregated_model_validation task metrics... <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\">aggregator.py</a>:694\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740bef3550>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO dice_coef: 0.4546 <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\">aggregator.py</a>:742\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740be87610>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO Saved the best model with score 0.454620 <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\">aggregator.py</a>:749\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740bef3b20>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">[17:53:03] INFO Saving round 9 model... <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\">aggregator.py</a>:783\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740bee4c10>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO Starting round 9... <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\">aggregator.py</a>:791\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740ad616a0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO All tasks completed on two for round 8... <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\">collaborator.py</a>:148\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f742003f400>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO Waiting for tasks... <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\">collaborator.py</a>:156\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740bdb9fd0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO Sending tasks to collaborator one for round 9 <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\">aggregator.py</a>:276\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f748dc85e20>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO Received the following tasks: ['aggregated_model_validation', 'train', 'locally_tuned_model_validation'] <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\">collaborator.py</a>:144\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740bf14280>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">[17:53:06] INFO Sending metric for task aggregated_model_validation, round number 9: dice_coef 0.47473019169222924 <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\">collaborator.py</a>:365\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f73fb224e20>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO Collaborator one is sending task results for aggregated_model_validation, round 9 <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\">aggregator.py</a>:428\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740be5b160>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">[17:53:28] INFO Sending metric for task train, round number 9: soft_dice_loss 0.5441964268684387 <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\">collaborator.py</a>:365\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740bf15280>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO Collaborator one is sending task results for train, round 9 <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\">aggregator.py</a>:428\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740bd5c940>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">[17:53:30] INFO Sending metric for task locally_tuned_model_validation, round number 9: dice_coef 0.49820235467726187 <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\">collaborator.py</a>:365\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740bf76cd0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO Collaborator one is sending task results for locally_tuned_model_validation, round 9 <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\">aggregator.py</a>:428\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740bdac8b0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO All tasks completed on one for round 9... <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\">collaborator.py</a>:148\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740b5375b0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO Waiting for tasks... <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\">collaborator.py</a>:156\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740bcbb130>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO Sending tasks to collaborator two for round 9 <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\">aggregator.py</a>:276\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740bcbb730>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO Received the following tasks: ['aggregated_model_validation', 'train', 'locally_tuned_model_validation'] <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\">collaborator.py</a>:144\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740bcbb730>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">[17:53:33] INFO Sending metric for task aggregated_model_validation, round number 9: dice_coef 0.3963722812552606 <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\">collaborator.py</a>:365\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740bea5670>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO Collaborator two is sending task results for aggregated_model_validation, round 9 <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\">aggregator.py</a>:428\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740b4bddc0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">[17:53:55] INFO Sending metric for task train, round number 9: soft_dice_loss 0.5431901216506958 <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\">collaborator.py</a>:365\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740bf14c10>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO Collaborator two is sending task results for train, round 9 <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\">aggregator.py</a>:428\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f748dc85100>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">[17:53:58] INFO Sending metric for task locally_tuned_model_validation, round number 9: dice_coef 0.4178659223741101 <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\">collaborator.py</a>:365\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740ad61d30>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO Collaborator two is sending task results for locally_tuned_model_validation, round 9 <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\">aggregator.py</a>:428\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740bee4a30>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO train task metrics... <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\">aggregator.py</a>:694\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740bdc6040>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO soft_dice_loss: 0.5437 <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\">aggregator.py</a>:742\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740ad61550>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">[17:54:04] INFO locally_tuned_model_validation task metrics... <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\">aggregator.py</a>:694\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740acbfe20>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO dice_coef: 0.4580 <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\">aggregator.py</a>:742\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740bef3eb0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO aggregated_model_validation task metrics... <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\">aggregator.py</a>:694\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740b501040>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO dice_coef: 0.4356 <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\">aggregator.py</a>:742\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740bef3a30>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO Saving round 10 model... <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\">aggregator.py</a>:783\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740bef3c70>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO Starting round 10... <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\">aggregator.py</a>:791\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740bea5670>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO All tasks completed on two for round 9... <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\">collaborator.py</a>:148\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740bea26d0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO Waiting for tasks... <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\">collaborator.py</a>:156\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740bea26d0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO Sending tasks to collaborator one for round 10 <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\">aggregator.py</a>:276\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740be4c4c0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO Received the following tasks: ['aggregated_model_validation', 'train', 'locally_tuned_model_validation'] <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\">collaborator.py</a>:144\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740bf14c10>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">[17:54:07] INFO Sending metric for task aggregated_model_validation, round number 10: dice_coef 0.49992890127243533 <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\">collaborator.py</a>:365\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740bcbbbb0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO Collaborator one is sending task results for aggregated_model_validation, round 10 <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\">aggregator.py</a>:428\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740be87730>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">[17:54:29] INFO Sending metric for task train, round number 10: soft_dice_loss 0.5380003452301025 <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\">collaborator.py</a>:365\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740be87d30>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO Collaborator one is sending task results for train, round 10 <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\">aggregator.py</a>:428\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740bea9bb0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">[17:54:32] INFO Sending metric for task locally_tuned_model_validation, round number 10: dice_coef 0.46978584989424677 <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\">collaborator.py</a>:365\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740bf62070>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO Collaborator one is sending task results for locally_tuned_model_validation, round 10 <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\">aggregator.py</a>:428\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740bee4ee0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO All tasks completed on one for round 10... <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\">collaborator.py</a>:148\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740bee4250>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO Waiting for tasks... <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\">collaborator.py</a>:156\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740bee4580>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO Sending tasks to collaborator two for round 10 <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\">aggregator.py</a>:276\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740bee4a30>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO Received the following tasks: ['aggregated_model_validation', 'train', 'locally_tuned_model_validation'] <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\">collaborator.py</a>:144\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740bf62490>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">[17:54:35] INFO Sending metric for task aggregated_model_validation, round number 10: dice_coef 0.42785738312428995 <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\">collaborator.py</a>:365\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740bf76340>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO Collaborator two is sending task results for aggregated_model_validation, round 10 <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\">aggregator.py</a>:428\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f740acbffd0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "ename": "KeyboardInterrupt", + "evalue": "", + "output_type": "error", + "traceback": [ + "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m", + "\u001b[0;31mKeyboardInterrupt\u001b[0m Traceback (most recent call last)", + "\u001b[0;32m<ipython-input-13-92562d71b5a6>\u001b[0m in \u001b[0;36m<module>\u001b[0;34m\u001b[0m\n\u001b[1;32m 1\u001b[0m \u001b[0;31m# Run experiment, return trained FederatedModel\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m----> 2\u001b[0;31m final_fl_model = fx.run_experiment(\n\u001b[0m\u001b[1;32m 3\u001b[0m collaborators, override_config={'aggregator.settings.rounds_to_train': 30})\n", + "\u001b[0;32m~/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/native/native.py\u001b[0m in \u001b[0;36mrun_experiment\u001b[0;34m(collaborator_dict, override_config)\u001b[0m\n\u001b[1;32m 262\u001b[0m \u001b[0mcollaborator\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mcollaborators\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0mcol\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 263\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 264\u001b[0;31m \u001b[0mcollaborator\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mrun_simulation\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m 265\u001b[0m \u001b[0mmodel_states\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0mcol\u001b[0m\u001b[0;34m]\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mcollaborator\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mtask_runner\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mget_tensor_dict\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mwith_opt_vars\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0;32mTrue\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 266\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n", + "\u001b[0;32m~/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\u001b[0m in \u001b[0;36mrun_simulation\u001b[0;34m(self)\u001b[0m\n\u001b[1;32m 145\u001b[0m 'Received the following tasks: {}'.format(tasks))\n\u001b[1;32m 146\u001b[0m \u001b[0;32mfor\u001b[0m \u001b[0mtask\u001b[0m \u001b[0;32min\u001b[0m \u001b[0mtasks\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 147\u001b[0;31m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mdo_task\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mtask\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mround_number\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m 148\u001b[0m self.logger.info(\n\u001b[1;32m 149\u001b[0m 'All tasks completed on {} for round {}...'.format(\n", + "\u001b[0;32m~/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\u001b[0m in \u001b[0;36mdo_task\u001b[0;34m(self, task, round_number)\u001b[0m\n\u001b[1;32m 196\u001b[0m \u001b[0;31m# now we have whatever the model needs to do the task\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 197\u001b[0m \u001b[0mfunc\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mgetattr\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mtask_runner\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mfunc_name\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 198\u001b[0;31m global_output_tensor_dict, local_output_tensor_dict = func(\n\u001b[0m\u001b[1;32m 199\u001b[0m \u001b[0mcol_name\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mcollaborator_name\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 200\u001b[0m \u001b[0mround_num\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mround_number\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n", + "\u001b[0;32m~/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/federated/task/runner_pt.py\u001b[0m in \u001b[0;36mtrain_batches\u001b[0;34m(self, col_name, round_num, input_tensor_dict, num_batches, use_tqdm, **kwargs)\u001b[0m\n\u001b[1;32m 159\u001b[0m \u001b[0mloss\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mbackward\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 160\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0moptimizer\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mstep\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 161\u001b[0;31m \u001b[0mlosses\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mappend\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mloss\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mdetach\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mcpu\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mnumpy\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m 162\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 163\u001b[0m \u001b[0;31m# Output metric tensors (scalar)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n", + "\u001b[0;31mKeyboardInterrupt\u001b[0m: " + ] + } + ], "source": [ "# Run experiment, return trained FederatedModel\n", "final_fl_model = fx.run_experiment(\n", @@ -550,7 +7077,7 @@ "name": "python", "nbconvert_exporter": "python", "pygments_lexer": "ipython3", - "version": "3.6.12" + "version": "3.8.8" } }, "nbformat": 4, diff --git a/openfl-tutorials/Federated_Pytorch_MNIST_Tutorial.ipynb b/openfl-tutorials/Federated_Pytorch_MNIST_Tutorial.ipynb index d42ef87e5f..8189afe719 100644 --- a/openfl-tutorials/Federated_Pytorch_MNIST_Tutorial.ipynb +++ b/openfl-tutorials/Federated_Pytorch_MNIST_Tutorial.ipynb @@ -9,17 +9,128 @@ }, { "cell_type": "code", - "execution_count": null, + "execution_count": 1, "metadata": {}, - "outputs": [], + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "Processing /home/itrushkin/repos/openfl-fork\n", + "Requirement already satisfied: torch in /home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages (1.8.1)\n", + "Requirement already satisfied: torchvision in /home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages (0.9.1)\n", + "Requirement already satisfied: Click>=7.0 in /home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages (from openfl==1.0.1) (7.1.2)\n", + "Requirement already satisfied: PyYAML>=5.4.1 in /home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages (from openfl==1.0.1) (5.4.1)\n", + "Requirement already satisfied: numpy in /home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages (from openfl==1.0.1) (1.18.5)\n", + "Requirement already satisfied: pandas in /home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages (from openfl==1.0.1) (1.2.3)\n", + "Requirement already satisfied: protobuf in /home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages (from openfl==1.0.1) (3.15.6)\n", + "Requirement already satisfied: grpcio==1.30.0 in /home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages (from openfl==1.0.1) (1.30.0)\n", + "Requirement already satisfied: grpcio-tools==1.30.0 in /home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages (from openfl==1.0.1) (1.30.0)\n", + "Requirement already satisfied: rich==9.1.0 in /home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages (from openfl==1.0.1) (9.1.0)\n", + "Requirement already satisfied: tqdm in /home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages (from openfl==1.0.1) (4.59.0)\n", + "Requirement already satisfied: scikit-learn in /home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages (from openfl==1.0.1) (0.24.1)\n", + "Requirement already satisfied: docker in /home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages (from openfl==1.0.1) (4.4.4)\n", + "Requirement already satisfied: jupyter in /home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages (from openfl==1.0.1) (1.0.0)\n", + "Requirement already satisfied: ipykernel in /home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages (from openfl==1.0.1) (5.5.0)\n", + "Requirement already satisfied: flatten_json in /home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages (from openfl==1.0.1) (0.1.13)\n", + "Requirement already satisfied: cryptography>=3.4.6 in /home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages (from openfl==1.0.1) (3.4.6)\n", + "Requirement already satisfied: six>=1.5.2 in /home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages (from grpcio==1.30.0->openfl==1.0.1) (1.15.0)\n", + "Requirement already satisfied: typing-extensions<4.0.0,>=3.7.4 in /home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages (from rich==9.1.0->openfl==1.0.1) (3.7.4.3)\n", + "Requirement already satisfied: pygments<3.0.0,>=2.6.0 in /home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages (from rich==9.1.0->openfl==1.0.1) (2.8.1)\n", + "Requirement already satisfied: colorama<0.5.0,>=0.4.0 in /home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages (from rich==9.1.0->openfl==1.0.1) (0.4.4)\n", + "Requirement already satisfied: commonmark<0.10.0,>=0.9.0 in /home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages (from rich==9.1.0->openfl==1.0.1) (0.9.1)\n", + "Requirement already satisfied: cffi>=1.12 in /home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages (from cryptography>=3.4.6->openfl==1.0.1) (1.14.5)\n", + "Requirement already satisfied: pycparser in /home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages (from cffi>=1.12->cryptography>=3.4.6->openfl==1.0.1) (2.20)\n", + "Requirement already satisfied: pillow>=4.1.1 in /home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages (from torchvision) (8.1.2)\n", + "Requirement already satisfied: websocket-client>=0.32.0 in /home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages (from docker->openfl==1.0.1) (0.58.0)\n", + "Requirement already satisfied: requests!=2.18.0,>=2.14.2 in /home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages (from docker->openfl==1.0.1) (2.25.1)\n", + "Requirement already satisfied: chardet<5,>=3.0.2 in /home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages (from requests!=2.18.0,>=2.14.2->docker->openfl==1.0.1) (4.0.0)\n", + "Requirement already satisfied: certifi>=2017.4.17 in /home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages (from requests!=2.18.0,>=2.14.2->docker->openfl==1.0.1) (2020.12.5)\n", + "Requirement already satisfied: idna<3,>=2.5 in /home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages (from requests!=2.18.0,>=2.14.2->docker->openfl==1.0.1) (2.10)\n", + "Requirement already satisfied: urllib3<1.27,>=1.21.1 in /home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages (from requests!=2.18.0,>=2.14.2->docker->openfl==1.0.1) (1.26.4)\n", + "Requirement already satisfied: tornado>=4.2 in /home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages (from ipykernel->openfl==1.0.1) (6.1)\n", + "Requirement already satisfied: jupyter-client in /home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages (from ipykernel->openfl==1.0.1) (6.1.12)\n", + "Requirement already satisfied: traitlets>=4.1.0 in /home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages (from ipykernel->openfl==1.0.1) (5.0.5)\n", + "Requirement already satisfied: ipython>=5.0.0 in /home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages (from ipykernel->openfl==1.0.1) (7.21.0)\n", + "Requirement already satisfied: decorator in /home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages (from ipython>=5.0.0->ipykernel->openfl==1.0.1) (4.4.2)\n", + "Requirement already satisfied: setuptools>=18.5 in /home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages (from ipython>=5.0.0->ipykernel->openfl==1.0.1) (54.1.2)\n", + "Requirement already satisfied: jedi>=0.16 in /home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages (from ipython>=5.0.0->ipykernel->openfl==1.0.1) (0.18.0)\n", + "Requirement already satisfied: backcall in /home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages (from ipython>=5.0.0->ipykernel->openfl==1.0.1) (0.2.0)\n", + "Requirement already satisfied: pickleshare in /home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages (from ipython>=5.0.0->ipykernel->openfl==1.0.1) (0.7.5)\n", + "Requirement already satisfied: pexpect>4.3 in /home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages (from ipython>=5.0.0->ipykernel->openfl==1.0.1) (4.8.0)\n", + "Requirement already satisfied: prompt-toolkit!=3.0.0,!=3.0.1,<3.1.0,>=2.0.0 in /home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages (from ipython>=5.0.0->ipykernel->openfl==1.0.1) (3.0.18)\n", + "Requirement already satisfied: parso<0.9.0,>=0.8.0 in /home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages (from jedi>=0.16->ipython>=5.0.0->ipykernel->openfl==1.0.1) (0.8.1)\n", + "Requirement already satisfied: ptyprocess>=0.5 in /home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages (from pexpect>4.3->ipython>=5.0.0->ipykernel->openfl==1.0.1) (0.7.0)\n", + "Requirement already satisfied: wcwidth in /home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages (from prompt-toolkit!=3.0.0,!=3.0.1,<3.1.0,>=2.0.0->ipython>=5.0.0->ipykernel->openfl==1.0.1) (0.2.5)\n", + "Requirement already satisfied: ipython-genutils in /home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages (from traitlets>=4.1.0->ipykernel->openfl==1.0.1) (0.2.0)\n", + "Requirement already satisfied: notebook in /home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages (from jupyter->openfl==1.0.1) (6.3.0)\n", + "Requirement already satisfied: jupyter-console in /home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages (from jupyter->openfl==1.0.1) (6.3.0)\n", + "Requirement already satisfied: ipywidgets in /home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages (from jupyter->openfl==1.0.1) (7.6.3)\n", + "Requirement already satisfied: nbconvert in /home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages (from jupyter->openfl==1.0.1) (6.0.7)\n", + "Requirement already satisfied: qtconsole in /home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages (from jupyter->openfl==1.0.1) (5.0.3)\n", + "Requirement already satisfied: jupyterlab-widgets>=1.0.0 in /home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages (from ipywidgets->jupyter->openfl==1.0.1) (1.0.0)\n", + "Requirement already satisfied: nbformat>=4.2.0 in /home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages (from ipywidgets->jupyter->openfl==1.0.1) (5.1.2)\n", + "Requirement already satisfied: widgetsnbextension~=3.5.0 in /home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages (from ipywidgets->jupyter->openfl==1.0.1) (3.5.1)\n", + "Requirement already satisfied: jupyter-core in /home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages (from nbformat>=4.2.0->ipywidgets->jupyter->openfl==1.0.1) (4.7.1)\n", + "Requirement already satisfied: jsonschema!=2.5.0,>=2.4 in /home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages (from nbformat>=4.2.0->ipywidgets->jupyter->openfl==1.0.1) (3.2.0)\n", + "Requirement already satisfied: pyrsistent>=0.14.0 in /home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages (from jsonschema!=2.5.0,>=2.4->nbformat>=4.2.0->ipywidgets->jupyter->openfl==1.0.1) (0.17.3)\n", + "Requirement already satisfied: attrs>=17.4.0 in /home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages (from jsonschema!=2.5.0,>=2.4->nbformat>=4.2.0->ipywidgets->jupyter->openfl==1.0.1) (20.3.0)\n" + ] + }, + { + "name": "stdout", + "output_type": "stream", + "text": [ + "Requirement already satisfied: jinja2 in /home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages (from notebook->jupyter->openfl==1.0.1) (2.11.3)\n", + "Requirement already satisfied: Send2Trash>=1.5.0 in /home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages (from notebook->jupyter->openfl==1.0.1) (1.5.0)\n", + "Requirement already satisfied: terminado>=0.8.3 in /home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages (from notebook->jupyter->openfl==1.0.1) (0.9.3)\n", + "Requirement already satisfied: pyzmq>=17 in /home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages (from notebook->jupyter->openfl==1.0.1) (22.0.3)\n", + "Requirement already satisfied: prometheus-client in /home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages (from notebook->jupyter->openfl==1.0.1) (0.9.0)\n", + "Requirement already satisfied: argon2-cffi in /home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages (from notebook->jupyter->openfl==1.0.1) (20.1.0)\n", + "Requirement already satisfied: python-dateutil>=2.1 in /home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages (from jupyter-client->ipykernel->openfl==1.0.1) (2.8.1)\n", + "Requirement already satisfied: MarkupSafe>=0.23 in /home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages (from jinja2->notebook->jupyter->openfl==1.0.1) (1.1.1)\n", + "Requirement already satisfied: entrypoints>=0.2.2 in /home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages (from nbconvert->jupyter->openfl==1.0.1) (0.3)\n", + "Requirement already satisfied: nbclient<0.6.0,>=0.5.0 in /home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages (from nbconvert->jupyter->openfl==1.0.1) (0.5.3)\n", + "Requirement already satisfied: defusedxml in /home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages (from nbconvert->jupyter->openfl==1.0.1) (0.7.1)\n", + "Requirement already satisfied: bleach in /home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages (from nbconvert->jupyter->openfl==1.0.1) (3.3.0)\n", + "Requirement already satisfied: jupyterlab-pygments in /home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages (from nbconvert->jupyter->openfl==1.0.1) (0.1.2)\n", + "Requirement already satisfied: testpath in /home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages (from nbconvert->jupyter->openfl==1.0.1) (0.4.4)\n", + "Requirement already satisfied: mistune<2,>=0.8.1 in /home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages (from nbconvert->jupyter->openfl==1.0.1) (0.8.4)\n", + "Requirement already satisfied: pandocfilters>=1.4.1 in /home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages (from nbconvert->jupyter->openfl==1.0.1) (1.4.3)\n", + "Requirement already satisfied: nest-asyncio in /home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages (from nbclient<0.6.0,>=0.5.0->nbconvert->jupyter->openfl==1.0.1) (1.5.1)\n", + "Requirement already satisfied: async-generator in /home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages (from nbclient<0.6.0,>=0.5.0->nbconvert->jupyter->openfl==1.0.1) (1.10)\n", + "Requirement already satisfied: packaging in /home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages (from bleach->nbconvert->jupyter->openfl==1.0.1) (20.9)\n", + "Requirement already satisfied: webencodings in /home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages (from bleach->nbconvert->jupyter->openfl==1.0.1) (0.5.1)\n", + "Requirement already satisfied: pyparsing>=2.0.2 in /home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages (from packaging->bleach->nbconvert->jupyter->openfl==1.0.1) (2.4.7)\n", + "Requirement already satisfied: pytz>=2017.3 in /home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages (from pandas->openfl==1.0.1) (2021.1)\n", + "Requirement already satisfied: qtpy in /home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages (from qtconsole->jupyter->openfl==1.0.1) (1.9.0)\n", + "Requirement already satisfied: scipy>=0.19.1 in /home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages (from scikit-learn->openfl==1.0.1) (1.6.1)\n", + "Requirement already satisfied: threadpoolctl>=2.0.0 in /home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages (from scikit-learn->openfl==1.0.1) (2.1.0)\n", + "Requirement already satisfied: joblib>=0.11 in /home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages (from scikit-learn->openfl==1.0.1) (1.0.1)\n", + "Building wheels for collected packages: openfl\n", + " Building wheel for openfl (setup.py) ... \u001b[?25ldone\n", + "\u001b[?25h Created wheel for openfl: filename=openfl-1.0.1-py3-none-any.whl size=4329637 sha256=3c7a0cc242c9c32265142a0c4b0efa4c76cad2db2da12cbcc929c0805919c5bf\n", + " Stored in directory: /home/itrushkin/.cache/pip/wheels/ed/ad/82/f26baafab4be32d052fb76a6f56c9a76ce6903cc3eed7bb394\n", + "Successfully built openfl\n", + "Installing collected packages: openfl\n", + " Attempting uninstall: openfl\n", + " Found existing installation: openfl 1.0.1\n", + " Uninstalling openfl-1.0.1:\n", + " Successfully uninstalled openfl-1.0.1\n", + "Successfully installed openfl-1.0.1\n", + "\u001b[33mWARNING: You are using pip version 21.0.1; however, version 21.1 is available.\n", + "You should consider upgrading via the '/home/itrushkin/.virtualenvs/openfl_research/bin/python -m pip install --upgrade pip' command.\u001b[0m\n" + ] + } + ], "source": [ "#Install dependencies if not already installed\n", - "!pip install torch torchvision" + "!pip install torch torchvision .." ] }, { "cell_type": "code", - "execution_count": null, + "execution_count": 2, "metadata": {}, "outputs": [], "source": [ @@ -44,9 +155,108 @@ }, { "cell_type": "code", - "execution_count": null, + "execution_count": 3, "metadata": {}, - "outputs": [], + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "Creating Workspace Directories\n", + "Creating Workspace Templates\n", + "Successfully installed packages from /home/itrushkin/.local/workspace/requirements.txt.\n", + "\n", + "New workspace directory structure:\n", + "workspace\n", + "├── plan\n", + "│ ├── defaults\n", + "│ │ ├── data_loader.yaml\n", + "│ │ ├── network.yaml\n", + "│ │ ├── tasks_keras.yaml\n", + "│ │ ├── collaborator.yaml\n", + "│ │ ├── assigner.yaml\n", + "│ │ ├── task_runner.yaml\n", + "│ │ ├── defaults\n", + "│ │ ├── tasks_tensorflow.yaml\n", + "│ │ ├── aggregator.yaml\n", + "│ │ ├── tasks_torch.yaml\n", + "│ │ └── tasks_fast_estimator.yaml\n", + "│ ├── plan.yaml\n", + "│ ├── data.yaml\n", + "│ └── cols.yaml\n", + "├── logs\n", + "├── agg_to_col_two_signed_cert.zip\n", + "├── data\n", + "│ └── MNIST\n", + "│ ├── raw\n", + "│ └── processed\n", + "├── cert\n", + "│ ├── ca\n", + "│ │ ├── signing-ca.crt\n", + "│ │ ├── root-ca.crt\n", + "│ │ ├── signing-ca\n", + "│ │ ├── signing-ca.csr\n", + "│ │ └── root-ca\n", + "│ ├── server\n", + "│ │ ├── agg_nnlicv611.inn.intel.com.csr\n", + "│ │ ├── agg_nnlicv611.inn.intel.com.key\n", + "│ │ └── agg_nnlicv611.inn.intel.com.crt\n", + "│ ├── client\n", + "│ │ ├── col_two.crt\n", + "│ │ ├── col_one.crt\n", + "│ │ ├── col_one.key\n", + "│ │ └── col_two.key\n", + "│ ├── cert_chain.crt\n", + "│ ├── col_one.csr\n", + "│ └── col_two.csr\n", + "├── code\n", + "│ ├── __init__.py\n", + "│ ├── mnist_utils.py\n", + "│ ├── ptmnist_inmemory.py\n", + "│ └── pt_cnn.py\n", + "├── final_pytorch_model\n", + "├── .workspace\n", + "├── requirements.txt\n", + "├── agg_to_col_one_signed_cert.zip\n", + "└── save\n", + " ├── torch_cnn_mnist_init.pbuf\n", + " ├── torch_cnn_mnist_best.pbuf\n", + " └── torch_cnn_mnist_last.pbuf\n", + "\n", + "15 directories, 39 files\n", + "Setting Up Certificate Authority...\n", + "\n", + "1. Create Root CA\n", + "1.1 Create Directories\n", + "1.2 Create Database\n", + "1.3 Create CA Request and Certificate\n", + "2. Create Signing Certificate\n", + "2.1 Create Directories\n", + "2.2 Create Database\n", + "2.3 Create Signing Certificate CSR\n", + "2.4 Sign Signing Certificate CSR\n", + "3 Create Certificate Chain\n", + "\n", + "Done.\n", + "Creating AGGREGATOR certificate key pair with following settings: CN=\u001b[31mnnlicv611.inn.intel.com\u001b[0m, SAN=\u001b[31mDNS:nnlicv611.inn.intel.com\u001b[0m\n", + " Writing AGGREGATOR certificate key pair to: \u001b[32mcert/server\u001b[0m\n", + "The CSR Hash for file \u001b[32mserver/agg_nnlicv611.inn.intel.com.csr\u001b[0m = \u001b[31m269da4be6e73f7a492ef3ff73a71eae14eaef3a228427542d0226e29e279222711f8bad6f0e0b62686c19d723c2f46eb\u001b[0m\n", + " Signing AGGREGATOR certificate\n", + "Creating COLLABORATOR certificate key pair with following settings: CN=\u001b[31mone\u001b[0m, SAN=\u001b[31mDNS:one\u001b[0m\n", + " Moving COLLABORATOR certificate to: \u001b[32mcert/col_one\u001b[0m\n", + "The CSR Hash for file \u001b[32mcol_one.csr\u001b[0m = \u001b[31m2c758f295d8b7a321b7af8fcd15c097e327cd53dbafe3393cbe8a6d71f728bf2f1175859e3f90477ace2b022c9b34d6a\u001b[0m\n", + " Signing COLLABORATOR certificate\n", + "\n", + "Registering \u001b[32mone\u001b[0m in \u001b[32mplan/cols.yaml\u001b[0m\n", + "Creating COLLABORATOR certificate key pair with following settings: CN=\u001b[31mtwo\u001b[0m, SAN=\u001b[31mDNS:two\u001b[0m\n", + " Moving COLLABORATOR certificate to: \u001b[32mcert/col_two\u001b[0m\n", + "The CSR Hash for file \u001b[32mcol_two.csr\u001b[0m = \u001b[31m7ff429d1e8baa9f2a6cfb9b26624ea9bbf78c3cd4bca2990061488dabffdd165ded3d51af1e2334469ede41b11af42b8\u001b[0m\n", + " Signing COLLABORATOR certificate\n", + "\n", + "Registering \u001b[32mtwo\u001b[0m in \u001b[32mplan/cols.yaml\u001b[0m\n" + ] + } + ], "source": [ "#Setup default workspace, logging, etc.\n", "fx.init('torch_cnn_mnist')" @@ -61,9 +271,24 @@ }, { "cell_type": "code", - "execution_count": null, + "execution_count": 4, "metadata": {}, - "outputs": [], + "outputs": [ + { + "name": "stderr", + "output_type": "stream", + "text": [ + "/home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/torchvision/datasets/mnist.py:64: UserWarning: train_data has been renamed data\n", + " warnings.warn(\"train_data has been renamed data\")\n", + "/home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/torchvision/datasets/mnist.py:54: UserWarning: train_labels has been renamed targets\n", + " warnings.warn(\"train_labels has been renamed targets\")\n", + "/home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/torchvision/datasets/mnist.py:69: UserWarning: test_data has been renamed data\n", + " warnings.warn(\"test_data has been renamed data\")\n", + "/home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/torchvision/datasets/mnist.py:59: UserWarning: test_labels has been renamed targets\n", + " warnings.warn(\"test_labels has been renamed targets\")\n" + ] + } + ], "source": [ "def one_hot(labels, classes):\n", " return np.eye(classes)[labels]\n", @@ -88,7 +313,7 @@ }, { "cell_type": "code", - "execution_count": null, + "execution_count": 5, "metadata": {}, "outputs": [], "source": [ @@ -126,9 +351,23 @@ }, { "cell_type": "code", - "execution_count": null, + "execution_count": 6, "metadata": {}, - "outputs": [], + "outputs": [ + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">[17:00:50] WARNING tried to remove tensor: __opt_state_needed not present in the tensor dict <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/utilities/utils.py\">utils.py</a>:85\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f1e44b9a3d0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + } + ], "source": [ "\n", "#Create a federated model using the pytorch class, lambda optimizer function, and loss function\n", @@ -144,9 +383,36 @@ }, { "cell_type": "code", - "execution_count": null, + "execution_count": 7, "metadata": {}, - "outputs": [], + "outputs": [ + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> WARNING tried to remove tensor: __opt_state_needed not present in the tensor dict <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/utilities/utils.py\">utils.py</a>:85\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f1e5cf39340>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> WARNING tried to remove tensor: __opt_state_needed not present in the tensor dict <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/utilities/utils.py\">utils.py</a>:85\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f1e44b44e20>" + ] + }, + "metadata": {}, + "output_type": "display_data" + } + ], "source": [ "collaborator_models = fl_model.setup(num_collaborators=2)\n", "collaborators = {'one':collaborator_models[0],'two':collaborator_models[1]}#, 'three':collaborator_models[2]}" @@ -154,9 +420,25 @@ }, { "cell_type": "code", - "execution_count": null, + "execution_count": 8, "metadata": {}, - "outputs": [], + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "Original training data size: 60000\n", + "Original validation data size: 10000\n", + "\n", + "Collaborator one's training data size: 30000\n", + "Collaborator one's validation data size: 5000\n", + "\n", + "Collaborator two's training data size: 30000\n", + "Collaborator two's validation data size: 5000\n", + "\n" + ] + } + ], "source": [ "#Original MNIST dataset\n", "print(f'Original training data size: {len(train_images)}')\n", @@ -184,13 +466,78 @@ }, { "cell_type": "code", - "execution_count": null, + "execution_count": 9, "metadata": {}, - "outputs": [], + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "{\n", + " \"aggregator.settings.best_state_path\": \"save/torch_cnn_mnist_best.pbuf\",\n", + " \"aggregator.settings.db_store_rounds\": 1,\n", + " \"aggregator.settings.init_state_path\": \"save/torch_cnn_mnist_init.pbuf\",\n", + " \"aggregator.settings.last_state_path\": \"save/torch_cnn_mnist_last.pbuf\",\n", + " \"aggregator.settings.rounds_to_train\": 10,\n", + " \"aggregator.template\": \"openfl.component.Aggregator\",\n", + " \"assigner.settings.task_groups\": [\n", + " {\n", + " \"name\": \"train_and_validate\",\n", + " \"percentage\": 1.0,\n", + " \"tasks\": [\n", + " \"aggregated_model_validation\",\n", + " \"train\",\n", + " \"locally_tuned_model_validation\"\n", + " ]\n", + " }\n", + " ],\n", + " \"assigner.template\": \"openfl.component.RandomGroupedAssigner\",\n", + " \"collaborator.settings.db_store_rounds\": 1,\n", + " \"collaborator.settings.delta_updates\": false,\n", + " \"collaborator.settings.opt_treatment\": \"RESET\",\n", + " \"collaborator.template\": \"openfl.component.Collaborator\",\n", + " \"data_loader.settings.batch_size\": 256,\n", + " \"data_loader.settings.collaborator_count\": 2,\n", + " \"data_loader.settings.data_group_name\": \"mnist\",\n", + " \"data_loader.template\": \"code.ptmnist_inmemory.PyTorchMNISTInMemory\",\n", + " \"network.settings.agg_addr\": \"nnlicv611.inn.intel.com\",\n", + " \"network.settings.agg_port\": 56041,\n", + " \"network.settings.cert_folder\": \"cert\",\n", + " \"network.settings.client_reconnect_interval\": 5,\n", + " \"network.settings.disable_client_auth\": false,\n", + " \"network.settings.disable_tls\": false,\n", + " \"network.settings.hash_salt\": \"auto\",\n", + " \"network.template\": \"openfl.federation.Network\",\n", + " \"task_runner.settings\": {},\n", + " \"task_runner.template\": \"code.pt_cnn.PyTorchCNN\",\n", + " \"tasks.aggregated_model_validation.function\": \"validate\",\n", + " \"tasks.aggregated_model_validation.kwargs\": {\n", + " \"apply\": \"global\",\n", + " \"metrics\": [\n", + " \"acc\"\n", + " ]\n", + " },\n", + " \"tasks.locally_tuned_model_validation.function\": \"validate\",\n", + " \"tasks.locally_tuned_model_validation.kwargs\": {\n", + " \"apply\": \"local\",\n", + " \"metrics\": [\n", + " \"acc\"\n", + " ]\n", + " },\n", + " \"tasks.settings\": {},\n", + " \"tasks.train.function\": \"train_batches\",\n", + " \"tasks.train.kwargs\": {\n", + " \"metrics\": [\n", + " \"loss\"\n", + " ]\n", + " }\n", + "}\n" + ] + } + ], "source": [ " #Get the current values of the plan. Each of these can be overridden\n", - "import json\n", - "print(json.dumps(fx.get_plan(), indent=4, sort_keys=True))" + "print(fx.get_plan())" ] }, { @@ -202,9 +549,2168 @@ }, { "cell_type": "code", - "execution_count": null, + "execution_count": 10, "metadata": {}, - "outputs": [], + "outputs": [ + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO Updating aggregator.settings.rounds_to_train to 5... <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/native/native.py\">native.py</a>:82\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f1e44b07040>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO Building 🡆 Object NoCompressionPipeline from openfl.pipelines Module. <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/federated/plan/plan.py\">plan.py</a>:162\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f1e44b076d0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO Settings 🡆 {} <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/federated/plan/plan.py\">plan.py</a>:165\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f1e44b077c0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO Override 🡆 {} <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/federated/plan/plan.py\">plan.py</a>:167\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f1e44b07940>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO Creating Initial Weights File 🠆 save/torch_cnn_mnist_init.pbuf <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/native/native.py\">native.py</a>:240\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f1d58ece5e0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO Starting Experiment... <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/native/native.py\">native.py</a>:244\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f1d58ece550>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO Building 🡆 Object RandomGroupedAssigner from openfl.component Module. <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/federated/plan/plan.py\">plan.py</a>:162\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f1d58ece730>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO Settings 🡆 {'task_groups': [{'name': 'train_and_validate', 'percentage': 1.0, 'tasks': ['aggregated_model_validation', 'train', <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/federated/plan/plan.py\">plan.py</a>:165\n", + " 'locally_tuned_model_validation']}], 'authorized_cols': ['one', 'two'], 'rounds_to_train': 10, 'tasks': {'settings': {}, \n", + " 'defaults': 'plan/defaults/tasks_torch.yaml', 'locally_tuned_model_validation': {'function': 'validate', 'kwargs': {'apply': \n", + " 'local', 'metrics': ['acc']}}, 'train': {'function': 'train_batches', 'kwargs': {'metrics': ['loss']}}, \n", + " 'aggregated_model_validation': {'function': 'validate', 'kwargs': {'apply': 'global', 'metrics': ['acc']}}}} \n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f1d58ece7f0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO Override 🡆 {'defaults': 'plan/defaults/assigner.yaml'} <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/federated/plan/plan.py\">plan.py</a>:167\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f1d58ece400>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO Building 🡆 Object Aggregator from openfl.component Module. <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/federated/plan/plan.py\">plan.py</a>:162\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f1d58e837c0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO Settings 🡆 {'init_state_path': 'save/torch_cnn_mnist_init.pbuf', 'last_state_path': 'save/torch_cnn_mnist_last.pbuf', <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/federated/plan/plan.py\">plan.py</a>:165\n", + " 'rounds_to_train': 5, 'best_state_path': 'save/torch_cnn_mnist_best.pbuf', 'db_store_rounds': 1, 'aggregator_uuid': \n", + " 'aggregator_plan.yaml_acf3e6ef', 'federation_uuid': 'plan.yaml_acf3e6ef', 'authorized_cols': ['one', 'two'], 'assigner': \n", + " <openfl.component.assigner.random_grouped_assigner.RandomGroupedAssigner object at 0x7f1e44b07550>} \n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f1d58e83850>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO Override 🡆 {'defaults': 'plan/defaults/aggregator.yaml'} <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/federated/plan/plan.py\">plan.py</a>:167\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f1d58e836d0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO Building 🡆 Object Collaborator from openfl.component Module. <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/federated/plan/plan.py\">plan.py</a>:162\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f1d58e30af0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO Settings 🡆 {'opt_treatment': 'RESET', 'delta_updates': False, 'db_store_rounds': 1, 'collaborator_name': 'one', <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/federated/plan/plan.py\">plan.py</a>:165\n", + " 'aggregator_uuid': 'aggregator_plan.yaml_acf3e6ef', 'federation_uuid': 'plan.yaml_acf3e6ef', 'task_runner': \n", + " <openfl.federated.task.fl_model.FederatedModel object at 0x7f1e44b9a5b0>, 'tensor_pipe': \n", + " <openfl.pipelines.no_compression_pipeline.NoCompressionPipeline object at 0x7f1e44b07be0>, 'task_config': {'settings': {}, \n", + " 'defaults': 'plan/defaults/tasks_torch.yaml', 'locally_tuned_model_validation': {'function': 'validate', 'kwargs': {'apply': \n", + " 'local', 'metrics': ['acc']}}, 'train': {'function': 'train_batches', 'kwargs': {'metrics': ['loss']}}, \n", + " 'aggregated_model_validation': {'function': 'validate', 'kwargs': {'apply': 'global', 'metrics': ['acc']}}}, 'client': \n", + " <openfl.component.aggregator.aggregator.Aggregator object at 0x7f1d58ece520>} \n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f1d58e30a60>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO Override 🡆 {'defaults': 'plan/defaults/collaborator.yaml'} <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/federated/plan/plan.py\">plan.py</a>:167\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f1d58e302e0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO Building 🡆 Object Collaborator from openfl.component Module. <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/federated/plan/plan.py\">plan.py</a>:162\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f1d58e30f70>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO Settings 🡆 {'opt_treatment': 'RESET', 'delta_updates': False, 'db_store_rounds': 1, 'collaborator_name': 'two', <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/federated/plan/plan.py\">plan.py</a>:165\n", + " 'aggregator_uuid': 'aggregator_plan.yaml_acf3e6ef', 'federation_uuid': 'plan.yaml_acf3e6ef', 'task_runner': \n", + " <openfl.federated.task.fl_model.FederatedModel object at 0x7f1e44b9ac40>, 'tensor_pipe': \n", + " <openfl.pipelines.no_compression_pipeline.NoCompressionPipeline object at 0x7f1e44b07be0>, 'task_config': {'settings': {}, \n", + " 'defaults': 'plan/defaults/tasks_torch.yaml', 'locally_tuned_model_validation': {'function': 'validate', 'kwargs': {'apply': \n", + " 'local', 'metrics': ['acc']}}, 'train': {'function': 'train_batches', 'kwargs': {'metrics': ['loss']}}, \n", + " 'aggregated_model_validation': {'function': 'validate', 'kwargs': {'apply': 'global', 'metrics': ['acc']}}}, 'client': \n", + " <openfl.component.aggregator.aggregator.Aggregator object at 0x7f1d58ece520>, 'defaults': 'plan/defaults/collaborator.yaml'} \n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f1d58e30fa0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO Override 🡆 {'defaults': 'plan/defaults/collaborator.yaml'} <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/federated/plan/plan.py\">plan.py</a>:167\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f1d58e309d0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO Waiting for tasks... <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\">collaborator.py</a>:156\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f1d58e83af0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO Sending tasks to collaborator one for round 0 <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\">aggregator.py</a>:276\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f1d58e83a90>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO Received the following tasks: ['aggregated_model_validation', 'train', 'locally_tuned_model_validation'] <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\">collaborator.py</a>:144\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f1d58e30250>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "name": "stderr", + "output_type": "stream", + "text": [ + "/home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/federated/task/runner_pt.py:287: UserWarning: The given NumPy array is not writeable, and PyTorch does not support non-writeable tensors. This means you can write to the underlying (supposedly non-writeable) NumPy array using the tensor. You may want to copy the array to protect its data or make it writeable before converting it to a tensor. This type of warning will be suppressed for the rest of this program. (Triggered internally at /pytorch/torch/csrc/utils/tensor_numpy.cpp:143.)\n", + " new_state[k] = pt.from_numpy(tensor_dict.pop(k)).to(device)\n", + "/home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/federated/task/runner_pt.py:100: UserWarning: To copy construct from a tensor, it is recommended to use sourceTensor.clone().detach() or sourceTensor.clone().detach().requires_grad_(True), rather than torch.tensor(sourceTensor).\n", + " data, target = pt.tensor(data).to(self.device), pt.tensor(\n" + ] + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">[17:00:52] INFO Sending metric for task aggregated_model_validation, round number 0: acc 0.1226 <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\">collaborator.py</a>:365\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f1d58e30d00>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO Collaborator one is sending task results for aggregated_model_validation, round 0 <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\">aggregator.py</a>:428\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f1d58e30eb0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "name": "stderr", + "output_type": "stream", + "text": [ + "/home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/federated/task/runner_pt.py:154: UserWarning: To copy construct from a tensor, it is recommended to use sourceTensor.clone().detach() or sourceTensor.clone().detach().requires_grad_(True), rather than torch.tensor(sourceTensor).\n", + " data, target = pt.tensor(data).to(self.device), pt.tensor(\n" + ] + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">[17:00:56] INFO Sending metric for task train, round number 0: cross_entropy 0.11886847764253616 <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\">collaborator.py</a>:365\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f1d58e178e0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO Collaborator one is sending task results for train, round 0 <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\">aggregator.py</a>:428\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f1d58e17f10>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO Sending metric for task locally_tuned_model_validation, round number 0: acc 0.954 <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\">collaborator.py</a>:365\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f1d58e17490>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO Collaborator one is sending task results for locally_tuned_model_validation, round 0 <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\">aggregator.py</a>:428\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f1d58e17760>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO All tasks completed on one for round 0... <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\">collaborator.py</a>:148\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f1d58e17430>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO Waiting for tasks... <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\">collaborator.py</a>:156\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f1d58e17c10>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO Sending tasks to collaborator two for round 0 <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\">aggregator.py</a>:276\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f1d58e17880>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO Received the following tasks: ['aggregated_model_validation', 'train', 'locally_tuned_model_validation'] <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\">collaborator.py</a>:144\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f1d58e17880>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO Sending metric for task aggregated_model_validation, round number 0: acc 0.1176 <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\">collaborator.py</a>:365\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f1d58e17760>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO Collaborator two is sending task results for aggregated_model_validation, round 0 <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\">aggregator.py</a>:428\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f1d58e83bb0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">[17:01:01] INFO Sending metric for task train, round number 0: cross_entropy 0.11838725954294205 <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\">collaborator.py</a>:365\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f1d58e30c70>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO Collaborator two is sending task results for train, round 0 <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\">aggregator.py</a>:428\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f1d58e30dc0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO Sending metric for task locally_tuned_model_validation, round number 0: acc 0.953 <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\">collaborator.py</a>:365\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f1d54711340>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO Collaborator two is sending task results for locally_tuned_model_validation, round 0 <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\">aggregator.py</a>:428\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f1d547110d0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO aggregated_model_validation task metrics... <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\">aggregator.py</a>:694\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f1d54711070>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO acc: 0.1201 <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\">aggregator.py</a>:742\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f1d58e83d90>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO Saved the best model with score 0.120100 <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\">aggregator.py</a>:749\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f1d54711220>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO train task metrics... <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\">aggregator.py</a>:694\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f1d547111f0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO cross_entropy: 0.1186 <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\">aggregator.py</a>:742\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f1d58e83550>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO locally_tuned_model_validation task metrics... <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\">aggregator.py</a>:694\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f1d54705b80>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO acc: 0.9535 <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\">aggregator.py</a>:742\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f1d54711220>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO Saving round 1 model... <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\">aggregator.py</a>:783\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f1d58e83700>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO Starting round 1... <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\">aggregator.py</a>:791\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f1d5472d7f0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO All tasks completed on two for round 0... <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\">collaborator.py</a>:148\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f1d54705f40>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO Waiting for tasks... <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\">collaborator.py</a>:156\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f1d5472d4c0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO Sending tasks to collaborator one for round 1 <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\">aggregator.py</a>:276\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f1d5472dc70>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO Received the following tasks: ['aggregated_model_validation', 'train', 'locally_tuned_model_validation'] <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\">collaborator.py</a>:144\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f1d58e30100>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO Sending metric for task aggregated_model_validation, round number 1: acc 0.952 <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\">collaborator.py</a>:365\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f1d58e30c40>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO Collaborator one is sending task results for aggregated_model_validation, round 1 <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\">aggregator.py</a>:428\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f1d58e30e50>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">[17:01:05] INFO Sending metric for task train, round number 1: cross_entropy 0.08371877670288086 <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\">collaborator.py</a>:365\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f1d546e2580>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO Collaborator one is sending task results for train, round 1 <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\">aggregator.py</a>:428\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f1d546e2d60>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO Sending metric for task locally_tuned_model_validation, round number 1: acc 0.9558 <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\">collaborator.py</a>:365\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f1d547118b0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO Collaborator one is sending task results for locally_tuned_model_validation, round 1 <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\">aggregator.py</a>:428\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f1d546e20a0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO All tasks completed on one for round 1... <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\">collaborator.py</a>:148\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f1d546e25e0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO Waiting for tasks... <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\">collaborator.py</a>:156\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f1d546e29a0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO Sending tasks to collaborator two for round 1 <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\">aggregator.py</a>:276\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f1d546e25e0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO Received the following tasks: ['aggregated_model_validation', 'train', 'locally_tuned_model_validation'] <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\">collaborator.py</a>:144\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f1d546e23a0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO Sending metric for task aggregated_model_validation, round number 1: acc 0.957 <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\">collaborator.py</a>:365\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f1d58e83940>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO Collaborator two is sending task results for aggregated_model_validation, round 1 <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\">aggregator.py</a>:428\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f1d58e83df0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">[17:01:09] INFO Sending metric for task train, round number 1: cross_entropy 0.08402730524539948 <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\">collaborator.py</a>:365\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f1d54705fd0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO Collaborator two is sending task results for train, round 1 <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\">aggregator.py</a>:428\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f1d54705b80>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO Sending metric for task locally_tuned_model_validation, round number 1: acc 0.9676 <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\">collaborator.py</a>:365\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f1d58e17e50>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO Collaborator two is sending task results for locally_tuned_model_validation, round 1 <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\">aggregator.py</a>:428\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f1d58e171f0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO aggregated_model_validation task metrics... <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\">aggregator.py</a>:694\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f1d546d84f0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">[17:01:10] INFO acc: 0.9545 <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\">aggregator.py</a>:742\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f1d546d8730>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO Saved the best model with score 0.954500 <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\">aggregator.py</a>:749\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f1d546d8df0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO train task metrics... <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\">aggregator.py</a>:694\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f1d546e2d90>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO cross_entropy: 0.0839 <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\">aggregator.py</a>:742\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f1d546e2070>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO locally_tuned_model_validation task metrics... <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\">aggregator.py</a>:694\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f1d546e25e0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO acc: 0.9617 <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\">aggregator.py</a>:742\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f1d546e2700>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO Saving round 2 model... <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\">aggregator.py</a>:783\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f1d54707b50>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO Starting round 2... <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\">aggregator.py</a>:791\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f1d546dccd0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO All tasks completed on two for round 1... <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\">collaborator.py</a>:148\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f1d54705f40>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO Waiting for tasks... <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\">collaborator.py</a>:156\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f1d546d8190>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO Sending tasks to collaborator one for round 2 <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\">aggregator.py</a>:276\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f1d58e30d00>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO Received the following tasks: ['aggregated_model_validation', 'train', 'locally_tuned_model_validation'] <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\">collaborator.py</a>:144\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f1d58e30e50>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO Sending metric for task aggregated_model_validation, round number 2: acc 0.9632 <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\">collaborator.py</a>:365\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f1d58e30130>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO Collaborator one is sending task results for aggregated_model_validation, round 2 <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\">aggregator.py</a>:428\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f1d546e29d0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">[17:01:14] INFO Sending metric for task train, round number 2: cross_entropy 0.07936933636665344 <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\">collaborator.py</a>:365\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f1d546d8190>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO Collaborator one is sending task results for train, round 2 <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\">aggregator.py</a>:428\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f1d58e30fa0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO Sending metric for task locally_tuned_model_validation, round number 2: acc 0.9772 <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\">collaborator.py</a>:365\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f1d5472dc40>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO Collaborator one is sending task results for locally_tuned_model_validation, round 2 <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\">aggregator.py</a>:428\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f1d58e17640>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO All tasks completed on one for round 2... <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\">collaborator.py</a>:148\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f1d54707e20>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO Waiting for tasks... <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\">collaborator.py</a>:156\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f1d58e177f0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO Sending tasks to collaborator two for round 2 <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\">aggregator.py</a>:276\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f1d58e171f0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO Received the following tasks: ['aggregated_model_validation', 'train', 'locally_tuned_model_validation'] <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\">collaborator.py</a>:144\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f1d58e171f0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO Sending metric for task aggregated_model_validation, round number 2: acc 0.9678 <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\">collaborator.py</a>:365\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f1d547118b0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO Collaborator two is sending task results for aggregated_model_validation, round 2 <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\">aggregator.py</a>:428\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f1d54711070>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">[17:01:19] INFO Sending metric for task train, round number 2: cross_entropy 0.07971804589033127 <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\">collaborator.py</a>:365\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f1d546e29d0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO Collaborator two is sending task results for train, round 2 <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\">aggregator.py</a>:428\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f1d54711a90>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO Sending metric for task locally_tuned_model_validation, round number 2: acc 0.9778 <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\">collaborator.py</a>:365\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f1d546bf550>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO Collaborator two is sending task results for locally_tuned_model_validation, round 2 <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\">aggregator.py</a>:428\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f1d546bf9d0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO aggregated_model_validation task metrics... <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\">aggregator.py</a>:694\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f1d546f8070>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO acc: 0.9655 <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\">aggregator.py</a>:742\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f1d546bf850>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO Saved the best model with score 0.965500 <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\">aggregator.py</a>:749\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f1d546f8700>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO train task metrics... <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\">aggregator.py</a>:694\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f1d546f8580>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO cross_entropy: 0.0795 <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\">aggregator.py</a>:742\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f1d58e171f0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO locally_tuned_model_validation task metrics... <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\">aggregator.py</a>:694\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f1d54696760>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO acc: 0.9775 <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\">aggregator.py</a>:742\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f1d58e17880>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO Saving round 3 model... <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\">aggregator.py</a>:783\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f1d58e17ee0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO Starting round 3... <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\">aggregator.py</a>:791\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f1d546f8fd0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO All tasks completed on two for round 2... <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\">collaborator.py</a>:148\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f1d546d8d30>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO Waiting for tasks... <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\">collaborator.py</a>:156\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f1d546bfee0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO Sending tasks to collaborator one for round 3 <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\">aggregator.py</a>:276\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f1d58e30e50>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO Received the following tasks: ['aggregated_model_validation', 'train', 'locally_tuned_model_validation'] <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\">collaborator.py</a>:144\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f1d58e30130>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO Sending metric for task aggregated_model_validation, round number 3: acc 0.9794 <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\">collaborator.py</a>:365\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f1d58e30fa0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO Collaborator one is sending task results for aggregated_model_validation, round 3 <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\">aggregator.py</a>:428\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f1d546bfc10>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">[17:01:23] INFO Sending metric for task train, round number 3: cross_entropy 0.07742393016815186 <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\">collaborator.py</a>:365\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f1d54707b50>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO Collaborator one is sending task results for train, round 3 <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\">aggregator.py</a>:428\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f1d547118b0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO Sending metric for task locally_tuned_model_validation, round number 3: acc 0.9756 <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\">collaborator.py</a>:365\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f1d546c6400>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO Collaborator one is sending task results for locally_tuned_model_validation, round 3 <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\">aggregator.py</a>:428\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f1d546c6310>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO All tasks completed on one for round 3... <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\">collaborator.py</a>:148\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f1d546e2df0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">[17:01:24] INFO Waiting for tasks... <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\">collaborator.py</a>:156\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f1d546e2df0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO Sending tasks to collaborator two for round 3 <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\">aggregator.py</a>:276\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f1d546c6040>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO Received the following tasks: ['aggregated_model_validation', 'train', 'locally_tuned_model_validation'] <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\">collaborator.py</a>:144\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f1d54696d00>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO Sending metric for task aggregated_model_validation, round number 3: acc 0.9806 <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\">collaborator.py</a>:365\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f1d546c6bb0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO Collaborator two is sending task results for aggregated_model_validation, round 3 <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\">aggregator.py</a>:428\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f1d546c6370>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">[17:01:28] INFO Sending metric for task train, round number 3: cross_entropy 0.07761752605438232 <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\">collaborator.py</a>:365\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f1d546bfcd0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO Collaborator two is sending task results for train, round 3 <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\">aggregator.py</a>:428\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f1d546e2700>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO Sending metric for task locally_tuned_model_validation, round number 3: acc 0.9834 <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\">collaborator.py</a>:365\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f1d546c6400>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO Collaborator two is sending task results for locally_tuned_model_validation, round 3 <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\">aggregator.py</a>:428\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f1d546c6310>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO aggregated_model_validation task metrics... <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\">aggregator.py</a>:694\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f1d546a8370>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO acc: 0.9800 <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\">aggregator.py</a>:742\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f1d546c6880>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO Saved the best model with score 0.980000 <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\">aggregator.py</a>:749\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f1d546a8610>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO train task metrics... <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\">aggregator.py</a>:694\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f1d546a8b20>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO cross_entropy: 0.0775 <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\">aggregator.py</a>:742\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f1d546c6520>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO locally_tuned_model_validation task metrics... <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\">aggregator.py</a>:694\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f1d546b8be0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO acc: 0.9795 <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\">aggregator.py</a>:742\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f1d546b8bb0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO Saving round 4 model... <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\">aggregator.py</a>:783\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f1d546c6e20>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO Starting round 4... <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\">aggregator.py</a>:791\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f1d546b8400>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO All tasks completed on two for round 3... <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\">collaborator.py</a>:148\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f1d54707880>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO Waiting for tasks... <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\">collaborator.py</a>:156\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f1d54696d00>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO Sending tasks to collaborator one for round 4 <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\">aggregator.py</a>:276\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f1d58e17880>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO Received the following tasks: ['aggregated_model_validation', 'train', 'locally_tuned_model_validation'] <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\">collaborator.py</a>:144\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f1d546f8dc0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO Sending metric for task aggregated_model_validation, round number 4: acc 0.9792 <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\">collaborator.py</a>:365\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f1d5468c4c0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO Collaborator one is sending task results for aggregated_model_validation, round 4 <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\">aggregator.py</a>:428\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f1d5468cdc0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">[17:01:33] INFO Sending metric for task train, round number 4: cross_entropy 0.07607017457485199 <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\">collaborator.py</a>:365\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f1d5469c340>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO Collaborator one is sending task results for train, round 4 <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\">aggregator.py</a>:428\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f1d5469ce50>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO Sending metric for task locally_tuned_model_validation, round number 4: acc 0.9804 <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\">collaborator.py</a>:365\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f1d58e17880>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO Collaborator one is sending task results for locally_tuned_model_validation, round 4 <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\">aggregator.py</a>:428\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f1d58e17880>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO All tasks completed on one for round 4... <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\">collaborator.py</a>:148\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f1d546c6370>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO Waiting for tasks... <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\">collaborator.py</a>:156\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f1d546e2700>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO Sending tasks to collaborator two for round 4 <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\">aggregator.py</a>:276\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f1d546b8df0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO Received the following tasks: ['aggregated_model_validation', 'train', 'locally_tuned_model_validation'] <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\">collaborator.py</a>:144\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f1d58e83790>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO Sending metric for task aggregated_model_validation, round number 4: acc 0.9812 <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\">collaborator.py</a>:365\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f1d546e2940>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO Collaborator two is sending task results for aggregated_model_validation, round 4 <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\">aggregator.py</a>:428\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f1d54705f40>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">[17:01:37] INFO Sending metric for task train, round number 4: cross_entropy 0.07631511241197586 <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\">collaborator.py</a>:365\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f1d546a80a0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO Collaborator two is sending task results for train, round 4 <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\">aggregator.py</a>:428\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f1d546a80a0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO Sending metric for task locally_tuned_model_validation, round number 4: acc 0.9836 <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\">collaborator.py</a>:365\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f1d546c6370>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO Collaborator two is sending task results for locally_tuned_model_validation, round 4 <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\">aggregator.py</a>:428\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f1d546b1760>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO aggregated_model_validation task metrics... <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\">aggregator.py</a>:694\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f1d546b1eb0>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO acc: 0.9802 <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\">aggregator.py</a>:742\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f1d5469c940>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO Saved the best model with score 0.980200 <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\">aggregator.py</a>:749\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f1d5469ca00>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO train task metrics... <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\">aggregator.py</a>:694\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f1d546b1b20>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO cross_entropy: 0.0762 <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\">aggregator.py</a>:742\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f1d54711a90>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">[17:01:38] INFO locally_tuned_model_validation task metrics... <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\">aggregator.py</a>:694\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f1d546b1790>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO acc: 0.9820 <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\">aggregator.py</a>:742\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f1d5469cc70>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO Saving round 5 model... <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\">aggregator.py</a>:783\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f1d5469ce50>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO Experiment Completed. Cleaning up... <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/aggregator/aggregator.py\">aggregator.py</a>:789\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f1d5469de50>" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"> INFO All tasks completed on two for round 4... <a href=\"file:///home/itrushkin/.virtualenvs/openfl_research/lib/python3.8/site-packages/openfl/component/collaborator/collaborator.py\">collaborator.py</a>:148\n", + "</pre>\n" + ], + "text/plain": [ + "<rich.jupyter.JupyterRenderable at 0x7f1d546d8d30>" + ] + }, + "metadata": {}, + "output_type": "display_data" + } + ], "source": [ "#Run experiment, return trained FederatedModel\n", "final_fl_model = fx.run_experiment(collaborators,{'aggregator.settings.rounds_to_train':5})" @@ -212,7 +2718,7 @@ }, { "cell_type": "code", - "execution_count": null, + "execution_count": 11, "metadata": {}, "outputs": [], "source": [ diff --git a/openfl-tutorials/Federated_Pytorch_MNIST_custom_aggregation_Tutorial.ipynb b/openfl-tutorials/Federated_Pytorch_MNIST_custom_aggregation_Tutorial.ipynb new file mode 100644 index 0000000000..d635a51d30 --- /dev/null +++ b/openfl-tutorials/Federated_Pytorch_MNIST_custom_aggregation_Tutorial.ipynb @@ -0,0 +1,430 @@ +{ + "cells": [ + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "# Federated PyTorch MNIST Tutorial" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "#Install dependencies if not already installed\n", + "!pip install torch torchvision" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "import numpy as np\n", + "import torch\n", + "import torch.nn as nn\n", + "import torch.nn.functional as F\n", + "import torch.optim as optim\n", + "\n", + "import torchvision\n", + "import torchvision.transforms as transforms\n", + "import openfl.native as fx\n", + "from openfl.federated import FederatedModel,FederatedDataSet\n", + "\n", + "torch.manual_seed(0)\n", + "np.random.seed(0)" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "After importing the required packages, the next step is setting up our openfl workspace. To do this, simply run the `fx.init()` command as follows:" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "#Setup default workspace, logging, etc.\n", + "fx.init('torch_cnn_mnist')" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "Now we are ready to define our dataset and model to perform federated learning on. The dataset should be composed of a numpy arrayWe start with a simple fully connected model that is trained on the MNIST dataset. " + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "def one_hot(labels, classes):\n", + " return np.eye(classes)[labels]\n", + "\n", + "transform = transforms.Compose(\n", + " [transforms.ToTensor(),\n", + " transforms.Normalize((0.5, 0.5, 0.5), (0.5, 0.5, 0.5))])\n", + "\n", + "trainset = torchvision.datasets.MNIST(root='./data', train=True,\n", + " download=True, transform=transform)\n", + "\n", + "train_images,train_labels = trainset.train_data, np.array(trainset.train_labels)\n", + "train_images = torch.from_numpy(np.expand_dims(train_images, axis=1)).float()\n", + "train_labels = one_hot(train_labels,10)\n", + "\n", + "validset = torchvision.datasets.MNIST(root='./data', train=False,\n", + " download=True, transform=transform)\n", + "\n", + "valid_images,valid_labels = validset.test_data, np.array(validset.test_labels)\n", + "valid_images = torch.from_numpy(np.expand_dims(valid_images, axis=1)).float()\n", + "valid_labels = one_hot(valid_labels,10)" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "feature_shape = train_images.shape[1]\n", + "classes = 10\n", + "\n", + "fl_data = FederatedDataSet(train_images,train_labels,valid_images,valid_labels,batch_size=32,num_classes=classes)\n", + "\n", + "class Net(nn.Module):\n", + " def __init__(self):\n", + " super(Net, self).__init__()\n", + " self.conv1 = nn.Conv2d(1, 16, 3)\n", + " self.pool = nn.MaxPool2d(2, 2)\n", + " self.conv2 = nn.Conv2d(16, 32, 3)\n", + " self.fc1 = nn.Linear(32 * 5 * 5, 32)\n", + " self.fc2 = nn.Linear(32, 84)\n", + " self.fc3 = nn.Linear(84, 10)\n", + "\n", + " def forward(self, x):\n", + " x = self.pool(F.relu(self.conv1(x)))\n", + " x = self.pool(F.relu(self.conv2(x)))\n", + " x = x.view(x.size(0),-1)\n", + " x = F.relu(self.fc1(x))\n", + " x = F.relu(self.fc2(x))\n", + " x = self.fc3(x)\n", + " return x\n", + " \n", + "optimizer = lambda x: optim.Adam(x, lr=1e-4)\n", + "\n", + "def cross_entropy(output, target):\n", + " \"\"\"Binary cross-entropy metric\n", + " \"\"\"\n", + " return F.cross_entropy(input=output,target=torch.argmax(target, dim=1))" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "\n", + "#Create a federated model using the pytorch class, lambda optimizer function, and loss function\n", + "fl_model = FederatedModel(build_model=Net,optimizer=optimizer,loss_fn=cross_entropy,data_loader=fl_data)" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "The `FederatedModel` object is a wrapper around your Keras, Tensorflow or PyTorch model that makes it compatible with openfl. It provides built in federated training and validation functions that we will see used below. Using it's `setup` function, collaborator models and datasets can be automatically defined for the experiment. " + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "collaborator_models = fl_model.setup(num_collaborators=10)\n", + "collaborators = {str(i): collaborator_models[i] for i in range(10)}#, 'three':collaborator_models[2]}" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "#Original MNIST dataset\n", + "print(f'Original training data size: {len(train_images)}')\n", + "print(f'Original validation data size: {len(valid_images)}\\n')\n", + "\n", + "#Collaborator one's data\n", + "print(f'Collaborator one\\'s training data size: {len(collaborator_models[0].data_loader.X_train)}')\n", + "print(f'Collaborator one\\'s validation data size: {len(collaborator_models[0].data_loader.X_valid)}\\n')\n", + "\n", + "#Collaborator two's data\n", + "print(f'Collaborator two\\'s training data size: {len(collaborator_models[1].data_loader.X_train)}')\n", + "print(f'Collaborator two\\'s validation data size: {len(collaborator_models[1].data_loader.X_valid)}\\n')\n", + "\n", + "#Collaborator three's data\n", + "#print(f'Collaborator three\\'s training data size: {len(collaborator_models[2].data_loader.X_train)}')\n", + "#print(f'Collaborator three\\'s validation data size: {len(collaborator_models[2].data_loader.X_valid)}')" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "We can see the current plan values by running the `fx.get_plan()` function" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + " #Get the current values of the plan. Each of these can be overridden\n", + "print(fx.get_plan())" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "Now we are ready to run our experiment. If we want to pass in custom plan settings, we can easily do that with the `override_config` parameter" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "from openfl.component.aggregation_functions import AggregationFunctionInterface\n", + "import numpy as np\n", + "\n", + "class ExponentialSmoothingAveraging(AggregationFunctionInterface):\n", + " \"\"\"\n", + " Averaging via exponential smoothing.\n", + " \n", + " In order to use this mechanism properly you should specify `aggregator.settings.db_store_rounds` \n", + " in `override_config` keyword argument of `run_experiment` function. \n", + " It should be equal to the number of rounds you want to include in smoothing window.\n", + " \n", + " Args:\n", + " alpha(float): Smoothing term.\n", + " \"\"\"\n", + " def __init__(self, alpha=0.9):\n", + " self.alpha = alpha\n", + " \n", + " def call(self,\n", + " local_tensors,\n", + " db_iterator,\n", + " tensor_name,\n", + " fl_round,\n", + " tags):\n", + " \"\"\"Aggregate tensors.\n", + "\n", + " Args:\n", + " local_tensors(list[openfl.utilities.LocalTensor]): List of local tensors to aggregate.\n", + " db_iterator: iterator over history of all tensors.\n", + " Columns: ['tensor_name', 'round', 'tags', 'nparray']\n", + " tensor_name: name of the tensor\n", + " fl_round: round number\n", + " tags: tuple of tags for this tensor\n", + " \"\"\"\n", + " tensors, weights = zip(*[(x.tensor, x.weight) for x in local_tensors])\n", + " tensors, weights = np.array(tensors), np.array(weights)\n", + " average = np.average(tensors, weights=weights, axis=0)\n", + " if 'metric' in tags:\n", + " return average\n", + " previous_tensor_values = []\n", + " for record in db_iterator:\n", + " if (\n", + " record['tensor_name'] == tensor_name\n", + " and 'aggregated' in record['tags']\n", + " and 'delta' not in record['tags']\n", + " ):\n", + " previous_tensor_values.append(record['nparray'])\n", + " for i, x in enumerate(previous_tensor_values):\n", + " previous_tensor_values[i] = x * self.alpha * (1 - self.alpha) ** i\n", + " smoothing_term = np.sum(previous_tensor_values, axis=0)\n", + " return self.alpha * average + (1 - self.alpha) * smoothing_term" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "from openfl.component.aggregation_functions import AggregationFunctionInterface\n", + "import numpy as np\n", + "\n", + "class ClippedAveraging(AggregationFunctionInterface):\n", + " def __init__(self, ratio):\n", + " \"\"\"Average clipped tensors.\n", + " \n", + " Args:\n", + " ratio(float): Ratio to multiply with a tensor for clipping\n", + " \"\"\"\n", + " self.ratio = ratio\n", + " \n", + " def call(self,\n", + " local_tensors,\n", + " db_iterator,\n", + " tensor_name,\n", + " fl_round,\n", + " *__):\n", + " \"\"\"Aggregate tensors.\n", + "\n", + " Args:\n", + " local_tensors(list[openfl.utilities.LocalTensor]): List of local tensors to aggregate.\n", + " db_iterator: iterator over history of all tensors.\n", + " Columns: ['tensor_name', 'round', 'tags', 'nparray']\n", + " tensor_name: name of the tensor\n", + " fl_round: round number\n", + " tags: tuple of tags for this tensor\n", + " \"\"\"\n", + " clipped_tensors = []\n", + " previous_tensor_value = None\n", + " for record in db_iterator:\n", + " if (\n", + " record['round'] == (fl_round - 1)\n", + " and record['tensor_name'] == tensor_name\n", + " and 'aggregated' in record['tags']\n", + " and 'delta' not in record['tags']\n", + " ):\n", + " previous_tensor_value = record['nparray']\n", + " weights = []\n", + " for local_tensor in local_tensors:\n", + " prev_tensor = previous_tensor_value if previous_tensor_value is not None else local_tensor.tensor\n", + " delta = local_tensor.tensor - prev_tensor\n", + " new_tensor = prev_tensor + delta * self.ratio\n", + " clipped_tensors.append(new_tensor)\n", + " weights.append(local_tensor.weight)\n", + "\n", + " return np.average(clipped_tensors, weights=weights, axis=0)" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "from openfl.component.aggregation_functions import AggregationFunctionInterface\n", + "\n", + "class ConditionalThresholdAveraging(AggregationFunctionInterface):\n", + " def __init__(self, threshold_fn, metric_name='acc', tags=['metric', 'validate_local']):\n", + " \"\"\"Average tensors by metric value on previous round.\n", + " If no tensors match threshold condition, a simple weighted averaging will be performed.\n", + " \n", + " Args:\n", + " threshold_fn(callable): function to define a threshold for each round.\n", + " Has single argument `round_number`. \n", + " Returns threshold value above which collaborators are allowed to participate in aggregation.\n", + " metric_name(str): name of the metric to trace. Can be either 'acc' or 'loss'.\n", + " tags(Tuple[str]): tags of the metric tensor.\n", + " \"\"\"\n", + " self.metric_name = metric_name\n", + " self.threshold_fn = threshold_fn\n", + " self.tags = tags\n", + " self.logged_round = -1\n", + " \n", + " def call(self,\n", + " local_tensors,\n", + " db_iterator,\n", + " tensor_name,\n", + " fl_round,\n", + " *__):\n", + " \"\"\"Aggregate tensors.\n", + "\n", + " Args:\n", + " local_tensors(list[openfl.utilities.LocalTensor]): List of local tensors to aggregate.\n", + " db_iterator: iterator over history of all tensors.\n", + " Columns: ['tensor_name', 'round', 'tags', 'nparray']\n", + " tensor_name: name of the tensor\n", + " fl_round: round number\n", + " tags: tuple of tags for this tensor\n", + " \"\"\"\n", + " selected_tensors = []\n", + " selected_weights = []\n", + " for record in db_iterator:\n", + " for local_tensor in local_tensors:\n", + " tags = set(self.tags + [local_tensor.col_name])\n", + " if (\n", + " tags <= set(record['tags']) \n", + " and record['round'] == fl_round\n", + " and record['tensor_name'] == self.metric_name\n", + " and record['nparray'] >= self.threshold_fn(fl_round)\n", + " ):\n", + " selected_tensors.append(local_tensor.tensor)\n", + " selected_weights.append(local_tensor.weight)\n", + " if not selected_tensors:\n", + " if self.logged_round < fl_round:\n", + " fx.logger.warning('No collaborators match threshold condition. Performing simple averaging...')\n", + " selected_tensors = [local_tensor.tensor for tensor in local_tensors]\n", + " selected_weights = [local_tensor.weight for tensor in local_tensors]\n", + " if self.logged_round < fl_round:\n", + " self.logged_round += 1\n", + " return np.average(selected_tensors, weights=selected_weights, axis=0)" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "#Run experiment, return trained FederatedModel\n", + "final_fl_model = fx.run_experiment(collaborators,\n", + " {\n", + " 'aggregator.settings.rounds_to_train':5,\n", + " 'aggregator.settings.db_store_rounds':5,\n", + " 'tasks.train.aggregation_type': ConditionalThresholdAveraging(lambda round_num: 0.85 + 0.03 * round_num)\n", + " })" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "#Save final model\n", + "final_fl_model.save_native('final_pytorch_model')" + ] + } + ], + "metadata": { + "kernelspec": { + "display_name": "Python 3", + "language": "python", + "name": "python3" + }, + "language_info": { + "codemirror_mode": { + "name": "ipython", + "version": 3 + }, + "file_extension": ".py", + "mimetype": "text/x-python", + "name": "python", + "nbconvert_exporter": "python", + "pygments_lexer": "ipython3", + "version": "3.8.8" + } + }, + "nbformat": 4, + "nbformat_minor": 4 +} diff --git a/openfl/component/aggregation_functions/__init__.py b/openfl/component/aggregation_functions/__init__.py new file mode 100644 index 0000000000..f0d8247f67 --- /dev/null +++ b/openfl/component/aggregation_functions/__init__.py @@ -0,0 +1,11 @@ +# Copyright (C) 2020-2021 Intel Corporation +# SPDX-License-Identifier: Apache-2.0 + +"""Aggregation functions package.""" + +from .median import Median +from .weighted_average import WeightedAverage +from .geometric_median import GeometricMedian +from .interface import AggregationFunctionInterface + +__all__ = ['Median', 'WeightedAverage', 'GeometricMedian', 'AggregationFunctionInterface'] diff --git a/openfl/component/aggregation_functions/geometric_median.py b/openfl/component/aggregation_functions/geometric_median.py new file mode 100644 index 0000000000..5f19c1adfc --- /dev/null +++ b/openfl/component/aggregation_functions/geometric_median.py @@ -0,0 +1,62 @@ +# Copyright (C) 2020-2021 Intel Corporation +# SPDX-License-Identifier: Apache-2.0 + +"""Geometric median module.""" + +from .interface import AggregationFunctionInterface +import numpy as np +from .weighted_average import weighted_average + + +def _geometric_median_objective(median, tensors, weights): + """Compute geometric median objective.""" + return sum([w * _l2dist(median, x) for w, x in zip(weights, tensors)]) + + +def geometric_median(tensors, weights, maxiter=4, eps=1e-5, ftol=1e-6): + """Compute geometric median of tensors with weights using Weiszfeld's Algorithm.""" + weights = np.asarray(weights) / sum(weights) + median = weighted_average(tensors, weights) + num_oracle_calls = 1 + + obj_val = _geometric_median_objective(median, tensors, weights) + + for _ in range(maxiter): + prev_obj_val = obj_val + weights = np.asarray([w / max(eps, _l2dist(median, x)) for w, x in zip(weights, tensors)]) + weights = weights / weights.sum() + median = weighted_average(tensors, weights) + num_oracle_calls += 1 + obj_val = _geometric_median_objective(median, tensors, weights) + if abs(prev_obj_val - obj_val) < ftol * obj_val: + break + return median + + +def _l2dist(p1, p2): + """L2 distance between p1, p2, each of which is a list of nd-arrays.""" + if p1.ndim != p2.ndim: + raise RuntimeError('Tensor shapes should be equal') + if p1.ndim < 2: + return _l2dist(*[np.expand_dims(x, axis=0) for x in [p1, p2]]) + return np.linalg.norm([np.linalg.norm(x1 - x2) for x1, x2 in zip(p1, p2)]) + + +class GeometricMedian(AggregationFunctionInterface): + """Geometric median aggregation.""" + + def call(self, local_tensors, *_) -> np.ndarray: + """Aggregate tensors. + + Args: + agg_tensor_dict: Dict of (collaborator name, tensor) pairs to aggregate. + weights: array of floats representing data partition (sum up to 1) + db_iterator: iterator over history of all tensors. + Columns: ['tensor_name', 'round', 'tags', 'nparray'] + tensor_name: name of the tensor + fl_round: round number + tags: tuple of tags for this tensor + """ + tensors, weights = zip(*[(x.tensor, x.weight) for x in local_tensors]) + tensors, weights = np.array(tensors), np.array(weights) + return geometric_median(tensors, weights) diff --git a/openfl/component/aggregation_functions/interface.py b/openfl/component/aggregation_functions/interface.py new file mode 100644 index 0000000000..bdc3a62686 --- /dev/null +++ b/openfl/component/aggregation_functions/interface.py @@ -0,0 +1,41 @@ +# Copyright (C) 2020-2021 Intel Corporation +# SPDX-License-Identifier: Apache-2.0 +"""Aggregation function interface module.""" +from typing import Iterator, Tuple, List +from openfl.utilities import LocalTensor +import numpy as np +from abc import ABC, abstractmethod + +import pandas as pd + + +class AggregationFunctionInterface(ABC): + """Interface for specifying aggregation function.""" + + @abstractmethod + def call(self, + local_tensors: List[LocalTensor], + weights: np.ndarray, + db_iterator: Iterator[pd.Series], + tensor_name: str, + fl_round: int, + tags: Tuple[str]) -> np.ndarray: + """Aggregate tensors. + + Args: + local_tensors(list[openfl.utilities.LocalTensor]): List of local tensors to aggregate. + db_iterator: iterator over history of all tensors. + Columns: ['tensor_name', 'round', 'tags', 'nparray'] + tensor_name: name of the tensor + fl_round: round number + tags: tuple of tags for this tensor + """ + raise NotImplementedError + + def __call__(self, local_tensors, + db_iterator, + tensor_name, + fl_round, + tags): + """Use magic function for ease.""" + return self.call(local_tensors, db_iterator, tensor_name, fl_round, tags) diff --git a/openfl/component/aggregation_functions/median.py b/openfl/component/aggregation_functions/median.py new file mode 100644 index 0000000000..81ef01c05c --- /dev/null +++ b/openfl/component/aggregation_functions/median.py @@ -0,0 +1,26 @@ +# Copyright (C) 2020-2021 Intel Corporation +# SPDX-License-Identifier: Apache-2.0 + +"""Median module.""" + +from .interface import AggregationFunctionInterface +import numpy as np + + +class Median(AggregationFunctionInterface): + """Median aggregation.""" + + def call(self, local_tensors, *_): + """Aggregate tensors. + + Args: + agg_tensor_dict: Dict of (collaborator name, tensor) pairs to aggregate. + weights: array of floats representing data partition (sum up to 1) + db_iterator: iterator over history of all tensors. + Columns: ['tensor_name', 'round', 'tags', 'nparray'] + tensor_name: name of the tensor + fl_round: round number + tags: tuple of tags for this tensor + """ + tensors = np.array([x.tensor for x in local_tensors]) + return np.median(tensors, axis=0) diff --git a/openfl/component/aggregation_functions/weighted_average.py b/openfl/component/aggregation_functions/weighted_average.py new file mode 100644 index 0000000000..e7ab596b4d --- /dev/null +++ b/openfl/component/aggregation_functions/weighted_average.py @@ -0,0 +1,31 @@ +# Copyright (C) 2020-2021 Intel Corporation +# SPDX-License-Identifier: Apache-2.0 + +"""Federated averaging module.""" + +from .interface import AggregationFunctionInterface +import numpy as np + + +def weighted_average(tensors, weights): + """Compute average.""" + return np.average(tensors, weights=weights, axis=0) + + +class WeightedAverage(AggregationFunctionInterface): + """Weighted average aggregation.""" + + def call(self, local_tensors, *_): + """Aggregate tensors. + + Args: + agg_tensor_dict: Dict of (collaborator name, tensor) pairs to aggregate. + weights: array of floats representing data partition (sum up to 1) + db_iterator: iterator over history of all tensors. + Columns: ['tensor_name', 'round', 'tags', 'nparray'] + tensor_name: name of the tensor + fl_round: round number + tags: tuple of tags for this tensor + """ + tensors, weights = zip(*[(x.tensor, x.weight) for x in local_tensors]) + return weighted_average(tensors, weights) diff --git a/openfl/component/aggregator/aggregator.py b/openfl/component/aggregator/aggregator.py index 9efc5ec159..719cf9257a 100644 --- a/openfl/component/aggregator/aggregator.py +++ b/openfl/component/aggregator/aggregator.py @@ -741,7 +741,7 @@ def _compute_validation_related_task_metrics(self, task_name): # collaborator in our subset, and apply the correct # transformations to the tensorkey to resolve the aggregated # tensor for that round - agg_functions = self.assigner.get_aggregation_type_for_task(task_name) + agg_function = self.assigner.get_aggregation_type_for_task(task_name) task_key = TaskResultKey(task_name, collaborators_for_task[0], self.round_number) for tensor_key in self.collaborator_tasks_results[task_key]: tensor_name, origin, round_number, report, tags = tensor_key @@ -752,8 +752,8 @@ def _compute_validation_related_task_metrics(self, task_name): new_tags = tuple(list(tags[:-1])) agg_tensor_key = TensorKey(tensor_name, origin, round_number, report, new_tags) agg_tensor_name, agg_origin, agg_round_number, agg_report, agg_tags = agg_tensor_key - agg_results, agg_metadata_dict = self.tensor_db.get_aggregated_tensor( - agg_tensor_key, collaborator_weight_dict, agg_functions) + agg_results = self.tensor_db.get_aggregated_tensor( + agg_tensor_key, collaborator_weight_dict, aggregation_function=agg_function) if report: # Print the aggregated metric if agg_results is None: @@ -761,16 +761,12 @@ def _compute_validation_related_task_metrics(self, task_name): 'Aggregated metric {} could not be collected for round {}. ' 'Skipping reporting for this round'.format( agg_tensor_name, self.round_number)) - if agg_functions is not None: + if agg_function: self.logger.info('{0} {1}:\t{2:.4f}'.format( - agg_functions[0], agg_tensor_name, agg_results) + agg_function, agg_tensor_name, agg_results) ) else: self.logger.info('{0}:\t{1:.4f}'.format(agg_tensor_name, agg_results)) - for met in agg_metadata_dict: - self.logger.info('{0} {1}:\t{2:.4f}'.format( - met, agg_tensor_name, agg_metadata_dict[met]) - ) # TODO Add all of the logic for saving the model based # on best accuracy, lowest loss, etc. if 'validate_agg' in tags: diff --git a/openfl/databases/tensor_db.py b/openfl/databases/tensor_db.py index 6e439deb38..ff251a7404 100644 --- a/openfl/databases/tensor_db.py +++ b/openfl/databases/tensor_db.py @@ -8,7 +8,9 @@ from threading import Lock -from openfl.utilities import TensorKey +from openfl.utilities import TensorKey, LocalTensor +from openfl.component.aggregation_functions import (WeightedAverage, Median, GeometricMedian, + AggregationFunctionInterface) class TensorDB: @@ -20,6 +22,12 @@ class TensorDB: collaborator and aggregator has its own TensorDB. """ + aggregation_fns = { + 'weighted_average': WeightedAverage(), + 'median': Median(), + 'geometric_median': GeometricMedian() + } + def __init__(self): """Initialize.""" self.tensor_db = pd.DataFrame([], columns=[ @@ -101,7 +109,7 @@ def get_tensor_from_cache(self, tensor_key): return np.array(df['nparray'].iloc[0]) def get_aggregated_tensor(self, tensor_key, collaborator_weight_dict, - aggregation_functions): + aggregation_function=None): """ Determine whether all of the collaborator tensors are present for a given tensor key. @@ -113,16 +121,25 @@ def get_aggregated_tensor(self, tensor_key, collaborator_weight_dict, compute weighted average of all collaborators collaborator_weight_dict: List of collaborator names in federation and their respective weights - aggregation_functions: Call the underlying numpy aggregation + aggregation_function: Call the underlying numpy aggregation function. Default is just the weighted - average. ONLY THE FIRST FUNCTION will get - recorded as the aggregated tensor - + average. Returns: weighted_nparray if all collaborator values are present None if not all values are present """ + if aggregation_function is None: + aggregation_function = 'weighted_average' + if aggregation_function in self.aggregation_fns: + aggregation_function = self.aggregation_fns[aggregation_function] + elif not isinstance(aggregation_function, AggregationFunctionInterface): + raise NotImplementedError( + 'Aggregation function should either ' + + f'implement {AggregationFunctionInterface.__name__} interface ' + + 'or be one of ' + + f'{list(TensorDB.aggregation_fns.keys())}') + if len(collaborator_weight_dict) != 0: assert (np.abs( 1.0 - sum(collaborator_weight_dict.values()) @@ -130,9 +147,6 @@ def get_aggregated_tensor(self, tensor_key, collaborator_weight_dict, 'Collaborator weights do not sum to 1.0:' \ ' {}'.format(collaborator_weight_dict) - if aggregation_functions is None: - aggregation_functions = ['weighted_average'] - collaborator_names = collaborator_weight_dict.keys() agg_tensor_dict = {} @@ -164,41 +178,30 @@ def get_aggregated_tensor(self, tensor_key, collaborator_weight_dict, col, TensorKey( tensor_name, origin, report, fl_round, new_tags ))) - return None, {} + return None else: agg_tensor_dict[col] = raw_df.iloc[0] # agg_tensor_dict[col] = agg_tensor_dict[col] # * collaborator_weight_dict[col] - - concat_nparray = np.array(list(agg_tensor_dict.values())) - - aggregated_tensorkey_is_set = False - agg_metadata_dict = {} - - for aggregation_function in aggregation_functions: - if aggregation_function == 'weighted_average': - agg_nparray = np.average( - concat_nparray, - weights=np.array(list(collaborator_weight_dict.values())), - axis=0 - ) - else: - agg_func = getattr(np, aggregation_function, None) - if callable(agg_func): - agg_nparray = agg_func(concat_nparray, axis=0) - else: - raise KeyError( - '{} is not a valid numpy function'.format( - aggregation_function)) - - # TODO This enforces that only the first aggregation function - # will be associated with - if not aggregated_tensorkey_is_set: - # Cache aggregated tensor in TensorDB - self.cache_tensor({tensor_key: agg_nparray}) - aggregated_tensorkey_is_set = True - primary_agg = agg_nparray - else: - agg_metadata_dict[aggregation_function] = np.array(agg_nparray) - - return np.array(primary_agg), agg_metadata_dict + local_tensors = [LocalTensor(col_name=col_name, + tensor=agg_tensor_dict[col_name], + weight=collaborator_weight_dict[col_name]) + for col_name in collaborator_names] + + db_iterator = self._iterate() + agg_nparray = aggregation_function(local_tensors, + db_iterator, + tensor_name, + fl_round, + tags) + self.cache_tensor({tensor_key: agg_nparray}) + + return np.array(agg_nparray) + + def _iterate(self, order_by='round', ascending=False): + columns = ['round', 'nparray', 'tensor_name', 'tags'] + + for _, row in self.tensor_db[columns]\ + .sort_values(by=order_by, ascending=ascending)\ + .iterrows(): + yield row diff --git a/openfl/federated/plan/plan.py b/openfl/federated/plan/plan.py index 0a0b27a1da..8062f34c8d 100644 --- a/openfl/federated/plan/plan.py +++ b/openfl/federated/plan/plan.py @@ -49,11 +49,11 @@ def ignore_aliases(self, data): if frozen_yaml_path.exists(): Plan.logger.info(f"{yaml_path.name} is already frozen") return - frozen_yaml_path.write_text(dump(config, Dumper=NoAliasDumper)) + frozen_yaml_path.write_text(dump(config)) frozen_yaml_path.chmod(0o400) Plan.logger.info(f"{yaml_path.name} frozen successfully") else: - yaml_path.write_text(dump(config, Dumper=NoAliasDumper)) + yaml_path.write_text(dump(config)) @staticmethod def Parse(plan_config_path: Path, cols_config_path: Path = None, diff --git a/openfl/native/native.py b/openfl/native/native.py index 7e3e7a55fc..09e60f04f4 100644 --- a/openfl/native/native.py +++ b/openfl/native/native.py @@ -24,7 +24,7 @@ WORKSPACE_PREFIX = os.path.join(os.path.expanduser('~'), '.local', 'workspace') -def setup_plan(save=True): +def setup_plan(log_level='CRITICAL'): """ Dump the plan with all defaults + overrides set. @@ -38,42 +38,20 @@ def setup_plan(save=True): plan_config = 'plan/plan.yaml' cols_config = 'plan/cols.yaml' data_config = 'plan/data.yaml' + + getLogger().setLevel(log_level) plan = Plan.Parse(plan_config_path=Path(plan_config), cols_config_path=Path(cols_config), - data_config_path=Path(data_config)) - if save: - Plan.Dump(Path(plan_config), plan.config) + data_config_path=Path(data_config), + resolve=False) + getLogger().setLevel('INFO') return plan -def get_plan(return_complete=False): - """ - Return the flattened dictionary associated with the plan. - - To read the output in a human readable format, we recommend interpreting it - as follows: - - ``` - print(json.dumps(fx.get_plan(), indent=4, sort_keys=True)) - ``` - - Args: - return_complete : bool (default=False) - By default will not print the default file locations for each of - the templates - - Returns: - plan : dict - flattened dictionary of the current plan - """ - getLogger().setLevel('CRITICAL') - - plan_config = setup_plan().config - - getLogger().setLevel('INFO') - - flattened_config = flatten_preserve_lists(plan_config, '.')[0] +def flatten(config, return_complete=False): + """Flatten nested config.""" + flattened_config = flatten_preserve_lists(config, '.')[0] if not return_complete: keys_to_remove = [ k for k, v in flattened_config.items() @@ -98,16 +76,18 @@ def update_plan(override_config): Returns: None """ - plan_path = 'plan/plan.yaml' - flat_plan_config = get_plan(return_complete=True) + plan = setup_plan() + flat_plan_config = flatten(plan.config, return_complete=True) for k, v in override_config.items(): if k in flat_plan_config: - flat_plan_config[k] = v logger.info(f'Updating {k} to {v}... ') else: - logger.info(f'Key {k} not found in plan. Ignoring... ') - plan_config = unflatten(flat_plan_config, '.') - Plan.Dump(Path(plan_path), plan_config) + # TODO: We probably need to validate the new key somehow + logger.warn(f'Did not find {k} in config. Make sure it should exist. Creating...') + flat_plan_config[k] = v + plan.config = unflatten(flat_plan_config, '.') + plan.resolve() + return plan def unflatten(config, separator='.'): @@ -236,19 +216,7 @@ def run_experiment(collaborator_dict, override_config={}): path.insert(0, str(work)) # Update the plan if necessary - if len(override_config) > 0: - update_plan(override_config) - - # TODO: Fix this implementation. The full plan parsing is reused here, - # but the model and data will be overwritten based on user specifications - plan_config = 'plan/plan.yaml' - cols_config = 'plan/cols.yaml' - data_config = 'plan/data.yaml' - - plan = Plan.Parse(plan_config_path=Path(plan_config), - cols_config_path=Path(cols_config), - data_config_path=Path(data_config)) - + plan = update_plan(override_config) # Overwrite plan values plan.authorized_cols = list(collaborator_dict) tensor_pipe = plan.get_tensor_pipe() @@ -296,3 +264,11 @@ def run_experiment(collaborator_dict, override_config={}): model.rebuild_model( rounds_to_train - 1, aggregator.last_tensor_dict, validation=True) return model + + +def get_plan(indent=4, sort_keys=True): + """Get string representation of current Plan.""" + import json + plan = setup_plan() + flat_plan_config = flatten(plan.config) + return json.dumps(flat_plan_config, indent=indent, sort_keys=sort_keys) diff --git a/openfl/utilities/types.py b/openfl/utilities/types.py index a9699246c7..1d752f4d72 100644 --- a/openfl/utilities/types.py +++ b/openfl/utilities/types.py @@ -9,3 +9,4 @@ TaskResultKey = namedtuple('TaskResultKey', ['task_name', 'owner', 'round_number']) Metric = namedtuple('Metric', ['name', 'value']) +LocalTensor = namedtuple('LocalTensor', ['col_name', 'tensor', 'weight']) diff --git a/setup.py b/setup.py index 8bafe8ec3b..f606b16b48 100644 --- a/setup.py +++ b/setup.py @@ -44,6 +44,7 @@ 'openfl-workspace', 'openfl-docker', 'openfl-tutorials', + 'openfl.component.aggregation_functions' ], include_package_data=True, install_requires=[

Simple aggregation schemes on aggregator: geometric mean + median It would be great to have not only **_Avg_** operator for tensor aggregation, but also few other simple operations like **_Geometric Mean_**, **_Median_**, etc.

@alexey-gruzdev, we do have some of this in place: I believe we interpret one of the flplan args as the aggregation function. If the name doesn't match our pre-defined strings, we assume it is a numpy function and invoke it. I'd like the end-result to be the same as others: you can declare arbitrary python classes in the flplan that provide the functionality, along with init kwargs, the function to invoke, and kwargs for that function. Essentially, our same generic task declaration method we use for the collaborators. This is one of the reasons Patrick and I have been talking about making the aggregator essentially "just another task runner". This would also have a nice advantage in that we wouldn't have to hard-code the logic that chooses which tensors are inputs for the aggregation function. Instead, it would use the same generic logic that determines the tensor input requirements for any task. @msheller that's great, if we have it in place - could you provide a simple example for the reference? Ideally, for better extendibility it would be great to have a python function to be passed to Python API entities & being used by OpenFL further. ``` def custom_aggregation(a: torch.Tensor, b: torch.Tensor) -> torch.Tensor: return (a + b)/2 fl_model = fx.run_experiment(collaborators, agg_function=custom_aggregation, override_config={'aggregator.settings.rounds_to_train':5}) ``` > you can declare arbitrary python classes in the flplan that provide the functionality, along with init kwargs, the function to invoke, and kwargs for that function Maybe I am wrong, however from Python user's perspective - is it a right place to define such kind of functionalities in FLPlan (.yaml)? @alexey-gruzdev This is a feature I added a few months ago. A limited set of custom aggregation types are supported at the task level. They can be called in the following way: ``` plan.yaml ... tasks: aggregated_model_validation: aggregation_type: median function: validate kwargs: apply: global batch_size: 32 metrics: - accuracy ... ``` and you can call it through the current Python API like this: ``` fl_model = fx.run_experiment(collaborators, agg_function=custom_aggregation, override_config={'aggregator.settings.rounds_to_train':5, 'tasks.aggregated_model_validation.aggregation_type': 'median'}) ``` These aggregation types are currently limited to simple numpy functions (see [here](https://github.com/intel/openfl/blob/develop/openfl/databases/tensor_db.py#L186-L192)). As both of you suggest, we should provide a way to define custom functions that can be applied to model weights and task metrics. There is likely a need to apply a specific aggregation function to certain task output, and possibly another for how to combine model weights. Because of this level of specificity my view is that the plan is the appropriate place to embed this information, but we could extend our existing python APIs so that the user has a cleaner way for expressing the function they want to use without modifying the plan directly.

2021-04-27T12:30:14

0.0

[]

securefederatedai/openfl

securefederatedai__openfl-28

be9ea6ebf16f80c7ed84a775e61269c36375e984

diff --git a/openfl/native/native.py b/openfl/native/native.py index d085335351..702116ad89 100644 --- a/openfl/native/native.py +++ b/openfl/native/native.py @@ -263,7 +263,7 @@ def run_experiment(collaborator_dict, override_config={}): rounds_to_train = plan.config['aggregator']['settings']['rounds_to_train'] tensor_dict, holdout_params = split_tensor_dict_for_holdouts( logger, - plan.runner_.get_tensor_dict(False) + model.get_tensor_dict(False) ) model_snap = utils.construct_model_proto(tensor_dict=tensor_dict, @@ -285,7 +285,7 @@ def run_experiment(collaborator_dict, override_config={}): # Create the collaborators collaborators = { collaborator: create_collaborator( - plan, collaborator, model, aggregator + plan, collaborator, collaborator_dict[collaborator], aggregator ) for collaborator in plan.authorized_cols } @@ -293,14 +293,9 @@ def run_experiment(collaborator_dict, override_config={}): for col in plan.authorized_cols: collaborator = collaborators[col] - model.set_data_loader(collaborator_dict[col].data_loader) - - if round_num != 0: - model.rebuild_model(round_num, model_states[col]) collaborator.run_simulation() - - model_states[col] = model.get_tensor_dict(with_opt_vars=True) + model_states[col] = collaborator.task_runner.get_tensor_dict(with_opt_vars=True) # Set the weights for the final model model.rebuild_model(

Same validation metric values across collaborators in Jupyter notebooks **Describe the bug** When launching the federated training in Jupyter notebooks of `openfl-tutorials` folder, I noticed that different collaborators achieve the same validation metric values. That could be possible if collaborators had the same data, but the data is randomly split. Looks like there is an issue of defining the Data Loader for each collaborator. **To Reproduce** Steps to reproduce the behavior: 1. Open `openfl-tutorials/Federated_PyTorch_UNET_Tutorial.ipynb` in Jupyter. 2. Execute all cells. **Expected behavior** Collaborators have different metrics due to random data split. **Screenshot** ![image](https://user-images.githubusercontent.com/76161256/110151259-53b6a080-7df1-11eb-86bc-7ff5f787a4e9.png)

2021-03-09T11:30:51

0.0

[]

TeamLEGWORK/LEGWORK

TeamLEGWORK__LEGWORK-112

3edb13b81d67d15977eead0a3e1849a4415f6fea

diff --git a/changelog.md b/changelog.md index f6cf0df..94da070 100644 --- a/changelog.md +++ b/changelog.md @@ -166,6 +166,10 @@ having two when we could just set `f_dom=2 f_orb` - Ensure `theta` is a co-latitude in `amplitude_modulation` ## 0.4.6 -*TW 13/04/22* +*TW 13/04/23* - [Issue [#109](https://github.com/TeamLEGWORK/LEGWORK/issues/109)] Upgrade `numpy` version - - Upgrade required to avoid error in `np.nan_to_num` usage - thanks to Jakob Stegmann for raising this \ No newline at end of file + - Upgrade required to avoid error in `np.nan_to_num` usage - thanks to Jakob Stegmann for raising this + +## 0.4.7 +*TW 12/10/23* +- [Issue [#111](https://github.com/TeamLEGWORK/LEGWORK/issues/111)] Allow `plot_sources_on_sc` to change the underlying sensitivity curve directly with `sc_vis_settings` diff --git a/legwork/_version.py b/legwork/_version.py index 3dd3d2d..a34b2f6 100644 --- a/legwork/_version.py +++ b/legwork/_version.py @@ -1,1 +1,1 @@ -__version__ = "0.4.6" +__version__ = "0.4.7" diff --git a/legwork/evol.py b/legwork/evol.py index ae64836..10092cf 100644 --- a/legwork/evol.py +++ b/legwork/evol.py @@ -15,7 +15,7 @@ 'determine_stationarity'] -@jit +@jit(nopython=True) def de_dt(e, times, beta, c_0): # pragma: no cover """Compute eccentricity time derivative diff --git a/legwork/source.py b/legwork/source.py index 000f2c7..1f70beb 100644 --- a/legwork/source.py +++ b/legwork/source.py @@ -1192,7 +1192,7 @@ def plot_source_variables(self, xstr, ystr=None, which_sources=None, return vis.plot_1D_dist(x=x[which_sources], weights=weights, **kwargs) def plot_sources_on_sc(self, snr_cutoff=0, fig=None, ax=None, show=True, - label="Stationary sources", **kwargs): # pragma: no cover + label="Stationary sources", sc_vis_settings={}, **kwargs): # pragma: no cover """Plot all sources in the class on the sensitivity curve Parameters @@ -1214,6 +1214,10 @@ def plot_sources_on_sc(self, snr_cutoff=0, fig=None, ax=None, show=True, label : `str` Label to use for the plotted points + sc_vis_settings : `dict` + Dictionary of parameters to pass to :meth:`~legwork.visualisation.plot_sensitivity_curve`, + e.g. {"y_quantity": "h_c"} will plot characteristic strain instead of ASD + **kwargs : `various` Keyword arguments to be passed to plotting functions @@ -1250,7 +1254,7 @@ def plot_sources_on_sc(self, snr_cutoff=0, fig=None, ax=None, show=True, weights = self.weights[stat] if self.weights is not None else None fig, ax = vis.plot_sources_on_sc(f_dom=f_dom, snr=self.snr[stat], weights=weights, snr_cutoff=snr_cutoff, show=show, fig=fig, ax=ax, - label=label, **self._sc_params, **kwargs) + label=label, sc_vis_settings=sc_vis_settings, **self._sc_params, **kwargs) # show warnings for evolving sources circ_evol = self.get_source_mask(circular=True, stationary=False) diff --git a/legwork/visualisation.py b/legwork/visualisation.py index 0bd367c..4a72ba6 100644 --- a/legwork/visualisation.py +++ b/legwork/visualisation.py @@ -374,7 +374,7 @@ def plot_sensitivity_curve(frequency_range=None, y_quantity="ASD", fig=None, ax= def plot_sources_on_sc(f_dom, snr, weights=None, snr_cutoff=0, t_obs="auto", instrument="LISA", custom_psd=None, L="auto", approximate_R=False, - confusion_noise="auto", fig=None, ax=None, show=True, **kwargs): + confusion_noise="auto", fig=None, ax=None, show=True, sc_vis_settings={}, **kwargs): """Overlay *stationary* sources on the LISA sensitivity curve. Each source is plotted at its max snr harmonic frequency such that that its height above the curve is @@ -429,6 +429,10 @@ def plot_sources_on_sc(f_dom, snr, weights=None, snr_cutoff=0, t_obs="auto", show : `boolean` Whether to immediately show the plot or only return the Figure and Axis + sc_vis_settings : `dict` + Dictionary of parameters to pass to :meth:`~legwork.visualisation.plot_sensitivity_curve`, + e.g. {"y_quantity": "h_c"} will plot characteristic strain instead of ASD + **kwargs : `various` This function is a wrapper on :func:`legwork.visualisation.plot_2D_dist` and each kwarg is passed directly to this function. For example, you can write `disttype="kde"` for a kde density plot @@ -446,7 +450,7 @@ def plot_sources_on_sc(f_dom, snr, weights=None, snr_cutoff=0, t_obs="auto", if fig is None or ax is None: fig, ax = plot_sensitivity_curve(show=False, t_obs=t_obs, instrument=instrument, L=L, custom_psd=custom_psd, approximate_R=approximate_R, - confusion_noise=confusion_noise) + confusion_noise=confusion_noise, **sc_vis_settings) # work out which binaries are above the cutoff detectable = snr > snr_cutoff @@ -455,12 +459,14 @@ def plot_sources_on_sc(f_dom, snr, weights=None, snr_cutoff=0, t_obs="auto", return fig, ax # calculate asd that makes it so height above curve is snr - asd = snr[detectable] * np.sqrt(psd.power_spectral_density(f_dom[detectable])) + asd = (snr[detectable] * np.sqrt(psd.power_spectral_density(f_dom[detectable]))).to(u.Hz**(-1/2)) + h_c = asd * np.sqrt(f_dom[detectable]) + use_h_c = ("y_quantity" in sc_vis_settings and sc_vis_settings["y_quantity"] == "h_c") # plot either a scatter or density plot of the detectable binaries ylims = ax.get_ylim() weights = weights[detectable] if weights is not None else None - fig, ax = plot_2D_dist(x=f_dom[detectable], y=asd.to(u.Hz**(-1/2)), weights=weights, + fig, ax = plot_2D_dist(x=f_dom[detectable], y=h_c if use_h_c else asd, weights=weights, fig=fig, ax=ax, show=False, **kwargs) ax.set_ylim(ylims)

y axis of the frequency versus ASD plot In the plot_sensitive_curve function, it provides an option of y-axis to be "h_c", while in the plot_source_on_sc function, there is no such an option. Sometimes, it would be necessary to have a such plot. Therefore, it would be good if the plot_source_on_sce function. Best wishes, Hailiang Chen

2023-10-12T15:36:57

0.0

[]

TeamLEGWORK/LEGWORK

TeamLEGWORK__LEGWORK-72

f95affca969255a21e4d84afa01170882e2d543e

diff --git a/changelog.md b/changelog.md index 7734470..0c4b755 100644 --- a/changelog.md +++ b/changelog.md @@ -76,4 +76,15 @@ Small changes to visualisation code and updates to tutorials/demos with the new ## 0.1.7 *TW 05/01/21* -- Link GitHub releases to Zenodo \ No newline at end of file +- Link GitHub releases to Zenodo + +## 0.2.0 +*TW 05/01/21* +- A couple of changes to how confusion noise is handled + - End user can now access confusion noise functions directly through `get_confusion_noise` + - Added confusion noise fit from Huang+20 to be used with TianQin + - Added confusion noise fit from Thiele+21 which is based on a WDWD population with a metallicity dependent binary fraction +- TianQin psd function now include the confusion noise +- Change defaults used in `source` and `psd` + - Often defaults for arm length, observation time and confusion noise were previously LISA related, LEGWORK now automatically works out the defaults based on what instrument is chosen +- Bug fix: in `visualisation` avoid mixing floats with Quantities when filling in a sensitivity curve diff --git a/legwork/_version.py b/legwork/_version.py index 283b03a..d3ec452 100644 --- a/legwork/_version.py +++ b/legwork/_version.py @@ -1,1 +1,1 @@ -__version__ = "0.1.7" \ No newline at end of file +__version__ = "0.2.0" diff --git a/legwork/psd.py b/legwork/psd.py index 5d2c7a3..6a36029 100644 --- a/legwork/psd.py +++ b/legwork/psd.py @@ -88,10 +88,11 @@ def lisa_psd(f, t_obs=4 * u.yr, L=2.5e9 * u.m, approximate_R=False, confusion_no approximate_R : `boolean` Whether to approximate the response function (default: no) - confusion_noise : `various` - Galactic confusion noise. Acceptable inputs are 'robson19' (the confusion noise from Robson+19), - `None` (don't include confusion noise) or a custom function that gives the confusion noise at each - frequency for a given mission length where it would be called by running `noise(f, t_obs)` + confusion_noise : `various` + Galactic confusion noise. Acceptable inputs are one of the values listed in + :meth:`legwork.psd.get_confusion_noise` or a custom function that gives the confusion noise at each + frequency for a given mission length where it would be called by running `noise(f, t_obs)` and return + a value with units of inverse Hertz Returns ------- @@ -116,23 +117,6 @@ def Poms(f): def Pacc(f): return (3e-15)**2 * (1 + (0.4e-3 / f)**2) * (1 + (f / (8e-3))**4) - # galactic confusion noise (Robson+ Eq. 14) - def Sc(f, t_obs): - - # parameters from Robson+ 2019 Table 1 - lengths = np.array([0.5, 1.0, 2.0, 4.0]) * u.yr - alpha = [0.133, 0.171, 0.165, 0.138] - beta = [243.0, 292.0, 299.0, -221.0] - kappa = [482.0, 1020.0, 611.0, 521.0] - gamma = [917.0, 1680.0, 1340.0, 1680.0] - fk = [2.58e-3, 2.15e-3, 1.73e-3, 1.13e-3] - - # find index of the closest length to inputted observation time - ind = np.abs(t_obs - lengths).argmin() - - return 9e-45 * f**(-7 / 3.) * np.exp(-f**(alpha[ind]) + beta[ind] * f * np.sin(kappa[ind] * f)) \ - * (1 + np.tanh(gamma[ind] * (fk[ind] - f))) - # calculate response function (either exactly or with approximation) fstar = (const.c / (2 * np.pi * L)).to(u.Hz).value if approximate_R: @@ -140,15 +124,11 @@ def Sc(f, t_obs): else: R = load_response_function(f, fstar) - # use confusion noise from Robson+19 - if confusion_noise == "robson19": - cn = Sc(f, t_obs) - # use a custom function for the confusion noise - elif confusion_noise is not None: - cn = confusion_noise(f, t_obs) - # don't include any confusion noise + # get confusion noise + if isinstance(confusion_noise, str) or confusion_noise is None: + cn = get_confusion_noise(f=f * u.Hz, t_obs=t_obs, model=confusion_noise).value else: - cn = np.zeros(np.shape(f)) if isinstance(f, (list, np.ndarray)) else 0.0 + cn = confusion_noise(f, t_obs).value L = L.to(u.m).value @@ -160,7 +140,7 @@ def Sc(f, t_obs): return psd / u.Hz -def tianqin_psd(f, L=np.sqrt(3) * 1e5 * u.km, t_obs=None, approximate_R=None, confusion_noise=None): +def tianqin_psd(f, L=np.sqrt(3) * 1e5 * u.km, t_obs=5 * u.yr, approximate_R=None, confusion_noise="huang20"): """Calculates the effective TianQin power spectral density sensitivity curve Using Eq. 13 from Huang+20, calculate the effective TianQin PSD for the sensitivity curve @@ -178,13 +158,16 @@ def tianqin_psd(f, L=np.sqrt(3) * 1e5 * u.km, t_obs=None, approximate_R=None, co Arm length t_obs : `float` - Ignored for this function + Observation time (default 5 years) approximate_R : `boolean` Ignored for this function - confusion_noise : `various` - Ignored for this function + confusion_noise : `various` + Galactic confusion noise. Acceptable inputs are one of the values listed in + :meth:`legwork.psd.get_confusion_noise` or a custom function that gives the confusion noise at each + frequency for a given mission length where it would be called by running `noise(f, t_obs)` and return + a value with units of inverse Hertz Returns ------- @@ -196,11 +179,19 @@ def tianqin_psd(f, L=np.sqrt(3) * 1e5 * u.km, t_obs=None, approximate_R=None, co Sx = 1e-24 * u.m**2 * u.Hz**(-1) psd = 10 / (3 * L**2) * (4 * Sa / (2 * np.pi * f)**4 * (1 + (1e-4 * u.Hz / f)) + Sx) \ * (1 + 0.6 * (f / fstar)**2) + + # get confusion noise + if isinstance(confusion_noise, str) or confusion_noise is None: + cn = get_confusion_noise(f=f, t_obs=t_obs, model=confusion_noise) + else: + cn = confusion_noise(f, t_obs) + psd += cn + return psd.to(u.Hz**(-1)) -def power_spectral_density(f, instrument="LISA", custom_psd=None, t_obs=4 * u.yr, L=None, - approximate_R=False, confusion_noise="robson19"): +def power_spectral_density(f, instrument="LISA", custom_psd=None, t_obs="auto", L="auto", + approximate_R=False, confusion_noise="auto"): """Calculates the effective power spectral density for all instruments. Parameters @@ -216,7 +207,7 @@ def power_spectral_density(f, instrument="LISA", custom_psd=None, t_obs=4 * u.yr even if it ignores some. t_obs : `float` - Observation time (default 4 years) + Observation time (default 4 years for LISA and 5 years for TianQin) L : `float` LISA arm length in metres @@ -224,10 +215,12 @@ def power_spectral_density(f, instrument="LISA", custom_psd=None, t_obs=4 * u.yr approximate_R : `boolean` Whether to approximate the response function (default: no) - confusion_noise : `various` - Galactic confusion noise. Acceptable inputs are 'robson19' (the confusion noise from Robson+19), - `None` (don't include confusion noise) or a custom function that gives the confusion noise at each - frequency for a given mission length where it would be called by running `noise(f, t_obs)` + confusion_noise : `various` + Galactic confusion noise. Acceptable inputs are either one of the values listed in + :meth:`legwork.psd.get_confusion_noise`, "auto" (automatically selects confusion noise based on + `instrument` - 'robson19' if LISA and 'huang20' if TianQin), or a custom function that gives the + confusion noise at each frequency for a given mission length where it would be called by running + `noise(f, t_obs)` and return a value with units of inverse Hertz Returns ------- @@ -235,12 +228,20 @@ def power_spectral_density(f, instrument="LISA", custom_psd=None, t_obs=4 * u.yr Effective power strain spectral density """ if instrument == "LISA": - if L is None: - L = 2.5e9 * u.m + # update any auto values to be instrument specific + L = 2.5e9 * u.m if L == "auto" else L + confusion_noise = "robson19" if confusion_noise == "auto" else confusion_noise + t_obs = 4 * u.yr if t_obs == "auto" else t_obs + + # calculate psd psd = lisa_psd(f=f, L=L, t_obs=t_obs, approximate_R=approximate_R, confusion_noise=confusion_noise) elif instrument == "TianQin": - if L is None: - L = np.sqrt(3) * 1e5 * u.km + # update any auto values to be instrument specific + L = np.sqrt(3) * 1e5 * u.km if L == "auto" else L + confusion_noise = "huang20" if confusion_noise == "auto" else confusion_noise + t_obs = 5 * u.yr if t_obs == "auto" else t_obs + + # calculate psd psd = tianqin_psd(f=f, L=L, t_obs=t_obs, approximate_R=approximate_R, confusion_noise=confusion_noise) elif instrument == "custom": psd = custom_psd(f=f, L=L, t_obs=t_obs, approximate_R=approximate_R, confusion_noise=confusion_noise) @@ -248,3 +249,160 @@ def power_spectral_density(f, instrument="LISA", custom_psd=None, t_obs=4 * u.yr raise ValueError("instrument: `{}` not recognised".format(instrument)) return psd + + +def get_confusion_noise_robson19(f, t_obs=4 * u.yr): + """Calculate the confusion noise using the model from Robson+19 Eq. 14 and Table 1 + + Also note that this fit is designed based on LISA sensitivity and so it is likely not sensible to apply + it to TianQin or other missions. + + Parameters + ---------- + f : `float/array` + Frequencies at which to calculate the confusion noise, must have units of frequency + t_obs : `float`, optional + Mission length, parameters are defined for 0.5, 1, 2 and 4 years, the closest mission length to the + one inputted will be used. By default 4 years. + + Returns + ------- + confusion_noise : `float/array` + The confusion noise at each frequency + """ + # erase the units for speed + f = f.to(u.Hz).value + + # parameters from Robson+ 2019 Table 1 + lengths = np.array([0.5, 1.0, 2.0, 4.0]) * u.yr + alpha = [0.133, 0.171, 0.165, 0.138] + beta = [243.0, 292.0, 299.0, -221.0] + kappa = [482.0, 1020.0, 611.0, 521.0] + gamma = [917.0, 1680.0, 1340.0, 1680.0] + fk = [2.58e-3, 2.15e-3, 1.73e-3, 1.13e-3] + + # find index of the closest length to inputted observation time + ind = np.abs(t_obs - lengths).argmin() + + confusion_noise = 9e-45 * f**(-7 / 3.) * np.exp(-f**(alpha[ind]) + beta[ind] + * f * np.sin(kappa[ind] * f))\ + * (1 + np.tanh(gamma[ind] * (fk[ind] - f))) * u.Hz**(-1) + return confusion_noise + + +def get_confusion_noise_huang20(f, t_obs=5 * u.yr): + """Calculate the confusion noise using the model from Huang+20 Table II. Note that we set the confusion + noise to be exactly 0 outside of the range [1e-4, 1] Hz as the fits are not designed to be used outside + of this range. + + Also note that this fit is designed based on TianQin sensitivity and so it is likely not sensible to apply + it to LISA or other missions. + + Parameters + ---------- + f : `float/array` + Frequencies at which to calculate the confusion noise, must have units of frequency + t_obs : `float`, optional + Mission length, parameters are defined for 0.5, 1, 2, 4 and 5 years, the closest mission length to the + one inputted will be used. By default 5 years. + + Returns + ------- + confusion_noise : `float/array` + The confusion noise at each frequency + """ + + # define the mission lengths and corresponding coefficients + lengths = np.array([0.5, 1.0, 2.0, 4.0, 5.0]) * u.yr + coefficients = np.array([ + [-18.6, -18.6, -18.6, -18.6, -18.6], + [-1.22, -1.13, -1.45, -1.43, -1.51], + [0.009, -0.945, 0.315, -0.687, -0.71], + [-1.87, -1.02, -1.19, 0.24, -1.13], + [0.65, 4.05, -4.48, -0.15, -0.83], + [3.6, -4.5, 10.8, -1.8, 13.2], + [-4.6, -0.5, -9.4, -3.2, -19.1] + ]) + + # find the nearest mission length + ind = np.abs(t_obs - lengths).argmin() + + # start with no confusion noise and add each coefficient using Huang+20 Table 1 + f = f.to(u.Hz).value + confusion_noise = np.zeros_like(f) + x = np.log10(f / 1e-3) + for i in range(len(coefficients)): + xi = x**i + ai = coefficients[i][ind] + confusion_noise += ai * xi + confusion_noise = 10**(confusion_noise) * u.Hz**(-1/2) + + # remove confusion noise outside of TianQin regime (fit doesn't apply outside of regime) + confusion_noise[np.logical_or(f < 1e-4, f > 1e0)] = 0.0 * u.Hz**(-1/2) + + # square the result as Huang+20 given the sensitivity not psd + return confusion_noise**2 + + +def get_confusion_noise_thiele21(f): + """Calculate the confusion noise using the model from Thiele+20 Eq. 16 and Table 1. This fit uses a + metallicity-dependent binary fraction. + + Note: This fit only applies to LISA and only when the mission length is 4 years. + + Parameters + ---------- + f : `float/array` + Frequencies at which to calculate the confusion noise, must have units of frequency + + Returns + ------- + confusion_noise : `float/array` + The confusion noise at each frequency + """ + x = np.log10(f.to(u.Hz).value) + t_obs = 4 * u.yr + coefficients = [-540.315450, -554.061115, -233.748834, -44.021504, -3.112634] + confusion_noise = 10**np.poly1d(coefficients[::-1])(x) * t_obs.to(u.s).value + return confusion_noise * u.Hz**(-1) + + +def get_confusion_noise(f, model, t_obs="auto"): + """Calculate the confusion noise for a particular model + + Parameters + ---------- + f : `float/array` + Frequencies at which to calculate the confusion noise, must have units of frequency + model : str, optional + Which model to use for the confusion noise. Must be one of 'robson19', 'huang20', 'thiele21' or None. + t_obs : `float`, optional + Mission length. Default is 4 years for robson19 and thiele21 and 5 years for huang20. + + Returns + ------- + confusion_noise : `float/array` + The confusion noise at each frequency. + + Raises + ------ + ValueError + When a model other than those defined above is used. + """ + if model == "robson19": + t_obs = 4 * u.yr if t_obs == "auto" else t_obs + return get_confusion_noise_robson19(f=f, t_obs=t_obs) + elif model == "huang20": + t_obs = 5 * u.yr if t_obs == "auto" else t_obs + return get_confusion_noise_huang20(f=f, t_obs=t_obs) + elif model == "thiele21": + if t_obs == 4 * u.yr or t_obs == "auto": + return get_confusion_noise_thiele21(f=f) + else: + error = "Invalid mission length: Thiele+21 confusion noise is only fit for `t_obs=4*u.yr`" + raise ValueError(error) + elif model is None: + f = f.to(u.Hz).value + return np.zeros_like(f) * u.Hz**(-1) if isinstance(f, (list, np.ndarray)) else 0.0 * u.Hz**(-1) + else: + raise ValueError("confusion noise model: `{}` not recognised".format(model)) diff --git a/legwork/source.py b/legwork/source.py index c33afd6..0e2ec42 100644 --- a/legwork/source.py +++ b/legwork/source.py @@ -76,7 +76,7 @@ class Source(): sc_params : `dict` Parameters for interpolated sensitivity curve. Include any of ``instrument``, ``custom_psd``, ``t_obs``, ``L``, ``approximate_R`` and ``confusion_noise``. Default values are: "LISA", None, - 4 years, 2.5e9, 19.09e-3, False and 'robson19'. + "auto", "auto", False and "auto". Attributes ---------- @@ -170,10 +170,10 @@ def __init__(self, m_1, m_2, ecc, dist, n_proc=1, f_orb=None, a=None, position=N default_sc_params = { "instrument": "LISA", "custom_psd": None, - "t_obs": 4 * u.yr, - "L": 2.5e9 * u.m, + "t_obs": "auto", + "L": "auto", "approximate_R": False, - "confusion_noise": 'robson19' + "confusion_noise": 'auto' } default_sc_params.update(sc_params) self._sc_params = default_sc_params @@ -347,10 +347,10 @@ def update_sc_params(self, sc_params): default_sc_params = { "instrument": "LISA", "custom_psd": None, - "t_obs": 4 * u.yr, - "L": 2.5e9 * u.m, + "t_obs": "auto", + "L": "auto", "approximate_R": False, - "confusion_noise": "robson19" + "confusion_noise": "auto" } if sc_params is not None: default_sc_params.update(sc_params) @@ -358,7 +358,7 @@ def update_sc_params(self, sc_params): self._sc_params = default_sc_params self.set_sc() - def get_source_mask(self, circular=None, stationary=None, t_obs=4 * u.yr): + def get_source_mask(self, circular=None, stationary=None, t_obs=None): """Produce a mask of the sources. Create a mask based on whether binaries are circular or eccentric and stationary or evolving. @@ -373,7 +373,7 @@ def get_source_mask(self, circular=None, stationary=None, t_obs=4 * u.yr): ``None`` means either, ``True`` means only stationary binaries and ``False`` means only evolving t_obs : `float` - Observation time + Observation time, default is value from `self._sc_params` Returns ------- @@ -392,6 +392,9 @@ def get_source_mask(self, circular=None, stationary=None, t_obs=4 * u.yr): if stationary is None: stat_mask = np.repeat(True, self.n_sources) elif stationary is True or stationary is False: + t_obs = self._sc_params["t_obs"] if t_obs is None else t_obs + if t_obs == "auto": + t_obs = 4 * u.yr if self._sc_params["instrument"] == "LISA" else 5 * u.yr stat_mask = evol.determine_stationarity(m_c=self.m_c, f_orb_i=self.f_orb, t_evol=t_obs, ecc_i=self.ecc, stat_tol=self.stat_tol) if stationary is False: @@ -603,14 +606,14 @@ def get_snr(self, t_obs=None, instrument=None, custom_psd=None, n_step=100, verbose=verbose) return snr - def get_snr_stationary(self, t_obs=4 * u.yr, instrument="LISA", custom_psd=None, which_sources=None, + def get_snr_stationary(self, t_obs=None, instrument=None, custom_psd=None, which_sources=None, verbose=False): """Computes the SNR assuming a stationary binary Parameters ---------- t_obs : `array` - Observation duration (default: 4 years) + Observation duration (default: follow sc_params) instrument : `{{ 'LISA', 'TianQin', 'custom' }}` Instrument to observe with. If 'custom' then ``custom_psd`` must be supplied. @@ -633,6 +636,11 @@ def get_snr_stationary(self, t_obs=4 * u.yr, instrument="LISA", custom_psd=None, if which_sources is None: which_sources = np.repeat(True, self.n_sources) + instrument = self._sc_params["instrument"] if instrument is None else instrument + t_obs = self._sc_params["t_obs"] if t_obs is None else t_obs + if t_obs == "auto": + t_obs = 4 * u.yr if instrument == "LISA" else 5 * u.yr + insp_sources = np.logical_and(which_sources, np.logical_not(self.merged)) snr = np.zeros(self.n_sources) e_mask = np.logical_and(self.ecc > self.ecc_tol, insp_sources) @@ -703,14 +711,14 @@ def get_snr_stationary(self, t_obs=4 * u.yr, instrument="LISA", custom_psd=None, return snr[which_sources] - def get_snr_evolving(self, t_obs, instrument="LISA", custom_psd=None, n_step=100, which_sources=None, + def get_snr_evolving(self, t_obs=None, instrument=None, custom_psd=None, n_step=100, which_sources=None, verbose=False): """Computes the SNR assuming an evolving binary Parameters ---------- t_obs : `array` - Observation duration (default: 4 years) + Observation duration (default: follow sc_params) instrument : `{{ 'LISA', 'TianQin', 'custom' }}` Instrument to observe with. If 'custom' then ``custom_psd`` must be supplied. @@ -738,6 +746,11 @@ def get_snr_evolving(self, t_obs, instrument="LISA", custom_psd=None, n_step=100 if which_sources is None: which_sources = np.repeat(True, self.n_sources) + instrument = self._sc_params["instrument"] if instrument is None else instrument + t_obs = self._sc_params["t_obs"] if t_obs is None else t_obs + if t_obs == "auto": + t_obs = 4 * u.yr if instrument == "LISA" else 5 * u.yr + insp_sources = np.logical_and(which_sources, np.logical_not(self.merged)) e_mask = np.logical_and(self.ecc > self.ecc_tol, insp_sources) c_mask = np.logical_and(self.ecc <= self.ecc_tol, insp_sources) @@ -1125,7 +1138,7 @@ def plot_sources_on_sc(self, snr_cutoff=0, fig=None, ax=None, show=True, **kwarg class Stationary(Source): """Subclass for sources that are stationary""" - def get_snr(self, t_obs=4 * u.yr, instrument="LISA", custom_psd=None, verbose=False): + def get_snr(self, t_obs=None, instrument=None, custom_psd=None, verbose=False): self.snr = self.get_snr_stationary(t_obs=t_obs, instrument=instrument, custom_psd=custom_psd, verbose=verbose) return self.snr @@ -1134,7 +1147,7 @@ def get_snr(self, t_obs=4 * u.yr, instrument="LISA", custom_psd=None, verbose=Fa class Evolving(Source): """Subclass for sources that are evolving""" - def get_snr(self, t_obs=4 * u.yr, instrument="LISA", custom_psd=None, n_step=100, verbose=False): + def get_snr(self, t_obs=None, instrument=None, custom_psd=None, n_step=100, verbose=False): self.snr = self.get_snr_evolving(t_obs=t_obs, n_step=n_step, instrument=instrument, custom_psd=custom_psd, verbose=verbose) return self.snr diff --git a/legwork/visualisation.py b/legwork/visualisation.py index aa5b7b4..90b510a 100644 --- a/legwork/visualisation.py +++ b/legwork/visualisation.py @@ -311,7 +311,7 @@ def plot_sensitivity_curve(frequency_range=None, y_quantity="ASD", fig=None, ax= # plot the curve and fill if needed ax.loglog(frequency_range, noise_amplitude, color=color, label=label, linewidth=linewidth) if fill: - ax.fill_between(frequency_range, 0, noise_amplitude, alpha=alpha, color=color) + ax.fill_between(frequency_range, 0 * noise_amplitude.unit, noise_amplitude, alpha=alpha, color=color) # adjust labels, sizes and frequency limits to plot is flush to the edges ax.set_xlabel(r'Frequency [$\rm Hz$]') @@ -329,9 +329,9 @@ def plot_sensitivity_curve(frequency_range=None, y_quantity="ASD", fig=None, ax= return fig, ax -def plot_sources_on_sc_circ_stat(f_orb, h_0_2, snr, weights=None, snr_cutoff=0, t_obs=4 * u.yr, - instrument="LISA", custom_psd=None, L=2.5e9 * u.m, approximate_R=False, - confusion_noise=True, fig=None, ax=None, show=True, **kwargs): +def plot_sources_on_sc_circ_stat(f_orb, h_0_2, snr, weights=None, snr_cutoff=0, t_obs="auto", + instrument="LISA", custom_psd=None, L="auto", approximate_R=False, + confusion_noise="auto", fig=None, ax=None, show=True, **kwargs): """Overlay circular/stationary sources on the LISA sensitivity curve. Each source is plotted at its gravitational wave frequency (n = 2) such that its height above the curve @@ -362,18 +362,20 @@ def plot_sources_on_sc_circ_stat(f_orb, h_0_2, snr, weights=None, snr_cutoff=0, even if it ignores some. t_obs : `float` - Observation time (default 4 years) + Observation time (default "auto") L : `float` - LISA arm length in metres + Arm length approximate_R : `boolean` Whether to approximate the response function (default: no) - confusion_noise : `various` - Galactic confusion noise. Acceptable inputs are 'robson19' (the confusion noise from Robson+19), - `None` (don't include confusion noise) or a custom function that gives the confusion noise at each - frequency for a given mission length where it would be called by running `noise(f, t_obs)` + confusion_noise : `various` + Galactic confusion noise. Acceptable inputs are either one of the values listed in + :meth:`legwork.psd.get_confusion_noise`, "auto" (automatically selects confusion noise based on + `instrument` - 'robson19' if LISA and 'huang20' if TianQin), or a custom function that gives the + confusion noise at each frequency for a given mission length where it would be called by running + `noise(f, t_obs)` and return a value with units of inverse Hertz fig: `matplotlib Figure` A figure on which to plot the distribution. Both `ax` and `fig` must be supplied for either to be used @@ -425,9 +427,9 @@ def plot_sources_on_sc_circ_stat(f_orb, h_0_2, snr, weights=None, snr_cutoff=0, return fig, ax -def plot_sources_on_sc_ecc_stat(f_dom, snr, weights=None, snr_cutoff=0, t_obs=4 * u.yr, - instrument="LISA", custom_psd=None, L=2.5e9 * u.m, approximate_R=False, - confusion_noise=True, fig=None, ax=None, show=True, **kwargs): +def plot_sources_on_sc_ecc_stat(f_dom, snr, weights=None, snr_cutoff=0, t_obs="auto", + instrument="LISA", custom_psd=None, L="auto", approximate_R=False, + confusion_noise="auto", fig=None, ax=None, show=True, **kwargs): """Overlay eccentric/stationary sources on the LISA sensitivity curve. Each source is plotted at its max snr harmonic frequency such that that its height above the curve is @@ -455,18 +457,20 @@ def plot_sources_on_sc_ecc_stat(f_dom, snr, weights=None, snr_cutoff=0, t_obs=4 even if it ignores some. t_obs : `float` - Observation time (default 4 years) + Observation time (default auto) L : `float` - LISA arm length in metres + Arm length in metres approximate_R : `boolean` Whether to approximate the response function (default: no) - confusion_noise : `various` - Galactic confusion noise. Acceptable inputs are 'robson19' (the confusion noise from Robson+19), - `None` (don't include confusion noise) or a custom function that gives the confusion noise at each - frequency for a given mission length where it would be called by running `noise(f, t_obs)` + confusion_noise : `various` + Galactic confusion noise. Acceptable inputs are either one of the values listed in + :meth:`legwork.psd.get_confusion_noise`, "auto" (automatically selects confusion noise based on + `instrument` - 'robson19' if LISA and 'huang20' if TianQin), or a custom function that gives the + confusion noise at each frequency for a given mission length where it would be called by running + `noise(f, t_obs)` and return a value with units of inverse Hertz fig: `matplotlib Figure` A figure on which to plot the distribution. Both `ax` and `fig` must be supplied for either to be used

Visualisation function mixes quantity with dimensionless constants This doesn't always produce an error, it seems to depend on your version of astropy/python but either way needs to be fixed. I'll sort it out in the next PR, the exact function is `plot_sensitivity_curve` when `fill=True`. A quick fix for this in the meantime is to include this at the top of any files that use this plotting function ``` from astropy.visualization import quantity_support quantity_support() ```

2022-01-05T17:55:39

0.0

[]

PyFilesystem/pyfilesystem2

PyFilesystem__pyfilesystem2-518

a6ea045e766c76bae2e2fde19294c8275773ffde

diff --git a/CHANGELOG.md b/CHANGELOG.md index ab0d827c..8a3f73be 100644 --- a/CHANGELOG.md +++ b/CHANGELOG.md @@ -13,6 +13,8 @@ and this project adheres to [Semantic Versioning](http://semver.org/). - Support more lenient usernames and group names in FTP servers ([#507](https://github.com/PyFilesystem/pyfilesystem2/pull/507)). Closes [#506](https://github.com/PyFilesystem/pyfilesystem2/issues/506). +- Removed dependency on pytz ([#518](https://github.com/PyFilesystem/pyfilesystem2/pull/518)). + Closes [#516](https://github.com/PyFilesystem/pyfilesystem2/issues/518). ### Fixed diff --git a/fs/_ftp_parse.py b/fs/_ftp_parse.py index 0061f875..42c0720e 100644 --- a/fs/_ftp_parse.py +++ b/fs/_ftp_parse.py @@ -3,17 +3,20 @@ from __future__ import unicode_literals import unicodedata -import datetime import re import time +from datetime import datetime -from pytz import UTC +try: + from datetime import timezone +except ImportError: + from ._tzcompat import timezone # type: ignore from .enums import ResourceType from .permissions import Permissions -EPOCH_DT = datetime.datetime.fromtimestamp(0, UTC) +EPOCH_DT = datetime.fromtimestamp(0, timezone.utc) RE_LINUX = re.compile( @@ -98,7 +101,7 @@ def _parse_time(t, formats): day = _t.tm_mday hour = _t.tm_hour minutes = _t.tm_min - dt = datetime.datetime(year, month, day, hour, minutes, tzinfo=UTC) + dt = datetime(year, month, day, hour, minutes, tzinfo=timezone.utc) epoch_time = (dt - EPOCH_DT).total_seconds() return epoch_time diff --git a/fs/_tzcompat.py b/fs/_tzcompat.py new file mode 100644 index 00000000..282859b2 --- /dev/null +++ b/fs/_tzcompat.py @@ -0,0 +1,30 @@ +"""Compatibility shim for python2's lack of datetime.timezone. + +This is the example code from the Python 2 documentation: +https://docs.python.org/2.7/library/datetime.html#tzinfo-objects +""" + +from datetime import tzinfo, timedelta + + +ZERO = timedelta(0) + + +class UTC(tzinfo): + """UTC""" + + def utcoffset(self, dt): + return ZERO + + def tzname(self, dt): + return "UTC" + + def dst(self, dt): + return ZERO + + +utc = UTC() + + +class timezone: + utc = utc diff --git a/fs/time.py b/fs/time.py index cdf06061..52a59a77 100644 --- a/fs/time.py +++ b/fs/time.py @@ -7,17 +7,16 @@ import typing from calendar import timegm from datetime import datetime -from pytz import UTC, timezone + +try: + from datetime import timezone +except ImportError: + from ._tzcompat import timezone # type: ignore if typing.TYPE_CHECKING: from typing import Optional -utcfromtimestamp = datetime.utcfromtimestamp -utclocalize = UTC.localize -GMT = timezone("GMT") - - def datetime_to_epoch(d): # type: (datetime) -> int """Convert datetime to epoch.""" @@ -39,4 +38,6 @@ def epoch_to_datetime(t): # noqa: D103 def epoch_to_datetime(t): # type: (Optional[int]) -> Optional[datetime] """Convert epoch time to a UTC datetime.""" - return utclocalize(utcfromtimestamp(t)) if t is not None else None + if t is None: + return None + return datetime.fromtimestamp(t, tz=timezone.utc) diff --git a/setup.cfg b/setup.cfg index d6aa8943..06dac8f0 100644 --- a/setup.cfg +++ b/setup.cfg @@ -43,7 +43,6 @@ setup_requires = setuptools >=38.3.0 install_requires = appdirs~=1.4.3 - pytz setuptools six ~=1.10 enum34 ~=1.1.6 ; python_version < '3.4'

Move away from pytz pytz is known for doing funky things that aren't really compatible with the rest of the python ecosystem and in some cases simply don't make sense. Their timezone objects aren't even compatible with actual Python tzinfo objects. Here's a discussion from matplotlib where they decided to drop it in favour of dateutil, which has also been recommended by the python documentation for years now: https://github.com/matplotlib/matplotlib/issues/10443/ Alternatively (if you're willing to drop support for Python 3.8 and below), Python 3.9 has a new built-in zoneinfo module in the stdlib that provides the common use case of just loading timezone objects by name with accurate timezone info.

This article provides a great writeup of why pytz is dangerous: https://blog.ganssle.io/articles/2018/03/pytz-fastest-footgun.html > This article provides a great writeup of why pytz is dangerous: > > https://blog.ganssle.io/articles/2018/03/pytz-fastest-footgun.html Interesting. `pytz` doesn't [appear to be used](https://github.com/PyFilesystem/pyfilesystem2/search?q=pytz) in too many places in PyFilesystem, so do you wanna create a suitable PR? Sure!

2022-01-28T06:21:03

0.0

[]

PyFilesystem/pyfilesystem2

PyFilesystem__pyfilesystem2-475

b88f43204960d69db6efb96e01f52b4f9c9c551e

diff --git a/CHANGELOG.md b/CHANGELOG.md index 62cc085c..f7b98bc1 100644 --- a/CHANGELOG.md +++ b/CHANGELOG.md @@ -40,6 +40,8 @@ and this project adheres to [Semantic Versioning](http://semver.org/). - `WrapCachedDir.isdir` and `WrapCachedDir.isfile` raising a `ResourceNotFound` error on non-existing path ([#470](https://github.com/PyFilesystem/pyfilesystem2/pull/470)). - `FTPFS` not listing certain entries with sticky/SUID/SGID permissions set by Linux server ([#473](https://github.com/PyFilesystem/pyfilesystem2/pull/473)). Closes [#451](https://github.com/PyFilesystem/pyfilesystem2/issues/451). +- `scandir` iterator not being closed explicitly in `OSFS.scandir`, occasionally causing a `ResourceWarning` + to be thrown. Closes [#311](https://github.com/PyFilesystem/pyfilesystem2/issues/311). ## [2.4.12] - 2021-01-14 diff --git a/fs/osfs.py b/fs/osfs.py index 3b35541c..5beb16bf 100644 --- a/fs/osfs.py +++ b/fs/osfs.py @@ -475,6 +475,7 @@ def _scandir(self, path, namespaces=None): # type: (Text, Optional[Collection[Text]]) -> Iterator[Info] self.check() namespaces = namespaces or () + requires_stat = not {"details", "stat", "access"}.isdisjoint(namespaces) _path = self.validatepath(path) if _WINDOWS_PLATFORM: sys_path = os.path.join( @@ -483,39 +484,47 @@ def _scandir(self, path, namespaces=None): else: sys_path = self._to_sys_path(_path) # type: ignore with convert_os_errors("scandir", path, directory=True): - for dir_entry in scandir(sys_path): - info = { - "basic": { - "name": fsdecode(dir_entry.name), - "is_dir": dir_entry.is_dir(), - } - } - if "details" in namespaces: - stat_result = dir_entry.stat() - info["details"] = self._make_details_from_stat(stat_result) - if "stat" in namespaces: - stat_result = dir_entry.stat() - info["stat"] = { - k: getattr(stat_result, k) - for k in dir(stat_result) - if k.startswith("st_") - } - if "lstat" in namespaces: - lstat_result = dir_entry.stat(follow_symlinks=False) - info["lstat"] = { - k: getattr(lstat_result, k) - for k in dir(lstat_result) - if k.startswith("st_") + scandir_iter = scandir(sys_path) + try: + for dir_entry in scandir_iter: + info = { + "basic": { + "name": fsdecode(dir_entry.name), + "is_dir": dir_entry.is_dir(), + } } - if "link" in namespaces: - info["link"] = self._make_link_info( - os.path.join(sys_path, dir_entry.name) - ) - if "access" in namespaces: - stat_result = dir_entry.stat() - info["access"] = self._make_access_from_stat(stat_result) - - yield Info(info) + if requires_stat: + stat_result = dir_entry.stat() + if "details" in namespaces: + info["details"] = self._make_details_from_stat( + stat_result + ) + if "stat" in namespaces: + info["stat"] = { + k: getattr(stat_result, k) + for k in dir(stat_result) + if k.startswith("st_") + } + if "access" in namespaces: + info["access"] = self._make_access_from_stat( + stat_result + ) + if "lstat" in namespaces: + lstat_result = dir_entry.stat(follow_symlinks=False) + info["lstat"] = { + k: getattr(lstat_result, k) + for k in dir(lstat_result) + if k.startswith("st_") + } + if "link" in namespaces: + info["link"] = self._make_link_info( + os.path.join(sys_path, dir_entry.name) + ) + + yield Info(info) + finally: + if sys.version_info >= (3, 6): + scandir_iter.close() else:

ResourceWarning: unclosed scandir iterator Getting this warning under Python 3.7 for scandir(): /home/ajung/.buildout/eggs/fs-2.4.8-py3.7.egg/fs/osfs.py:521: ResourceWarning: unclosed scandir iterator <posix.ScandirIterator object at 0x7f61d00596b0> yield Info(info)

Do you have some code to reproduce this? This will happen if you have an iterator to scandir that isn't full exhausted when it goes out of scope. It's not immediately obvious why it wouldn't close the native scandir iterator at the same time.

2021-03-27T03:21:07

0.0

[]

PyFilesystem/pyfilesystem2

PyFilesystem__pyfilesystem2-474

60b0a6347e3bbcee4a2a6261a5e1b3166c45e192

diff --git a/CHANGELOG.md b/CHANGELOG.md index aacded93..62cc085c 100644 --- a/CHANGELOG.md +++ b/CHANGELOG.md @@ -13,6 +13,8 @@ and this project adheres to [Semantic Versioning](http://semver.org/). - Added FTP over TLS (FTPS) support to FTPFS. Closes [#437](https://github.com/PyFilesystem/pyfilesystem2/issues/437), [#449](https://github.com/PyFilesystem/pyfilesystem2/pull/449). +- `PathError` now supports wrapping an exception using the `exc` argument. + Closes [#453](https://github.com/PyFilesystem/pyfilesystem2/issues/453). ### Changed diff --git a/fs/errors.py b/fs/errors.py index 25625e28..2448c7a6 100644 --- a/fs/errors.py +++ b/fs/errors.py @@ -139,13 +139,14 @@ class PathError(FSError): default_message = "path '{path}' is invalid" - def __init__(self, path, msg=None): # noqa: D107 - # type: (Text, Optional[Text]) -> None + def __init__(self, path, msg=None, exc=None): # noqa: D107 + # type: (Text, Optional[Text], Optional[Exception]) -> None self.path = path + self.exc = exc super(PathError, self).__init__(msg=msg) def __reduce__(self): - return type(self), (self.path, self._msg) + return type(self), (self.path, self._msg, self.exc) class NoSysPath(PathError):

convert_os_errors error when converting errno.ENAMETOOLONG to PathError Using convert_os_errors to convert errno.ENAMETOOLONG to PathError results in unexpected keyword arguement 'exc' error. Seems that [this line](https://github.com/PyFilesystem/pyfilesystem2/blob/8784fd61797fac654511f16fd50c38c8a1675f16/fs/error_tools.py#L89) will set exc=exc_value however PathError does not take in exc parameter. The following example code: ``` import errno from fs.error_tools import convert_os_errors from fs import errors as fserrors try: from fs import open_fs f = open_fs('/tmp') f.open('x'*500, mode='w') except OSError as e: with convert_os_errors("foo", "test"): raise e ``` results in: TypeError: __init__() got an unexpected keyword argument 'exc'

2021-03-27T02:24:49

0.0

[]

PyFilesystem/pyfilesystem2

PyFilesystem__pyfilesystem2-473

4efe083ebb1a43aafd0e171b3b20d6b546c86756

diff --git a/CHANGELOG.md b/CHANGELOG.md index c99ffcc3..aacded93 100644 --- a/CHANGELOG.md +++ b/CHANGELOG.md @@ -25,7 +25,7 @@ and this project adheres to [Semantic Versioning](http://semver.org/). test suites. - `FSTestCases` now builds the large data required for `upload` and `download` tests only once in order to reduce the total testing time. -- `MemoryFS.move` and `MemoryFS.movedir` will now avoid copying data. +- `MemoryFS.move` and `MemoryFS.movedir` will now avoid copying data. Closes [#452](https://github.com/PyFilesystem/pyfilesystem2/issues/452). ### Fixed @@ -36,6 +36,8 @@ and this project adheres to [Semantic Versioning](http://semver.org/). - Avoid creating a new connection on every call of `FTPFS.upload`. Closes [#455](https://github.com/PyFilesystem/pyfilesystem2/issues/455). - `WrapReadOnly.removetree` not raising a `ResourceReadOnly` when called. Closes [#468](https://github.com/PyFilesystem/pyfilesystem2/issues/468). - `WrapCachedDir.isdir` and `WrapCachedDir.isfile` raising a `ResourceNotFound` error on non-existing path ([#470](https://github.com/PyFilesystem/pyfilesystem2/pull/470)). +- `FTPFS` not listing certain entries with sticky/SUID/SGID permissions set by Linux server ([#473](https://github.com/PyFilesystem/pyfilesystem2/pull/473)). + Closes [#451](https://github.com/PyFilesystem/pyfilesystem2/issues/451). ## [2.4.12] - 2021-01-14 diff --git a/fs/_ftp_parse.py b/fs/_ftp_parse.py index 9e15a265..a9088ab4 100644 --- a/fs/_ftp_parse.py +++ b/fs/_ftp_parse.py @@ -19,7 +19,8 @@ RE_LINUX = re.compile( r""" ^ - ([ldrwx-]{10}) + ([-dlpscbD]) + ([r-][w-][xsS-][r-][w-][xsS-][r-][w-][xtT-][\.\+]?) \s+? (\d+) \s+? @@ -110,14 +111,14 @@ def _decode_linux_time(mtime): def decode_linux(line, match): - perms, links, uid, gid, size, mtime, name = match.groups() - is_link = perms.startswith("l") - is_dir = perms.startswith("d") or is_link + ty, perms, links, uid, gid, size, mtime, name = match.groups() + is_link = ty == "l" + is_dir = ty == "d" or is_link if is_link: name, _, _link_name = name.partition("->") name = name.strip() _link_name = _link_name.strip() - permissions = Permissions.parse(perms[1:]) + permissions = Permissions.parse(perms) mtime_epoch = _decode_linux_time(mtime)

FTPFS does not list certain folders It seems that the pub folder is not listed and igrnoed. ftplib does show the folder. This is with python 3.8. ```python import fs fs.__version__ # 2.4.12 from fs.ftpfs import FTPFS a = FTP(host='ftp.ensemblgenomes.org', user="anonymous", passwd="") a.dir() """" drwxr-sr-x 72 ftp ftp 8192 Feb 16 02:00 pub -rw-r--r-- 1 ftp ftp 25 Jan 06 22:41 robots.txt """" from ftplib import FTP a = FTPFS(host='ftp.ensemblgenomes.org') a.listdir('.') # ['robots.txt'] ```

Hmm, could be an issue with the #439, which changed the FTP list parsing. I'll have a look when I can. Thanks @althonos!

2021-03-26T18:30:40

0.0

[]

PyFilesystem/pyfilesystem2

PyFilesystem__pyfilesystem2-472

99d5314cd15a36c408e3ce90be25a4aa09bb8391

diff --git a/CHANGELOG.md b/CHANGELOG.md index 0b80ff81..c99ffcc3 100644 --- a/CHANGELOG.md +++ b/CHANGELOG.md @@ -20,11 +20,13 @@ and this project adheres to [Semantic Versioning](http://semver.org/). Closes [#445](https://github.com/PyFilesystem/pyfilesystem2/pull/445). - Migrate continuous integration from Travis-CI to GitHub Actions and introduce several linters again in the build steps ([#448](https://github.com/PyFilesystem/pyfilesystem2/pull/448)). - Closes [#446](https://github.com/PyFilesystem/pyfilesystem2/pull/446). + Closes [#446](https://github.com/PyFilesystem/pyfilesystem2/issues/446). - Stop requiring `pytest` to run tests, allowing any test runner supporting `unittest`-style test suites. - `FSTestCases` now builds the large data required for `upload` and `download` tests only once in order to reduce the total testing time. +- `MemoryFS.move` and `MemoryFS.movedir` will now avoid copying data. + Closes [#452](https://github.com/PyFilesystem/pyfilesystem2/issues/452). ### Fixed diff --git a/fs/memoryfs.py b/fs/memoryfs.py index 30e9c21f..7965a135 100644 --- a/fs/memoryfs.py +++ b/fs/memoryfs.py @@ -38,6 +38,7 @@ SupportsInt, Union, Text, + Tuple, ) from .base import _OpendirFactory from .info import RawInfo @@ -274,6 +275,10 @@ def remove_entry(self, name): # type: (Text) -> None del self._dir[name] + def clear(self): + # type: () -> None + self._dir.clear() + def __contains__(self, name): # type: (object) -> bool return name in self._dir @@ -294,6 +299,21 @@ def remove_open_file(self, memory_file): # type: (_MemoryFile) -> None self._open_files.remove(memory_file) + def to_info(self, namespaces=None): + # type: (Optional[Collection[Text]]) -> Info + namespaces = namespaces or () + info = {"basic": {"name": self.name, "is_dir": self.is_dir}} + if "details" in namespaces: + info["details"] = { + "_write": ["accessed", "modified"], + "type": int(self.resource_type), + "size": self.size, + "accessed": self.accessed_time, + "modified": self.modified_time, + "created": self.created_time, + } + return Info(info) + @six.python_2_unicode_compatible class MemoryFS(FS): @@ -368,33 +388,24 @@ def close(self): def getinfo(self, path, namespaces=None): # type: (Text, Optional[Collection[Text]]) -> Info - namespaces = namespaces or () _path = self.validatepath(path) dir_entry = self._get_dir_entry(_path) if dir_entry is None: raise errors.ResourceNotFound(path) - info = {"basic": {"name": dir_entry.name, "is_dir": dir_entry.is_dir}} - if "details" in namespaces: - info["details"] = { - "_write": ["accessed", "modified"], - "type": int(dir_entry.resource_type), - "size": dir_entry.size, - "accessed": dir_entry.accessed_time, - "modified": dir_entry.modified_time, - "created": dir_entry.created_time, - } - return Info(info) + return dir_entry.to_info(namespaces=namespaces) def listdir(self, path): # type: (Text) -> List[Text] self.check() _path = self.validatepath(path) with self._lock: + # locate and validate the entry corresponding to the given path dir_entry = self._get_dir_entry(_path) if dir_entry is None: raise errors.ResourceNotFound(path) if not dir_entry.is_dir: raise errors.DirectoryExpected(path) + # return the filenames in the order they were created return dir_entry.list() if typing.TYPE_CHECKING: @@ -433,6 +444,46 @@ def makedir( parent_dir.set_entry(dir_name, new_dir) return self.opendir(path) + def move(self, src_path, dst_path, overwrite=False): + src_dir, src_name = split(self.validatepath(src_path)) + dst_dir, dst_name = split(self.validatepath(dst_path)) + + with self._lock: + src_dir_entry = self._get_dir_entry(src_dir) + if src_dir_entry is None or src_name not in src_dir_entry: + raise errors.ResourceNotFound(src_path) + src_entry = src_dir_entry.get_entry(src_name) + if src_entry.is_dir: + raise errors.FileExpected(src_path) + + dst_dir_entry = self._get_dir_entry(dst_dir) + if dst_dir_entry is None: + raise errors.ResourceNotFound(dst_path) + elif not overwrite and dst_name in dst_dir_entry: + raise errors.DestinationExists(dst_path) + + dst_dir_entry.set_entry(dst_name, src_entry) + src_dir_entry.remove_entry(src_name) + + def movedir(self, src_path, dst_path, create=False): + src_dir, src_name = split(self.validatepath(src_path)) + dst_dir, dst_name = split(self.validatepath(dst_path)) + + with self._lock: + src_dir_entry = self._get_dir_entry(src_dir) + if src_dir_entry is None or src_name not in src_dir_entry: + raise errors.ResourceNotFound(src_path) + src_entry = src_dir_entry.get_entry(src_name) + if not src_entry.is_dir: + raise errors.DirectoryExpected(src_path) + + dst_dir_entry = self._get_dir_entry(dst_dir) + if dst_dir_entry is None or (not create and dst_name not in dst_dir_entry): + raise errors.ResourceNotFound(dst_path) + + dst_dir_entry.set_entry(dst_name, src_entry) + src_dir_entry.remove_entry(src_name) + def openbin(self, path, mode="r", buffering=-1, **options): # type: (Text, Text, int, **Any) -> BinaryIO _mode = Mode(mode) @@ -499,12 +550,29 @@ def remove(self, path): def removedir(self, path): # type: (Text) -> None + # make sure we are not removing root _path = self.validatepath(path) - if _path == "/": raise errors.RemoveRootError() + # make sure the directory is empty + if not self.isempty(path): + raise errors.DirectoryNotEmpty(path) + # we can now delegate to removetree since we confirmed that + # * path exists (isempty) + # * path is a folder (isempty) + # * path is not root + self.removetree(_path) + + def removetree(self, path): + # type: (Text) -> None + _path = self.validatepath(path) with self._lock: + + if _path == "/": + self.root.clear() + return + dir_path, file_name = split(_path) parent_dir_entry = self._get_dir_entry(dir_path) @@ -515,11 +583,34 @@ def removedir(self, path): if not dir_dir_entry.is_dir: raise errors.DirectoryExpected(path) - if len(dir_dir_entry): - raise errors.DirectoryNotEmpty(path) - parent_dir_entry.remove_entry(file_name) + def scandir( + self, + path, # type: Text + namespaces=None, # type: Optional[Collection[Text]] + page=None, # type: Optional[Tuple[int, int]] + ): + # type: (...) -> Iterator[Info] + self.check() + _path = self.validatepath(path) + with self._lock: + # locate and validate the entry corresponding to the given path + dir_entry = self._get_dir_entry(_path) + if dir_entry is None: + raise errors.ResourceNotFound(path) + if not dir_entry.is_dir: + raise errors.DirectoryExpected(path) + # if paging was requested, slice the filenames + filenames = dir_entry.list() + if page is not None: + start, end = page + filenames = filenames[start:end] + # yield info with the right namespaces + for name in filenames: + entry = typing.cast(_DirEntry, dir_entry.get_entry(name)) + yield entry.to_info(namespaces=namespaces) + def setinfo(self, path, info): # type: (Text, RawInfo) -> None _path = self.validatepath(path)

How to rename files in MemFS? Hi, there doesn't seem to be a way to rename files in MemFS without copying them around. Looks like MemFS tries to apply `os.rename()` if the file can be seen by the OS (which is not the case for MemFS). The only remaining way is to use something like `fs.move` that duplicates the file before removing the old one. On systems with limited memory this obviously doesn't work. Example: a MemFS with a total size of 10 GB and a single file of 8 GB lying on it. How to rename this file?

Hmmm, interesting point... https://docs.pyfilesystem.org/en/latest/guide.html#moving-and-copying says "These move and copy methods are optimized where possible, and depending on the implementation, they may be more performant than reading and writing files.". However it looks like MemFS doesn't have optimised `move` or `movedir` methods and instead just drops through to the base `fs.move`, which as you note does a copy-then-delete. I'm not familiar enough with the MemFS to know if there's a good reason _why_ it doesn't provide optimised `move` / `movedir` methods though (and I don't have time to look into it ATM). But feel free to experiment and put a PR together... :wink: MemFs certainly could do a rename operation rather than a copy. It just hasn’t been implemented.

2021-03-26T14:06:18

0.0

[]

PyFilesystem/pyfilesystem2

PyFilesystem__pyfilesystem2-457

64d7a52e1f3037f33937f91e9db812cafdbbafc0

diff --git a/CHANGELOG.md b/CHANGELOG.md index 31cb320c..4dfe9058 100644 --- a/CHANGELOG.md +++ b/CHANGELOG.md @@ -31,6 +31,7 @@ and this project adheres to [Semantic Versioning](http://semver.org/). - Make `FTPFile`, `MemoryFile` and `RawWrapper` accept [`array.array`](https://docs.python.org/3/library/array.html) arguments for the `write` and `writelines` methods, as expected by their base class [`io.RawIOBase`](https://docs.python.org/3/library/io.html#io.RawIOBase). - Various documentation issues, including `MemoryFS` docstring not rendering properly. +- Avoid creating a new connection on every call of `FTPFS.upload`. Closes [#455](https://github.com/PyFilesystem/pyfilesystem2/issues/455). ## [2.4.12] - 2021-01-14 diff --git a/fs/ftpfs.py b/fs/ftpfs.py index 3d8b0944..c5929ee1 100644 --- a/fs/ftpfs.py +++ b/fs/ftpfs.py @@ -814,11 +814,10 @@ def upload(self, path, file, chunk_size=None, **options): # type: (Text, BinaryIO, Optional[int], **Any) -> None _path = self.validatepath(path) with self._lock: - with self._manage_ftp() as ftp: - with ftp_errors(self, path): - ftp.storbinary( - str("STOR ") + _encode(_path, self.ftp.encoding), file - ) + with ftp_errors(self, path): + self.ftp.storbinary( + str("STOR ") + _encode(_path, self.ftp.encoding), file + ) def writebytes(self, path, contents): # type: (Text, ByteString) -> None

fs.ftpfs.FTPFS.upload opens new connections Hi, I'm looking at this function: https://github.com/PyFilesystem/pyfilesystem2/blob/64d7a52e1f3037f33937f91e9db812cafdbbafc0/fs/ftpfs.py#L813 `_manage_ftp` always opens a new FTP connection, together with user login, "FEAT" command and so on. This is done everytime one wants to copy a file to an FTPFS. Is there a good reason why that code doesn't just use `self._ftp`, the connection that should be open for the lifetime of the FTPFS object? Otherwise, I'd like to create a PR to fix that.

Hmmm, true, I'm actually wondering too. I know it makes sense to open a new connection when opening a new file, to guarantee the file and the `FTPFS` can be used in parallel, but i don't think it makes sense for the shortcut methods (like `upload` or `readbytes`) to create a new connection. @willmcgugan , any opinion? Maybe there was a reason for doing so when you wrote the code that I am unaware of. > I know it makes sense to open a new connection when opening a new file, to guarantee the file and the FTPFS can be used in parallel, but i don't think it makes sense for the shortcut methods (like upload or readbytes) to create a new connection. Yeah, FTP offers no way of multiplexing file downloads that I know of, so multiple instances are the only way to achieve that. Re shortcut methods, I guess we could re-use the main connection used for listing directories, but there would still need to be one per thread to make it thread-safe. I wouldn't be surprised if there were wins that could be made for copying files sequentially. Especially for many small files. But in general I would expect downloading in parallel to eclipse those wins... > but there would still need to be one per thread to make it thread-safe. Wouldn't locking the `FTPFS` when accessing the main connection be sufficient though? Sure. That would limit performance though. Do we do that already? I don’t recall. In the snippet linked by @atollk the lock is acquired before `_manage_ftp` is called, yes. In that case re-using the connection may be beneficial. @atollk you’re welcome to tackle that, assuming @althonos approves. Fine by me! There are probably other places in `fs.ftpfs` where this could be beneficial, but we can figure that out if you want to open a PR.

2021-03-19T08:15:22

0.0

[]

NREL/routee-compass

NREL__routee-compass-260

7ca47d112b954cd668d05c7539d908462d5bf986

diff --git a/rust/routee-compass-core/src/algorithm/component/scc.rs b/rust/routee-compass-core/src/algorithm/component/scc.rs index e6de4381..8cc5e10d 100644 --- a/rust/routee-compass-core/src/algorithm/component/scc.rs +++ b/rust/routee-compass-core/src/algorithm/component/scc.rs @@ -1,5 +1,5 @@ -use crate::model::road_network::graph::Graph; -use crate::model::road_network::{graph_error::GraphError, vertex_id::VertexId}; +use crate::model::network::graph::Graph; +use crate::model::network::{network_error::NetworkError, vertex_id::VertexId}; use std::collections::HashSet; /// Conducts a depth-first search (DFS) on a directed graph. @@ -21,23 +21,23 @@ use std::collections::HashSet; /// pub fn depth_first_search( graph: &Graph, - vertex: VertexId, + vertex: &VertexId, visited: &mut HashSet<VertexId>, stack: &mut Vec<VertexId>, -) -> Result<(), GraphError> { - if visited.contains(&vertex) { +) -> Result<(), NetworkError> { + if visited.contains(vertex) { return Ok(()); } - visited.insert(vertex); + visited.insert(*vertex); - let edges = graph.out_edges(vertex)?; + let edges = graph.out_edges(vertex); for edge in edges { - let dst = graph.dst_vertex_id(edge)?; - depth_first_search(graph, dst, visited, stack)?; + let dst = graph.dst_vertex_id(&edge)?; + depth_first_search(graph, &dst, visited, stack)?; } - stack.push(vertex); + stack.push(*vertex); Ok(()) } @@ -61,23 +61,23 @@ pub fn depth_first_search( /// pub fn reverse_depth_first_search( graph: &Graph, - vertex: VertexId, + vertex: &VertexId, visited: &mut HashSet<VertexId>, stack: &mut Vec<VertexId>, -) -> Result<(), GraphError> { - if visited.contains(&vertex) { +) -> Result<(), NetworkError> { + if visited.contains(vertex) { return Ok(()); } - visited.insert(vertex); + visited.insert(*vertex); - let edges = graph.in_edges(vertex)?; + let edges = graph.in_edges(vertex); for edge in edges { - let src = graph.src_vertex_id(edge)?; - reverse_depth_first_search(graph, src, visited, stack)?; + let src = graph.src_vertex_id(&edge)?; + reverse_depth_first_search(graph, &src, visited, stack)?; } - stack.push(vertex); + stack.push(*vertex); Ok(()) } @@ -94,14 +94,15 @@ pub fn reverse_depth_first_search( /// /// Returns an error if the `graph` has an issue like a non-existing vertex. /// -pub fn all_strongly_connected_componenets(graph: &Graph) -> Result<Vec<Vec<VertexId>>, GraphError> { +pub fn all_strongly_connected_componenets( + graph: &Graph, +) -> Result<Vec<Vec<VertexId>>, NetworkError> { let mut visited: HashSet<VertexId> = HashSet::new(); let mut container: Vec<VertexId> = Vec::new(); - let mut result: Vec<Vec<VertexId>> = Vec::new(); for vertex_id in graph.vertex_ids() { - depth_first_search(graph, vertex_id, &mut visited, &mut container)?; + depth_first_search(graph, &vertex_id, &mut visited, &mut container)?; } visited.clear(); @@ -112,7 +113,7 @@ pub fn all_strongly_connected_componenets(graph: &Graph) -> Result<Vec<Vec<Verte } let mut component: Vec<VertexId> = Vec::new(); - reverse_depth_first_search(graph, vertex_id, &mut visited, &mut component)?; + reverse_depth_first_search(graph, &vertex_id, &mut visited, &mut component)?; result.push(component); } @@ -131,7 +132,7 @@ pub fn all_strongly_connected_componenets(graph: &Graph) -> Result<Vec<Vec<Verte /// /// Returns an error if the `graph` has an issue like a non-existing vertex. /// -pub fn largest_strongly_connected_component(graph: &Graph) -> Result<Vec<VertexId>, GraphError> { +pub fn largest_strongly_connected_component(graph: &Graph) -> Result<Vec<VertexId>, NetworkError> { let components = all_strongly_connected_componenets(graph)?; let mut largest_component: Vec<VertexId> = Vec::new(); @@ -149,7 +150,7 @@ pub fn largest_strongly_connected_component(graph: &Graph) -> Result<Vec<VertexI mod tests { use super::*; use crate::{ - model::property::{edge::Edge, vertex::Vertex}, + model::network::{Edge, Vertex}, util::compact_ordered_hash_map::CompactOrderedHashMap, }; diff --git a/rust/routee-compass-core/src/algorithm/search/a_star/a_star_algorithm.rs b/rust/routee-compass-core/src/algorithm/search/a_star/a_star_algorithm.rs index 0bcceecf..641d130a 100644 --- a/rust/routee-compass-core/src/algorithm/search/a_star/a_star_algorithm.rs +++ b/rust/routee-compass-core/src/algorithm/search/a_star/a_star_algorithm.rs @@ -4,8 +4,8 @@ use crate::algorithm::search::search_error::SearchError; use crate::algorithm::search::search_instance::SearchInstance; use crate::algorithm::search::search_result::SearchResult; use crate::algorithm::search::search_tree_branch::SearchTreeBranch; -use crate::model::road_network::edge_id::EdgeId; -use crate::model::road_network::vertex_id::VertexId; +use crate::model::network::edge_id::EdgeId; +use crate::model::network::vertex_id::VertexId; use crate::model::unit::as_f64::AsF64; use crate::model::unit::cost::ReverseCost; use crate::model::unit::Cost; @@ -61,7 +61,7 @@ pub fn run_a_star( let last_edge_id = get_last_traversed_edge_id(&current_vertex_id, &source, &solution)?; let last_edge = match last_edge_id { - Some(id) => Some(si.directed_graph.get_edge(id)?), + Some(id) => Some(si.directed_graph.get_edge(&id)?), None => None, }; @@ -72,7 +72,7 @@ pub fn run_a_star( solution .get(&current_vertex_id) .ok_or_else(|| { - SearchError::InternalSearchError(format!( + SearchError::InternalError(format!( "expected vertex id {} missing from solution", current_vertex_id )) @@ -83,9 +83,9 @@ pub fn run_a_star( }; // visit all neighbors of this source vertex - let incident_edge_iterator = direction.get_incident_edges(&current_vertex_id, si)?; + let incident_edge_iterator = direction.get_incident_edges(&current_vertex_id, si); for edge_id in incident_edge_iterator { - let e = si.directed_graph.get_edge(*edge_id)?; + let e = si.directed_graph.get_edge(edge_id)?; let terminal_vertex_id = direction.terminal_vertex_id(e); let key_vertex_id = direction.tree_key_vertex_id(e); @@ -188,8 +188,8 @@ pub fn run_a_star_edge_oriented( si: &SearchInstance, ) -> Result<SearchResult, SearchError> { // 1. guard against edge conditions (src==dst, src.dst_v == dst.src_v) - let e1_src = si.directed_graph.src_vertex_id(source)?; - let e1_dst = si.directed_graph.dst_vertex_id(source)?; + let e1_src = si.directed_graph.src_vertex_id(&source)?; + let e1_dst = si.directed_graph.dst_vertex_id(&source)?; let src_et = EdgeTraversal { edge_id: source, access_cost: Cost::ZERO, @@ -217,8 +217,8 @@ pub fn run_a_star_edge_oriented( Ok(updated) } Some(target_edge) => { - let e2_src = si.directed_graph.src_vertex_id(target_edge)?; - let e2_dst = si.directed_graph.dst_vertex_id(target_edge)?; + let e2_src = si.directed_graph.src_vertex_id(&target_edge)?; + let e2_dst = si.directed_graph.dst_vertex_id(&target_edge)?; if source == target_edge { Ok(SearchResult::default()) @@ -254,12 +254,17 @@ pub fn run_a_star_edge_oriented( } = run_a_star(e1_dst, Some(e2_src), direction, weight_factor, si)?; if tree.is_empty() { - return Err(SearchError::NoPathExists(e1_dst, e2_src)); + return Err(SearchError::NoPathExistsBetweenVertices(e1_dst, e2_src)); } let final_state = &tree .get(&e2_src) - .ok_or_else(|| SearchError::VertexMissingFromSearchTree(e2_src))? + .ok_or_else(|| { + SearchError::InternalError(format!( + "resulting tree missing vertex {} expected via backtrack", + e2_src + )) + })? .edge_traversal .result_state; let dst_et = EdgeTraversal { @@ -313,7 +318,10 @@ fn advance_search( target: Option<VertexId>, ) -> Result<Option<VertexId>, SearchError> { match (cost.pop(), target) { - (None, Some(target_vertex_id)) => Err(SearchError::NoPathExists(source, target_vertex_id)), + (None, Some(target_vertex_id)) => Err(SearchError::NoPathExistsBetweenVertices( + source, + target_vertex_id, + )), (None, None) => Ok(None), (Some((current_v, _)), Some(target_v)) if current_v == target_v => Ok(None), (Some((current_vertex_id, _)), _) => Ok(Some(current_vertex_id)), @@ -347,7 +355,7 @@ fn get_last_traversed_edge_id( let edge_id = tree .get(this_vertex_id) .ok_or_else(|| { - SearchError::InternalSearchError(format!( + SearchError::InternalError(format!( "expected vertex id {} missing from solution", this_vertex_id )) @@ -368,10 +376,10 @@ mod tests { use crate::model::cost::cost_model::CostModel; use crate::model::cost::vehicle::vehicle_cost_rate::VehicleCostRate; use crate::model::frontier::default::no_restriction::NoRestriction; - use crate::model::property::edge::Edge; - use crate::model::property::vertex::Vertex; - use crate::model::road_network::edge_id::EdgeId; - use crate::model::road_network::graph::Graph; + use crate::model::network::edge_id::EdgeId; + use crate::model::network::graph::Graph; + use crate::model::network::Edge; + use crate::model::network::Vertex; use crate::model::state::state_feature::StateFeature; use crate::model::state::state_model::StateModel; use crate::model::termination::termination_model::TerminationModel; diff --git a/rust/routee-compass-core/src/algorithm/search/a_star/bidirectional_a_star_algorithm.rs b/rust/routee-compass-core/src/algorithm/search/a_star/bidirectional_ops.rs similarity index 89% rename from rust/routee-compass-core/src/algorithm/search/a_star/bidirectional_a_star_algorithm.rs rename to rust/routee-compass-core/src/algorithm/search/a_star/bidirectional_ops.rs index 882adf9e..398f7731 100644 --- a/rust/routee-compass-core/src/algorithm/search/a_star/bidirectional_a_star_algorithm.rs +++ b/rust/routee-compass-core/src/algorithm/search/a_star/bidirectional_ops.rs @@ -23,12 +23,7 @@ pub fn reorient_reverse_route( ) -> Result<Vec<EdgeTraversal>, SearchError> { // get the final edge id and state for the forward traversal let (final_fwd_edge_id, mut acc_state) = match fwd_route.last() { - None => ( - None, - si.state_model - .initial_state() - .map_err(SearchError::StateError)?, - ), + None => (None, si.state_model.initial_state()?), Some(last_edge) => (Some(last_edge.edge_id), last_edge.result_state.clone()), }; @@ -46,7 +41,7 @@ pub fn reorient_reverse_route( let mut result: Vec<EdgeTraversal> = Vec::with_capacity(rev_route.len()); for (prev_opt, next_opt) in edge_ids.iter().tuple_windows() { let next = next_opt.ok_or_else(|| { - SearchError::InternalSearchError(String::from("next_opt should never be None")) + SearchError::InternalError(String::from("next_opt should never be None")) })?; let et = EdgeTraversal::forward_traversal(next, *prev_opt, &acc_state, si)?; acc_state = et.result_state.clone(); @@ -72,7 +67,7 @@ pub fn route_contains_loop( ) -> Result<bool, SearchError> { let src_vertices = route .iter() - .map(|e| si.directed_graph.src_vertex_id(e.edge_id)) + .map(|e| si.directed_graph.src_vertex_id(&e.edge_id)) .collect::<Result<Vec<_>, _>>()?; Ok(src_vertices.iter().unique().collect_vec().len() < src_vertices.len()) } diff --git a/rust/routee-compass-core/src/algorithm/search/a_star/mod.rs b/rust/routee-compass-core/src/algorithm/search/a_star/mod.rs index 7177a9bb..c60a7533 100644 --- a/rust/routee-compass-core/src/algorithm/search/a_star/mod.rs +++ b/rust/routee-compass-core/src/algorithm/search/a_star/mod.rs @@ -1,2 +1,2 @@ pub mod a_star_algorithm; -pub mod bidirectional_a_star_algorithm; +pub mod bidirectional_ops; diff --git a/rust/routee-compass-core/src/algorithm/search/backtrack.rs b/rust/routee-compass-core/src/algorithm/search/backtrack.rs index 307f0e20..8670a278 100644 --- a/rust/routee-compass-core/src/algorithm/search/backtrack.rs +++ b/rust/routee-compass-core/src/algorithm/search/backtrack.rs @@ -1,7 +1,7 @@ use super::{ edge_traversal::EdgeTraversal, search_error::SearchError, search_tree_branch::SearchTreeBranch, }; -use crate::model::road_network::{edge_id::EdgeId, graph::Graph, vertex_id::VertexId}; +use crate::model::network::{edge_id::EdgeId, graph::Graph, vertex_id::VertexId}; use std::{ collections::{HashMap, HashSet}, sync::Arc, @@ -24,12 +24,16 @@ pub fn vertex_oriented_route( } let traversal = solution .get(&this_vertex) - .ok_or_else(|| SearchError::VertexMissingFromSearchTree(this_vertex))?; + .ok_or(SearchError::InternalError(format!( + "resulting tree missing vertex {} expected via backtrack", + this_vertex + )))?; let first_visit = visited.insert(traversal.edge_traversal.edge_id); if !first_visit { - return Err(SearchError::LoopInSearchResult( - traversal.edge_traversal.edge_id, - )); + return Err(SearchError::InternalError(format!( + "loop in search result, edge {} visited more than once", + traversal.edge_traversal.edge_id + ))); } result.push(traversal.edge_traversal.clone()); this_vertex = traversal.terminal_vertex; @@ -45,11 +49,7 @@ pub fn edge_oriented_route( solution: &HashMap<VertexId, SearchTreeBranch>, graph: Arc<Graph>, ) -> Result<Vec<EdgeTraversal>, SearchError> { - let o_v = graph - .src_vertex_id(source_id) - .map_err(SearchError::GraphError)?; - let d_v = graph - .dst_vertex_id(target_id) - .map_err(SearchError::GraphError)?; + let o_v = graph.src_vertex_id(&source_id)?; + let d_v = graph.dst_vertex_id(&target_id)?; vertex_oriented_route(o_v, d_v, solution) } diff --git a/rust/routee-compass-core/src/algorithm/search/direction.rs b/rust/routee-compass-core/src/algorithm/search/direction.rs index 6c49c337..d51bd5ae 100644 --- a/rust/routee-compass-core/src/algorithm/search/direction.rs +++ b/rust/routee-compass-core/src/algorithm/search/direction.rs @@ -1,11 +1,8 @@ use super::{ edge_traversal::EdgeTraversal, search_error::SearchError, search_instance::SearchInstance, }; -use crate::model::{ - property::edge::Edge, - road_network::{edge_id::EdgeId, graph_error::GraphError, vertex_id::VertexId}, - traversal::state::state_variable::StateVar, -}; +use crate::model::network::{Edge, EdgeId, VertexId}; +use crate::model::traversal::state::state_variable::StateVar; use serde::{Deserialize, Serialize}; #[derive(Copy, Clone, Serialize, Deserialize, Default)] @@ -21,10 +18,10 @@ impl Direction { &'a self, vertex_id: &VertexId, si: &'a SearchInstance, - ) -> Result<Box<dyn Iterator<Item = &EdgeId> + 'a>, GraphError> { + ) -> Box<dyn Iterator<Item = &'a EdgeId> + 'a> { match self { - Direction::Forward => si.directed_graph.out_edges_iter(*vertex_id), - Direction::Reverse => si.directed_graph.in_edges_iter(*vertex_id), + Direction::Forward => si.directed_graph.out_edges_iter(vertex_id), + Direction::Reverse => si.directed_graph.in_edges_iter(vertex_id), } } diff --git a/rust/routee-compass-core/src/algorithm/search/edge_traversal.rs b/rust/routee-compass-core/src/algorithm/search/edge_traversal.rs index 786ce83a..675975ed 100644 --- a/rust/routee-compass-core/src/algorithm/search/edge_traversal.rs +++ b/rust/routee-compass-core/src/algorithm/search/edge_traversal.rs @@ -1,6 +1,6 @@ use super::search_error::SearchError; use super::search_instance::SearchInstance; -use crate::model::road_network::edge_id::EdgeId; +use crate::model::network::edge_id::EdgeId; use crate::model::traversal::state::state_variable::StateVar; use crate::model::traversal::state::traversal_state::TraversalState; use crate::model::unit::Cost; @@ -39,7 +39,6 @@ impl EdgeTraversal { /// /// # Arguments /// - /// * `next_edge_id` - the edge to traverse /// * `prev_edge_id_opt` - the previously traversed edge, if exists, for access costs /// * `prev_state` - the state before traversal, positioned closer to the destination @@ -58,22 +57,13 @@ impl EdgeTraversal { let mut access_cost = Cost::ZERO; // find this traversal in the graph - let traversal_trajectory = si - .directed_graph - .edge_triplet_attrs(next_edge_id) - .map_err(SearchError::GraphError)?; + let traversal_trajectory = si.directed_graph.edge_triplet(&next_edge_id)?; // perform access traversal for (v2)-[next]->(v3) // access cost for (v1)-[prev]->(v2)-[next]->(v3) if let Some(prev_edge_id) = prev_edge_id_opt { - let e1 = si - .directed_graph - .get_edge(prev_edge_id) - .map_err(SearchError::GraphError)?; - let v1 = si - .directed_graph - .get_vertex(e1.src_vertex_id) - .map_err(SearchError::GraphError)?; + let e1 = si.directed_graph.get_edge(&prev_edge_id)?; + let v1 = si.directed_graph.get_vertex(&e1.src_vertex_id)?; let (v2, e2, v3) = traversal_trajectory; let access_trajectory = (v1, e1, v2, e2, v3); @@ -83,20 +73,20 @@ impl EdgeTraversal { let ac = si .cost_model - .access_cost(e1, e2, prev_state, &result_state) - .map_err(SearchError::CostError)?; + .access_cost(e1, e2, prev_state, &result_state)?; access_cost = access_cost + ac; } - si.traversal_model - .traverse_edge(traversal_trajectory, &mut result_state, &si.state_model) - .map_err(SearchError::TraversalModelFailure)?; + si.traversal_model.traverse_edge( + traversal_trajectory, + &mut result_state, + &si.state_model, + )?; let (_, edge, _) = traversal_trajectory; let total_cost = si .cost_model - .traversal_cost(edge, prev_state, &result_state) - .map_err(SearchError::CostError)?; + .traversal_cost(edge, prev_state, &result_state)?; let traversal_cost = total_cost - access_cost; let result = EdgeTraversal { @@ -137,22 +127,13 @@ impl EdgeTraversal { let mut access_cost = Cost::ZERO; // find this traversal in the graph - let traversal_trajectory = si - .directed_graph - .edge_triplet_attrs(prev_edge_id) - .map_err(SearchError::GraphError)?; + let traversal_trajectory = si.directed_graph.edge_triplet(&prev_edge_id)?; // perform access traversal for (v1)-[prev]->(v2) // access cost for (v1)-[prev]->(v2)-[next]->(v3) if let Some(next_edge_id) = next_edge_id_opt { - let e2 = si - .directed_graph - .get_edge(next_edge_id) - .map_err(SearchError::GraphError)?; - let v3 = si - .directed_graph - .get_vertex(e2.dst_vertex_id) - .map_err(SearchError::GraphError)?; + let e2 = si.directed_graph.get_edge(&next_edge_id)?; + let v3 = si.directed_graph.get_vertex(&e2.dst_vertex_id)?; let (v1, e1, v2) = traversal_trajectory; let access_trajectory = (v1, e1, v2, e2, v3); @@ -162,20 +143,20 @@ impl EdgeTraversal { let ac = si .cost_model - .access_cost(e1, e2, prev_state, &result_state) - .map_err(SearchError::CostError)?; + .access_cost(e1, e2, prev_state, &result_state)?; access_cost = access_cost + ac; } - si.traversal_model - .traverse_edge(traversal_trajectory, &mut result_state, &si.state_model) - .map_err(SearchError::TraversalModelFailure)?; + si.traversal_model.traverse_edge( + traversal_trajectory, + &mut result_state, + &si.state_model, + )?; let (_, edge, _) = traversal_trajectory; let total_cost = si .cost_model - .traversal_cost(edge, prev_state, &result_state) - .map_err(SearchError::CostError)?; + .traversal_cost(edge, prev_state, &result_state)?; let traversal_cost = total_cost - access_cost; let result = EdgeTraversal { diff --git a/rust/routee-compass-core/src/algorithm/search/ksp/ksp_query.rs b/rust/routee-compass-core/src/algorithm/search/ksp/ksp_query.rs index d9714d95..4b4093ef 100644 --- a/rust/routee-compass-core/src/algorithm/search/ksp/ksp_query.rs +++ b/rust/routee-compass-core/src/algorithm/search/ksp/ksp_query.rs @@ -1,6 +1,4 @@ -use crate::{ - algorithm::search::search_error::SearchError, model::road_network::vertex_id::VertexId, -}; +use crate::{algorithm::search::search_error::SearchError, model::network::vertex_id::VertexId}; #[derive(Debug, Clone)] pub struct KspQuery<'a> { diff --git a/rust/routee-compass-core/src/algorithm/search/ksp/single_via_paths_algorithm.rs b/rust/routee-compass-core/src/algorithm/search/ksp/single_via_paths_algorithm.rs index 5f6bc9b7..5ce9ef7c 100644 --- a/rust/routee-compass-core/src/algorithm/search/ksp/single_via_paths_algorithm.rs +++ b/rust/routee-compass-core/src/algorithm/search/ksp/single_via_paths_algorithm.rs @@ -3,12 +3,12 @@ use itertools::Itertools; use super::{ksp_query::KspQuery, ksp_termination_criteria::KspTerminationCriteria}; use crate::{ algorithm::search::{ - a_star::bidirectional_a_star_algorithm, backtrack, direction::Direction, - edge_traversal::EdgeTraversal, search_algorithm::SearchAlgorithm, - search_algorithm_result::SearchAlgorithmResult, search_error::SearchError, - search_instance::SearchInstance, util::route_similarity_function::RouteSimilarityFunction, + a_star::bidirectional_ops, backtrack, direction::Direction, edge_traversal::EdgeTraversal, + search_algorithm::SearchAlgorithm, search_algorithm_result::SearchAlgorithmResult, + search_error::SearchError, search_instance::SearchInstance, + util::route_similarity_function::RouteSimilarityFunction, }, - model::{road_network::vertex_id::VertexId, unit::cost::ReverseCost}, + model::{network::vertex_id::VertexId, unit::cost::ReverseCost}, util::priority_queue::InternalPriorityQueue, }; use std::collections::HashMap; @@ -45,23 +45,23 @@ pub fn run( si, )?; if fwd_trees.len() != 1 { - Err(SearchError::InternalSearchError(format!( + Err(SearchError::InternalError(format!( "ksp solver fwd trees count should be exactly 1, found {}", fwd_trees.len() )))?; } if rev_trees.len() != 1 { - Err(SearchError::InternalSearchError(format!( + Err(SearchError::InternalError(format!( "ksp solver rev trees count should be exactly 1, found {}", rev_trees.len() )))?; } - let fwd_tree = fwd_trees.first().ok_or_else(|| { - SearchError::InternalSearchError(String::from("cannot retrieve fwd tree 0")) - })?; - let rev_tree = rev_trees.first().ok_or_else(|| { - SearchError::InternalSearchError(String::from("cannot retrieve rev tree 0")) - })?; + let fwd_tree = fwd_trees + .first() + .ok_or_else(|| SearchError::InternalError(String::from("cannot retrieve fwd tree 0")))?; + let rev_tree = rev_trees + .first() + .ok_or_else(|| SearchError::InternalError(String::from("cannot retrieve rev tree 0")))?; // find intersection vertices let rev_vertices = rev_trees.iter().flatten().collect::<HashMap<_, _>>(); @@ -114,15 +114,12 @@ pub fn run( intersection_vertex_id, rev_tree, )?; - let rev_route = bidirectional_a_star_algorithm::reorient_reverse_route( - &fwd_route, - &rev_route_backward, - si, - )?; + let rev_route = + bidirectional_ops::reorient_reverse_route(&fwd_route, &rev_route_backward, si)?; let this_route = fwd_route.into_iter().chain(rev_route).collect::<Vec<_>>(); // test loop - if bidirectional_a_star_algorithm::route_contains_loop(&this_route, si)? { + if bidirectional_ops::route_contains_loop(&this_route, si)? { log::debug!("ksp:{} contains loop", ksp_it); accept_route = false; } diff --git a/rust/routee-compass-core/src/algorithm/search/ksp/yens_algorithm.rs b/rust/routee-compass-core/src/algorithm/search/ksp/yens_algorithm.rs index d0525ab0..b480cfc0 100644 --- a/rust/routee-compass-core/src/algorithm/search/ksp/yens_algorithm.rs +++ b/rust/routee-compass-core/src/algorithm/search/ksp/yens_algorithm.rs @@ -6,7 +6,7 @@ use crate::{ search_instance::SearchInstance, util::edge_cut_frontier_model::EdgeCutFrontierModel, util::route_similarity_function::RouteSimilarityFunction, }, - model::{road_network::edge_id::EdgeId, unit::Cost}, + model::{network::edge_id::EdgeId, unit::Cost}, }; use itertools::Itertools; use std::{collections::HashSet, sync::Arc}; @@ -53,7 +53,7 @@ pub fn run( accepted .last() .cloned() - .ok_or(SearchError::InternalSearchError(String::from( + .ok_or(SearchError::InternalError(String::from( "at least one route should be in routes", )))?; @@ -65,12 +65,12 @@ pub fn run( let spur_edge_traversal = root_path .last() - .ok_or(SearchError::InternalSearchError(String::from( + .ok_or(SearchError::InternalError(String::from( "root path is empty", )))?; let spur_vertex_id = si .directed_graph - .get_edge(spur_edge_traversal.edge_id)? + .get_edge(&spur_edge_traversal.edge_id)? .dst_vertex_id; // cut frontier edges based on previous paths with matching root path @@ -149,7 +149,7 @@ pub fn run( fn get_first_route(res: &SearchAlgorithmResult) -> Result<&Vec<EdgeTraversal>, SearchError> { res.routes .first() - .ok_or(SearchError::InternalSearchError(String::from( + .ok_or(SearchError::InternalError(String::from( "no empty results should be stored in routes", ))) } diff --git a/rust/routee-compass-core/src/algorithm/search/mod.rs b/rust/routee-compass-core/src/algorithm/search/mod.rs index aaef9d27..989db776 100644 --- a/rust/routee-compass-core/src/algorithm/search/mod.rs +++ b/rust/routee-compass-core/src/algorithm/search/mod.rs @@ -1,5 +1,5 @@ use self::search_tree_branch::SearchTreeBranch; -use crate::model::road_network::vertex_id::VertexId; +use crate::model::network::vertex_id::VertexId; use std::collections::HashMap; pub mod a_star; diff --git a/rust/routee-compass-core/src/algorithm/search/search_algorithm.rs b/rust/routee-compass-core/src/algorithm/search/search_algorithm.rs index a09c9253..34fe869f 100644 --- a/rust/routee-compass-core/src/algorithm/search/search_algorithm.rs +++ b/rust/routee-compass-core/src/algorithm/search/search_algorithm.rs @@ -9,7 +9,7 @@ use super::search_instance::SearchInstance; use super::search_tree_branch::SearchTreeBranch; use super::util::route_similarity_function::RouteSimilarityFunction; use super::{a_star::a_star_algorithm, direction::Direction}; -use crate::model::road_network::{edge_id::EdgeId, vertex_id::VertexId}; +use crate::model::network::{edge_id::EdgeId, vertex_id::VertexId}; use crate::model::unit::Cost; use serde::{Deserialize, Serialize}; @@ -182,8 +182,8 @@ pub fn run_edge_oriented( si: &SearchInstance, ) -> Result<SearchAlgorithmResult, SearchError> { // 1. guard against edge conditions (src==dst, src.dst_v == dst.src_v) - let e1_src = si.directed_graph.src_vertex_id(source)?; - let e1_dst = si.directed_graph.dst_vertex_id(source)?; + let e1_src = si.directed_graph.src_vertex_id(&source)?; + let e1_dst = si.directed_graph.dst_vertex_id(&source)?; let src_et = EdgeTraversal { edge_id: source, access_cost: Cost::ZERO, @@ -218,8 +218,8 @@ pub fn run_edge_oriented( Ok(updated) } Some(target_edge) => { - let e2_src = si.directed_graph.src_vertex_id(target_edge)?; - let e2_dst = si.directed_graph.dst_vertex_id(target_edge)?; + let e2_src = si.directed_graph.src_vertex_id(&target_edge)?; + let e2_dst = si.directed_graph.dst_vertex_id(&target_edge)?; if source == target_edge { Ok(SearchAlgorithmResult::default()) @@ -258,14 +258,14 @@ pub fn run_edge_oriented( } = alg.run_vertex_oriented(e1_dst, Some(e2_src), query, direction, si)?; if trees.is_empty() { - return Err(SearchError::NoPathExists(e1_dst, e2_src)); + return Err(SearchError::NoPathExistsBetweenVertices(e1_dst, e2_src)); } // it is possible that the search already found these vertices. one major edge // case is when the trip starts with a u-turn. for route in routes.iter_mut() { let final_state = route.last().ok_or_else(|| { - SearchError::InternalSearchError(String::from("found empty result route")) + SearchError::InternalError(String::from("found empty result route")) })?; let dst_et = EdgeTraversal { diff --git a/rust/routee-compass-core/src/algorithm/search/search_algorithm_result.rs b/rust/routee-compass-core/src/algorithm/search/search_algorithm_result.rs index cd14a504..93fb6e4c 100644 --- a/rust/routee-compass-core/src/algorithm/search/search_algorithm_result.rs +++ b/rust/routee-compass-core/src/algorithm/search/search_algorithm_result.rs @@ -1,5 +1,5 @@ use super::{edge_traversal::EdgeTraversal, search_tree_branch::SearchTreeBranch}; -use crate::model::road_network::vertex_id::VertexId; +use crate::model::network::vertex_id::VertexId; use std::collections::HashMap; #[derive(Default)] diff --git a/rust/routee-compass-core/src/algorithm/search/search_error.rs b/rust/routee-compass-core/src/algorithm/search/search_error.rs index 0777d61b..fad6be4d 100644 --- a/rust/routee-compass-core/src/algorithm/search/search_error.rs +++ b/rust/routee-compass-core/src/algorithm/search/search_error.rs @@ -1,43 +1,60 @@ use crate::model::{ access::access_model_error::AccessModelError, - cost::cost_error::CostError, + cost::cost_model_error::CostModelError, frontier::frontier_model_error::FrontierModelError, - road_network::{edge_id::EdgeId, graph_error::GraphError, vertex_id::VertexId}, - state::state_error::StateError, + network::{edge_id::EdgeId, network_error::NetworkError, vertex_id::VertexId}, + state::state_model_error::StateModelError, termination::termination_model_error::TerminationModelError, traversal::traversal_model_error::TraversalModelError, }; #[derive(thiserror::Error, Debug)] pub enum SearchError { - #[error("distance heuristic can only be provided when there is a target")] - DistanceHeuristicWithNoTarget, - #[error(transparent)] - StateError(#[from] StateError), - #[error(transparent)] - GraphError(#[from] GraphError), - #[error(transparent)] - TerminationModelError(#[from] TerminationModelError), - #[error(transparent)] - TraversalModelFailure(#[from] TraversalModelError), - #[error(transparent)] - AccessModelFailure(#[from] AccessModelError), - #[error(transparent)] - FrontierModelFailure(#[from] FrontierModelError), - #[error(transparent)] - CostError(#[from] CostError), - #[error("loop in search result revisits edge {0}")] - LoopInSearchResult(EdgeId), + #[error("failure building search algorithm: {0}")] + BuildError(String), + #[error("The search failed due to state model error. The state model is responsible for updates to the state of the search at each increment of traversal. Please review the [state] section of your Compass configuration. Source: {source}")] + StateFailure { + #[from] + source: StateModelError, + }, + #[error("The search failed due to a road network error. Please review the [graph] section of your Compass configuration. Source: {source}")] + NetworkFailure { + #[from] + source: NetworkError, + }, + #[error("The search failed due to termination model error. Please review the [termination] section of your Compass Configuration to make changes. Source: {source}")] + TerminationModelFailure { + #[from] + source: TerminationModelError, + }, + #[error("The search failed due to traversal model error. The traversal model performs edge traversals by updating the various search dimensions, based on the state and cost models. Please review the [traversal] section of your Compass Configuration. Source: {source}")] + TraversalModelFailure { + #[from] + source: TraversalModelError, + }, + #[error("The search failed due to access model error. The access model performs edge-to-edge transitions by updating the various search dimensions, based on the state and cost models. Please review the [access] section of your Compass Configuration. Source: {source}")] + AccessModelFailure { + #[from] + source: AccessModelError, + }, + #[error("The search failed due to frontier model error. The frontier model restricts access to edges. Please review the [frontier] section of your Compass Configuration. Source: {source}")] + FrontierModelFailure { + #[from] + source: FrontierModelError, + }, + #[error("The search failed due to cost model error. The cost model interprets a delta of search state dimensions as having a cost value, which is minimized by the search. Please see the [cost] section of your Compass Configuration and additionally any query-time overrides. Source: {source}")] + CostFailure { + #[from] + source: CostModelError, + }, #[error("query terminated due to {0}")] QueryTerminated(String), #[error("no path exists between vertices {0} and {1}")] - NoPathExists(VertexId, VertexId), - #[error("search tree is missing linked vertex {0}")] - VertexMissingFromSearchTree(VertexId), + NoPathExistsBetweenVertices(VertexId, VertexId), + #[error("no path exists between edges {0} and {1}")] + NoPathExistsBetweenEdges(EdgeId, EdgeId), #[error("error accessing shared read-only dataset: {0}")] ReadOnlyPoisonError(String), - #[error("failure building search algorithm: {0}")] - BuildError(String), #[error("internal error due to search logic: {0}")] - InternalSearchError(String), + InternalError(String), } diff --git a/rust/routee-compass-core/src/algorithm/search/search_instance.rs b/rust/routee-compass-core/src/algorithm/search/search_instance.rs index b8d50e8b..e71e665c 100644 --- a/rust/routee-compass-core/src/algorithm/search/search_instance.rs +++ b/rust/routee-compass-core/src/algorithm/search/search_instance.rs @@ -3,7 +3,7 @@ use crate::model::{ access::access_model::AccessModel, cost::cost_model::CostModel, frontier::frontier_model::FrontierModel, - road_network::{graph::Graph, vertex_id::VertexId}, + network::{graph::Graph, vertex_id::VertexId}, state::state_model::StateModel, termination::termination_model::TerminationModel, traversal::{state::state_variable::StateVar, traversal_model::TraversalModel}, @@ -32,8 +32,8 @@ impl SearchInstance { dst: VertexId, state: &[StateVar], ) -> Result<Cost, SearchError> { - let src = self.directed_graph.get_vertex(src)?; - let dst = self.directed_graph.get_vertex(dst)?; + let src = self.directed_graph.get_vertex(&src)?; + let dst = self.directed_graph.get_vertex(&dst)?; let mut dst_state = state.to_vec(); self.traversal_model diff --git a/rust/routee-compass-core/src/algorithm/search/search_result.rs b/rust/routee-compass-core/src/algorithm/search/search_result.rs index c3538f31..e414fe92 100644 --- a/rust/routee-compass-core/src/algorithm/search/search_result.rs +++ b/rust/routee-compass-core/src/algorithm/search/search_result.rs @@ -1,5 +1,5 @@ use super::search_tree_branch::SearchTreeBranch; -use crate::model::road_network::vertex_id::VertexId; +use crate::model::network::vertex_id::VertexId; use std::collections::HashMap; #[derive(Default)] diff --git a/rust/routee-compass-core/src/algorithm/search/search_tree_branch.rs b/rust/routee-compass-core/src/algorithm/search/search_tree_branch.rs index 1845de44..92409d34 100644 --- a/rust/routee-compass-core/src/algorithm/search/search_tree_branch.rs +++ b/rust/routee-compass-core/src/algorithm/search/search_tree_branch.rs @@ -1,5 +1,5 @@ use crate::{ - algorithm::search::edge_traversal::EdgeTraversal, model::road_network::vertex_id::VertexId, + algorithm::search::edge_traversal::EdgeTraversal, model::network::vertex_id::VertexId, }; use allocative::Allocative; use serde::Serialize; diff --git a/rust/routee-compass-core/src/algorithm/search/util/edge_cut_frontier_model.rs b/rust/routee-compass-core/src/algorithm/search/util/edge_cut_frontier_model.rs index d243ab68..58b591fb 100644 --- a/rust/routee-compass-core/src/algorithm/search/util/edge_cut_frontier_model.rs +++ b/rust/routee-compass-core/src/algorithm/search/util/edge_cut_frontier_model.rs @@ -1,7 +1,9 @@ +use crate::model::{ + frontier::frontier_model::FrontierModel, + network::{Edge, EdgeId}, +}; use std::{collections::HashSet, sync::Arc}; -use crate::model::{frontier::frontier_model::FrontierModel, road_network::edge_id::EdgeId}; - /// A wrapper of the user-generated FrontierModel which prohibits traversals /// on selected edges. algorithms can create this wrapper with a set of "cut edges" /// and the search will not allow traversal of these edges. @@ -25,9 +27,9 @@ impl EdgeCutFrontierModel { impl FrontierModel for EdgeCutFrontierModel { fn valid_frontier( &self, - edge: &crate::model::property::edge::Edge, + edge: &Edge, state: &[crate::model::traversal::state::state_variable::StateVar], - previous_edge: Option<&crate::model::property::edge::Edge>, + previous_edge: Option<&Edge>, state_model: &crate::model::state::state_model::StateModel, ) -> Result<bool, crate::model::frontier::frontier_model_error::FrontierModelError> { if self.cut_edges.contains(&edge.edge_id) { diff --git a/rust/routee-compass-core/src/algorithm/search/util/route_similarity_function.rs b/rust/routee-compass-core/src/algorithm/search/util/route_similarity_function.rs index babad35e..cc219898 100644 --- a/rust/routee-compass-core/src/algorithm/search/util/route_similarity_function.rs +++ b/rust/routee-compass-core/src/algorithm/search/util/route_similarity_function.rs @@ -1,12 +1,11 @@ -use std::collections::{HashMap, HashSet}; - use crate::{ algorithm::search::{ edge_traversal::EdgeTraversal, search_error::SearchError, search_instance::SearchInstance, }, - model::{road_network::edge_id::EdgeId, unit::as_f64::AsF64}, + model::{network::edge_id::EdgeId, unit::as_f64::AsF64}, }; use serde::{Deserialize, Serialize}; +use std::collections::{HashMap, HashSet}; #[derive(Serialize, Deserialize, Default, Clone)] #[serde(rename_all = "snake_case", tag = "type")] @@ -21,6 +20,8 @@ pub enum RouteSimilarityFunction { }, } +type DistanceFunction<'a> = Box<dyn Fn(&'_ EdgeId) -> Result<f64, SearchError> + 'a>; + impl RouteSimilarityFunction { /// tests for similarity between two paths. /// @@ -79,15 +80,15 @@ impl RouteSimilarityFunction { match self { RouteSimilarityFunction::AcceptAll => Ok(0.0), RouteSimilarityFunction::EdgeIdCosineSimilarity { threshold: _ } => { - let unit_dist_fn = Box::new(|_| Ok(1.0)); + let unit_dist_fn = Box::new(|_: &EdgeId| Ok(1.0)); cos_similarity(a, b, unit_dist_fn) } RouteSimilarityFunction::DistanceWeightedCosineSimilarity { threshold: _ } => { - let dist_fn = Box::new(|edge_id| { + let dist_fn = Box::new(|edge_id: &EdgeId| { si.directed_graph .get_edge(edge_id) .map(|edge| edge.distance.as_f64()) - .map_err(SearchError::GraphError) + .map_err(SearchError::from) }); cos_similarity(a, b, dist_fn) } @@ -106,18 +107,18 @@ impl RouteSimilarityFunction { /// # Returns /// /// the cosine similarity of the routes, a value from -1 to 1 -fn cos_similarity<'a>( +fn cos_similarity( a: &[&EdgeTraversal], b: &[&EdgeTraversal], - dist_fn: Box<dyn Fn(EdgeId) -> Result<f64, SearchError> + 'a>, + dist_fn: DistanceFunction<'_>, ) -> Result<f64, SearchError> { let a_map = a .iter() - .map(|e| dist_fn(e.edge_id).map(|dist| (e.edge_id, dist))) + .map(|e| dist_fn(&e.edge_id).map(|dist| (e.edge_id, dist))) .collect::<Result<HashMap<_, _>, _>>()?; let b_map = b .iter() - .map(|e| dist_fn(e.edge_id).map(|dist| (e.edge_id, dist))) + .map(|e| dist_fn(&e.edge_id).map(|dist| (e.edge_id, dist))) .collect::<Result<HashMap<_, _>, _>>()?; let numer: f64 = a_map diff --git a/rust/routee-compass-core/src/doc.md b/rust/routee-compass-core/src/doc.md index e39a9208..bbb2ee72 100644 --- a/rust/routee-compass-core/src/doc.md +++ b/rust/routee-compass-core/src/doc.md @@ -20,8 +20,8 @@ This can be broken up into the following modules: RouteE Compass takes a layered approach to modeling the road network. At the core is the [Graph] model. The edges and vertices are stored in `vec`s and carry the minimal data required for most search applications: -- [`crate::model::property::edge::Edge`] records store distance in meters -- [`crate::model::property::vertex::Vertex`] records store coordinate locations, assumed in WGS84 projection +- [`crate::model::road_network::Edge`] records store distance in meters +- [`crate::model::road_network::Vertex`] records store coordinate locations, assumed in WGS84 projection A forward and reverse adjacency list describes connectivity in the graph using the indices of the edges and vertices in their respective `vec`s. diff --git a/rust/routee-compass-core/src/model/access/access_model.rs b/rust/routee-compass-core/src/model/access/access_model.rs index f6f0597e..02254ebf 100644 --- a/rust/routee-compass-core/src/model/access/access_model.rs +++ b/rust/routee-compass-core/src/model/access/access_model.rs @@ -1,6 +1,6 @@ use super::access_model_error::AccessModelError; use crate::model::{ - property::{edge::Edge, vertex::Vertex}, + network::{Edge, Vertex}, state::{state_feature::StateFeature, state_model::StateModel}, traversal::state::state_variable::StateVar, }; diff --git a/rust/routee-compass-core/src/model/access/access_model_error.rs b/rust/routee-compass-core/src/model/access/access_model_error.rs index 1787a77d..b28307ec 100644 --- a/rust/routee-compass-core/src/model/access/access_model_error.rs +++ b/rust/routee-compass-core/src/model/access/access_model_error.rs @@ -1,13 +1,14 @@ -use crate::model::state::state_error::StateError; +use crate::model::state::state_model_error::StateModelError; #[derive(thiserror::Error, Debug)] pub enum AccessModelError { #[error("error while executing access model {name}: {error}")] RuntimeError { name: String, error: String }, - #[error(transparent)] - StateError(#[from] StateError), - #[error(transparent)] - SerdeJsonError(#[from] serde_json::Error), + #[error("access model failed due to state transition error: {source}")] + StateError { + #[from] + source: StateModelError, + }, #[error("{0}")] BuildError(String), } diff --git a/rust/routee-compass-core/src/model/access/default/combined_model.rs b/rust/routee-compass-core/src/model/access/default/combined_model.rs index 118bc62f..ad350a2b 100644 --- a/rust/routee-compass-core/src/model/access/default/combined_model.rs +++ b/rust/routee-compass-core/src/model/access/default/combined_model.rs @@ -3,7 +3,7 @@ use crate::model::{ access_model::AccessModel, access_model_error::AccessModelError, access_model_service::AccessModelService, }, - property::{edge::Edge, vertex::Vertex}, + network::{Edge, Vertex}, state::{state_feature::StateFeature, state_model::StateModel}, traversal::state::state_variable::StateVar, }; diff --git a/rust/routee-compass-core/src/model/access/default/no_access_model.rs b/rust/routee-compass-core/src/model/access/default/no_access_model.rs index 236adc53..812ab2c6 100644 --- a/rust/routee-compass-core/src/model/access/default/no_access_model.rs +++ b/rust/routee-compass-core/src/model/access/default/no_access_model.rs @@ -1,9 +1,11 @@ -use std::sync::Arc; - -use crate::model::access::{ - access_model::AccessModel, access_model_builder::AccessModelBuilder, - access_model_service::AccessModelService, +use crate::model::{ + access::{ + access_model::AccessModel, access_model_builder::AccessModelBuilder, + access_model_service::AccessModelService, + }, + network::{Edge, Vertex}, }; +use std::sync::Arc; #[derive(Clone, Debug)] pub struct NoAccessModel {} @@ -15,15 +17,9 @@ impl AccessModel for NoAccessModel { fn access_edge( &self, - _traversal: ( - &crate::model::property::vertex::Vertex, - &crate::model::property::edge::Edge, - &crate::model::property::vertex::Vertex, - &crate::model::property::edge::Edge, - &crate::model::property::vertex::Vertex, - ), - _state: &mut Vec<crate::model::traversal::state::state_variable::StateVar>, - _state_model: &crate::model::state::state_model::StateModel, + _: (&Vertex, &Edge, &Vertex, &Edge, &Vertex), + _: &mut Vec<crate::model::traversal::state::state_variable::StateVar>, + _: &crate::model::state::state_model::StateModel, ) -> Result<(), crate::model::access::access_model_error::AccessModelError> { Ok(()) } diff --git a/rust/routee-compass-core/src/model/access/default/turn_delays/turn_delay_access_model.rs b/rust/routee-compass-core/src/model/access/default/turn_delays/turn_delay_access_model.rs index a019c2ab..917d19ac 100644 --- a/rust/routee-compass-core/src/model/access/default/turn_delays/turn_delay_access_model.rs +++ b/rust/routee-compass-core/src/model/access/default/turn_delays/turn_delay_access_model.rs @@ -1,7 +1,7 @@ use super::turn_delay_access_model_engine::TurnDelayAccessModelEngine; use crate::model::{ access::{access_model::AccessModel, access_model_error::AccessModelError}, - property::{edge::Edge, vertex::Vertex}, + network::{Edge, Vertex}, state::{state_feature::StateFeature, state_model::StateModel}, traversal::state::state_variable::StateVar, }; diff --git a/rust/routee-compass-core/src/model/access/default/turn_delays/turn_delay_access_model_engine.rs b/rust/routee-compass-core/src/model/access/default/turn_delays/turn_delay_access_model_engine.rs index 7fe963fa..32e8ab15 100644 --- a/rust/routee-compass-core/src/model/access/default/turn_delays/turn_delay_access_model_engine.rs +++ b/rust/routee-compass-core/src/model/access/default/turn_delays/turn_delay_access_model_engine.rs @@ -1,12 +1,9 @@ -use crate::model::access::access_model_error::AccessModelError; -use crate::model::property::edge::Edge; -use crate::model::property::vertex::Vertex; -use crate::model::road_network::edge_id::EdgeId; -use crate::model::unit::{Time, TimeUnit}; - use super::edge_heading::EdgeHeading; use super::turn::Turn; use super::turn_delay_model::TurnDelayModel; +use crate::model::access::access_model_error::AccessModelError; +use crate::model::network::{Edge, EdgeId, Vertex}; +use crate::model::unit::{Time, TimeUnit}; pub struct TurnDelayAccessModelEngine { pub edge_headings: Box<[EdgeHeading]>, diff --git a/rust/routee-compass-core/src/model/cost/cost_aggregation.rs b/rust/routee-compass-core/src/model/cost/cost_aggregation.rs index e0094239..968be6ba 100644 --- a/rust/routee-compass-core/src/model/cost/cost_aggregation.rs +++ b/rust/routee-compass-core/src/model/cost/cost_aggregation.rs @@ -1,7 +1,7 @@ use crate::model::unit::{as_f64::AsF64, Cost}; use serde::{Deserialize, Serialize}; -use super::cost_error::CostError; +use super::cost_model_error::CostModelError; #[derive(Deserialize, Serialize, Clone, Copy, Debug, Default)] #[serde(rename_all = "snake_case")] @@ -29,8 +29,8 @@ impl CostAggregation { pub fn agg_iter<'a>( &self, - costs: impl Iterator<Item = Result<(&'a String, Cost), CostError>>, - ) -> Result<Cost, CostError> { + costs: impl Iterator<Item = Result<(&'a String, Cost), CostModelError>>, + ) -> Result<Cost, CostModelError> { match self { CostAggregation::Sum => { let mut sum = Cost::ZERO; diff --git a/rust/routee-compass-core/src/model/cost/cost_model.rs b/rust/routee-compass-core/src/model/cost/cost_model.rs index 44bfd6c4..99b3578d 100644 --- a/rust/routee-compass-core/src/model/cost/cost_model.rs +++ b/rust/routee-compass-core/src/model/cost/cost_model.rs @@ -2,8 +2,8 @@ use super::cost_aggregation::CostAggregation; use super::cost_ops; use super::network::network_cost_rate::NetworkCostRate; use super::vehicle::vehicle_cost_rate::VehicleCostRate; -use crate::model::cost::cost_error::CostError; -use crate::model::property::edge::Edge; +use crate::model::cost::cost_model_error::CostModelError; +use crate::model::network::Edge; use crate::model::state::state_model::StateModel; use crate::model::traversal::state::state_variable::StateVar; use crate::model::unit::Cost; @@ -52,7 +52,7 @@ impl CostModel { network_rate_mapping: Arc<HashMap<String, NetworkCostRate>>, cost_aggregation: CostAggregation, state_model: Arc<StateModel>, - ) -> Result<CostModel, CostError> { + ) -> Result<CostModel, CostModelError> { let mut indices = vec![]; let mut weights = vec![]; let mut vehicle_rates = vec![]; @@ -75,7 +75,7 @@ impl CostModel { } if weights.iter().sum::<f64>() == 0.0 { - return Err(CostError::InvalidCostVariables); + return Err(CostModelError::InvalidCostVariables); } Ok(CostModel { feature_indices: indices, @@ -102,7 +102,7 @@ impl CostModel { edge: &Edge, prev_state: &[StateVar], next_state: &[StateVar], - ) -> Result<Cost, CostError> { + ) -> Result<Cost, CostModelError> { let vehicle_cost = cost_ops::calculate_vehicle_costs( (prev_state, next_state), &self.feature_indices, @@ -146,7 +146,7 @@ impl CostModel { next_edge: &Edge, prev_state: &[StateVar], next_state: &[StateVar], - ) -> Result<Cost, CostError> { + ) -> Result<Cost, CostModelError> { let vehicle_cost = cost_ops::calculate_vehicle_costs( (prev_state, next_state), &self.feature_indices, @@ -184,7 +184,7 @@ impl CostModel { &self, src_state: &[StateVar], dst_state: &[StateVar], - ) -> Result<Cost, CostError> { + ) -> Result<Cost, CostModelError> { let vehicle_cost = cost_ops::calculate_vehicle_costs( (src_state, dst_state), &self.feature_indices, @@ -207,14 +207,14 @@ impl CostModel { /// A JSON serialized version of the state. This does not need to include /// additional details such as the units (kph, hours, etc), which can be /// summarized in the serialize_state_info method. - pub fn serialize_cost(&self, state: &[StateVar]) -> Result<serde_json::Value, CostError> { + pub fn serialize_cost(&self, state: &[StateVar]) -> Result<serde_json::Value, CostModelError> { let mut state_variable_costs = self .feature_indices .iter() .map(move |(name, idx)| { let state_var = state .get(*idx) - .ok_or_else(|| CostError::StateIndexOutOfBounds(*idx, name.clone()))?; + .ok_or_else(|| CostModelError::StateIndexOutOfBounds(*idx, name.clone()))?; let rate = self.vehicle_rates.get(*idx).ok_or_else(|| { let alternatives = self @@ -224,7 +224,7 @@ impl CostModel { .map(|(n, _)| n.to_string()) .collect::<Vec<_>>() .join(","); - CostError::StateVariableNotFound( + CostModelError::StateVariableNotFound( name.clone(), String::from("vehicle cost rates while serializing cost"), alternatives, @@ -233,7 +233,7 @@ impl CostModel { let cost = rate.map_value(*state_var); Ok((name.clone(), cost)) }) - .collect::<Result<HashMap<String, Cost>, CostError>>()?; + .collect::<Result<HashMap<String, Cost>, CostModelError>>()?; let total_cost = state_variable_costs .values() @@ -255,27 +255,27 @@ impl CostModel { /// /// JSON containing information such as the units (kph, hours, etc) or other /// traversal info (charge events, days traveled, etc) - pub fn serialize_cost_info(&self) -> Result<serde_json::Value, CostError> { + pub fn serialize_cost_info(&self) -> Result<serde_json::Value, CostModelError> { let mut result = serde_json::Map::with_capacity(self.feature_indices.len()); for (name, index) in self.feature_indices.iter() { let weight = self .weights .get(*index) - .ok_or(CostError::CostVectorOutOfBounds( + .ok_or(CostModelError::CostVectorOutOfBounds( *index, String::from(Self::WEIGHTS), ))?; let veh_rate = self.vehicle_rates .get(*index) - .ok_or(CostError::CostVectorOutOfBounds( + .ok_or(CostModelError::CostVectorOutOfBounds( *index, String::from(Self::VEHICLE_RATES), ))?; let net_rate = self.network_rates .get(*index) - .ok_or(CostError::CostVectorOutOfBounds( + .ok_or(CostModelError::CostVectorOutOfBounds( *index, String::from(Self::NETWORK_RATES), ))?; diff --git a/rust/routee-compass-core/src/model/cost/cost_error.rs b/rust/routee-compass-core/src/model/cost/cost_model_error.rs similarity index 82% rename from rust/routee-compass-core/src/model/cost/cost_error.rs rename to rust/routee-compass-core/src/model/cost/cost_model_error.rs index 71a9cc7a..6e67aa42 100644 --- a/rust/routee-compass-core/src/model/cost/cost_error.rs +++ b/rust/routee-compass-core/src/model/cost/cost_model_error.rs @@ -1,10 +1,7 @@ #[derive(thiserror::Error, Debug)] -pub enum CostError { - #[error("failure reading CSV: {source}")] - CsvIoError { - #[from] - source: csv::Error, - }, +pub enum CostModelError { + #[error("{0}")] + BuildError(String), #[error("invalid cost model configuration: {0}")] InvalidConfiguration(String), #[error( diff --git a/rust/routee-compass-core/src/model/cost/cost_ops.rs b/rust/routee-compass-core/src/model/cost/cost_ops.rs index aa6bbd39..db54b22b 100644 --- a/rust/routee-compass-core/src/model/cost/cost_ops.rs +++ b/rust/routee-compass-core/src/model/cost/cost_ops.rs @@ -1,8 +1,8 @@ use super::{ - cost_aggregation::CostAggregation, cost_error::CostError, + cost_aggregation::CostAggregation, cost_model_error::CostModelError, network::network_cost_rate::NetworkCostRate, vehicle::vehicle_cost_rate::VehicleCostRate, }; -use crate::model::{property::edge::Edge, traversal::state::state_variable::StateVar, unit::Cost}; +use crate::model::{network::Edge, traversal::state::state_variable::StateVar, unit::Cost}; /// steps through each state variable and assigns vehicle costs related to that variable /// due to an edge access + traversal event. @@ -17,25 +17,25 @@ pub fn calculate_vehicle_costs( weights: &[f64], rates: &[VehicleCostRate], cost_aggregation: &CostAggregation, -) -> Result<Cost, CostError> { +) -> Result<Cost, CostModelError> { let (prev_state, next_state) = state_sequence; let costs = indices.iter().map(|(name, state_idx)| { // compute delta let prev_state_var = prev_state .get(*state_idx) - .ok_or_else(|| CostError::StateIndexOutOfBounds(*state_idx, name.clone()))?; + .ok_or_else(|| CostModelError::StateIndexOutOfBounds(*state_idx, name.clone()))?; let next_state_var = next_state .get(*state_idx) - .ok_or_else(|| CostError::StateIndexOutOfBounds(*state_idx, name.clone()))?; + .ok_or_else(|| CostModelError::StateIndexOutOfBounds(*state_idx, name.clone()))?; let delta: StateVar = *next_state_var - *prev_state_var; // collect weight and vehicle cost rate let mapping = rates.get(*state_idx).ok_or_else(|| { - CostError::CostVectorOutOfBounds(*state_idx, String::from("vehicle_rates")) + CostModelError::CostVectorOutOfBounds(*state_idx, String::from("vehicle_rates")) + })?; + let weight = weights.get(*state_idx).ok_or_else(|| { + CostModelError::CostVectorOutOfBounds(*state_idx, String::from("weights")) })?; - let weight = weights - .get(*state_idx) - .ok_or_else(|| CostError::CostVectorOutOfBounds(*state_idx, String::from("weights")))?; // compute cost let delta_cost = mapping.map_value(delta); @@ -53,22 +53,22 @@ pub fn calculate_network_traversal_costs( weights: &[f64], rates: &[NetworkCostRate], cost_aggregation: &CostAggregation, -) -> Result<Cost, CostError> { +) -> Result<Cost, CostModelError> { let (prev_state, next_state) = state_sequence; let costs = indices.iter().map(|(name, state_idx)| { let prev_state_var = prev_state .get(*state_idx) - .ok_or_else(|| CostError::StateIndexOutOfBounds(*state_idx, name.clone()))?; + .ok_or_else(|| CostModelError::StateIndexOutOfBounds(*state_idx, name.clone()))?; let next_state_var = next_state .get(*state_idx) - .ok_or_else(|| CostError::StateIndexOutOfBounds(*state_idx, name.clone()))?; + .ok_or_else(|| CostModelError::StateIndexOutOfBounds(*state_idx, name.clone()))?; // determine weight and access cost for this state feature - let weight = weights - .get(*state_idx) - .ok_or_else(|| CostError::CostVectorOutOfBounds(*state_idx, String::from("weights")))?; + let weight = weights.get(*state_idx).ok_or_else(|| { + CostModelError::CostVectorOutOfBounds(*state_idx, String::from("weights")) + })?; let rate = rates.get(*state_idx).ok_or_else(|| { - CostError::CostVectorOutOfBounds(*state_idx, String::from("network_cost_rate")) + CostModelError::CostVectorOutOfBounds(*state_idx, String::from("network_cost_rate")) })?; let access_cost = rate.traversal_cost(*prev_state_var, *next_state_var, edge)?; let cost = access_cost * weight; @@ -85,7 +85,7 @@ pub fn calculate_network_access_costs( weights: &[f64], rates: &[NetworkCostRate], cost_aggregation: &CostAggregation, -) -> Result<Cost, CostError> { +) -> Result<Cost, CostModelError> { let (prev_state, next_state) = state_sequence; let (prev_edge, next_edge) = edge_sequence; let costs = indices.iter().map(|(name, idx)| match rates.get(*idx) { @@ -93,10 +93,10 @@ pub fn calculate_network_access_costs( Some(m) => { let prev_state_var = prev_state .get(*idx) - .ok_or_else(|| CostError::StateIndexOutOfBounds(*idx, name.clone()))?; + .ok_or_else(|| CostModelError::StateIndexOutOfBounds(*idx, name.clone()))?; let next_state_var = next_state .get(*idx) - .ok_or_else(|| CostError::StateIndexOutOfBounds(*idx, name.clone()))?; + .ok_or_else(|| CostModelError::StateIndexOutOfBounds(*idx, name.clone()))?; let access_cost = m.access_cost(*prev_state_var, *next_state_var, prev_edge, next_edge)?; let coefficient = weights.get(*idx).unwrap_or(&1.0); diff --git a/rust/routee-compass-core/src/model/cost/mod.rs b/rust/routee-compass-core/src/model/cost/mod.rs index 6fee0752..ee9ff94f 100644 --- a/rust/routee-compass-core/src/model/cost/mod.rs +++ b/rust/routee-compass-core/src/model/cost/mod.rs @@ -1,6 +1,6 @@ pub mod cost_aggregation; -pub mod cost_error; pub mod cost_model; +pub mod cost_model_error; pub mod cost_ops; pub mod network; pub mod vehicle; diff --git a/rust/routee-compass-core/src/model/cost/network/network_access_cost_row.rs b/rust/routee-compass-core/src/model/cost/network/network_access_cost_row.rs index d073af28..1f1bcb26 100644 --- a/rust/routee-compass-core/src/model/cost/network/network_access_cost_row.rs +++ b/rust/routee-compass-core/src/model/cost/network/network_access_cost_row.rs @@ -1,4 +1,4 @@ -use crate::model::{road_network::edge_id::EdgeId, unit::Cost}; +use crate::model::{network::edge_id::EdgeId, unit::Cost}; use serde::{Deserialize, Serialize}; #[derive(Serialize, Deserialize)] diff --git a/rust/routee-compass-core/src/model/cost/network/network_cost_rate.rs b/rust/routee-compass-core/src/model/cost/network/network_cost_rate.rs index e77cdee0..80303232 100644 --- a/rust/routee-compass-core/src/model/cost/network/network_cost_rate.rs +++ b/rust/routee-compass-core/src/model/cost/network/network_cost_rate.rs @@ -1,7 +1,7 @@ -use crate::model::cost::cost_error::CostError; -use crate::model::property::edge::Edge; +use crate::model::network::Edge; +use crate::model::traversal::state::state_variable::StateVar; use crate::model::unit::Cost; -use crate::model::{road_network::edge_id::EdgeId, traversal::state::state_variable::StateVar}; +use crate::model::{cost::cost_model_error::CostModelError, network::EdgeId}; use serde::{Deserialize, Serialize}; use std::collections::HashMap; @@ -30,7 +30,7 @@ impl NetworkCostRate { _prev_state_var: StateVar, _next_state_var: StateVar, edge: &Edge, - ) -> Result<Cost, CostError> { + ) -> Result<Cost, CostModelError> { match self { NetworkCostRate::Zero => Ok(Cost::ZERO), NetworkCostRate::EdgeEdgeLookup { lookup: _ } => Ok(Cost::ZERO), @@ -42,7 +42,7 @@ impl NetworkCostRate { let mapped = mappings .iter() .map(|f| f.traversal_cost(_prev_state_var, _next_state_var, edge)) - .collect::<Result<Vec<Cost>, CostError>>()?; + .collect::<Result<Vec<Cost>, CostModelError>>()?; let cost = mapped.iter().fold(Cost::ZERO, |a, b| a + *b); Ok(cost) @@ -58,7 +58,7 @@ impl NetworkCostRate { /// * `next_state_var` - the state variable after accessing the next edge origin /// * `prev_edge` - the edge traversed to reach the next_edge (or none if at origin) /// * `next_edge` - the edge we are attempting to access (not yet traversed) - + /// /// # Result /// /// the Cost value for that state, a real number that is aggregated with @@ -69,7 +69,7 @@ impl NetworkCostRate { _next_state_var: StateVar, prev_edge: &Edge, next_edge: &Edge, - ) -> Result<Cost, CostError> { + ) -> Result<Cost, CostModelError> { match self { NetworkCostRate::Zero => Ok(Cost::ZERO), NetworkCostRate::EdgeLookup { lookup: _ } => Ok(Cost::ZERO), @@ -83,7 +83,7 @@ impl NetworkCostRate { let mapped = mappings .iter() .map(|f| f.access_cost(_prev_state_var, _next_state_var, prev_edge, next_edge)) - .collect::<Result<Vec<Cost>, CostError>>()?; + .collect::<Result<Vec<Cost>, CostModelError>>()?; let cost = mapped.iter().fold(Cost::ZERO, |a, b| a + *b); Ok(cost) diff --git a/rust/routee-compass-core/src/model/cost/network/network_cost_rate_builder.rs b/rust/routee-compass-core/src/model/cost/network/network_cost_rate_builder.rs index dbc32000..f421d7e6 100644 --- a/rust/routee-compass-core/src/model/cost/network/network_cost_rate_builder.rs +++ b/rust/routee-compass-core/src/model/cost/network/network_cost_rate_builder.rs @@ -1,5 +1,5 @@ use super::network_cost_rate::NetworkCostRate; -use crate::model::cost::cost_error::CostError; +use crate::model::cost::cost_model_error::CostModelError; use crate::{ model::cost::network::{ network_access_cost_row::NetworkAccessUtilityRow, @@ -22,14 +22,19 @@ pub enum NetworkCostRateBuilder { } impl NetworkCostRateBuilder { - pub fn build(&self) -> Result<NetworkCostRate, CostError> { + pub fn build(&self) -> Result<NetworkCostRate, CostModelError> { use NetworkCostRate as NCM; use NetworkCostRateBuilder as Builder; match self { Builder::EdgeLookupBuilder { cost_input_file } => { let lookup = read_utils::from_csv::<NetworkTraversalUtilityRow>(cost_input_file, true, None) - .map_err(|source| CostError::CsvIoError { source })? + .map_err(|source| { + CostModelError::BuildError(format!( + "failure reading file {}: {}", + cost_input_file, source + )) + })? .iter() .map(|row| (row.edge_id, row.cost)) .collect::<HashMap<_, _>>(); @@ -38,7 +43,12 @@ impl NetworkCostRateBuilder { Builder::EdgeEdgeLookupBuilder { cost_input_file } => { let lookup = read_utils::from_csv::<NetworkAccessUtilityRow>(cost_input_file, true, None) - .map_err(|source| CostError::CsvIoError { source })? + .map_err(|source| { + CostModelError::BuildError(format!( + "failure reading file {}: {}", + cost_input_file, source + )) + })? .iter() .map(|row| ((row.source, row.destination), row.cost)) .collect::<HashMap<_, _>>(); @@ -49,7 +59,7 @@ impl NetworkCostRateBuilder { let mappings = builders .iter() .map(|b| b.build()) - .collect::<Result<Vec<_>, CostError>>()?; + .collect::<Result<Vec<_>, CostModelError>>()?; Ok(NCM::Combined(mappings)) } } diff --git a/rust/routee-compass-core/src/model/cost/network/network_traversal_cost_row.rs b/rust/routee-compass-core/src/model/cost/network/network_traversal_cost_row.rs index 550e42cd..88cc4aa9 100644 --- a/rust/routee-compass-core/src/model/cost/network/network_traversal_cost_row.rs +++ b/rust/routee-compass-core/src/model/cost/network/network_traversal_cost_row.rs @@ -1,4 +1,4 @@ -use crate::model::{road_network::edge_id::EdgeId, unit::Cost}; +use crate::model::{network::edge_id::EdgeId, unit::Cost}; use serde::{Deserialize, Serialize}; #[derive(Serialize, Deserialize)] diff --git a/rust/routee-compass-core/src/model/frontier/frontier_model.rs b/rust/routee-compass-core/src/model/frontier/frontier_model.rs index 71734df7..135704d7 100644 --- a/rust/routee-compass-core/src/model/frontier/frontier_model.rs +++ b/rust/routee-compass-core/src/model/frontier/frontier_model.rs @@ -1,7 +1,6 @@ use super::frontier_model_error::FrontierModelError; use crate::model::{ - property::edge::Edge, state::state_model::StateModel, - traversal::state::state_variable::StateVar, + network::Edge, state::state_model::StateModel, traversal::state::state_variable::StateVar, }; /// Validates edge and traversal states. Provides an API for removing edges from diff --git a/rust/routee-compass-core/src/model/frontier/frontier_model_error.rs b/rust/routee-compass-core/src/model/frontier/frontier_model_error.rs index d177a33b..bdee0c42 100644 --- a/rust/routee-compass-core/src/model/frontier/frontier_model_error.rs +++ b/rust/routee-compass-core/src/model/frontier/frontier_model_error.rs @@ -2,6 +2,6 @@ pub enum FrontierModelError { #[error("failure building frontier model: {0}")] BuildError(String), - #[error("edge id {0} missing from frontier model file")] - MissingIndex(String), + #[error("{0}")] + FrontierModelError(String), } diff --git a/rust/routee-compass-core/src/model/mod.rs b/rust/routee-compass-core/src/model/mod.rs index cf79cd37..c2922796 100644 --- a/rust/routee-compass-core/src/model/mod.rs +++ b/rust/routee-compass-core/src/model/mod.rs @@ -1,8 +1,7 @@ pub mod access; pub mod cost; pub mod frontier; -pub mod property; -pub mod road_network; +pub mod network; pub mod state; pub mod termination; pub mod traversal; diff --git a/rust/routee-compass-core/src/model/property/edge.rs b/rust/routee-compass-core/src/model/network/edge.rs similarity index 93% rename from rust/routee-compass-core/src/model/property/edge.rs rename to rust/routee-compass-core/src/model/network/edge.rs index 6a7a44b4..2f3bd972 100644 --- a/rust/routee-compass-core/src/model/property/edge.rs +++ b/rust/routee-compass-core/src/model/network/edge.rs @@ -1,4 +1,4 @@ -use crate::model::road_network::{edge_id::EdgeId, vertex_id::VertexId}; +use crate::model::network::{edge_id::EdgeId, vertex_id::VertexId}; use crate::model::unit::Distance; use allocative::Allocative; use serde::{Deserialize, Serialize}; diff --git a/rust/routee-compass-core/src/model/road_network/edge_id.rs b/rust/routee-compass-core/src/model/network/edge_id.rs similarity index 100% rename from rust/routee-compass-core/src/model/road_network/edge_id.rs rename to rust/routee-compass-core/src/model/network/edge_id.rs diff --git a/rust/routee-compass-core/src/model/road_network/edge_loader.rs b/rust/routee-compass-core/src/model/network/edge_loader.rs similarity index 84% rename from rust/routee-compass-core/src/model/road_network/edge_loader.rs rename to rust/routee-compass-core/src/model/network/edge_loader.rs index 3e20243f..95c3c6f5 100644 --- a/rust/routee-compass-core/src/model/road_network/edge_loader.rs +++ b/rust/routee-compass-core/src/model/network/edge_loader.rs @@ -1,10 +1,5 @@ -use crate::{ - model::{ - property::edge::Edge, - road_network::{edge_id::EdgeId, graph_error::GraphError, vertex_id::VertexId}, - }, - util::{compact_ordered_hash_map::CompactOrderedHashMap, fs::read_utils}, -}; +use super::{Edge, EdgeId, NetworkError, VertexId}; +use crate::util::{compact_ordered_hash_map::CompactOrderedHashMap, fs::read_utils}; use kdam::Bar; use kdam::BarExt; use std::{collections::HashSet, path::PathBuf}; @@ -22,7 +17,7 @@ pub struct EdgeLoaderConfig { } impl TryFrom<EdgeLoaderConfig> for EdgeLoader { - type Error = GraphError; + type Error = NetworkError; fn try_from(c: EdgeLoaderConfig) -> Result<Self, Self::Error> { let mut adj: Vec<CompactOrderedHashMap<EdgeId, VertexId>> = @@ -35,7 +30,9 @@ impl TryFrom<EdgeLoaderConfig> for EdgeLoader { .animation("fillup") .desc("edge list") .build() - .map_err(|e| GraphError::ProgressBarBuildError(String::from("edge list"), e))?; + .map_err(|e| { + NetworkError::InternalError(format!("could not build progress bar: {}", e)) + })?; let mut missing_vertices: HashSet<VertexId> = HashSet::new(); let cb = Box::new(|edge: &Edge| { @@ -61,7 +58,7 @@ impl TryFrom<EdgeLoaderConfig> for EdgeLoader { let edges = read_utils::from_csv(&c.edge_list_csv, true, Some(cb))?; - println!(); + eprintln!(); let result = EdgeLoader { edges, adj: adj.into_boxed_slice(), diff --git a/rust/routee-compass-core/src/model/road_network/graph.rs b/rust/routee-compass-core/src/model/network/graph.rs similarity index 74% rename from rust/routee-compass-core/src/model/road_network/graph.rs rename to rust/routee-compass-core/src/model/network/graph.rs index efe53dbc..402b55bc 100644 --- a/rust/routee-compass-core/src/model/road_network/graph.rs +++ b/rust/routee-compass-core/src/model/network/graph.rs @@ -1,14 +1,12 @@ +use super::{Edge, EdgeId, NetworkError, Vertex, VertexId}; use crate::algorithm::search::direction::Direction; -use crate::model::property::edge::Edge; -use crate::model::property::vertex::Vertex; -use crate::model::road_network::graph_error::GraphError; -use crate::model::road_network::{edge_id::EdgeId, vertex_id::VertexId}; use crate::util::compact_ordered_hash_map::CompactOrderedHashMap; use std::path::Path; use super::graph_loader::graph_from_files; use allocative::Allocative; +use itertools::Itertools; /// Road network topology represented as an adjacency list. /// The `EdgeId` and `VertexId` values correspond to edge and @@ -61,7 +59,7 @@ impl Graph { n_edges: Option<usize>, n_vertices: Option<usize>, verbose: Option<bool>, - ) -> Result<Graph, GraphError> { + ) -> Result<Graph, NetworkError> { graph_from_files(edge_list_csv, vertex_list_csv, n_edges, n_vertices, verbose) } /// number of edges in the Graph @@ -99,9 +97,9 @@ impl Graph { /// # Returns /// /// The associated `Edge` or an error if the id is missing - pub fn get_edge(&self, edge_id: EdgeId) -> Result<&Edge, GraphError> { + pub fn get_edge(&self, edge_id: &EdgeId) -> Result<&Edge, NetworkError> { match self.edges.get(edge_id.0) { - None => Err(GraphError::EdgeAttributeNotFound { edge_id }), + None => Err(NetworkError::EdgeNotFound(*edge_id)), Some(edge) => Ok(edge), } } @@ -115,9 +113,9 @@ impl Graph { /// # Returns /// /// The associated `Vertex` or an error if the id is missing - pub fn get_vertex(&self, vertex_id: VertexId) -> Result<&Vertex, GraphError> { + pub fn get_vertex(&self, vertex_id: &VertexId) -> Result<&Vertex, NetworkError> { match self.vertices.get(vertex_id.0) { - None => Err(GraphError::VertexAttributeNotFound { vertex_id }), + None => Err(NetworkError::VertexNotFound(*vertex_id)), Some(vertex) => Ok(vertex), } } @@ -132,26 +130,17 @@ impl Graph { /// /// A list of `EdgeIds` for outbound edges that leave this `VertexId`, or an error /// if the vertex is missing from the Graph adjacency matrix. - pub fn out_edges(&self, src: VertexId) -> Result<Vec<EdgeId>, GraphError> { - match self.adj.get(src.0) { - None => Err(GraphError::VertexWithoutOutEdges { vertex_id: src }), - Some(out_map) => { - let edge_ids = out_map.keys().cloned().collect(); - Ok(edge_ids) - } - } + pub fn out_edges(&self, src: &VertexId) -> Vec<EdgeId> { + self.out_edges_iter(src).cloned().collect_vec() } pub fn out_edges_iter<'a>( &'a self, - src: VertexId, - ) -> Result<Box<dyn Iterator<Item = &EdgeId> + 'a>, GraphError> { + src: &VertexId, + ) -> Box<dyn Iterator<Item = &'a EdgeId> + 'a> { match self.adj.get(src.0) { - None => Err(GraphError::VertexWithoutOutEdges { vertex_id: src }), - Some(out_map) => { - let edge_ids = out_map.keys(); - Ok(edge_ids) - } + None => Box::new(std::iter::empty()), + Some(out_map) => out_map.keys(), } } @@ -165,26 +154,17 @@ impl Graph { /// /// A list of `EdgeIds` for inbound edges that arrive at this `VertexId`, or an error /// if the vertex is missing from the Graph adjacency matrix. - pub fn in_edges(&self, dst: VertexId) -> Result<Vec<EdgeId>, GraphError> { - match self.rev.get(dst.0) { - None => Err(GraphError::VertexWithoutInEdges { vertex_id: dst }), - Some(in_map) => { - let edge_ids = in_map.keys().cloned().collect(); - Ok(edge_ids) - } - } + pub fn in_edges(&self, dst: &VertexId) -> Vec<EdgeId> { + self.in_edges_iter(dst).cloned().collect_vec() } pub fn in_edges_iter<'a>( &'a self, - dst: VertexId, - ) -> Result<Box<dyn Iterator<Item = &EdgeId> + 'a>, GraphError> { + dst: &VertexId, + ) -> Box<dyn Iterator<Item = &'a EdgeId> + 'a> { match self.rev.get(dst.0) { - None => Err(GraphError::VertexWithoutInEdges { vertex_id: dst }), - Some(in_map) => { - let edge_ids = in_map.keys(); - Ok(edge_ids) - } + None => Box::new(std::iter::empty()), + Some(out_map) => out_map.keys(), } } @@ -197,7 +177,7 @@ impl Graph { /// # Returns /// /// The source `VertexId` of an `Edge` or an error if the edge is missing - pub fn src_vertex_id(&self, edge_id: EdgeId) -> Result<VertexId, GraphError> { + pub fn src_vertex_id(&self, edge_id: &EdgeId) -> Result<VertexId, NetworkError> { self.get_edge(edge_id).map(|e| e.src_vertex_id) } @@ -210,7 +190,7 @@ impl Graph { /// # Returns /// /// The destination `VertexId` of an `Edge` or an error if the edge is missing - pub fn dst_vertex_id(&self, edge_id: EdgeId) -> Result<VertexId, GraphError> { + pub fn dst_vertex_id(&self, edge_id: &EdgeId) -> Result<VertexId, NetworkError> { self.get_edge(edge_id).map(|e| e.dst_vertex_id) } @@ -225,17 +205,35 @@ impl Graph { /// # Returns /// /// The incident `EdgeId`s or an error if the vertex is not connected. - pub fn incident_edges( - &self, - vertex_id: VertexId, - direction: Direction, - ) -> Result<Vec<EdgeId>, GraphError> { + pub fn incident_edges(&self, vertex_id: &VertexId, direction: &Direction) -> Vec<EdgeId> { match direction { Direction::Forward => self.out_edges(vertex_id), Direction::Reverse => self.in_edges(vertex_id), } } + /// helper function to give incident edges to a vertex based on a + /// traversal direction. + /// + /// # Arguments + /// + /// * `vertex_id` - vertex to find edges which connect to it + /// * `direction` - whether to find out edges (Forward) or in edges (Reverse) + /// + /// # Returns + /// + /// The incident `EdgeId`s or an error if the vertex is not connected. + pub fn incident_edges_iter<'a>( + &'a self, + vertex_id: &VertexId, + direction: &Direction, + ) -> Box<dyn Iterator<Item = &'a EdgeId> + 'a> { + match direction { + Direction::Forward => self.out_edges_iter(vertex_id), + Direction::Reverse => self.in_edges_iter(vertex_id), + } + } + /// helper function to give the incident vertex to an edge based on a /// traversal direction. /// @@ -249,9 +247,9 @@ impl Graph { /// The incident `VertexId` of an edge or an error if the edge is missing pub fn incident_vertex( &self, - edge_id: EdgeId, - direction: Direction, - ) -> Result<VertexId, GraphError> { + edge_id: &EdgeId, + direction: &Direction, + ) -> Result<VertexId, NetworkError> { match direction { Direction::Forward => self.dst_vertex_id(edge_id), Direction::Reverse => self.src_vertex_id(edge_id), @@ -268,13 +266,13 @@ impl Graph { /// /// The triplet of attributes surrounding one `Edge` or an error if /// any id is invalid. - pub fn edge_triplet_attrs( + pub fn edge_triplet( &self, - edge_id: EdgeId, - ) -> Result<(&Vertex, &Edge, &Vertex), GraphError> { + edge_id: &EdgeId, + ) -> Result<(&Vertex, &Edge, &Vertex), NetworkError> { let edge = self.get_edge(edge_id)?; - let src = self.get_vertex(edge.src_vertex_id)?; - let dst = self.get_vertex(edge.dst_vertex_id)?; + let src = self.get_vertex(&edge.src_vertex_id)?; + let dst = self.get_vertex(&edge.dst_vertex_id)?; Ok((src, edge, dst)) } @@ -294,16 +292,15 @@ impl Graph { /// /// For each edge connected to this `vertex_id` traversable via the provided `direction`, /// a triplet of the `EdgeId` and it's connecting `VertexId`s. - pub fn incident_triplets( + pub fn incident_triplet_ids( &self, - vertex_id: VertexId, - direction: Direction, - ) -> Result<Vec<(VertexId, EdgeId, VertexId)>, GraphError> { - self.incident_edges(vertex_id, direction)? - .into_iter() + vertex_id: &VertexId, + direction: &Direction, + ) -> Result<Vec<(VertexId, EdgeId, VertexId)>, NetworkError> { + self.incident_edges_iter(vertex_id, direction) .map(|edge_id| { let terminal_vid = self.incident_vertex(edge_id, direction)?; - Ok((vertex_id, edge_id, terminal_vid)) + Ok((*vertex_id, *edge_id, terminal_vid)) }) .collect() } @@ -325,15 +322,15 @@ impl Graph { /// a triplet of the `Edge` and it's connecting `Vertex`s. pub fn incident_triplet_attributes( &self, - vertex_id: VertexId, - direction: Direction, - ) -> Result<Vec<(&Vertex, &Edge, &Vertex)>, GraphError> { - self.incident_triplets(vertex_id, direction)? + vertex_id: &VertexId, + direction: &Direction, + ) -> Result<Vec<(&Vertex, &Edge, &Vertex)>, NetworkError> { + self.incident_triplet_ids(vertex_id, direction)? .iter() .map(|(src_id, edge_id, dst_id)| { - let src = self.get_vertex(*src_id)?; - let edge = self.get_edge(*edge_id)?; - let dst = self.get_vertex(*dst_id)?; + let src = self.get_vertex(src_id)?; + let edge = self.get_edge(edge_id)?; + let dst = self.get_vertex(dst_id)?; Ok((src, edge, dst)) }) .collect() diff --git a/rust/routee-compass-core/src/model/road_network/graph_loader.rs b/rust/routee-compass-core/src/model/network/graph_loader.rs similarity index 76% rename from rust/routee-compass-core/src/model/road_network/graph_loader.rs rename to rust/routee-compass-core/src/model/network/graph_loader.rs index bdbd9941..a7f44211 100644 --- a/rust/routee-compass-core/src/model/road_network/graph_loader.rs +++ b/rust/routee-compass-core/src/model/network/graph_loader.rs @@ -1,15 +1,11 @@ -use std::path::Path; - -use log::warn; - -use crate::{model::property::vertex::Vertex, util::fs::fs_utils::line_count}; - use super::{ edge_loader::{EdgeLoader, EdgeLoaderConfig}, - graph::Graph, - graph_error::GraphError, vertex_loader::VertexLoaderConfig, + Graph, NetworkError, Vertex, }; +use crate::util::fs::fs_utils::line_count; +use log::warn; +use std::path::Path; pub fn graph_from_files<P: AsRef<Path>>( edge_list_csv: &P, @@ -17,7 +13,7 @@ pub fn graph_from_files<P: AsRef<Path>>( n_edges: Option<usize>, n_vertices: Option<usize>, verbose: Option<bool>, -) -> Result<Graph, GraphError> { +) -> Result<Graph, NetworkError> { let verbose = verbose.unwrap_or(false); let n_edges = match n_edges { Some(n) => n, @@ -63,7 +59,7 @@ pub fn graph_from_files<P: AsRef<Path>>( Ok(graph) } -fn get_n_edges<P: AsRef<Path>>(edge_list_csv: &P) -> Result<usize, GraphError> { +fn get_n_edges<P: AsRef<Path>>(edge_list_csv: &P) -> Result<usize, NetworkError> { // check if the extension is .gz let is_gzip = edge_list_csv .as_ref() @@ -73,14 +69,16 @@ fn get_n_edges<P: AsRef<Path>>(edge_list_csv: &P) -> Result<usize, GraphError> { .unwrap_or(false); let n = line_count(edge_list_csv, is_gzip)?; if n < 1 { - return Err(GraphError::EmptyFileSource { - filename: edge_list_csv.as_ref().to_path_buf(), - }); + let msg = match edge_list_csv.as_ref().to_str() { + Some(path_str) => format!("edge file {} is empty", path_str), + None => String::from("edge file is empty"), + }; + return Err(NetworkError::DatasetError(msg)); } Ok(n - 1) // drop count of header line } -fn get_n_vertices<P: AsRef<Path>>(vertex_list_csv: &P) -> Result<usize, GraphError> { +fn get_n_vertices<P: AsRef<Path>>(vertex_list_csv: &P) -> Result<usize, NetworkError> { let is_gzip = vertex_list_csv .as_ref() .to_path_buf() @@ -89,9 +87,11 @@ fn get_n_vertices<P: AsRef<Path>>(vertex_list_csv: &P) -> Result<usize, GraphErr .unwrap_or(false); let n = line_count(vertex_list_csv, is_gzip)?; if n < 1 { - return Err(GraphError::EmptyFileSource { - filename: vertex_list_csv.as_ref().to_path_buf(), - }); + let msg = match vertex_list_csv.as_ref().to_str() { + Some(path_str) => format!("vertex file {} is empty", path_str), + None => String::from("vertex file is empty"), + }; + return Err(NetworkError::DatasetError(msg)); } Ok(n - 1) // drop count of header line } diff --git a/rust/routee-compass-core/src/model/network/mod.rs b/rust/routee-compass-core/src/model/network/mod.rs new file mode 100644 index 00000000..82e63b5d --- /dev/null +++ b/rust/routee-compass-core/src/model/network/mod.rs @@ -0,0 +1,16 @@ +pub mod edge; +pub mod edge_id; +pub mod edge_loader; +pub mod graph; +pub mod graph_loader; +pub mod network_error; +pub mod vertex; +pub mod vertex_id; +pub mod vertex_loader; + +pub use edge::Edge; +pub use edge_id::EdgeId; +pub use graph::Graph; +pub use network_error::NetworkError; +pub use vertex::Vertex; +pub use vertex_id::VertexId; diff --git a/rust/routee-compass-core/src/model/network/network_error.rs b/rust/routee-compass-core/src/model/network/network_error.rs new file mode 100644 index 00000000..96c4ccae --- /dev/null +++ b/rust/routee-compass-core/src/model/network/network_error.rs @@ -0,0 +1,25 @@ +use crate::model::network::{edge_id::EdgeId, vertex_id::VertexId}; + +#[derive(thiserror::Error, Debug)] +pub enum NetworkError { + #[error("edge attribute not found for edge {0}")] + EdgeNotFound(EdgeId), + #[error("vertex attribute not found for vertex {0}")] + VertexNotFound(VertexId), + #[error("Error with graph attribute {0}: {1}")] + AttributeError(String, String), + #[error("error with provided dataset: {0}")] + DatasetError(String), + #[error("failure reading graph data from file: {source}")] + IOError { + #[from] + source: std::io::Error, + }, + #[error("failure reading graph data from CSV: {source}")] + CsvError { + #[from] + source: csv::Error, + }, + #[error("{0}")] + InternalError(String), +} diff --git a/rust/routee-compass-core/src/model/property/vertex.rs b/rust/routee-compass-core/src/model/network/vertex.rs similarity index 94% rename from rust/routee-compass-core/src/model/property/vertex.rs rename to rust/routee-compass-core/src/model/network/vertex.rs index f0d8138a..8c711453 100644 --- a/rust/routee-compass-core/src/model/property/vertex.rs +++ b/rust/routee-compass-core/src/model/network/vertex.rs @@ -1,4 +1,6 @@ -use crate::{model::road_network::vertex_id::VertexId, util::geo::coord::InternalCoord}; +use std::fmt::Display; + +use crate::{model::network::vertex_id::VertexId, util::geo::coord::InternalCoord}; use allocative::Allocative; use geo::coord; use serde::de; @@ -32,6 +34,12 @@ impl Vertex { } } +impl Display for Vertex { + fn fmt(&self, f: &mut std::fmt::Formatter<'_>) -> std::fmt::Result { + write!(f, "Vertex {} ({},{})", self.vertex_id, self.x(), self.y()) + } +} + const VERTEX_ID: &str = "vertex_id"; const X_COORDINATE: &str = "x"; const Y_COORDINATE: &str = "y"; @@ -132,7 +140,7 @@ mod tests { use geo::Coord; use crate::{ - model::{property::vertex::Vertex, road_network::vertex_id::VertexId}, + model::network::{Vertex, VertexId}, util::geo::coord::InternalCoord, }; @@ -141,7 +149,7 @@ mod tests { let filepath = PathBuf::from(env!("CARGO_MANIFEST_DIR")) .join("src") .join("model") - .join("property") + .join("network") .join("test") .join("vertices.csv"); diff --git a/rust/routee-compass-core/src/model/road_network/vertex_id.rs b/rust/routee-compass-core/src/model/network/vertex_id.rs similarity index 100% rename from rust/routee-compass-core/src/model/road_network/vertex_id.rs rename to rust/routee-compass-core/src/model/network/vertex_id.rs diff --git a/rust/routee-compass-core/src/model/road_network/vertex_loader.rs b/rust/routee-compass-core/src/model/network/vertex_loader.rs similarity index 75% rename from rust/routee-compass-core/src/model/road_network/vertex_loader.rs rename to rust/routee-compass-core/src/model/network/vertex_loader.rs index 9b2aa55d..c97dc0a5 100644 --- a/rust/routee-compass-core/src/model/road_network/vertex_loader.rs +++ b/rust/routee-compass-core/src/model/network/vertex_loader.rs @@ -1,7 +1,6 @@ use std::path::PathBuf; -use crate::model::property::vertex::Vertex; -use crate::model::road_network::graph_error::GraphError; +use super::{NetworkError, Vertex}; use crate::util::fs::read_utils; use kdam::{Bar, BarExt}; @@ -11,7 +10,7 @@ pub struct VertexLoaderConfig { } impl TryFrom<VertexLoaderConfig> for Box<[Vertex]> { - type Error = GraphError; + type Error = NetworkError; fn try_from(conf: VertexLoaderConfig) -> Result<Self, Self::Error> { let mut processed: usize = 0; @@ -20,7 +19,9 @@ impl TryFrom<VertexLoaderConfig> for Box<[Vertex]> { .animation("fillup") .desc("vertex list") .build() - .map_err(|e| GraphError::ProgressBarBuildError(String::from("vertex list"), e))?; + .map_err(|e| { + NetworkError::InternalError(format!("could not build progress bar: {}", e)) + })?; let cb = Box::new(|_v: &Vertex| { let _ = pb.update(1); @@ -28,7 +29,7 @@ impl TryFrom<VertexLoaderConfig> for Box<[Vertex]> { }); let result: Box<[Vertex]> = read_utils::from_csv(&conf.vertex_list_csv, true, Some(cb))?; - println!(); + eprintln!(); Ok(result) } } diff --git a/rust/routee-compass-core/src/model/property/mod.rs b/rust/routee-compass-core/src/model/property/mod.rs deleted file mode 100644 index dbda5dc7..00000000 --- a/rust/routee-compass-core/src/model/property/mod.rs +++ /dev/null @@ -1,2 +0,0 @@ -pub mod edge; -pub mod vertex; diff --git a/rust/routee-compass-core/src/model/road_network/graph_error.rs b/rust/routee-compass-core/src/model/road_network/graph_error.rs deleted file mode 100644 index bc89eb51..00000000 --- a/rust/routee-compass-core/src/model/road_network/graph_error.rs +++ /dev/null @@ -1,39 +0,0 @@ -use std::path::PathBuf; - -use crate::model::road_network::{edge_id::EdgeId, vertex_id::VertexId}; - -#[derive(thiserror::Error, Debug)] -pub enum GraphError { - #[error("edge {edge_id} not found")] - EdgeIdNotFound { edge_id: EdgeId }, - #[error("edge attribute not found for edge {edge_id}")] - EdgeAttributeNotFound { edge_id: EdgeId }, - #[error("vertex {vertex_id} not found")] - VertexIdNotFound { vertex_id: VertexId }, - #[error("vertex attribute not found for vertex {vertex_id}")] - VertexAttributeNotFound { vertex_id: VertexId }, - #[error("vertex without out edges in graph")] - VertexWithoutOutEdges { vertex_id: VertexId }, - #[error("vertex without in edges in graph")] - VertexWithoutInEdges { vertex_id: VertexId }, - #[error("error in test setup")] - TestError, - #[error("Error with graph attribute {0}: {1}")] - AttributeError(String, String), - #[error("{filename} file source was empty")] - EmptyFileSource { filename: PathBuf }, - #[error("failure reading TomTom graph: {source}")] - IOError { - #[from] - source: std::io::Error, - }, - #[error("csv error: {source}")] - CsvError { - #[from] - source: csv::Error, - }, - #[error("internal error: adjacency list missing vertex {0}")] - AdjacencyVertexMissing(VertexId), - #[error("error creating progress bar for {0}: {1}")] - ProgressBarBuildError(String, String), -} diff --git a/rust/routee-compass-core/src/model/road_network/mod.rs b/rust/routee-compass-core/src/model/road_network/mod.rs deleted file mode 100644 index 2e41655a..00000000 --- a/rust/routee-compass-core/src/model/road_network/mod.rs +++ /dev/null @@ -1,7 +0,0 @@ -pub mod edge_id; -pub mod edge_loader; -pub mod graph; -pub mod graph_error; -pub mod graph_loader; -pub mod vertex_id; -pub mod vertex_loader; diff --git a/rust/routee-compass-core/src/model/state/custom_feature_format.rs b/rust/routee-compass-core/src/model/state/custom_feature_format.rs index 2243a3e5..e347d0ac 100644 --- a/rust/routee-compass-core/src/model/state/custom_feature_format.rs +++ b/rust/routee-compass-core/src/model/state/custom_feature_format.rs @@ -1,6 +1,6 @@ use std::fmt::Display; -use super::{state_error::StateError, unit_codec_name::UnitCodecType}; +use super::{state_model_error::StateModelError, unit_codec_name::UnitCodecType}; use crate::model::traversal::state::state_variable::StateVar; use ordered_float::OrderedFloat; use serde::{Deserialize, Serialize}; @@ -63,7 +63,7 @@ impl CustomFeatureFormat { } } - pub fn initial(&self) -> Result<StateVar, StateError> { + pub fn initial(&self) -> Result<StateVar, StateModelError> { match self { CustomFeatureFormat::FloatingPoint { initial } => self.encode_f64(initial), CustomFeatureFormat::SignedInteger { initial } => self.encode_i64(initial), @@ -72,37 +72,37 @@ impl CustomFeatureFormat { } } - pub fn encode_f64(&self, value: &f64) -> Result<StateVar, StateError> { + pub fn encode_f64(&self, value: &f64) -> Result<StateVar, StateModelError> { match self { CustomFeatureFormat::FloatingPoint { initial: _ } => Ok(StateVar(*value)), - _ => Err(StateError::EncodeError( + _ => Err(StateModelError::EncodeError( UnitCodecType::FloatingPoint.to_string(), self.name(), )), } } - pub fn encode_i64(&self, value: &i64) -> Result<StateVar, StateError> { + pub fn encode_i64(&self, value: &i64) -> Result<StateVar, StateModelError> { match self { CustomFeatureFormat::SignedInteger { initial: _ } => Ok(StateVar(*value as f64)), - _ => Err(StateError::EncodeError( + _ => Err(StateModelError::EncodeError( UnitCodecType::SignedInteger.to_string(), self.name(), )), } } - pub fn encode_u64(&self, value: &u64) -> Result<StateVar, StateError> { + pub fn encode_u64(&self, value: &u64) -> Result<StateVar, StateModelError> { match self { CustomFeatureFormat::UnsignedInteger { initial: _ } => Ok(StateVar(*value as f64)), - _ => Err(StateError::EncodeError( + _ => Err(StateModelError::EncodeError( UnitCodecType::UnsignedInteger.to_string(), self.name(), )), } } - pub fn encode_bool(&self, value: &bool) -> Result<StateVar, StateError> { + pub fn encode_bool(&self, value: &bool) -> Result<StateVar, StateModelError> { match self { CustomFeatureFormat::Boolean { initial: _ } => { if *value { @@ -111,38 +111,38 @@ impl CustomFeatureFormat { Ok(StateVar(0.0)) } } - _ => Err(StateError::EncodeError( + _ => Err(StateModelError::EncodeError( UnitCodecType::Boolean.to_string(), self.name(), )), } } - pub fn decode_f64(&self, value: &StateVar) -> Result<f64, StateError> { + pub fn decode_f64(&self, value: &StateVar) -> Result<f64, StateModelError> { match self { CustomFeatureFormat::FloatingPoint { initial: _ } => Ok(value.0), - _ => Err(StateError::DecodeError( + _ => Err(StateModelError::DecodeError( *value, UnitCodecType::FloatingPoint.to_string(), self.name(), )), } } - pub fn decode_i64(&self, value: &StateVar) -> Result<i64, StateError> { + pub fn decode_i64(&self, value: &StateVar) -> Result<i64, StateModelError> { match self { CustomFeatureFormat::SignedInteger { initial: _ } => Ok(value.0 as i64), - _ => Err(StateError::DecodeError( + _ => Err(StateModelError::DecodeError( *value, UnitCodecType::SignedInteger.to_string(), self.name(), )), } } - pub fn decode_u64(&self, value: &StateVar) -> Result<u64, StateError> { + pub fn decode_u64(&self, value: &StateVar) -> Result<u64, StateModelError> { match self { CustomFeatureFormat::UnsignedInteger { initial: _ } => { if value < &StateVar::ZERO { - Err(StateError::ValueError( + Err(StateModelError::ValueError( *value, UnitCodecType::UnsignedInteger.to_string(), )) @@ -150,14 +150,14 @@ impl CustomFeatureFormat { Ok(value.0 as u64) } } - _ => Err(StateError::DecodeError( + _ => Err(StateModelError::DecodeError( *value, UnitCodecType::UnsignedInteger.to_string(), self.name(), )), } } - pub fn decode_bool(&self, value: &StateVar) -> Result<bool, StateError> { + pub fn decode_bool(&self, value: &StateVar) -> Result<bool, StateModelError> { match self { CustomFeatureFormat::Boolean { initial: _ } => { let is_false = value.0 == 0.0; @@ -167,7 +167,7 @@ impl CustomFeatureFormat { Ok(true) } } - _ => Err(StateError::DecodeError( + _ => Err(StateModelError::DecodeError( *value, UnitCodecType::Boolean.to_string(), self.name(), diff --git a/rust/routee-compass-core/src/model/state/mod.rs b/rust/routee-compass-core/src/model/state/mod.rs index 8514bab2..e4ef7d2e 100644 --- a/rust/routee-compass-core/src/model/state/mod.rs +++ b/rust/routee-compass-core/src/model/state/mod.rs @@ -1,7 +1,7 @@ pub mod custom_feature_format; pub mod indexed_state_feature; -pub mod state_error; pub mod state_feature; pub mod state_model; +pub mod state_model_error; pub mod unit_codec_name; pub mod update_operation; diff --git a/rust/routee-compass-core/src/model/state/state_feature.rs b/rust/routee-compass-core/src/model/state/state_feature.rs index dc3f5edc..dbda3cc1 100644 --- a/rust/routee-compass-core/src/model/state/state_feature.rs +++ b/rust/routee-compass-core/src/model/state/state_feature.rs @@ -1,6 +1,6 @@ use std::fmt::Display; -use super::{custom_feature_format::CustomFeatureFormat, state_error::StateError}; +use super::{custom_feature_format::CustomFeatureFormat, state_model_error::StateModelError}; use crate::model::{traversal::state::state_variable::StateVar, unit}; use serde::{Deserialize, Serialize}; @@ -195,7 +195,7 @@ impl StateFeature { } } - pub fn get_initial(&self) -> Result<StateVar, StateError> { + pub fn get_initial(&self) -> Result<StateVar, StateModelError> { match self { StateFeature::Distance { distance_unit: _, @@ -217,53 +217,53 @@ impl StateFeature { } } - pub fn get_distance_unit(&self) -> Result<unit::DistanceUnit, StateError> { + pub fn get_distance_unit(&self) -> Result<unit::DistanceUnit, StateModelError> { match self { StateFeature::Distance { distance_unit: unit, initial: _, } => Ok(*unit), - _ => Err(StateError::UnexpectedFeatureUnit( + _ => Err(StateModelError::UnexpectedFeatureUnit( String::from("distance"), self.get_feature_type(), )), } } - pub fn get_time_unit(&self) -> Result<unit::TimeUnit, StateError> { + pub fn get_time_unit(&self) -> Result<unit::TimeUnit, StateModelError> { match self { StateFeature::Time { time_unit: unit, initial: _, } => Ok(*unit), - _ => Err(StateError::UnexpectedFeatureUnit( + _ => Err(StateModelError::UnexpectedFeatureUnit( String::from("time"), self.get_feature_type(), )), } } - pub fn get_energy_unit(&self) -> Result<unit::EnergyUnit, StateError> { + pub fn get_energy_unit(&self) -> Result<unit::EnergyUnit, StateModelError> { match self { StateFeature::Energy { energy_unit, initial: _, } => Ok(*energy_unit), - _ => Err(StateError::UnexpectedFeatureUnit( + _ => Err(StateModelError::UnexpectedFeatureUnit( String::from("energy"), self.get_feature_type(), )), } } - pub fn get_custom_feature_format(&self) -> Result<&CustomFeatureFormat, StateError> { + pub fn get_custom_feature_format(&self) -> Result<&CustomFeatureFormat, StateModelError> { match self { StateFeature::Custom { r#type: _, unit: _, format, } => Ok(format), - _ => Err(StateError::UnexpectedFeatureUnit( + _ => Err(StateModelError::UnexpectedFeatureUnit( self.get_feature_unit_name(), self.get_feature_type(), )), diff --git a/rust/routee-compass-core/src/model/state/state_model.rs b/rust/routee-compass-core/src/model/state/state_model.rs index cb8a9f0e..93ee9d66 100644 --- a/rust/routee-compass-core/src/model/state/state_model.rs +++ b/rust/routee-compass-core/src/model/state/state_model.rs @@ -1,6 +1,6 @@ use super::{ - custom_feature_format::CustomFeatureFormat, state_error::StateError, - state_feature::StateFeature, update_operation::UpdateOperation, + custom_feature_format::CustomFeatureFormat, state_feature::StateFeature, + state_model_error::StateModelError, update_operation::UpdateOperation, }; use crate::util::compact_ordered_hash_map::CompactOrderedHashMap; use crate::{ @@ -44,7 +44,10 @@ impl StateModel { /// /// # Arguments /// * `query` - JSON search query contents containing state model information - pub fn extend(&self, entries: Vec<(String, StateFeature)>) -> Result<StateModel, StateError> { + pub fn extend( + &self, + entries: Vec<(String, StateFeature)>, + ) -> Result<StateModel, StateModelError> { let mut map = self .0 .iter() @@ -64,7 +67,7 @@ impl StateModel { .iter() .map(|(k, old, new)| format!("{} old: {} | new: {}", k, old, new)) .join(", "); - Err(StateError::BuildError(format!( + Err(StateModelError::BuildError(format!( "new state features overwriting existing: {}", msg ))) @@ -105,7 +108,7 @@ impl StateModel { /// # Returns /// /// an initialized, "zero"-valued traversal state, or an error - pub fn initial_state(&self) -> Result<Vec<StateVar>, StateError> { + pub fn initial_state(&self) -> Result<Vec<StateVar>, StateModelError> { self.0 .iter() .map(|(_, feature)| { @@ -130,7 +133,7 @@ impl StateModel { state: &[StateVar], name: &String, unit: &DistanceUnit, - ) -> Result<Distance, StateError> { + ) -> Result<Distance, StateModelError> { let value = self.get_state_variable(state, name)?; let feature = self.get_feature(name)?; let result = feature.get_distance_unit()?.convert(&value.into(), unit); @@ -151,7 +154,7 @@ impl StateModel { state: &[StateVar], name: &String, unit: &TimeUnit, - ) -> Result<Time, StateError> { + ) -> Result<Time, StateModelError> { let value = self.get_state_variable(state, name)?; let feature = self.get_feature(name)?; let result = feature.get_time_unit()?.convert(&value.into(), unit); @@ -172,7 +175,7 @@ impl StateModel { state: &[StateVar], name: &String, unit: &EnergyUnit, - ) -> Result<Energy, StateError> { + ) -> Result<Energy, StateModelError> { let value = self.get_state_variable(state, name)?; let feature = self.get_feature(name)?; let result = feature.get_energy_unit()?.convert(&value.into(), unit); @@ -187,7 +190,11 @@ impl StateModel { /// # Returns /// /// the expected value or an error - pub fn get_custom_f64(&self, state: &[StateVar], name: &String) -> Result<f64, StateError> { + pub fn get_custom_f64( + &self, + state: &[StateVar], + name: &String, + ) -> Result<f64, StateModelError> { let (value, format) = self.get_custom_state_variable(state, name)?; let result = format.decode_f64(&value)?; Ok(result) @@ -201,7 +208,11 @@ impl StateModel { /// # Returns /// /// the expected value or an error - pub fn get_custom_i64(&self, state: &[StateVar], name: &String) -> Result<i64, StateError> { + pub fn get_custom_i64( + &self, + state: &[StateVar], + name: &String, + ) -> Result<i64, StateModelError> { let (value, format) = self.get_custom_state_variable(state, name)?; let result = format.decode_i64(&value)?; Ok(result) @@ -215,7 +226,11 @@ impl StateModel { /// # Returns /// /// the expected value or an error - pub fn get_custom_u64(&self, state: &[StateVar], name: &String) -> Result<u64, StateError> { + pub fn get_custom_u64( + &self, + state: &[StateVar], + name: &String, + ) -> Result<u64, StateModelError> { let (value, format) = self.get_custom_state_variable(state, name)?; let result = format.decode_u64(&value)?; Ok(result) @@ -229,7 +244,11 @@ impl StateModel { /// # Returns /// /// the expected value or an error - pub fn get_custom_bool(&self, state: &[StateVar], name: &String) -> Result<bool, StateError> { + pub fn get_custom_bool( + &self, + state: &[StateVar], + name: &String, + ) -> Result<bool, StateModelError> { let (value, format) = self.get_custom_state_variable(state, name)?; let result = format.decode_bool(&value)?; Ok(result) @@ -250,7 +269,7 @@ impl StateModel { &self, state: &[StateVar], name: &String, - ) -> Result<(StateVar, &CustomFeatureFormat), StateError> { + ) -> Result<(StateVar, &CustomFeatureFormat), StateModelError> { let value = self.get_state_variable(state, name)?; let feature = self.get_feature(name)?; let format = feature.get_custom_feature_format()?; @@ -273,7 +292,7 @@ impl StateModel { prev: &[StateVar], next: &[StateVar], name: &String, - ) -> Result<StateVar, StateError> { + ) -> Result<StateVar, StateModelError> { let prev_val = self.get_state_variable(prev, name)?; let next_val = self.get_state_variable(next, name)?; Ok(next_val - prev_val) @@ -286,7 +305,7 @@ impl StateModel { name: &String, distance: &Distance, from_unit: &DistanceUnit, - ) -> Result<(), StateError> { + ) -> Result<(), StateModelError> { let prev_distance = self.get_distance(state, name, from_unit)?; let next_distance = prev_distance + *distance; self.set_distance(state, name, &next_distance, from_unit) @@ -299,7 +318,7 @@ impl StateModel { name: &String, time: &Time, from_unit: &TimeUnit, - ) -> Result<(), StateError> { + ) -> Result<(), StateModelError> { let prev_time = self.get_time(state, name, from_unit)?; let next_time = prev_time + *time; self.set_time(state, name, &next_time, from_unit) @@ -312,7 +331,7 @@ impl StateModel { name: &String, energy: &Energy, from_unit: &EnergyUnit, - ) -> Result<(), StateError> { + ) -> Result<(), StateModelError> { let prev_energy = self.get_energy(state, name, from_unit)?; let next_energy = prev_energy + *energy; self.set_energy(state, name, &next_energy, from_unit) @@ -324,7 +343,7 @@ impl StateModel { name: &String, distance: &Distance, from_unit: &DistanceUnit, - ) -> Result<(), StateError> { + ) -> Result<(), StateModelError> { let feature = self.get_feature(name)?; let to_unit = feature.get_distance_unit()?; let value = from_unit.convert(distance, &to_unit); @@ -337,7 +356,7 @@ impl StateModel { name: &String, time: &Time, from_unit: &TimeUnit, - ) -> Result<(), StateError> { + ) -> Result<(), StateModelError> { let feature = self.get_feature(name)?; let to_unit = feature.get_time_unit()?; let value = from_unit.convert(time, &to_unit); @@ -350,7 +369,7 @@ impl StateModel { name: &String, energy: &Energy, from_unit: &EnergyUnit, - ) -> Result<(), StateError> { + ) -> Result<(), StateModelError> { let feature = self.get_feature(name)?; let to_unit = feature.get_energy_unit()?; let value = from_unit.convert(energy, &to_unit); @@ -362,7 +381,7 @@ impl StateModel { state: &mut [StateVar], name: &String, value: &f64, - ) -> Result<(), StateError> { + ) -> Result<(), StateModelError> { let feature = self.get_feature(name)?; let format = feature.get_custom_feature_format()?; let encoded_value = format.encode_f64(value)?; @@ -374,7 +393,7 @@ impl StateModel { state: &mut [StateVar], name: &String, value: &i64, - ) -> Result<(), StateError> { + ) -> Result<(), StateModelError> { let feature = self.get_feature(name)?; let format = feature.get_custom_feature_format()?; let encoded_value = format.encode_i64(value)?; @@ -386,7 +405,7 @@ impl StateModel { state: &mut [StateVar], name: &String, value: &u64, - ) -> Result<(), StateError> { + ) -> Result<(), StateModelError> { let feature = self.get_feature(name)?; let format = feature.get_custom_feature_format()?; let encoded_value = format.encode_u64(value)?; @@ -398,7 +417,7 @@ impl StateModel { state: &mut [StateVar], name: &String, value: &bool, - ) -> Result<(), StateError> { + ) -> Result<(), StateModelError> { let feature = self.get_feature(name)?; let format = feature.get_custom_feature_format()?; let encoded_value = format.encode_bool(value)?; @@ -443,9 +462,9 @@ impl StateModel { self.0.iter().map(|(k, _)| k.clone()).join(",") } - fn get_feature(&self, feature_name: &String) -> Result<&StateFeature, StateError> { + fn get_feature(&self, feature_name: &String) -> Result<&StateFeature, StateModelError> { self.0.get(feature_name).ok_or_else(|| { - StateError::UnknownStateVariableName(feature_name.clone(), self.get_names()) + StateModelError::UnknownStateVariableName(feature_name.clone(), self.get_names()) }) } @@ -454,13 +473,12 @@ impl StateModel { &self, state: &[StateVar], name: &String, - ) -> Result<StateVar, StateError> { - let idx = self - .0 - .get_index(name) - .ok_or_else(|| StateError::UnknownStateVariableName(name.clone(), self.get_names()))?; + ) -> Result<StateVar, StateModelError> { + let idx = self.0.get_index(name).ok_or_else(|| { + StateModelError::UnknownStateVariableName(name.clone(), self.get_names()) + })?; let value = state.get(idx).ok_or_else(|| { - StateError::RuntimeError(format!( + StateModelError::RuntimeError(format!( "state index {} for {} is out of range for state vector with {} entries", idx, name, @@ -476,14 +494,16 @@ impl StateModel { name: &String, value: &StateVar, op: UpdateOperation, - ) -> Result<(), StateError> { - let index = self - .0 - .get_index(name) - .ok_or_else(|| StateError::UnknownStateVariableName(name.clone(), self.get_names()))?; + ) -> Result<(), StateModelError> { + let index = self.0.get_index(name).ok_or_else(|| { + StateModelError::UnknownStateVariableName(name.clone(), self.get_names()) + })?; let prev = state .get(index) - .ok_or(StateError::InvalidStateVariableIndex(index, state.len()))?; + .ok_or(StateModelError::InvalidStateVariableIndex( + index, + state.len(), + ))?; let updated = op.perform_operation(prev, value); state[index] = updated; Ok(()) @@ -491,7 +511,7 @@ impl StateModel { } impl<'a> TryFrom<&'a serde_json::Value> for StateModel { - type Error = StateError; + type Error = StateModelError; /// builds a new state model from a JSON array of deserialized StateFeatures. /// the size of the JSON object matches the size of the feature vector. downstream @@ -526,11 +546,11 @@ impl<'a> TryFrom<&'a serde_json::Value> for StateModel { /// } /// } /// ``` - fn try_from(json: &'a serde_json::Value) -> Result<StateModel, StateError> { + fn try_from(json: &'a serde_json::Value) -> Result<StateModel, StateModelError> { let tuples = json .as_object() .ok_or_else(|| { - StateError::BuildError(String::from( + StateModelError::BuildError(String::from( "expected state model configuration to be a JSON object {}", )) })? @@ -538,7 +558,7 @@ impl<'a> TryFrom<&'a serde_json::Value> for StateModel { .map(|(feature_name, feature_json)| { let feature = serde_json::from_value::<StateFeature>(feature_json.clone()) .map_err(|e| { - StateError::BuildError(format!( + StateModelError::BuildError(format!( "unable to parse state feature row with name '{}' contents '{}' due to: {}", feature_name.clone(), feature_json.clone(), diff --git a/rust/routee-compass-core/src/model/state/state_error.rs b/rust/routee-compass-core/src/model/state/state_model_error.rs similarity index 97% rename from rust/routee-compass-core/src/model/state/state_error.rs rename to rust/routee-compass-core/src/model/state/state_model_error.rs index 8f39c607..22b5b606 100644 --- a/rust/routee-compass-core/src/model/state/state_error.rs +++ b/rust/routee-compass-core/src/model/state/state_model_error.rs @@ -1,7 +1,7 @@ use crate::model::traversal::state::state_variable::StateVar; #[derive(thiserror::Error, Debug)] -pub enum StateError { +pub enum StateModelError { #[error("attempting to encode {0} as state variable when value is a {1}")] EncodeError(String, String), #[error("attempting to decode {0} as a {1} when codec expects a {2}")] diff --git a/rust/routee-compass-core/src/model/state/update_operation.rs b/rust/routee-compass-core/src/model/state/update_operation.rs index 07f7ddae..bcea3751 100644 --- a/rust/routee-compass-core/src/model/state/update_operation.rs +++ b/rust/routee-compass-core/src/model/state/update_operation.rs @@ -1,8 +1,8 @@ use crate::model::traversal::state::state_variable::StateVar; - -/// describes an arbitrary state update operation. // pub(crate) type GenericStateUpdateOp = Box<dyn Fn(&StateVar, &StateVar) -> StateVar>; +/// describes an arbitrary state update operation. +/// /// represents the type of arithmetic operation used to update a state variable. /// the specific index of a state variable is hidden via the StateModel, which /// makes life harder, but protects against all sorts of indexing errors. diff --git a/rust/routee-compass-core/src/model/traversal/default/distance_traversal_model.rs b/rust/routee-compass-core/src/model/traversal/default/distance_traversal_model.rs index e347d1f6..d75e64f6 100644 --- a/rust/routee-compass-core/src/model/traversal/default/distance_traversal_model.rs +++ b/rust/routee-compass-core/src/model/traversal/default/distance_traversal_model.rs @@ -1,12 +1,12 @@ +use crate::model::network::{Edge, Vertex}; use crate::model::state::state_feature::StateFeature; use crate::model::state::state_model::StateModel; use crate::model::traversal::traversal_model::TraversalModel; +use crate::model::traversal::{ + state::state_variable::StateVar, traversal_model_error::TraversalModelError, +}; use crate::model::unit::DistanceUnit; use crate::model::unit::BASE_DISTANCE_UNIT; -use crate::model::{ - property::{edge::Edge, vertex::Vertex}, - traversal::{state::state_variable::StateVar, traversal_model_error::TraversalModelError}, -}; use crate::util::geo::haversine; /// A simple traversal model that uses the edge distance as the cost of traversal. @@ -49,7 +49,12 @@ impl TraversalModel for DistanceTraversalModel { let (src, dst) = od; let distance = haversine::coord_distance(&src.coordinate, &dst.coordinate, self.distance_unit) - .map_err(TraversalModelError::NumericError)?; + .map_err(|e| { + TraversalModelError::TraversalModelFailure(format!( + "could not compute haversine distance between {} and {}: {}", + src, dst, e + )) + })?; state_model.add_distance( state, &Self::DISTANCE.into(), diff --git a/rust/routee-compass-core/src/model/traversal/default/speed_traversal_engine.rs b/rust/routee-compass-core/src/model/traversal/default/speed_traversal_engine.rs index 93ef4363..8539c0db 100644 --- a/rust/routee-compass-core/src/model/traversal/default/speed_traversal_engine.rs +++ b/rust/routee-compass-core/src/model/traversal/default/speed_traversal_engine.rs @@ -25,10 +25,11 @@ impl SpeedTraversalEngine { let speed_table: Box<[Speed]> = read_utils::read_raw_file(speed_table_path, read_decoders::default, None).map_err( |e| { - TraversalModelError::FileReadError( - speed_table_path.as_ref().to_path_buf(), - e.to_string(), - ) + TraversalModelError::BuildError(format!( + "cannot read {} due to {}", + speed_table_path.as_ref().to_str().unwrap_or_default(), + e, + )) }, )?; let max_speed = get_max_speed(&speed_table)?; diff --git a/rust/routee-compass-core/src/model/traversal/default/speed_traversal_model.rs b/rust/routee-compass-core/src/model/traversal/default/speed_traversal_model.rs index 28d7ee20..706d3a9b 100644 --- a/rust/routee-compass-core/src/model/traversal/default/speed_traversal_model.rs +++ b/rust/routee-compass-core/src/model/traversal/default/speed_traversal_model.rs @@ -1,11 +1,11 @@ use super::speed_traversal_engine::SpeedTraversalEngine; -use crate::model::road_network::edge_id::EdgeId; +use crate::model::network::edge_id::EdgeId; +use crate::model::network::{Edge, Vertex}; use crate::model::state::state_feature::StateFeature; use crate::model::state::state_model::StateModel; use crate::model::traversal::traversal_model::TraversalModel; use crate::model::unit::{Distance, Time, BASE_DISTANCE_UNIT}; use crate::model::{ - property::{edge::Edge, vertex::Vertex}, traversal::{state::state_variable::StateVar, traversal_model_error::TraversalModelError}, unit::Speed, }; @@ -66,7 +66,12 @@ impl TraversalModel for SpeedTraversalModel { let (src, dst) = od; let distance = haversine::coord_distance(&src.coordinate, &dst.coordinate, self.engine.distance_unit) - .map_err(TraversalModelError::NumericError)?; + .map_err(|e| { + TraversalModelError::TraversalModelFailure(format!( + "could not compute haversine distance between {} and {}: {}", + src, dst, e + )) + })?; if distance == Distance::ZERO { return Ok(()); @@ -118,11 +123,10 @@ impl TraversalModel for SpeedTraversalModel { /// look up a speed from the speed table pub fn get_speed(speed_table: &[Speed], edge_id: EdgeId) -> Result<Speed, TraversalModelError> { let speed: &Speed = speed_table.get(edge_id.as_usize()).ok_or_else(|| { - TraversalModelError::MissingIdInTabularCostFunction( - format!("{}", edge_id), - String::from("EdgeId"), - String::from("speed table"), - ) + TraversalModelError::TraversalModelFailure(format!( + "could not find expected index {} in speed table", + edge_id + )) })?; Ok(*speed) } @@ -130,11 +134,8 @@ pub fn get_speed(speed_table: &[Speed], edge_id: EdgeId) -> Result<Speed, Traver #[cfg(test)] mod tests { use super::*; + use crate::model::network::{Edge, EdgeId, Vertex, VertexId}; use crate::model::unit::{Distance, DistanceUnit, SpeedUnit, TimeUnit}; - use crate::model::{ - property::{edge::Edge, vertex::Vertex}, - road_network::{edge_id::EdgeId, vertex_id::VertexId}, - }; use crate::util::geo::coord::InternalCoord; use geo::coord; use std::path::PathBuf; diff --git a/rust/routee-compass-core/src/model/traversal/traversal_model.rs b/rust/routee-compass-core/src/model/traversal/traversal_model.rs index 53547170..9fd889e4 100644 --- a/rust/routee-compass-core/src/model/traversal/traversal_model.rs +++ b/rust/routee-compass-core/src/model/traversal/traversal_model.rs @@ -1,6 +1,6 @@ use super::state::state_variable::StateVar; use super::traversal_model_error::TraversalModelError; -use crate::model::property::{edge::Edge, vertex::Vertex}; +use crate::model::network::{Edge, Vertex}; use crate::model::state::state_feature::StateFeature; use crate::model::state::state_model::StateModel; diff --git a/rust/routee-compass-core/src/model/traversal/traversal_model_error.rs b/rust/routee-compass-core/src/model/traversal/traversal_model_error.rs index ecba24af..8b5d393f 100644 --- a/rust/routee-compass-core/src/model/traversal/traversal_model_error.rs +++ b/rust/routee-compass-core/src/model/traversal/traversal_model_error.rs @@ -1,32 +1,34 @@ -use super::state::traversal_state::TraversalState; -use crate::model::road_network::graph_error::GraphError; -use crate::model::state::state_error::StateError; +use crate::model::network::network_error::NetworkError; +use crate::model::state::state_model_error::StateModelError; use crate::model::unit::UnitError; use crate::util::cache_policy::cache_error::CacheError; -use std::path::PathBuf; #[derive(thiserror::Error, Debug)] pub enum TraversalModelError { - #[error("failure building traversal model from file {0}: {1}")] - FileReadError(PathBuf, String), #[error("failure building traversal model: {0}")] BuildError(String), - #[error("numeric error during calculation: {0}")] - NumericError(String), - #[error("index {0} for {1} not found on search state {2:?}")] - StateVectorIndexOutOfBounds(usize, String, TraversalState), - #[error("id {0} for id type {1} not found in tabular edge cost function {2}")] - MissingIdInTabularCostFunction(String, String, String), + #[error("{0}")] + TraversalModelFailure(String), #[error("internal error: {0}")] InternalError(String), - #[error(transparent)] - TraversalUnitsError(#[from] UnitError), - #[error(transparent)] - CacheError(#[from] CacheError), - #[error(transparent)] - GraphError(#[from] GraphError), - #[error(transparent)] - StateError(#[from] StateError), - #[error("prediction model failed with error {0}")] - PredictionModel(String), + #[error("failure executing traversal model due to numeric units: {source}")] + UnitsFailure { + #[from] + source: UnitError, + }, + #[error("failure executing traversal model due to cache: {source}")] + CacheFailure { + #[from] + source: CacheError, + }, + #[error("failure executing traversal model due to network: {source}")] + NetworkFailure { + #[from] + source: NetworkError, + }, + #[error("failure executing traversal model due to search state: {source}")] + StateError { + #[from] + source: StateModelError, + }, } diff --git a/rust/routee-compass-core/src/util/fs/read_utils.rs b/rust/routee-compass-core/src/util/fs/read_utils.rs index 9a039eeb..39f0b386 100644 --- a/rust/routee-compass-core/src/util/fs/read_utils.rs +++ b/rust/routee-compass-core/src/util/fs/read_utils.rs @@ -34,13 +34,12 @@ where .trim(csv::Trim::Fields) .from_reader(r) .into_deserialize::<T>() - .map(move |r| { - if let Ok(t) = &r { + .inspect(move |r| { + if let Ok(t) = r { if let Some(cb) = &mut row_callback { cb(t); } } - r }); Ok(Box::new(reader)) diff --git a/rust/routee-compass-core/src/util/io_utils.rs b/rust/routee-compass-core/src/util/io_utils.rs index 347a2fa7..f42d4c51 100644 --- a/rust/routee-compass-core/src/util/io_utils.rs +++ b/rust/routee-compass-core/src/util/io_utils.rs @@ -26,7 +26,7 @@ pub fn with_progress_bar<R, E>( let result = closure(cb); - println!(); // create a newline once the progress bar is complete + eprintln!(); // create a newline once the progress bar is complete result } diff --git a/rust/routee-compass-core/src/util/multiset.rs b/rust/routee-compass-core/src/util/multiset.rs index fb2ecc6c..ec6a3d74 100644 --- a/rust/routee-compass-core/src/util/multiset.rs +++ b/rust/routee-compass-core/src/util/multiset.rs @@ -19,7 +19,7 @@ where } } -impl<'a, T> Iterator for MultiSet<'a, T> +impl<T> Iterator for MultiSet<'_, T> where T: Clone + Copy, { diff --git a/rust/routee-compass-macros/src/lib.rs b/rust/routee-compass-macros/src/lib.rs index defbc6f4..bd3b0b2b 100644 --- a/rust/routee-compass-macros/src/lib.rs +++ b/rust/routee-compass-macros/src/lib.rs @@ -50,21 +50,11 @@ pub fn pybindings(_attr: TokenStream, item: TokenStream) -> TokenStream { )) }) } - fn graph_get_out_edge_ids(&self, vertex_id: usize) -> PyResult<Vec<usize>> { - CompassAppBindings::graph_get_out_edge_ids(self, vertex_id).map_err(|e| { - PyException::new_err(format!( - "error retrieving out edge ids for vertex_id {}: {}", - vertex_id, e - )) - }) + fn graph_get_out_edge_ids(&self, vertex_id: usize) -> Vec<usize> { + CompassAppBindings::graph_get_out_edge_ids(self, vertex_id) } - fn graph_get_in_edge_ids(&self, vertex_id: usize) -> PyResult<Vec<usize>> { - CompassAppBindings::graph_get_in_edge_ids(self, vertex_id).map_err(|e| { - PyException::new_err(format!( - "error retrieving in edge ids for vertex_id {}: {}", - vertex_id, e - )) - }) + fn graph_get_in_edge_ids(&self, vertex_id: usize) -> Vec<usize> { + CompassAppBindings::graph_get_in_edge_ids(self, vertex_id) } #[staticmethod] pub fn _from_config_toml_string( diff --git a/rust/routee-compass-powertrain/src/routee/energy_model_ops.rs b/rust/routee-compass-powertrain/src/routee/energy_model_ops.rs index eb3238ce..236b3dc6 100644 --- a/rust/routee-compass-powertrain/src/routee/energy_model_ops.rs +++ b/rust/routee-compass-powertrain/src/routee/energy_model_ops.rs @@ -1,5 +1,5 @@ use routee_compass_core::model::{ - access::default::turn_delays::edge_heading::EdgeHeading, road_network::edge_id::EdgeId, + access::default::turn_delays::edge_heading::EdgeHeading, network::edge_id::EdgeId, traversal::traversal_model_error::TraversalModelError, unit::Grade, }; @@ -14,11 +14,10 @@ pub fn get_grade( None => Ok(Grade::ZERO), Some(gt) => { let grade: &Grade = gt.get(edge_id.as_usize()).ok_or_else(|| { - TraversalModelError::MissingIdInTabularCostFunction( - format!("{}", edge_id), - String::from("EdgeId"), - String::from("grade table"), - ) + TraversalModelError::TraversalModelFailure(format!( + "missing index {} from grade table", + edge_id + )) })?; Ok(*grade) } @@ -31,11 +30,10 @@ pub fn get_headings( edge_id: EdgeId, ) -> Result<EdgeHeading, TraversalModelError> { let heading: &EdgeHeading = headings_table.get(edge_id.as_usize()).ok_or_else(|| { - TraversalModelError::MissingIdInTabularCostFunction( - format!("{}", edge_id), - String::from("EdgeId"), - String::from("headings table"), - ) + TraversalModelError::TraversalModelFailure(format!( + "missing index {} from headings table", + edge_id + )) })?; Ok(*heading) } diff --git a/rust/routee-compass-powertrain/src/routee/energy_model_service.rs b/rust/routee-compass-powertrain/src/routee/energy_model_service.rs index 5153c113..faeac742 100644 --- a/rust/routee-compass-powertrain/src/routee/energy_model_service.rs +++ b/rust/routee-compass-powertrain/src/routee/energy_model_service.rs @@ -38,7 +38,11 @@ impl EnergyModelService { let grade_table: Arc<Option<Box<[Grade]>>> = match grade_table_path_option { Some(gtp) => Arc::new(Some( read_utils::read_raw_file(gtp, read_decoders::default, None).map_err(|e| { - TraversalModelError::FileReadError(gtp.as_ref().to_path_buf(), e.to_string()) + TraversalModelError::BuildError(format!( + "failure reading grade table {} due to {}", + gtp.as_ref().to_str().unwrap_or_default(), + e + )) })?, )), None => Arc::new(None), diff --git a/rust/routee-compass-powertrain/src/routee/energy_traversal_model.rs b/rust/routee-compass-powertrain/src/routee/energy_traversal_model.rs index 61743a58..7c2d400e 100644 --- a/rust/routee-compass-powertrain/src/routee/energy_traversal_model.rs +++ b/rust/routee-compass-powertrain/src/routee/energy_traversal_model.rs @@ -1,8 +1,7 @@ use super::energy_model_ops::get_grade; use super::energy_model_service::EnergyModelService; use super::vehicle::vehicle_type::VehicleType; -use routee_compass_core::model::property::edge::Edge; -use routee_compass_core::model::property::vertex::Vertex; +use routee_compass_core::model::network::{Edge, Vertex}; use routee_compass_core::model::state::state_feature::StateFeature; use routee_compass_core::model::state::state_model::StateModel; use routee_compass_core::model::traversal::state::state_variable::StateVar; @@ -92,7 +91,12 @@ impl TraversalModel for EnergyTraversalModel { &dst.coordinate, self.energy_model_service.distance_unit, ) - .map_err(TraversalModelError::NumericError)?; + .map_err(|e| { + TraversalModelError::TraversalModelFailure(format!( + "could not compute haversine distance between {} and {}: {}", + src, dst, e + )) + })?; if distance == Distance::ZERO { return Ok(()); @@ -162,8 +166,7 @@ mod tests { use geo::coord; use routee_compass_core::{ model::{ - property::{edge::Edge, vertex::Vertex}, - road_network::{edge_id::EdgeId, vertex_id::VertexId}, + network::{Edge, EdgeId, Vertex, VertexId}, traversal::default::{ speed_traversal_engine::SpeedTraversalEngine, speed_traversal_service::SpeedLookupService, diff --git a/rust/routee-compass-powertrain/src/routee/prediction/interpolation/interpolation_speed_grade_model.rs b/rust/routee-compass-powertrain/src/routee/prediction/interpolation/interpolation_speed_grade_model.rs index 279f17aa..52385cfb 100644 --- a/rust/routee-compass-powertrain/src/routee/prediction/interpolation/interpolation_speed_grade_model.rs +++ b/rust/routee-compass-powertrain/src/routee/prediction/interpolation/interpolation_speed_grade_model.rs @@ -36,28 +36,28 @@ impl PredictionModel for InterpolationSpeedGradeModel { let (min_speed, max_speed, min_grade, max_grade) = match &self.interpolator { interp::Interpolator::Interp2D(interp) => ( *interp.x.first().ok_or_else(|| { - TraversalModelError::PredictionModel( - "Could not get first x-value; are x-values empty?".to_string(), + TraversalModelError::TraversalModelFailure( + "Could not get first x-value from powertrain interpolation result; are x-values empty?".to_string(), ) })?, *interp.x.last().ok_or_else(|| { - TraversalModelError::PredictionModel( - "Could not get last x-value; are x-values empty?".to_string(), + TraversalModelError::TraversalModelFailure( + "Could not get last x-value from powertrain interpolation result; are x-values empty?".to_string(), ) })?, *interp.y.first().ok_or_else(|| { - TraversalModelError::PredictionModel( - "Could not get first y-value; are y-values empty?".to_string(), + TraversalModelError::TraversalModelFailure( + "Could not get first y-value from powertrain interpolation result; are y-values empty?".to_string(), ) })?, *interp.y.last().ok_or_else(|| { - TraversalModelError::PredictionModel( - "Could not get last y-value; are y-values empty?".to_string(), + TraversalModelError::TraversalModelFailure( + "Could not get last y-value from powertrain interpolation result; are y-values empty?".to_string(), ) })?, ), _ => { - return Err(TraversalModelError::PredictionModel( + return Err(TraversalModelError::TraversalModelFailure( "Only 2-D interpolators are currently supported".to_string(), )) } @@ -70,7 +70,10 @@ impl PredictionModel for InterpolationSpeedGradeModel { .interpolator .interpolate(&[speed_value, grade_value], &interp::Strategy::Linear) .map_err(|e| { - TraversalModelError::PredictionModel(format!("Failed to interpolate: {}", e)) + TraversalModelError::TraversalModelFailure(format!( + "Failed to interpolate speed/grade model output during prediction: {}", + e + )) })?; let energy_rate = EnergyRate::new(y); @@ -119,13 +122,11 @@ impl InterpolationSpeedGradeModel { for speed_value in speed_values.clone().into_iter() { let mut row: Vec<f64> = Vec::new(); for grade_value in grade_values.clone().into_iter() { - let (energy, _energy_unit) = model - .predict( - (Speed::new(speed_value), speed_unit), - (Grade::new(grade_value), grade_unit), - (distance, distance_unit), - ) - .map_err(|e| TraversalModelError::PredictionModel(e.to_string()))?; + let (energy, _energy_unit) = model.predict( + (Speed::new(speed_value), speed_unit), + (Grade::new(grade_value), grade_unit), + (distance, distance_unit), + )?; row.push(energy.as_f64()); } values.push(row); @@ -133,8 +134,8 @@ impl InterpolationSpeedGradeModel { let interpolator = interp::Interpolator::Interp2D( interp::Interp2D::new(speed_values, grade_values, values).map_err(|e| { - TraversalModelError::PredictionModel(format!( - "Failed to validate interpolation model: {}", + TraversalModelError::TraversalModelFailure(format!( + "Failed to build interpolation speed/grade model: {}", e )) })?, diff --git a/rust/routee-compass-powertrain/src/routee/prediction/prediction_model_ops.rs b/rust/routee-compass-powertrain/src/routee/prediction/prediction_model_ops.rs index 0aad686c..fe6467a7 100644 --- a/rust/routee-compass-powertrain/src/routee/prediction/prediction_model_ops.rs +++ b/rust/routee-compass-powertrain/src/routee/prediction/prediction_model_ops.rs @@ -113,7 +113,7 @@ pub fn find_min_energy_rate( (speed, SpeedUnit::MilesPerHour), (grade, GradeUnit::Percent), ) - .map_err(|e| TraversalModelError::PredictionModel(e.to_string()))?; + .map_err(|e| TraversalModelError::BuildError(format!("failure while executing grid search for minimum energy rate in prediction model: {}", e)))?; if energy_rate < minimum_energy_rate { minimum_energy_rate = energy_rate; } diff --git a/rust/routee-compass-powertrain/src/routee/prediction/smartcore/smartcore_speed_grade_model.rs b/rust/routee-compass-powertrain/src/routee/prediction/smartcore/smartcore_speed_grade_model.rs index 5d527f30..13296c69 100644 --- a/rust/routee-compass-powertrain/src/routee/prediction/smartcore/smartcore_speed_grade_model.rs +++ b/rust/routee-compass-powertrain/src/routee/prediction/smartcore/smartcore_speed_grade_model.rs @@ -27,10 +27,12 @@ impl PredictionModel for SmartcoreSpeedGradeModel { let speed_value = speed_unit.convert(&speed, &self.speed_unit).as_f64(); let grade_value = grade_unit.convert(&grade, &self.grade_unit).as_f64(); let x = DenseMatrix::from_2d_vec(&vec![vec![speed_value, grade_value]]); - let y = self - .rf - .predict(&x) - .map_err(|e| TraversalModelError::PredictionModel(e.to_string()))?; + let y = self.rf.predict(&x).map_err(|e| { + TraversalModelError::TraversalModelFailure(format!( + "failure running underlying Smartcore random forest energy prediction: {}", + e + )) + })?; let energy_rate = EnergyRate::new(y[0]); Ok((energy_rate, self.energy_rate_unit)) @@ -46,17 +48,19 @@ impl SmartcoreSpeedGradeModel { ) -> Result<Self, TraversalModelError> { // Load random forest binary file let rf_binary = std::fs::read(routee_model_path).map_err(|e| { - TraversalModelError::FileReadError( - routee_model_path.as_ref().to_path_buf(), - e.to_string(), - ) + TraversalModelError::BuildError(format!( + "failure reading smartcore binary text file {} due to {}", + routee_model_path.as_ref().to_str().unwrap_or_default(), + e + )) })?; let rf: RandomForestRegressor<f64, f64, DenseMatrix<f64>, Vec<f64>> = bincode::deserialize(&rf_binary).map_err(|e| { - TraversalModelError::FileReadError( - routee_model_path.as_ref().to_path_buf(), - e.to_string(), - ) + TraversalModelError::BuildError(format!( + "failure deserializing smartcore model {} due to {}", + routee_model_path.as_ref().to_str().unwrap_or_default(), + e + )) })?; Ok(SmartcoreSpeedGradeModel { rf, diff --git a/rust/routee-compass-powertrain/src/routee/vehicle/vehicle_ops.rs b/rust/routee-compass-powertrain/src/routee/vehicle/vehicle_ops.rs index 5c005683..fef1899a 100644 --- a/rust/routee-compass-powertrain/src/routee/vehicle/vehicle_ops.rs +++ b/rust/routee-compass-powertrain/src/routee/vehicle/vehicle_ops.rs @@ -1,5 +1,5 @@ use routee_compass_core::model::{ - state::{state_error::StateError, state_model::StateModel}, + state::{state_model::StateModel, state_model_error::StateModelError}, traversal::state::state_variable::StateVar, unit::{as_f64::AsF64, Energy}, }; @@ -22,7 +22,7 @@ pub fn update_soc_percent( delta: &Energy, max: &Energy, state_model: &StateModel, -) -> Result<(), StateError> { +) -> Result<(), StateModelError> { let start_soc = state_model.get_custom_f64(state, &feature_name.into())?; let start_battery = max.as_f64() * (start_soc / 100.0); let current_soc = soc_from_battery_and_delta(&Energy::new(start_battery), delta, max); diff --git a/rust/routee-compass/src/app/bindings.rs b/rust/routee-compass/src/app/bindings.rs index 4a87f90b..88d6a711 100644 --- a/rust/routee-compass/src/app/bindings.rs +++ b/rust/routee-compass/src/app/bindings.rs @@ -1,9 +1,10 @@ use std::str::FromStr; +use itertools::Itertools; use routee_compass_core::{ algorithm::search::direction::Direction, model::{ - road_network::{edge_id::EdgeId, vertex_id::VertexId}, + network::{edge_id::EdgeId, vertex_id::VertexId}, unit::{as_f64::AsF64, DistanceUnit}, }, }; @@ -101,7 +102,7 @@ pub trait CompassAppBindings { let edge_id_internal = EdgeId(edge_id); self.app() .search_app - .get_edge_origin(edge_id_internal) + .get_edge_origin(&edge_id_internal) .map(|o| o.0) } @@ -116,7 +117,7 @@ pub trait CompassAppBindings { let edge_id_internal = EdgeId(edge_id); self.app() .search_app - .get_edge_destination(edge_id_internal) + .get_edge_destination(&edge_id_internal) .map(|o| o.0) } @@ -150,7 +151,7 @@ pub trait CompassAppBindings { let edge_id_internal = EdgeId(edge_id); self.app() .search_app - .get_edge_distance(edge_id_internal, du_internal) + .get_edge_distance(&edge_id_internal, du_internal) .map(|o| o.as_f64()) } @@ -161,12 +162,14 @@ pub trait CompassAppBindings { /// /// # Returns /// * the ids of the edges incident to the vertex in the forward direction - fn graph_get_out_edge_ids(&self, vertex_id: usize) -> Result<Vec<usize>, CompassAppError> { + fn graph_get_out_edge_ids(&self, vertex_id: usize) -> Vec<usize> { let vertex_id_internal = VertexId(vertex_id); self.app() .search_app - .get_incident_edge_ids(vertex_id_internal, Direction::Forward) - .map(|es| es.iter().map(|e| e.0).collect()) + .get_incident_edge_ids(&vertex_id_internal, &Direction::Forward) + .into_iter() + .map(|e| e.0) + .collect_vec() } /// Get the ids of the edges incident to a vertex in the reverse direction @@ -176,12 +179,15 @@ pub trait CompassAppBindings { /// /// # Returns /// * the ids of the edges incident to the vertex in the reverse direction - fn graph_get_in_edge_ids(&self, vertex_id: usize) -> Result<Vec<usize>, CompassAppError> { + fn graph_get_in_edge_ids(&self, vertex_id: usize) -> Vec<usize> { let vertex_id_internal = VertexId(vertex_id); self.app() .search_app - .get_incident_edge_ids(vertex_id_internal, Direction::Reverse) - .map(|es| es.iter().map(|e| e.0).collect()) + .get_incident_edge_ids(&vertex_id_internal, &Direction::Reverse) + // .map(|es| es.iter().map(|e| e.0).collect()) + .into_iter() + .map(|e| e.0) + .collect_vec() } /// Runs a set of queries and returns the results diff --git a/rust/routee-compass/src/app/cli/cli_args.rs b/rust/routee-compass/src/app/cli/cli_args.rs index b1fa5214..5f920e1c 100644 --- a/rust/routee-compass/src/app/cli/cli_args.rs +++ b/rust/routee-compass/src/app/cli/cli_args.rs @@ -52,7 +52,7 @@ impl CliArgs { if c > 0 { Ok(Some(c as usize)) } else { - Err(CompassAppError::InvalidInput(format!( + Err(CompassAppError::CompassFailure(format!( "chunksize must be positive, found {}", c ))) diff --git a/rust/routee-compass/src/app/cli/run.rs b/rust/routee-compass/src/app/cli/run.rs index ff438d12..ba319fdb 100644 --- a/rust/routee-compass/src/app/cli/run.rs +++ b/rust/routee-compass/src/app/cli/run.rs @@ -43,7 +43,7 @@ pub fn command_line_runner( // read user file containing JSON query/queries let query_file = File::open(args.query_file.clone()).map_err(|_e| { - CompassAppError::NoInputFile(format!("Could not find query file {}", args.query_file)) + CompassAppError::BuildFailure(format!("Could not find query file {}", args.query_file)) })?; // execute queries on app @@ -70,8 +70,7 @@ fn run_json( run_config: Option<&Value>, ) -> Result<(), CompassAppError> { let reader = BufReader::new(query_file); - let user_json: serde_json::Value = - serde_json::from_reader(reader).map_err(CompassAppError::CodecError)?; + let user_json: serde_json::Value = serde_json::from_reader(reader)?; let user_queries = user_json.get_queries()?; let results = compass_app.run(user_queries, run_config)?; for result in results.iter() { @@ -106,7 +105,10 @@ fn run_newline_json( CompassConfigurationError::SerdeDeserializationError(e), )), }, - Err(e) => Either::Right(CompassAppError::IOError(e)), + Err(e) => Either::Right(CompassAppError::CompassFailure(format!( + "failed to parse query row due to: {}", + e + ))), }); // run Compass on this chunk of queries diff --git a/rust/routee-compass/src/app/compass/compass_app.rs b/rust/routee-compass/src/app/compass/compass_app.rs index d922706b..801c0e7d 100644 --- a/rust/routee-compass/src/app/compass/compass_app.rs +++ b/rust/routee-compass/src/app/compass/compass_app.rs @@ -351,7 +351,9 @@ impl CompassApp { .animation("fillup") .desc("input plugins") .build() - .map_err(CompassAppError::UXError)?; + .map_err(|e| { + CompassAppError::InternalError(format!("could not build progress bar: {}", e)) + })?; let input_pb_shared = Arc::new(Mutex::new(input_pb)); // input plugins need to be flattened, and queries that fail input processing need to be @@ -378,7 +380,7 @@ impl CompassApp { }) .unzip(); - println!(); + eprintln!(); // unpack input plugin results let (processed_inputs_nested, error_inputs_nested) = input_plugin_result; @@ -416,7 +418,9 @@ impl CompassApp { .animation("fillup") .desc("search") .build() - .map_err(CompassAppError::UXError)?; + .map_err(|e| { + CompassAppError::InternalError(format!("could not build progress bar: {}", e)) + })?; let search_pb_shared = Arc::new(Mutex::new(search_pb)); // run parallel searches as organized by the (optional) load balancing policy @@ -439,6 +443,7 @@ impl CompassApp { search_pb_shared, )?, }; + eprintln!(); let run_result = run_query_result.chain(error_inputs).collect(); Ok(run_result) @@ -572,7 +577,7 @@ pub fn apply_input_plugins( let mut plugin_state = serde_json::Value::Array(vec![query.clone()]); for plugin in plugins { let p = plugin.clone(); - let op: in_ops::ArrayOp = Rc::new(|q| p.process(q)); + let op: in_ops::InputArrayOp = Rc::new(|q| p.process(q)); in_ops::json_array_op(&mut plugin_state, op)? } let result = in_ops::json_array_flatten(&mut plugin_state)?; diff --git a/rust/routee-compass/src/app/compass/compass_app_error.rs b/rust/routee-compass/src/app/compass/compass_app_error.rs index ba1c5336..b28255d7 100644 --- a/rust/routee-compass/src/app/compass/compass_app_error.rs +++ b/rust/routee-compass/src/app/compass/compass_app_error.rs @@ -1,49 +1,88 @@ -use super::{ - compass_input_field::CompassInputField, - config::compass_configuration_error::CompassConfigurationError, +use super::config::compass_configuration_error::CompassConfigurationError; +use crate::plugin::{ + input::InputPluginError, output::OutputPluginError, plugin_error::PluginError, }; -use crate::plugin::plugin_error::PluginError; use config::ConfigError; use routee_compass_core::{ algorithm::search::search_error::SearchError, model::{ - frontier::frontier_model_error::FrontierModelError, road_network::graph_error::GraphError, - state::state_error::StateError, traversal::traversal_model_error::TraversalModelError, + access::access_model_error::AccessModelError, cost::cost_model_error::CostModelError, + frontier::frontier_model_error::FrontierModelError, network::network_error::NetworkError, + state::state_model_error::StateModelError, + termination::termination_model_error::TerminationModelError, + traversal::traversal_model_error::TraversalModelError, }, }; #[derive(thiserror::Error, Debug)] pub enum CompassAppError { + // TOP-LEVEL FAILURES + // errors observed during build and run of compass that are not due to the search modules + #[error("failure building compass app: {0}")] + BuildFailure(String), + #[error("failure while running app: {0}")] + CompassFailure(String), + #[error("internal error: {0}")] + InternalError(String), + #[error("error accessing shared read-only dataset: {0}")] + ReadOnlyPoisonError(String), + + // TRANSPARENT MODULE FAILURES + // failures from these modules are detailed enough to get surfaced directly to the user #[error(transparent)] - SearchError(#[from] SearchError), + ConfigFailure(#[from] ConfigError), #[error(transparent)] - FrontierModelError(#[from] FrontierModelError), + CompassConfigurationError(#[from] CompassConfigurationError), #[error(transparent)] - TraversalModelError(#[from] TraversalModelError), + SearchFailure(#[from] SearchError), #[error(transparent)] PluginError(#[from] PluginError), #[error(transparent)] - IOError(#[from] std::io::Error), + InputPluginFailure(#[from] InputPluginError), #[error(transparent)] - CodecError(#[from] serde_json::Error), - #[error(transparent)] - ConfigError(#[from] ConfigError), - #[error(transparent)] - GraphError(#[from] GraphError), - #[error(transparent)] - StateError(#[from] StateError), - #[error("Input file {0} missing")] - NoInputFile(String), - #[error(transparent)] - CompassConfigurationError(#[from] CompassConfigurationError), - #[error("a ux component caused a failure: {0}")] - UXError(String), - #[error("internal error: {0}")] - InternalError(String), - #[error("app input JSON missing field: {0}")] - MissingInputField(CompassInputField), - #[error("error accessing shared read-only dataset: {0}")] - ReadOnlyPoisonError(String), - #[error("error decoding input:\n{0}")] - InvalidInput(String), + OutputPluginFailure(#[from] OutputPluginError), + + // CONTEXTUALIZED MODULE FAILURES + // failures from these modules are happening outside of the context of running the search, + // which is clarified for the user and may help direct where to look to solve the problem. + #[error("While interacting with the state model outside of the context of search, an error occurred. Source: {source}")] + StateFailure { + #[from] + source: StateModelError, + }, + #[error("While interacting with the network model outside of the context of search, an error occurred. Source: {source}")] + NetworkFailure { + #[from] + source: NetworkError, + }, + #[error("While interacting with the termination model outside of the context of search, an error occurred. Source: {source}")] + TerminationModelFailure { + #[from] + source: TerminationModelError, + }, + #[error("While interacting with the traversal model outside of the context of search, an error occurred. Source: {source}")] + TraversalModelFailure { + #[from] + source: TraversalModelError, + }, + #[error("While interacting with the access model outside of the context of search, an error occurred. Source: {source}")] + AccessModelFailure { + #[from] + source: AccessModelError, + }, + #[error("While interacting with the frontier model outside of the context of search, an error occurred. Source: {source}")] + FrontierModelFailure { + #[from] + source: FrontierModelError, + }, + #[error("While interacting with the cost model outside of the context of search, an error occurred. Source: {source}")] + CostFailure { + #[from] + source: CostModelError, + }, + #[error("failure due to JSON: {source}")] + JsonError { + #[from] + source: serde_json::Error, + }, } diff --git a/rust/routee-compass/src/app/compass/compass_app_ops.rs b/rust/routee-compass/src/app/compass/compass_app_ops.rs index 2bb2325f..df358a4d 100644 --- a/rust/routee-compass/src/app/compass/compass_app_ops.rs +++ b/rust/routee-compass/src/app/compass/compass_app_ops.rs @@ -36,8 +36,7 @@ pub fn read_config_from_file(config_path: &Path) -> Result<Config, CompassAppErr CompassInputField::ConfigInputFile.to_string(), conf_file_string, )? - .build() - .map_err(CompassAppError::ConfigError)?; + .build()?; Ok(config) } @@ -74,8 +73,7 @@ pub fn read_config_from_string( CompassInputField::ConfigInputFile.to_string(), original_file_path, )? - .build() - .map_err(CompassAppError::ConfigError)?; + .build()?; Ok(config) } diff --git a/rust/routee-compass/src/app/compass/compass_json_extensions.rs b/rust/routee-compass/src/app/compass/compass_json_extensions.rs index 3c0ad2a1..ee8211bb 100644 --- a/rust/routee-compass/src/app/compass/compass_json_extensions.rs +++ b/rust/routee-compass/src/app/compass/compass_json_extensions.rs @@ -19,11 +19,11 @@ impl CompassJsonExtensions for serde_json::Value { serde_json::Value::Array(vec) => Ok(vec.to_owned()), _ => { let msg = String::from("user JSON argument must be an object or an array"); - Err(CompassAppError::InvalidInput(msg)) + Err(CompassAppError::CompassFailure(msg)) } }, }, - _ => Err(CompassAppError::InvalidInput(String::from( + _ => Err(CompassAppError::CompassFailure(String::from( "expected object, object with queries, or array input", ))), } diff --git a/rust/routee-compass/src/app/compass/config/compass_configuration_error.rs b/rust/routee-compass/src/app/compass/config/compass_configuration_error.rs index 729e3203..4b8b7065 100644 --- a/rust/routee-compass/src/app/compass/config/compass_configuration_error.rs +++ b/rust/routee-compass/src/app/compass/config/compass_configuration_error.rs @@ -3,7 +3,7 @@ use config::ConfigError; use routee_compass_core::{ model::{ access::access_model_error::AccessModelError, - frontier::frontier_model_error::FrontierModelError, road_network::graph_error::GraphError, + frontier::frontier_model_error::FrontierModelError, network::network_error::NetworkError, traversal::traversal_model_error::TraversalModelError, }, util::{cache_policy::cache_error::CacheError, conversion::conversion_error::ConversionError}, @@ -53,7 +53,7 @@ pub enum CompassConfigurationError { #[error("{0}")] InsertError(String), #[error(transparent)] - GraphError(#[from] GraphError), + GraphError(#[from] NetworkError), #[error(transparent)] ConfigError(#[from] ConfigError), #[error(transparent)] diff --git a/rust/routee-compass/src/app/compass/config/frontier_model/combined/combined_model.rs b/rust/routee-compass/src/app/compass/config/frontier_model/combined/combined_model.rs index 9fda3248..75f5154d 100644 --- a/rust/routee-compass/src/app/compass/config/frontier_model/combined/combined_model.rs +++ b/rust/routee-compass/src/app/compass/config/frontier_model/combined/combined_model.rs @@ -1,6 +1,6 @@ use routee_compass_core::model::{ frontier::{frontier_model::FrontierModel, frontier_model_error::FrontierModelError}, - property::edge::Edge, + network::Edge, state::state_model::StateModel, traversal::state::state_variable::StateVar, }; diff --git a/rust/routee-compass/src/app/compass/config/frontier_model/road_class/road_class_model.rs b/rust/routee-compass/src/app/compass/config/frontier_model/road_class/road_class_model.rs index dd2ab068..abcf7a89 100644 --- a/rust/routee-compass/src/app/compass/config/frontier_model/road_class/road_class_model.rs +++ b/rust/routee-compass/src/app/compass/config/frontier_model/road_class/road_class_model.rs @@ -1,7 +1,7 @@ use super::road_class_service::RoadClassFrontierService; use routee_compass_core::model::{ frontier::{frontier_model::FrontierModel, frontier_model_error::FrontierModelError}, - property::edge::Edge, + network::Edge, state::state_model::StateModel, traversal::state::state_variable::StateVar, }; @@ -26,7 +26,12 @@ impl FrontierModel for RoadClassFrontierModel { .service .road_class_lookup .get(edge.edge_id.0) - .ok_or_else(|| FrontierModelError::MissingIndex(format!("{}", edge.edge_id))) + .ok_or_else(|| { + FrontierModelError::FrontierModelError(format!( + "edge id {} missing from frontier model file", + edge.edge_id + )) + }) .map(|road_class| road_classes.contains(road_class)), } } diff --git a/rust/routee-compass/src/app/compass/config/frontier_model/road_class/road_class_parser.rs b/rust/routee-compass/src/app/compass/config/frontier_model/road_class/road_class_parser.rs index 06df970d..dce3fe82 100644 --- a/rust/routee-compass/src/app/compass/config/frontier_model/road_class/road_class_parser.rs +++ b/rust/routee-compass/src/app/compass/config/frontier_model/road_class/road_class_parser.rs @@ -25,7 +25,7 @@ impl RoadClassParser { if self.mapping.is_empty() { // if we don't have a road class mapping then we fail let value_string = value.to_string(); - Err(CompassAppError::InvalidInput( + Err(CompassAppError::CompassFailure( formatdoc! {r#" Could not parse incoming query valid road_classes of {value_string} as an array of integers and this FrontierModel does not specify a mapping of string to integer. @@ -42,7 +42,7 @@ impl RoadClassParser { .map(|s| { self.mapping.get(s).cloned().ok_or_else(|| { let valid_mapping: Vec<String> = self.mapping.keys().cloned().collect(); - CompassAppError::InvalidInput( + CompassAppError::CompassFailure( format!("Could not find road class mapping for incoming value {}. Here are the road classes I have: {:?}", s, valid_mapping) ) }) diff --git a/rust/routee-compass/src/app/compass/config/frontier_model/turn_restrictions/turn_restriction_model.rs b/rust/routee-compass/src/app/compass/config/frontier_model/turn_restrictions/turn_restriction_model.rs index c719fd32..fbd489da 100644 --- a/rust/routee-compass/src/app/compass/config/frontier_model/turn_restrictions/turn_restriction_model.rs +++ b/rust/routee-compass/src/app/compass/config/frontier_model/turn_restrictions/turn_restriction_model.rs @@ -1,6 +1,6 @@ use routee_compass_core::model::{ frontier::{frontier_model::FrontierModel, frontier_model_error::FrontierModelError}, - property::edge::Edge, + network::Edge, state::state_model::StateModel, traversal::state::state_variable::StateVar, }; diff --git a/rust/routee-compass/src/app/compass/config/frontier_model/turn_restrictions/turn_restriction_service.rs b/rust/routee-compass/src/app/compass/config/frontier_model/turn_restrictions/turn_restriction_service.rs index 1953b3ef..4a7250a7 100644 --- a/rust/routee-compass/src/app/compass/config/frontier_model/turn_restrictions/turn_restriction_service.rs +++ b/rust/routee-compass/src/app/compass/config/frontier_model/turn_restrictions/turn_restriction_service.rs @@ -4,7 +4,7 @@ use routee_compass_core::model::{ frontier_model::FrontierModel, frontier_model_error::FrontierModelError, frontier_model_service::FrontierModelService, }, - road_network::edge_id::EdgeId, + network::edge_id::EdgeId, state::state_model::StateModel, }; use serde::Deserialize; diff --git a/rust/routee-compass/src/app/compass/config/frontier_model/vehicle_restrictions/vehicle_restriction_builder.rs b/rust/routee-compass/src/app/compass/config/frontier_model/vehicle_restrictions/vehicle_restriction_builder.rs index 683e4e56..ac36407b 100644 --- a/rust/routee-compass/src/app/compass/config/frontier_model/vehicle_restrictions/vehicle_restriction_builder.rs +++ b/rust/routee-compass/src/app/compass/config/frontier_model/vehicle_restrictions/vehicle_restriction_builder.rs @@ -8,7 +8,7 @@ use routee_compass_core::{ frontier_model_builder::FrontierModelBuilder, frontier_model_error::FrontierModelError, frontier_model_service::FrontierModelService, }, - road_network::edge_id::EdgeId, + network::edge_id::EdgeId, }, util::fs::read_utils, }; diff --git a/rust/routee-compass/src/app/compass/config/frontier_model/vehicle_restrictions/vehicle_restriction_model.rs b/rust/routee-compass/src/app/compass/config/frontier_model/vehicle_restrictions/vehicle_restriction_model.rs index 9ad3295f..77a6a43f 100644 --- a/rust/routee-compass/src/app/compass/config/frontier_model/vehicle_restrictions/vehicle_restriction_model.rs +++ b/rust/routee-compass/src/app/compass/config/frontier_model/vehicle_restrictions/vehicle_restriction_model.rs @@ -4,7 +4,7 @@ use super::{ }; use routee_compass_core::model::{ frontier::{frontier_model::FrontierModel, frontier_model_error::FrontierModelError}, - property::edge::Edge, + network::Edge, state::state_model::StateModel, traversal::state::state_variable::StateVar, }; diff --git a/rust/routee-compass/src/app/compass/config/frontier_model/vehicle_restrictions/vehicle_restriction_row.rs b/rust/routee-compass/src/app/compass/config/frontier_model/vehicle_restrictions/vehicle_restriction_row.rs index 428eaa33..834a5725 100644 --- a/rust/routee-compass/src/app/compass/config/frontier_model/vehicle_restrictions/vehicle_restriction_row.rs +++ b/rust/routee-compass/src/app/compass/config/frontier_model/vehicle_restrictions/vehicle_restriction_row.rs @@ -1,5 +1,5 @@ use routee_compass_core::model::{ - frontier::frontier_model_error::FrontierModelError, road_network::edge_id::EdgeId, + frontier::frontier_model_error::FrontierModelError, network::edge_id::EdgeId, }; use serde::Deserialize; diff --git a/rust/routee-compass/src/app/compass/config/frontier_model/vehicle_restrictions/vehicle_restriction_service.rs b/rust/routee-compass/src/app/compass/config/frontier_model/vehicle_restrictions/vehicle_restriction_service.rs index de97ab2c..d9092dd5 100644 --- a/rust/routee-compass/src/app/compass/config/frontier_model/vehicle_restrictions/vehicle_restriction_service.rs +++ b/rust/routee-compass/src/app/compass/config/frontier_model/vehicle_restrictions/vehicle_restriction_service.rs @@ -7,7 +7,7 @@ use routee_compass_core::model::{ frontier_model::FrontierModel, frontier_model_error::FrontierModelError, frontier_model_service::FrontierModelService, }, - road_network::edge_id::EdgeId, + network::edge_id::EdgeId, state::state_model::StateModel, }; use std::{collections::HashMap, sync::Arc}; diff --git a/rust/routee-compass/src/app/compass/config/graph_builder.rs b/rust/routee-compass/src/app/compass/config/graph_builder.rs index dc4f15aa..f74860c5 100644 --- a/rust/routee-compass/src/app/compass/config/graph_builder.rs +++ b/rust/routee-compass/src/app/compass/config/graph_builder.rs @@ -1,4 +1,4 @@ -use routee_compass_core::model::road_network::graph::Graph; +use routee_compass_core::model::network::graph::Graph; use crate::app::compass::config::compass_configuration_field::CompassConfigurationField; diff --git a/rust/routee-compass/src/app/compass/response/response_output_format_json.rs b/rust/routee-compass/src/app/compass/response/response_output_format_json.rs index d3baa8af..a3eb3dd8 100644 --- a/rust/routee-compass/src/app/compass/response/response_output_format_json.rs +++ b/rust/routee-compass/src/app/compass/response/response_output_format_json.rs @@ -21,9 +21,9 @@ pub fn format_response( newline_delimited: bool, ) -> Result<String, CompassAppError> { if newline_delimited { - serde_json::to_string(response).map_err(CompassAppError::CodecError) + serde_json::to_string(response).map_err(CompassAppError::from) } else { - let row = serde_json::to_string_pretty(response).map_err(CompassAppError::CodecError)?; + let row = serde_json::to_string_pretty(response).map_err(CompassAppError::from)?; Ok(row) } } diff --git a/rust/routee-compass/src/app/compass/response/response_output_policy.rs b/rust/routee-compass/src/app/compass/response/response_output_policy.rs index d79e2825..66699c25 100644 --- a/rust/routee-compass/src/app/compass/response/response_output_policy.rs +++ b/rust/routee-compass/src/app/compass/response/response_output_policy.rs @@ -44,7 +44,7 @@ impl ResponseOutputPolicy { let file_shareable = Arc::new(Mutex::new(file)); let iterations_per_flush = match file_flush_rate { - Some(rate) if *rate <= 0 => Err(CompassAppError::InvalidInput(format!( + Some(rate) if *rate <= 0 => Err(CompassAppError::CompassFailure(format!( "file policy iterations_per_flush must be positive, found {}", rate ))), diff --git a/rust/routee-compass/src/app/compass/response/response_sink.rs b/rust/routee-compass/src/app/compass/response/response_sink.rs index d829292d..08cb8396 100644 --- a/rust/routee-compass/src/app/compass/response/response_sink.rs +++ b/rust/routee-compass/src/app/compass/response/response_sink.rs @@ -25,7 +25,7 @@ impl ResponseSink { match self { ResponseSink::None => Ok(()), ResponseSink::File { - filename: _, + filename, file, format, delimiter: _, @@ -48,10 +48,20 @@ impl ResponseSink { })?; let output_row = format.format_response(response)?; - writeln!(file_attained, "{}", output_row).map_err(CompassAppError::IOError)?; + writeln!(file_attained, "{}", output_row).map_err(|e| { + CompassAppError::InternalError(format!( + "failure writing to {}: {}", + filename, e + )) + })?; *it_attained += 1; if *it_attained % iterations_per_flush == 0 { - file_attained.flush().map_err(CompassAppError::IOError)?; + file_attained.flush().map_err(|e| { + CompassAppError::InternalError(format!( + "failure flushing output to {}: {}", + filename, e + )) + })?; } Ok(()) @@ -87,7 +97,12 @@ impl ResponseSink { let final_contents = format .final_file_contents() .unwrap_or_else(|| String::from("")); - writeln!(file_attained, "{}", final_contents).map_err(CompassAppError::IOError)?; + writeln!(file_attained, "{}", final_contents).map_err(|e| { + CompassAppError::InternalError(format!( + "failure writing final contents to {}: {}", + filename, e + )) + })?; Ok(filename.clone()) } diff --git a/rust/routee-compass/src/app/compass/response/write_mode.rs b/rust/routee-compass/src/app/compass/response/write_mode.rs index 7e2cfb7e..4734f426 100644 --- a/rust/routee-compass/src/app/compass/response/write_mode.rs +++ b/rust/routee-compass/src/app/compass/response/write_mode.rs @@ -54,12 +54,21 @@ fn write_header(path: &Path, format: &ResponseOutputFormat) -> Result<(), Compas let header = format .initial_file_contents() .unwrap_or_else(|| String::from("")); - std::fs::write(path, header).map_err(CompassAppError::IOError) + std::fs::write(path, header).map_err(|e| { + CompassAppError::InternalError(format!( + "failure writing to {}: {}", + path.to_str().unwrap_or_default(), + e + )) + }) } fn open_append(path: &Path) -> Result<File, CompassAppError> { - OpenOptions::new() - .append(true) - .open(path) - .map_err(CompassAppError::IOError) + OpenOptions::new().append(true).open(path).map_err(|e| { + CompassAppError::InternalError(format!( + "failure opening file {} in append mode: {}", + path.to_str().unwrap_or_default(), + e + )) + }) } diff --git a/rust/routee-compass/src/app/geom/geom_app.rs b/rust/routee-compass/src/app/geom/geom_app.rs index 5a51dd1b..e1ad4983 100644 --- a/rust/routee-compass/src/app/geom/geom_app.rs +++ b/rust/routee-compass/src/app/geom/geom_app.rs @@ -24,14 +24,21 @@ impl TryFrom<&GeomAppConfig> for GeomApp { fn try_from(conf: &GeomAppConfig) -> Result<Self, Self::Error> { let count = fs_utils::line_count(conf.edge_file.clone(), fs_utils::is_gzip(&conf.edge_file)) - .map_err(CompassAppError::IOError)?; + .map_err(|e| { + CompassAppError::BuildFailure(format!( + "failure reading edge file {}: {}", + conf.edge_file, e + )) + })?; let mut pb = Bar::builder() .total(count) .animation("fillup") .desc("geometry file") .build() - .map_err(CompassAppError::UXError)?; + .map_err(|e| { + CompassAppError::InternalError(format!("could not build progress bar: {}", e)) + })?; let cb = Box::new(|| { let _ = pb.update(1); @@ -42,9 +49,13 @@ impl TryFrom<&GeomAppConfig> for GeomApp { Ok(result) }; - let geoms = read_utils::read_raw_file(&conf.edge_file, op, Some(cb)) - .map_err(CompassAppError::IOError)?; - println!(); + let geoms = read_utils::read_raw_file(&conf.edge_file, op, Some(cb)).map_err(|e| { + CompassAppError::BuildFailure(format!( + "failure reading edge file {}: {}", + conf.edge_file, e + )) + })?; + eprintln!(); let app = GeomApp { geoms }; Ok(app) } @@ -54,15 +65,21 @@ impl GeomApp { /// run the GeomApp. reads each line of a file, which is expected to be a number coorelating to /// some EdgeId. looks up the geometry for that EdgeId. pub fn run(&self, file: String) -> Result<Box<[LineString<f32>]>, CompassAppError> { - let count = fs_utils::line_count(file.clone(), fs_utils::is_gzip(&file)) - .map_err(CompassAppError::IOError)?; + let count = fs_utils::line_count(file.clone(), fs_utils::is_gzip(&file)).map_err(|e| { + CompassAppError::BuildFailure(format!( + "failure reading geometry index input file {}: {}", + file, e + )) + })?; let mut pb = Bar::builder() .total(count) .animation("fillup") .desc("edge id list") .build() - .map_err(CompassAppError::UXError)?; + .map_err(|e| { + CompassAppError::InternalError(format!("could not build progress bar: {}", e)) + })?; let cb = Box::new(|| { let _ = pb.update(1); @@ -81,8 +98,14 @@ impl GeomApp { result }; - let result: Box<[LineString<f32>]> = - read_utils::read_raw_file(&file, op, Some(cb)).map_err(CompassAppError::IOError)?; + let result: Box<[LineString<f32>]> = read_utils::read_raw_file(&file, op, Some(cb)) + .map_err(|e| { + CompassAppError::BuildFailure(format!( + "failure reading linestring file {}: {}", + file, e + )) + })?; + eprintln!(); Ok(result) } } diff --git a/rust/routee-compass/src/app/search/search_app.rs b/rust/routee-compass/src/app/search/search_app.rs index 834a22b5..5e99f621 100644 --- a/rust/routee-compass/src/app/search/search_app.rs +++ b/rust/routee-compass/src/app/search/search_app.rs @@ -5,7 +5,7 @@ use crate::{ config::cost_model::cost_model_service::CostModelService, search_orientation::SearchOrientation, }, - plugin::input::input_json_extensions::InputJsonExtensions, + plugin::{input::input_json_extensions::InputJsonExtensions, plugin_error::PluginError}, }; use chrono::Local; use routee_compass_core::{ @@ -16,7 +16,7 @@ use routee_compass_core::{ }, model::{ access::access_model_service::AccessModelService, - frontier::frontier_model_service::FrontierModelService, road_network::graph::Graph, + frontier::frontier_model_service::FrontierModelService, network::graph::Graph, state::state_model::StateModel, termination::termination_model::TerminationModel, traversal::traversal_model_service::TraversalModelService, }, @@ -112,35 +112,35 @@ impl SearchApp { &self, query: &serde_json::Value, ) -> Result<(SearchAlgorithmResult, SearchInstance), CompassAppError> { - let o = query - .get_origin_vertex() - .map_err(CompassAppError::PluginError)?; - let d = query - .get_destination_vertex() - .map_err(CompassAppError::PluginError)?; + let o = query.get_origin_vertex().map_err(|e| { + CompassAppError::PluginError(PluginError::InputPluginFailed { source: e }) + })?; + let d = query.get_destination_vertex().map_err(|e| { + CompassAppError::PluginError(PluginError::InputPluginFailed { source: e }) + })?; let search_instance = self.build_search_instance(query)?; self.search_algorithm .run_vertex_oriented(o, d, query, &Direction::Forward, &search_instance) .map(|search_result| (search_result, search_instance)) - .map_err(CompassAppError::SearchError) + .map_err(CompassAppError::SearchFailure) } pub fn run_edge_oriented( &self, query: &serde_json::Value, ) -> Result<(SearchAlgorithmResult, SearchInstance), CompassAppError> { - let o = query - .get_origin_edge() - .map_err(CompassAppError::PluginError)?; - let d_opt = query - .get_destination_edge() - .map_err(CompassAppError::PluginError)?; + let o = query.get_origin_edge().map_err(|e| { + CompassAppError::PluginError(PluginError::InputPluginFailed { source: e }) + })?; + let d_opt = query.get_destination_edge().map_err(|e| { + CompassAppError::PluginError(PluginError::InputPluginFailed { source: e }) + })?; let search_instance = self.build_search_instance(query)?; self.search_algorithm .run_edge_oriented(o, d_opt, query, &Direction::Forward, &search_instance) .map(|search_result| (search_result, search_instance)) - .map_err(CompassAppError::SearchError) + .map_err(CompassAppError::SearchFailure) } /// builds the assets that will run the search for this query instance. diff --git a/rust/routee-compass/src/app/search/search_app_graph_ops.rs b/rust/routee-compass/src/app/search/search_app_graph_ops.rs index 93438469..0d0802a1 100644 --- a/rust/routee-compass/src/app/search/search_app_graph_ops.rs +++ b/rust/routee-compass/src/app/search/search_app_graph_ops.rs @@ -2,51 +2,38 @@ use super::search_app::SearchApp; use crate::app::compass::compass_app_error::CompassAppError; use routee_compass_core::{ algorithm::search::direction::Direction, - model::road_network::{edge_id::EdgeId, vertex_id::VertexId}, + model::network::{edge_id::EdgeId, vertex_id::VertexId}, model::unit::{Distance, DistanceUnit}, }; pub trait SearchAppGraphOps { - fn get_edge_origin(&self, edge_id: EdgeId) -> Result<VertexId, CompassAppError>; - fn get_edge_destination(&self, edge_id: EdgeId) -> Result<VertexId, CompassAppError>; + fn get_edge_origin(&self, edge_id: &EdgeId) -> Result<VertexId, CompassAppError>; + fn get_edge_destination(&self, edge_id: &EdgeId) -> Result<VertexId, CompassAppError>; fn get_edge_distance( &self, - edge_id: EdgeId, + edge_id: &EdgeId, distance_unit: Option<DistanceUnit>, ) -> Result<Distance, CompassAppError>; - fn get_incident_edge_ids( - &self, - vertex_id: VertexId, - direction: Direction, - ) -> Result<Vec<EdgeId>, CompassAppError>; + fn get_incident_edge_ids(&self, vertex_id: &VertexId, direction: &Direction) -> Vec<EdgeId>; } impl SearchAppGraphOps for SearchApp { - fn get_edge_origin(&self, edge_id: EdgeId) -> Result<VertexId, CompassAppError> { - let edge = self - .directed_graph - .get_edge(edge_id) - .map_err(CompassAppError::GraphError)?; + fn get_edge_origin(&self, edge_id: &EdgeId) -> Result<VertexId, CompassAppError> { + let edge = self.directed_graph.get_edge(edge_id)?; Ok(edge.src_vertex_id) } - fn get_edge_destination(&self, edge_id: EdgeId) -> Result<VertexId, CompassAppError> { - let edge = self - .directed_graph - .get_edge(edge_id) - .map_err(CompassAppError::GraphError)?; + fn get_edge_destination(&self, edge_id: &EdgeId) -> Result<VertexId, CompassAppError> { + let edge = self.directed_graph.get_edge(edge_id)?; Ok(edge.dst_vertex_id) } fn get_edge_distance( &self, - edge_id: EdgeId, + edge_id: &EdgeId, distance_unit: Option<DistanceUnit>, ) -> Result<Distance, CompassAppError> { - let edge = self - .directed_graph - .get_edge(edge_id) - .map_err(CompassAppError::GraphError)?; + let edge = self.directed_graph.get_edge(edge_id)?; let result_base = edge.distance; let result = match distance_unit { Some(du) => DistanceUnit::Meters.convert(&result_base, &du), @@ -55,15 +42,7 @@ impl SearchAppGraphOps for SearchApp { Ok(result) } - fn get_incident_edge_ids( - &self, - vertex_id: VertexId, - direction: Direction, - ) -> Result<Vec<EdgeId>, CompassAppError> { - let incident_edges = self - .directed_graph - .incident_edges(vertex_id, direction) - .map_err(CompassAppError::GraphError)?; - Ok(incident_edges) + fn get_incident_edge_ids(&self, vertex_id: &VertexId, direction: &Direction) -> Vec<EdgeId> { + self.directed_graph.incident_edges(vertex_id, direction) } } diff --git a/rust/routee-compass/src/app/search/search_app_ops.rs b/rust/routee-compass/src/app/search/search_app_ops.rs index 7a2ac161..4791237b 100644 --- a/rust/routee-compass/src/app/search/search_app_ops.rs +++ b/rust/routee-compass/src/app/search/search_app_ops.rs @@ -3,7 +3,7 @@ use std::{collections::HashMap, sync::Arc}; use itertools::Itertools; use routee_compass_core::model::{ access::access_model::AccessModel, - state::{state_error::StateError, state_feature::StateFeature}, + state::{state_feature::StateFeature, state_model_error::StateModelError}, traversal::traversal_model::TraversalModel, }; @@ -22,7 +22,7 @@ pub fn collect_features( query: &serde_json::Value, traversal_model: Arc<dyn TraversalModel>, access_model: Arc<dyn AccessModel>, -) -> Result<Vec<(String, StateFeature)>, StateError> { +) -> Result<Vec<(String, StateFeature)>, StateModelError> { // prepare the set of features for this state model let model_features = traversal_model .state_features() @@ -34,17 +34,17 @@ pub fn collect_features( // state features. let user_features_option: Option<HashMap<String, StateFeature>> = query .get_config_serde_optional(&"state_features", &"query") - .map_err(|e| StateError::BuildError(e.to_string()))?; + .map_err(|e| StateModelError::BuildError(e.to_string()))?; let user_features = user_features_option .unwrap_or_default() .into_iter() .map(|(name, feature)| match model_features.get(&name) { None => { let fnames = model_features.keys().join(","); - Err(StateError::UnknownStateVariableName(name, fnames)) + Err(StateModelError::UnknownStateVariableName(name, fnames)) } Some(existing) if existing.get_feature_type() != feature.get_feature_type() => { - Err(StateError::UnexpectedFeatureType( + Err(StateModelError::UnexpectedFeatureType( existing.get_feature_type(), feature.get_feature_type(), )) diff --git a/rust/routee-compass/src/app/search/search_app_result.rs b/rust/routee-compass/src/app/search/search_app_result.rs index 3f71ccb2..c5461cee 100644 --- a/rust/routee-compass/src/app/search/search_app_result.rs +++ b/rust/routee-compass/src/app/search/search_app_result.rs @@ -2,7 +2,7 @@ use allocative::Allocative; use routee_compass_core::{ algorithm::search::{edge_traversal::EdgeTraversal, search_tree_branch::SearchTreeBranch}, - model::road_network::vertex_id::VertexId, + model::network::vertex_id::VertexId, }; use std::{collections::HashMap, time::Duration}; diff --git a/rust/routee-compass/src/plugin/input/default/debug/debug_plugin.rs b/rust/routee-compass/src/plugin/input/default/debug/debug_plugin.rs index 44a6abda..dbfd5daf 100644 --- a/rust/routee-compass/src/plugin/input/default/debug/debug_plugin.rs +++ b/rust/routee-compass/src/plugin/input/default/debug/debug_plugin.rs @@ -1,10 +1,11 @@ -use crate::plugin::{input::input_plugin::InputPlugin, plugin_error::PluginError}; +use crate::plugin::input::{input_plugin::InputPlugin, InputPluginError}; pub struct DebugInputPlugin {} impl InputPlugin for DebugInputPlugin { - fn process(&self, input: &mut serde_json::Value) -> Result<(), PluginError> { - let string = serde_json::to_string_pretty(input).map_err(PluginError::JsonError)?; + fn process(&self, input: &mut serde_json::Value) -> Result<(), InputPluginError> { + let string = serde_json::to_string_pretty(input) + .map_err(|e| InputPluginError::JsonError { source: e })?; println!("{}", string); Ok(()) } diff --git a/rust/routee-compass/src/plugin/input/default/edge_rtree/edge_rtree_input_plugin.rs b/rust/routee-compass/src/plugin/input/default/edge_rtree/edge_rtree_input_plugin.rs index eddfd61e..286e52c9 100644 --- a/rust/routee-compass/src/plugin/input/default/edge_rtree/edge_rtree_input_plugin.rs +++ b/rust/routee-compass/src/plugin/input/default/edge_rtree/edge_rtree_input_plugin.rs @@ -10,14 +10,13 @@ use crate::{ }, }, }, - plugin::{ - input::{input_json_extensions::InputJsonExtensions, input_plugin::InputPlugin}, - plugin_error::PluginError, + plugin::input::{ + input_json_extensions::InputJsonExtensions, input_plugin::InputPlugin, InputPluginError, }, }; use geo_types::Coord; use routee_compass_core::{ - model::road_network::edge_id::EdgeId, + model::network::edge_id::EdgeId, model::unit::{as_f64::AsF64, Distance, DistanceUnit, BASE_DISTANCE_UNIT}, util::{ fs::{read_decoders, read_utils}, @@ -49,9 +48,9 @@ pub struct EdgeRtreeInputPlugin { impl InputPlugin for EdgeRtreeInputPlugin { /// finds the nearest edge ids to the user-provided origin and destination coordinates. /// optionally restricts the search to a subset of road classes tagged by the user. - fn process(&self, query: &mut serde_json::Value) -> Result<(), PluginError> { + fn process(&self, query: &mut serde_json::Value) -> Result<(), InputPluginError> { let road_classes = self.road_class_parser.read_query(query).map_err(|e| { - PluginError::InputError(format!( + InputPluginError::InputPluginFailed(format!( "Unable to apply EdgeRtree Input Plugin due to:\n\n{}", e )) @@ -188,7 +187,7 @@ fn search( road_classes: &Option<HashSet<u8>>, vehicle_restrictions: &Option<HashMap<EdgeId, Vec<VehicleRestriction>>>, vehicle_parameters: &Option<VehicleParameters>, -) -> Result<Option<EdgeId>, PluginError> { +) -> Result<Option<EdgeId>, InputPluginError> { let point = geo::Point(coord); for (record, distance_meters) in rtree.nearest_neighbor_iter_with_distance_2(&point) { if !within_tolerance(tolerance, &distance_meters) { @@ -197,7 +196,7 @@ fn search( let valid_class = match (road_classes, road_class_lookup) { (Some(valid_classes), Some(lookup)) => { let this_class = lookup.get(record.edge_id.0).ok_or_else(|| { - PluginError::PluginFailed(format!( + InputPluginError::InputPluginFailed(format!( "edge rtree road class file missing edge {}", record.edge_id )) @@ -227,12 +226,15 @@ fn search( } /// helper to build a matching error response -fn matching_error(coord: &Coord<f32>, tolerance: Option<(Distance, DistanceUnit)>) -> PluginError { +fn matching_error( + coord: &Coord<f32>, + tolerance: Option<(Distance, DistanceUnit)>, +) -> InputPluginError { let mut message = format!("unable to match coordinate {:?} to network edge", coord); if let Some((dist, unit)) = tolerance { message.push_str(&format!(" within tolerance of {} {}", dist, unit)) }; - PluginError::PluginFailed(message) + InputPluginError::InputPluginFailed(message) } /// helper to test if some distance in meters is within the optionally-provided tolerance diff --git a/rust/routee-compass/src/plugin/input/default/edge_rtree/edge_rtree_record.rs b/rust/routee-compass/src/plugin/input/default/edge_rtree/edge_rtree_record.rs index 672bc36d..f0c67c62 100644 --- a/rust/routee-compass/src/plugin/input/default/edge_rtree/edge_rtree_record.rs +++ b/rust/routee-compass/src/plugin/input/default/edge_rtree/edge_rtree_record.rs @@ -1,5 +1,5 @@ use geo::{Centroid, LineString, Point}; -use routee_compass_core::model::road_network::edge_id::EdgeId; +use routee_compass_core::model::network::edge_id::EdgeId; use rstar::{PointDistance, RTreeObject, AABB}; pub struct EdgeRtreeRecord { diff --git a/rust/routee-compass/src/plugin/input/default/grid_search/plugin.rs b/rust/routee-compass/src/plugin/input/default/grid_search/plugin.rs index 9eb18f12..fb62246b 100644 --- a/rust/routee-compass/src/plugin/input/default/grid_search/plugin.rs +++ b/rust/routee-compass/src/plugin/input/default/grid_search/plugin.rs @@ -1,7 +1,7 @@ use crate::plugin::input::input_field::InputField; use crate::plugin::input::input_json_extensions::InputJsonExtensions; use crate::plugin::input::input_plugin::InputPlugin; -use crate::plugin::plugin_error::PluginError; +use crate::plugin::input::InputPluginError; use routee_compass_core::util::multiset::MultiSet; /// Builds an input plugin that duplicates queries if array-valued fields are present @@ -9,23 +9,23 @@ use routee_compass_core::util::multiset::MultiSet; pub struct GridSearchPlugin {} impl InputPlugin for GridSearchPlugin { - fn process(&self, input: &mut serde_json::Value) -> Result<(), PluginError> { + fn process(&self, input: &mut serde_json::Value) -> Result<(), InputPluginError> { match input.get_grid_search() { None => Ok(()), Some(grid_search_input) => { // prevent recursion due to nested grid search keys let recurses = serde_json::to_string(grid_search_input) - .map_err(PluginError::JsonError)? + .map_err(|e| InputPluginError::JsonError { source: e })? .contains("grid_search"); if recurses { - return Err(PluginError::PluginFailed(String::from( + return Err(InputPluginError::InputPluginFailed(String::from( "grid search section cannot contain the string 'grid_search'", ))); } - let map = grid_search_input - .as_object() - .ok_or_else(|| PluginError::UnexpectedQueryStructure(format!("{:?}", input)))?; + let map = grid_search_input.as_object().ok_or_else(|| { + InputPluginError::UnexpectedQueryStructure(format!("{:?}", input)) + })?; let mut keys: Vec<String> = vec![]; let mut multiset_input: Vec<Vec<serde_json::Value>> = vec![]; let mut multiset_indices: Vec<Vec<usize>> = vec![]; @@ -42,7 +42,9 @@ impl InputPlugin for GridSearchPlugin { // let remove_key = InputField::GridSearch.to_str(); let mut initial_map = input .as_object() - .ok_or_else(|| PluginError::UnexpectedQueryStructure(format!("{:?}", input)))? + .ok_or_else(|| { + InputPluginError::UnexpectedQueryStructure(format!("{:?}", input)) + })? .clone(); initial_map.remove(InputField::GridSearch.to_str()); let initial = serde_json::json!(initial_map); diff --git a/rust/routee-compass/src/plugin/input/default/inject/inject_builder.rs b/rust/routee-compass/src/plugin/input/default/inject/inject_builder.rs index 1949a805..b2c8a45a 100644 --- a/rust/routee-compass/src/plugin/input/default/inject/inject_builder.rs +++ b/rust/routee-compass/src/plugin/input/default/inject/inject_builder.rs @@ -20,7 +20,9 @@ impl InputPluginBuilder for InjectPluginBuilder { let value_string = parameters.get_config_string(&"value", &"inject")?; let format: InjectFormat = parameters.get_config_serde(&"format", &"inject")?; let value = format.to_json(&value_string)?; - let plugin = InjectInputPlugin::new(key, value); + let overwrite: Option<bool> = + parameters.get_config_serde_optional(&"overwrite", &"inject")?; + let plugin = InjectInputPlugin::new(key, value, overwrite); Ok(Arc::new(plugin)) } } diff --git a/rust/routee-compass/src/plugin/input/default/inject/inject_format.rs b/rust/routee-compass/src/plugin/input/default/inject/inject_format.rs index 0abdde00..ca16a714 100644 --- a/rust/routee-compass/src/plugin/input/default/inject/inject_format.rs +++ b/rust/routee-compass/src/plugin/input/default/inject/inject_format.rs @@ -1,11 +1,7 @@ +use crate::app::compass::config::compass_configuration_error::CompassConfigurationError; use routee_compass_core::util::serde::serde_ops; use serde::{Deserialize, Serialize}; -use crate::{ - app::compass::config::compass_configuration_error::CompassConfigurationError, - plugin::plugin_error::PluginError, -}; - #[derive(Debug, Serialize, Deserialize, Clone)] #[serde(rename_all = "snake_case")] pub enum InjectFormat { @@ -20,19 +16,19 @@ impl InjectFormat { InjectFormat::String => { let decode_result = serde_ops::string_deserialize(value); decode_result.map_err(|e| { - CompassConfigurationError::PluginError(PluginError::InputError(format!( + CompassConfigurationError::UserConfigurationError(format!( "could not deserialize inject value as string: {}", e - ))) + )) }) } InjectFormat::Json => { let result = serde_json::from_str(value); result.map_err(|e| { - CompassConfigurationError::PluginError(PluginError::InputError(format!( + CompassConfigurationError::UserConfigurationError(format!( "could not deserialize inject value as JSON: {}", e - ))) + )) }) } InjectFormat::Toml => todo!(), diff --git a/rust/routee-compass/src/plugin/input/default/inject/inject_plugin.rs b/rust/routee-compass/src/plugin/input/default/inject/inject_plugin.rs index 7aebc12d..596b3a58 100644 --- a/rust/routee-compass/src/plugin/input/default/inject/inject_plugin.rs +++ b/rust/routee-compass/src/plugin/input/default/inject/inject_plugin.rs @@ -1,18 +1,38 @@ -use crate::plugin::{input::input_plugin::InputPlugin, plugin_error::PluginError}; +use crate::plugin::input::{input_plugin::InputPlugin, InputPluginError}; pub struct InjectInputPlugin { key: String, value: serde_json::Value, + overwrite: bool, } impl InjectInputPlugin { - pub fn new(key: String, value: serde_json::Value) -> InjectInputPlugin { - InjectInputPlugin { key, value } + pub fn new( + key: String, + value: serde_json::Value, + overwrite: Option<bool>, + ) -> InjectInputPlugin { + InjectInputPlugin { + key, + value, + overwrite: overwrite.unwrap_or(true), + } } } impl InputPlugin for InjectInputPlugin { - fn process(&self, input: &mut serde_json::Value) -> Result<(), PluginError> { + fn process(&self, input: &mut serde_json::Value) -> Result<(), InputPluginError> { + if !self.overwrite { + if let Some(obj) = input.as_object() { + if obj.contains_key(&self.key) { + return Err(InputPluginError::InputPluginFailed(format!("inject plugin for key {} has a no-overwrite policy but encountered query with existing value", self.key))); + } + } else { + return Err(InputPluginError::UnexpectedQueryStructure(String::from( + "query is not a JSON object", + ))); + } + } input[self.key.clone()] = self.value.clone(); Ok(()) } diff --git a/rust/routee-compass/src/plugin/input/default/load_balancer/custom_weight_type.rs b/rust/routee-compass/src/plugin/input/default/load_balancer/custom_weight_type.rs index a0fae675..41db2230 100644 --- a/rust/routee-compass/src/plugin/input/default/load_balancer/custom_weight_type.rs +++ b/rust/routee-compass/src/plugin/input/default/load_balancer/custom_weight_type.rs @@ -4,12 +4,11 @@ use serde::{Deserialize, Serialize}; use crate::{ app::compass::config::config_json_extension::ConfigJsonExtensions, - plugin::{input::input_field::InputField, plugin_error::PluginError}, + plugin::input::{input_field::InputField, InputPluginError}, }; #[derive(Debug, Serialize, Deserialize, Clone)] #[serde(rename_all = "snake_case", tag = "type")] - pub enum CustomWeightType { /// a weight value found on each query which can be used directly as it is a numeric /// field. will use provided column_name or fall back to InputField::QueryWeightEstimate. @@ -27,7 +26,7 @@ pub enum CustomWeightType { } impl CustomWeightType { - pub fn get_weight(&self, query: &serde_json::Value) -> Result<f64, PluginError> { + pub fn get_weight(&self, query: &serde_json::Value) -> Result<f64, InputPluginError> { match self { CustomWeightType::Numeric { column_name } => { let col: String = column_name @@ -35,7 +34,12 @@ impl CustomWeightType { .unwrap_or(InputField::QueryWeightEstimate.to_string()); let value = query .get_config_f64(&col, &"custom_weight_type") - .map_err(|_| PluginError::ParseError(col, String::from("String")))?; + .map_err(|_| { + InputPluginError::QueryFieldHasInvalidType( + InputField::Custom(col), + String::from("String"), + ) + })?; Ok(value) } CustomWeightType::Categorical { @@ -48,11 +52,16 @@ impl CustomWeightType { .unwrap_or(InputField::QueryWeightEstimate.to_string()); let categorical_value = query .get_config_string(&col, &"custom_weight_type") - .map_err(|_| PluginError::ParseError(col.clone(), String::from("String")))?; + .map_err(|_| { + InputPluginError::QueryFieldHasInvalidType( + InputField::Custom(col), + String::from("String"), + ) + })?; match (mapping.get(&categorical_value), default) { (Some(result), _) => Ok(*result), (None, Some(fallback)) => Ok(*fallback), - _ => Err(PluginError::InputError(format!("load balancing categorical {} not found in mapping and no default was provided", categorical_value))), + _ => Err(InputPluginError::InputPluginFailed(format!("load balancing categorical {} not found in mapping and no default was provided", categorical_value))), } } } diff --git a/rust/routee-compass/src/plugin/input/default/load_balancer/plugin.rs b/rust/routee-compass/src/plugin/input/default/load_balancer/plugin.rs index 5bf587c2..56c3ae4d 100644 --- a/rust/routee-compass/src/plugin/input/default/load_balancer/plugin.rs +++ b/rust/routee-compass/src/plugin/input/default/load_balancer/plugin.rs @@ -1,14 +1,14 @@ use super::weight_heuristic::WeightHeuristic; use crate::plugin::input::input_json_extensions::InputJsonExtensions; use crate::plugin::input::input_plugin::InputPlugin; -use crate::plugin::plugin_error::PluginError; +use crate::plugin::input::InputPluginError; pub struct LoadBalancerPlugin { pub heuristic: WeightHeuristic, } impl InputPlugin for LoadBalancerPlugin { - fn process(&self, query: &mut serde_json::Value) -> Result<(), PluginError> { + fn process(&self, query: &mut serde_json::Value) -> Result<(), InputPluginError> { let w = self.heuristic.estimate_weight(query)?; let _updated = query.clone(); query.add_query_weight_estimate(w)?; diff --git a/rust/routee-compass/src/plugin/input/default/load_balancer/weight_heuristic.rs b/rust/routee-compass/src/plugin/input/default/load_balancer/weight_heuristic.rs index 77077fad..f3d90786 100644 --- a/rust/routee-compass/src/plugin/input/default/load_balancer/weight_heuristic.rs +++ b/rust/routee-compass/src/plugin/input/default/load_balancer/weight_heuristic.rs @@ -1,5 +1,5 @@ use super::custom_weight_type::CustomWeightType; -use crate::plugin::{input::input_json_extensions::InputJsonExtensions, plugin_error::PluginError}; +use crate::plugin::{input::input_json_extensions::InputJsonExtensions, input::InputPluginError}; use routee_compass_core::model::unit::{as_f64::AsF64, DistanceUnit}; use routee_compass_core::util::geo::haversine; use serde::{Deserialize, Serialize}; @@ -18,19 +18,19 @@ pub enum WeightHeuristic { } impl WeightHeuristic { - pub fn estimate_weight(&self, query: &serde_json::Value) -> Result<f64, PluginError> { + pub fn estimate_weight(&self, query: &serde_json::Value) -> Result<f64, InputPluginError> { match self { WeightHeuristic::Haversine => { let o = query.get_origin_coordinate()?; let d_option = query.get_destination_coordinate()?; match d_option { - None => Err(PluginError::InputError(String::from( + None => Err(InputPluginError::InputPluginFailed(String::from( "cannot estimate search size without destination coordinate", ))), Some(d) => haversine::coord_distance(&o, &d, DistanceUnit::Kilometers) .map(|d| d.as_f64()) .map_err(|s| { - PluginError::PluginFailed(format!( + InputPluginError::InputPluginFailed(format!( "failed calculating load balancing weight value due to {}", s )) diff --git a/rust/routee-compass/src/plugin/input/default/vertex_rtree/plugin.rs b/rust/routee-compass/src/plugin/input/default/vertex_rtree/plugin.rs index e08c5395..68c94535 100644 --- a/rust/routee-compass/src/plugin/input/default/vertex_rtree/plugin.rs +++ b/rust/routee-compass/src/plugin/input/default/vertex_rtree/plugin.rs @@ -1,12 +1,12 @@ use std::path::Path; -use crate::plugin::input::input_json_extensions::InputJsonExtensions; use crate::plugin::input::input_plugin::InputPlugin; +use crate::plugin::input::{input_json_extensions::InputJsonExtensions, InputPluginError}; use crate::plugin::plugin_error::PluginError; use geo::{coord, Coord}; use routee_compass_core::{ model::unit::{Distance, DistanceUnit, BASE_DISTANCE_UNIT}, - model::{property::vertex::Vertex, road_network::graph::Graph}, + model::{network::graph::Graph, network::Vertex}, util::{fs::read_utils, geo::haversine}, }; use rstar::{PointDistance, RTree, RTreeObject, AABB}; @@ -110,7 +110,9 @@ impl RTreePlugin { distance_unit: Option<DistanceUnit>, ) -> Result<Self, PluginError> { let vertices: Box<[Vertex]> = - read_utils::from_csv(&vertex_file, true, None).map_err(PluginError::CsvReadError)?; + read_utils::from_csv(&vertex_file, true, None).map_err(|e| { + InputPluginError::BuildFailed(format!("failure reading vertex file: {}", e)) + })?; let vertex_rtree = VertexRTree::new(vertices.to_vec()); let tolerance = match (tolerance_distance, distance_unit) { (None, None) => None, @@ -136,12 +138,12 @@ impl InputPlugin for RTreePlugin { /// /// * either vertex ids for the nearest coordinates to the the origin (and optionally destination), /// or, an error if not found or not within tolerance - fn process(&self, query: &mut serde_json::Value) -> Result<(), PluginError> { + fn process(&self, query: &mut serde_json::Value) -> Result<(), InputPluginError> { let src_coord = query.get_origin_coordinate()?; let dst_coord_option = query.get_destination_coordinate()?; let src_vertex = self.vertex_rtree.nearest_vertex(src_coord).ok_or_else(|| { - PluginError::PluginFailed(format!( + InputPluginError::InputPluginFailed(format!( "nearest vertex not found for origin coordinate {:?}", src_coord )) @@ -154,7 +156,7 @@ impl InputPlugin for RTreePlugin { None => {} Some(dst_coord) => { let dst_vertex = self.vertex_rtree.nearest_vertex(dst_coord).ok_or_else(|| { - PluginError::PluginFailed(format!( + InputPluginError::InputPluginFailed(format!( "nearest vertex not found for destination coordinate {:?}", dst_coord )) @@ -187,14 +189,14 @@ fn validate_tolerance( src: &Coord<f32>, dst: &Coord<f32>, tolerance: &Option<(Distance, DistanceUnit)>, -) -> Result<(), PluginError> { +) -> Result<(), InputPluginError> { match tolerance { Some((tolerance_distance, tolerance_distance_unit)) => { - let distance_meters = - haversine::coord_distance_meters(src, dst).map_err(PluginError::PluginFailed)?; + let distance_meters = haversine::coord_distance_meters(src, dst) + .map_err(InputPluginError::InputPluginFailed)?; let distance = DistanceUnit::Meters.convert(&distance_meters, tolerance_distance_unit); if &distance >= tolerance_distance { - Err(PluginError::PluginFailed( + Err(InputPluginError::InputPluginFailed( format!( "coord {:?} nearest vertex coord is {:?} which is {} {} away, exceeding the distance tolerance of {} {}", src, diff --git a/rust/routee-compass/src/plugin/input/input_field.rs b/rust/routee-compass/src/plugin/input/input_field.rs index cb475cca..69e2967c 100644 --- a/rust/routee-compass/src/plugin/input/input_field.rs +++ b/rust/routee-compass/src/plugin/input/input_field.rs @@ -1,5 +1,7 @@ +use serde::{Deserialize, Serialize}; use std::fmt::Display; +#[derive(Debug, Serialize, Deserialize)] pub enum InputField { OriginX, OriginY, @@ -11,10 +13,11 @@ pub enum InputField { DestinationEdge, GridSearch, QueryWeightEstimate, + Custom(String), } impl InputField { - pub fn to_str(&self) -> &'static str { + pub fn to_str(&self) -> &str { use InputField as I; match self { I::OriginX => "origin_x", @@ -27,6 +30,7 @@ impl InputField { I::DestinationEdge => "destination_edge", I::GridSearch => "grid_search", I::QueryWeightEstimate => "query_weight_estimate", + I::Custom(field) => field, } } } diff --git a/rust/routee-compass/src/plugin/input/input_json_extensions.rs b/rust/routee-compass/src/plugin/input/input_json_extensions.rs index b38d7770..48822dbd 100644 --- a/rust/routee-compass/src/plugin/input/input_json_extensions.rs +++ b/rust/routee-compass/src/plugin/input/input_json_extensions.rs @@ -1,70 +1,69 @@ -use super::input_field::InputField; -use crate::plugin::plugin_error::PluginError; +use crate::{app::compass::compass_app_error::CompassAppError, plugin::plugin_error::PluginError}; + +use super::{InputField, InputPluginError}; use geo; -use routee_compass_core::model::road_network::{edge_id::EdgeId, vertex_id::VertexId}; +use routee_compass_core::model::network::{edge_id::EdgeId, vertex_id::VertexId}; use serde_json::{self, json}; pub trait InputJsonExtensions { - fn get_origin_coordinate(&self) -> Result<geo::Coord<f32>, PluginError>; - fn get_destination_coordinate(&self) -> Result<Option<geo::Coord<f32>>, PluginError>; - fn add_origin_vertex(&mut self, vertex_id: VertexId) -> Result<(), PluginError>; - fn add_destination_vertex(&mut self, vertex_id: VertexId) -> Result<(), PluginError>; - fn add_origin_edge(&mut self, edge_id: EdgeId) -> Result<(), PluginError>; - fn add_destination_edge(&mut self, edge_id: EdgeId) -> Result<(), PluginError>; - fn get_origin_vertex(&self) -> Result<VertexId, PluginError>; - fn get_destination_vertex(&self) -> Result<Option<VertexId>, PluginError>; - fn get_origin_edge(&self) -> Result<EdgeId, PluginError>; - fn get_destination_edge(&self) -> Result<Option<EdgeId>, PluginError>; + fn get_origin_coordinate(&self) -> Result<geo::Coord<f32>, InputPluginError>; + fn get_destination_coordinate(&self) -> Result<Option<geo::Coord<f32>>, InputPluginError>; + fn add_origin_vertex(&mut self, vertex_id: VertexId) -> Result<(), InputPluginError>; + fn add_destination_vertex(&mut self, vertex_id: VertexId) -> Result<(), InputPluginError>; + fn add_origin_edge(&mut self, edge_id: EdgeId) -> Result<(), InputPluginError>; + fn add_destination_edge(&mut self, edge_id: EdgeId) -> Result<(), InputPluginError>; + fn get_origin_vertex(&self) -> Result<VertexId, InputPluginError>; + fn get_destination_vertex(&self) -> Result<Option<VertexId>, InputPluginError>; + fn get_origin_edge(&self) -> Result<EdgeId, InputPluginError>; + fn get_destination_edge(&self) -> Result<Option<EdgeId>, InputPluginError>; fn get_grid_search(&self) -> Option<&serde_json::Value>; - fn add_query_weight_estimate(&mut self, weight: f64) -> Result<(), PluginError>; - fn get_query_weight_estimate(&self) -> Result<Option<f64>, PluginError>; + fn add_query_weight_estimate(&mut self, weight: f64) -> Result<(), InputPluginError>; + fn get_query_weight_estimate(&self) -> Result<Option<f64>, CompassAppError>; } impl InputJsonExtensions for serde_json::Value { - fn get_origin_coordinate(&self) -> Result<geo::Coord<f32>, PluginError> { + fn get_origin_coordinate(&self) -> Result<geo::Coord<f32>, InputPluginError> { let origin_x = self - .get(InputField::OriginX.to_string()) - .ok_or_else(|| PluginError::MissingField(InputField::OriginX.to_string()))? + .get(InputField::OriginX.to_str()) + .ok_or(InputPluginError::MissingExpectedQueryField( + InputField::OriginX, + ))? .as_f64() .ok_or_else(|| { - PluginError::ParseError(InputField::OriginX.to_string(), String::from("f64")) + InputPluginError::QueryFieldHasInvalidType(InputField::OriginX, String::from("f64")) })?; let origin_y = self - .get(InputField::OriginY.to_string()) - .ok_or_else(|| PluginError::MissingField(InputField::OriginY.to_string()))? + .get(InputField::OriginY.to_str()) + .ok_or(InputPluginError::MissingExpectedQueryField( + InputField::OriginY, + ))? .as_f64() .ok_or_else(|| { - PluginError::ParseError(InputField::OriginY.to_string(), String::from("f64")) + InputPluginError::QueryFieldHasInvalidType(InputField::OriginY, String::from("f64")) })?; Ok(geo::Coord::from((origin_x as f32, origin_y as f32))) } - fn get_destination_coordinate(&self) -> Result<Option<geo::Coord<f32>>, PluginError> { - let x_field = InputField::DestinationX.to_string(); - let y_field = InputField::DestinationY.to_string(); - let x_opt = self.get(&x_field); - let y_opt = self.get(&y_field); + fn get_destination_coordinate(&self) -> Result<Option<geo::Coord<f32>>, InputPluginError> { + let x_field = InputField::DestinationX; + let y_field = InputField::DestinationY; + let x_opt = self.get(x_field.to_str()); + let y_opt = self.get(y_field.to_str()); match (x_opt, y_opt) { (None, None) => Ok(None), - (None, Some(_)) => Err(PluginError::MissingField(format!( - "{} provided without {}", - &y_field, &x_field - ))), - (Some(_), None) => Err(PluginError::MissingField(format!( - "{} provided without {}", - &x_field, &y_field - ))), + (None, Some(_)) => Err(InputPluginError::MissingQueryFieldPair(y_field, x_field)), + (Some(_), None) => Err(InputPluginError::MissingQueryFieldPair(x_field, y_field)), (Some(x_json), Some(y_json)) => { - let x = x_json - .as_f64() - .ok_or_else(|| PluginError::ParseError(x_field.clone(), String::from("f64")))?; - let y = y_json - .as_f64() - .ok_or_else(|| PluginError::ParseError(y_field.clone(), String::from("f64")))?; + let x = x_json.as_f64().ok_or_else(|| { + InputPluginError::QueryFieldHasInvalidType(x_field, String::from("f64")) + })?; + let y = y_json.as_f64().ok_or_else(|| { + InputPluginError::QueryFieldHasInvalidType(y_field, String::from("f64")) + })?; Ok(Some(geo::Coord::from((x as f32, y as f32)))) } } } - fn add_origin_vertex(&mut self, vertex_id: VertexId) -> Result<(), PluginError> { + fn add_origin_vertex(&mut self, vertex_id: VertexId) -> Result<(), InputPluginError> { match self { serde_json::Value::Object(map) => { map.insert( @@ -73,12 +72,12 @@ impl InputJsonExtensions for serde_json::Value { ); Ok(()) } - _ => Err(PluginError::InputError(String::from( + _ => Err(InputPluginError::UnexpectedQueryStructure(String::from( "InputQuery is not a JSON object", ))), } } - fn add_destination_vertex(&mut self, vertex_id: VertexId) -> Result<(), PluginError> { + fn add_destination_vertex(&mut self, vertex_id: VertexId) -> Result<(), InputPluginError> { match self { serde_json::Value::Object(map) => { map.insert( @@ -87,60 +86,73 @@ impl InputJsonExtensions for serde_json::Value { ); Ok(()) } - _ => Err(PluginError::InputError(String::from( + _ => Err(InputPluginError::UnexpectedQueryStructure(String::from( "InputQuery is not a JSON object", ))), } } - fn get_origin_vertex(&self) -> Result<VertexId, PluginError> { - self.get(InputField::OriginVertex.to_string()) - .ok_or_else(|| PluginError::MissingField(InputField::OriginVertex.to_string()))? + fn get_origin_vertex(&self) -> Result<VertexId, InputPluginError> { + self.get(InputField::OriginVertex.to_str()) + .ok_or(InputPluginError::MissingExpectedQueryField( + InputField::OriginVertex, + ))? .as_u64() .map(|v| VertexId(v as usize)) .ok_or_else(|| { - PluginError::ParseError(InputField::OriginVertex.to_string(), String::from("u64")) + InputPluginError::QueryFieldHasInvalidType( + InputField::OriginVertex, + String::from("u64"), + ) }) } - fn get_destination_vertex(&self) -> Result<Option<VertexId>, PluginError> { - match self.get(InputField::DestinationVertex.to_string()) { + fn get_destination_vertex(&self) -> Result<Option<VertexId>, InputPluginError> { + match self.get(InputField::DestinationVertex.to_str()) { None => Ok(None), Some(v) => v .as_u64() .map(|v| Some(VertexId(v as usize))) .ok_or_else(|| { - PluginError::ParseError( - InputField::DestinationVertex.to_string(), + InputPluginError::QueryFieldHasInvalidType( + InputField::DestinationVertex, String::from("u64"), ) }), } } - fn get_origin_edge(&self) -> Result<EdgeId, PluginError> { - self.get(InputField::OriginEdge.to_string()) - .ok_or_else(|| PluginError::MissingField(InputField::OriginEdge.to_string()))? + fn get_origin_edge(&self) -> Result<EdgeId, InputPluginError> { + self.get(InputField::OriginEdge.to_str()) + .ok_or(InputPluginError::MissingExpectedQueryField( + InputField::OriginEdge, + ))? .as_u64() .map(|v| EdgeId(v as usize)) .ok_or_else(|| { - PluginError::ParseError(InputField::OriginEdge.to_string(), String::from("u64")) + InputPluginError::QueryFieldHasInvalidType( + InputField::OriginEdge, + String::from("u64"), + ) }) } - fn get_destination_edge(&self) -> Result<Option<EdgeId>, PluginError> { - match self.get(InputField::DestinationEdge.to_string()) { + fn get_destination_edge(&self) -> Result<Option<EdgeId>, InputPluginError> { + match self.get(InputField::DestinationEdge.to_str()) { None => Ok(None), Some(v) => v.as_u64().map(|v| Some(EdgeId(v as usize))).ok_or_else(|| { - PluginError::ParseError(InputField::OriginEdge.to_string(), String::from("u64")) + InputPluginError::QueryFieldHasInvalidType( + InputField::OriginEdge, + String::from("u64"), + ) }), } } fn get_grid_search(&self) -> Option<&serde_json::Value> { - self.get(InputField::GridSearch.to_string()) + self.get(InputField::GridSearch.to_str()) } - fn add_origin_edge(&mut self, edge_id: EdgeId) -> Result<(), PluginError> { + fn add_origin_edge(&mut self, edge_id: EdgeId) -> Result<(), InputPluginError> { match self { serde_json::Value::Object(map) => { map.insert( @@ -149,13 +161,13 @@ impl InputJsonExtensions for serde_json::Value { ); Ok(()) } - _ => Err(PluginError::InputError(String::from( + _ => Err(InputPluginError::UnexpectedQueryStructure(String::from( "InputQuery is not a JSON object", ))), } } - fn add_destination_edge(&mut self, edge_id: EdgeId) -> Result<(), PluginError> { + fn add_destination_edge(&mut self, edge_id: EdgeId) -> Result<(), InputPluginError> { match self { serde_json::Value::Object(map) => { map.insert( @@ -164,32 +176,34 @@ impl InputJsonExtensions for serde_json::Value { ); Ok(()) } - _ => Err(PluginError::InputError(String::from( + _ => Err(InputPluginError::UnexpectedQueryStructure(String::from( "InputQuery is not a JSON object", ))), } } - fn add_query_weight_estimate(&mut self, weight: f64) -> Result<(), PluginError> { + fn add_query_weight_estimate(&mut self, weight: f64) -> Result<(), InputPluginError> { match self { serde_json::Value::Object(map) => { map.insert(InputField::QueryWeightEstimate.to_string(), json!(weight)); Ok(()) } - _ => Err(PluginError::InputError(String::from( + _ => Err(InputPluginError::UnexpectedQueryStructure(String::from( "InputQuery is not a JSON object", ))), } } - fn get_query_weight_estimate(&self) -> Result<Option<f64>, PluginError> { - match self.get(InputField::QueryWeightEstimate.to_string()) { + fn get_query_weight_estimate(&self) -> Result<Option<f64>, CompassAppError> { + match self.get(InputField::QueryWeightEstimate.to_str()) { None => Ok(None), Some(v) => v.as_f64().map(Some).ok_or_else(|| { - PluginError::ParseError( - InputField::QueryWeightEstimate.to_string(), + let ipe = InputPluginError::QueryFieldHasInvalidType( + InputField::QueryWeightEstimate, String::from("f64"), - ) + ); + let pe = PluginError::InputPluginFailed { source: ipe }; + CompassAppError::PluginError(pe) }), } } @@ -201,16 +215,16 @@ impl InputJsonExtensions for serde_json::Value { // value: &serde_json::Value, // field: InputField, // op: DecodeOp<T>, -// ) -> Result<T, PluginError> { +// ) -> Result<T, InputPluginError> { // let at_field = value.get(field.to_string()); // match at_field { -// None => Err(PluginError::MissingField(field.to_string())), -// Some(v) => op(v).ok_or_else( ||PluginError::ParseError(field.to_string(), ())), +// None => Err(InputPluginError::MissingField(field.to_string())), +// Some(v) => op(v).ok_or_else( ||InputPluginError::QueryFieldHasInvalidType(field.to_string(), ())), // }; // } -// fn get_f64(v: &serde_json::Value) -> Result<f64, PluginError> { +// fn get_f64(v: &serde_json::Value) -> Result<f64, InputPluginError> { // get_from_json(v, field, |v| v.as_f64()) -// v.as_f64().ok_or_else( ||PluginError::ParseError((), ()) +// v.as_f64().ok_or_else( ||InputPluginError::QueryFieldHasInvalidType((), ()) // } diff --git a/rust/routee-compass/src/plugin/input/input_plugin.rs b/rust/routee-compass/src/plugin/input/input_plugin.rs index 49cfb748..401b9dd9 100644 --- a/rust/routee-compass/src/plugin/input/input_plugin.rs +++ b/rust/routee-compass/src/plugin/input/input_plugin.rs @@ -1,4 +1,4 @@ -use crate::plugin::plugin_error::PluginError; +use super::InputPluginError; /// Performs some kind of pre-processing on a user query input. The input JSON is available /// to the plugin as a reference. The plugin produces a vector of zero to many JSON objects that will @@ -26,5 +26,5 @@ pub trait InputPlugin: Send + Sync { /// # Returns /// /// A `Vec` of JSON values to replace the input JSON, or an error - fn process(&self, input: &mut serde_json::Value) -> Result<(), PluginError>; + fn process(&self, input: &mut serde_json::Value) -> Result<(), InputPluginError>; } diff --git a/rust/routee-compass/src/plugin/input/input_plugin_error.rs b/rust/routee-compass/src/plugin/input/input_plugin_error.rs new file mode 100644 index 00000000..a0bb4bc9 --- /dev/null +++ b/rust/routee-compass/src/plugin/input/input_plugin_error.rs @@ -0,0 +1,24 @@ +use super::input_field::InputField; + +#[derive(thiserror::Error, Debug)] +pub enum InputPluginError { + #[error("failure building input plugin: {0}")] + BuildFailed(String), + #[error("required query field '{0}' not found")] + MissingExpectedQueryField(InputField), + #[error("{0} provided without {1}")] + MissingQueryFieldPair(InputField, InputField), + #[error("required query field '{0}' is not of type {1}")] + QueryFieldHasInvalidType(InputField, String), + #[error("expected query to be a json object '{{}}' but found {0}")] + UnexpectedQueryStructure(String), + #[error("plugin experienced JSON error: {source}")] + JsonError { + #[from] + source: serde_json::Error, + }, + #[error("failure running plugin: {0}")] + InputPluginFailed(String), + #[error("unexpected error: {0}")] + InternalError(String), +} diff --git a/rust/routee-compass/src/plugin/input/input_plugin_ops.rs b/rust/routee-compass/src/plugin/input/input_plugin_ops.rs index f2cceb96..a94eb008 100644 --- a/rust/routee-compass/src/plugin/input/input_plugin_ops.rs +++ b/rust/routee-compass/src/plugin/input/input_plugin_ops.rs @@ -1,8 +1,8 @@ -use std::rc::Rc; - -use crate::plugin::plugin_error::PluginError; use indoc::indoc; use serde_json::{json, Value}; +use std::rc::Rc; + +use super::InputPluginError; /// helper to return errors as JSON response objects which include the /// original request along with the error message @@ -62,13 +62,13 @@ pub fn package_invariant_error( } } -pub type ArrayOp<'a> = Rc<dyn Fn(&mut Value) -> Result<(), PluginError> + 'a>; +pub type InputArrayOp<'a> = Rc<dyn Fn(&mut Value) -> Result<(), InputPluginError> + 'a>; /// executes an operation on an input query. maintains the invariant that /// input queries should always remain wrapped in a top-level JSON Array /// so that we can perform operations like grid search, which transform a /// single query into multiple child queries. -pub fn json_array_op<'a>(query: &'a mut Value, op: ArrayOp<'a>) -> Result<(), Value> { +pub fn json_array_op<'a>(query: &'a mut Value, op: InputArrayOp<'a>) -> Result<(), Value> { match query { Value::Array(queries) => { for q in queries.iter_mut() { diff --git a/rust/routee-compass/src/plugin/input/mod.rs b/rust/routee-compass/src/plugin/input/mod.rs index 2dcabdb5..bd438a45 100644 --- a/rust/routee-compass/src/plugin/input/mod.rs +++ b/rust/routee-compass/src/plugin/input/mod.rs @@ -2,4 +2,8 @@ pub mod default; pub mod input_field; pub mod input_json_extensions; pub mod input_plugin; +pub mod input_plugin_error; pub mod input_plugin_ops; + +pub use input_field::InputField; +pub use input_plugin_error::InputPluginError; diff --git a/rust/routee-compass/src/plugin/output/default/summary/plugin.rs b/rust/routee-compass/src/plugin/output/default/summary/plugin.rs index c6375bea..4a66479f 100644 --- a/rust/routee-compass/src/plugin/output/default/summary/plugin.rs +++ b/rust/routee-compass/src/plugin/output/default/summary/plugin.rs @@ -2,7 +2,7 @@ use crate::app::{ compass::compass_app_error::CompassAppError, search::search_app_result::SearchAppResult, }; use crate::plugin::output::output_plugin::OutputPlugin; -use crate::plugin::plugin_error::PluginError; +use crate::plugin::output::OutputPluginError; use routee_compass_core::algorithm::search::search_instance::SearchInstance; use routee_compass_core::util::duration_extension::DurationExtension; use serde_json::{self, json}; @@ -16,7 +16,7 @@ impl OutputPlugin for SummaryOutputPlugin { &self, output: &mut serde_json::Value, search_result: &Result<(SearchAppResult, SearchInstance), CompassAppError>, - ) -> Result<(), PluginError> { + ) -> Result<(), OutputPluginError> { match search_result { Err(_e) => Ok(()), Ok((result, _)) => { diff --git a/rust/routee-compass/src/plugin/output/default/traversal/builder.rs b/rust/routee-compass/src/plugin/output/default/traversal/builder.rs index 4911472c..a5036d1a 100644 --- a/rust/routee-compass/src/plugin/output/default/traversal/builder.rs +++ b/rust/routee-compass/src/plugin/output/default/traversal/builder.rs @@ -6,7 +6,7 @@ use crate::{ builders::OutputPluginBuilder, compass_configuration_error::CompassConfigurationError, config_json_extension::ConfigJsonExtensions, }, - plugin::output::output_plugin::OutputPlugin, + plugin::{output::output_plugin::OutputPlugin, plugin_error::PluginError}, }; /// Builds a plugin that can generate traversal outputs. @@ -49,7 +49,8 @@ impl OutputPluginBuilder for TraversalPluginBuilder { let tree: Option<TraversalOutputFormat> = parameters.get_config_serde_optional(&"tree", &parent_key)?; - let geom_plugin = TraversalPlugin::from_file(&geometry_filename, route, tree)?; + let geom_plugin = TraversalPlugin::from_file(&geometry_filename, route, tree) + .map_err(|e| PluginError::OutputPluginFailed { source: e })?; Ok(Arc::new(geom_plugin)) } } diff --git a/rust/routee-compass/src/plugin/output/default/traversal/json_extensions.rs b/rust/routee-compass/src/plugin/output/default/traversal/json_extensions.rs index bd0d24fd..d3069ba5 100644 --- a/rust/routee-compass/src/plugin/output/default/traversal/json_extensions.rs +++ b/rust/routee-compass/src/plugin/output/default/traversal/json_extensions.rs @@ -1,7 +1,6 @@ +use crate::plugin::{input::InputField, output::OutputPluginError}; use std::fmt::Display; -use crate::plugin::plugin_error::PluginError; - pub enum TraversalJsonField { RouteOutput, TreeOutput, @@ -23,18 +22,22 @@ impl Display for TraversalJsonField { } pub trait TraversalJsonExtensions { - fn get_route_geometry_wkt(&self) -> Result<String, PluginError>; + fn get_route_geometry_wkt(&self) -> Result<String, OutputPluginError>; } impl TraversalJsonExtensions for serde_json::Value { - fn get_route_geometry_wkt(&self) -> Result<String, PluginError> { + fn get_route_geometry_wkt(&self) -> Result<String, OutputPluginError> { let geometry = self .get(TraversalJsonField::RouteOutput.as_str()) - .ok_or_else(|| PluginError::MissingField(TraversalJsonField::RouteOutput.to_string()))? - .as_str() .ok_or_else(|| { - PluginError::ParseError( + OutputPluginError::MissingExpectedQueryField(InputField::Custom( TraversalJsonField::RouteOutput.to_string(), + )) + })? + .as_str() + .ok_or_else(|| { + OutputPluginError::QueryFieldHasInvalidType( + InputField::Custom(TraversalJsonField::RouteOutput.to_string()), String::from("string"), ) })? diff --git a/rust/routee-compass/src/plugin/output/default/traversal/plugin.rs b/rust/routee-compass/src/plugin/output/default/traversal/plugin.rs index ef44ebb3..45529bc1 100644 --- a/rust/routee-compass/src/plugin/output/default/traversal/plugin.rs +++ b/rust/routee-compass/src/plugin/output/default/traversal/plugin.rs @@ -3,7 +3,7 @@ use super::traversal_output_format::TraversalOutputFormat; use crate::app::compass::compass_app_error::CompassAppError; use crate::app::search::search_app_result::SearchAppResult; use crate::plugin::output::output_plugin::OutputPlugin; -use crate::plugin::plugin_error::PluginError; +use crate::plugin::output::OutputPluginError; use geo::LineString; use kdam::Bar; use kdam::BarExt; @@ -28,9 +28,13 @@ impl TraversalPlugin { filename: &P, route: Option<TraversalOutputFormat>, tree: Option<TraversalOutputFormat>, - ) -> Result<TraversalPlugin, PluginError> { + ) -> Result<TraversalPlugin, OutputPluginError> { let count = fs_utils::line_count(filename, fs_utils::is_gzip(filename)).map_err(|e| { - PluginError::FileReadError(filename.as_ref().to_path_buf(), e.to_string()) + OutputPluginError::BuildFailed(format!( + "failure reading line count for file {}: {}", + filename.as_ref().to_str().unwrap_or_default(), + e + )) })?; let mut pb = Bar::builder() @@ -38,16 +42,20 @@ impl TraversalPlugin { .animation("fillup") .desc("geometry file") .build() - .map_err(PluginError::InternalError)?; + .map_err(OutputPluginError::InternalError)?; let cb = Box::new(|| { let _ = pb.update(1); }); let geoms = read_raw_file(filename, geo_io_utils::parse_wkt_linestring, Some(cb)).map_err(|e| { - PluginError::FileReadError(filename.as_ref().to_path_buf(), e.to_string()) + OutputPluginError::BuildFailed(format!( + "failure reading geometry file {}: {}", + filename.as_ref().to_str().unwrap_or_default(), + e + )) })?; - println!(); + eprintln!(); let route_key = TraversalJsonField::RouteOutput.to_string(); let tree_key = TraversalJsonField::TreeOutput.to_string(); @@ -66,7 +74,7 @@ impl OutputPlugin for TraversalPlugin { &self, output: &mut serde_json::Value, search_result: &Result<(SearchAppResult, SearchInstance), CompassAppError>, - ) -> Result<(), PluginError> { + ) -> Result<(), OutputPluginError> { match search_result { Err(_) => Ok(()), Ok((result, si)) => { @@ -80,7 +88,7 @@ impl OutputPlugin for TraversalPlugin { construct_route_output(route, si, &route_args, &self.geoms) }) .collect::<Result<Vec<_>, _>>() - .map_err(PluginError::PluginFailed)?; + .map_err(OutputPluginError::OutputPluginFailed)?; // vary the type of value stored at the route key. if there is // no route, store 'null'. if one, store an output object. if diff --git a/rust/routee-compass/src/plugin/output/default/traversal/traversal_ops.rs b/rust/routee-compass/src/plugin/output/default/traversal/traversal_ops.rs index 8960dffd..9d904090 100644 --- a/rust/routee-compass/src/plugin/output/default/traversal/traversal_ops.rs +++ b/rust/routee-compass/src/plugin/output/default/traversal/traversal_ops.rs @@ -1,30 +1,35 @@ -use crate::plugin::plugin_error::PluginError; +use crate::plugin::output::OutputPluginError; use geo::{LineString, MultiLineString, Point}; use geo_types::MultiPoint; use geojson::feature::Id; use geojson::{Feature, FeatureCollection}; use routee_compass_core::algorithm::search::edge_traversal::EdgeTraversal; use routee_compass_core::algorithm::search::search_tree_branch::SearchTreeBranch; -use routee_compass_core::model::road_network::vertex_id::VertexId; +use routee_compass_core::model::network::vertex_id::VertexId; use routee_compass_core::util::geo::geo_io_utils; use std::collections::HashMap; pub fn create_tree_geojson( tree: &HashMap<VertexId, SearchTreeBranch>, geoms: &[LineString<f32>], -) -> Result<serde_json::Value, PluginError> { +) -> Result<serde_json::Value, OutputPluginError> { let features = tree .values() .map(|t| { let row_result = geoms .get(t.edge_traversal.edge_id.0) .cloned() - .ok_or_else(|| PluginError::EdgeGeometryMissing(t.edge_traversal.edge_id)) + .ok_or_else(|| { + OutputPluginError::OutputPluginFailed(format!( + "geometry table missing edge id {}", + t.edge_traversal.edge_id + )) + }) .and_then(|g| create_geojson_feature(&t.edge_traversal, g)); row_result }) - .collect::<Result<Vec<_>, PluginError>>()?; + .collect::<Result<Vec<_>, OutputPluginError>>()?; // let result_json = serde_json::to_value(features)?;/ let feature_collection = FeatureCollection { bbox: None, @@ -38,19 +43,24 @@ pub fn create_tree_geojson( pub fn create_route_geojson( route: &[EdgeTraversal], geoms: &[LineString<f32>], -) -> Result<serde_json::Value, PluginError> { +) -> Result<serde_json::Value, OutputPluginError> { let features = route .iter() .map(|t| { let row_result = geoms .get(t.edge_id.0) .cloned() - .ok_or_else(|| PluginError::EdgeGeometryMissing(t.edge_id)) + .ok_or_else(|| { + OutputPluginError::OutputPluginFailed(format!( + "geometry table missing edge id {}", + t.edge_id + )) + }) .and_then(|g| create_geojson_feature(t, g)); row_result }) - .collect::<Result<Vec<_>, PluginError>>()?; + .collect::<Result<Vec<_>, OutputPluginError>>()?; // let result_json = serde_json::to_value(features)?;/ let feature_collection = FeatureCollection { bbox: None, @@ -64,14 +74,14 @@ pub fn create_route_geojson( pub fn create_geojson_feature( t: &EdgeTraversal, g: LineString<f32>, -) -> Result<Feature, PluginError> { +) -> Result<Feature, OutputPluginError> { let props = match serde_json::to_value(t).map(|v| v.as_object().cloned()) { - Ok(None) => Err(PluginError::InternalError(format!( + Ok(None) => Err(OutputPluginError::InternalError(format!( "serialized EdgeTraversal was not a JSON object for {}", t ))), Ok(Some(obj)) => Ok(obj), - Err(err) => Err(PluginError::JsonError(err)), + Err(err) => Err(OutputPluginError::JsonError { source: err }), }?; let id = Id::Number(serde_json::Number::from(t.edge_id.0)); @@ -89,24 +99,26 @@ pub fn create_geojson_feature( pub fn create_edge_geometry( edge: &EdgeTraversal, geoms: &[LineString<f32>], -) -> Result<LineString<f32>, PluginError> { - geoms - .get(edge.edge_id.0) - .cloned() - .ok_or_else(|| PluginError::EdgeGeometryMissing(edge.edge_id)) +) -> Result<LineString<f32>, OutputPluginError> { + geoms.get(edge.edge_id.0).cloned().ok_or_else(|| { + OutputPluginError::OutputPluginFailed(format!( + "geometry table missing edge id {}", + edge.edge_id + )) + }) } pub fn create_branch_geometry( branch: &SearchTreeBranch, geoms: &[LineString<f32>], -) -> Result<LineString<f32>, PluginError> { +) -> Result<LineString<f32>, OutputPluginError> { create_edge_geometry(&branch.edge_traversal, geoms) } pub fn create_route_linestring( route: &[EdgeTraversal], geoms: &[LineString<f32>], -) -> Result<LineString<f32>, PluginError> { +) -> Result<LineString<f32>, OutputPluginError> { let edge_ids = route .iter() .map(|traversal| traversal.edge_id) @@ -115,12 +127,15 @@ pub fn create_route_linestring( let edge_linestrings = edge_ids .iter() .map(|eid| { - let geom = geoms - .get(eid.0) - .ok_or_else(|| PluginError::EdgeGeometryMissing(*eid)); + let geom = geoms.get(eid.0).ok_or_else(|| { + OutputPluginError::OutputPluginFailed(format!( + "geometry table missing edge id {}", + *eid + )) + }); geom }) - .collect::<Result<Vec<&LineString<f32>>, PluginError>>()?; + .collect::<Result<Vec<&LineString<f32>>, OutputPluginError>>()?; let geometry = geo_io_utils::concat_linestrings(edge_linestrings); Ok(geometry) } @@ -128,7 +143,7 @@ pub fn create_route_linestring( pub fn create_tree_multilinestring( tree: &HashMap<VertexId, SearchTreeBranch>, geoms: &[LineString<f32>], -) -> Result<MultiLineString<f32>, PluginError> { +) -> Result<MultiLineString<f32>, OutputPluginError> { let edge_ids = tree .values() .map(|traversal| traversal.edge_traversal.edge_id) @@ -137,12 +152,15 @@ pub fn create_tree_multilinestring( let tree_linestrings = edge_ids .iter() .map(|eid| { - let geom = geoms - .get(eid.0) - .ok_or_else(|| PluginError::EdgeGeometryMissing(*eid)); + let geom = geoms.get(eid.0).ok_or_else(|| { + OutputPluginError::OutputPluginFailed(format!( + "geometry table missing edge id {}", + *eid + )) + }); geom.cloned() }) - .collect::<Result<Vec<LineString<f32>>, PluginError>>()?; + .collect::<Result<Vec<LineString<f32>>, OutputPluginError>>()?; let geometry = MultiLineString::new(tree_linestrings); Ok(geometry) } @@ -150,7 +168,7 @@ pub fn create_tree_multilinestring( pub fn create_tree_multipoint( tree: &HashMap<VertexId, SearchTreeBranch>, geoms: &[LineString<f64>], -) -> Result<MultiPoint, PluginError> { +) -> Result<MultiPoint, OutputPluginError> { let edge_ids = tree .values() .map(|traversal| traversal.edge_traversal.edge_id) @@ -161,10 +179,15 @@ pub fn create_tree_multipoint( .map(|eid| { let geom = geoms .get(eid.0) - .ok_or_else(|| PluginError::EdgeGeometryMissing(*eid)) + .ok_or_else(|| { + OutputPluginError::OutputPluginFailed(format!( + "geometry table missing edge id {}", + *eid + )) + }) .map(|l| { l.points().last().ok_or_else(|| { - PluginError::InputError(format!( + OutputPluginError::OutputPluginFailed(format!( "linestring is invalid for edge_id {}", eid )) @@ -177,7 +200,7 @@ pub fn create_tree_multipoint( Err(e) => Err(e), } }) - .collect::<Result<Vec<Point>, PluginError>>()?; + .collect::<Result<Vec<Point>, OutputPluginError>>()?; let geometry = MultiPoint::new(tree_destinations); Ok(geometry) } diff --git a/rust/routee-compass/src/plugin/output/default/traversal/traversal_output_format.rs b/rust/routee-compass/src/plugin/output/default/traversal/traversal_output_format.rs index bed69658..3be5b340 100644 --- a/rust/routee-compass/src/plugin/output/default/traversal/traversal_output_format.rs +++ b/rust/routee-compass/src/plugin/output/default/traversal/traversal_output_format.rs @@ -1,11 +1,11 @@ use std::collections::HashMap; use super::traversal_ops as ops; -use crate::plugin::plugin_error::PluginError; +use crate::plugin::output::OutputPluginError; use geo::{CoordFloat, Geometry, LineString}; use routee_compass_core::{ algorithm::search::{edge_traversal::EdgeTraversal, search_tree_branch::SearchTreeBranch}, - model::road_network::vertex_id::VertexId, + model::network::vertex_id::VertexId, }; use serde::{Deserialize, Serialize}; use wkt::ToWkt; @@ -30,7 +30,7 @@ impl TraversalOutputFormat { &self, route: &Vec<EdgeTraversal>, geoms: &[LineString<f32>], - ) -> Result<serde_json::Value, PluginError> { + ) -> Result<serde_json::Value, OutputPluginError> { match self { TraversalOutputFormat::Wkt => { let route_geometry = ops::create_route_linestring(route, geoms)?; @@ -64,7 +64,7 @@ impl TraversalOutputFormat { &self, tree: &HashMap<VertexId, SearchTreeBranch>, geoms: &[LineString<f32>], - ) -> Result<serde_json::Value, PluginError> { + ) -> Result<serde_json::Value, OutputPluginError> { match self { TraversalOutputFormat::Wkt => { let route_geometry = ops::create_tree_multilinestring(tree, geoms)?; @@ -99,9 +99,9 @@ impl TraversalOutputFormat { fn geometry_to_wkb_string<T: CoordFloat + Into<f64>>( geometry: &Geometry<T>, -) -> Result<String, PluginError> { +) -> Result<String, OutputPluginError> { let bytes = wkb::geom_to_wkb(geometry).map_err(|e| { - PluginError::PluginFailed(format!( + OutputPluginError::OutputPluginFailed(format!( "failed to generate wkb for geometry '{:?}' - {:?}", geometry, e )) @@ -122,9 +122,7 @@ mod test { use geo::{coord, LineString}; use routee_compass_core::{ algorithm::search::edge_traversal::EdgeTraversal, - model::{ - road_network::edge_id::EdgeId, traversal::state::state_variable::StateVar, unit::Cost, - }, + model::{network::edge_id::EdgeId, traversal::state::state_variable::StateVar, unit::Cost}, }; use std::time::Duration; diff --git a/rust/routee-compass/src/plugin/output/default/uuid/builder.rs b/rust/routee-compass/src/plugin/output/default/uuid/builder.rs index c20515e2..820d5c5c 100644 --- a/rust/routee-compass/src/plugin/output/default/uuid/builder.rs +++ b/rust/routee-compass/src/plugin/output/default/uuid/builder.rs @@ -5,7 +5,7 @@ use crate::{ builders::OutputPluginBuilder, compass_configuration_error::CompassConfigurationError, config_json_extension::ConfigJsonExtensions, }, - plugin::output::output_plugin::OutputPlugin, + plugin::{output::output_plugin::OutputPlugin, plugin_error::PluginError}, }; use super::plugin::UUIDOutputPlugin; @@ -19,8 +19,10 @@ impl OutputPluginBuilder for UUIDOutputPluginBuilder { ) -> Result<Arc<dyn OutputPlugin>, CompassConfigurationError> { let uuid_filename = parameters.get_config_path(&"uuid_input_file", &"uuid")?; - let uuid_plugin = UUIDOutputPlugin::from_file(&uuid_filename) - .map_err(CompassConfigurationError::PluginError)?; + let uuid_plugin = UUIDOutputPlugin::from_file(&uuid_filename).map_err(|e| { + let pe = PluginError::OutputPluginFailed { source: e }; + CompassConfigurationError::PluginError(pe) + })?; Ok(Arc::new(uuid_plugin)) } } diff --git a/rust/routee-compass/src/plugin/output/default/uuid/mod.rs b/rust/routee-compass/src/plugin/output/default/uuid/mod.rs index 668d9c48..fd557c7b 100644 --- a/rust/routee-compass/src/plugin/output/default/uuid/mod.rs +++ b/rust/routee-compass/src/plugin/output/default/uuid/mod.rs @@ -1,3 +1,3 @@ pub mod builder; -pub mod json_extensions; +pub mod output_json_extensions; pub mod plugin; diff --git a/rust/routee-compass/src/plugin/output/default/uuid/json_extensions.rs b/rust/routee-compass/src/plugin/output/default/uuid/output_json_extensions.rs similarity index 64% rename from rust/routee-compass/src/plugin/output/default/uuid/json_extensions.rs rename to rust/routee-compass/src/plugin/output/default/uuid/output_json_extensions.rs index 7943e8bc..9a671d12 100644 --- a/rust/routee-compass/src/plugin/output/default/uuid/json_extensions.rs +++ b/rust/routee-compass/src/plugin/output/default/uuid/output_json_extensions.rs @@ -1,8 +1,8 @@ use std::fmt::Display; -use routee_compass_core::model::road_network::vertex_id::VertexId; +use routee_compass_core::model::network::vertex_id::VertexId; -use crate::plugin::plugin_error::PluginError; +use crate::plugin::{input::InputField, output::OutputPluginError}; pub enum UUIDJsonField { Request, @@ -12,6 +12,12 @@ pub enum UUIDJsonField { DestinationVertexUUID, } +impl From<UUIDJsonField> for InputField { + fn from(value: UUIDJsonField) -> Self { + InputField::Custom(value.to_string()) + } +} + impl UUIDJsonField { pub fn as_str(&self) -> &'static str { match self { @@ -31,46 +37,52 @@ impl Display for UUIDJsonField { } pub trait UUIDJsonExtensions { - fn get_od_vertex_ids(&self) -> Result<(VertexId, VertexId), PluginError>; + fn get_od_vertex_ids(&self) -> Result<(VertexId, VertexId), OutputPluginError>; fn add_od_uuids( &mut self, origin_uuid: String, destination_uuid: String, - ) -> Result<(), PluginError>; + ) -> Result<(), OutputPluginError>; } impl UUIDJsonExtensions for serde_json::Value { - fn get_od_vertex_ids(&self) -> Result<(VertexId, VertexId), PluginError> { + fn get_od_vertex_ids(&self) -> Result<(VertexId, VertexId), OutputPluginError> { let request = self .get(UUIDJsonField::Request.as_str()) - .ok_or_else(|| PluginError::MissingField(UUIDJsonField::Request.to_string()))? + .ok_or_else(|| { + OutputPluginError::MissingExpectedQueryField(UUIDJsonField::Request.into()) + })? .as_object() .ok_or_else(|| { - PluginError::ParseError( - UUIDJsonField::Request.to_string(), + OutputPluginError::QueryFieldHasInvalidType( + UUIDJsonField::Request.into(), String::from("json object"), ) })?; let origin_vertex_id = request .get(&UUIDJsonField::OriginVertexId.to_string()) - .ok_or_else(|| PluginError::MissingField(UUIDJsonField::OriginVertexId.to_string()))? + .ok_or_else(|| { + OutputPluginError::MissingExpectedQueryField(UUIDJsonField::OriginVertexId.into()) + })? .as_u64() .ok_or_else(|| { - PluginError::ParseError( - UUIDJsonField::OriginVertexId.to_string(), + OutputPluginError::QueryFieldHasInvalidType( + UUIDJsonField::OriginVertexId.into(), String::from("u64"), ) })?; let destination_vertex_id = request .get(&UUIDJsonField::DestinationVertexId.to_string()) .ok_or_else(|| { - PluginError::MissingField(UUIDJsonField::DestinationVertexId.to_string()) + OutputPluginError::MissingExpectedQueryField( + UUIDJsonField::DestinationVertexId.into(), + ) })? .as_u64() .ok_or_else(|| { - PluginError::ParseError( - UUIDJsonField::DestinationVertexId.to_string(), + OutputPluginError::QueryFieldHasInvalidType( + UUIDJsonField::DestinationVertexId.into(), String::from("u64"), ) })?; @@ -83,14 +95,16 @@ impl UUIDJsonExtensions for serde_json::Value { &mut self, origin_uuid: String, destination_uuid: String, - ) -> Result<(), PluginError> { + ) -> Result<(), OutputPluginError> { let request = self .get_mut(UUIDJsonField::Request.as_str()) - .ok_or_else(|| PluginError::MissingField(UUIDJsonField::Request.to_string()))? + .ok_or_else(|| { + OutputPluginError::MissingExpectedQueryField(UUIDJsonField::Request.into()) + })? .as_object_mut() .ok_or_else(|| { - PluginError::ParseError( - UUIDJsonField::Request.to_string(), + OutputPluginError::QueryFieldHasInvalidType( + UUIDJsonField::Request.into(), String::from("json object"), ) })?; diff --git a/rust/routee-compass/src/plugin/output/default/uuid/plugin.rs b/rust/routee-compass/src/plugin/output/default/uuid/plugin.rs index 05623e94..be0ba1c6 100644 --- a/rust/routee-compass/src/plugin/output/default/uuid/plugin.rs +++ b/rust/routee-compass/src/plugin/output/default/uuid/plugin.rs @@ -1,8 +1,8 @@ -use super::json_extensions::UUIDJsonExtensions; +use super::output_json_extensions::UUIDJsonExtensions; use crate::app::compass::compass_app_error::CompassAppError; use crate::app::search::search_app_result::SearchAppResult; -use crate::plugin::output::default::uuid::json_extensions::UUIDJsonField; -use crate::plugin::{output::output_plugin::OutputPlugin, plugin_error::PluginError}; +use crate::plugin::output::default::uuid::output_json_extensions::UUIDJsonField; +use crate::plugin::output::{OutputPlugin, OutputPluginError}; use kdam::Bar; use kdam::BarExt; @@ -17,9 +17,13 @@ pub struct UUIDOutputPlugin { } impl UUIDOutputPlugin { - pub fn from_file<P: AsRef<Path>>(filename: &P) -> Result<UUIDOutputPlugin, PluginError> { + pub fn from_file<P: AsRef<Path>>(filename: &P) -> Result<UUIDOutputPlugin, OutputPluginError> { let count = fs_utils::line_count(filename, fs_utils::is_gzip(filename)).map_err(|e| { - PluginError::FileReadError(filename.as_ref().to_path_buf(), e.to_string()) + OutputPluginError::BuildFailed(format!( + "failure reading UUID file {}: {}", + filename.as_ref().to_str().unwrap_or_default(), + e + )) })?; let mut pb = Bar::builder() @@ -27,16 +31,20 @@ impl UUIDOutputPlugin { .animation("fillup") .desc("uuid file") .build() - .map_err(PluginError::InternalError)?; + .map_err(OutputPluginError::InternalError)?; let cb = Box::new(|| { let _ = pb.update(1); }); let uuids = read_raw_file(filename, |_idx, row| Ok(row), Some(cb)).map_err(|e| { - PluginError::FileReadError(filename.as_ref().to_path_buf(), e.to_string()) + OutputPluginError::BuildFailed(format!( + "failure reading UUID file {}: {}", + filename.as_ref().to_str().unwrap_or_default(), + e + )) })?; - println!(); + eprintln!(); let o_key = UUIDJsonField::OriginVertexUUID.to_string(); let d_key = UUIDJsonField::DestinationVertexUUID.to_string(); @@ -53,21 +61,27 @@ impl OutputPlugin for UUIDOutputPlugin { &self, output: &mut serde_json::Value, search_result: &Result<(SearchAppResult, SearchInstance), CompassAppError>, - ) -> Result<(), PluginError> { + ) -> Result<(), OutputPluginError> { match search_result { Err(_) => Ok(()), Ok(_) => { let (origin_vertex_id, destination_vertex_id) = output.get_od_vertex_ids()?; - let origin_uuid = self - .uuids - .get(origin_vertex_id.0) - .cloned() - .ok_or_else(|| PluginError::UUIDMissing(origin_vertex_id.0))?; + let origin_uuid = self.uuids.get(origin_vertex_id.0).cloned().ok_or_else(|| { + OutputPluginError::OutputPluginFailed(format!( + "UUID lookup table missing vertex index {}", + origin_vertex_id.0 + )) + })?; let destination_uuid = self .uuids .get(destination_vertex_id.0) .cloned() - .ok_or_else(|| PluginError::UUIDMissing(destination_vertex_id.0))?; + .ok_or_else(|| { + OutputPluginError::OutputPluginFailed(format!( + "UUID lookup table missing vertex index {}", + destination_vertex_id.0 + )) + })?; output[&self.o_key] = serde_json::Value::String(origin_uuid); output[&self.d_key] = serde_json::Value::String(destination_uuid); diff --git a/rust/routee-compass/src/plugin/output/mod.rs b/rust/routee-compass/src/plugin/output/mod.rs index acc6f390..28f0084a 100644 --- a/rust/routee-compass/src/plugin/output/mod.rs +++ b/rust/routee-compass/src/plugin/output/mod.rs @@ -1,3 +1,7 @@ pub mod default; pub mod output_plugin; +pub mod output_plugin_error; pub mod output_plugin_ops; + +pub use output_plugin::OutputPlugin; +pub use output_plugin_error::OutputPluginError; diff --git a/rust/routee-compass/src/plugin/output/output_plugin.rs b/rust/routee-compass/src/plugin/output/output_plugin.rs index de116ba5..263ed8b8 100644 --- a/rust/routee-compass/src/plugin/output/output_plugin.rs +++ b/rust/routee-compass/src/plugin/output/output_plugin.rs @@ -1,8 +1,7 @@ -use routee_compass_core::algorithm::search::search_instance::SearchInstance; - +use super::output_plugin_error::OutputPluginError; use crate::app::compass::compass_app_error::CompassAppError; use crate::app::search::search_app_result::SearchAppResult; -use crate::plugin::plugin_error::PluginError; +use routee_compass_core::algorithm::search::search_instance::SearchInstance; /// Performs some kind of post-processing on a search result. The result JSON is available /// to the plugin as a reference which was potentially modified upstream by another output @@ -40,5 +39,5 @@ pub trait OutputPlugin: Send + Sync { &self, output: &mut serde_json::Value, result: &Result<(SearchAppResult, SearchInstance), CompassAppError>, - ) -> Result<(), PluginError>; + ) -> Result<(), OutputPluginError>; } diff --git a/rust/routee-compass/src/plugin/output/output_plugin_error.rs b/rust/routee-compass/src/plugin/output/output_plugin_error.rs new file mode 100644 index 00000000..cb41a5e5 --- /dev/null +++ b/rust/routee-compass/src/plugin/output/output_plugin_error.rs @@ -0,0 +1,24 @@ +use crate::plugin::input::InputField; + +#[derive(thiserror::Error, Debug)] +pub enum OutputPluginError { + #[error("failure building input plugin: {0}")] + BuildFailed(String), + #[error("required query field '{0}' not found")] + MissingExpectedQueryField(InputField), + #[error("{0} provided without {1}")] + MissingQueryFieldPair(InputField, InputField), + #[error("required query field '{0}' is not of type {1}")] + QueryFieldHasInvalidType(InputField, String), + #[error("expected query to be a json object '{{}}' but found {0}")] + UnexpectedQueryStructure(String), + #[error("plugin experienced JSON error: {source}")] + JsonError { + #[from] + source: serde_json::Error, + }, + #[error("failure running plugin: {0}")] + OutputPluginFailed(String), + #[error("unexpected error: {0}")] + InternalError(String), +} diff --git a/rust/routee-compass/src/plugin/plugin_error.rs b/rust/routee-compass/src/plugin/plugin_error.rs index 786ef4b7..b7f7a588 100644 --- a/rust/routee-compass/src/plugin/plugin_error.rs +++ b/rust/routee-compass/src/plugin/plugin_error.rs @@ -1,37 +1,29 @@ -use std::path::PathBuf; - -use routee_compass_core::{ - algorithm::search::search_error::SearchError, model::road_network::edge_id::EdgeId, -}; +use super::input::InputPluginError; +use super::output::OutputPluginError; #[derive(thiserror::Error, Debug)] pub enum PluginError { - #[error("failed to parse {0} as {1}")] - ParseError(String, String), - #[error("missing field {0}")] - MissingField(String), - #[error("error with parsing query inputs:\n{0}")] - InputError(String), - #[error("error with building plugin")] - BuildError, - #[error("{0}")] - PluginFailed(String), - #[error("unable to read file {0} due to {1}")] - FileReadError(PathBuf, String), - #[error(transparent)] - JsonError(#[from] serde_json::Error), - #[error(transparent)] - GeoJsonError(#[from] geojson::Error), - #[error(transparent)] - CsvReadError(#[from] csv::Error), - #[error("geometry missing for edge id {0}")] - EdgeGeometryMissing(EdgeId), - #[error("uuid missing for edge id {0}")] - UUIDMissing(usize), - #[error(transparent)] - SearchError(#[from] SearchError), + #[error("failure building plugin: {0}")] + BuildFailed(String), + #[error("required query field '{0}' for plugin {1} not found")] + MissingExpectedQueryField(String, String), + #[error("failure running input plugin: {source}")] + InputPluginFailed { + #[from] + source: InputPluginError, + }, + #[error("failure running output plugin: {source}")] + OutputPluginFailed { + #[from] + source: OutputPluginError, + }, + #[error("plugin experienced JSON error: {source}")] + JsonError { + #[from] + source: serde_json::Error, + }, #[error("expected query to be a json object '{{}}' but found {0}")] UnexpectedQueryStructure(String), - #[error("unexpected error {0}")] + #[error("unexpected error: {0}")] InternalError(String), }

simplify error variants when we started _this journey_ we picked specified taxonomies for traversal model errors, frontier model errors, graph errors and the like, such as `TraversalModelError::MissingIdInTabularCostFunction`. but as we solidify RouteE Compass as a dependency of other downstream projects, those variants are not general enough. what is instead missing is a standardized taxonomy of error types. for example, the following variants (incomplete set): ```rust pub enum TraversalModelError { #[error("failure building traversal model from file {0}: {1}")] FileReadError(PathBuf, String), #[error("failure building traversal model: {0}")] BuildError(String), #[error("numeric error during calculation: {0}")] NumericError(String), #[error("index {0} for {1} not found on search state {2:?}")] StateVectorIndexOutOfBounds(usize, String, TraversalState), #[error("id {0} for id type {1} not found in tabular edge cost function {2}")] MissingIdInTabularCostFunction(String, String, String), #[error("prediction model failed with error {0}")] PredictionModel(String), } ``` could be reduced to something like ```rust pub enum TraversalModelError { #[error("failure during build for model {1} due to {0}")] BuildErrorForModel(String, String), #[error("failure during build for service {1} due to {0}")] BuildErrorForService(String, String), #[error("failure during build for builder {1} due to {0}")] BuildErrorForBuilder(String, String), #[error("runtime error for model {1} due to {0}")] RuntimeErrorForModel(String, String) } ``` no transparent errors should bubble to top of app attempting to run a compass app and i was getting this error: ``` [2024-04-08T22:49:11Z ERROR mep_routee_compass] invalid type: map, expected variant identifier ``` this was due to the fact that my configuration TOML's output policy had double braces instead of single: ```toml [[response_output_policy]] # should be [...] type = "file" ``` the error is vague because CompassAppError has a transparent ConfigError (from the config crate): ```rust #[error(transparent)] ConfigError(#[from] ConfigError), ``` Compass' top-level error types should not be transparent unless they do not require context. this should instead return a BuildError or similar where the developer includes the contextual information: ```rust #[error("error building component {0} of type {1}: {2}")] BuildError(String, String, String), ```

2024-12-02T17:21:28

0.0

[]

NREL/routee-compass

NREL__routee-compass-168

9fa285c2ac27a9efcacba5658e7e9040c38ffb5d

diff --git a/docs/notebooks/open_street_maps_example.ipynb b/docs/notebooks/open_street_maps_example.ipynb index e73c5f8f..84831098 100644 --- a/docs/notebooks/open_street_maps_example.ipynb +++ b/docs/notebooks/open_street_maps_example.ipynb @@ -86,7 +86,21 @@ "output_type": "stream", "text": [ "processing graph topology and speeds\n", - "adding grade information\n", + "adding grade information\n" + ] + }, + { + "name": "stderr", + "output_type": "stream", + "text": [ + "/Users/rfitzger/anaconda3/envs/routee-rust/lib/python3.9/site-packages/osgeo/gdal.py:287: FutureWarning: Neither gdal.UseExceptions() nor gdal.DontUseExceptions() has been explicitly called. In GDAL 4.0, exceptions will be enabled by default.\n", + " warnings.warn(\n" + ] + }, + { + "name": "stdout", + "output_type": "stream", + "text": [ "processing vertices\n", "processing edges\n", "writing vertex files\n", @@ -119,7 +133,7 @@ }, { "cell_type": "code", - "execution_count": 4, + "execution_count": 5, "id": "edad6c15-c641-4010-b097-7d07aa86c1e7", "metadata": {}, "outputs": [ @@ -127,6 +141,9 @@ "name": "stdout", "output_type": "stream", "text": [ + "\n", + "\n", + "\n", "\n" ] }, @@ -134,16 +151,7 @@ "name": "stderr", "output_type": "stream", "text": [ - "uuid file: 100%|██████████| 16992/16992 [00:00<00:00, 6624988.00it/s]/s]" - ] - }, - { - "name": "stdout", - "output_type": "stream", - "text": [ - "\n", - "\n", - "\n" + "uuid file: 100%|██████████| 16991/16991 [00:00<00:00, 4182921.75it/s]/s]" ] } ], @@ -177,7 +185,7 @@ }, { "cell_type": "code", - "execution_count": 193, + "execution_count": 6, "id": "f01086d0-231f-4013-bc24-97f9c56670bd", "metadata": {}, "outputs": [], @@ -189,7 +197,7 @@ " \"destination_x\": -104.975360,\n", " \"destination_y\": 39.693005,\n", " \"model_name\": \"2016_TOYOTA_Camry_4cyl_2WD\",\n", - " \"vehicle_state_variable_rates\": {\n", + " \"vehicle_rates\": {\n", " \"distance\": {\"type\": \"factor\", \"factor\": 0.655},\n", " \"time\": {\"type\": \"factor\", \"factor\": 0.33},\n", " \"energy_liquid\": {\"type\": \"factor\", \"factor\": 3.0}, \n", @@ -198,7 +206,7 @@ " \"test_cases\": [\n", " {\n", " \"name\": \"least_time\",\n", - " \"state_variable_coefficients\": {\n", + " \"weights\": {\n", " \"distance\": 0,\n", " \"time\": 1,\n", " \"energy_liquid\": 0\n", @@ -206,7 +214,7 @@ " },\n", " {\n", " \"name\": \"least_energy\",\n", - " \"state_variable_coefficients\": {\n", + " \"weights\": {\n", " \"distance\": 0,\n", " \"time\": 0,\n", " \"energy_liquid\": 1\n", @@ -214,7 +222,7 @@ " },\n", " {\n", " \"name\": \"least_cost\",\n", - " \"state_variable_coefficients\": {\n", + " \"weights\": {\n", " \"distance\": 1,\n", " \"time\": 1,\n", " \"energy_liquid\": 1\n", @@ -241,7 +249,7 @@ }, { "cell_type": "code", - "execution_count": 176, + "execution_count": 7, "id": "c6e9361b-5ffd-4ec5-a09a-d51e46752ebf", "metadata": {}, "outputs": [ @@ -256,7 +264,7 @@ "name": "stderr", "output_type": "stream", "text": [ - "search: 100%|██████████| 3/3 [00:00<00:00, 58.36it/s]7.79it/s]" + "search: 100%|██████████| 3/3 [00:00<00:00, 35.65it/s]2.69it/s]" ] } ], @@ -277,7 +285,7 @@ }, { "cell_type": "code", - "execution_count": 177, + "execution_count": 7, "id": "2c51774c-acff-4569-a820-4be5ba5aa51b", "metadata": {}, "outputs": [], @@ -302,7 +310,7 @@ }, { "cell_type": "code", - "execution_count": 178, + "execution_count": 8, "id": "1825ee52", "metadata": {}, "outputs": [], @@ -325,7 +333,7 @@ }, { "cell_type": "code", - "execution_count": 179, + "execution_count": 9, "id": "961c910c-63ec-4c79-9522-19f24f151215", "metadata": {}, "outputs": [ @@ -336,8 +344,8 @@ "{\n", " \"distance\": 9.152878807061885,\n", " \"energy_electric\": 0.0,\n", - " \"energy_liquid\": 0.3200989780246874,\n", - " \"time\": 10.605213626907442\n", + " \"energy_liquid\": 0.32011129944536876,\n", + " \"time\": 10.905464892278664\n", "}\n" ] } @@ -356,7 +364,7 @@ }, { "cell_type": "code", - "execution_count": 180, + "execution_count": 10, "id": "804b8d54-43d8-4c57-889d-e23f947892ee", "metadata": {}, "outputs": [ @@ -401,7 +409,7 @@ }, { "cell_type": "code", - "execution_count": 181, + "execution_count": 11, "id": "d4501094", "metadata": {}, "outputs": [ @@ -412,9 +420,9 @@ "{\n", " \"distance\": 5.995135618625535,\n", " \"energy_electric\": 0.0,\n", - " \"energy_liquid\": 0.9602969340740621,\n", - " \"time\": 3.499720496879456,\n", - " \"total_cost\": 10.455153049579053\n", + " \"energy_liquid\": 0.9603338983361063,\n", + " \"time\": 3.598803414451959,\n", + " \"total_cost\": 10.5542729314136\n", "}\n" ] } @@ -437,7 +445,7 @@ }, { "cell_type": "code", - "execution_count": 182, + "execution_count": 12, "id": "788183c8", "metadata": {}, "outputs": [ @@ -508,7 +516,7 @@ }, { "cell_type": "code", - "execution_count": 183, + "execution_count": 13, "id": "e2c6cb39-49aa-4697-9785-fa839dc16a86", "metadata": {}, "outputs": [ @@ -519,15 +527,15 @@ "{\n", " \"distance\": 6.733433188572938,\n", " \"energy_electric\": 0.0,\n", - " \"energy_liquid\": 0.26646891776649895,\n", - " \"time\": 12.657830096916395\n", + " \"energy_liquid\": 0.2664853255906838,\n", + " \"time\": 12.777312852016575\n", "}\n", "{\n", " \"distance\": 4.4103987385152745,\n", " \"energy_electric\": 0.0,\n", - " \"energy_liquid\": 0.7994067532994968,\n", - " \"time\": 4.177083931982411,\n", - " \"total_cost\": 9.386889423797182\n", + " \"energy_liquid\": 0.7994559767720513,\n", + " \"time\": 4.21651324116547,\n", + " \"total_cost\": 9.426367956452795\n", "}\n" ] } @@ -547,7 +555,7 @@ }, { "cell_type": "code", - "execution_count": 184, + "execution_count": 14, "id": "6283c273", "metadata": {}, "outputs": [ @@ -556,9 +564,9 @@ "output_type": "stream", "text": [ " - distance: 2.42 miles further with time-optimal\n", - " - time: 2.05 minutes longer with energy-optimal\n", + " - time: 1.87 minutes longer with energy-optimal\n", " - energy: 0.05 gallons_gasoline more with time-optimal\n", - " - cost: $1.07 more with time-optimal\n" + " - cost: $1.13 more with time-optimal\n" ] } ], @@ -587,7 +595,7 @@ }, { "cell_type": "code", - "execution_count": 185, + "execution_count": 15, "id": "791e814f-333c-43f4-ae84-36b9be517548", "metadata": {}, "outputs": [ @@ -714,7 +722,7 @@ }, { "cell_type": "code", - "execution_count": 186, + "execution_count": 16, "id": "30257102-07ee-4338-9c8c-41676e1995ed", "metadata": {}, "outputs": [], @@ -732,7 +740,7 @@ }, { "cell_type": "code", - "execution_count": 187, + "execution_count": 17, "id": "1ac13aae-32fc-4c10-b888-6373b175655a", "metadata": {}, "outputs": [ @@ -766,7 +774,7 @@ " <meta name="viewport" content="width=device-width,\n", " initial-scale=1.0, maximum-scale=1.0, user-scalable=no" />\n", " <style>\n", - " #map_a5cb9760638531fd5b14e75a46b3b1f1 {\n", + " #map_b6cf708d54aa4599b756d3ce62516c9c {\n", " position: relative;\n", " width: 100.0%;\n", " height: 100.0%;\n", @@ -780,14 +788,14 @@ "<body>\n", " \n", " \n", - " <div class="folium-map" id="map_a5cb9760638531fd5b14e75a46b3b1f1" ></div>\n", + " <div class="folium-map" id="map_b6cf708d54aa4599b756d3ce62516c9c" ></div>\n", " \n", "</body>\n", "<script>\n", " \n", " \n", - " var map_a5cb9760638531fd5b14e75a46b3b1f1 = L.map(\n", - " "map_a5cb9760638531fd5b14e75a46b3b1f1",\n", + " var map_b6cf708d54aa4599b756d3ce62516c9c = L.map(\n", + " "map_b6cf708d54aa4599b756d3ce62516c9c",\n", " {\n", " center: [39.73823928833008, -105.01406860351562],\n", " crs: L.CRS.EPSG3857,\n", @@ -801,19 +809,19 @@ "\n", " \n", " \n", - " var tile_layer_9b8abbe2aac9b29cd6e7073b6627118d = L.tileLayer(\n", + " var tile_layer_91198d8453ecb8ddf13da3c43cd11146 = L.tileLayer(\n", " "https://{s}.tile.openstreetmap.org/{z}/{x}/{y}.png",\n", " {"attribution": "Data by \\u0026copy; \\u003ca target=\\"_blank\\" href=\\"http://openstreetmap.org\\"\\u003eOpenStreetMap\\u003c/a\\u003e, under \\u003ca target=\\"_blank\\" href=\\"http://www.openstreetmap.org/copyright\\"\\u003eODbL\\u003c/a\\u003e.", "detectRetina": false, "maxNativeZoom": 18, "maxZoom": 18, "minZoom": 0, "noWrap": false, "opacity": 1, "subdomains": "abc", "tms": false}\n", - " ).addTo(map_a5cb9760638531fd5b14e75a46b3b1f1);\n", + " ).addTo(map_b6cf708d54aa4599b756d3ce62516c9c);\n", " \n", " \n", - " var poly_line_b64ee0679bc07d7a55e8ab1075c0227e = L.polyline(\n", + " var poly_line_0633987bf2e82ded548a26dc464c4978 = L.polyline(\n", " [[39.77902603149414, -104.96916961669922], [39.77896499633789, -104.96910095214844], [39.77878189086914, -104.96887969970703], [39.778751373291016, -104.96884155273438], [39.778587341308594, -104.9686508178711], [39.778480529785156, -104.96852111816406], [39.77839660644531, -104.96841430664062], [39.77846908569336, -104.96832275390625], [39.77865982055664, -104.96807098388672], [39.778709411621094, -104.96791076660156], [39.77894592285156, -104.96761322021484], [39.77906799316406, -104.96745300292969], [39.77915954589844, -104.96732330322266], [39.7791862487793, -104.96728515625], [39.779232025146484, -104.96722412109375], [39.77926254272461, -104.96719360351562], [39.779293060302734, -104.9671630859375], [39.77932357788086, -104.96714782714844], [39.77938461303711, -104.96710968017578], [39.779441833496094, -104.96707916259766], [39.77951431274414, -104.96704864501953], [39.7796745300293, -104.96698760986328], [39.77983093261719, -104.9669418334961], [39.78031921386719, -104.96683502197266], [39.780364990234375, -104.9668197631836], [39.78050231933594, -104.966796875], [39.78049850463867, -104.96688079833984], [39.78049850463867, -104.96698760986328], [39.78049850463867, -104.96705627441406], [39.78049087524414, -104.96717834472656], [39.78046798706055, -104.96759033203125], [39.78045654296875, -104.96772003173828], [39.78042984008789, -104.96797180175781], [39.78041458129883, -104.96810913085938], [39.780372619628906, -104.96839904785156], [39.78034210205078, -104.96859741210938], [39.780296325683594, -104.96888732910156], [39.7802734375, -104.96903228759766], [39.78025436401367, -104.96917724609375], [39.780250549316406, -104.96920776367188], [39.780235290527344, -104.96934509277344], [39.78020095825195, -104.96965789794922], [39.78001403808594, -104.9715347290039], [39.77994155883789, -104.97315216064453], [39.77980422973633, -104.97496032714844], [39.7797737121582, -104.97691345214844], [39.7797737121582, -104.9772720336914], [39.7797737121582, -104.97753143310547], [39.7797737121582, -104.97764587402344], [39.779781341552734, -104.97801971435547], [39.77980041503906, -104.97874450683594], [39.77981948852539, -104.97929382324219], [39.77983093261719, -104.97946166992188], [39.77986526489258, -104.98001861572266], [39.77991485595703, -104.98080444335938], [39.779991149902344, -104.98214721679688], [39.78001403808594, -104.98335266113281], [39.7800178527832, -104.98450469970703], [39.78002166748047, -104.98558044433594], [39.780033111572266, -104.98602294921875], [39.7800407409668, -104.98613739013672], [39.780059814453125, -104.98633575439453], [39.780086517333984, -104.9865493774414], [39.780128479003906, -104.98677825927734], [39.780174255371094, -104.98695373535156], [39.78023910522461, -104.98715209960938], [39.780540466308594, -104.98787689208984], [39.78069305419922, -104.98828887939453], [39.78076934814453, -104.98860168457031], [39.780784606933594, -104.98871612548828], [39.780792236328125, -104.98896026611328], [39.78078079223633, -104.98912048339844], [39.780765533447266, -104.98928833007812], [39.78071975708008, -104.98950958251953], [39.78061294555664, -104.98975372314453], [39.78052520751953, -104.98993682861328], [39.78038787841797, -104.99011993408203], [39.78028869628906, -104.99024963378906], [39.780174255371094, -104.99034881591797], [39.77997589111328, -104.990478515625], [39.77995300292969, -104.99049377441406], [39.77978515625, -104.99056243896484], [39.77960205078125, -104.99060821533203], [39.7794189453125, -104.99060821533203], [39.77922439575195, -104.9905776977539], [39.778968811035156, -104.99048614501953], [39.77858352661133, -104.9903335571289], [39.778541564941406, -104.99031066894531], [39.77811050415039, -104.99011993408203], [39.777706146240234, -104.98995208740234], [39.7775764465332, -104.98990631103516], [39.77741241455078, -104.98985290527344], [39.77730941772461, -104.98983001708984], [39.7772102355957, -104.98980712890625], [39.777130126953125, -104.98979949951172], [39.77692794799805, -104.98979187011719], [39.77660369873047, -104.98979187011719], [39.77629089355469, -104.98980712890625], [39.77605056762695, -104.98982238769531], [39.77583694458008, -104.98986053466797], [39.77559280395508, -104.98991394042969], [39.7751579284668, -104.99003601074219], [39.774871826171875, -104.99012756347656], [39.77482986450195, -104.9901351928711], [39.774696350097656, -104.99018096923828], [39.774539947509766, -104.99022674560547], [39.774383544921875, -104.99026489257812], [39.77424621582031, -104.99028778076172], [39.77410888671875, -104.99029541015625], [39.773887634277344, -104.99030303955078], [39.77339553833008, -104.99031829833984], [39.773284912109375, -104.99031829833984], [39.77314376831055, -104.99032592773438], [39.773006439208984, -104.99034118652344], [39.77288055419922, -104.99036407470703], [39.77272033691406, -104.99040222167969], [39.77251434326172, -104.99046325683594], [39.77234649658203, -104.99051666259766], [39.77214431762695, -104.99059295654297], [39.77194595336914, -104.99068450927734], [39.77174377441406, -104.99079895019531], [39.77142333984375, -104.99098205566406], [39.77122116088867, -104.9911117553711], [39.77073287963867, -104.9914321899414], [39.7704963684082, -104.99164581298828], [39.770259857177734, -104.99187469482422], [39.77001190185547, -104.99214172363281], [39.769779205322266, -104.99241638183594], [39.76954650878906, -104.99272918701172], [39.76931381225586, -104.99305725097656], [39.76912307739258, -104.99334716796875], [39.76897048950195, -104.99359893798828], [39.768775939941406, -104.99393463134766], [39.768592834472656, -104.99432373046875], [39.76849365234375, -104.99454498291016], [39.76826477050781, -104.9951171875], [39.768192291259766, -104.99535369873047], [39.76802062988281, -104.99580383300781], [39.767581939697266, -104.99694061279297], [39.767154693603516, -104.99806213378906], [39.766963958740234, -104.9985580444336], [39.76682662963867, -104.9989013671875], [39.7667121887207, -104.99915313720703], [39.76656723022461, -104.99944305419922], [39.76641845703125, -104.99970245361328], [39.766273498535156, -104.99990844726562], [39.766109466552734, -105.00015258789062], [39.7659912109375, -105.00031280517578], [39.7658805847168, -105.0004653930664], [39.765750885009766, -105.00062561035156], [39.765567779541016, -105.00081634521484], [39.76536560058594, -105.00100708007812], [39.76518249511719, -105.00117492675781], [39.76498794555664, -105.00135803222656], [39.763919830322266, -105.002197265625], [39.76262283325195, -105.00331115722656], [39.76240539550781, -105.00350189208984], [39.762203216552734, -105.0036849975586], [39.76198959350586, -105.00389099121094], [39.76179885864258, -105.00407409667969], [39.7616081237793, -105.0042724609375], [39.761444091796875, -105.00445556640625], [39.761260986328125, -105.00466918945312], [39.76103973388672, -105.00492858886719], [39.76044845581055, -105.00575256347656], [39.759456634521484, -105.0072021484375], [39.75861740112305, -105.0083999633789], [39.758094787597656, -105.00910949707031], [39.75727844238281, -105.0101547241211], [39.75709915161133, -105.01038360595703], [39.756736755371094, -105.0108871459961], [39.75618362426758, -105.0116195678711], [39.75572967529297, -105.01225280761719], [39.75527572631836, -105.01285552978516], [39.75492858886719, -105.01334381103516], [39.7547492980957, -105.01359558105469], [39.75423812866211, -105.01427459716797], [39.75400924682617, -105.01457214355469], [39.753841400146484, -105.0147705078125], [39.75370788574219, -105.01492309570312], [39.75352096557617, -105.01513671875], [39.75336837768555, -105.01529693603516], [39.753173828125, -105.01548767089844], [39.75297164916992, -105.01567840576172], [39.752811431884766, -105.01581573486328], [39.75261688232422, -105.01597595214844], [39.75239944458008, -105.0161361694336], [39.75222396850586, -105.01627349853516], [39.752037048339844, -105.01639556884766], [39.751853942871094, -105.0165023803711], [39.75165939331055, -105.01661682128906], [39.75148391723633, -105.01670837402344], [39.75117874145508, -105.0168685913086], [39.75090026855469, -105.01699829101562], [39.75063705444336, -105.01710510253906], [39.74983215332031, -105.01744079589844], [39.74938201904297, -105.01763916015625], [39.74889373779297, -105.01783752441406], [39.748531341552734, -105.01799011230469], [39.74820327758789, -105.01811218261719], [39.74799728393555, -105.01818084716797], [39.74775695800781, -105.01825714111328], [39.74755096435547, -105.01830291748047], [39.74735641479492, -105.01834106445312], [39.747108459472656, -105.01836395263672], [39.74689865112305, -105.01837158203125], [39.74666976928711, -105.01836395263672], [39.74650955200195, -105.01835632324219], [39.74633026123047, -105.01832580566406], [39.746131896972656, -105.0182876586914], [39.74597930908203, -105.01824951171875], [39.74581527709961, -105.01819610595703], [39.745609283447266, -105.01811218261719], [39.74538040161133, -105.01800537109375], [39.74514389038086, -105.01786804199219], [39.744937896728516, -105.01773834228516], [39.744747161865234, -105.01759338378906], [39.74456787109375, -105.0174560546875], [39.74444580078125, -105.01734161376953], [39.743892669677734, -105.01679992675781], [39.74342727661133, -105.0163345336914], [39.743038177490234, -105.015869140625], [39.742897033691406, -105.01570892333984], [39.74272537231445, -105.0155258178711], [39.7425422668457, -105.01532745361328], [39.74235916137695, -105.01516723632812], [39.7421875, -105.01500701904297], [39.74201965332031, -105.0148696899414], [39.74187469482422, -105.01476287841797], [39.74171447753906, -105.0146484375], [39.74155807495117, -105.01456451416016], [39.741329193115234, -105.01445007324219], [39.74116134643555, -105.01438903808594], [39.74099349975586, -105.01432037353516], [39.74075698852539, -105.01425170898438], [39.7405891418457, -105.01421356201172], [39.740386962890625, -105.01416778564453], [39.740230560302734, -105.01415252685547], [39.74014663696289, -105.0141372680664], [39.73987579345703, -105.01412963867188], [39.73881530761719, -105.01406860351562], [39.73861312866211, -105.0140609741211], [39.73841094970703, -105.0140609741211], [39.73823928833008, -105.01406860351562], [39.73807907104492, -105.01408386230469], [39.7379035949707, -105.01409912109375], [39.73770523071289, -105.01412963867188], [39.737457275390625, -105.01416778564453], [39.737152099609375, -105.01428985595703], [39.73687744140625, -105.01436614990234], [39.7366828918457, -105.01441955566406], [39.73640823364258, -105.01451873779297], [39.73619079589844, -105.01460266113281], [39.73607635498047, -105.01464080810547], [39.735904693603516, -105.01471710205078], [39.73575973510742, -105.01478576660156], [39.73533630371094, -105.0149917602539], [39.73477554321289, -105.01528930664062], [39.734619140625, -105.01536560058594], [39.73448181152344, -105.01542663574219], [39.73434066772461, -105.01548767089844], [39.73418426513672, -105.01554870605469], [39.73397445678711, -105.015625], [39.733829498291016, -105.01566314697266], [39.73370361328125, -105.01570129394531], [39.73356246948242, -105.01573944091797], [39.733436584472656, -105.0157699584961], [39.73330307006836, -105.01579284667969], [39.73316955566406, -105.01582336425781], [39.73301315307617, -105.01583862304688], [39.73285675048828, -105.01585388183594], [39.73271942138672, -105.01586151123047], [39.732574462890625, -105.015869140625], [39.73243713378906, -105.01587677001953], [39.732303619384766, -105.01588439941406], [39.73218536376953, -105.01588439941406], [39.73207092285156, -105.01587677001953], [39.73195266723633, -105.015869140625], [39.73179244995117, -105.01585388183594], [39.73155212402344, -105.01580047607422], [39.73124313354492, -105.01573181152344], [39.731075286865234, -105.01568603515625], [39.7308464050293, -105.01561737060547], [39.730567932128906, -105.01553344726562], [39.730201721191406, -105.0154037475586], [39.729949951171875, -105.01531219482422], [39.72952651977539, -105.01518249511719], [39.729122161865234, -105.0150375366211], [39.728843688964844, -105.01493835449219], [39.72844314575195, -105.01477813720703], [39.72832107543945, -105.01473236083984], [39.728145599365234, -105.01466369628906], [39.727943420410156, -105.01458740234375], [39.727439880371094, -105.01441192626953], [39.72711181640625, -105.01427459716797], [39.72694778442383, -105.01421356201172], [39.726749420166016, -105.01412200927734], [39.726558685302734, -105.0140380859375], [39.726375579833984, -105.01394653320312], [39.72616195678711, -105.01382446289062], [39.72585678100586, -105.01362609863281], [39.725624084472656, -105.01348114013672], [39.72536087036133, -105.01329803466797], [39.72511672973633, -105.01311492919922], [39.72493362426758, -105.01297760009766], [39.7247200012207, -105.01280212402344], [39.72451400756836, -105.01262664794922], [39.72426986694336, -105.01239776611328], [39.72408676147461, -105.01221466064453], [39.723899841308594, -105.01201629638672], [39.72377395629883, -105.01188659667969], [39.72359848022461, -105.01168060302734], [39.72343826293945, -105.0114974975586], [39.72330093383789, -105.01133728027344], [39.723079681396484, -105.01106262207031], [39.7227897644043, -105.01066589355469], [39.72258377075195, -105.01038360595703], [39.722251892089844, -105.0099105834961], [39.7220573425293, -105.0096435546875], [39.721920013427734, -105.00946807861328], [39.721744537353516, -105.00924682617188], [39.721649169921875, -105.00914001464844], [39.7214241027832, -105.00888061523438], [39.72117233276367, -105.00861358642578], [39.720741271972656, -105.0081787109375], [39.720523834228516, -105.00794219970703], [39.7200813293457, -105.0074691772461], [39.71998977661133, -105.00737762451172], [39.71975326538086, -105.00711822509766], [39.71941375732422, -105.00675201416016], [39.719242095947266, -105.0065689086914], [39.719085693359375, -105.00640106201172], [39.718936920166016, -105.00623321533203], [39.718807220458984, -105.00608825683594], [39.718658447265625, -105.00591278076172], [39.71849060058594, -105.00570678710938], [39.7183723449707, -105.00556182861328], [39.71826171875, -105.00541687011719], [39.71812057495117, -105.00521850585938], [39.71792221069336, -105.00492858886719], [39.71759796142578, -105.00445556640625], [39.717166900634766, -105.0038070678711], [39.716880798339844, -105.00337982177734], [39.71671676635742, -105.00314331054688], [39.716552734375, -105.00292205810547], [39.7164306640625, -105.00276184082031], [39.71633529663086, -105.00264739990234], [39.71624755859375, -105.00254821777344], [39.71617889404297, -105.00247192382812], [39.71608352661133, -105.00237274169922], [39.715999603271484, -105.00228118896484], [39.71593475341797, -105.0022201538086], [39.715885162353516, -105.0021743774414], [39.715755462646484, -105.00206756591797], [39.715660095214844, -105.00198364257812], [39.715538024902344, -105.00189208984375], [39.71543884277344, -105.00181579589844], [39.71533203125, -105.00173950195312], [39.7152214050293, -105.00166320800781], [39.71499252319336, -105.00151824951172], [39.714324951171875, -105.0010757446289], [39.71399688720703, -105.00086212158203], [39.71346664428711, -105.00052642822266], [39.71318435668945, -105.00035095214844], [39.71266174316406, -105.00003051757812], [39.71223449707031, -104.9997787475586], [39.71159362792969, -104.99940490722656], [39.71141052246094, -104.99930572509766], [39.710933685302734, -104.99901580810547], [39.71067810058594, -104.99886322021484], [39.710357666015625, -104.99867248535156], [39.71013641357422, -104.99855041503906], [39.70988464355469, -104.99840545654297], [39.70966720581055, -104.99829864501953], [39.7094841003418, -104.99820709228516], [39.70940017700195, -104.99816131591797], [39.709136962890625, -104.99803924560547], [39.708702087402344, -104.99784088134766], [39.70827865600586, -104.99764251708984], [39.7078857421875, -104.99746704101562], [39.707275390625, -104.9971923828125], [39.70710372924805, -104.99711608886719], [39.7068977355957, -104.99701690673828], [39.706809997558594, -104.99697875976562], [39.706729888916016, -104.99693298339844], [39.70663070678711, -104.99688720703125], [39.706539154052734, -104.99683380126953], [39.70646667480469, -104.99678802490234], [39.70637130737305, -104.9967269897461], [39.70630645751953, -104.99668884277344], [39.70623779296875, -104.99664306640625], [39.70616912841797, -104.99658966064453], [39.706085205078125, -104.99652862548828], [39.70596694946289, -104.9964370727539], [39.705814361572266, -104.9963150024414], [39.7056999206543, -104.9962158203125], [39.70561218261719, -104.99613189697266], [39.70552062988281, -104.99604034423828], [39.705467224121094, -104.99598693847656], [39.70537567138672, -104.99588775634766], [39.705299377441406, -104.99580383300781], [39.705169677734375, -104.99565124511719], [39.70507049560547, -104.99552917480469], [39.704959869384766, -104.9953842163086], [39.704837799072266, -104.99520874023438], [39.70473098754883, -104.99505615234375], [39.704627990722656, -104.99488067626953], [39.704551696777344, -104.99474334716797], [39.704471588134766, -104.99459838867188], [39.70440673828125, -104.99447631835938], [39.7043342590332, -104.99433135986328], [39.704261779785156, -104.99417877197266], [39.70420455932617, -104.99404907226562], [39.70415115356445, -104.99392700195312], [39.704097747802734, -104.99378204345703], [39.70404815673828, -104.99363708496094], [39.70399475097656, -104.99347686767578], [39.70393753051758, -104.9933090209961], [39.70388412475586, -104.99313354492188], [39.703792572021484, -104.99284362792969], [39.70363235473633, -104.9923324584961], [39.703487396240234, -104.99188232421875], [39.703304290771484, -104.99127197265625], [39.703189849853516, -104.99092102050781], [39.70305633544922, -104.990478515625], [39.702980041503906, -104.99024200439453], [39.70287322998047, -104.989990234375], [39.70277404785156, -104.98977661132812], [39.70265197753906, -104.98954010009766], [39.70252990722656, -104.98933410644531], [39.70241928100586, -104.98914337158203], [39.70228576660156, -104.98894500732422], [39.702110290527344, -104.98870849609375], [39.70179748535156, -104.9883041381836], [39.70148849487305, -104.98790740966797], [39.700843811035156, -104.987060546875], [39.700706481933594, -104.98689270019531], [39.70027542114258, -104.98641967773438], [39.700077056884766, -104.9861831665039], [39.69994354248047, -104.98603057861328], [39.699771881103516, -104.98583984375], [39.69960021972656, -104.98567962646484], [39.699462890625, -104.98554229736328], [39.699344635009766, -104.98542022705078], [39.69913101196289, -104.98521423339844], [39.698974609375, -104.98504638671875], [39.69866180419922, -104.9847412109375], [39.6981086730957, -104.98410034179688], [39.69713592529297, -104.98299407958984], [39.696075439453125, -104.98184967041016], [39.69560623168945, -104.9813003540039], [39.69524383544922, -104.98087310791016], [39.69488525390625, -104.98043823242188], [39.694786071777344, -104.98033905029297], [39.6947021484375, -104.98025512695312], [39.69462585449219, -104.98019409179688], [39.6945915222168, -104.98016357421875], [39.69448471069336, -104.98009490966797], [39.69437789916992, -104.98002624511719], [39.6942024230957, -104.97992706298828], [39.69398498535156, -104.97977447509766], [39.69377899169922, -104.97953033447266], [39.69367599487305, -104.97943115234375], [39.69364929199219, -104.97940063476562], [39.69361114501953, -104.97936248779297], [39.69355010986328, -104.97931671142578], [39.6934814453125, -104.9792709350586], [39.69330978393555, -104.97918701171875], [39.69325637817383, -104.97914123535156], [39.693199157714844, -104.97909545898438], [39.693145751953125, -104.97903442382812], [39.693092346191406, -104.97897338867188], [39.69303512573242, -104.97886657714844], [39.69300079345703, -104.97877502441406], [39.69298553466797, -104.97871398925781], [39.69298553466797, -104.97850036621094], [39.69298553466797, -104.97827911376953], [39.69298553466797, -104.97792053222656], [39.6929931640625, -104.9774398803711], [39.6929931640625, -104.9773941040039], [39.6929931640625, -104.9772720336914], [39.692989349365234, -104.97704315185547], [39.692989349365234, -104.97694396972656], [39.692989349365234, -104.97685241699219], [39.69298553466797, -104.97636413574219], [39.6929817199707, -104.97577667236328]],\n", " {"bubblingMouseEvents": true, "color": "red", "dashArray": null, "dashOffset": null, "fill": false, "fillColor": "red", "fillOpacity": 0.2, "fillRule": "evenodd", "lineCap": "round", "lineJoin": "round", "noClip": false, "opacity": 0.8, "smoothFactor": 1.0, "stroke": true, "weight": 10}\n", - " ).addTo(map_a5cb9760638531fd5b14e75a46b3b1f1);\n", + " ).addTo(map_b6cf708d54aa4599b756d3ce62516c9c);\n", " \n", " \n", - " poly_line_b64ee0679bc07d7a55e8ab1075c0227e.bindTooltip(\n", + " poly_line_0633987bf2e82ded548a26dc464c4978.bindTooltip(\n", " `<div>\n", " Shortest Time\n", " </div>`,\n", @@ -821,19 +829,19 @@ " );\n", " \n", " \n", - " var marker_68f779ab1d817630f938af7a6c95a2b2 = L.marker(\n", + " var marker_51e8e84125a77a2fb180e6f323d27b85 = L.marker(\n", " [39.77902603149414, -104.96916961669922],\n", " {}\n", - " ).addTo(map_a5cb9760638531fd5b14e75a46b3b1f1);\n", + " ).addTo(map_b6cf708d54aa4599b756d3ce62516c9c);\n", " \n", " \n", - " var icon_cad02f21805c4a5539d3cc33a1463c3a = L.AwesomeMarkers.icon(\n", + " var icon_5fe7f896fc566755b249e564107262ca = L.AwesomeMarkers.icon(\n", " {"extraClasses": "fa-rotate-0", "icon": "circle", "iconColor": "white", "markerColor": "green", "prefix": "fa"}\n", " );\n", - " marker_68f779ab1d817630f938af7a6c95a2b2.setIcon(icon_cad02f21805c4a5539d3cc33a1463c3a);\n", + " marker_51e8e84125a77a2fb180e6f323d27b85.setIcon(icon_5fe7f896fc566755b249e564107262ca);\n", " \n", " \n", - " marker_68f779ab1d817630f938af7a6c95a2b2.bindTooltip(\n", + " marker_51e8e84125a77a2fb180e6f323d27b85.bindTooltip(\n", " `<div>\n", " Origin\n", " </div>`,\n", @@ -841,19 +849,19 @@ " );\n", " \n", " \n", - " var marker_a4d31af8d6a545d85c81795dd2d38ffe = L.marker(\n", + " var marker_b3dbf16a30f95c6a552145a7c117e057 = L.marker(\n", " [39.6929817199707, -104.97577667236328],\n", " {}\n", - " ).addTo(map_a5cb9760638531fd5b14e75a46b3b1f1);\n", + " ).addTo(map_b6cf708d54aa4599b756d3ce62516c9c);\n", " \n", " \n", - " var icon_2a7dbd406a91da668bcdaab591f6b9e1 = L.AwesomeMarkers.icon(\n", + " var icon_05c025ef9bca364192e682fa3b0133d4 = L.AwesomeMarkers.icon(\n", " {"extraClasses": "fa-rotate-0", "icon": "circle", "iconColor": "white", "markerColor": "red", "prefix": "fa"}\n", " );\n", - " marker_a4d31af8d6a545d85c81795dd2d38ffe.setIcon(icon_2a7dbd406a91da668bcdaab591f6b9e1);\n", + " marker_b3dbf16a30f95c6a552145a7c117e057.setIcon(icon_05c025ef9bca364192e682fa3b0133d4);\n", " \n", " \n", - " marker_a4d31af8d6a545d85c81795dd2d38ffe.bindTooltip(\n", + " marker_b3dbf16a30f95c6a552145a7c117e057.bindTooltip(\n", " `<div>\n", " Destination\n", " </div>`,\n", @@ -861,13 +869,13 @@ " );\n", " \n", " \n", - " var poly_line_812484b9eb27ef423783fc9826e4c79d = L.polyline(\n", + " var poly_line_b88057c0260a68bf7b601e9b657ca271 = L.polyline(\n", " [[39.77902603149414, -104.96916961669922], [39.77896499633789, -104.96910095214844], [39.77878189086914, -104.96887969970703], [39.778751373291016, -104.96884155273438], [39.778587341308594, -104.9686508178711], [39.778480529785156, -104.96852111816406], [39.77839660644531, -104.96841430664062], [39.778324127197266, -104.96851348876953], [39.778228759765625, -104.96863555908203], [39.778076171875, -104.96883392333984], [39.777626037597656, -104.96941375732422], [39.7774772644043, -104.96961212158203], [39.77703857421875, -104.97017669677734], [39.776737213134766, -104.97056579589844], [39.77666473388672, -104.97066497802734], [39.7766227722168, -104.97071838378906], [39.776371002197266, -104.97103881835938], [39.77608108520508, -104.9714126586914], [39.775978088378906, -104.97154998779297], [39.774810791015625, -104.97306060791016], [39.77475357055664, -104.97313690185547], [39.77467346191406, -104.97323608398438], [39.77459716796875, -104.97333526611328], [39.773738861083984, -104.97444915771484], [39.77363586425781, -104.97457885742188], [39.773353576660156, -104.97494506835938], [39.77303695678711, -104.97535705566406], [39.77297592163086, -104.9754409790039], [39.77290344238281, -104.97553253173828], [39.772544860839844, -104.97599029541016], [39.772335052490234, -104.97626495361328], [39.77207946777344, -104.97659301757812], [39.771976470947266, -104.97673034667969], [39.771705627441406, -104.97708129882812], [39.77125549316406, -104.97766876220703], [39.77039337158203, -104.9787826538086], [39.77033233642578, -104.9788589477539], [39.770263671875, -104.97895050048828], [39.769996643066406, -104.97929382324219], [39.76996994018555, -104.9793930053711], [39.76955032348633, -104.97993469238281], [39.769287109375, -104.98027801513672], [39.7691764831543, -104.98041534423828], [39.769039154052734, -104.98059844970703], [39.76881408691406, -104.98088836669922], [39.768733978271484, -104.98099517822266], [39.76864242553711, -104.98110961914062], [39.76858901977539, -104.9811782836914], [39.76840591430664, -104.98141479492188], [39.7678337097168, -104.98216247558594], [39.76764678955078, -104.9823989868164], [39.76701736450195, -104.98321533203125], [39.766876220703125, -104.9833984375], [39.766693115234375, -104.98363494873047], [39.766624450683594, -104.98372650146484], [39.76657485961914, -104.9837875366211], [39.76653289794922, -104.98384857177734], [39.766178131103516, -104.98429870605469], [39.766021728515625, -104.98450469970703], [39.76557159423828, -104.98507690429688], [39.76506423950195, -104.98573303222656], [39.76464080810547, -104.98627471923828], [39.764495849609375, -104.98646545410156], [39.76442337036133, -104.98655700683594], [39.764347076416016, -104.98664855957031], [39.763763427734375, -104.98738861083984], [39.76366424560547, -104.98750305175781], [39.763572692871094, -104.98760223388672], [39.76346969604492, -104.98770141601562], [39.76337814331055, -104.98778533935547], [39.76326370239258, -104.98786926269531], [39.76316452026367, -104.9879379272461], [39.76304626464844, -104.98800659179688], [39.762916564941406, -104.98807525634766], [39.762813568115234, -104.98812103271484], [39.76266860961914, -104.98816680908203], [39.76252746582031, -104.98819732666016], [39.762413024902344, -104.98820495605469], [39.762290954589844, -104.98821258544922], [39.76216125488281, -104.98821258544922], [39.762062072753906, -104.98820495605469], [39.76179885864258, -104.9881591796875], [39.76161575317383, -104.98812103271484], [39.76142501831055, -104.98808288574219], [39.76121520996094, -104.98804473876953], [39.75983428955078, -104.98776245117188], [39.75953674316406, -104.98770141601562], [39.75946807861328, -104.98763275146484], [39.75936508178711, -104.98761749267578], [39.7588996887207, -104.987548828125], [39.758426666259766, -104.98747253417969], [39.758270263671875, -104.9874496459961], [39.75812911987305, -104.98744201660156], [39.75796127319336, -104.98744201660156], [39.75767517089844, -104.9874267578125], [39.75761032104492, -104.9874267578125], [39.75736618041992, -104.98741912841797], [39.757179260253906, -104.98741149902344], [39.757102966308594, -104.98741149902344], [39.756961822509766, -104.98741149902344], [39.756805419921875, -104.9874038696289], [39.756404876708984, -104.9874038696289], [39.75601577758789, -104.9874038696289], [39.75578308105469, -104.98739624023438], [39.75563049316406, -104.98739624023438], [39.755496978759766, -104.98739624023438], [39.75507736206055, -104.98738861083984], [39.75468063354492, -104.98738861083984], [39.75453186035156, -104.98738098144531], [39.754398345947266, -104.98738098144531], [39.754364013671875, -104.98738098144531], [39.75428009033203, -104.98738861083984], [39.75388717651367, -104.98738861083984], [39.753604888916016, -104.98738861083984], [39.75343704223633, -104.98738861083984], [39.7532844543457, -104.98738861083984], [39.75323486328125, -104.98738861083984], [39.75295639038086, -104.98738861083984], [39.752933502197266, -104.98738861083984], [39.752662658691406, -104.98738861083984], [39.752262115478516, -104.98738861083984], [39.75224304199219, -104.98738098144531], [39.751827239990234, -104.98738098144531], [39.75173568725586, -104.98738098144531], [39.751583099365234, -104.98738098144531], [39.75148391723633, -104.98738098144531], [39.751014709472656, -104.98738098144531], [39.750850677490234, -104.98738098144531], [39.75082015991211, -104.98738098144531], [39.750736236572266, -104.98738098144531], [39.750389099121094, -104.98737335205078], [39.75023651123047, -104.98737335205078], [39.7501106262207, -104.98737335205078], [39.7497673034668, -104.98737335205078], [39.74958419799805, -104.98737335205078], [39.74951171875, -104.98737335205078], [39.749412536621094, -104.98737335205078], [39.74927520751953, -104.98738098144531], [39.749168395996094, -104.98738098144531], [39.74892044067383, -104.98738098144531], [39.74868392944336, -104.98738098144531], [39.74851989746094, -104.98738098144531], [39.74822235107422, -104.98738098144531], [39.74802017211914, -104.98738098144531], [39.7476692199707, -104.98738098144531], [39.74758529663086, -104.98738098144531], [39.74751281738281, -104.98738098144531], [39.7474365234375, -104.98738098144531], [39.747379302978516, -104.98738098144531], [39.74735641479492, -104.98738098144531], [39.74730682373047, -104.98738098144531], [39.74697494506836, -104.98738098144531], [39.746944427490234, -104.98738098144531], [39.74678039550781, -104.98738861083984], [39.74667739868164, -104.98738861083984], [39.74653244018555, -104.98738861083984], [39.746482849121094, -104.98738861083984], [39.74626541137695, -104.98738861083984], [39.74618148803711, -104.98738861083984], [39.74605178833008, -104.98738861083984], [39.745948791503906, -104.98738861083984], [39.74554443359375, -104.98738861083984], [39.745418548583984, -104.98738861083984], [39.74506378173828, -104.98739624023438], [39.74480438232422, -104.98739624023438], [39.74473571777344, -104.9874038696289], [39.74430847167969, -104.98739624023438], [39.743526458740234, -104.98739624023438], [39.74336242675781, -104.98739624023438], [39.74320602416992, -104.98739624023438], [39.74302291870117, -104.98739624023438], [39.74254608154297, -104.98739624023438], [39.74245834350586, -104.98739624023438], [39.74220275878906, -104.98739624023438], [39.74214172363281, -104.98738861083984], [39.74182891845703, -104.98738861083984], [39.74177932739258, -104.98738861083984], [39.7416877746582, -104.98738861083984], [39.74163818359375, -104.98738098144531], [39.74156188964844, -104.98738861083984], [39.74149703979492, -104.98738861083984], [39.741416931152344, -104.98738861083984], [39.74137496948242, -104.98738861083984], [39.74129867553711, -104.98738861083984], [39.7410888671875, -104.98738861083984], [39.74089050292969, -104.98739624023438], [39.740875244140625, -104.98739624023438], [39.74071502685547, -104.98739624023438], [39.740631103515625, -104.98739624023438], [39.740478515625, -104.98739624023438], [39.74021530151367, -104.9874038696289], [39.74014663696289, -104.9874038696289], [39.74000930786133, -104.9874038696289], [39.73992919921875, -104.9874038696289], [39.73912048339844, -104.98739624023438], [39.73855209350586, -104.98739624023438], [39.73845672607422, -104.98739624023438], [39.73835372924805, -104.98739624023438], [39.73801040649414, -104.98739624023438], [39.73710250854492, -104.98738861083984], [39.73695373535156, -104.98738861083984], [39.736900329589844, -104.98738861083984], [39.736846923828125, -104.98738861083984], [39.73676300048828, -104.98738861083984], [39.73644256591797, -104.98739624023438], [39.73640441894531, -104.98739624023438], [39.736244201660156, -104.98739624023438], [39.73619079589844, -104.98739624023438], [39.73557662963867, -104.98739624023438], [39.73536682128906, -104.98739624023438], [39.73527526855469, -104.98739624023438], [39.73518753051758, -104.98739624023438], [39.73509216308594, -104.98739624023438], [39.73466110229492, -104.98739624023438], [39.734039306640625, -104.9874038696289], [39.73394775390625, -104.98739624023438], [39.733848571777344, -104.9874038696289], [39.73375701904297, -104.9874038696289], [39.73365020751953, -104.9874038696289], [39.73337173461914, -104.9874038696289], [39.733306884765625, -104.9874038696289], [39.732810974121094, -104.9874038696289], [39.732444763183594, -104.98741149902344], [39.732147216796875, -104.98741149902344], [39.7320556640625, -104.98741149902344], [39.731964111328125, -104.98741149902344], [39.73185348510742, -104.98741149902344], [39.73183059692383, -104.98741149902344], [39.73168182373047, -104.98741149902344], [39.73122024536133, -104.98741912841797], [39.731178283691406, -104.98741912841797], [39.73095703125, -104.98741912841797], [39.73054885864258, -104.98741912841797], [39.73045349121094, -104.98741912841797], [39.73037338256836, -104.98741912841797], [39.730010986328125, -104.98741912841797], [39.729522705078125, -104.98741912841797], [39.72943115234375, -104.98741912841797], [39.72917175292969, -104.98741912841797], [39.72908020019531, -104.98741912841797], [39.7289924621582, -104.98741912841797], [39.728668212890625, -104.98741912841797], [39.72862243652344, -104.98741912841797], [39.72829055786133, -104.98741912841797], [39.727420806884766, -104.98741912841797], [39.727298736572266, -104.98741912841797], [39.72718048095703, -104.98741912841797], [39.72665786743164, -104.9874267578125], [39.72663497924805, -104.9874267578125], [39.726600646972656, -104.9874267578125], [39.726375579833984, -104.98743438720703], [39.726226806640625, -104.98744201660156], [39.726112365722656, -104.98744201660156], [39.72574996948242, -104.98745727539062], [39.72563934326172, -104.98746490478516], [39.72555923461914, -104.98746490478516], [39.725223541259766, -104.98747253417969], [39.7251091003418, -104.98747253417969], [39.724884033203125, -104.98748016357422], [39.72444152832031, -104.98748779296875], [39.72422409057617, -104.98748779296875], [39.72414779663086, -104.98748779296875], [39.72405242919922, -104.98748779296875], [39.72385025024414, -104.98748779296875], [39.72378921508789, -104.98748779296875], [39.72360610961914, -104.98748779296875], [39.72319793701172, -104.98750305175781], [39.723026275634766, -104.98751068115234], [39.722869873046875, -104.98751068115234], [39.72262191772461, -104.98751068115234], [39.72255325317383, -104.98751068115234], [39.72245788574219, -104.98751068115234], [39.72215270996094, -104.98750305175781], [39.72206497192383, -104.98750305175781], [39.72195816040039, -104.98749542236328], [39.7216911315918, -104.98748779296875], [39.72103500366211, -104.98748779296875], [39.720943450927734, -104.98748779296875], [39.720863342285156, -104.98748779296875], [39.71990966796875, -104.98748779296875], [39.71987533569336, -104.98748779296875], [39.719451904296875, -104.98748779296875], [39.7193603515625, -104.98748779296875], [39.71928787231445, -104.98748779296875], [39.71902847290039, -104.98748779296875], [39.71844482421875, -104.98748779296875], [39.71834945678711, -104.98748779296875], [39.7182731628418, -104.98748779296875], [39.71819305419922, -104.98748779296875], [39.71754455566406, -104.98748779296875], [39.717464447021484, -104.98748779296875], [39.7174186706543, -104.98748779296875], [39.716644287109375, -104.98748779296875], [39.71656799316406, -104.98748779296875], [39.71649932861328, -104.98748779296875], [39.716224670410156, -104.98749542236328], [39.71613311767578, -104.98749542236328], [39.71575927734375, -104.98750305175781], [39.71568298339844, -104.98750305175781], [39.715606689453125, -104.98750305175781], [39.71514892578125, -104.98751831054688], [39.714847564697266, -104.98753356933594], [39.71476364135742, -104.98753356933594], [39.71470260620117, -104.98753356933594], [39.713897705078125, -104.987548828125], [39.713294982910156, -104.987548828125], [39.71319580078125, -104.987548828125], [39.71306610107422, -104.98755645751953], [39.71297836303711, -104.98755645751953], [39.71291732788086, -104.98755645751953], [39.712486267089844, -104.98756408691406], [39.71209716796875, -104.9875717163086], [39.71201705932617, -104.98757934570312], [39.71174240112305, -104.98757934570312], [39.711578369140625, -104.98758697509766], [39.711280822753906, -104.98759460449219], [39.71114730834961, -104.98759460449219], [39.71101379394531, -104.98759460449219], [39.71085739135742, -104.98759460449219], [39.71025085449219, -104.98760986328125], [39.71002960205078, -104.98760986328125], [39.70989227294922, -104.98761749267578], [39.70948028564453, -104.98761749267578], [39.70942306518555, -104.98760986328125], [39.70934295654297, -104.98761749267578], [39.709259033203125, -104.98761749267578], [39.70914840698242, -104.98762512207031], [39.7091064453125, -104.98762512207031], [39.70846939086914, -104.9876480102539], [39.70835876464844, -104.9876480102539], [39.70762634277344, -104.9876480102539], [39.707542419433594, -104.9876480102539], [39.707462310791016, -104.98764038085938], [39.7066535949707, -104.98760223388672], [39.70616912841797, -104.98758697509766], [39.705841064453125, -104.9875717163086], [39.70573806762695, -104.98756408691406], [39.705665588378906, -104.98755645751953], [39.705406188964844, -104.98754119873047], [39.705223083496094, -104.98753356933594], [39.705101013183594, -104.9875259399414], [39.704872131347656, -104.9875259399414], [39.7047119140625, -104.9875259399414], [39.704627990722656, -104.98751831054688], [39.70394515991211, -104.98751068115234], [39.703880310058594, -104.98751068115234], [39.70368194580078, -104.98750305175781], [39.70335006713867, -104.98750305175781], [39.70302963256836, -104.98750305175781], [39.7028694152832, -104.98750305175781], [39.70270538330078, -104.98751831054688], [39.70259475708008, -104.9875259399414], [39.702491760253906, -104.98753356933594], [39.70227813720703, -104.987548828125], [39.7021369934082, -104.98755645751953], [39.70204162597656, -104.98756408691406], [39.700843811035156, -104.98760986328125], [39.700782775878906, -104.98749542236328], [39.70073318481445, -104.9874038696289], [39.700687408447266, -104.98731994628906], [39.700660705566406, -104.98727416992188], [39.70061111450195, -104.9871826171875], [39.70054626464844, -104.98707580566406], [39.70036315917969, -104.98675537109375], [39.70024490356445, -104.986572265625], [39.70008850097656, -104.98636627197266], [39.700042724609375, -104.98631286621094], [39.69866180419922, -104.9847412109375], [39.6981086730957, -104.98410034179688], [39.69713592529297, -104.98299407958984], [39.696075439453125, -104.98184967041016], [39.69560623168945, -104.9813003540039], [39.69524383544922, -104.98087310791016], [39.69488525390625, -104.98043823242188], [39.694786071777344, -104.98033905029297], [39.6947021484375, -104.98025512695312], [39.69462585449219, -104.98019409179688], [39.6945915222168, -104.98016357421875], [39.69448471069336, -104.98009490966797], [39.69437789916992, -104.98002624511719], [39.6942024230957, -104.97992706298828], [39.69398498535156, -104.97977447509766], [39.69377899169922, -104.97953033447266], [39.69367599487305, -104.97943115234375], [39.69364929199219, -104.97940063476562], [39.69361114501953, -104.97936248779297], [39.69355010986328, -104.97931671142578], [39.6934814453125, -104.9792709350586], [39.69330978393555, -104.97918701171875], [39.69325637817383, -104.97914123535156], [39.693199157714844, -104.97909545898438], [39.693145751953125, -104.97903442382812], [39.693092346191406, -104.97897338867188], [39.69303512573242, -104.97886657714844], [39.69300079345703, -104.97877502441406], [39.69298553466797, -104.97871398925781], [39.69298553466797, -104.97850036621094], [39.69298553466797, -104.97827911376953], [39.69298553466797, -104.97792053222656], [39.6929931640625, -104.9774398803711], [39.6929931640625, -104.9773941040039], [39.6929931640625, -104.9772720336914], [39.692989349365234, -104.97704315185547], [39.692989349365234, -104.97694396972656], [39.692989349365234, -104.97685241699219], [39.69298553466797, -104.97636413574219], [39.6929817199707, -104.97577667236328]],\n", " {"bubblingMouseEvents": true, "color": "green", "dashArray": null, "dashOffset": null, "fill": false, "fillColor": "green", "fillOpacity": 0.2, "fillRule": "evenodd", "lineCap": "round", "lineJoin": "round", "noClip": false, "opacity": 0.8, "smoothFactor": 1.0, "stroke": true, "weight": 10}\n", - " ).addTo(map_a5cb9760638531fd5b14e75a46b3b1f1);\n", + " ).addTo(map_b6cf708d54aa4599b756d3ce62516c9c);\n", " \n", " \n", - " poly_line_812484b9eb27ef423783fc9826e4c79d.bindTooltip(\n", + " poly_line_b88057c0260a68bf7b601e9b657ca271.bindTooltip(\n", " `<div>\n", " Least Energy\n", " </div>`,\n", @@ -875,19 +883,19 @@ " );\n", " \n", " \n", - " var marker_001ab4478d92ad31bfd4defff343e8ea = L.marker(\n", + " var marker_51449498af2992f67a03ee1a4743b9b3 = L.marker(\n", " [39.77902603149414, -104.96916961669922],\n", " {}\n", - " ).addTo(map_a5cb9760638531fd5b14e75a46b3b1f1);\n", + " ).addTo(map_b6cf708d54aa4599b756d3ce62516c9c);\n", " \n", " \n", - " var icon_2ab048e082a6330258a4a74b57a185dc = L.AwesomeMarkers.icon(\n", + " var icon_0d220eb467754e9f97fb970647ef051a = L.AwesomeMarkers.icon(\n", " {"extraClasses": "fa-rotate-0", "icon": "circle", "iconColor": "white", "markerColor": "green", "prefix": "fa"}\n", " );\n", - " marker_001ab4478d92ad31bfd4defff343e8ea.setIcon(icon_2ab048e082a6330258a4a74b57a185dc);\n", + " marker_51449498af2992f67a03ee1a4743b9b3.setIcon(icon_0d220eb467754e9f97fb970647ef051a);\n", " \n", " \n", - " marker_001ab4478d92ad31bfd4defff343e8ea.bindTooltip(\n", + " marker_51449498af2992f67a03ee1a4743b9b3.bindTooltip(\n", " `<div>\n", " Origin\n", " </div>`,\n", @@ -895,19 +903,19 @@ " );\n", " \n", " \n", - " var marker_f1c968e33b632da1bdd369a1c4c1d2eb = L.marker(\n", + " var marker_0844385ced7afed6079e777212cdc756 = L.marker(\n", " [39.6929817199707, -104.97577667236328],\n", " {}\n", - " ).addTo(map_a5cb9760638531fd5b14e75a46b3b1f1);\n", + " ).addTo(map_b6cf708d54aa4599b756d3ce62516c9c);\n", " \n", " \n", - " var icon_8497e7de14e7e225b487c736ec56ea06 = L.AwesomeMarkers.icon(\n", + " var icon_c89925817012d0ad601284ad1f049e9e = L.AwesomeMarkers.icon(\n", " {"extraClasses": "fa-rotate-0", "icon": "circle", "iconColor": "white", "markerColor": "red", "prefix": "fa"}\n", " );\n", - " marker_f1c968e33b632da1bdd369a1c4c1d2eb.setIcon(icon_8497e7de14e7e225b487c736ec56ea06);\n", + " marker_0844385ced7afed6079e777212cdc756.setIcon(icon_c89925817012d0ad601284ad1f049e9e);\n", " \n", " \n", - " marker_f1c968e33b632da1bdd369a1c4c1d2eb.bindTooltip(\n", + " marker_0844385ced7afed6079e777212cdc756.bindTooltip(\n", " `<div>\n", " Destination\n", " </div>`,\n", @@ -915,13 +923,13 @@ " );\n", " \n", " \n", - " var poly_line_6b1b8ea217704dede06972c2196f427f = L.polyline(\n", - " [[39.77902603149414, -104.96916961669922], [39.77896499633789, -104.96910095214844], [39.77878189086914, -104.96887969970703], [39.778751373291016, -104.96884155273438], [39.778587341308594, -104.9686508178711], [39.778480529785156, -104.96852111816406], [39.77839660644531, -104.96841430664062], [39.778324127197266, -104.96851348876953], [39.778228759765625, -104.96863555908203], [39.778076171875, -104.96883392333984], [39.777626037597656, -104.96941375732422], [39.7774772644043, -104.96961212158203], [39.77703857421875, -104.97017669677734], [39.776737213134766, -104.97056579589844], [39.77666473388672, -104.97066497802734], [39.7766227722168, -104.97071838378906], [39.776371002197266, -104.97103881835938], [39.77608108520508, -104.9714126586914], [39.775978088378906, -104.97154998779297], [39.774810791015625, -104.97306060791016], [39.77475357055664, -104.97313690185547], [39.77467346191406, -104.97323608398438], [39.77459716796875, -104.97333526611328], [39.773738861083984, -104.97444915771484], [39.77363586425781, -104.97457885742188], [39.773353576660156, -104.97494506835938], [39.77303695678711, -104.97535705566406], [39.77297592163086, -104.9754409790039], [39.77287673950195, -104.97531127929688], [39.77263259887695, -104.9749984741211], [39.77257537841797, -104.97492218017578], [39.77229690551758, -104.97456359863281], [39.77223587036133, -104.9744873046875], [39.77199935913086, -104.97418212890625], [39.77191162109375, -104.97406768798828], [39.77155685424805, -104.9736099243164], [39.771270751953125, -104.97322845458984], [39.7706184387207, -104.97236633300781], [39.77056884765625, -104.97230529785156], [39.77048110961914, -104.97219848632812], [39.77042770385742, -104.9721450805664], [39.7703742980957, -104.97209167480469], [39.77033233642578, -104.97205352783203], [39.770233154296875, -104.97198486328125], [39.770118713378906, -104.9721450805664], [39.76996994018555, -104.97232818603516], [39.76978302001953, -104.97257232666016], [39.76957321166992, -104.97283172607422], [39.76926803588867, -104.97325134277344], [39.76918411254883, -104.97335815429688], [39.76904296875, -104.97335815429688], [39.76840591430664, -104.97335052490234], [39.76824188232422, -104.97335052490234], [39.7681770324707, -104.97335052490234], [39.7680549621582, -104.97335052490234], [39.7677001953125, -104.97335052490234], [39.767547607421875, -104.97335052490234], [39.7670783996582, -104.97335052490234], [39.76704788208008, -104.97335052490234], [39.766971588134766, -104.97335052490234], [39.76690673828125, -104.97335052490234], [39.76670455932617, -104.97335052490234], [39.76658248901367, -104.97335052490234], [39.766502380371094, -104.97335052490234], [39.76637649536133, -104.97335052490234], [39.766136169433594, -104.97335052490234], [39.76605987548828, -104.97335052490234], [39.765865325927734, -104.97335052490234], [39.76577377319336, -104.97335052490234], [39.76567459106445, -104.97335052490234], [39.76523208618164, -104.97335052490234], [39.76495361328125, -104.97335052490234], [39.764835357666016, -104.97335052490234], [39.7646598815918, -104.97335052490234], [39.764591217041016, -104.97335052490234], [39.7645149230957, -104.97335052490234], [39.764427185058594, -104.97335052490234], [39.76421356201172, -104.97335052490234], [39.764034271240234, -104.97335052490234], [39.76380920410156, -104.97335052490234], [39.763580322265625, -104.97335052490234], [39.76325988769531, -104.97335052490234], [39.76316833496094, -104.97339630126953], [39.763118743896484, -104.97339630126953], [39.76306915283203, -104.97335052490234], [39.76280212402344, -104.97335815429688], [39.762454986572266, -104.97335815429688], [39.76237869262695, -104.97335815429688], [39.76214599609375, -104.97335052490234], [39.762054443359375, -104.97335052490234], [39.76194381713867, -104.97335052490234], [39.76179122924805, -104.97335052490234], [39.76169967651367, -104.97335052490234], [39.76160430908203, -104.97335052490234], [39.76118087768555, -104.97335052490234], [39.76103973388672, -104.97335052490234], [39.76084518432617, -104.97335052490234], [39.76076126098633, -104.97335052490234], [39.76066589355469, -104.97335052490234], [39.760440826416016, -104.97335052490234], [39.76033401489258, -104.97335052490234], [39.759613037109375, -104.97335052490234], [39.7595100402832, -104.97335052490234], [39.75939178466797, -104.97339630126953], [39.759151458740234, -104.97339630126953], [39.75912094116211, -104.97335052490234], [39.759010314941406, -104.97335052490234], [39.758827209472656, -104.97335052490234], [39.75831604003906, -104.97335052490234], [39.75823211669922, -104.97335052490234], [39.758174896240234, -104.97335052490234], [39.75764083862305, -104.97335815429688], [39.7572021484375, -104.97335815429688], [39.75708770751953, -104.97335815429688], [39.75699234008789, -104.97335815429688], [39.75688171386719, -104.97335815429688], [39.75673294067383, -104.97335815429688], [39.75627899169922, -104.97335815429688], [39.75574493408203, -104.97335815429688], [39.75540542602539, -104.97335815429688], [39.75531005859375, -104.97335815429688], [39.75495147705078, -104.97335815429688], [39.75469970703125, -104.97335815429688], [39.754547119140625, -104.97335815429688], [39.75442886352539, -104.97335815429688], [39.753578186035156, -104.97335815429688], [39.75333786010742, -104.97335815429688], [39.75325012207031, -104.97335815429688], [39.75314712524414, -104.97335815429688], [39.75208282470703, -104.97335815429688], [39.751991271972656, -104.97335815429688], [39.751895904541016, -104.97335815429688], [39.75139236450195, -104.9733657836914], [39.75082778930664, -104.9733657836914], [39.750732421875, -104.9733657836914], [39.75063705444336, -104.9733657836914], [39.750423431396484, -104.9733657836914], [39.750282287597656, -104.9733657836914], [39.75018310546875, -104.9733657836914], [39.749576568603516, -104.9733657836914], [39.74947738647461, -104.9733657836914], [39.74938201904297, -104.9733657836914], [39.74884033203125, -104.9733657836914], [39.74821853637695, -104.97337341308594], [39.74812698364258, -104.97337341308594], [39.748016357421875, -104.97337341308594], [39.747798919677734, -104.97337341308594], [39.74757766723633, -104.97337341308594], [39.74747085571289, -104.97337341308594], [39.74736785888672, -104.97337341308594], [39.74721908569336, -104.97337341308594], [39.747066497802734, -104.97337341308594], [39.746952056884766, -104.97337341308594], [39.746864318847656, -104.97337341308594], [39.74663162231445, -104.97337341308594], [39.746551513671875, -104.97338104248047], [39.746482849121094, -104.97338104248047], [39.7460823059082, -104.97338104248047], [39.74594497680664, -104.97338104248047], [39.74583435058594, -104.97338104248047], [39.74509048461914, -104.97340393066406], [39.745052337646484, -104.97340393066406], [39.745018005371094, -104.97344207763672], [39.744972229003906, -104.97344970703125], [39.744937896728516, -104.97344970703125], [39.744834899902344, -104.97339630126953], [39.744834899902344, -104.97355651855469], [39.744834899902344, -104.97415924072266], [39.74483871459961, -104.97454833984375], [39.74483871459961, -104.97476196289062], [39.74464416503906, -104.97476196289062], [39.744606018066406, -104.97476196289062], [39.74433517456055, -104.97476959228516], [39.74381637573242, -104.97476959228516], [39.74373245239258, -104.97476959228516], [39.743377685546875, -104.97477722167969], [39.74324417114258, -104.97477722167969], [39.743133544921875, -104.97477722167969], [39.7426872253418, -104.97477722167969], [39.74175262451172, -104.97477722167969], [39.74166488647461, -104.97477722167969], [39.74156951904297, -104.97477722167969], [39.7412109375, -104.97478485107422], [39.740718841552734, -104.97479248046875], [39.74055862426758, -104.97479248046875], [39.74032211303711, -104.97479248046875], [39.74012756347656, -104.97480010986328], [39.740020751953125, -104.97480010986328], [39.740020751953125, -104.9746322631836], [39.740020751953125, -104.97451782226562], [39.74002456665039, -104.97415924072266], [39.73991775512695, -104.97415924072266], [39.73970031738281, -104.97415924072266], [39.73955535888672, -104.97415161132812], [39.739315032958984, -104.97415161132812], [39.73896789550781, -104.97415161132812], [39.738609313964844, -104.97415161132812], [39.73850631713867, -104.9741439819336], [39.73841857910156, -104.9741439819336], [39.73833465576172, -104.9741439819336], [39.736942291259766, -104.9741439819336], [39.736881256103516, -104.9741439819336], [39.73681640625, -104.9741439819336], [39.73567581176758, -104.97412109375], [39.735252380371094, -104.97411346435547], [39.735164642333984, -104.97411346435547], [39.735069274902344, -104.97411346435547], [39.73379898071289, -104.97410583496094], [39.73370361328125, -104.97410583496094], [39.733612060546875, -104.97410583496094], [39.73240280151367, -104.97410583496094], [39.732357025146484, -104.97410583496094], [39.73214340209961, -104.97410583496094], [39.7320556640625, -104.97410583496094], [39.73196029663086, -104.97410583496094], [39.731422424316406, -104.97410583496094], [39.73128890991211, -104.97410583496094], [39.730960845947266, -104.97410583496094], [39.73088455200195, -104.97410583496094], [39.73067855834961, -104.97410583496094], [39.73054504394531, -104.97410583496094], [39.7304573059082, -104.97410583496094], [39.730369567871094, -104.97410583496094], [39.73027420043945, -104.97410583496094], [39.73003387451172, -104.97410583496094], [39.729976654052734, -104.97410583496094], [39.729896545410156, -104.97410583496094], [39.72964859008789, -104.97410583496094], [39.7294807434082, -104.97410583496094], [39.729434967041016, -104.97410583496094], [39.72917175292969, -104.97410583496094], [39.72907257080078, -104.97410583496094], [39.72898864746094, -104.97410583496094], [39.727378845214844, -104.9740982055664], [39.72727966308594, -104.9740982055664], [39.727195739746094, -104.9740982055664], [39.7259635925293, -104.9740982055664], [39.72566604614258, -104.97409057617188], [39.7255859375, -104.97409057617188], [39.72549819946289, -104.97409057617188], [39.725337982177734, -104.97409057617188], [39.7252082824707, -104.97409057617188], [39.724693298339844, -104.97409057617188], [39.72453689575195, -104.97409057617188], [39.72406768798828, -104.97408294677734], [39.72374725341797, -104.97408294677734], [39.72247314453125, -104.97408294677734], [39.72218322753906, -104.97408294677734], [39.72205352783203, -104.97408294677734], [39.721946716308594, -104.97408294677734], [39.72013473510742, -104.97406768798828], [39.719642639160156, -104.97406005859375], [39.71931457519531, -104.97406005859375], [39.71918869018555, -104.97406005859375], [39.718971252441406, -104.97405242919922], [39.71888732910156, -104.97404479980469], [39.71880340576172, -104.97403717041016], [39.71875, -104.97402954101562], [39.71870422363281, -104.9740219116211], [39.718650817871094, -104.97401428222656], [39.71857452392578, -104.97399139404297], [39.718505859375, -104.97396850585938], [39.71845245361328, -104.97393798828125], [39.71842956542969, -104.97392272949219], [39.71839141845703, -104.97390747070312], [39.71834945678711, -104.97386932373047], [39.718318939208984, -104.97382354736328], [39.71824264526367, -104.97368621826172], [39.71818923950195, -104.97358703613281], [39.71816635131836, -104.97354888916016], [39.71815490722656, -104.9735336303711], [39.71811294555664, -104.97348022460938], [39.71806335449219, -104.97343444824219], [39.71802520751953, -104.97341918945312], [39.717979431152344, -104.97340393066406], [39.717952728271484, -104.97339630126953], [39.71792221069336, -104.973388671875], [39.717899322509766, -104.973388671875], [39.71767807006836, -104.973388671875], [39.716922760009766, -104.973388671875], [39.71653366088867, -104.97339630126953], [39.71645736694336, -104.97339630126953], [39.71589660644531, -104.97339630126953], [39.71546936035156, -104.97341918945312], [39.71541976928711, -104.97341918945312], [39.71538543701172, -104.97342681884766], [39.71536636352539, -104.97342681884766], [39.71527862548828, -104.97342681884766], [39.7148551940918, -104.97344970703125], [39.71479415893555, -104.97344970703125], [39.71472930908203, -104.97345733642578], [39.714656829833984, -104.97344970703125], [39.71451950073242, -104.97344207763672], [39.71427917480469, -104.9734115600586], [39.712989807128906, -104.97340393066406], [39.71291732788086, -104.97340393066406], [39.71285629272461, -104.97340393066406], [39.71247482299805, -104.97340393066406], [39.711551666259766, -104.97340393066406], [39.71137237548828, -104.97340393066406], [39.711204528808594, -104.97340393066406], [39.71110534667969, -104.97340393066406], [39.711021423339844, -104.97340393066406], [39.7109489440918, -104.97340393066406], [39.710845947265625, -104.97340393066406], [39.709381103515625, -104.973388671875], [39.70929718017578, -104.973388671875], [39.707820892333984, -104.973388671875], [39.707557678222656, -104.973388671875], [39.70749282836914, -104.973388671875], [39.707420349121094, -104.973388671875], [39.70706558227539, -104.97338104248047], [39.70566940307617, -104.97337341308594], [39.70392990112305, -104.973388671875], [39.70386505126953, -104.973388671875], [39.70379638671875, -104.973388671875], [39.70212173461914, -104.97339630126953], [39.70206069946289, -104.97339630126953], [39.70030212402344, -104.97339630126953], [39.70021057128906, -104.97340393066406], [39.700138092041016, -104.97339630126953], [39.698429107666016, -104.97340393066406], [39.69660949707031, -104.9734115600586], [39.69625473022461, -104.97341918945312], [39.694740295410156, -104.97341918945312], [39.694149017333984, -104.97343444824219], [39.69339370727539, -104.97343444824219], [39.69325256347656, -104.97343444824219], [39.69306564331055, -104.97342681884766], [39.6929817199707, -104.97342681884766], [39.6929817199707, -104.97357177734375], [39.6929817199707, -104.97403717041016], [39.6929817199707, -104.974609375], [39.69298553466797, -104.97517395019531], [39.6929817199707, -104.97577667236328]],\n", + " var poly_line_cb6d42712bef935f49dac78318d38137 = L.polyline(\n", + " [[39.77902603149414, -104.96916961669922], [39.77896499633789, -104.96910095214844], [39.77878189086914, -104.96887969970703], [39.778751373291016, -104.96884155273438], [39.778587341308594, -104.9686508178711], [39.778480529785156, -104.96852111816406], [39.77839660644531, -104.96841430664062], [39.778324127197266, -104.96851348876953], [39.778228759765625, -104.96863555908203], [39.778076171875, -104.96883392333984], [39.777626037597656, -104.96941375732422], [39.7774772644043, -104.96961212158203], [39.77703857421875, -104.97017669677734], [39.776737213134766, -104.97056579589844], [39.77666473388672, -104.97066497802734], [39.7766227722168, -104.97071838378906], [39.776371002197266, -104.97103881835938], [39.77608108520508, -104.9714126586914], [39.775978088378906, -104.97154998779297], [39.774810791015625, -104.97306060791016], [39.77475357055664, -104.97313690185547], [39.77467346191406, -104.97323608398438], [39.77459716796875, -104.97333526611328], [39.773738861083984, -104.97444915771484], [39.77363586425781, -104.97457885742188], [39.773353576660156, -104.97494506835938], [39.77303695678711, -104.97535705566406], [39.77297592163086, -104.9754409790039], [39.77287673950195, -104.97531127929688], [39.77263259887695, -104.9749984741211], [39.77257537841797, -104.97492218017578], [39.77229690551758, -104.97456359863281], [39.77223587036133, -104.9744873046875], [39.77199935913086, -104.97418212890625], [39.77191162109375, -104.97406768798828], [39.77157974243164, -104.9736328125], [39.77155303955078, -104.9736328125], [39.77154541015625, -104.97361755371094], [39.7712516784668, -104.97325897216797], [39.77125549316406, -104.97321319580078], [39.77090072631836, -104.97274017333984], [39.7706184387207, -104.97236633300781], [39.77056884765625, -104.97230529785156], [39.77048110961914, -104.97219848632812], [39.77042770385742, -104.9721450805664], [39.7703742980957, -104.97209167480469], [39.77033233642578, -104.97205352783203], [39.770233154296875, -104.97198486328125], [39.770118713378906, -104.9721450805664], [39.76996994018555, -104.97232818603516], [39.76978302001953, -104.97257232666016], [39.76957321166992, -104.97283172607422], [39.76926803588867, -104.97325134277344], [39.76918411254883, -104.97335815429688], [39.76904296875, -104.97335815429688], [39.76840591430664, -104.97335052490234], [39.76824188232422, -104.97335052490234], [39.7681770324707, -104.97335052490234], [39.7680549621582, -104.97335052490234], [39.7677001953125, -104.97335052490234], [39.767547607421875, -104.97335052490234], [39.7670783996582, -104.97335052490234], [39.76704788208008, -104.97335052490234], [39.766971588134766, -104.97335052490234], [39.76690673828125, -104.97335052490234], [39.76670455932617, -104.97335052490234], [39.76658248901367, -104.97335052490234], [39.766502380371094, -104.97335052490234], [39.76637649536133, -104.97335052490234], [39.766136169433594, -104.97335052490234], [39.76605987548828, -104.97335052490234], [39.765865325927734, -104.97335052490234], [39.76577377319336, -104.97335052490234], [39.76567459106445, -104.97335052490234], [39.76523208618164, -104.97335052490234], [39.76495361328125, -104.97335052490234], [39.764835357666016, -104.97335052490234], [39.7646598815918, -104.97335052490234], [39.764591217041016, -104.97335052490234], [39.7645149230957, -104.97335052490234], [39.764427185058594, -104.97335052490234], [39.76421356201172, -104.97335052490234], [39.764034271240234, -104.97335052490234], [39.76380920410156, -104.97335052490234], [39.763580322265625, -104.97335052490234], [39.76325988769531, -104.97335052490234], [39.76316833496094, -104.97339630126953], [39.763118743896484, -104.97339630126953], [39.76306915283203, -104.97335052490234], [39.762874603271484, -104.97335815429688], [39.762454986572266, -104.97335815429688], [39.76237869262695, -104.97335815429688], [39.76214599609375, -104.97335052490234], [39.762054443359375, -104.97335052490234], [39.76194381713867, -104.97335052490234], [39.76179122924805, -104.97335052490234], [39.76169967651367, -104.97335052490234], [39.76160430908203, -104.97335052490234], [39.76118087768555, -104.97335052490234], [39.76103973388672, -104.97335052490234], [39.76084518432617, -104.97335052490234], [39.76076126098633, -104.97335052490234], [39.76066589355469, -104.97335052490234], [39.760440826416016, -104.97335052490234], [39.76033401489258, -104.97335052490234], [39.759613037109375, -104.97335052490234], [39.7595100402832, -104.97335052490234], [39.75939178466797, -104.97339630126953], [39.759151458740234, -104.97339630126953], [39.75912094116211, -104.97335052490234], [39.759010314941406, -104.97335052490234], [39.758827209472656, -104.97335052490234], [39.75831604003906, -104.97335052490234], [39.75823211669922, -104.97335052490234], [39.758174896240234, -104.97335052490234], [39.75764083862305, -104.97335815429688], [39.7572021484375, -104.97335815429688], [39.75708770751953, -104.97335815429688], [39.75699234008789, -104.97335815429688], [39.75688171386719, -104.97335815429688], [39.75673294067383, -104.97335815429688], [39.75627899169922, -104.97335815429688], [39.75574493408203, -104.97335815429688], [39.75540542602539, -104.97335815429688], [39.75531005859375, -104.97335815429688], [39.75495147705078, -104.97335815429688], [39.75469970703125, -104.97335815429688], [39.754547119140625, -104.97335815429688], [39.75442886352539, -104.97335815429688], [39.753578186035156, -104.97335815429688], [39.75333786010742, -104.97335815429688], [39.75325012207031, -104.97335815429688], [39.75314712524414, -104.97335815429688], [39.75208282470703, -104.97335815429688], [39.751991271972656, -104.97335815429688], [39.751895904541016, -104.97335815429688], [39.75139236450195, -104.9733657836914], [39.75082778930664, -104.9733657836914], [39.750732421875, -104.9733657836914], [39.75063705444336, -104.9733657836914], [39.750423431396484, -104.9733657836914], [39.750282287597656, -104.9733657836914], [39.75018310546875, -104.9733657836914], [39.749576568603516, -104.9733657836914], [39.74947738647461, -104.9733657836914], [39.74938201904297, -104.9733657836914], [39.74884033203125, -104.9733657836914], [39.74821853637695, -104.97337341308594], [39.74812698364258, -104.97337341308594], [39.748016357421875, -104.97337341308594], [39.747798919677734, -104.97337341308594], [39.74757766723633, -104.97337341308594], [39.74747085571289, -104.97337341308594], [39.74736785888672, -104.97337341308594], [39.74721908569336, -104.97337341308594], [39.747066497802734, -104.97337341308594], [39.746952056884766, -104.97337341308594], [39.746864318847656, -104.97337341308594], [39.74663162231445, -104.97337341308594], [39.746551513671875, -104.97338104248047], [39.746482849121094, -104.97338104248047], [39.7460823059082, -104.97338104248047], [39.74594497680664, -104.97338104248047], [39.74583435058594, -104.97338104248047], [39.74509048461914, -104.97340393066406], [39.745052337646484, -104.97340393066406], [39.745018005371094, -104.97344207763672], [39.744972229003906, -104.97344970703125], [39.744937896728516, -104.97344970703125], [39.744834899902344, -104.97339630126953], [39.744834899902344, -104.97355651855469], [39.744834899902344, -104.97415924072266], [39.74483871459961, -104.97454833984375], [39.74483871459961, -104.97476196289062], [39.74464416503906, -104.97476196289062], [39.744606018066406, -104.97476196289062], [39.74433517456055, -104.97476959228516], [39.74381637573242, -104.97476959228516], [39.74373245239258, -104.97476959228516], [39.743377685546875, -104.97477722167969], [39.74324417114258, -104.97477722167969], [39.743133544921875, -104.97477722167969], [39.7426872253418, -104.97477722167969], [39.74175262451172, -104.97477722167969], [39.74166488647461, -104.97477722167969], [39.74156951904297, -104.97477722167969], [39.7412109375, -104.97478485107422], [39.740718841552734, -104.97479248046875], [39.74055862426758, -104.97479248046875], [39.74032211303711, -104.97479248046875], [39.74012756347656, -104.97480010986328], [39.740020751953125, -104.97480010986328], [39.740020751953125, -104.9746322631836], [39.740020751953125, -104.97451782226562], [39.74002456665039, -104.97415924072266], [39.73991775512695, -104.97415924072266], [39.73970031738281, -104.97415924072266], [39.73955535888672, -104.97415161132812], [39.739315032958984, -104.97415161132812], [39.73896789550781, -104.97415161132812], [39.738609313964844, -104.97415161132812], [39.73850631713867, -104.9741439819336], [39.73841857910156, -104.9741439819336], [39.73833465576172, -104.9741439819336], [39.736942291259766, -104.9741439819336], [39.736881256103516, -104.9741439819336], [39.73681640625, -104.9741439819336], [39.73567581176758, -104.97412109375], [39.735252380371094, -104.97411346435547], [39.735164642333984, -104.97411346435547], [39.735069274902344, -104.97411346435547], [39.73379898071289, -104.97410583496094], [39.73370361328125, -104.97410583496094], [39.733612060546875, -104.97410583496094], [39.73240280151367, -104.97410583496094], [39.732357025146484, -104.97410583496094], [39.73214340209961, -104.97410583496094], [39.7320556640625, -104.97410583496094], [39.73196029663086, -104.97410583496094], [39.731422424316406, -104.97410583496094], [39.73128890991211, -104.97410583496094], [39.730960845947266, -104.97410583496094], [39.73088455200195, -104.97410583496094], [39.73067855834961, -104.97410583496094], [39.73054504394531, -104.97410583496094], [39.7304573059082, -104.97410583496094], [39.730369567871094, -104.97410583496094], [39.73027420043945, -104.97410583496094], [39.73003387451172, -104.97410583496094], [39.729976654052734, -104.97410583496094], [39.729896545410156, -104.97410583496094], [39.72964859008789, -104.97410583496094], [39.7294807434082, -104.97410583496094], [39.729434967041016, -104.97410583496094], [39.72917175292969, -104.97410583496094], [39.72907257080078, -104.97410583496094], [39.72898864746094, -104.97410583496094], [39.727378845214844, -104.9740982055664], [39.72727966308594, -104.9740982055664], [39.727195739746094, -104.9740982055664], [39.7259635925293, -104.9740982055664], [39.72566604614258, -104.97409057617188], [39.7255859375, -104.97409057617188], [39.72549819946289, -104.97409057617188], [39.725337982177734, -104.97409057617188], [39.7252082824707, -104.97409057617188], [39.724693298339844, -104.97409057617188], [39.72453689575195, -104.97409057617188], [39.72406768798828, -104.97408294677734], [39.72374725341797, -104.97408294677734], [39.72247314453125, -104.97408294677734], [39.72218322753906, -104.97408294677734], [39.72205352783203, -104.97408294677734], [39.721946716308594, -104.97408294677734], [39.72013473510742, -104.97406768798828], [39.719642639160156, -104.97406005859375], [39.71931457519531, -104.97406005859375], [39.71918869018555, -104.97406005859375], [39.718971252441406, -104.97405242919922], [39.71888732910156, -104.97404479980469], [39.71880340576172, -104.97403717041016], [39.71875, -104.97402954101562], [39.71870422363281, -104.9740219116211], [39.718650817871094, -104.97401428222656], [39.71857452392578, -104.97399139404297], [39.718505859375, -104.97396850585938], [39.71845245361328, -104.97393798828125], [39.71842956542969, -104.97392272949219], [39.71839141845703, -104.97390747070312], [39.71834945678711, -104.97386932373047], [39.718318939208984, -104.97382354736328], [39.71824264526367, -104.97368621826172], [39.71818923950195, -104.97358703613281], [39.71816635131836, -104.97354888916016], [39.71815490722656, -104.9735336303711], [39.71811294555664, -104.97348022460938], [39.71806335449219, -104.97343444824219], [39.71802520751953, -104.97341918945312], [39.717979431152344, -104.97340393066406], [39.717952728271484, -104.97339630126953], [39.71792221069336, -104.973388671875], [39.717899322509766, -104.973388671875], [39.71767807006836, -104.973388671875], [39.716922760009766, -104.973388671875], [39.71653366088867, -104.97339630126953], [39.71645736694336, -104.97339630126953], [39.71589660644531, -104.97339630126953], [39.71546936035156, -104.97341918945312], [39.71541976928711, -104.97341918945312], [39.71538543701172, -104.97342681884766], [39.71536636352539, -104.97342681884766], [39.71527862548828, -104.97342681884766], [39.7148551940918, -104.97344970703125], [39.71479415893555, -104.97344970703125], [39.71472930908203, -104.97345733642578], [39.714656829833984, -104.97344970703125], [39.71451950073242, -104.97344207763672], [39.71427917480469, -104.9734115600586], [39.712989807128906, -104.97340393066406], [39.71291732788086, -104.97340393066406], [39.71285629272461, -104.97340393066406], [39.71247482299805, -104.97340393066406], [39.711551666259766, -104.97340393066406], [39.71137237548828, -104.97340393066406], [39.711204528808594, -104.97340393066406], [39.71110534667969, -104.97340393066406], [39.711021423339844, -104.97340393066406], [39.7109489440918, -104.97340393066406], [39.710845947265625, -104.97340393066406], [39.709381103515625, -104.973388671875], [39.70929718017578, -104.973388671875], [39.707820892333984, -104.973388671875], [39.707557678222656, -104.973388671875], [39.70749282836914, -104.973388671875], [39.707420349121094, -104.973388671875], [39.70706558227539, -104.97338104248047], [39.70566940307617, -104.97337341308594], [39.70392990112305, -104.973388671875], [39.70386505126953, -104.973388671875], [39.70379638671875, -104.973388671875], [39.70212173461914, -104.97339630126953], [39.70206069946289, -104.97339630126953], [39.70030212402344, -104.97339630126953], [39.70021057128906, -104.97340393066406], [39.700138092041016, -104.97339630126953], [39.698429107666016, -104.97340393066406], [39.69660949707031, -104.9734115600586], [39.69625473022461, -104.97341918945312], [39.694740295410156, -104.97341918945312], [39.694149017333984, -104.97343444824219], [39.69339370727539, -104.97343444824219], [39.69325256347656, -104.97343444824219], [39.69306564331055, -104.97342681884766], [39.6929817199707, -104.97342681884766], [39.6929817199707, -104.97357177734375], [39.6929817199707, -104.97403717041016], [39.6929817199707, -104.974609375], [39.69298553466797, -104.97517395019531], [39.6929817199707, -104.97577667236328]],\n", " {"bubblingMouseEvents": true, "color": "blue", "dashArray": null, "dashOffset": null, "fill": false, "fillColor": "blue", "fillOpacity": 0.2, "fillRule": "evenodd", "lineCap": "round", "lineJoin": "round", "noClip": false, "opacity": 0.8, "smoothFactor": 1.0, "stroke": true, "weight": 10}\n", - " ).addTo(map_a5cb9760638531fd5b14e75a46b3b1f1);\n", + " ).addTo(map_b6cf708d54aa4599b756d3ce62516c9c);\n", " \n", " \n", - " poly_line_6b1b8ea217704dede06972c2196f427f.bindTooltip(\n", + " poly_line_cb6d42712bef935f49dac78318d38137.bindTooltip(\n", " `<div>\n", " Least Cost\n", " </div>`,\n", @@ -929,19 +937,19 @@ " );\n", " \n", " \n", - " var marker_aa49883cffaf0f523687a8cb8e017e8d = L.marker(\n", + " var marker_d7e8d51bdfa0b414359dba4b6850173d = L.marker(\n", " [39.77902603149414, -104.96916961669922],\n", " {}\n", - " ).addTo(map_a5cb9760638531fd5b14e75a46b3b1f1);\n", + " ).addTo(map_b6cf708d54aa4599b756d3ce62516c9c);\n", " \n", " \n", - " var icon_fbf3a814c59b3c7dc86707db6d25d53a = L.AwesomeMarkers.icon(\n", + " var icon_777c8c4fb41598cadb9bc5aa82e9fc8e = L.AwesomeMarkers.icon(\n", " {"extraClasses": "fa-rotate-0", "icon": "circle", "iconColor": "white", "markerColor": "green", "prefix": "fa"}\n", " );\n", - " marker_aa49883cffaf0f523687a8cb8e017e8d.setIcon(icon_fbf3a814c59b3c7dc86707db6d25d53a);\n", + " marker_d7e8d51bdfa0b414359dba4b6850173d.setIcon(icon_777c8c4fb41598cadb9bc5aa82e9fc8e);\n", " \n", " \n", - " marker_aa49883cffaf0f523687a8cb8e017e8d.bindTooltip(\n", + " marker_d7e8d51bdfa0b414359dba4b6850173d.bindTooltip(\n", " `<div>\n", " Origin\n", " </div>`,\n", @@ -949,19 +957,19 @@ " );\n", " \n", " \n", - " var marker_0652f42a9bda08b402dea3440ab7709e = L.marker(\n", + " var marker_cd94423e998833f6c8afd0bb042a8e37 = L.marker(\n", " [39.6929817199707, -104.97577667236328],\n", " {}\n", - " ).addTo(map_a5cb9760638531fd5b14e75a46b3b1f1);\n", + " ).addTo(map_b6cf708d54aa4599b756d3ce62516c9c);\n", " \n", " \n", - " var icon_2e33d13a04a5446fc4e8dc8f2295adc2 = L.AwesomeMarkers.icon(\n", + " var icon_67a4dd8906dac98fef7f4c548ce0a4fe = L.AwesomeMarkers.icon(\n", " {"extraClasses": "fa-rotate-0", "icon": "circle", "iconColor": "white", "markerColor": "red", "prefix": "fa"}\n", " );\n", - " marker_0652f42a9bda08b402dea3440ab7709e.setIcon(icon_2e33d13a04a5446fc4e8dc8f2295adc2);\n", + " marker_cd94423e998833f6c8afd0bb042a8e37.setIcon(icon_67a4dd8906dac98fef7f4c548ce0a4fe);\n", " \n", " \n", - " marker_0652f42a9bda08b402dea3440ab7709e.bindTooltip(\n", + " marker_cd94423e998833f6c8afd0bb042a8e37.bindTooltip(\n", " `<div>\n", " Destination\n", " </div>`,\n", @@ -972,10 +980,10 @@ "</html>\" style=\"position:absolute;width:100%;height:100%;left:0;top:0;border:none !important;\" allowfullscreen webkitallowfullscreen mozallowfullscreen></iframe></div></div>" ], "text/plain": [ - "<folium.folium.Map at 0x129306dd0>" + "<folium.folium.Map at 0x1b03568b0>" ] }, - "execution_count": 187, + "execution_count": 17, "metadata": {}, "output_type": "execute_result" } @@ -997,7 +1005,7 @@ }, { "cell_type": "code", - "execution_count": 188, + "execution_count": 18, "id": "aa1e6b11", "metadata": {}, "outputs": [ @@ -1031,7 +1039,7 @@ " <meta name="viewport" content="width=device-width,\n", " initial-scale=1.0, maximum-scale=1.0, user-scalable=no" />\n", " <style>\n", - " #map_63d14900899a70183befbf30d354febd {\n", + " #map_77bceaf15f8f92685ff07972bb265db5 {\n", " position: relative;\n", " width: 100.0%;\n", " height: 100.0%;\n", @@ -1045,14 +1053,14 @@ "<body>\n", " \n", " \n", - " <div class="folium-map" id="map_63d14900899a70183befbf30d354febd" ></div>\n", + " <div class="folium-map" id="map_77bceaf15f8f92685ff07972bb265db5" ></div>\n", " \n", "</body>\n", "<script>\n", " \n", " \n", - " var map_63d14900899a70183befbf30d354febd = L.map(\n", - " "map_63d14900899a70183befbf30d354febd",\n", + " var map_77bceaf15f8f92685ff07972bb265db5 = L.map(\n", + " "map_77bceaf15f8f92685ff07972bb265db5",\n", " {\n", " center: [39.73823928833008, -105.01406860351562],\n", " crs: L.CRS.EPSG3857,\n", @@ -1066,39 +1074,39 @@ "\n", " \n", " \n", - " var tile_layer_9092626b7ec75f73d91deb0b7833c4fb = L.tileLayer(\n", + " var tile_layer_63693c84a48315790e7524f1adf3fbe4 = L.tileLayer(\n", " "https://{s}.tile.openstreetmap.org/{z}/{x}/{y}.png",\n", " {"attribution": "Data by \\u0026copy; \\u003ca target=\\"_blank\\" href=\\"http://openstreetmap.org\\"\\u003eOpenStreetMap\\u003c/a\\u003e, under \\u003ca target=\\"_blank\\" href=\\"http://www.openstreetmap.org/copyright\\"\\u003eODbL\\u003c/a\\u003e.", "detectRetina": false, "maxNativeZoom": 18, "maxZoom": 18, "minZoom": 0, "noWrap": false, "opacity": 1, "subdomains": "abc", "tms": false}\n", - " ).addTo(map_63d14900899a70183befbf30d354febd);\n", + " ).addTo(map_77bceaf15f8f92685ff07972bb265db5);\n", " \n", " \n", - " var poly_line_a70438f127ac39c09a7541b9864285e7 = L.polyline(\n", + " var poly_line_3096e7d1c76ca60404f4c47b9117a47c = L.polyline(\n", " [[39.77902603149414, -104.96916961669922], [39.77896499633789, -104.96910095214844], [39.77878189086914, -104.96887969970703], [39.778751373291016, -104.96884155273438], [39.778587341308594, -104.9686508178711], [39.778480529785156, -104.96852111816406], [39.77839660644531, -104.96841430664062], [39.77846908569336, -104.96832275390625], [39.77865982055664, -104.96807098388672], [39.778709411621094, -104.96791076660156], [39.77894592285156, -104.96761322021484], [39.77906799316406, -104.96745300292969], [39.77915954589844, -104.96732330322266], [39.7791862487793, -104.96728515625], [39.779232025146484, -104.96722412109375], [39.77926254272461, -104.96719360351562], [39.779293060302734, -104.9671630859375], [39.77932357788086, -104.96714782714844], [39.77938461303711, -104.96710968017578], [39.779441833496094, -104.96707916259766], [39.77951431274414, -104.96704864501953], [39.7796745300293, -104.96698760986328], [39.77983093261719, -104.9669418334961], [39.78031921386719, -104.96683502197266], [39.780364990234375, -104.9668197631836], [39.78050231933594, -104.966796875], [39.78049850463867, -104.96688079833984], [39.78049850463867, -104.96698760986328], [39.78049850463867, -104.96705627441406], [39.78049087524414, -104.96717834472656], [39.78046798706055, -104.96759033203125], [39.78045654296875, -104.96772003173828], [39.78042984008789, -104.96797180175781], [39.78041458129883, -104.96810913085938], [39.780372619628906, -104.96839904785156], [39.78034210205078, -104.96859741210938], [39.780296325683594, -104.96888732910156], [39.7802734375, -104.96903228759766], [39.78025436401367, -104.96917724609375], [39.780250549316406, -104.96920776367188], [39.780235290527344, -104.96934509277344], [39.78020095825195, -104.96965789794922], [39.78001403808594, -104.9715347290039], [39.77994155883789, -104.97315216064453], [39.77980422973633, -104.97496032714844], [39.7797737121582, -104.97691345214844], [39.7797737121582, -104.9772720336914], [39.7797737121582, -104.97753143310547], [39.7797737121582, -104.97764587402344], [39.779781341552734, -104.97801971435547], [39.77980041503906, -104.97874450683594], [39.77981948852539, -104.97929382324219], [39.77983093261719, -104.97946166992188], [39.77986526489258, -104.98001861572266], [39.77991485595703, -104.98080444335938], [39.779991149902344, -104.98214721679688], [39.78001403808594, -104.98335266113281], [39.7800178527832, -104.98450469970703], [39.78002166748047, -104.98558044433594], [39.780033111572266, -104.98602294921875], [39.7800407409668, -104.98613739013672], [39.780059814453125, -104.98633575439453], [39.780086517333984, -104.9865493774414], [39.780128479003906, -104.98677825927734], [39.780174255371094, -104.98695373535156], [39.78023910522461, -104.98715209960938], [39.780540466308594, -104.98787689208984], [39.78069305419922, -104.98828887939453], [39.78076934814453, -104.98860168457031], [39.780784606933594, -104.98871612548828], [39.780792236328125, -104.98896026611328], [39.78078079223633, -104.98912048339844], [39.780765533447266, -104.98928833007812], [39.78071975708008, -104.98950958251953], [39.78061294555664, -104.98975372314453], [39.78052520751953, -104.98993682861328], [39.78038787841797, -104.99011993408203], [39.78028869628906, -104.99024963378906], [39.780174255371094, -104.99034881591797], [39.77997589111328, -104.990478515625], [39.77995300292969, -104.99049377441406], [39.77978515625, -104.99056243896484], [39.77960205078125, -104.99060821533203], [39.7794189453125, -104.99060821533203], [39.77922439575195, -104.9905776977539], [39.778968811035156, -104.99048614501953], [39.77858352661133, -104.9903335571289], [39.778541564941406, -104.99031066894531], [39.77811050415039, -104.99011993408203], [39.777706146240234, -104.98995208740234], [39.7775764465332, -104.98990631103516], [39.77741241455078, -104.98985290527344], [39.77730941772461, -104.98983001708984], [39.7772102355957, -104.98980712890625], [39.777130126953125, -104.98979949951172], [39.77692794799805, -104.98979187011719], [39.77660369873047, -104.98979187011719], [39.77629089355469, -104.98980712890625], [39.77605056762695, -104.98982238769531], [39.77583694458008, -104.98986053466797], [39.77559280395508, -104.98991394042969], [39.7751579284668, -104.99003601074219], [39.774871826171875, -104.99012756347656], [39.77482986450195, -104.9901351928711], [39.774696350097656, -104.99018096923828], [39.774539947509766, -104.99022674560547], [39.774383544921875, -104.99026489257812], [39.77424621582031, -104.99028778076172], [39.77410888671875, -104.99029541015625], [39.773887634277344, -104.99030303955078], [39.77339553833008, -104.99031829833984], [39.773284912109375, -104.99031829833984], [39.77314376831055, -104.99032592773438], [39.773006439208984, -104.99034118652344], [39.77288055419922, -104.99036407470703], [39.77272033691406, -104.99040222167969], [39.77251434326172, -104.99046325683594], [39.77234649658203, -104.99051666259766], [39.77214431762695, -104.99059295654297], [39.77194595336914, -104.99068450927734], [39.77174377441406, -104.99079895019531], [39.77142333984375, -104.99098205566406], [39.77122116088867, -104.9911117553711], [39.77073287963867, -104.9914321899414], [39.7704963684082, -104.99164581298828], [39.770259857177734, -104.99187469482422], [39.77001190185547, -104.99214172363281], [39.769779205322266, -104.99241638183594], [39.76954650878906, -104.99272918701172], [39.76931381225586, -104.99305725097656], [39.76912307739258, -104.99334716796875], [39.76897048950195, -104.99359893798828], [39.768775939941406, -104.99393463134766], [39.768592834472656, -104.99432373046875], [39.76849365234375, -104.99454498291016], [39.76826477050781, -104.9951171875], [39.768192291259766, -104.99535369873047], [39.76802062988281, -104.99580383300781], [39.767581939697266, -104.99694061279297], [39.767154693603516, -104.99806213378906], [39.766963958740234, -104.9985580444336], [39.76682662963867, -104.9989013671875], [39.7667121887207, -104.99915313720703], [39.76656723022461, -104.99944305419922], [39.76641845703125, -104.99970245361328], [39.766273498535156, -104.99990844726562], [39.766109466552734, -105.00015258789062], [39.7659912109375, -105.00031280517578], [39.7658805847168, -105.0004653930664], [39.765750885009766, -105.00062561035156], [39.765567779541016, -105.00081634521484], [39.76536560058594, -105.00100708007812], [39.76518249511719, -105.00117492675781], [39.76498794555664, -105.00135803222656], [39.763919830322266, -105.002197265625], [39.76262283325195, -105.00331115722656], [39.76240539550781, -105.00350189208984], [39.762203216552734, -105.0036849975586], [39.76198959350586, -105.00389099121094], [39.76179885864258, -105.00407409667969], [39.7616081237793, -105.0042724609375], [39.761444091796875, -105.00445556640625], [39.761260986328125, -105.00466918945312], [39.76103973388672, -105.00492858886719], [39.76044845581055, -105.00575256347656], [39.759456634521484, -105.0072021484375], [39.75861740112305, -105.0083999633789], [39.758094787597656, -105.00910949707031], [39.75727844238281, -105.0101547241211], [39.75709915161133, -105.01038360595703], [39.756736755371094, -105.0108871459961], [39.75618362426758, -105.0116195678711], [39.75572967529297, -105.01225280761719], [39.75527572631836, -105.01285552978516], [39.75492858886719, -105.01334381103516], [39.7547492980957, -105.01359558105469], [39.75423812866211, -105.01427459716797], [39.75400924682617, -105.01457214355469], [39.753841400146484, -105.0147705078125], [39.75370788574219, -105.01492309570312], [39.75352096557617, -105.01513671875], [39.75336837768555, -105.01529693603516], [39.753173828125, -105.01548767089844], [39.75297164916992, -105.01567840576172], [39.752811431884766, -105.01581573486328], [39.75261688232422, -105.01597595214844], [39.75239944458008, -105.0161361694336], [39.75222396850586, -105.01627349853516], [39.752037048339844, -105.01639556884766], [39.751853942871094, -105.0165023803711], [39.75165939331055, -105.01661682128906], [39.75148391723633, -105.01670837402344], [39.75117874145508, -105.0168685913086], [39.75090026855469, -105.01699829101562], [39.75063705444336, -105.01710510253906], [39.74983215332031, -105.01744079589844], [39.74938201904297, -105.01763916015625], [39.74889373779297, -105.01783752441406], [39.748531341552734, -105.01799011230469], [39.74820327758789, -105.01811218261719], [39.74799728393555, -105.01818084716797], [39.74775695800781, -105.01825714111328], [39.74755096435547, -105.01830291748047], [39.74735641479492, -105.01834106445312], [39.747108459472656, -105.01836395263672], [39.74689865112305, -105.01837158203125], [39.74666976928711, -105.01836395263672], [39.74650955200195, -105.01835632324219], [39.74633026123047, -105.01832580566406], [39.746131896972656, -105.0182876586914], [39.74597930908203, -105.01824951171875], [39.74581527709961, -105.01819610595703], [39.745609283447266, -105.01811218261719], [39.74538040161133, -105.01800537109375], [39.74514389038086, -105.01786804199219], [39.744937896728516, -105.01773834228516], [39.744747161865234, -105.01759338378906], [39.74456787109375, -105.0174560546875], [39.74444580078125, -105.01734161376953], [39.743892669677734, -105.01679992675781], [39.74342727661133, -105.0163345336914], [39.743038177490234, -105.015869140625], [39.742897033691406, -105.01570892333984], [39.74272537231445, -105.0155258178711], [39.7425422668457, -105.01532745361328], [39.74235916137695, -105.01516723632812], [39.7421875, -105.01500701904297], [39.74201965332031, -105.0148696899414], [39.74187469482422, -105.01476287841797], [39.74171447753906, -105.0146484375], [39.74155807495117, -105.01456451416016], [39.741329193115234, -105.01445007324219], [39.74116134643555, -105.01438903808594], [39.74099349975586, -105.01432037353516], [39.74075698852539, -105.01425170898438], [39.7405891418457, -105.01421356201172], [39.740386962890625, -105.01416778564453], [39.740230560302734, -105.01415252685547], [39.74014663696289, -105.0141372680664], [39.73987579345703, -105.01412963867188], [39.73881530761719, -105.01406860351562], [39.73861312866211, -105.0140609741211], [39.73841094970703, -105.0140609741211], [39.73823928833008, -105.01406860351562], [39.73807907104492, -105.01408386230469], [39.7379035949707, -105.01409912109375], [39.73770523071289, -105.01412963867188], [39.737457275390625, -105.01416778564453], [39.737152099609375, -105.01428985595703], [39.73687744140625, -105.01436614990234], [39.7366828918457, -105.01441955566406], [39.73640823364258, -105.01451873779297], [39.73619079589844, -105.01460266113281], [39.73607635498047, -105.01464080810547], [39.735904693603516, -105.01471710205078], [39.73575973510742, -105.01478576660156], [39.73533630371094, -105.0149917602539], [39.73477554321289, -105.01528930664062], [39.734619140625, -105.01536560058594], [39.73448181152344, -105.01542663574219], [39.73434066772461, -105.01548767089844], [39.73418426513672, -105.01554870605469], [39.73397445678711, -105.015625], [39.733829498291016, -105.01566314697266], [39.73370361328125, -105.01570129394531], [39.73356246948242, -105.01573944091797], [39.733436584472656, -105.0157699584961], [39.73330307006836, -105.01579284667969], [39.73316955566406, -105.01582336425781], [39.73301315307617, -105.01583862304688], [39.73285675048828, -105.01585388183594], [39.73271942138672, -105.01586151123047], [39.732574462890625, -105.015869140625], [39.73243713378906, -105.01587677001953], [39.732303619384766, -105.01588439941406], [39.73218536376953, -105.01588439941406], [39.73207092285156, -105.01587677001953], [39.73195266723633, -105.015869140625], [39.73179244995117, -105.01585388183594], [39.73155212402344, -105.01580047607422], [39.73124313354492, -105.01573181152344], [39.731075286865234, -105.01568603515625], [39.7308464050293, -105.01561737060547], [39.730567932128906, -105.01553344726562], [39.730201721191406, -105.0154037475586], [39.729949951171875, -105.01531219482422], [39.72952651977539, -105.01518249511719], [39.729122161865234, -105.0150375366211], [39.728843688964844, -105.01493835449219], [39.72844314575195, -105.01477813720703], [39.72832107543945, -105.01473236083984], [39.728145599365234, -105.01466369628906], [39.727943420410156, -105.01458740234375], [39.727439880371094, -105.01441192626953], [39.72711181640625, -105.01427459716797], [39.72694778442383, -105.01421356201172], [39.726749420166016, -105.01412200927734], [39.726558685302734, -105.0140380859375], [39.726375579833984, -105.01394653320312], [39.72616195678711, -105.01382446289062], [39.72585678100586, -105.01362609863281], [39.725624084472656, -105.01348114013672], [39.72536087036133, -105.01329803466797], [39.72511672973633, -105.01311492919922], [39.72493362426758, -105.01297760009766], [39.7247200012207, -105.01280212402344], [39.72451400756836, -105.01262664794922], [39.72426986694336, -105.01239776611328], [39.72408676147461, -105.01221466064453], [39.723899841308594, -105.01201629638672], [39.72377395629883, -105.01188659667969], [39.72359848022461, -105.01168060302734], [39.72343826293945, -105.0114974975586], [39.72330093383789, -105.01133728027344], [39.723079681396484, -105.01106262207031], [39.7227897644043, -105.01066589355469], [39.72258377075195, -105.01038360595703], [39.722251892089844, -105.0099105834961], [39.7220573425293, -105.0096435546875], [39.721920013427734, -105.00946807861328], [39.721744537353516, -105.00924682617188], [39.721649169921875, -105.00914001464844], [39.7214241027832, -105.00888061523438], [39.72117233276367, -105.00861358642578], [39.720741271972656, -105.0081787109375], [39.720523834228516, -105.00794219970703], [39.7200813293457, -105.0074691772461], [39.71998977661133, -105.00737762451172], [39.71975326538086, -105.00711822509766], [39.71941375732422, -105.00675201416016], [39.719242095947266, -105.0065689086914], [39.719085693359375, -105.00640106201172], [39.718936920166016, -105.00623321533203], [39.718807220458984, -105.00608825683594], [39.718658447265625, -105.00591278076172], [39.71849060058594, -105.00570678710938], [39.7183723449707, -105.00556182861328], [39.71826171875, -105.00541687011719], [39.71812057495117, -105.00521850585938], [39.71792221069336, -105.00492858886719], [39.71759796142578, -105.00445556640625], [39.717166900634766, -105.0038070678711], [39.716880798339844, -105.00337982177734], [39.71671676635742, -105.00314331054688], [39.716552734375, -105.00292205810547], [39.7164306640625, -105.00276184082031], [39.71633529663086, -105.00264739990234], [39.71624755859375, -105.00254821777344], [39.71617889404297, -105.00247192382812], [39.71608352661133, -105.00237274169922], [39.715999603271484, -105.00228118896484], [39.71593475341797, -105.0022201538086], [39.715885162353516, -105.0021743774414], [39.715755462646484, -105.00206756591797], [39.715660095214844, -105.00198364257812], [39.715538024902344, -105.00189208984375], [39.71543884277344, -105.00181579589844], [39.71533203125, -105.00173950195312], [39.7152214050293, -105.00166320800781], [39.71499252319336, -105.00151824951172], [39.714324951171875, -105.0010757446289], [39.71399688720703, -105.00086212158203], [39.71346664428711, -105.00052642822266], [39.71318435668945, -105.00035095214844], [39.71266174316406, -105.00003051757812], [39.71223449707031, -104.9997787475586], [39.71159362792969, -104.99940490722656], [39.71141052246094, -104.99930572509766], [39.710933685302734, -104.99901580810547], [39.71067810058594, -104.99886322021484], [39.710357666015625, -104.99867248535156], [39.71013641357422, -104.99855041503906], [39.70988464355469, -104.99840545654297], [39.70966720581055, -104.99829864501953], [39.7094841003418, -104.99820709228516], [39.70940017700195, -104.99816131591797], [39.709136962890625, -104.99803924560547], [39.708702087402344, -104.99784088134766], [39.70827865600586, -104.99764251708984], [39.7078857421875, -104.99746704101562], [39.707275390625, -104.9971923828125], [39.70710372924805, -104.99711608886719], [39.7068977355957, -104.99701690673828], [39.706809997558594, -104.99697875976562], [39.706729888916016, -104.99693298339844], [39.70663070678711, -104.99688720703125], [39.706539154052734, -104.99683380126953], [39.70646667480469, -104.99678802490234], [39.70637130737305, -104.9967269897461], [39.70630645751953, -104.99668884277344], [39.70623779296875, -104.99664306640625], [39.70616912841797, -104.99658966064453], [39.706085205078125, -104.99652862548828], [39.70596694946289, -104.9964370727539], [39.705814361572266, -104.9963150024414], [39.7056999206543, -104.9962158203125], [39.70561218261719, -104.99613189697266], [39.70552062988281, -104.99604034423828], [39.705467224121094, -104.99598693847656], [39.70537567138672, -104.99588775634766], [39.705299377441406, -104.99580383300781], [39.705169677734375, -104.99565124511719], [39.70507049560547, -104.99552917480469], [39.704959869384766, -104.9953842163086], [39.704837799072266, -104.99520874023438], [39.70473098754883, -104.99505615234375], [39.704627990722656, -104.99488067626953], [39.704551696777344, -104.99474334716797], [39.704471588134766, -104.99459838867188], [39.70440673828125, -104.99447631835938], [39.7043342590332, -104.99433135986328], [39.704261779785156, -104.99417877197266], [39.70420455932617, -104.99404907226562], [39.70415115356445, -104.99392700195312], [39.704097747802734, -104.99378204345703], [39.70404815673828, -104.99363708496094], [39.70399475097656, -104.99347686767578], [39.70393753051758, -104.9933090209961], [39.70388412475586, -104.99313354492188], [39.703792572021484, -104.99284362792969], [39.70363235473633, -104.9923324584961], [39.703487396240234, -104.99188232421875], [39.703304290771484, -104.99127197265625], [39.703189849853516, -104.99092102050781], [39.70305633544922, -104.990478515625], [39.702980041503906, -104.99024200439453], [39.70287322998047, -104.989990234375], [39.70277404785156, -104.98977661132812], [39.70265197753906, -104.98954010009766], [39.70252990722656, -104.98933410644531], [39.70241928100586, -104.98914337158203], [39.70228576660156, -104.98894500732422], [39.702110290527344, -104.98870849609375], [39.70179748535156, -104.9883041381836], [39.70148849487305, -104.98790740966797], [39.700843811035156, -104.987060546875], [39.700706481933594, -104.98689270019531], [39.70027542114258, -104.98641967773438], [39.700077056884766, -104.9861831665039], [39.69994354248047, -104.98603057861328], [39.699771881103516, -104.98583984375], [39.69960021972656, -104.98567962646484], [39.699462890625, -104.98554229736328], [39.699344635009766, -104.98542022705078], [39.69913101196289, -104.98521423339844], [39.698974609375, -104.98504638671875], [39.69866180419922, -104.9847412109375], [39.6981086730957, -104.98410034179688], [39.69713592529297, -104.98299407958984], [39.696075439453125, -104.98184967041016], [39.69560623168945, -104.9813003540039], [39.69524383544922, -104.98087310791016], [39.69488525390625, -104.98043823242188], [39.694786071777344, -104.98033905029297], [39.6947021484375, -104.98025512695312], [39.69462585449219, -104.98019409179688], [39.6945915222168, -104.98016357421875], [39.69448471069336, -104.98009490966797], [39.69437789916992, -104.98002624511719], [39.6942024230957, -104.97992706298828], [39.69398498535156, -104.97977447509766], [39.69377899169922, -104.97953033447266], [39.69367599487305, -104.97943115234375], [39.69364929199219, -104.97940063476562], [39.69361114501953, -104.97936248779297], [39.69355010986328, -104.97931671142578], [39.6934814453125, -104.9792709350586], [39.69330978393555, -104.97918701171875], [39.69325637817383, -104.97914123535156], [39.693199157714844, -104.97909545898438], [39.693145751953125, -104.97903442382812], [39.693092346191406, -104.97897338867188], [39.69303512573242, -104.97886657714844], [39.69300079345703, -104.97877502441406], [39.69298553466797, -104.97871398925781], [39.69298553466797, -104.97850036621094], [39.69298553466797, -104.97827911376953], [39.69298553466797, -104.97792053222656], [39.6929931640625, -104.9774398803711], [39.6929931640625, -104.9773941040039], [39.6929931640625, -104.9772720336914], [39.692989349365234, -104.97704315185547], [39.692989349365234, -104.97694396972656], [39.692989349365234, -104.97685241699219], [39.69298553466797, -104.97636413574219], [39.6929817199707, -104.97577667236328]],\n", " {"bubblingMouseEvents": true, "color": "#eff821", "dashArray": null, "dashOffset": null, "fill": false, "fillColor": "#eff821", "fillOpacity": 0.2, "fillRule": "evenodd", "lineCap": "round", "lineJoin": "round", "noClip": false, "opacity": 0.8, "smoothFactor": 1.0, "stroke": true, "weight": 10}\n", - " ).addTo(map_63d14900899a70183befbf30d354febd);\n", + " ).addTo(map_77bceaf15f8f92685ff07972bb265db5);\n", " \n", " \n", - " poly_line_a70438f127ac39c09a7541b9864285e7.bindTooltip(\n", + " poly_line_3096e7d1c76ca60404f4c47b9117a47c.bindTooltip(\n", " `<div>\n", - " 0.3200989780246874\n", + " 0.32011129944536876\n", " </div>`,\n", " {"sticky": true}\n", " );\n", " \n", " \n", - " var marker_32cb0a5c3e87861426d1338199f84c15 = L.marker(\n", + " var marker_b408369e32baf49a93b267bb1c26ab1e = L.marker(\n", " [39.77902603149414, -104.96916961669922],\n", " {}\n", - " ).addTo(map_63d14900899a70183befbf30d354febd);\n", + " ).addTo(map_77bceaf15f8f92685ff07972bb265db5);\n", " \n", " \n", - " var icon_125348ab18ac3db4155a4629ca19372d = L.AwesomeMarkers.icon(\n", + " var icon_6bdb7e8c90867a43a909c4b03b0f45ff = L.AwesomeMarkers.icon(\n", " {"extraClasses": "fa-rotate-0", "icon": "circle", "iconColor": "white", "markerColor": "green", "prefix": "fa"}\n", " );\n", - " marker_32cb0a5c3e87861426d1338199f84c15.setIcon(icon_125348ab18ac3db4155a4629ca19372d);\n", + " marker_b408369e32baf49a93b267bb1c26ab1e.setIcon(icon_6bdb7e8c90867a43a909c4b03b0f45ff);\n", " \n", " \n", - " marker_32cb0a5c3e87861426d1338199f84c15.bindTooltip(\n", + " marker_b408369e32baf49a93b267bb1c26ab1e.bindTooltip(\n", " `<div>\n", " Origin\n", " </div>`,\n", @@ -1106,19 +1114,19 @@ " );\n", " \n", " \n", - " var marker_1dd53c8b4ad0e849af3d27a233f7eb8d = L.marker(\n", + " var marker_5e58e74755fc0f7be486049c35336a3f = L.marker(\n", " [39.6929817199707, -104.97577667236328],\n", " {}\n", - " ).addTo(map_63d14900899a70183befbf30d354febd);\n", + " ).addTo(map_77bceaf15f8f92685ff07972bb265db5);\n", " \n", " \n", - " var icon_1a6efdb2b1623d9092fbc338dec339f1 = L.AwesomeMarkers.icon(\n", + " var icon_0a710de33c2e19f55db9d8a13e270ae9 = L.AwesomeMarkers.icon(\n", " {"extraClasses": "fa-rotate-0", "icon": "circle", "iconColor": "white", "markerColor": "red", "prefix": "fa"}\n", " );\n", - " marker_1dd53c8b4ad0e849af3d27a233f7eb8d.setIcon(icon_1a6efdb2b1623d9092fbc338dec339f1);\n", + " marker_5e58e74755fc0f7be486049c35336a3f.setIcon(icon_0a710de33c2e19f55db9d8a13e270ae9);\n", " \n", " \n", - " marker_1dd53c8b4ad0e849af3d27a233f7eb8d.bindTooltip(\n", + " marker_5e58e74755fc0f7be486049c35336a3f.bindTooltip(\n", " `<div>\n", " Destination\n", " </div>`,\n", @@ -1126,33 +1134,33 @@ " );\n", " \n", " \n", - " var poly_line_dc18b3d37d1d8229eeaee4b38b80d4ef = L.polyline(\n", - " [[39.77902603149414, -104.96916961669922], [39.77896499633789, -104.96910095214844], [39.77878189086914, -104.96887969970703], [39.778751373291016, -104.96884155273438], [39.778587341308594, -104.9686508178711], [39.778480529785156, -104.96852111816406], [39.77839660644531, -104.96841430664062], [39.778324127197266, -104.96851348876953], [39.778228759765625, -104.96863555908203], [39.778076171875, -104.96883392333984], [39.777626037597656, -104.96941375732422], [39.7774772644043, -104.96961212158203], [39.77703857421875, -104.97017669677734], [39.776737213134766, -104.97056579589844], [39.77666473388672, -104.97066497802734], [39.7766227722168, -104.97071838378906], [39.776371002197266, -104.97103881835938], [39.77608108520508, -104.9714126586914], [39.775978088378906, -104.97154998779297], [39.774810791015625, -104.97306060791016], [39.77475357055664, -104.97313690185547], [39.77467346191406, -104.97323608398438], [39.77459716796875, -104.97333526611328], [39.773738861083984, -104.97444915771484], [39.77363586425781, -104.97457885742188], [39.773353576660156, -104.97494506835938], [39.77303695678711, -104.97535705566406], [39.77297592163086, -104.9754409790039], [39.77287673950195, -104.97531127929688], [39.77263259887695, -104.9749984741211], [39.77257537841797, -104.97492218017578], [39.77229690551758, -104.97456359863281], [39.77223587036133, -104.9744873046875], [39.77199935913086, -104.97418212890625], [39.77191162109375, -104.97406768798828], [39.77155685424805, -104.9736099243164], [39.771270751953125, -104.97322845458984], [39.7706184387207, -104.97236633300781], [39.77056884765625, -104.97230529785156], [39.77048110961914, -104.97219848632812], [39.77042770385742, -104.9721450805664], [39.7703742980957, -104.97209167480469], [39.77033233642578, -104.97205352783203], [39.770233154296875, -104.97198486328125], [39.770118713378906, -104.9721450805664], [39.76996994018555, -104.97232818603516], [39.76978302001953, -104.97257232666016], [39.76957321166992, -104.97283172607422], [39.76926803588867, -104.97325134277344], [39.76918411254883, -104.97335815429688], [39.76904296875, -104.97335815429688], [39.76840591430664, -104.97335052490234], [39.76824188232422, -104.97335052490234], [39.7681770324707, -104.97335052490234], [39.7680549621582, -104.97335052490234], [39.7677001953125, -104.97335052490234], [39.767547607421875, -104.97335052490234], [39.7670783996582, -104.97335052490234], [39.76704788208008, -104.97335052490234], [39.766971588134766, -104.97335052490234], [39.76690673828125, -104.97335052490234], [39.76670455932617, -104.97335052490234], [39.76658248901367, -104.97335052490234], [39.766502380371094, -104.97335052490234], [39.76637649536133, -104.97335052490234], [39.766136169433594, -104.97335052490234], [39.76605987548828, -104.97335052490234], [39.765865325927734, -104.97335052490234], [39.76577377319336, -104.97335052490234], [39.76567459106445, -104.97335052490234], [39.76523208618164, -104.97335052490234], [39.76495361328125, -104.97335052490234], [39.764835357666016, -104.97335052490234], [39.7646598815918, -104.97335052490234], [39.764591217041016, -104.97335052490234], [39.7645149230957, -104.97335052490234], [39.764427185058594, -104.97335052490234], [39.76421356201172, -104.97335052490234], [39.764034271240234, -104.97335052490234], [39.76380920410156, -104.97335052490234], [39.763580322265625, -104.97335052490234], [39.76325988769531, -104.97335052490234], [39.76316833496094, -104.97339630126953], [39.763118743896484, -104.97339630126953], [39.76306915283203, -104.97335052490234], [39.76280212402344, -104.97335815429688], [39.762454986572266, -104.97335815429688], [39.76237869262695, -104.97335815429688], [39.76214599609375, -104.97335052490234], [39.762054443359375, -104.97335052490234], [39.76194381713867, -104.97335052490234], [39.76179122924805, -104.97335052490234], [39.76169967651367, -104.97335052490234], [39.76160430908203, -104.97335052490234], [39.76118087768555, -104.97335052490234], [39.76103973388672, -104.97335052490234], [39.76084518432617, -104.97335052490234], [39.76076126098633, -104.97335052490234], [39.76066589355469, -104.97335052490234], [39.760440826416016, -104.97335052490234], [39.76033401489258, -104.97335052490234], [39.759613037109375, -104.97335052490234], [39.7595100402832, -104.97335052490234], [39.75939178466797, -104.97339630126953], [39.759151458740234, -104.97339630126953], [39.75912094116211, -104.97335052490234], [39.759010314941406, -104.97335052490234], [39.758827209472656, -104.97335052490234], [39.75831604003906, -104.97335052490234], [39.75823211669922, -104.97335052490234], [39.758174896240234, -104.97335052490234], [39.75764083862305, -104.97335815429688], [39.7572021484375, -104.97335815429688], [39.75708770751953, -104.97335815429688], [39.75699234008789, -104.97335815429688], [39.75688171386719, -104.97335815429688], [39.75673294067383, -104.97335815429688], [39.75627899169922, -104.97335815429688], [39.75574493408203, -104.97335815429688], [39.75540542602539, -104.97335815429688], [39.75531005859375, -104.97335815429688], [39.75495147705078, -104.97335815429688], [39.75469970703125, -104.97335815429688], [39.754547119140625, -104.97335815429688], [39.75442886352539, -104.97335815429688], [39.753578186035156, -104.97335815429688], [39.75333786010742, -104.97335815429688], [39.75325012207031, -104.97335815429688], [39.75314712524414, -104.97335815429688], [39.75208282470703, -104.97335815429688], [39.751991271972656, -104.97335815429688], [39.751895904541016, -104.97335815429688], [39.75139236450195, -104.9733657836914], [39.75082778930664, -104.9733657836914], [39.750732421875, -104.9733657836914], [39.75063705444336, -104.9733657836914], [39.750423431396484, -104.9733657836914], [39.750282287597656, -104.9733657836914], [39.75018310546875, -104.9733657836914], [39.749576568603516, -104.9733657836914], [39.74947738647461, -104.9733657836914], [39.74938201904297, -104.9733657836914], [39.74884033203125, -104.9733657836914], [39.74821853637695, -104.97337341308594], [39.74812698364258, -104.97337341308594], [39.748016357421875, -104.97337341308594], [39.747798919677734, -104.97337341308594], [39.74757766723633, -104.97337341308594], [39.74747085571289, -104.97337341308594], [39.74736785888672, -104.97337341308594], [39.74721908569336, -104.97337341308594], [39.747066497802734, -104.97337341308594], [39.746952056884766, -104.97337341308594], [39.746864318847656, -104.97337341308594], [39.74663162231445, -104.97337341308594], [39.746551513671875, -104.97338104248047], [39.746482849121094, -104.97338104248047], [39.7460823059082, -104.97338104248047], [39.74594497680664, -104.97338104248047], [39.74583435058594, -104.97338104248047], [39.74509048461914, -104.97340393066406], [39.745052337646484, -104.97340393066406], [39.745018005371094, -104.97344207763672], [39.744972229003906, -104.97344970703125], [39.744937896728516, -104.97344970703125], [39.744834899902344, -104.97339630126953], [39.744834899902344, -104.97355651855469], [39.744834899902344, -104.97415924072266], [39.74483871459961, -104.97454833984375], [39.74483871459961, -104.97476196289062], [39.74464416503906, -104.97476196289062], [39.744606018066406, -104.97476196289062], [39.74433517456055, -104.97476959228516], [39.74381637573242, -104.97476959228516], [39.74373245239258, -104.97476959228516], [39.743377685546875, -104.97477722167969], [39.74324417114258, -104.97477722167969], [39.743133544921875, -104.97477722167969], [39.7426872253418, -104.97477722167969], [39.74175262451172, -104.97477722167969], [39.74166488647461, -104.97477722167969], [39.74156951904297, -104.97477722167969], [39.7412109375, -104.97478485107422], [39.740718841552734, -104.97479248046875], [39.74055862426758, -104.97479248046875], [39.74032211303711, -104.97479248046875], [39.74012756347656, -104.97480010986328], [39.740020751953125, -104.97480010986328], [39.740020751953125, -104.9746322631836], [39.740020751953125, -104.97451782226562], [39.74002456665039, -104.97415924072266], [39.73991775512695, -104.97415924072266], [39.73970031738281, -104.97415924072266], [39.73955535888672, -104.97415161132812], [39.739315032958984, -104.97415161132812], [39.73896789550781, -104.97415161132812], [39.738609313964844, -104.97415161132812], [39.73850631713867, -104.9741439819336], [39.73841857910156, -104.9741439819336], [39.73833465576172, -104.9741439819336], [39.736942291259766, -104.9741439819336], [39.736881256103516, -104.9741439819336], [39.73681640625, -104.9741439819336], [39.73567581176758, -104.97412109375], [39.735252380371094, -104.97411346435547], [39.735164642333984, -104.97411346435547], [39.735069274902344, -104.97411346435547], [39.73379898071289, -104.97410583496094], [39.73370361328125, -104.97410583496094], [39.733612060546875, -104.97410583496094], [39.73240280151367, -104.97410583496094], [39.732357025146484, -104.97410583496094], [39.73214340209961, -104.97410583496094], [39.7320556640625, -104.97410583496094], [39.73196029663086, -104.97410583496094], [39.731422424316406, -104.97410583496094], [39.73128890991211, -104.97410583496094], [39.730960845947266, -104.97410583496094], [39.73088455200195, -104.97410583496094], [39.73067855834961, -104.97410583496094], [39.73054504394531, -104.97410583496094], [39.7304573059082, -104.97410583496094], [39.730369567871094, -104.97410583496094], [39.73027420043945, -104.97410583496094], [39.73003387451172, -104.97410583496094], [39.729976654052734, -104.97410583496094], [39.729896545410156, -104.97410583496094], [39.72964859008789, -104.97410583496094], [39.7294807434082, -104.97410583496094], [39.729434967041016, -104.97410583496094], [39.72917175292969, -104.97410583496094], [39.72907257080078, -104.97410583496094], [39.72898864746094, -104.97410583496094], [39.727378845214844, -104.9740982055664], [39.72727966308594, -104.9740982055664], [39.727195739746094, -104.9740982055664], [39.7259635925293, -104.9740982055664], [39.72566604614258, -104.97409057617188], [39.7255859375, -104.97409057617188], [39.72549819946289, -104.97409057617188], [39.725337982177734, -104.97409057617188], [39.7252082824707, -104.97409057617188], [39.724693298339844, -104.97409057617188], [39.72453689575195, -104.97409057617188], [39.72406768798828, -104.97408294677734], [39.72374725341797, -104.97408294677734], [39.72247314453125, -104.97408294677734], [39.72218322753906, -104.97408294677734], [39.72205352783203, -104.97408294677734], [39.721946716308594, -104.97408294677734], [39.72013473510742, -104.97406768798828], [39.719642639160156, -104.97406005859375], [39.71931457519531, -104.97406005859375], [39.71918869018555, -104.97406005859375], [39.718971252441406, -104.97405242919922], [39.71888732910156, -104.97404479980469], [39.71880340576172, -104.97403717041016], [39.71875, -104.97402954101562], [39.71870422363281, -104.9740219116211], [39.718650817871094, -104.97401428222656], [39.71857452392578, -104.97399139404297], [39.718505859375, -104.97396850585938], [39.71845245361328, -104.97393798828125], [39.71842956542969, -104.97392272949219], [39.71839141845703, -104.97390747070312], [39.71834945678711, -104.97386932373047], [39.718318939208984, -104.97382354736328], [39.71824264526367, -104.97368621826172], [39.71818923950195, -104.97358703613281], [39.71816635131836, -104.97354888916016], [39.71815490722656, -104.9735336303711], [39.71811294555664, -104.97348022460938], [39.71806335449219, -104.97343444824219], [39.71802520751953, -104.97341918945312], [39.717979431152344, -104.97340393066406], [39.717952728271484, -104.97339630126953], [39.71792221069336, -104.973388671875], [39.717899322509766, -104.973388671875], [39.71767807006836, -104.973388671875], [39.716922760009766, -104.973388671875], [39.71653366088867, -104.97339630126953], [39.71645736694336, -104.97339630126953], [39.71589660644531, -104.97339630126953], [39.71546936035156, -104.97341918945312], [39.71541976928711, -104.97341918945312], [39.71538543701172, -104.97342681884766], [39.71536636352539, -104.97342681884766], [39.71527862548828, -104.97342681884766], [39.7148551940918, -104.97344970703125], [39.71479415893555, -104.97344970703125], [39.71472930908203, -104.97345733642578], [39.714656829833984, -104.97344970703125], [39.71451950073242, -104.97344207763672], [39.71427917480469, -104.9734115600586], [39.712989807128906, -104.97340393066406], [39.71291732788086, -104.97340393066406], [39.71285629272461, -104.97340393066406], [39.71247482299805, -104.97340393066406], [39.711551666259766, -104.97340393066406], [39.71137237548828, -104.97340393066406], [39.711204528808594, -104.97340393066406], [39.71110534667969, -104.97340393066406], [39.711021423339844, -104.97340393066406], [39.7109489440918, -104.97340393066406], [39.710845947265625, -104.97340393066406], [39.709381103515625, -104.973388671875], [39.70929718017578, -104.973388671875], [39.707820892333984, -104.973388671875], [39.707557678222656, -104.973388671875], [39.70749282836914, -104.973388671875], [39.707420349121094, -104.973388671875], [39.70706558227539, -104.97338104248047], [39.70566940307617, -104.97337341308594], [39.70392990112305, -104.973388671875], [39.70386505126953, -104.973388671875], [39.70379638671875, -104.973388671875], [39.70212173461914, -104.97339630126953], [39.70206069946289, -104.97339630126953], [39.70030212402344, -104.97339630126953], [39.70021057128906, -104.97340393066406], [39.700138092041016, -104.97339630126953], [39.698429107666016, -104.97340393066406], [39.69660949707031, -104.9734115600586], [39.69625473022461, -104.97341918945312], [39.694740295410156, -104.97341918945312], [39.694149017333984, -104.97343444824219], [39.69339370727539, -104.97343444824219], [39.69325256347656, -104.97343444824219], [39.69306564331055, -104.97342681884766], [39.6929817199707, -104.97342681884766], [39.6929817199707, -104.97357177734375], [39.6929817199707, -104.97403717041016], [39.6929817199707, -104.974609375], [39.69298553466797, -104.97517395019531], [39.6929817199707, -104.97577667236328]],\n", - " {"bubblingMouseEvents": true, "color": "#7e03a7", "dashArray": null, "dashOffset": null, "fill": false, "fillColor": "#7e03a7", "fillOpacity": 0.2, "fillRule": "evenodd", "lineCap": "round", "lineJoin": "round", "noClip": false, "opacity": 0.8, "smoothFactor": 1.0, "stroke": true, "weight": 10}\n", - " ).addTo(map_63d14900899a70183befbf30d354febd);\n", + " var poly_line_467195630d68d76e5593e7251fc8958f = L.polyline(\n", + " [[39.77902603149414, -104.96916961669922], [39.77896499633789, -104.96910095214844], [39.77878189086914, -104.96887969970703], [39.778751373291016, -104.96884155273438], [39.778587341308594, -104.9686508178711], [39.778480529785156, -104.96852111816406], [39.77839660644531, -104.96841430664062], [39.778324127197266, -104.96851348876953], [39.778228759765625, -104.96863555908203], [39.778076171875, -104.96883392333984], [39.777626037597656, -104.96941375732422], [39.7774772644043, -104.96961212158203], [39.77703857421875, -104.97017669677734], [39.776737213134766, -104.97056579589844], [39.77666473388672, -104.97066497802734], [39.7766227722168, -104.97071838378906], [39.776371002197266, -104.97103881835938], [39.77608108520508, -104.9714126586914], [39.775978088378906, -104.97154998779297], [39.774810791015625, -104.97306060791016], [39.77475357055664, -104.97313690185547], [39.77467346191406, -104.97323608398438], [39.77459716796875, -104.97333526611328], [39.773738861083984, -104.97444915771484], [39.77363586425781, -104.97457885742188], [39.773353576660156, -104.97494506835938], [39.77303695678711, -104.97535705566406], [39.77297592163086, -104.9754409790039], [39.77287673950195, -104.97531127929688], [39.77263259887695, -104.9749984741211], [39.77257537841797, -104.97492218017578], [39.77229690551758, -104.97456359863281], [39.77223587036133, -104.9744873046875], [39.77199935913086, -104.97418212890625], [39.77191162109375, -104.97406768798828], [39.77157974243164, -104.9736328125], [39.77155303955078, -104.9736328125], [39.77154541015625, -104.97361755371094], [39.7712516784668, -104.97325897216797], [39.77125549316406, -104.97321319580078], [39.77090072631836, -104.97274017333984], [39.7706184387207, -104.97236633300781], [39.77056884765625, -104.97230529785156], [39.77048110961914, -104.97219848632812], [39.77042770385742, -104.9721450805664], [39.7703742980957, -104.97209167480469], [39.77033233642578, -104.97205352783203], [39.770233154296875, -104.97198486328125], [39.770118713378906, -104.9721450805664], [39.76996994018555, -104.97232818603516], [39.76978302001953, -104.97257232666016], [39.76957321166992, -104.97283172607422], [39.76926803588867, -104.97325134277344], [39.76918411254883, -104.97335815429688], [39.76904296875, -104.97335815429688], [39.76840591430664, -104.97335052490234], [39.76824188232422, -104.97335052490234], [39.7681770324707, -104.97335052490234], [39.7680549621582, -104.97335052490234], [39.7677001953125, -104.97335052490234], [39.767547607421875, -104.97335052490234], [39.7670783996582, -104.97335052490234], [39.76704788208008, -104.97335052490234], [39.766971588134766, -104.97335052490234], [39.76690673828125, -104.97335052490234], [39.76670455932617, -104.97335052490234], [39.76658248901367, -104.97335052490234], [39.766502380371094, -104.97335052490234], [39.76637649536133, -104.97335052490234], [39.766136169433594, -104.97335052490234], [39.76605987548828, -104.97335052490234], [39.765865325927734, -104.97335052490234], [39.76577377319336, -104.97335052490234], [39.76567459106445, -104.97335052490234], [39.76523208618164, -104.97335052490234], [39.76495361328125, -104.97335052490234], [39.764835357666016, -104.97335052490234], [39.7646598815918, -104.97335052490234], [39.764591217041016, -104.97335052490234], [39.7645149230957, -104.97335052490234], [39.764427185058594, -104.97335052490234], [39.76421356201172, -104.97335052490234], [39.764034271240234, -104.97335052490234], [39.76380920410156, -104.97335052490234], [39.763580322265625, -104.97335052490234], [39.76325988769531, -104.97335052490234], [39.76316833496094, -104.97339630126953], [39.763118743896484, -104.97339630126953], [39.76306915283203, -104.97335052490234], [39.762874603271484, -104.97335815429688], [39.762454986572266, -104.97335815429688], [39.76237869262695, -104.97335815429688], [39.76214599609375, -104.97335052490234], [39.762054443359375, -104.97335052490234], [39.76194381713867, -104.97335052490234], [39.76179122924805, -104.97335052490234], [39.76169967651367, -104.97335052490234], [39.76160430908203, -104.97335052490234], [39.76118087768555, -104.97335052490234], [39.76103973388672, -104.97335052490234], [39.76084518432617, -104.97335052490234], [39.76076126098633, -104.97335052490234], [39.76066589355469, -104.97335052490234], [39.760440826416016, -104.97335052490234], [39.76033401489258, -104.97335052490234], [39.759613037109375, -104.97335052490234], [39.7595100402832, -104.97335052490234], [39.75939178466797, -104.97339630126953], [39.759151458740234, -104.97339630126953], [39.75912094116211, -104.97335052490234], [39.759010314941406, -104.97335052490234], [39.758827209472656, -104.97335052490234], [39.75831604003906, -104.97335052490234], [39.75823211669922, -104.97335052490234], [39.758174896240234, -104.97335052490234], [39.75764083862305, -104.97335815429688], [39.7572021484375, -104.97335815429688], [39.75708770751953, -104.97335815429688], [39.75699234008789, -104.97335815429688], [39.75688171386719, -104.97335815429688], [39.75673294067383, -104.97335815429688], [39.75627899169922, -104.97335815429688], [39.75574493408203, -104.97335815429688], [39.75540542602539, -104.97335815429688], [39.75531005859375, -104.97335815429688], [39.75495147705078, -104.97335815429688], [39.75469970703125, -104.97335815429688], [39.754547119140625, -104.97335815429688], [39.75442886352539, -104.97335815429688], [39.753578186035156, -104.97335815429688], [39.75333786010742, -104.97335815429688], [39.75325012207031, -104.97335815429688], [39.75314712524414, -104.97335815429688], [39.75208282470703, -104.97335815429688], [39.751991271972656, -104.97335815429688], [39.751895904541016, -104.97335815429688], [39.75139236450195, -104.9733657836914], [39.75082778930664, -104.9733657836914], [39.750732421875, -104.9733657836914], [39.75063705444336, -104.9733657836914], [39.750423431396484, -104.9733657836914], [39.750282287597656, -104.9733657836914], [39.75018310546875, -104.9733657836914], [39.749576568603516, -104.9733657836914], [39.74947738647461, -104.9733657836914], [39.74938201904297, -104.9733657836914], [39.74884033203125, -104.9733657836914], [39.74821853637695, -104.97337341308594], [39.74812698364258, -104.97337341308594], [39.748016357421875, -104.97337341308594], [39.747798919677734, -104.97337341308594], [39.74757766723633, -104.97337341308594], [39.74747085571289, -104.97337341308594], [39.74736785888672, -104.97337341308594], [39.74721908569336, -104.97337341308594], [39.747066497802734, -104.97337341308594], [39.746952056884766, -104.97337341308594], [39.746864318847656, -104.97337341308594], [39.74663162231445, -104.97337341308594], [39.746551513671875, -104.97338104248047], [39.746482849121094, -104.97338104248047], [39.7460823059082, -104.97338104248047], [39.74594497680664, -104.97338104248047], [39.74583435058594, -104.97338104248047], [39.74509048461914, -104.97340393066406], [39.745052337646484, -104.97340393066406], [39.745018005371094, -104.97344207763672], [39.744972229003906, -104.97344970703125], [39.744937896728516, -104.97344970703125], [39.744834899902344, -104.97339630126953], [39.744834899902344, -104.97355651855469], [39.744834899902344, -104.97415924072266], [39.74483871459961, -104.97454833984375], [39.74483871459961, -104.97476196289062], [39.74464416503906, -104.97476196289062], [39.744606018066406, -104.97476196289062], [39.74433517456055, -104.97476959228516], [39.74381637573242, -104.97476959228516], [39.74373245239258, -104.97476959228516], [39.743377685546875, -104.97477722167969], [39.74324417114258, -104.97477722167969], [39.743133544921875, -104.97477722167969], [39.7426872253418, -104.97477722167969], [39.74175262451172, -104.97477722167969], [39.74166488647461, -104.97477722167969], [39.74156951904297, -104.97477722167969], [39.7412109375, -104.97478485107422], [39.740718841552734, -104.97479248046875], [39.74055862426758, -104.97479248046875], [39.74032211303711, -104.97479248046875], [39.74012756347656, -104.97480010986328], [39.740020751953125, -104.97480010986328], [39.740020751953125, -104.9746322631836], [39.740020751953125, -104.97451782226562], [39.74002456665039, -104.97415924072266], [39.73991775512695, -104.97415924072266], [39.73970031738281, -104.97415924072266], [39.73955535888672, -104.97415161132812], [39.739315032958984, -104.97415161132812], [39.73896789550781, -104.97415161132812], [39.738609313964844, -104.97415161132812], [39.73850631713867, -104.9741439819336], [39.73841857910156, -104.9741439819336], [39.73833465576172, -104.9741439819336], [39.736942291259766, -104.9741439819336], [39.736881256103516, -104.9741439819336], [39.73681640625, -104.9741439819336], [39.73567581176758, -104.97412109375], [39.735252380371094, -104.97411346435547], [39.735164642333984, -104.97411346435547], [39.735069274902344, -104.97411346435547], [39.73379898071289, -104.97410583496094], [39.73370361328125, -104.97410583496094], [39.733612060546875, -104.97410583496094], [39.73240280151367, -104.97410583496094], [39.732357025146484, -104.97410583496094], [39.73214340209961, -104.97410583496094], [39.7320556640625, -104.97410583496094], [39.73196029663086, -104.97410583496094], [39.731422424316406, -104.97410583496094], [39.73128890991211, -104.97410583496094], [39.730960845947266, -104.97410583496094], [39.73088455200195, -104.97410583496094], [39.73067855834961, -104.97410583496094], [39.73054504394531, -104.97410583496094], [39.7304573059082, -104.97410583496094], [39.730369567871094, -104.97410583496094], [39.73027420043945, -104.97410583496094], [39.73003387451172, -104.97410583496094], [39.729976654052734, -104.97410583496094], [39.729896545410156, -104.97410583496094], [39.72964859008789, -104.97410583496094], [39.7294807434082, -104.97410583496094], [39.729434967041016, -104.97410583496094], [39.72917175292969, -104.97410583496094], [39.72907257080078, -104.97410583496094], [39.72898864746094, -104.97410583496094], [39.727378845214844, -104.9740982055664], [39.72727966308594, -104.9740982055664], [39.727195739746094, -104.9740982055664], [39.7259635925293, -104.9740982055664], [39.72566604614258, -104.97409057617188], [39.7255859375, -104.97409057617188], [39.72549819946289, -104.97409057617188], [39.725337982177734, -104.97409057617188], [39.7252082824707, -104.97409057617188], [39.724693298339844, -104.97409057617188], [39.72453689575195, -104.97409057617188], [39.72406768798828, -104.97408294677734], [39.72374725341797, -104.97408294677734], [39.72247314453125, -104.97408294677734], [39.72218322753906, -104.97408294677734], [39.72205352783203, -104.97408294677734], [39.721946716308594, -104.97408294677734], [39.72013473510742, -104.97406768798828], [39.719642639160156, -104.97406005859375], [39.71931457519531, -104.97406005859375], [39.71918869018555, -104.97406005859375], [39.718971252441406, -104.97405242919922], [39.71888732910156, -104.97404479980469], [39.71880340576172, -104.97403717041016], [39.71875, -104.97402954101562], [39.71870422363281, -104.9740219116211], [39.718650817871094, -104.97401428222656], [39.71857452392578, -104.97399139404297], [39.718505859375, -104.97396850585938], [39.71845245361328, -104.97393798828125], [39.71842956542969, -104.97392272949219], [39.71839141845703, -104.97390747070312], [39.71834945678711, -104.97386932373047], [39.718318939208984, -104.97382354736328], [39.71824264526367, -104.97368621826172], [39.71818923950195, -104.97358703613281], [39.71816635131836, -104.97354888916016], [39.71815490722656, -104.9735336303711], [39.71811294555664, -104.97348022460938], [39.71806335449219, -104.97343444824219], [39.71802520751953, -104.97341918945312], [39.717979431152344, -104.97340393066406], [39.717952728271484, -104.97339630126953], [39.71792221069336, -104.973388671875], [39.717899322509766, -104.973388671875], [39.71767807006836, -104.973388671875], [39.716922760009766, -104.973388671875], [39.71653366088867, -104.97339630126953], [39.71645736694336, -104.97339630126953], [39.71589660644531, -104.97339630126953], [39.71546936035156, -104.97341918945312], [39.71541976928711, -104.97341918945312], [39.71538543701172, -104.97342681884766], [39.71536636352539, -104.97342681884766], [39.71527862548828, -104.97342681884766], [39.7148551940918, -104.97344970703125], [39.71479415893555, -104.97344970703125], [39.71472930908203, -104.97345733642578], [39.714656829833984, -104.97344970703125], [39.71451950073242, -104.97344207763672], [39.71427917480469, -104.9734115600586], [39.712989807128906, -104.97340393066406], [39.71291732788086, -104.97340393066406], [39.71285629272461, -104.97340393066406], [39.71247482299805, -104.97340393066406], [39.711551666259766, -104.97340393066406], [39.71137237548828, -104.97340393066406], [39.711204528808594, -104.97340393066406], [39.71110534667969, -104.97340393066406], [39.711021423339844, -104.97340393066406], [39.7109489440918, -104.97340393066406], [39.710845947265625, -104.97340393066406], [39.709381103515625, -104.973388671875], [39.70929718017578, -104.973388671875], [39.707820892333984, -104.973388671875], [39.707557678222656, -104.973388671875], [39.70749282836914, -104.973388671875], [39.707420349121094, -104.973388671875], [39.70706558227539, -104.97338104248047], [39.70566940307617, -104.97337341308594], [39.70392990112305, -104.973388671875], [39.70386505126953, -104.973388671875], [39.70379638671875, -104.973388671875], [39.70212173461914, -104.97339630126953], [39.70206069946289, -104.97339630126953], [39.70030212402344, -104.97339630126953], [39.70021057128906, -104.97340393066406], [39.700138092041016, -104.97339630126953], [39.698429107666016, -104.97340393066406], [39.69660949707031, -104.9734115600586], [39.69625473022461, -104.97341918945312], [39.694740295410156, -104.97341918945312], [39.694149017333984, -104.97343444824219], [39.69339370727539, -104.97343444824219], [39.69325256347656, -104.97343444824219], [39.69306564331055, -104.97342681884766], [39.6929817199707, -104.97342681884766], [39.6929817199707, -104.97357177734375], [39.6929817199707, -104.97403717041016], [39.6929817199707, -104.974609375], [39.69298553466797, -104.97517395019531], [39.6929817199707, -104.97577667236328]],\n", + " {"bubblingMouseEvents": true, "color": "#8104a7", "dashArray": null, "dashOffset": null, "fill": false, "fillColor": "#8104a7", "fillOpacity": 0.2, "fillRule": "evenodd", "lineCap": "round", "lineJoin": "round", "noClip": false, "opacity": 0.8, "smoothFactor": 1.0, "stroke": true, "weight": 10}\n", + " ).addTo(map_77bceaf15f8f92685ff07972bb265db5);\n", " \n", " \n", - " poly_line_dc18b3d37d1d8229eeaee4b38b80d4ef.bindTooltip(\n", + " poly_line_467195630d68d76e5593e7251fc8958f.bindTooltip(\n", " `<div>\n", - " 0.28001659029997283\n", + " 0.280412929797509\n", " </div>`,\n", " {"sticky": true}\n", " );\n", " \n", " \n", - " var marker_3bdf654f47407dce4feca9fa68867609 = L.marker(\n", + " var marker_f8fb9f6234222553ddb96ee8b37c8ae1 = L.marker(\n", " [39.77902603149414, -104.96916961669922],\n", " {}\n", - " ).addTo(map_63d14900899a70183befbf30d354febd);\n", + " ).addTo(map_77bceaf15f8f92685ff07972bb265db5);\n", " \n", " \n", - " var icon_7f8da654b577358175378d695d3554c9 = L.AwesomeMarkers.icon(\n", + " var icon_727ad313e66aff1a5fd555ee887af284 = L.AwesomeMarkers.icon(\n", " {"extraClasses": "fa-rotate-0", "icon": "circle", "iconColor": "white", "markerColor": "green", "prefix": "fa"}\n", " );\n", - " marker_3bdf654f47407dce4feca9fa68867609.setIcon(icon_7f8da654b577358175378d695d3554c9);\n", + " marker_f8fb9f6234222553ddb96ee8b37c8ae1.setIcon(icon_727ad313e66aff1a5fd555ee887af284);\n", " \n", " \n", - " marker_3bdf654f47407dce4feca9fa68867609.bindTooltip(\n", + " marker_f8fb9f6234222553ddb96ee8b37c8ae1.bindTooltip(\n", " `<div>\n", " Origin\n", " </div>`,\n", @@ -1160,19 +1168,19 @@ " );\n", " \n", " \n", - " var marker_544ae1d7c714f4dfb886e68f5b2b7f9f = L.marker(\n", + " var marker_e81ff4662acbccd08209491db4657bd7 = L.marker(\n", " [39.6929817199707, -104.97577667236328],\n", " {}\n", - " ).addTo(map_63d14900899a70183befbf30d354febd);\n", + " ).addTo(map_77bceaf15f8f92685ff07972bb265db5);\n", " \n", " \n", - " var icon_2f6f228696d0f9e6b923b6575750ad35 = L.AwesomeMarkers.icon(\n", + " var icon_0081b7beb646617f32b06a6aaa6096ad = L.AwesomeMarkers.icon(\n", " {"extraClasses": "fa-rotate-0", "icon": "circle", "iconColor": "white", "markerColor": "red", "prefix": "fa"}\n", " );\n", - " marker_544ae1d7c714f4dfb886e68f5b2b7f9f.setIcon(icon_2f6f228696d0f9e6b923b6575750ad35);\n", + " marker_e81ff4662acbccd08209491db4657bd7.setIcon(icon_0081b7beb646617f32b06a6aaa6096ad);\n", " \n", " \n", - " marker_544ae1d7c714f4dfb886e68f5b2b7f9f.bindTooltip(\n", + " marker_e81ff4662acbccd08209491db4657bd7.bindTooltip(\n", " `<div>\n", " Destination\n", " </div>`,\n", @@ -1180,33 +1188,33 @@ " );\n", " \n", " \n", - " var poly_line_2d4942578ecd49eeb510a47f63676b70 = L.polyline(\n", + " var poly_line_26538eb8c516090da23fe9fc05784049 = L.polyline(\n", " [[39.77902603149414, -104.96916961669922], [39.77896499633789, -104.96910095214844], [39.77878189086914, -104.96887969970703], [39.778751373291016, -104.96884155273438], [39.778587341308594, -104.9686508178711], [39.778480529785156, -104.96852111816406], [39.77839660644531, -104.96841430664062], [39.778324127197266, -104.96851348876953], [39.778228759765625, -104.96863555908203], [39.778076171875, -104.96883392333984], [39.777626037597656, -104.96941375732422], [39.7774772644043, -104.96961212158203], [39.77703857421875, -104.97017669677734], [39.776737213134766, -104.97056579589844], [39.77666473388672, -104.97066497802734], [39.7766227722168, -104.97071838378906], [39.776371002197266, -104.97103881835938], [39.77608108520508, -104.9714126586914], [39.775978088378906, -104.97154998779297], [39.774810791015625, -104.97306060791016], [39.77475357055664, -104.97313690185547], [39.77467346191406, -104.97323608398438], [39.77459716796875, -104.97333526611328], [39.773738861083984, -104.97444915771484], [39.77363586425781, -104.97457885742188], [39.773353576660156, -104.97494506835938], [39.77303695678711, -104.97535705566406], [39.77297592163086, -104.9754409790039], [39.77290344238281, -104.97553253173828], [39.772544860839844, -104.97599029541016], [39.772335052490234, -104.97626495361328], [39.77207946777344, -104.97659301757812], [39.771976470947266, -104.97673034667969], [39.771705627441406, -104.97708129882812], [39.77125549316406, -104.97766876220703], [39.77039337158203, -104.9787826538086], [39.77033233642578, -104.9788589477539], [39.770263671875, -104.97895050048828], [39.769996643066406, -104.97929382324219], [39.76996994018555, -104.9793930053711], [39.76955032348633, -104.97993469238281], [39.769287109375, -104.98027801513672], [39.7691764831543, -104.98041534423828], [39.769039154052734, -104.98059844970703], [39.76881408691406, -104.98088836669922], [39.768733978271484, -104.98099517822266], [39.76864242553711, -104.98110961914062], [39.76858901977539, -104.9811782836914], [39.76840591430664, -104.98141479492188], [39.7678337097168, -104.98216247558594], [39.76764678955078, -104.9823989868164], [39.76701736450195, -104.98321533203125], [39.766876220703125, -104.9833984375], [39.766693115234375, -104.98363494873047], [39.766624450683594, -104.98372650146484], [39.76657485961914, -104.9837875366211], [39.76653289794922, -104.98384857177734], [39.766178131103516, -104.98429870605469], [39.766021728515625, -104.98450469970703], [39.76557159423828, -104.98507690429688], [39.76506423950195, -104.98573303222656], [39.76464080810547, -104.98627471923828], [39.764495849609375, -104.98646545410156], [39.76442337036133, -104.98655700683594], [39.764347076416016, -104.98664855957031], [39.763763427734375, -104.98738861083984], [39.76366424560547, -104.98750305175781], [39.763572692871094, -104.98760223388672], [39.76346969604492, -104.98770141601562], [39.76337814331055, -104.98778533935547], [39.76326370239258, -104.98786926269531], [39.76316452026367, -104.9879379272461], [39.76304626464844, -104.98800659179688], [39.762916564941406, -104.98807525634766], [39.762813568115234, -104.98812103271484], [39.76266860961914, -104.98816680908203], [39.76252746582031, -104.98819732666016], [39.762413024902344, -104.98820495605469], [39.762290954589844, -104.98821258544922], [39.76216125488281, -104.98821258544922], [39.762062072753906, -104.98820495605469], [39.76179885864258, -104.9881591796875], [39.76161575317383, -104.98812103271484], [39.76142501831055, -104.98808288574219], [39.76121520996094, -104.98804473876953], [39.75983428955078, -104.98776245117188], [39.75953674316406, -104.98770141601562], [39.75946807861328, -104.98763275146484], [39.75936508178711, -104.98761749267578], [39.7588996887207, -104.987548828125], [39.758426666259766, -104.98747253417969], [39.758270263671875, -104.9874496459961], [39.75812911987305, -104.98744201660156], [39.75796127319336, -104.98744201660156], [39.75767517089844, -104.9874267578125], [39.75761032104492, -104.9874267578125], [39.75736618041992, -104.98741912841797], [39.757179260253906, -104.98741149902344], [39.757102966308594, -104.98741149902344], [39.756961822509766, -104.98741149902344], [39.756805419921875, -104.9874038696289], [39.756404876708984, -104.9874038696289], [39.75601577758789, -104.9874038696289], [39.75578308105469, -104.98739624023438], [39.75563049316406, -104.98739624023438], [39.755496978759766, -104.98739624023438], [39.75507736206055, -104.98738861083984], [39.75468063354492, -104.98738861083984], [39.75453186035156, -104.98738098144531], [39.754398345947266, -104.98738098144531], [39.754364013671875, -104.98738098144531], [39.75428009033203, -104.98738861083984], [39.75388717651367, -104.98738861083984], [39.753604888916016, -104.98738861083984], [39.75343704223633, -104.98738861083984], [39.7532844543457, -104.98738861083984], [39.75323486328125, -104.98738861083984], [39.75295639038086, -104.98738861083984], [39.752933502197266, -104.98738861083984], [39.752662658691406, -104.98738861083984], [39.752262115478516, -104.98738861083984], [39.75224304199219, -104.98738098144531], [39.751827239990234, -104.98738098144531], [39.75173568725586, -104.98738098144531], [39.751583099365234, -104.98738098144531], [39.75148391723633, -104.98738098144531], [39.751014709472656, -104.98738098144531], [39.750850677490234, -104.98738098144531], [39.75082015991211, -104.98738098144531], [39.750736236572266, -104.98738098144531], [39.750389099121094, -104.98737335205078], [39.75023651123047, -104.98737335205078], [39.7501106262207, -104.98737335205078], [39.7497673034668, -104.98737335205078], [39.74958419799805, -104.98737335205078], [39.74951171875, -104.98737335205078], [39.749412536621094, -104.98737335205078], [39.74927520751953, -104.98738098144531], [39.749168395996094, -104.98738098144531], [39.74892044067383, -104.98738098144531], [39.74868392944336, -104.98738098144531], [39.74851989746094, -104.98738098144531], [39.74822235107422, -104.98738098144531], [39.74802017211914, -104.98738098144531], [39.7476692199707, -104.98738098144531], [39.74758529663086, -104.98738098144531], [39.74751281738281, -104.98738098144531], [39.7474365234375, -104.98738098144531], [39.747379302978516, -104.98738098144531], [39.74735641479492, -104.98738098144531], [39.74730682373047, -104.98738098144531], [39.74697494506836, -104.98738098144531], [39.746944427490234, -104.98738098144531], [39.74678039550781, -104.98738861083984], [39.74667739868164, -104.98738861083984], [39.74653244018555, -104.98738861083984], [39.746482849121094, -104.98738861083984], [39.74626541137695, -104.98738861083984], [39.74618148803711, -104.98738861083984], [39.74605178833008, -104.98738861083984], [39.745948791503906, -104.98738861083984], [39.74554443359375, -104.98738861083984], [39.745418548583984, -104.98738861083984], [39.74506378173828, -104.98739624023438], [39.74480438232422, -104.98739624023438], [39.74473571777344, -104.9874038696289], [39.74430847167969, -104.98739624023438], [39.743526458740234, -104.98739624023438], [39.74336242675781, -104.98739624023438], [39.74320602416992, -104.98739624023438], [39.74302291870117, -104.98739624023438], [39.74254608154297, -104.98739624023438], [39.74245834350586, -104.98739624023438], [39.74220275878906, -104.98739624023438], [39.74214172363281, -104.98738861083984], [39.74182891845703, -104.98738861083984], [39.74177932739258, -104.98738861083984], [39.7416877746582, -104.98738861083984], [39.74163818359375, -104.98738098144531], [39.74156188964844, -104.98738861083984], [39.74149703979492, -104.98738861083984], [39.741416931152344, -104.98738861083984], [39.74137496948242, -104.98738861083984], [39.74129867553711, -104.98738861083984], [39.7410888671875, -104.98738861083984], [39.74089050292969, -104.98739624023438], [39.740875244140625, -104.98739624023438], [39.74071502685547, -104.98739624023438], [39.740631103515625, -104.98739624023438], [39.740478515625, -104.98739624023438], [39.74021530151367, -104.9874038696289], [39.74014663696289, -104.9874038696289], [39.74000930786133, -104.9874038696289], [39.73992919921875, -104.9874038696289], [39.73912048339844, -104.98739624023438], [39.73855209350586, -104.98739624023438], [39.73845672607422, -104.98739624023438], [39.73835372924805, -104.98739624023438], [39.73801040649414, -104.98739624023438], [39.73710250854492, -104.98738861083984], [39.73695373535156, -104.98738861083984], [39.736900329589844, -104.98738861083984], [39.736846923828125, -104.98738861083984], [39.73676300048828, -104.98738861083984], [39.73644256591797, -104.98739624023438], [39.73640441894531, -104.98739624023438], [39.736244201660156, -104.98739624023438], [39.73619079589844, -104.98739624023438], [39.73557662963867, -104.98739624023438], [39.73536682128906, -104.98739624023438], [39.73527526855469, -104.98739624023438], [39.73518753051758, -104.98739624023438], [39.73509216308594, -104.98739624023438], [39.73466110229492, -104.98739624023438], [39.734039306640625, -104.9874038696289], [39.73394775390625, -104.98739624023438], [39.733848571777344, -104.9874038696289], [39.73375701904297, -104.9874038696289], [39.73365020751953, -104.9874038696289], [39.73337173461914, -104.9874038696289], [39.733306884765625, -104.9874038696289], [39.732810974121094, -104.9874038696289], [39.732444763183594, -104.98741149902344], [39.732147216796875, -104.98741149902344], [39.7320556640625, -104.98741149902344], [39.731964111328125, -104.98741149902344], [39.73185348510742, -104.98741149902344], [39.73183059692383, -104.98741149902344], [39.73168182373047, -104.98741149902344], [39.73122024536133, -104.98741912841797], [39.731178283691406, -104.98741912841797], [39.73095703125, -104.98741912841797], [39.73054885864258, -104.98741912841797], [39.73045349121094, -104.98741912841797], [39.73037338256836, -104.98741912841797], [39.730010986328125, -104.98741912841797], [39.729522705078125, -104.98741912841797], [39.72943115234375, -104.98741912841797], [39.72917175292969, -104.98741912841797], [39.72908020019531, -104.98741912841797], [39.7289924621582, -104.98741912841797], [39.728668212890625, -104.98741912841797], [39.72862243652344, -104.98741912841797], [39.72829055786133, -104.98741912841797], [39.727420806884766, -104.98741912841797], [39.727298736572266, -104.98741912841797], [39.72718048095703, -104.98741912841797], [39.72665786743164, -104.9874267578125], [39.72663497924805, -104.9874267578125], [39.726600646972656, -104.9874267578125], [39.726375579833984, -104.98743438720703], [39.726226806640625, -104.98744201660156], [39.726112365722656, -104.98744201660156], [39.72574996948242, -104.98745727539062], [39.72563934326172, -104.98746490478516], [39.72555923461914, -104.98746490478516], [39.725223541259766, -104.98747253417969], [39.7251091003418, -104.98747253417969], [39.724884033203125, -104.98748016357422], [39.72444152832031, -104.98748779296875], [39.72422409057617, -104.98748779296875], [39.72414779663086, -104.98748779296875], [39.72405242919922, -104.98748779296875], [39.72385025024414, -104.98748779296875], [39.72378921508789, -104.98748779296875], [39.72360610961914, -104.98748779296875], [39.72319793701172, -104.98750305175781], [39.723026275634766, -104.98751068115234], [39.722869873046875, -104.98751068115234], [39.72262191772461, -104.98751068115234], [39.72255325317383, -104.98751068115234], [39.72245788574219, -104.98751068115234], [39.72215270996094, -104.98750305175781], [39.72206497192383, -104.98750305175781], [39.72195816040039, -104.98749542236328], [39.7216911315918, -104.98748779296875], [39.72103500366211, -104.98748779296875], [39.720943450927734, -104.98748779296875], [39.720863342285156, -104.98748779296875], [39.71990966796875, -104.98748779296875], [39.71987533569336, -104.98748779296875], [39.719451904296875, -104.98748779296875], [39.7193603515625, -104.98748779296875], [39.71928787231445, -104.98748779296875], [39.71902847290039, -104.98748779296875], [39.71844482421875, -104.98748779296875], [39.71834945678711, -104.98748779296875], [39.7182731628418, -104.98748779296875], [39.71819305419922, -104.98748779296875], [39.71754455566406, -104.98748779296875], [39.717464447021484, -104.98748779296875], [39.7174186706543, -104.98748779296875], [39.716644287109375, -104.98748779296875], [39.71656799316406, -104.98748779296875], [39.71649932861328, -104.98748779296875], [39.716224670410156, -104.98749542236328], [39.71613311767578, -104.98749542236328], [39.71575927734375, -104.98750305175781], [39.71568298339844, -104.98750305175781], [39.715606689453125, -104.98750305175781], [39.71514892578125, -104.98751831054688], [39.714847564697266, -104.98753356933594], [39.71476364135742, -104.98753356933594], [39.71470260620117, -104.98753356933594], [39.713897705078125, -104.987548828125], [39.713294982910156, -104.987548828125], [39.71319580078125, -104.987548828125], [39.71306610107422, -104.98755645751953], [39.71297836303711, -104.98755645751953], [39.71291732788086, -104.98755645751953], [39.712486267089844, -104.98756408691406], [39.71209716796875, -104.9875717163086], [39.71201705932617, -104.98757934570312], [39.71174240112305, -104.98757934570312], [39.711578369140625, -104.98758697509766], [39.711280822753906, -104.98759460449219], [39.71114730834961, -104.98759460449219], [39.71101379394531, -104.98759460449219], [39.71085739135742, -104.98759460449219], [39.71025085449219, -104.98760986328125], [39.71002960205078, -104.98760986328125], [39.70989227294922, -104.98761749267578], [39.70948028564453, -104.98761749267578], [39.70942306518555, -104.98760986328125], [39.70934295654297, -104.98761749267578], [39.709259033203125, -104.98761749267578], [39.70914840698242, -104.98762512207031], [39.7091064453125, -104.98762512207031], [39.70846939086914, -104.9876480102539], [39.70835876464844, -104.9876480102539], [39.70762634277344, -104.9876480102539], [39.707542419433594, -104.9876480102539], [39.707462310791016, -104.98764038085938], [39.7066535949707, -104.98760223388672], [39.70616912841797, -104.98758697509766], [39.705841064453125, -104.9875717163086], [39.70573806762695, -104.98756408691406], [39.705665588378906, -104.98755645751953], [39.705406188964844, -104.98754119873047], [39.705223083496094, -104.98753356933594], [39.705101013183594, -104.9875259399414], [39.704872131347656, -104.9875259399414], [39.7047119140625, -104.9875259399414], [39.704627990722656, -104.98751831054688], [39.70394515991211, -104.98751068115234], [39.703880310058594, -104.98751068115234], [39.70368194580078, -104.98750305175781], [39.70335006713867, -104.98750305175781], [39.70302963256836, -104.98750305175781], [39.7028694152832, -104.98750305175781], [39.70270538330078, -104.98751831054688], [39.70259475708008, -104.9875259399414], [39.702491760253906, -104.98753356933594], [39.70227813720703, -104.987548828125], [39.7021369934082, -104.98755645751953], [39.70204162597656, -104.98756408691406], [39.700843811035156, -104.98760986328125], [39.700782775878906, -104.98749542236328], [39.70073318481445, -104.9874038696289], [39.700687408447266, -104.98731994628906], [39.700660705566406, -104.98727416992188], [39.70061111450195, -104.9871826171875], [39.70054626464844, -104.98707580566406], [39.70036315917969, -104.98675537109375], [39.70024490356445, -104.986572265625], [39.70008850097656, -104.98636627197266], [39.700042724609375, -104.98631286621094], [39.69866180419922, -104.9847412109375], [39.6981086730957, -104.98410034179688], [39.69713592529297, -104.98299407958984], [39.696075439453125, -104.98184967041016], [39.69560623168945, -104.9813003540039], [39.69524383544922, -104.98087310791016], [39.69488525390625, -104.98043823242188], [39.694786071777344, -104.98033905029297], [39.6947021484375, -104.98025512695312], [39.69462585449219, -104.98019409179688], [39.6945915222168, -104.98016357421875], [39.69448471069336, -104.98009490966797], [39.69437789916992, -104.98002624511719], [39.6942024230957, -104.97992706298828], [39.69398498535156, -104.97977447509766], [39.69377899169922, -104.97953033447266], [39.69367599487305, -104.97943115234375], [39.69364929199219, -104.97940063476562], [39.69361114501953, -104.97936248779297], [39.69355010986328, -104.97931671142578], [39.6934814453125, -104.9792709350586], [39.69330978393555, -104.97918701171875], [39.69325637817383, -104.97914123535156], [39.693199157714844, -104.97909545898438], [39.693145751953125, -104.97903442382812], [39.693092346191406, -104.97897338867188], [39.69303512573242, -104.97886657714844], [39.69300079345703, -104.97877502441406], [39.69298553466797, -104.97871398925781], [39.69298553466797, -104.97850036621094], [39.69298553466797, -104.97827911376953], [39.69298553466797, -104.97792053222656], [39.6929931640625, -104.9774398803711], [39.6929931640625, -104.9773941040039], [39.6929931640625, -104.9772720336914], [39.692989349365234, -104.97704315185547], [39.692989349365234, -104.97694396972656], [39.692989349365234, -104.97685241699219], [39.69298553466797, -104.97636413574219], [39.6929817199707, -104.97577667236328]],\n", " {"bubblingMouseEvents": true, "color": "#0c0786", "dashArray": null, "dashOffset": null, "fill": false, "fillColor": "#0c0786", "fillOpacity": 0.2, "fillRule": "evenodd", "lineCap": "round", "lineJoin": "round", "noClip": false, "opacity": 0.8, "smoothFactor": 1.0, "stroke": true, "weight": 10}\n", - " ).addTo(map_63d14900899a70183befbf30d354febd);\n", + " ).addTo(map_77bceaf15f8f92685ff07972bb265db5);\n", " \n", " \n", - " poly_line_2d4942578ecd49eeb510a47f63676b70.bindTooltip(\n", + " poly_line_26538eb8c516090da23fe9fc05784049.bindTooltip(\n", " `<div>\n", - " 0.26646891776649895\n", + " 0.2664853255906838\n", " </div>`,\n", " {"sticky": true}\n", " );\n", " \n", " \n", - " var marker_b9ee2d187851ba8e0a0ff1154ede17fc = L.marker(\n", + " var marker_50f1588a12c520bb262fc8c55a243907 = L.marker(\n", " [39.77902603149414, -104.96916961669922],\n", " {}\n", - " ).addTo(map_63d14900899a70183befbf30d354febd);\n", + " ).addTo(map_77bceaf15f8f92685ff07972bb265db5);\n", " \n", " \n", - " var icon_793a9186e619a2b12a3f732ea3a61bd5 = L.AwesomeMarkers.icon(\n", + " var icon_50d53abde7ef57cbeed74c98cc5cec60 = L.AwesomeMarkers.icon(\n", " {"extraClasses": "fa-rotate-0", "icon": "circle", "iconColor": "white", "markerColor": "green", "prefix": "fa"}\n", " );\n", - " marker_b9ee2d187851ba8e0a0ff1154ede17fc.setIcon(icon_793a9186e619a2b12a3f732ea3a61bd5);\n", + " marker_50f1588a12c520bb262fc8c55a243907.setIcon(icon_50d53abde7ef57cbeed74c98cc5cec60);\n", " \n", " \n", - " marker_b9ee2d187851ba8e0a0ff1154ede17fc.bindTooltip(\n", + " marker_50f1588a12c520bb262fc8c55a243907.bindTooltip(\n", " `<div>\n", " Origin\n", " </div>`,\n", @@ -1214,19 +1222,19 @@ " );\n", " \n", " \n", - " var marker_3d7b473615d4cde876195834b12934ae = L.marker(\n", + " var marker_b77a6be72ef94bcf92969d8a1c82f68f = L.marker(\n", " [39.6929817199707, -104.97577667236328],\n", " {}\n", - " ).addTo(map_63d14900899a70183befbf30d354febd);\n", + " ).addTo(map_77bceaf15f8f92685ff07972bb265db5);\n", " \n", " \n", - " var icon_c58caf8f919cb00c88bbd6461e14fecd = L.AwesomeMarkers.icon(\n", + " var icon_39a9568a303e01f77af42b73eab40e7f = L.AwesomeMarkers.icon(\n", " {"extraClasses": "fa-rotate-0", "icon": "circle", "iconColor": "white", "markerColor": "red", "prefix": "fa"}\n", " );\n", - " marker_3d7b473615d4cde876195834b12934ae.setIcon(icon_c58caf8f919cb00c88bbd6461e14fecd);\n", + " marker_b77a6be72ef94bcf92969d8a1c82f68f.setIcon(icon_39a9568a303e01f77af42b73eab40e7f);\n", " \n", " \n", - " marker_3d7b473615d4cde876195834b12934ae.bindTooltip(\n", + " marker_b77a6be72ef94bcf92969d8a1c82f68f.bindTooltip(\n", " `<div>\n", " Destination\n", " </div>`,\n", @@ -1237,10 +1245,10 @@ "</html>\" style=\"position:absolute;width:100%;height:100%;left:0;top:0;border:none !important;\" allowfullscreen webkitallowfullscreen mozallowfullscreen></iframe></div></div>" ], "text/plain": [ - "<folium.folium.Map at 0x1272b78d0>" + "<folium.folium.Map at 0x1a655fe20>" ] }, - "execution_count": 188, + "execution_count": 18, "metadata": {}, "output_type": "execute_result" } @@ -1488,7 +1496,7 @@ "name": "python", "nbconvert_exporter": "python", "pygments_lexer": "ipython3", - "version": "3.11.5" + "version": "3.9.15" } }, "nbformat": 4, diff --git a/python/nrel/routee/compass/resources/osm_default_distance.toml b/python/nrel/routee/compass/resources/osm_default_distance.toml index d93087b6..5e0fe982 100644 --- a/python/nrel/routee/compass/resources/osm_default_distance.toml +++ b/python/nrel/routee/compass/resources/osm_default_distance.toml @@ -30,10 +30,10 @@ sharp_left = 3.5 u_turn = 9.5 # based on 65.5 cents per mile 2023 IRS mileage rate, $/mile -[cost.vehicle_state_variable_rates.distance] +[cost.vehicle_rates.distance] type = "factor" factor = 0.655 -[cost.state_variable_coefficients] +[cost.weights] distance = 1 [plugin] diff --git a/python/nrel/routee/compass/resources/osm_default_energy.toml b/python/nrel/routee/compass/resources/osm_default_energy.toml index bf960dfc..6f602e33 100644 --- a/python/nrel/routee/compass/resources/osm_default_energy.toml +++ b/python/nrel/routee/compass/resources/osm_default_energy.toml @@ -24,26 +24,26 @@ output_plugins = [ { type = "uuid", uuid_input_file = "vertices-uuid-enumerated.txt.gz" }, ] # based on 65.5 cents per mile 2023 IRS mileage rate, $/mile -[cost.vehicle_state_variable_rates.distance] +[cost.vehicle_rates.distance] type = "factor" factor = 0.655 # based on $20/hr approximation of 2023 median hourly wages, $/second -[cost.vehicle_state_variable_rates.time] +[cost.vehicle_rates.time] type = "factor" factor = 0.333336 # based on AAA regular unleaded gas prices sampled 12/21/2023 -[cost.vehicle_state_variable_rates.energy_liquid] +[cost.vehicle_rates.energy_liquid] type = "factor" factor = 3.120 # based on $0.50/kWh approximation of DCFC charge rates, $/kWhtype = "factor" -[cost.vehicle_state_variable_rates.energy_electric] +[cost.vehicle_rates.energy_electric] type = "factor" factor = 0.50 -[cost.state_variable_coefficients] +[cost.weights] distance = 1 time = 1 energy_liquid = 1 diff --git a/python/nrel/routee/compass/resources/osm_default_speed.toml b/python/nrel/routee/compass/resources/osm_default_speed.toml index 3a3203bb..16a96e07 100644 --- a/python/nrel/routee/compass/resources/osm_default_speed.toml +++ b/python/nrel/routee/compass/resources/osm_default_speed.toml @@ -34,16 +34,16 @@ sharp_left = 3.5 u_turn = 9.5 # based on 65.5 cents per mile 2023 IRS mileage rate, $/mile -[cost.vehicle_state_variable_rates.distance] +[cost.vehicle_rates.distance] type = "factor" factor = 0.655 # based on $20/hr approximation of 2023 median hourly wages, $/second -[cost.vehicle_state_variable_rates.time] +[cost.vehicle_rates.time] type = "factor" factor = 0.333336 -[cost.state_variable_coefficients] +[cost.weights] distance = 1 time = 1 diff --git a/rust/routee-compass/Cargo.toml b/rust/routee-compass/Cargo.toml index 4f606c3c..c626e814 100644 --- a/rust/routee-compass/Cargo.toml +++ b/rust/routee-compass/Cargo.toml @@ -43,3 +43,4 @@ itertools = { workspace = true } ordered-float = { workspace = true } allocative = { workspace = true } indoc = { workspace = true } +ordered_hash_map = { version = "0.4.0", features = ["serde"] } diff --git a/rust/routee-compass/src/app/compass/compass_app.rs b/rust/routee-compass/src/app/compass/compass_app.rs index fa68339b..12a0bbcf 100644 --- a/rust/routee-compass/src/app/compass/compass_app.rs +++ b/rust/routee-compass/src/app/compass/compass_app.rs @@ -1,5 +1,5 @@ use super::response::response_output_policy::ResponseOutputPolicy; -use super::response::response_writer::ResponseWriter; +use super::response::response_sink::ResponseSink; use super::{ compass_app_ops as ops, config::compass_app_builder::CompassAppBuilder, search_orientation::SearchOrientation, @@ -507,7 +507,7 @@ pub fn run_batch_with_responses( search_orientation: &SearchOrientation, output_plugins: &[Arc<dyn OutputPlugin>], search_app: &SearchApp, - response_writer: &ResponseWriter, + response_writer: &ResponseSink, pb: Arc<Mutex<Bar>>, ) -> Result<Box<dyn Iterator<Item = Value>>, CompassAppError> { let run_query_result = load_balanced_inputs @@ -516,12 +516,12 @@ pub fn run_batch_with_responses( queries .iter() .map(|q| { - let response = + let mut response = run_single_query(q, search_orientation, output_plugins, search_app)?; if let Ok(mut pb_local) = pb.lock() { let _ = pb_local.update(1); } - response_writer.write_response(&response)?; + response_writer.write_response(&mut response)?; Ok(response) }) .collect::<Result<Vec<serde_json::Value>, CompassAppError>>() @@ -542,7 +542,7 @@ pub fn run_batch_without_responses( search_orientation: &SearchOrientation, output_plugins: &[Arc<dyn OutputPlugin>], search_app: &SearchApp, - response_writer: &ResponseWriter, + response_writer: &ResponseSink, pb: Arc<Mutex<Bar>>, ) -> Result<Box<dyn Iterator<Item = Value>>, CompassAppError> { // run the computations, discard values that do not trigger an error @@ -554,11 +554,12 @@ pub fn run_batch_without_responses( // within a for loop or for_each call, and map creates more allocations. open to other ideas! let initial: Result<(), CompassAppError> = Ok(()); let _ = queries.iter().fold(initial, |_, q| { - let response = run_single_query(q, search_orientation, output_plugins, search_app)?; + let mut response = + run_single_query(q, search_orientation, output_plugins, search_app)?; if let Ok(mut pb_local) = pb.lock() { let _ = pb_local.update(1); } - response_writer.write_response(&response)?; + response_writer.write_response(&mut response)?; Ok(()) }); Ok(()) diff --git a/rust/routee-compass/src/app/compass/config.default.toml b/rust/routee-compass/src/app/compass/config.default.toml index f2d74cba..2df35edf 100644 --- a/rust/routee-compass/src/app/compass/config.default.toml +++ b/rust/routee-compass/src/app/compass/config.default.toml @@ -2,12 +2,12 @@ parallelism = 2 search_orientation = "vertex" response_persistence_policy = "persist_response_in_memory" [response_output_policy] -type = "no_output" +type = "none" # # example file output policy # response_persistence_policy = "discard_response_from_memory" # [response_output_policy] -# type = "run_batch_file_sink" +# type = "file" # filename = "output.json" # format = { type = "json", newline_delimited = true } @@ -26,7 +26,7 @@ type = "no_access_model" [cost] cost_aggregation = "sum" -network_state_variable_rates = {} +network_rates = {} [frontier] type = "no_restriction" diff --git a/rust/routee-compass/src/app/compass/config/cost_model/cost_model_builder.rs b/rust/routee-compass/src/app/compass/config/cost_model/cost_model_builder.rs index 04a3c623..dfb2f841 100644 --- a/rust/routee-compass/src/app/compass/config/cost_model/cost_model_builder.rs +++ b/rust/routee-compass/src/app/compass/config/cost_model/cost_model_builder.rs @@ -19,19 +19,19 @@ impl CostModelBuilder { ) -> Result<CostModelService, CompassConfigurationError> { let parent_key = CompassConfigurationField::Cost.to_string(); let vehicle_rates: HashMap<String, VehicleCostRate> = - config.get_config_serde(&"vehicle_state_variable_rates", &parent_key)?; + config.get_config_serde(&"vehicle_rates", &parent_key)?; let network_rates: HashMap<String, NetworkCostRate> = - config.get_config_serde(&"network_state_variable_rates", &parent_key)?; + config.get_config_serde(&"network_rates", &parent_key)?; let coefficients: HashMap<String, f64> = - config.get_config_serde(&"state_variable_coefficients", &parent_key)?; + config.get_config_serde(&"weights", &parent_key)?; let cost_aggregation: CostAggregation = config.get_config_serde(&"cost_aggregation", &parent_key)?; let model = CostModelService { - vehicle_state_variable_rates: Arc::new(vehicle_rates), - network_state_variable_rates: Arc::new(network_rates), - state_variable_coefficients: Arc::new(coefficients), + vehicle_rates: Arc::new(vehicle_rates), + network_rates: Arc::new(network_rates), + weights: Arc::new(coefficients), cost_aggregation, }; Ok(model) diff --git a/rust/routee-compass/src/app/compass/config/cost_model/cost_model_service.rs b/rust/routee-compass/src/app/compass/config/cost_model/cost_model_service.rs index d605ce22..e863a3d1 100644 --- a/rust/routee-compass/src/app/compass/config/cost_model/cost_model_service.rs +++ b/rust/routee-compass/src/app/compass/config/cost_model/cost_model_service.rs @@ -13,9 +13,9 @@ use std::{ }; pub struct CostModelService { - pub vehicle_state_variable_rates: Arc<HashMap<String, VehicleCostRate>>, - pub network_state_variable_rates: Arc<HashMap<String, NetworkCostRate>>, - pub state_variable_coefficients: Arc<HashMap<String, f64>>, + pub vehicle_rates: Arc<HashMap<String, VehicleCostRate>>, + pub network_rates: Arc<HashMap<String, NetworkCostRate>>, + pub weights: Arc<HashMap<String, f64>>, pub cost_aggregation: CostAggregation, } @@ -52,12 +52,9 @@ impl CostModelService { // at minimum, we default to the "distance" traveled. // invariant: this hashmap dictates the list of keys for all subsequent CostModel hashmaps. let weights: Arc<HashMap<String, f64>> = query - .get_config_serde_optional::<HashMap<String, f64>>( - &"state_variable_coefficients", - &"cost_model", - )? + .get_config_serde_optional::<HashMap<String, f64>>(&"weights", &"cost_model")? .map(Arc::new) - .unwrap_or(self.state_variable_coefficients.clone()); + .unwrap_or(self.weights.clone()); // // union the requested state variables with those in the existing traversal model // // load only indices that appear in coefficients object @@ -88,12 +85,12 @@ impl CostModelService { // the user can append/replace rates from the query let vehicle_rates = query .get_config_serde_optional::<HashMap<String, VehicleCostRate>>( - &"vehicle_state_variable_rates", + &"vehicle_rates", &"cost_model", ) .map(|opt_rates| match opt_rates { Some(rates) => Arc::new(rates), - None => self.vehicle_state_variable_rates.clone(), + None => self.vehicle_rates.clone(), })?; let cost_aggregation: CostAggregation = query @@ -103,7 +100,7 @@ impl CostModelService { let model = CostModel::new( weights, vehicle_rates, - self.network_state_variable_rates.clone(), + self.network_rates.clone(), cost_aggregation, state_model, ) diff --git a/rust/routee-compass/src/app/compass/response/csv/csv_mapping.rs b/rust/routee-compass/src/app/compass/response/csv/csv_mapping.rs new file mode 100644 index 00000000..f7533590 --- /dev/null +++ b/rust/routee-compass/src/app/compass/response/csv/csv_mapping.rs @@ -0,0 +1,88 @@ +use itertools::Itertools; +use serde::{Deserialize, Serialize}; +use serde_json::json; + +#[derive(Serialize, Deserialize, Clone, Debug)] +#[serde(untagged)] +pub enum CsvMapping { + Path(String), + Sum { sum: Vec<Box<CsvMapping>> }, + Optional { optional: Box<CsvMapping> }, +} + +impl CsvMapping { + // const CSV_ERR_SEPARATOR: &'static str = "-"; + + // /// applies a user-specified JSON to CSV mapping. if the mapping fails, + // /// a helpful error message is returned. + pub fn apply_mapping(&self, json: &serde_json::Value) -> Result<serde_json::Value, String> { + match self { + CsvMapping::Path(p) => { + let split_path = p.split('.').collect_vec(); + traverse(json, &split_path) + } + CsvMapping::Sum { sum } => { + let (ok, err): (Vec<_>, Vec<_>) = + sum.iter().map(|m| m.apply_mapping(json)).partition_result(); + if !err.is_empty() { + let valid = if ok.is_empty() { + String::from("") + } else { + format!(" with these valid paths: {}", ok.iter().join(", ")) + }; + let invalid = err.join(", "); + let msg = format!("unable to sum invalid paths: {}{}", invalid, valid); + Err(msg) + } else { + let (nums, num_errs): (Vec<_>, Vec<_>) = ok + .iter() + .map(|v| match v { + serde_json::Value::Null => Ok(0.0), + serde_json::Value::Number(n) => { + n.as_f64().ok_or_else(|| format!("invalid number {}", v)) + } + _ => Err(format!("expected a number, found {}", v)), + }) + .partition_result(); + if !num_errs.is_empty() { + let valid = if nums.is_empty() { + String::from("") + } else { + format!(" with these valid numbers: {}", nums.iter().join(", ")) + }; + let invalid = num_errs.join(", "); + let msg = format!("unable to sum invalid numbers: {}{}", invalid, valid); + Err(msg) + } else { + Ok(json![nums.into_iter().sum::<f64>()]) + } + } + } + CsvMapping::Optional { optional } => match optional.apply_mapping(json).ok() { + Some(value) => Ok(value), + None => Ok(serde_json::Value::Null), + }, + } + } +} + +fn traverse(value: &serde_json::Value, path: &Vec<&str>) -> Result<serde_json::Value, String> { + let mut cursor = value; + let mut remaining = path.as_slice(); + while let Some(next) = remaining.first() { + match cursor.get(*next) { + None => { + let path_str = path.iter().join("."); + return Err(format!( + "could not find object {} in path {}", + next, path_str + )); + } + Some(child) => { + cursor = child; + remaining = &remaining[1..]; + } + } + } + Ok(cursor.clone()) +} diff --git a/rust/routee-compass/src/app/compass/response/csv/mod.rs b/rust/routee-compass/src/app/compass/response/csv/mod.rs new file mode 100644 index 00000000..245298cc --- /dev/null +++ b/rust/routee-compass/src/app/compass/response/csv/mod.rs @@ -0,0 +1,1 @@ +pub mod csv_mapping; diff --git a/rust/routee-compass/src/app/compass/response/mod.rs b/rust/routee-compass/src/app/compass/response/mod.rs index 3f0300cf..208500b2 100644 --- a/rust/routee-compass/src/app/compass/response/mod.rs +++ b/rust/routee-compass/src/app/compass/response/mod.rs @@ -1,5 +1,6 @@ +pub mod csv; pub mod response_output_format; pub mod response_output_format_json; pub mod response_output_policy; pub mod response_persistence_policy; -pub mod response_writer; +pub mod response_sink; diff --git a/rust/routee-compass/src/app/compass/response/response_output_format.rs b/rust/routee-compass/src/app/compass/response/response_output_format.rs index 521b9e7a..1a683972 100644 --- a/rust/routee-compass/src/app/compass/response/response_output_format.rs +++ b/rust/routee-compass/src/app/compass/response/response_output_format.rs @@ -1,11 +1,24 @@ -use super::response_output_format_json as json_ops; +use super::{csv::csv_mapping::CsvMapping, response_output_format_json as json_ops}; use crate::app::compass::compass_app_error::CompassAppError; +use itertools::Itertools; +use ordered_hash_map::OrderedHashMap; use serde::{Deserialize, Serialize}; +use serde_json::json; +use std::collections::HashMap; -#[derive(Deserialize, Serialize, Debug, Clone, Copy)] +#[derive(Deserialize, Serialize, Debug, Clone)] #[serde(rename_all = "snake_case", tag = "type")] pub enum ResponseOutputFormat { + /// writes outputs to a JSON file as either ECMA-404 JSON or as newline-delimited + /// JSON. Json { newline_delimited: bool }, + /// writes outputs to a CSV file given the specified mapping. by default, column + /// order matches the order of keys in the map, unless "sorted" is true, in which + /// case the fields are sorted lexicagraphically. + Csv { + mapping: OrderedHashMap<String, CsvMapping>, + sorted: bool, + }, } impl ResponseOutputFormat { @@ -14,6 +27,14 @@ impl ResponseOutputFormat { ResponseOutputFormat::Json { newline_delimited } => { json_ops::initial_file_contents(*newline_delimited) } + ResponseOutputFormat::Csv { mapping, sorted } => { + let header = if *sorted { + mapping.keys().sorted().join(",") + } else { + mapping.keys().rev().join(",") + }; + Some(format!("{}\n", header)) + } } } @@ -22,14 +43,54 @@ impl ResponseOutputFormat { ResponseOutputFormat::Json { newline_delimited } => { json_ops::final_file_contents(*newline_delimited) } + ResponseOutputFormat::Csv { + mapping: _, + sorted: _, + } => None, } } - pub fn format_response(&self, response: &serde_json::Value) -> Result<String, CompassAppError> { + pub fn format_response( + &self, + response: &mut serde_json::Value, + ) -> Result<String, CompassAppError> { match self { ResponseOutputFormat::Json { newline_delimited } => { json_ops::format_response(response, *newline_delimited) } + ResponseOutputFormat::Csv { mapping, sorted } => { + let mut errors: HashMap<String, String> = HashMap::new(); + let row = if *sorted { + mapping + .iter() + .sorted_by_key(|(k, _)| *k) + .map(|(k, v)| match v.apply_mapping(response) { + Ok(cell) => cell.to_string(), + Err(msg) => { + errors.insert(k.clone(), msg); + String::from("") + } + }) + .join(",") + } else { + mapping + .iter() + .rev() + .map(|(k, v)| match v.apply_mapping(response) { + Ok(cell) => cell.to_string(), + Err(msg) => { + errors.insert(k.clone(), msg); + String::from("") + } + }) + .join(",") + }; + + if !errors.is_empty() { + response["error"] = json![{"csv": json![errors]}]; + } + Ok(row) + } } } @@ -38,6 +99,10 @@ impl ResponseOutputFormat { ResponseOutputFormat::Json { newline_delimited } => { json_ops::delimiter(*newline_delimited) } + ResponseOutputFormat::Csv { + mapping: _, + sorted: _, + } => Some(String::from("\n")), } } } diff --git a/rust/routee-compass/src/app/compass/response/response_output_policy.rs b/rust/routee-compass/src/app/compass/response/response_output_policy.rs index cd64f8e7..305eef32 100644 --- a/rust/routee-compass/src/app/compass/response/response_output_policy.rs +++ b/rust/routee-compass/src/app/compass/response/response_output_policy.rs @@ -1,4 +1,4 @@ -use super::{response_output_format::ResponseOutputFormat, response_writer::ResponseWriter}; +use super::{response_output_format::ResponseOutputFormat, response_sink::ResponseSink}; use crate::app::compass::compass_app_error::CompassAppError; use serde::{Deserialize, Serialize}; use std::{ @@ -10,21 +10,24 @@ use std::{ #[derive(Deserialize, Serialize, Clone, Debug)] #[serde(rename_all = "snake_case", tag = "type")] pub enum ResponseOutputPolicy { - NoOutput, - RunBatchFileSink { + None, + File { filename: String, format: ResponseOutputFormat, }, + Combined { + policies: Vec<Box<ResponseOutputPolicy>>, + }, } impl ResponseOutputPolicy { /// creates an instance of a writer which writes responses to some destination. /// the act of building this writer may include writing initial content to some sink, /// such as a file header. - pub fn build(&self) -> Result<ResponseWriter, CompassAppError> { + pub fn build(&self) -> Result<ResponseSink, CompassAppError> { match self { - ResponseOutputPolicy::NoOutput => Ok(ResponseWriter::NoOutputWriter), - ResponseOutputPolicy::RunBatchFileSink { filename, format } => { + ResponseOutputPolicy::None => Ok(ResponseSink::None), + ResponseOutputPolicy::File { filename, format } => { let output_file_path = PathBuf::from(filename); // initialize the file @@ -39,13 +42,20 @@ impl ResponseOutputPolicy { // wrap the file in a mutex so we can share it between threads let file_shareable = Arc::new(Mutex::new(file)); - Ok(ResponseWriter::RunBatchFileSink { + Ok(ResponseSink::File { filename: filename.clone(), file: file_shareable, - format: *format, + format: format.clone(), delimiter: format.delimiter(), }) } + ResponseOutputPolicy::Combined { policies } => { + let policies = policies + .iter() + .map(|p| p.build().map(Box::new)) + .collect::<Result<Vec<_>, _>>()?; + Ok(ResponseSink::Combined(policies)) + } } } } diff --git a/rust/routee-compass/src/app/compass/response/response_writer.rs b/rust/routee-compass/src/app/compass/response/response_sink.rs similarity index 65% rename from rust/routee-compass/src/app/compass/response/response_writer.rs rename to rust/routee-compass/src/app/compass/response/response_sink.rs index 584c3205..bae4ad99 100644 --- a/rust/routee-compass/src/app/compass/response/response_writer.rs +++ b/rust/routee-compass/src/app/compass/response/response_sink.rs @@ -6,21 +6,23 @@ use std::{ sync::{Arc, Mutex}, }; -pub enum ResponseWriter { - NoOutputWriter, - RunBatchFileSink { +pub enum ResponseSink { + None, + File { filename: String, file: Arc<Mutex<File>>, format: ResponseOutputFormat, delimiter: Option<String>, }, + Combined(Vec<Box<ResponseSink>>), } -impl ResponseWriter { - pub fn write_response(&self, response: &serde_json::Value) -> Result<(), CompassAppError> { +impl ResponseSink { + /// uses a writer + pub fn write_response(&self, response: &mut serde_json::Value) -> Result<(), CompassAppError> { match self { - ResponseWriter::NoOutputWriter => Ok(()), - ResponseWriter::RunBatchFileSink { + ResponseSink::None => Ok(()), + ResponseSink::File { filename: _, file, format, @@ -34,25 +36,23 @@ impl ResponseWriter { )) })?; - // if write_delimiter { - // match delimiter { - // None => {} - // Some(delim) => writeln!(file_attained, "{}", delim) - // .map_err(CompassAppError::IOError)?, - // } - // } - let output_row = format.format_response(response)?; writeln!(file_attained, "{}", output_row).map_err(CompassAppError::IOError)?; Ok(()) } + ResponseSink::Combined(policies) => { + for policy in policies { + policy.write_response(response)?; + } + Ok(()) + } } } pub fn close(&self) -> Result<String, CompassAppError> { match self { - ResponseWriter::NoOutputWriter => Ok(String::from("")), - ResponseWriter::RunBatchFileSink { + ResponseSink::None => Ok(String::from("")), + ResponseSink::File { filename, file, format, @@ -73,6 +73,17 @@ impl ResponseWriter { Ok(filename.clone()) } + ResponseSink::Combined(policies) => { + let mut out_strs = vec![]; + for policy in policies { + let out_str = policy.close()?; + if !out_str.is_empty() { + out_strs.push(out_str); + } + } + + Ok(out_strs.join(",")) + } } } } diff --git a/rust/routee-compass/src/plugin/input/default/grid_search/plugin.rs b/rust/routee-compass/src/plugin/input/default/grid_search/plugin.rs index a21316e4..44f2317d 100644 --- a/rust/routee-compass/src/plugin/input/default/grid_search/plugin.rs +++ b/rust/routee-compass/src/plugin/input/default/grid_search/plugin.rs @@ -167,21 +167,21 @@ mod test { "grid_search":{ "model_name": ["2016_TOYOTA_Camry_4cyl_2WD","2017_CHEVROLET_Bolt"], "_ignore":[ - { "name":"d1", "state_variable_coefficients": { "distance":1, "time":0, "energy_electric":0 } }, - { "name":"t1", "state_variable_coefficients": { "distance":0, "time":1, "energy_electric":0 } }, - { "name":"e1", "state_variable_coefficients": { "distance":0, "time":0, "energy_electric":1 } } + { "name":"d1", "weights": { "distance":1, "time":0, "energy_electric":0 } }, + { "name":"t1", "weights": { "distance":0, "time":1, "energy_electric":0 } }, + { "name":"e1", "weights": { "distance":0, "time":0, "energy_electric":1 } } ] } }); let plugin = GridSearchPlugin {}; plugin.process(&mut input).unwrap(); let expected = vec![ - json![{"abc":123,"model_name":"2016_TOYOTA_Camry_4cyl_2WD","name":"d1","state_variable_coefficients":{"distance":1,"energy_electric":0,"time":0}}], - json![{"abc":123,"model_name":"2016_TOYOTA_Camry_4cyl_2WD","name":"t1","state_variable_coefficients":{"distance":0,"energy_electric":0,"time":1}}], - json![{"abc":123,"model_name":"2016_TOYOTA_Camry_4cyl_2WD","name":"e1","state_variable_coefficients":{"distance":0,"energy_electric":1,"time":0}}], - json![{"abc":123,"model_name":"2017_CHEVROLET_Bolt","name":"d1","state_variable_coefficients":{"distance":1,"energy_electric":0,"time":0}}], - json![{"abc":123,"model_name":"2017_CHEVROLET_Bolt","name":"t1","state_variable_coefficients":{"distance":0,"energy_electric":0,"time":1}}], - json![{"abc":123,"model_name":"2017_CHEVROLET_Bolt","name":"e1","state_variable_coefficients":{"distance":0,"energy_electric":1,"time":0}}], + json![{"abc":123,"model_name":"2016_TOYOTA_Camry_4cyl_2WD","name":"d1","weights":{"distance":1,"energy_electric":0,"time":0}}], + json![{"abc":123,"model_name":"2016_TOYOTA_Camry_4cyl_2WD","name":"t1","weights":{"distance":0,"energy_electric":0,"time":1}}], + json![{"abc":123,"model_name":"2016_TOYOTA_Camry_4cyl_2WD","name":"e1","weights":{"distance":0,"energy_electric":1,"time":0}}], + json![{"abc":123,"model_name":"2017_CHEVROLET_Bolt","name":"d1","weights":{"distance":1,"energy_electric":0,"time":0}}], + json![{"abc":123,"model_name":"2017_CHEVROLET_Bolt","name":"t1","weights":{"distance":0,"energy_electric":0,"time":1}}], + json![{"abc":123,"model_name":"2017_CHEVROLET_Bolt","name":"e1","weights":{"distance":0,"energy_electric":1,"time":0}}], ]; match input { serde_json::Value::Array(result) => assert_eq!(result, expected),

general CSV ResponseOutputPolicy a user should be able to specify a response output policy format targeting a CSV file. to make it general, a user should be allowed to specify the relevant fields to flatten onto the CSV row as a key/value mapping from destination keys to their dot-delimited JSON paths. for example: ```toml [response_output_policy] type = "run_batch_file_sink" filename = "mep_denver_output.csv" [response_output_policy.format] type = "csv" [response_output_policy.format.mapping] src_edge = "request.origin_edge_id" dst_edge = "request.destination_edge_id" cost = "cost.total_cost" ```

2024-03-28T16:59:48

0.0

[]

NREL/routee-compass

NREL__routee-compass-97

834e23e5b02cea4f6a7bb5c40b5d42aba458a50b

diff --git a/rust/routee-compass/src/app/compass/config/compass_app_builder.rs b/rust/routee-compass/src/app/compass/config/compass_app_builder.rs index 19c3b489..24182b5f 100644 --- a/rust/routee-compass/src/app/compass/config/compass_app_builder.rs +++ b/rust/routee-compass/src/app/compass/config/compass_app_builder.rs @@ -16,6 +16,7 @@ use super::{ use crate::plugin::{ input::{ default::{ + debug::debug_builder::DebugInputPluginBuilder, edge_rtree::edge_rtree_input_plugin_builder::EdgeRtreeInputPluginBuilder, grid_search::builder::GridSearchBuilder, inject::inject_builder::InjectPluginBuilder, load_balancer::builder::LoadBalancerBuilder, vertex_rtree::builder::VertexRTreeBuilder, @@ -150,12 +151,14 @@ impl CompassAppBuilder { let edge_rtree: Rc<dyn InputPluginBuilder> = Rc::new(EdgeRtreeInputPluginBuilder {}); let load_balancer: Rc<dyn InputPluginBuilder> = Rc::new(LoadBalancerBuilder {}); let inject: Rc<dyn InputPluginBuilder> = Rc::new(InjectPluginBuilder {}); + let debug: Rc<dyn InputPluginBuilder> = Rc::new(DebugInputPluginBuilder {}); let input_plugin_builders = HashMap::from([ (String::from("grid_search"), grid_search), (String::from("vertex_rtree"), vertex_tree), (String::from("edge_rtree"), edge_rtree), (String::from("load_balancer"), load_balancer), (String::from("inject"), inject), + (String::from("debug"), debug), ]); // Output plugin builders diff --git a/rust/routee-compass/src/plugin/input/default/debug/debug_builder.rs b/rust/routee-compass/src/plugin/input/default/debug/debug_builder.rs new file mode 100644 index 00000000..f82dad99 --- /dev/null +++ b/rust/routee-compass/src/plugin/input/default/debug/debug_builder.rs @@ -0,0 +1,20 @@ +use crate::{ + app::compass::config::{ + builders::InputPluginBuilder, compass_configuration_error::CompassConfigurationError, + }, + plugin::input::input_plugin::InputPlugin, +}; +use std::sync::Arc; + +use super::debug_plugin::DebugInputPlugin; + +pub struct DebugInputPluginBuilder {} + +impl InputPluginBuilder for DebugInputPluginBuilder { + fn build( + &self, + _parameters: &serde_json::Value, + ) -> Result<Arc<dyn InputPlugin>, CompassConfigurationError> { + Ok(Arc::new(DebugInputPlugin {})) + } +} diff --git a/rust/routee-compass/src/plugin/input/default/debug/debug_plugin.rs b/rust/routee-compass/src/plugin/input/default/debug/debug_plugin.rs new file mode 100644 index 00000000..3f2393f1 --- /dev/null +++ b/rust/routee-compass/src/plugin/input/default/debug/debug_plugin.rs @@ -0,0 +1,11 @@ +use crate::plugin::{input::input_plugin::InputPlugin, plugin_error::PluginError}; + +pub struct DebugInputPlugin {} + +impl InputPlugin for DebugInputPlugin { + fn process(&self, input: &serde_json::Value) -> Result<Vec<serde_json::Value>, PluginError> { + let string = serde_json::to_string_pretty(input).map_err(PluginError::JsonError)?; + println!("{}", string); + Ok(vec![input.clone()]) + } +} diff --git a/rust/routee-compass/src/plugin/input/default/debug/mod.rs b/rust/routee-compass/src/plugin/input/default/debug/mod.rs new file mode 100644 index 00000000..fcde563f --- /dev/null +++ b/rust/routee-compass/src/plugin/input/default/debug/mod.rs @@ -0,0 +1,2 @@ +pub mod debug_builder; +pub mod debug_plugin; diff --git a/rust/routee-compass/src/plugin/input/default/grid_search/plugin.rs b/rust/routee-compass/src/plugin/input/default/grid_search/plugin.rs index 18704bbc..51e93d92 100644 --- a/rust/routee-compass/src/plugin/input/default/grid_search/plugin.rs +++ b/rust/routee-compass/src/plugin/input/default/grid_search/plugin.rs @@ -1,7 +1,7 @@ use crate::plugin::input::input_field::InputField; use crate::plugin::input::input_json_extensions::InputJsonExtensions; use crate::plugin::input::input_plugin::InputPlugin; -use crate::plugin::plugin_error::PluginError as E; +use crate::plugin::plugin_error::PluginError; use routee_compass_core::util::multiset::MultiSet; /// Builds an input plugin that duplicates queries if array-valued fields are present @@ -9,13 +9,27 @@ use routee_compass_core::util::multiset::MultiSet; pub struct GridSearchPlugin {} impl InputPlugin for GridSearchPlugin { - fn process(&self, input: &serde_json::Value) -> Result<Vec<serde_json::Value>, E> { + fn process(&self, input: &serde_json::Value) -> Result<Vec<serde_json::Value>, PluginError> { match input.get_grid_search() { None => Ok(vec![input.clone()]), Some(grid_search_input) => { - let map = grid_search_input - .as_object() - .ok_or(E::UnexpectedQueryStructure(format!("{:?}", input)))?; + // prevent recursion due to nested grid search keys + let recurses = serde_json::to_string(grid_search_input) + .map_err(PluginError::JsonError)? + .contains("grid_search"); + if recurses { + return Err(PluginError::PluginFailed(String::from( + "grid search section cannot contain the string 'grid_search'", + ))); + } + + let map = + grid_search_input + .as_object() + .ok_or(PluginError::UnexpectedQueryStructure(format!( + "{:?}", + input + )))?; let mut keys: Vec<String> = vec![]; let mut multiset_input: Vec<Vec<serde_json::Value>> = vec![]; let mut multiset_indices: Vec<Vec<usize>> = vec![]; @@ -32,7 +46,10 @@ impl InputPlugin for GridSearchPlugin { // let remove_key = InputField::GridSearch.to_str(); let mut initial_map = input .as_object() - .ok_or(E::UnexpectedQueryStructure(format!("{:?}", input)))? + .ok_or(PluginError::UnexpectedQueryStructure(format!( + "{:?}", + input + )))? .clone(); initial_map.remove(InputField::GridSearch.to_str()); let initial = serde_json::json!(initial_map); @@ -154,4 +171,51 @@ mod test { ]; assert_eq!(result, expected); } + + #[test] + fn test_nested() { + let input = serde_json::json!({ + "abc": 123, + "grid_search":{ + "model_name": ["2016_TOYOTA_Camry_4cyl_2WD","2017_CHEVROLET_Bolt"], + "_ignore":[ + { "name":"d1", "state_variable_coefficients": { "distance":1, "time":0, "energy_electric":0 } }, + { "name":"t1", "state_variable_coefficients": { "distance":0, "time":1, "energy_electric":0 } }, + { "name":"e1", "state_variable_coefficients": { "distance":0, "time":0, "energy_electric":1 } } + ] + } + }); + let plugin = GridSearchPlugin {}; + let result = plugin + .process(&input) + .unwrap() + .iter() + .map(serde_json::to_string) + .collect::<Result<Vec<String>, serde_json::Error>>() + .unwrap(); + let expected = vec![ + String::from("{\"abc\":123,\"model_name\":\"2016_TOYOTA_Camry_4cyl_2WD\",\"name\":\"d1\",\"state_variable_coefficients\":{\"distance\":1,\"energy_electric\":0,\"time\":0}}"), + String::from("{\"abc\":123,\"model_name\":\"2016_TOYOTA_Camry_4cyl_2WD\",\"name\":\"t1\",\"state_variable_coefficients\":{\"distance\":0,\"energy_electric\":0,\"time\":1}}"), + String::from("{\"abc\":123,\"model_name\":\"2016_TOYOTA_Camry_4cyl_2WD\",\"name\":\"e1\",\"state_variable_coefficients\":{\"distance\":0,\"energy_electric\":1,\"time\":0}}"), + String::from("{\"abc\":123,\"model_name\":\"2017_CHEVROLET_Bolt\",\"name\":\"d1\",\"state_variable_coefficients\":{\"distance\":1,\"energy_electric\":0,\"time\":0}}"), + String::from("{\"abc\":123,\"model_name\":\"2017_CHEVROLET_Bolt\",\"name\":\"t1\",\"state_variable_coefficients\":{\"distance\":0,\"energy_electric\":0,\"time\":1}}"), + String::from("{\"abc\":123,\"model_name\":\"2017_CHEVROLET_Bolt\",\"name\":\"e1\",\"state_variable_coefficients\":{\"distance\":0,\"energy_electric\":1,\"time\":0}}"), + ]; + assert_eq!(result, expected); + } + + #[test] + pub fn test_handle_recursion() { + let input = serde_json::json!({ + "abc": 123, + "grid_search":{ + "grid_search": { + "foo": [ "a", "b" ] + } + } + }); + let plugin = GridSearchPlugin {}; + let result = plugin.process(&input); + assert!(result.is_err()); + } } diff --git a/rust/routee-compass/src/plugin/input/default/mod.rs b/rust/routee-compass/src/plugin/input/default/mod.rs index 37ed26c7..a6d22176 100644 --- a/rust/routee-compass/src/plugin/input/default/mod.rs +++ b/rust/routee-compass/src/plugin/input/default/mod.rs @@ -1,3 +1,4 @@ +pub mod debug; pub mod edge_rtree; pub mod grid_search; pub mod inject; diff --git a/rust/routee-compass/src/plugin/input/default/vertex_rtree/plugin.rs b/rust/routee-compass/src/plugin/input/default/vertex_rtree/plugin.rs index 6c2516db..6e3f7ae8 100644 --- a/rust/routee-compass/src/plugin/input/default/vertex_rtree/plugin.rs +++ b/rust/routee-compass/src/plugin/input/default/vertex_rtree/plugin.rs @@ -233,6 +233,7 @@ mod test { .join("plugin") .join("input") .join("default") + .join("vertex_rtree") .join("test") .join("rtree_vertices.csv"); @@ -241,6 +242,7 @@ mod test { .join("plugin") .join("input") .join("default") + .join("vertex_rtree") .join("test") .join("rtree_query.json"); let query_str = fs::read_to_string(query_filepath).unwrap();

endless loop when "grid_search" contains "grid_search" by mistake i injected a value at key "grid_search" which contained a value "grid_search" within it: ```json "grid_search": { "grid_search": { "_details": [ {}, {}, {}, {} ], } } ``` this causes an infinite loop. endless loop when "grid_search" contains "grid_search" by mistake i injected a value at key "grid_search" which contained a value "grid_search" within it: ```json "grid_search": { "grid_search": { "_details": [ {}, {}, {}, {} ], } } ``` this causes an infinite loop.

2024-01-03T23:17:11

0.0

[]

NREL/routee-compass

NREL__routee-compass-33

442f9a671677b2e59e1067d7efc23dfc9cf2e604

diff --git a/docs/notebooks/open_street_maps_example.ipynb b/docs/notebooks/open_street_maps_example.ipynb index 505fbf27..75979a9d 100644 --- a/docs/notebooks/open_street_maps_example.ipynb +++ b/docs/notebooks/open_street_maps_example.ipynb @@ -75,16 +75,7 @@ "execution_count": 3, "id": "2a99e267-0ba5-462f-83e4-e8428701597c", "metadata": {}, - "outputs": [ - { - "name": "stderr", - "output_type": "stream", - "text": [ - "/Users/nreinick/dev/miniconda3/envs/test/lib/python3.12/site-packages/osgeo/gdal.py:287: FutureWarning: Neither gdal.UseExceptions() nor gdal.DontUseExceptions() has been explicitly called. In GDAL 4.0, exceptions will be enabled by default.\n", - " warnings.warn(\n" - ] - } - ], + "outputs": [], "source": [ "generate_compass_dataset(g, output_directory=\"golden_co\", add_grade=True) " ] @@ -115,9 +106,6 @@ "name": "stdout", "output_type": "stream", "text": [ - "\n", - "\n", - "\n", "\n" ] }, @@ -125,7 +113,16 @@ "name": "stderr", "output_type": "stream", "text": [ - "uuid file: 100%|██████████| 746/746 [00:00<00:00, 3760516.50it/s]it/s]" + "uuid file: 100%|██████████| 746/746 [00:00<00:00, 4123301.00it/s]it/s]" + ] + }, + { + "name": "stdout", + "output_type": "stream", + "text": [ + "\n", + "\n", + "\n" ] } ], @@ -246,7 +243,7 @@ }, { "cell_type": "code", - "execution_count": 9, + "execution_count": 10, "id": "961c910c-63ec-4c79-9522-19f24f151215", "metadata": {}, "outputs": [ @@ -255,13 +252,13 @@ "text/plain": [ "{'distance': 4.23738036063421,\n", " 'distance_unit': 'miles',\n", - " 'energy': 0.14906186986507128,\n", - " 'energy_unit': 'gallons_gasoline',\n", " 'time': 8.501927726263151,\n", - " 'time_unit': 'minutes'}" + " 'time_unit': 'minutes',\n", + " 'vehicle': {'energy': 4.23738036063421},\n", + " 'vehicle_info': {'energy_unit': 'gallons_gasoline'}}" ] }, - "execution_count": 9, + "execution_count": 10, "metadata": {}, "output_type": "execute_result" } @@ -272,7 +269,7 @@ }, { "cell_type": "code", - "execution_count": 10, + "execution_count": 11, "id": "e2c6cb39-49aa-4697-9785-fa839dc16a86", "metadata": {}, "outputs": [ @@ -281,13 +278,13 @@ "text/plain": [ "{'distance': 4.013268489898491,\n", " 'distance_unit': 'miles',\n", - " 'energy': 0.14412252939511852,\n", - " 'energy_unit': 'gallons_gasoline',\n", " 'time': 8.978411394482656,\n", - " 'time_unit': 'minutes'}" + " 'time_unit': 'minutes',\n", + " 'vehicle': {'energy': 4.013268489898491},\n", + " 'vehicle_info': {'energy_unit': 'gallons_gasoline'}}" ] }, - "execution_count": 10, + "execution_count": 11, "metadata": {}, "output_type": "execute_result" } @@ -308,17 +305,17 @@ }, { "cell_type": "code", - "execution_count": 11, + "execution_count": 12, "id": "791e814f-333c-43f4-ae84-36b9be517548", "metadata": {}, "outputs": [ { "data": { "text/plain": [ - "dict_keys(['cost', 'request', 'route', 'route_edge_count', 'route_runtime', 'search_executed_time', 'search_runtime', 'total_runtime', 'traversal_summary', 'tree', 'tree_edge_count'])" + "dict_keys(['cost', 'request', 'route', 'route_edge_count', 'route_runtime', 'search_executed_time', 'search_runtime', 'total_runtime', 'traversal_summary', 'tree_edge_count'])" ] }, - "execution_count": 11, + "execution_count": 12, "metadata": {}, "output_type": "execute_result" } @@ -339,7 +336,7 @@ }, { "cell_type": "code", - "execution_count": 12, + "execution_count": 13, "id": "30257102-07ee-4338-9c8c-41676e1995ed", "metadata": {}, "outputs": [], @@ -357,7 +354,7 @@ }, { "cell_type": "code", - "execution_count": 13, + "execution_count": 14, "id": "1ac13aae-32fc-4c10-b888-6373b175655a", "metadata": {}, "outputs": [ @@ -378,7 +375,7 @@ " <style>html, body {width: 100%;height: 100%;margin: 0;padding: 0;}</style>\n", " <style>#map {position:absolute;top:0;bottom:0;right:0;left:0;}</style>\n", " <script src="https://cdn.jsdelivr.net/npm/[email protected]/dist/leaflet.js"></script>\n", - " <script src="https://code.jquery.com/jquery-3.7.1.min.js"></script>\n", + " <script src="https://code.jquery.com/jquery-1.12.4.min.js"></script>\n", " <script src="https://cdn.jsdelivr.net/npm/[email protected]/dist/js/bootstrap.bundle.min.js"></script>\n", " <script src="https://cdnjs.cloudflare.com/ajax/libs/Leaflet.awesome-markers/2.0.2/leaflet.awesome-markers.js"></script>\n", " <link rel="stylesheet" href="https://cdn.jsdelivr.net/npm/[email protected]/dist/leaflet.css"/>\n", @@ -391,7 +388,7 @@ " <meta name="viewport" content="width=device-width,\n", " initial-scale=1.0, maximum-scale=1.0, user-scalable=no" />\n", " <style>\n", - " #map_a2ba13c8f965257bcc81a5b9f1664801 {\n", + " #map_c7345d2ba3ee51ff5955d8b9d1dc20b6 {\n", " position: relative;\n", " width: 100.0%;\n", " height: 100.0%;\n", @@ -405,14 +402,14 @@ "<body>\n", " \n", " \n", - " <div class="folium-map" id="map_a2ba13c8f965257bcc81a5b9f1664801" ></div>\n", + " <div class="folium-map" id="map_c7345d2ba3ee51ff5955d8b9d1dc20b6" ></div>\n", " \n", "</body>\n", "<script>\n", " \n", " \n", - " var map_a2ba13c8f965257bcc81a5b9f1664801 = L.map(\n", - " "map_a2ba13c8f965257bcc81a5b9f1664801",\n", + " var map_c7345d2ba3ee51ff5955d8b9d1dc20b6 = L.map(\n", + " "map_c7345d2ba3ee51ff5955d8b9d1dc20b6",\n", " {\n", " center: [39.7465774, -105.2216276],\n", " crs: L.CRS.EPSG3857,\n", @@ -426,21 +423,19 @@ "\n", " \n", " \n", - " var tile_layer_cd7720600dfcd5d06923fef393e6de53 = L.tileLayer(\n", + " var tile_layer_cbd0b24e6b90b7467b78acb978f031a6 = L.tileLayer(\n", " "https://{s}.tile.openstreetmap.org/{z}/{x}/{y}.png",\n", " {"attribution": "Data by \\u0026copy; \\u003ca target=\\"_blank\\" href=\\"http://openstreetmap.org\\"\\u003eOpenStreetMap\\u003c/a\\u003e, under \\u003ca target=\\"_blank\\" href=\\"http://www.openstreetmap.org/copyright\\"\\u003eODbL\\u003c/a\\u003e.", "detectRetina": false, "maxNativeZoom": 18, "maxZoom": 18, "minZoom": 0, "noWrap": false, "opacity": 1, "subdomains": "abc", "tms": false}\n", - " );\n", + " ).addTo(map_c7345d2ba3ee51ff5955d8b9d1dc20b6);\n", " \n", " \n", - " tile_layer_cd7720600dfcd5d06923fef393e6de53.addTo(map_a2ba13c8f965257bcc81a5b9f1664801);\n", - " \n", - " var poly_line_ad1a043197792f1a02bcbfd31ef088d3 = L.polyline(\n", + " var poly_line_f65a6830adc436b28dd0922803db2edb = L.polyline(\n", " [[39.7273069, -105.1994169], [39.7272989, -105.1994535], [39.7272848, -105.1994868], [39.7272593, -105.1995218], [39.7272272, -105.1995461], [39.7271912, -105.1995576], [39.7271688, -105.199558], [39.7271461, -105.1995532], [39.7271142, -105.1995365], [39.7270868, -105.1995096], [39.7270657, -105.1994741], [39.7270519, -105.1994296], [39.7270483, -105.1993816], [39.727052, -105.1993539], [39.7270571, -105.1993247], [39.7270719, -105.1992916], [39.7270973, -105.1992566], [39.7271293, -105.1992323], [39.7271651, -105.1992207], [39.7271854, -105.1992226], [39.7272113, -105.199227], [39.7273496, -105.1991687], [39.7274185, -105.1991465], [39.7275326, -105.1990928], [39.7276828, -105.1990602], [39.7277455, -105.1990518], [39.7279702, -105.1990232], [39.7281558, -105.1990357], [39.7284384, -105.1990893], [39.7286639, -105.199177], [39.7288222, -105.1992773], [39.7289214, -105.199385], [39.7290292, -105.1995543], [39.7290506, -105.1995879], [39.7292157, -105.1998155], [39.7293094, -105.1998945], [39.7294079, -105.1999604], [39.7295437, -105.200014], [39.7296299, -105.2000721], [39.7297292, -105.2000757], [39.7297135, -105.2007081], [39.7297192, -105.2013106], [39.7298153, -105.2032041], [39.7298256, -105.2033784], [39.7298664, -105.2035708], [39.7298795, -105.203649], [39.7299267, -105.2038558], [39.7301082, -105.2042608], [39.7304911, -105.2048068], [39.7308741, -105.2053111], [39.7310676, -105.2055985], [39.7311933, -105.205824], [39.7312069, -105.2058543], [39.7313075, -105.2061158], [39.731348, -105.206248], [39.731386, -105.2064213], [39.7314237, -105.2067336], [39.731429, -105.2070565], [39.7314226, -105.207224], [39.7314076, -105.2074036], [39.7313611, -105.2077675], [39.7313265, -105.2079124], [39.7312831, -105.20806], [39.7312292, -105.2082086], [39.7311706, -105.208344], [39.7310965, -105.2084777], [39.7310049, -105.208616], [39.7309234, -105.2087235], [39.7308212, -105.2088303], [39.7306227, -105.2089819], [39.7304801, -105.2090736], [39.7303681, -105.2091988], [39.7303327, -105.2092597], [39.7303114, -105.2093181], [39.730299, -105.20939], [39.7302939, -105.2095126], [39.7303105, -105.2096804], [39.7304414, -105.2099267], [39.7305001, -105.2100344], [39.7305871, -105.2101842], [39.7307404, -105.2104481], [39.7309234, -105.2107229], [39.7310934, -105.2109639], [39.731238, -105.211145], [39.7314101, -105.2113464], [39.7315783, -105.2115336], [39.7317541, -105.2117064], [39.7319796, -105.2119089], [39.732359, -105.2122393], [39.7329187, -105.2127184], [39.7334557, -105.213188], [39.7338968, -105.2135676], [39.7342777, -105.2138968], [39.7347139, -105.2142786], [39.7354652, -105.2149301], [39.7356319, -105.2150733], [39.7358844, -105.2152902], [39.7363901, -105.2157275], [39.7369129, -105.2161859], [39.7377243, -105.2168831], [39.7390882, -105.2180682], [39.7397882, -105.2186718], [39.7403043, -105.2191078], [39.7405645, -105.2193336], [39.7406767, -105.2194225], [39.7415803, -105.2202142], [39.7416486, -105.2202727], [39.7417047, -105.2203151], [39.7421502, -105.2206932], [39.7425816, -105.221062], [39.7432687, -105.2215398], [39.7435797, -105.2217631], [39.7436721, -105.2218299], [39.7437366, -105.2218764], [39.7438794, -105.2219759], [39.7439546, -105.2220253], [39.7440082, -105.2220586], [39.7440819, -105.2221021], [39.7441377, -105.2221322], [39.7441808, -105.2221541], [39.7442796, -105.2222043], [39.7443357, -105.2222286], [39.7444062, -105.222259], [39.7444657, -105.2222871], [39.7444962, -105.2223029], [39.7445695, -105.2223431], [39.7446343, -105.2223808], [39.744683, -105.2224104], [39.7447215, -105.2224349], [39.7447658, -105.2224654], [39.744813, -105.222499], [39.7448547, -105.2225315], [39.7448881, -105.2225586], [39.7449328, -105.2225965], [39.7449815, -105.2226391], [39.7451542, -105.2227925], [39.7454784, -105.2230438], [39.7455034, -105.2230623], [39.7455159, -105.2230713], [39.7455403, -105.2230861], [39.7455548, -105.2230941], [39.7455766, -105.2231014], [39.7456018, -105.223106], [39.7456176, -105.2231071], [39.7456337, -105.223107], [39.7456638, -105.2231045], [39.7456814, -105.2231011], [39.7456983, -105.2230945], [39.7457224, -105.2230821], [39.7457338, -105.2230745], [39.7457588, -105.2230543], [39.7457709, -105.223043], [39.7457834, -105.2230284], [39.7458035, -105.2230027], [39.7458905, -105.2228787], [39.7460068, -105.2226624], [39.7461589, -105.2223894], [39.7463617, -105.2220126], [39.7464427, -105.2218672], [39.746482, -105.2217968], [39.7465202, -105.2217322], [39.7465523, -105.2216771], [39.7465774, -105.2216276], [39.7465928, -105.2215823], [39.7466004, -105.2215466], [39.7466042, -105.2215101], [39.7466044, -105.2214695], [39.746601, -105.2214368], [39.7465881, -105.2213862], [39.7465804, -105.2213567], [39.7465774, -105.2213321], [39.7465774, -105.2213048], [39.746581, -105.2212762], [39.7465874, -105.2212518], [39.7465978, -105.2212266], [39.7466125, -105.2212024], [39.7466266, -105.2211858], [39.7466485, -105.2211678], [39.7466708, -105.2211566], [39.746693, -105.2211511], [39.746715, -105.2211506], [39.7467203, -105.2211517], [39.7467798, -105.2211327], [39.7468101, -105.2211204], [39.7468347, -105.2211032], [39.7468673, -105.2210676], [39.7468863, -105.2210389], [39.7469079, -105.2209995], [39.7469752, -105.2208699], [39.7471073, -105.220655], [39.7471371, -105.2206129], [39.7471594, -105.2205845], [39.7472061, -105.2205376], [39.7473376, -105.2202962], [39.747357, -105.2202177], [39.7473682, -105.2201868], [39.7473848, -105.2201519], [39.7475089, -105.2199249], [39.7475861, -105.2197973], [39.7476254, -105.2197401], [39.7476917, -105.2196525], [39.7477724, -105.2195069], [39.7478875, -105.219299], [39.7479219, -105.2192372], [39.7479518, -105.219179], [39.7480957, -105.2189201], [39.7481337, -105.2188515], [39.7482567, -105.2186294], [39.7484851, -105.2182271], [39.7485167, -105.218164], [39.7485605, -105.2180881], [39.7490516, -105.217166], [39.7490955, -105.2170866], [39.7491263, -105.2170387], [39.7493736, -105.2165801], [39.7496841, -105.2160128], [39.7497482, -105.2160733], [39.7504593, -105.216753], [39.7505084, -105.2168007], [39.7505733, -105.2168567], [39.7509525, -105.2172063], [39.7512593, -105.2174884], [39.7513304, -105.2175506], [39.7513869, -105.217604], [39.7517458, -105.2179367], [39.7520945, -105.21826], [39.7521462, -105.2183047], [39.7522126, -105.218369], [39.7523012, -105.2184523], [39.7525052, -105.218642], [39.7527145, -105.2188366], [39.7528959, -105.2189993], [39.7529601, -105.2190649], [39.7530308, -105.219128], [39.7537033, -105.2197718], [39.753764, -105.2198302], [39.7538299, -105.2198915], [39.7542346, -105.2202638], [39.7545279, -105.2205433], [39.7545847, -105.2206031], [39.7546623, -105.2206728], [39.7553326, -105.2212752], [39.755407, -105.221342], [39.7554908, -105.2214211], [39.7561483, -105.2220415], [39.7562254, -105.2221143], [39.7563006, -105.2221853], [39.7566193, -105.222483], [39.7566315, -105.2224957], [39.7569475, -105.2227913], [39.7570561, -105.222895], [39.7574792, -105.2232693], [39.7575579, -105.2233409], [39.7576236, -105.2234019], [39.7578565, -105.2236199], [39.7579703, -105.2237258], [39.7580027, -105.2237561], [39.7583001, -105.2240544], [39.7583591, -105.2241084], [39.7587249, -105.2244446], [39.7588057, -105.22452], [39.7591609, -105.2248516], [39.7595089, -105.2251741], [39.7596887, -105.2253408], [39.7597554, -105.2254024], [39.7598802, -105.2255173], [39.7598939, -105.2255305], [39.7599658, -105.2255988], [39.7600186, -105.2256498], [39.7604008, -105.2259941], [39.7605066, -105.2260931], [39.7605924, -105.226174], [39.760667, -105.2262427], [39.7607946, -105.2263631], [39.7608753, -105.2264373], [39.7615829, -105.2271044], [39.7616266, -105.2271481], [39.7616788, -105.2272008], [39.762203, -105.2276943], [39.7623411, -105.2278299], [39.7630405, -105.2284565], [39.7632824, -105.2286685], [39.7637358, -105.2290679], [39.7640137, -105.229309], [39.7640954, -105.229381], [39.7641743, -105.229462], [39.76436, -105.2296495], [39.7645376, -105.2298658], [39.764824, -105.2301918], [39.7652051, -105.2307044], [39.7654711, -105.2310802], [39.7654801, -105.2310934], [39.7655308, -105.231168], [39.7655792, -105.231232], [39.7655918, -105.2312486], [39.7657719, -105.2315035], [39.7658973, -105.2316751], [39.7659962, -105.2318039], [39.7661018, -105.231924], [39.7662469, -105.2320614], [39.7664515, -105.2322588], [39.7671442, -105.2329283], [39.7673559, -105.2331244], [39.7674414, -105.2332053], [39.7675186, -105.2332845], [39.7675756, -105.2333522], [39.7676291, -105.2334243], [39.7676969, -105.2335369], [39.7677304, -105.233602], [39.767761, -105.2336649], [39.7677908, -105.233739], [39.7678172, -105.233812], [39.7678551, -105.2339422], [39.7678736, -105.2340294], [39.7678904, -105.2341257], [39.7679032, -105.2342329], [39.7679095, -105.2343359], [39.7679068, -105.2344027], [39.7679029, -105.234499], [39.7678875, -105.2346626], [39.7678651, -105.2347791], [39.7678384, -105.2348898], [39.7678006, -105.2350283], [39.7677517, -105.2352031], [39.7677119, -105.2353468]],\n", " {"bubblingMouseEvents": true, "color": "blue", "dashArray": null, "dashOffset": null, "fill": false, "fillColor": "blue", "fillOpacity": 0.2, "fillRule": "evenodd", "lineCap": "round", "lineJoin": "round", "noClip": false, "opacity": 0.8, "smoothFactor": 1.0, "stroke": true, "weight": 10}\n", - " ).addTo(map_a2ba13c8f965257bcc81a5b9f1664801);\n", + " ).addTo(map_c7345d2ba3ee51ff5955d8b9d1dc20b6);\n", " \n", " \n", - " poly_line_ad1a043197792f1a02bcbfd31ef088d3.bindTooltip(\n", + " poly_line_f65a6830adc436b28dd0922803db2edb.bindTooltip(\n", " `<div>\n", " Shortest Time\n", " </div>`,\n", @@ -448,19 +443,19 @@ " );\n", " \n", " \n", - " var marker_9aec00c6a27264acf91afd3f0996a66d = L.marker(\n", + " var marker_c509908f7e3039cb10134666c7e7930d = L.marker(\n", " [39.7273069, -105.1994169],\n", " {}\n", - " ).addTo(map_a2ba13c8f965257bcc81a5b9f1664801);\n", + " ).addTo(map_c7345d2ba3ee51ff5955d8b9d1dc20b6);\n", " \n", " \n", - " var icon_ddb004cd6627984db241463133fc8206 = L.AwesomeMarkers.icon(\n", + " var icon_fce1a97d514cb3efe68f0ab61c47881a = L.AwesomeMarkers.icon(\n", " {"extraClasses": "fa-rotate-0", "icon": "circle", "iconColor": "white", "markerColor": "green", "prefix": "fa"}\n", " );\n", - " marker_9aec00c6a27264acf91afd3f0996a66d.setIcon(icon_ddb004cd6627984db241463133fc8206);\n", + " marker_c509908f7e3039cb10134666c7e7930d.setIcon(icon_fce1a97d514cb3efe68f0ab61c47881a);\n", " \n", " \n", - " marker_9aec00c6a27264acf91afd3f0996a66d.bindTooltip(\n", + " marker_c509908f7e3039cb10134666c7e7930d.bindTooltip(\n", " `<div>\n", " Origin\n", " </div>`,\n", @@ -468,19 +463,19 @@ " );\n", " \n", " \n", - " var marker_2400252a991122b76a0f550d44f2471a = L.marker(\n", + " var marker_53fa08b69a274c2485a6a5bfe29337f8 = L.marker(\n", " [39.7677119, -105.2353468],\n", " {}\n", - " ).addTo(map_a2ba13c8f965257bcc81a5b9f1664801);\n", + " ).addTo(map_c7345d2ba3ee51ff5955d8b9d1dc20b6);\n", " \n", " \n", - " var icon_5275c70cca7a2265a72fb4c4f9fdd51d = L.AwesomeMarkers.icon(\n", + " var icon_a80a162478d0b051a78633e6cf82d5c2 = L.AwesomeMarkers.icon(\n", " {"extraClasses": "fa-rotate-0", "icon": "circle", "iconColor": "white", "markerColor": "red", "prefix": "fa"}\n", " );\n", - " marker_2400252a991122b76a0f550d44f2471a.setIcon(icon_5275c70cca7a2265a72fb4c4f9fdd51d);\n", + " marker_53fa08b69a274c2485a6a5bfe29337f8.setIcon(icon_a80a162478d0b051a78633e6cf82d5c2);\n", " \n", " \n", - " marker_2400252a991122b76a0f550d44f2471a.bindTooltip(\n", + " marker_53fa08b69a274c2485a6a5bfe29337f8.bindTooltip(\n", " `<div>\n", " Destination\n", " </div>`,\n", @@ -488,13 +483,13 @@ " );\n", " \n", " \n", - " var poly_line_f8164f878668d4d431bf9490b1c0c98b = L.polyline(\n", + " var poly_line_f7dfde4d70684478e2ba63e72b94ca5b = L.polyline(\n", " [[39.7273069, -105.1994169], [39.7272989, -105.1994535], [39.7272848, -105.1994868], [39.7272593, -105.1995218], [39.7272272, -105.1995461], [39.7271912, -105.1995576], [39.7271688, -105.199558], [39.7271461, -105.1995532], [39.7271142, -105.1995365], [39.7270868, -105.1995096], [39.7270657, -105.1994741], [39.7270519, -105.1994296], [39.7270483, -105.1993816], [39.727052, -105.1993539], [39.7270571, -105.1993247], [39.7270719, -105.1992916], [39.7270973, -105.1992566], [39.7271293, -105.1992323], [39.7271651, -105.1992207], [39.7271854, -105.1992226], [39.7272113, -105.199227], [39.7273496, -105.1991687], [39.7274185, -105.1991465], [39.7275326, -105.1990928], [39.7276828, -105.1990602], [39.7277455, -105.1990518], [39.7279702, -105.1990232], [39.7281558, -105.1990357], [39.7284384, -105.1990893], [39.7286639, -105.199177], [39.7288222, -105.1992773], [39.7289214, -105.199385], [39.7290292, -105.1995543], [39.7290506, -105.1995879], [39.7292157, -105.1998155], [39.7293094, -105.1998945], [39.7294079, -105.1999604], [39.7295437, -105.200014], [39.7296299, -105.2000721], [39.7296469, -105.1996789], [39.7296738, -105.1987284], [39.7296906, -105.1986794], [39.7297244, -105.1985856], [39.7297272, -105.1984193], [39.7299723, -105.1984198], [39.7302117, -105.1984227], [39.7303235, -105.1984285], [39.7304434, -105.198443], [39.7305651, -105.1984732], [39.7307182, -105.1985167], [39.7308487, -105.1985623], [39.7309909, -105.1986311], [39.731127, -105.1987144], [39.7312292, -105.1987897], [39.7313292, -105.1988731], [39.7314292, -105.1989611], [39.731524, -105.1990554], [39.7316089, -105.1991466], [39.7317085, -105.1992639], [39.7318575, -105.1994488], [39.7319522, -105.1995697], [39.7320423, -105.1996716], [39.7321372, -105.1997578], [39.7322615, -105.1998506], [39.7323697, -105.1999163], [39.7324712, -105.1999637], [39.7325898, -105.2000081], [39.732689, -105.2000309], [39.7327987, -105.2000627], [39.732945, -105.2001049], [39.7330635, -105.2001374], [39.7331838, -105.2001914], [39.7332899, -105.2002528], [39.7334312, -105.200363], [39.7335257, -105.2004544], [39.7335523, -105.2004835], [39.7336132, -105.2005503], [39.7336902, -105.2006555], [39.733767, -105.2007557], [39.7338411, -105.2008491], [39.7339359, -105.2009412], [39.7340357, -105.2010154], [39.7340742, -105.201042], [39.7341117, -105.2010678], [39.7341542, -105.2010901], [39.734211, -105.2011198], [39.7343173, -105.2011712], [39.7344056, -105.2012021], [39.7345704, -105.2012376], [39.7347726, -105.2012753], [39.7351329, -105.2013311], [39.7352203, -105.2013441], [39.7353243, -105.201352], [39.7354415, -105.2013441], [39.7355491, -105.2013268], [39.7357022, -105.2012901], [39.7358395, -105.201259], [39.7362429, -105.2011675], [39.7367157, -105.2010639], [39.7367857, -105.2010539], [39.7368531, -105.2010442], [39.7369645, -105.2010356], [39.7370755, -105.201044], [39.7371895, -105.2010632], [39.7373179, -105.201099], [39.737714, -105.2012295], [39.7379567, -105.2013009], [39.7384451, -105.2014404], [39.7386872, -105.2015167], [39.7395717, -105.2017905], [39.7397153, -105.2018369], [39.7398269, -105.2018673], [39.7399356, -105.2018843], [39.7400618, -105.2018961], [39.740184, -105.2018932], [39.7403133, -105.2018823], [39.7404288, -105.2018645], [39.7405518, -105.2018387], [39.740665, -105.2018073], [39.7407597, -105.2017713], [39.7408728, -105.2017196], [39.740992, -105.2016551], [39.7411268, -105.2015625], [39.7412599, -105.2014489], [39.7413718, -105.2013453], [39.7414758, -105.2012326], [39.741608, -105.2010691], [39.7416901, -105.200964], [39.7417903, -105.2008396], [39.7417994, -105.200826], [39.7419078, -105.2006985], [39.7419027, -105.2006189], [39.7419122, -105.200549], [39.7419496, -105.2004812], [39.7419894, -105.20044], [39.742047, -105.2004033], [39.7420816, -105.2004], [39.7421282, -105.2004131], [39.7421839, -105.200448], [39.7422234, -105.2004924], [39.7422448, -105.2005578], [39.7422566, -105.2006096], [39.7422556, -105.2006625], [39.7422421, -105.2007231], [39.7422712, -105.2008198], [39.7423227, -105.200901], [39.7428069, -105.2016549], [39.7429048, -105.2018114], [39.7429131, -105.2018245], [39.7431503, -105.2022035], [39.7434374, -105.2026477], [39.7435807, -105.2028659], [39.7437705, -105.2031843], [39.7441255, -105.2037193], [39.7444536, -105.2042289], [39.7445369, -105.204369], [39.7448087, -105.2048085], [39.7450389, -105.205197], [39.745164, -105.2054397], [39.7452617, -105.205642], [39.745359, -105.2058753], [39.7454413, -105.2060693], [39.7454698, -105.2061365], [39.7455586, -105.2063498], [39.7456394, -105.2066197], [39.7457343, -105.2068984], [39.7458096, -105.2071619], [39.7458432, -105.2072893], [39.7459057, -105.2075162], [39.7459547, -105.2077463], [39.7460104, -105.2080066], [39.7460312, -105.2081149], [39.7460508, -105.2082709], [39.7460836, -105.2084787], [39.7461092, -105.2086125], [39.7461165, -105.2087078], [39.7461128, -105.2088019], [39.7461078, -105.2089177], [39.7461163, -105.208995], [39.7461377, -105.2090672], [39.7461588, -105.2091147], [39.7462509, -105.2092159], [39.746281, -105.2092382], [39.7462978, -105.209256], [39.7463242, -105.209292], [39.7463458, -105.2093473], [39.746361, -105.2094251], [39.7463921, -105.2095768], [39.7464353, -105.2097133], [39.7465001, -105.2098285], [39.7467097, -105.21004], [39.7467814, -105.2101094], [39.7473357, -105.2106047], [39.7474554, -105.2107116], [39.7475615, -105.2108064], [39.7475958, -105.2108371], [39.747653, -105.2108907], [39.7481502, -105.2113567], [39.7483361, -105.2115309], [39.7483726, -105.2115651], [39.7484183, -105.2116079], [39.7484718, -105.2116584], [39.7487212, -105.2118951], [39.749238, -105.2123721], [39.7492814, -105.2124135], [39.7500059, -105.2130759], [39.7500454, -105.2131129], [39.7503812, -105.2134256], [39.7504627, -105.2135015], [39.7508219, -105.2138362], [39.7508792, -105.2138895], [39.7509348, -105.2139426], [39.7511093, -105.2141093], [39.7516271, -105.2146038], [39.7516716, -105.2146463], [39.751739, -105.2147085], [39.7524829, -105.2153953], [39.7525012, -105.2154123], [39.7532517, -105.2161101], [39.7533052, -105.2161598], [39.7540716, -105.2168674], [39.7541308, -105.2169221], [39.7544795, -105.2172508], [39.7546281, -105.2174522], [39.7546745, -105.2175037], [39.7548394, -105.2176869], [39.7549112, -105.2177871], [39.7548521, -105.2179433], [39.7546476, -105.2183047], [39.7544379, -105.2186805], [39.7544293, -105.218688], [39.7543723, -105.2187394], [39.7544376, -105.2188002], [39.7544872, -105.2188464], [39.7547934, -105.2191313], [39.7551071, -105.2194077], [39.7551368, -105.2194339], [39.755202, -105.2194913], [39.7552559, -105.2195411], [39.755615, -105.2198733], [39.755696, -105.2199479], [39.7559351, -105.2201693], [39.7559559, -105.220191], [39.7560109, -105.2202486], [39.7560703, -105.2203182], [39.756257, -105.2205368], [39.7562813, -105.2205598], [39.7564057, -105.2206774], [39.7565461, -105.2208058], [39.7567417, -105.2209442], [39.7566986, -105.2211001], [39.756638, -105.2213112], [39.7565732, -105.2214643], [39.7565209, -105.2215678], [39.7564595, -105.2216814], [39.7562914, -105.2219923], [39.7562254, -105.2221143], [39.7563006, -105.2221853], [39.7566193, -105.222483], [39.7566315, -105.2224957], [39.7569475, -105.2227913], [39.7570561, -105.222895], [39.7574792, -105.2232693], [39.7575579, -105.2233409], [39.7576236, -105.2234019], [39.7578565, -105.2236199], [39.7579703, -105.2237258], [39.7580027, -105.2237561], [39.7583001, -105.2240544], [39.7583591, -105.2241084], [39.7587249, -105.2244446], [39.7588057, -105.22452], [39.7591609, -105.2248516], [39.7595089, -105.2251741], [39.7596887, -105.2253408], [39.7597554, -105.2254024], [39.7598802, -105.2255173], [39.7598939, -105.2255305], [39.7599658, -105.2255988], [39.7600186, -105.2256498], [39.7604008, -105.2259941], [39.7605066, -105.2260931], [39.7605924, -105.226174], [39.760667, -105.2262427], [39.7607946, -105.2263631], [39.7608753, -105.2264373], [39.7615829, -105.2271044], [39.7616266, -105.2271481], [39.7616788, -105.2272008], [39.762203, -105.2276943], [39.7623411, -105.2278299], [39.7630405, -105.2284565], [39.7632824, -105.2286685], [39.7637358, -105.2290679], [39.7640137, -105.229309], [39.7640954, -105.229381], [39.7641743, -105.229462], [39.76436, -105.2296495], [39.7645376, -105.2298658], [39.764824, -105.2301918], [39.7652051, -105.2307044], [39.7654711, -105.2310802], [39.7654801, -105.2310934], [39.7655308, -105.231168], [39.7655792, -105.231232], [39.7655918, -105.2312486], [39.7657719, -105.2315035], [39.7658973, -105.2316751], [39.7659962, -105.2318039], [39.7661018, -105.231924], [39.7662469, -105.2320614], [39.7664515, -105.2322588], [39.7671442, -105.2329283], [39.7673559, -105.2331244], [39.7674414, -105.2332053], [39.7675186, -105.2332845], [39.7675756, -105.2333522], [39.7676291, -105.2334243], [39.7676969, -105.2335369], [39.7677304, -105.233602], [39.767761, -105.2336649], [39.7677908, -105.233739], [39.7678172, -105.233812], [39.7678551, -105.2339422], [39.7678736, -105.2340294], [39.7678904, -105.2341257], [39.7679032, -105.2342329], [39.7679095, -105.2343359], [39.7679068, -105.2344027], [39.7679029, -105.234499], [39.7678875, -105.2346626], [39.7678651, -105.2347791], [39.7678384, -105.2348898], [39.7678006, -105.2350283], [39.7677517, -105.2352031], [39.7677119, -105.2353468]],\n", " {"bubblingMouseEvents": true, "color": "green", "dashArray": null, "dashOffset": null, "fill": false, "fillColor": "green", "fillOpacity": 0.2, "fillRule": "evenodd", "lineCap": "round", "lineJoin": "round", "noClip": false, "opacity": 0.8, "smoothFactor": 1.0, "stroke": true, "weight": 10}\n", - " ).addTo(map_a2ba13c8f965257bcc81a5b9f1664801);\n", + " ).addTo(map_c7345d2ba3ee51ff5955d8b9d1dc20b6);\n", " \n", " \n", - " poly_line_f8164f878668d4d431bf9490b1c0c98b.bindTooltip(\n", + " poly_line_f7dfde4d70684478e2ba63e72b94ca5b.bindTooltip(\n", " `<div>\n", " Least Energy\n", " </div>`,\n", @@ -502,19 +497,19 @@ " );\n", " \n", " \n", - " var marker_e7903ecfe2687cdf35e58bb6187e05c3 = L.marker(\n", + " var marker_a8c28fbe383145e2e368ea5d04e0e725 = L.marker(\n", " [39.7273069, -105.1994169],\n", " {}\n", - " ).addTo(map_a2ba13c8f965257bcc81a5b9f1664801);\n", + " ).addTo(map_c7345d2ba3ee51ff5955d8b9d1dc20b6);\n", " \n", " \n", - " var icon_d00a95d2fc383df1ee01c11c1d9015a2 = L.AwesomeMarkers.icon(\n", + " var icon_f4b068a33dbb5d6e0551beafff0c091c = L.AwesomeMarkers.icon(\n", " {"extraClasses": "fa-rotate-0", "icon": "circle", "iconColor": "white", "markerColor": "green", "prefix": "fa"}\n", " );\n", - " marker_e7903ecfe2687cdf35e58bb6187e05c3.setIcon(icon_d00a95d2fc383df1ee01c11c1d9015a2);\n", + " marker_a8c28fbe383145e2e368ea5d04e0e725.setIcon(icon_f4b068a33dbb5d6e0551beafff0c091c);\n", " \n", " \n", - " marker_e7903ecfe2687cdf35e58bb6187e05c3.bindTooltip(\n", + " marker_a8c28fbe383145e2e368ea5d04e0e725.bindTooltip(\n", " `<div>\n", " Origin\n", " </div>`,\n", @@ -522,19 +517,19 @@ " );\n", " \n", " \n", - " var marker_a8605646854f3779e76608978a213f99 = L.marker(\n", + " var marker_896695b507afb28930fcb5a332f47cb2 = L.marker(\n", " [39.7677119, -105.2353468],\n", " {}\n", - " ).addTo(map_a2ba13c8f965257bcc81a5b9f1664801);\n", + " ).addTo(map_c7345d2ba3ee51ff5955d8b9d1dc20b6);\n", " \n", " \n", - " var icon_7e6ec847709d3163e63c9f64c56b3953 = L.AwesomeMarkers.icon(\n", + " var icon_3f03d1531d14f55d0f27e168b4b4da80 = L.AwesomeMarkers.icon(\n", " {"extraClasses": "fa-rotate-0", "icon": "circle", "iconColor": "white", "markerColor": "red", "prefix": "fa"}\n", " );\n", - " marker_a8605646854f3779e76608978a213f99.setIcon(icon_7e6ec847709d3163e63c9f64c56b3953);\n", + " marker_896695b507afb28930fcb5a332f47cb2.setIcon(icon_3f03d1531d14f55d0f27e168b4b4da80);\n", " \n", " \n", - " marker_a8605646854f3779e76608978a213f99.bindTooltip(\n", + " marker_896695b507afb28930fcb5a332f47cb2.bindTooltip(\n", " `<div>\n", " Destination\n", " </div>`,\n", @@ -545,10 +540,10 @@ "</html>\" style=\"position:absolute;width:100%;height:100%;left:0;top:0;border:none !important;\" allowfullscreen webkitallowfullscreen mozallowfullscreen></iframe></div></div>" ], "text/plain": [ - "<folium.folium.Map at 0x1a3fa3560>" + "<folium.folium.Map at 0x12c5a1dd0>" ] }, - "execution_count": 13, + "execution_count": 14, "metadata": {}, "output_type": "execute_result" } @@ -569,7 +564,7 @@ }, { "cell_type": "code", - "execution_count": 14, + "execution_count": 16, "id": "aa1e6b11", "metadata": {}, "outputs": [ @@ -590,7 +585,7 @@ " <style>html, body {width: 100%;height: 100%;margin: 0;padding: 0;}</style>\n", " <style>#map {position:absolute;top:0;bottom:0;right:0;left:0;}</style>\n", " <script src="https://cdn.jsdelivr.net/npm/[email protected]/dist/leaflet.js"></script>\n", - " <script src="https://code.jquery.com/jquery-3.7.1.min.js"></script>\n", + " <script src="https://code.jquery.com/jquery-1.12.4.min.js"></script>\n", " <script src="https://cdn.jsdelivr.net/npm/[email protected]/dist/js/bootstrap.bundle.min.js"></script>\n", " <script src="https://cdnjs.cloudflare.com/ajax/libs/Leaflet.awesome-markers/2.0.2/leaflet.awesome-markers.js"></script>\n", " <link rel="stylesheet" href="https://cdn.jsdelivr.net/npm/[email protected]/dist/leaflet.css"/>\n", @@ -603,7 +598,7 @@ " <meta name="viewport" content="width=device-width,\n", " initial-scale=1.0, maximum-scale=1.0, user-scalable=no" />\n", " <style>\n", - " #map_0a69fb0ffb4fdf07d938c59697d2d5b7 {\n", + " #map_08d0a92c80cee353042d813e9d897e9a {\n", " position: relative;\n", " width: 100.0%;\n", " height: 100.0%;\n", @@ -617,14 +612,14 @@ "<body>\n", " \n", " \n", - " <div class="folium-map" id="map_0a69fb0ffb4fdf07d938c59697d2d5b7" ></div>\n", + " <div class="folium-map" id="map_08d0a92c80cee353042d813e9d897e9a" ></div>\n", " \n", "</body>\n", "<script>\n", " \n", " \n", - " var map_0a69fb0ffb4fdf07d938c59697d2d5b7 = L.map(\n", - " "map_0a69fb0ffb4fdf07d938c59697d2d5b7",\n", + " var map_08d0a92c80cee353042d813e9d897e9a = L.map(\n", + " "map_08d0a92c80cee353042d813e9d897e9a",\n", " {\n", " center: [39.7465774, -105.2216276],\n", " crs: L.CRS.EPSG3857,\n", @@ -638,41 +633,39 @@ "\n", " \n", " \n", - " var tile_layer_bf7e6951595d6bd28e928637e222904d = L.tileLayer(\n", + " var tile_layer_3d8120759aee037fb6485ddcaae7eb08 = L.tileLayer(\n", " "https://{s}.tile.openstreetmap.org/{z}/{x}/{y}.png",\n", " {"attribution": "Data by \\u0026copy; \\u003ca target=\\"_blank\\" href=\\"http://openstreetmap.org\\"\\u003eOpenStreetMap\\u003c/a\\u003e, under \\u003ca target=\\"_blank\\" href=\\"http://www.openstreetmap.org/copyright\\"\\u003eODbL\\u003c/a\\u003e.", "detectRetina": false, "maxNativeZoom": 18, "maxZoom": 18, "minZoom": 0, "noWrap": false, "opacity": 1, "subdomains": "abc", "tms": false}\n", - " );\n", + " ).addTo(map_08d0a92c80cee353042d813e9d897e9a);\n", " \n", " \n", - " tile_layer_bf7e6951595d6bd28e928637e222904d.addTo(map_0a69fb0ffb4fdf07d938c59697d2d5b7);\n", - " \n", - " var poly_line_b2f52dbae2df737185eacbb7f7f1814f = L.polyline(\n", + " var poly_line_fce8b3c76b03f40e0dbe8d7422abf0e5 = L.polyline(\n", " [[39.7273069, -105.1994169], [39.7272989, -105.1994535], [39.7272848, -105.1994868], [39.7272593, -105.1995218], [39.7272272, -105.1995461], [39.7271912, -105.1995576], [39.7271688, -105.199558], [39.7271461, -105.1995532], [39.7271142, -105.1995365], [39.7270868, -105.1995096], [39.7270657, -105.1994741], [39.7270519, -105.1994296], [39.7270483, -105.1993816], [39.727052, -105.1993539], [39.7270571, -105.1993247], [39.7270719, -105.1992916], [39.7270973, -105.1992566], [39.7271293, -105.1992323], [39.7271651, -105.1992207], [39.7271854, -105.1992226], [39.7272113, -105.199227], [39.7273496, -105.1991687], [39.7274185, -105.1991465], [39.7275326, -105.1990928], [39.7276828, -105.1990602], [39.7277455, -105.1990518], [39.7279702, -105.1990232], [39.7281558, -105.1990357], [39.7284384, -105.1990893], [39.7286639, -105.199177], [39.7288222, -105.1992773], [39.7289214, -105.199385], [39.7290292, -105.1995543], [39.7290506, -105.1995879], [39.7292157, -105.1998155], [39.7293094, -105.1998945], [39.7294079, -105.1999604], [39.7295437, -105.200014], [39.7296299, -105.2000721], [39.7297292, -105.2000757], [39.7297135, -105.2007081], [39.7297192, -105.2013106], [39.7298153, -105.2032041], [39.7298256, -105.2033784], [39.7298664, -105.2035708], [39.7298795, -105.203649], [39.7299267, -105.2038558], [39.7301082, -105.2042608], [39.7304911, -105.2048068], [39.7308741, -105.2053111], [39.7310676, -105.2055985], [39.7311933, -105.205824], [39.7312069, -105.2058543], [39.7313075, -105.2061158], [39.731348, -105.206248], [39.731386, -105.2064213], [39.7314237, -105.2067336], [39.731429, -105.2070565], [39.7314226, -105.207224], [39.7314076, -105.2074036], [39.7313611, -105.2077675], [39.7313265, -105.2079124], [39.7312831, -105.20806], [39.7312292, -105.2082086], [39.7311706, -105.208344], [39.7310965, -105.2084777], [39.7310049, -105.208616], [39.7309234, -105.2087235], [39.7308212, -105.2088303], [39.7306227, -105.2089819], [39.7304801, -105.2090736], [39.7303681, -105.2091988], [39.7303327, -105.2092597], [39.7303114, -105.2093181], [39.730299, -105.20939], [39.7302939, -105.2095126], [39.7303105, -105.2096804], [39.7304414, -105.2099267], [39.7305001, -105.2100344], [39.7305871, -105.2101842], [39.7307404, -105.2104481], [39.7309234, -105.2107229], [39.7310934, -105.2109639], [39.731238, -105.211145], [39.7314101, -105.2113464], [39.7315783, -105.2115336], [39.7317541, -105.2117064], [39.7319796, -105.2119089], [39.732359, -105.2122393], [39.7329187, -105.2127184], [39.7334557, -105.213188], [39.7338968, -105.2135676], [39.7342777, -105.2138968], [39.7347139, -105.2142786], [39.7354652, -105.2149301], [39.7356319, -105.2150733], [39.7358844, -105.2152902], [39.7363901, -105.2157275], [39.7369129, -105.2161859], [39.7377243, -105.2168831], [39.7390882, -105.2180682], [39.7397882, -105.2186718], [39.7403043, -105.2191078], [39.7405645, -105.2193336], [39.7406767, -105.2194225], [39.7415803, -105.2202142], [39.7416486, -105.2202727], [39.7417047, -105.2203151], [39.7421502, -105.2206932], [39.7425816, -105.221062], [39.7432687, -105.2215398], [39.7435797, -105.2217631], [39.7436721, -105.2218299], [39.7437366, -105.2218764], [39.7438794, -105.2219759], [39.7439546, -105.2220253], [39.7440082, -105.2220586], [39.7440819, -105.2221021], [39.7441377, -105.2221322], [39.7441808, -105.2221541], [39.7442796, -105.2222043], [39.7443357, -105.2222286], [39.7444062, -105.222259], [39.7444657, -105.2222871], [39.7444962, -105.2223029], [39.7445695, -105.2223431], [39.7446343, -105.2223808], [39.744683, -105.2224104], [39.7447215, -105.2224349], [39.7447658, -105.2224654], [39.744813, -105.222499], [39.7448547, -105.2225315], [39.7448881, -105.2225586], [39.7449328, -105.2225965], [39.7449815, -105.2226391], [39.7451542, -105.2227925], [39.7454784, -105.2230438], [39.7455034, -105.2230623], [39.7455159, -105.2230713], [39.7455403, -105.2230861], [39.7455548, -105.2230941], [39.7455766, -105.2231014], [39.7456018, -105.223106], [39.7456176, -105.2231071], [39.7456337, -105.223107], [39.7456638, -105.2231045], [39.7456814, -105.2231011], [39.7456983, -105.2230945], [39.7457224, -105.2230821], [39.7457338, -105.2230745], [39.7457588, -105.2230543], [39.7457709, -105.223043], [39.7457834, -105.2230284], [39.7458035, -105.2230027], [39.7458905, -105.2228787], [39.7460068, -105.2226624], [39.7461589, -105.2223894], [39.7463617, -105.2220126], [39.7464427, -105.2218672], [39.746482, -105.2217968], [39.7465202, -105.2217322], [39.7465523, -105.2216771], [39.7465774, -105.2216276], [39.7465928, -105.2215823], [39.7466004, -105.2215466], [39.7466042, -105.2215101], [39.7466044, -105.2214695], [39.746601, -105.2214368], [39.7465881, -105.2213862], [39.7465804, -105.2213567], [39.7465774, -105.2213321], [39.7465774, -105.2213048], [39.746581, -105.2212762], [39.7465874, -105.2212518], [39.7465978, -105.2212266], [39.7466125, -105.2212024], [39.7466266, -105.2211858], [39.7466485, -105.2211678], [39.7466708, -105.2211566], [39.746693, -105.2211511], [39.746715, -105.2211506], [39.7467203, -105.2211517], [39.7467798, -105.2211327], [39.7468101, -105.2211204], [39.7468347, -105.2211032], [39.7468673, -105.2210676], [39.7468863, -105.2210389], [39.7469079, -105.2209995], [39.7469752, -105.2208699], [39.7471073, -105.220655], [39.7471371, -105.2206129], [39.7471594, -105.2205845], [39.7472061, -105.2205376], [39.7473376, -105.2202962], [39.747357, -105.2202177], [39.7473682, -105.2201868], [39.7473848, -105.2201519], [39.7475089, -105.2199249], [39.7475861, -105.2197973], [39.7476254, -105.2197401], [39.7476917, -105.2196525], [39.7477724, -105.2195069], [39.7478875, -105.219299], [39.7479219, -105.2192372], [39.7479518, -105.219179], [39.7480957, -105.2189201], [39.7481337, -105.2188515], [39.7482567, -105.2186294], [39.7484851, -105.2182271], [39.7485167, -105.218164], [39.7485605, -105.2180881], [39.7490516, -105.217166], [39.7490955, -105.2170866], [39.7491263, -105.2170387], [39.7493736, -105.2165801], [39.7496841, -105.2160128], [39.7497482, -105.2160733], [39.7504593, -105.216753], [39.7505084, -105.2168007], [39.7505733, -105.2168567], [39.7509525, -105.2172063], [39.7512593, -105.2174884], [39.7513304, -105.2175506], [39.7513869, -105.217604], [39.7517458, -105.2179367], [39.7520945, -105.21826], [39.7521462, -105.2183047], [39.7522126, -105.218369], [39.7523012, -105.2184523], [39.7525052, -105.218642], [39.7527145, -105.2188366], [39.7528959, -105.2189993], [39.7529601, -105.2190649], [39.7530308, -105.219128], [39.7537033, -105.2197718], [39.753764, -105.2198302], [39.7538299, -105.2198915], [39.7542346, -105.2202638], [39.7545279, -105.2205433], [39.7545847, -105.2206031], [39.7546623, -105.2206728], [39.7553326, -105.2212752], [39.755407, -105.221342], [39.7554908, -105.2214211], [39.7561483, -105.2220415], [39.7562254, -105.2221143], [39.7563006, -105.2221853], [39.7566193, -105.222483], [39.7566315, -105.2224957], [39.7569475, -105.2227913], [39.7570561, -105.222895], [39.7574792, -105.2232693], [39.7575579, -105.2233409], [39.7576236, -105.2234019], [39.7578565, -105.2236199], [39.7579703, -105.2237258], [39.7580027, -105.2237561], [39.7583001, -105.2240544], [39.7583591, -105.2241084], [39.7587249, -105.2244446], [39.7588057, -105.22452], [39.7591609, -105.2248516], [39.7595089, -105.2251741], [39.7596887, -105.2253408], [39.7597554, -105.2254024], [39.7598802, -105.2255173], [39.7598939, -105.2255305], [39.7599658, -105.2255988], [39.7600186, -105.2256498], [39.7604008, -105.2259941], [39.7605066, -105.2260931], [39.7605924, -105.226174], [39.760667, -105.2262427], [39.7607946, -105.2263631], [39.7608753, -105.2264373], [39.7615829, -105.2271044], [39.7616266, -105.2271481], [39.7616788, -105.2272008], [39.762203, -105.2276943], [39.7623411, -105.2278299], [39.7630405, -105.2284565], [39.7632824, -105.2286685], [39.7637358, -105.2290679], [39.7640137, -105.229309], [39.7640954, -105.229381], [39.7641743, -105.229462], [39.76436, -105.2296495], [39.7645376, -105.2298658], [39.764824, -105.2301918], [39.7652051, -105.2307044], [39.7654711, -105.2310802], [39.7654801, -105.2310934], [39.7655308, -105.231168], [39.7655792, -105.231232], [39.7655918, -105.2312486], [39.7657719, -105.2315035], [39.7658973, -105.2316751], [39.7659962, -105.2318039], [39.7661018, -105.231924], [39.7662469, -105.2320614], [39.7664515, -105.2322588], [39.7671442, -105.2329283], [39.7673559, -105.2331244], [39.7674414, -105.2332053], [39.7675186, -105.2332845], [39.7675756, -105.2333522], [39.7676291, -105.2334243], [39.7676969, -105.2335369], [39.7677304, -105.233602], [39.767761, -105.2336649], [39.7677908, -105.233739], [39.7678172, -105.233812], [39.7678551, -105.2339422], [39.7678736, -105.2340294], [39.7678904, -105.2341257], [39.7679032, -105.2342329], [39.7679095, -105.2343359], [39.7679068, -105.2344027], [39.7679029, -105.234499], [39.7678875, -105.2346626], [39.7678651, -105.2347791], [39.7678384, -105.2348898], [39.7678006, -105.2350283], [39.7677517, -105.2352031], [39.7677119, -105.2353468]],\n", " {"bubblingMouseEvents": true, "color": "#eff821", "dashArray": null, "dashOffset": null, "fill": false, "fillColor": "#eff821", "fillOpacity": 0.2, "fillRule": "evenodd", "lineCap": "round", "lineJoin": "round", "noClip": false, "opacity": 0.8, "smoothFactor": 1.0, "stroke": true, "weight": 10}\n", - " ).addTo(map_0a69fb0ffb4fdf07d938c59697d2d5b7);\n", + " ).addTo(map_08d0a92c80cee353042d813e9d897e9a);\n", " \n", " \n", - " poly_line_b2f52dbae2df737185eacbb7f7f1814f.bindTooltip(\n", + " poly_line_fce8b3c76b03f40e0dbe8d7422abf0e5.bindTooltip(\n", " `<div>\n", - " 0.14906186986507128\n", + " 4.23738036063421\n", " </div>`,\n", " {"sticky": true}\n", " );\n", " \n", " \n", - " var marker_594f011ae899e7390103390e1b1c4e44 = L.marker(\n", + " var marker_57a9fae5fd1181475e8678e53c2e738d = L.marker(\n", " [39.7273069, -105.1994169],\n", " {}\n", - " ).addTo(map_0a69fb0ffb4fdf07d938c59697d2d5b7);\n", + " ).addTo(map_08d0a92c80cee353042d813e9d897e9a);\n", " \n", " \n", - " var icon_bc3742f29d0bbe9069e73d45bf225cb0 = L.AwesomeMarkers.icon(\n", + " var icon_b7f64c69eda0aa3787f9d028a487e2d0 = L.AwesomeMarkers.icon(\n", " {"extraClasses": "fa-rotate-0", "icon": "circle", "iconColor": "white", "markerColor": "green", "prefix": "fa"}\n", " );\n", - " marker_594f011ae899e7390103390e1b1c4e44.setIcon(icon_bc3742f29d0bbe9069e73d45bf225cb0);\n", + " marker_57a9fae5fd1181475e8678e53c2e738d.setIcon(icon_b7f64c69eda0aa3787f9d028a487e2d0);\n", " \n", " \n", - " marker_594f011ae899e7390103390e1b1c4e44.bindTooltip(\n", + " marker_57a9fae5fd1181475e8678e53c2e738d.bindTooltip(\n", " `<div>\n", " Origin\n", " </div>`,\n", @@ -680,19 +673,19 @@ " );\n", " \n", " \n", - " var marker_12b03e286e45c752ef3ec596908d4c51 = L.marker(\n", + " var marker_f2ad9efdc5dfb0e30a8de14f167f2b0e = L.marker(\n", " [39.7677119, -105.2353468],\n", " {}\n", - " ).addTo(map_0a69fb0ffb4fdf07d938c59697d2d5b7);\n", + " ).addTo(map_08d0a92c80cee353042d813e9d897e9a);\n", " \n", " \n", - " var icon_c9a062431c29df954115c641bf7beb58 = L.AwesomeMarkers.icon(\n", + " var icon_6b5edaf0623d1591b034bea1de6b720e = L.AwesomeMarkers.icon(\n", " {"extraClasses": "fa-rotate-0", "icon": "circle", "iconColor": "white", "markerColor": "red", "prefix": "fa"}\n", " );\n", - " marker_12b03e286e45c752ef3ec596908d4c51.setIcon(icon_c9a062431c29df954115c641bf7beb58);\n", + " marker_f2ad9efdc5dfb0e30a8de14f167f2b0e.setIcon(icon_6b5edaf0623d1591b034bea1de6b720e);\n", " \n", " \n", - " marker_12b03e286e45c752ef3ec596908d4c51.bindTooltip(\n", + " marker_f2ad9efdc5dfb0e30a8de14f167f2b0e.bindTooltip(\n", " `<div>\n", " Destination\n", " </div>`,\n", @@ -700,33 +693,33 @@ " );\n", " \n", " \n", - " var poly_line_a48dac49b22022311b057855def06976 = L.polyline(\n", + " var poly_line_56d979cdb52e71e5e1c0fdd7c6f9b67b = L.polyline(\n", " [[39.7273069, -105.1994169], [39.7272989, -105.1994535], [39.7272848, -105.1994868], [39.7272593, -105.1995218], [39.7272272, -105.1995461], [39.7271912, -105.1995576], [39.7271688, -105.199558], [39.7271461, -105.1995532], [39.7271142, -105.1995365], [39.7270868, -105.1995096], [39.7270657, -105.1994741], [39.7270519, -105.1994296], [39.7270483, -105.1993816], [39.727052, -105.1993539], [39.7270571, -105.1993247], [39.7270719, -105.1992916], [39.7270973, -105.1992566], [39.7271293, -105.1992323], [39.7271651, -105.1992207], [39.7271854, -105.1992226], [39.7272113, -105.199227], [39.7273496, -105.1991687], [39.7274185, -105.1991465], [39.7275326, -105.1990928], [39.7276828, -105.1990602], [39.7277455, -105.1990518], [39.7279702, -105.1990232], [39.7281558, -105.1990357], [39.7284384, -105.1990893], [39.7286639, -105.199177], [39.7288222, -105.1992773], [39.7289214, -105.199385], [39.7290292, -105.1995543], [39.7290506, -105.1995879], [39.7292157, -105.1998155], [39.7293094, -105.1998945], [39.7294079, -105.1999604], [39.7295437, -105.200014], [39.7296299, -105.2000721], [39.7296469, -105.1996789], [39.7296738, -105.1987284], [39.7296906, -105.1986794], [39.7297244, -105.1985856], [39.7297272, -105.1984193], [39.7299723, -105.1984198], [39.7302117, -105.1984227], [39.7303235, -105.1984285], [39.7304434, -105.198443], [39.7305651, -105.1984732], [39.7307182, -105.1985167], [39.7308487, -105.1985623], [39.7309909, -105.1986311], [39.731127, -105.1987144], [39.7312292, -105.1987897], [39.7313292, -105.1988731], [39.7314292, -105.1989611], [39.731524, -105.1990554], [39.7316089, -105.1991466], [39.7317085, -105.1992639], [39.7318575, -105.1994488], [39.7319522, -105.1995697], [39.7320423, -105.1996716], [39.7321372, -105.1997578], [39.7322615, -105.1998506], [39.7323697, -105.1999163], [39.7324712, -105.1999637], [39.7325898, -105.2000081], [39.732689, -105.2000309], [39.7327987, -105.2000627], [39.732945, -105.2001049], [39.7330635, -105.2001374], [39.7331838, -105.2001914], [39.7332899, -105.2002528], [39.7334312, -105.200363], [39.7335257, -105.2004544], [39.7335523, -105.2004835], [39.7336132, -105.2005503], [39.7336902, -105.2006555], [39.733767, -105.2007557], [39.7338411, -105.2008491], [39.7339359, -105.2009412], [39.7340357, -105.2010154], [39.7340742, -105.201042], [39.7341117, -105.2010678], [39.7341542, -105.2010901], [39.734211, -105.2011198], [39.7343173, -105.2011712], [39.7344056, -105.2012021], [39.7345704, -105.2012376], [39.7347726, -105.2012753], [39.7351329, -105.2013311], [39.7352203, -105.2013441], [39.7353243, -105.201352], [39.7354415, -105.2013441], [39.7355491, -105.2013268], [39.7357022, -105.2012901], [39.7358395, -105.201259], [39.7362429, -105.2011675], [39.7367157, -105.2010639], [39.7367857, -105.2010539], [39.7368531, -105.2010442], [39.7369645, -105.2010356], [39.7370755, -105.201044], [39.7371895, -105.2010632], [39.7373179, -105.201099], [39.737714, -105.2012295], [39.7379567, -105.2013009], [39.7384451, -105.2014404], [39.7386872, -105.2015167], [39.7395717, -105.2017905], [39.7397153, -105.2018369], [39.7398269, -105.2018673], [39.7399356, -105.2018843], [39.7400618, -105.2018961], [39.740184, -105.2018932], [39.7403133, -105.2018823], [39.7404288, -105.2018645], [39.7405518, -105.2018387], [39.740665, -105.2018073], [39.7407597, -105.2017713], [39.7408728, -105.2017196], [39.740992, -105.2016551], [39.7411268, -105.2015625], [39.7412599, -105.2014489], [39.7413718, -105.2013453], [39.7414758, -105.2012326], [39.741608, -105.2010691], [39.7416901, -105.200964], [39.7417903, -105.2008396], [39.7417994, -105.200826], [39.7419078, -105.2006985], [39.7419027, -105.2006189], [39.7419122, -105.200549], [39.7419496, -105.2004812], [39.7419894, -105.20044], [39.742047, -105.2004033], [39.7420816, -105.2004], [39.7421282, -105.2004131], [39.7421839, -105.200448], [39.7422234, -105.2004924], [39.7422448, -105.2005578], [39.7422566, -105.2006096], [39.7422556, -105.2006625], [39.7422421, -105.2007231], [39.7422712, -105.2008198], [39.7423227, -105.200901], [39.7428069, -105.2016549], [39.7429048, -105.2018114], [39.7429131, -105.2018245], [39.7431503, -105.2022035], [39.7434374, -105.2026477], [39.7435807, -105.2028659], [39.7437705, -105.2031843], [39.7441255, -105.2037193], [39.7444536, -105.2042289], [39.7445369, -105.204369], [39.7448087, -105.2048085], [39.7450389, -105.205197], [39.745164, -105.2054397], [39.7452617, -105.205642], [39.745359, -105.2058753], [39.7454413, -105.2060693], [39.7454698, -105.2061365], [39.7455586, -105.2063498], [39.7456394, -105.2066197], [39.7457343, -105.2068984], [39.7458096, -105.2071619], [39.7458432, -105.2072893], [39.7459057, -105.2075162], [39.7459547, -105.2077463], [39.7460104, -105.2080066], [39.7460312, -105.2081149], [39.7460508, -105.2082709], [39.7460836, -105.2084787], [39.7461092, -105.2086125], [39.7461165, -105.2087078], [39.7461128, -105.2088019], [39.7461078, -105.2089177], [39.7461163, -105.208995], [39.7461377, -105.2090672], [39.7461588, -105.2091147], [39.7462509, -105.2092159], [39.746281, -105.2092382], [39.7462978, -105.209256], [39.7463242, -105.209292], [39.7463458, -105.2093473], [39.746361, -105.2094251], [39.7463921, -105.2095768], [39.7464353, -105.2097133], [39.7465001, -105.2098285], [39.7467097, -105.21004], [39.7467814, -105.2101094], [39.7473357, -105.2106047], [39.7474554, -105.2107116], [39.7475615, -105.2108064], [39.7475958, -105.2108371], [39.747653, -105.2108907], [39.7481502, -105.2113567], [39.7483361, -105.2115309], [39.7483726, -105.2115651], [39.7484183, -105.2116079], [39.7484718, -105.2116584], [39.7487212, -105.2118951], [39.749238, -105.2123721], [39.7492814, -105.2124135], [39.7500059, -105.2130759], [39.7500454, -105.2131129], [39.7503812, -105.2134256], [39.7504627, -105.2135015], [39.7508219, -105.2138362], [39.7508792, -105.2138895], [39.7509348, -105.2139426], [39.7511093, -105.2141093], [39.7516271, -105.2146038], [39.7516716, -105.2146463], [39.751739, -105.2147085], [39.7524829, -105.2153953], [39.7525012, -105.2154123], [39.7532517, -105.2161101], [39.7533052, -105.2161598], [39.7540716, -105.2168674], [39.7541308, -105.2169221], [39.7544795, -105.2172508], [39.7546281, -105.2174522], [39.7546745, -105.2175037], [39.7548394, -105.2176869], [39.7549112, -105.2177871], [39.7548521, -105.2179433], [39.7546476, -105.2183047], [39.7544379, -105.2186805], [39.7544293, -105.218688], [39.7543723, -105.2187394], [39.7544376, -105.2188002], [39.7544872, -105.2188464], [39.7547934, -105.2191313], [39.7551071, -105.2194077], [39.7551368, -105.2194339], [39.755202, -105.2194913], [39.7552559, -105.2195411], [39.755615, -105.2198733], [39.755696, -105.2199479], [39.7559351, -105.2201693], [39.7559559, -105.220191], [39.7560109, -105.2202486], [39.7560703, -105.2203182], [39.756257, -105.2205368], [39.7562813, -105.2205598], [39.7564057, -105.2206774], [39.7565461, -105.2208058], [39.7567417, -105.2209442], [39.7566986, -105.2211001], [39.756638, -105.2213112], [39.7565732, -105.2214643], [39.7565209, -105.2215678], [39.7564595, -105.2216814], [39.7562914, -105.2219923], [39.7562254, -105.2221143], [39.7563006, -105.2221853], [39.7566193, -105.222483], [39.7566315, -105.2224957], [39.7569475, -105.2227913], [39.7570561, -105.222895], [39.7574792, -105.2232693], [39.7575579, -105.2233409], [39.7576236, -105.2234019], [39.7578565, -105.2236199], [39.7579703, -105.2237258], [39.7580027, -105.2237561], [39.7583001, -105.2240544], [39.7583591, -105.2241084], [39.7587249, -105.2244446], [39.7588057, -105.22452], [39.7591609, -105.2248516], [39.7595089, -105.2251741], [39.7596887, -105.2253408], [39.7597554, -105.2254024], [39.7598802, -105.2255173], [39.7598939, -105.2255305], [39.7599658, -105.2255988], [39.7600186, -105.2256498], [39.7604008, -105.2259941], [39.7605066, -105.2260931], [39.7605924, -105.226174], [39.760667, -105.2262427], [39.7607946, -105.2263631], [39.7608753, -105.2264373], [39.7615829, -105.2271044], [39.7616266, -105.2271481], [39.7616788, -105.2272008], [39.762203, -105.2276943], [39.7623411, -105.2278299], [39.7630405, -105.2284565], [39.7632824, -105.2286685], [39.7637358, -105.2290679], [39.7640137, -105.229309], [39.7640954, -105.229381], [39.7641743, -105.229462], [39.76436, -105.2296495], [39.7645376, -105.2298658], [39.764824, -105.2301918], [39.7652051, -105.2307044], [39.7654711, -105.2310802], [39.7654801, -105.2310934], [39.7655308, -105.231168], [39.7655792, -105.231232], [39.7655918, -105.2312486], [39.7657719, -105.2315035], [39.7658973, -105.2316751], [39.7659962, -105.2318039], [39.7661018, -105.231924], [39.7662469, -105.2320614], [39.7664515, -105.2322588], [39.7671442, -105.2329283], [39.7673559, -105.2331244], [39.7674414, -105.2332053], [39.7675186, -105.2332845], [39.7675756, -105.2333522], [39.7676291, -105.2334243], [39.7676969, -105.2335369], [39.7677304, -105.233602], [39.767761, -105.2336649], [39.7677908, -105.233739], [39.7678172, -105.233812], [39.7678551, -105.2339422], [39.7678736, -105.2340294], [39.7678904, -105.2341257], [39.7679032, -105.2342329], [39.7679095, -105.2343359], [39.7679068, -105.2344027], [39.7679029, -105.234499], [39.7678875, -105.2346626], [39.7678651, -105.2347791], [39.7678384, -105.2348898], [39.7678006, -105.2350283], [39.7677517, -105.2352031], [39.7677119, -105.2353468]],\n", " {"bubblingMouseEvents": true, "color": "#0c0786", "dashArray": null, "dashOffset": null, "fill": false, "fillColor": "#0c0786", "fillOpacity": 0.2, "fillRule": "evenodd", "lineCap": "round", "lineJoin": "round", "noClip": false, "opacity": 0.8, "smoothFactor": 1.0, "stroke": true, "weight": 10}\n", - " ).addTo(map_0a69fb0ffb4fdf07d938c59697d2d5b7);\n", + " ).addTo(map_08d0a92c80cee353042d813e9d897e9a);\n", " \n", " \n", - " poly_line_a48dac49b22022311b057855def06976.bindTooltip(\n", + " poly_line_56d979cdb52e71e5e1c0fdd7c6f9b67b.bindTooltip(\n", " `<div>\n", - " 0.14412252939511852\n", + " 4.013268489898491\n", " </div>`,\n", " {"sticky": true}\n", " );\n", " \n", " \n", - " var marker_a42455ffdb4b3173ce10e967b7fe4e0d = L.marker(\n", + " var marker_8223aabc91c0f1ea02ec2ea5181c8453 = L.marker(\n", " [39.7273069, -105.1994169],\n", " {}\n", - " ).addTo(map_0a69fb0ffb4fdf07d938c59697d2d5b7);\n", + " ).addTo(map_08d0a92c80cee353042d813e9d897e9a);\n", " \n", " \n", - " var icon_182dc1421ab527e4215256615066eba2 = L.AwesomeMarkers.icon(\n", + " var icon_efff58254cff016e5572e3aa608b8d64 = L.AwesomeMarkers.icon(\n", " {"extraClasses": "fa-rotate-0", "icon": "circle", "iconColor": "white", "markerColor": "green", "prefix": "fa"}\n", " );\n", - " marker_a42455ffdb4b3173ce10e967b7fe4e0d.setIcon(icon_182dc1421ab527e4215256615066eba2);\n", + " marker_8223aabc91c0f1ea02ec2ea5181c8453.setIcon(icon_efff58254cff016e5572e3aa608b8d64);\n", " \n", " \n", - " marker_a42455ffdb4b3173ce10e967b7fe4e0d.bindTooltip(\n", + " marker_8223aabc91c0f1ea02ec2ea5181c8453.bindTooltip(\n", " `<div>\n", " Origin\n", " </div>`,\n", @@ -734,19 +727,19 @@ " );\n", " \n", " \n", - " var marker_3f60358a3318165e8ed01991c2970fe3 = L.marker(\n", + " var marker_e3fa3a2726000cb73f319b6758aa89a8 = L.marker(\n", " [39.7677119, -105.2353468],\n", " {}\n", - " ).addTo(map_0a69fb0ffb4fdf07d938c59697d2d5b7);\n", + " ).addTo(map_08d0a92c80cee353042d813e9d897e9a);\n", " \n", " \n", - " var icon_7a556a87128e1f5ad26d35ebcac3cf6e = L.AwesomeMarkers.icon(\n", + " var icon_1d3ff4766d7167f6aae206479f62d543 = L.AwesomeMarkers.icon(\n", " {"extraClasses": "fa-rotate-0", "icon": "circle", "iconColor": "white", "markerColor": "red", "prefix": "fa"}\n", " );\n", - " marker_3f60358a3318165e8ed01991c2970fe3.setIcon(icon_7a556a87128e1f5ad26d35ebcac3cf6e);\n", + " marker_e3fa3a2726000cb73f319b6758aa89a8.setIcon(icon_1d3ff4766d7167f6aae206479f62d543);\n", " \n", " \n", - " marker_3f60358a3318165e8ed01991c2970fe3.bindTooltip(\n", + " marker_e3fa3a2726000cb73f319b6758aa89a8.bindTooltip(\n", " `<div>\n", " Destination\n", " </div>`,\n", @@ -757,16 +750,16 @@ "</html>\" style=\"position:absolute;width:100%;height:100%;left:0;top:0;border:none !important;\" allowfullscreen webkitallowfullscreen mozallowfullscreen></iframe></div></div>" ], "text/plain": [ - "<folium.folium.Map at 0x1a3b215b0>" + "<folium.folium.Map at 0x12ca7f290>" ] }, - "execution_count": 14, + "execution_count": 16, "metadata": {}, "output_type": "execute_result" } ], "source": [ - "m = plot_routes_folium(results, value_fn=lambda r: r[\"traversal_summary\"][\"energy\"], color_map=\"plasma\")\n", + "m = plot_routes_folium(results, value_fn=lambda r: r[\"traversal_summary\"][\"vehicle\"][\"energy\"], color_map=\"plasma\")\n", "m" ] }, diff --git a/python/nrel/routee/compass/resources/models/2016_BMW_i3_REx_PHEV_Charge_Depleting.bin b/python/nrel/routee/compass/resources/models/2016_BMW_i3_REx_PHEV_Charge_Depleting.bin new file mode 100644 index 00000000..d6d6b993 Binary files /dev/null and b/python/nrel/routee/compass/resources/models/2016_BMW_i3_REx_PHEV_Charge_Depleting.bin differ diff --git a/python/nrel/routee/compass/resources/models/2016_BMW_i3_REx_PHEV_Charge_Sustaining.bin b/python/nrel/routee/compass/resources/models/2016_BMW_i3_REx_PHEV_Charge_Sustaining.bin new file mode 100644 index 00000000..373891f8 Binary files /dev/null and b/python/nrel/routee/compass/resources/models/2016_BMW_i3_REx_PHEV_Charge_Sustaining.bin differ diff --git a/python/nrel/routee/compass/resources/models/2016_CHEVROLET_Volt_Charge_Depleting.bin b/python/nrel/routee/compass/resources/models/2016_CHEVROLET_Volt_Charge_Depleting.bin new file mode 100644 index 00000000..32927154 Binary files /dev/null and b/python/nrel/routee/compass/resources/models/2016_CHEVROLET_Volt_Charge_Depleting.bin differ diff --git a/python/nrel/routee/compass/resources/models/2016_CHEVROLET_Volt_Charge_Sustaining.bin b/python/nrel/routee/compass/resources/models/2016_CHEVROLET_Volt_Charge_Sustaining.bin new file mode 100644 index 00000000..5745af3d Binary files /dev/null and b/python/nrel/routee/compass/resources/models/2016_CHEVROLET_Volt_Charge_Sustaining.bin differ diff --git a/python/nrel/routee/compass/resources/models/2016_FORD_C-MAX_(PHEV)_Charge_Depleting.bin b/python/nrel/routee/compass/resources/models/2016_FORD_C-MAX_(PHEV)_Charge_Depleting.bin new file mode 100644 index 00000000..a7bc8059 Binary files /dev/null and b/python/nrel/routee/compass/resources/models/2016_FORD_C-MAX_(PHEV)_Charge_Depleting.bin differ diff --git a/python/nrel/routee/compass/resources/models/2016_FORD_C-MAX_(PHEV)_Charge_Sustaining.bin b/python/nrel/routee/compass/resources/models/2016_FORD_C-MAX_(PHEV)_Charge_Sustaining.bin new file mode 100644 index 00000000..e283a16d Binary files /dev/null and b/python/nrel/routee/compass/resources/models/2016_FORD_C-MAX_(PHEV)_Charge_Sustaining.bin differ diff --git a/python/nrel/routee/compass/resources/models/2016_HYUNDAI_Sonata_PHEV_Charge_Depleting.bin b/python/nrel/routee/compass/resources/models/2016_HYUNDAI_Sonata_PHEV_Charge_Depleting.bin new file mode 100644 index 00000000..91446c6e Binary files /dev/null and b/python/nrel/routee/compass/resources/models/2016_HYUNDAI_Sonata_PHEV_Charge_Depleting.bin differ diff --git a/python/nrel/routee/compass/resources/models/2016_HYUNDAI_Sonata_PHEV_Charge_Sustaining.bin b/python/nrel/routee/compass/resources/models/2016_HYUNDAI_Sonata_PHEV_Charge_Sustaining.bin new file mode 100644 index 00000000..a47c1085 Binary files /dev/null and b/python/nrel/routee/compass/resources/models/2016_HYUNDAI_Sonata_PHEV_Charge_Sustaining.bin differ diff --git a/python/nrel/routee/compass/resources/models/2017_Prius_Prime_Charge_Depleting.bin b/python/nrel/routee/compass/resources/models/2017_Prius_Prime_Charge_Depleting.bin new file mode 100644 index 00000000..a55bd3eb Binary files /dev/null and b/python/nrel/routee/compass/resources/models/2017_Prius_Prime_Charge_Depleting.bin differ diff --git a/python/nrel/routee/compass/resources/models/2017_Prius_Prime_Charge_Sustaining.bin b/python/nrel/routee/compass/resources/models/2017_Prius_Prime_Charge_Sustaining.bin new file mode 100644 index 00000000..ed388588 Binary files /dev/null and b/python/nrel/routee/compass/resources/models/2017_Prius_Prime_Charge_Sustaining.bin differ diff --git a/python/nrel/routee/compass/resources/osm_default_energy.toml b/python/nrel/routee/compass/resources/osm_default_energy.toml index 7dae2fdb..a055ce36 100644 --- a/python/nrel/routee/compass/resources/osm_default_energy.toml +++ b/python/nrel/routee/compass/resources/osm_default_energy.toml @@ -17,14 +17,15 @@ output_plugins = [ ] [traversal] -type = "speed_grade_energy_model" +type = "energy_model" speed_table_input_file = "edges-posted-speed-enumerated.txt.gz" speed_table_speed_unit = "kilometers_per_hour" output_time_unit = "minutes" output_distance_unit = "miles" -[[traversal.energy_models]] +[[traversal.vehicles]] name = "2012_Ford_Focus" +type = "single_fuel" model_input_file = "models/2012_Ford_Focus.bin" model_type = "smartcore" speed_unit = "miles_per_hour" @@ -33,8 +34,9 @@ energy_rate_unit = "gallons_gasoline_per_mile" ideal_energy_rate = 0.02857143 real_world_energy_adjustment = 1.166 -[[traversal.energy_models]] +[[traversal.vehicles]] name = "2012_Ford_Fusion" +type = "single_fuel" model_input_file = "models/2012_Ford_Fusion.bin" model_type = "smartcore" speed_unit = "miles_per_hour" @@ -43,8 +45,9 @@ energy_rate_unit = "gallons_gasoline_per_mile" ideal_energy_rate = 0.02857143 real_world_energy_adjustment = 1.166 -[[traversal.energy_models]] +[[traversal.vehicles]] name = "2016_AUDI_A3_4cyl_2WD" +type = "single_fuel" model_input_file = "models/2016_AUDI_A3_4cyl_2WD.bin" model_type = "smartcore" speed_unit = "miles_per_hour" @@ -53,8 +56,9 @@ energy_rate_unit = "gallons_gasoline_per_mile" ideal_energy_rate = 0.02857143 real_world_energy_adjustment = 1.166 -[[traversal.energy_models]] +[[traversal.vehicles]] name = "2016_BMW_328d_4cyl_2WD" +type = "single_fuel" model_input_file = "models/2016_BMW_328d_4cyl_2WD.bin" model_type = "smartcore" speed_unit = "miles_per_hour" @@ -63,8 +67,9 @@ energy_rate_unit = "gallons_gasoline_per_mile" ideal_energy_rate = 0.02857143 real_world_energy_adjustment = 1.166 -[[traversal.energy_models]] +[[traversal.vehicles]] name = "2016_CHEVROLET_Malibu_4cyl_2WD" +type = "single_fuel" model_input_file = "models/2016_CHEVROLET_Malibu_4cyl_2WD.bin" model_type = "smartcore" speed_unit = "miles_per_hour" @@ -73,8 +78,9 @@ energy_rate_unit = "gallons_gasoline_per_mile" ideal_energy_rate = 0.02857143 real_world_energy_adjustment = 1.166 -[[traversal.energy_models]] +[[traversal.vehicles]] name = "2016_CHEVROLET_Spark_EV" +type = "single_fuel" model_input_file = "models/2016_CHEVROLET_Spark_EV.bin" model_type = "smartcore" speed_unit = "miles_per_hour" @@ -83,8 +89,9 @@ energy_rate_unit = "kilowatt_hours_per_mile" ideal_energy_rate = 0.2 real_world_energy_adjustment = 1.3958 -[[traversal.energy_models]] +[[traversal.vehicles]] name = "2016_FORD_C-MAX_HEV" +type = "single_fuel" model_input_file = "models/2016_FORD_C-MAX_HEV.bin" model_type = "smartcore" speed_unit = "miles_per_hour" @@ -93,8 +100,9 @@ energy_rate_unit = "gallons_gasoline_per_mile" ideal_energy_rate = 0.02 real_world_energy_adjustment = 1.1252 -[[traversal.energy_models]] +[[traversal.vehicles]] name = "2016_FORD_Escape_4cyl_2WD" +type = "single_fuel" model_input_file = "models/2016_FORD_Escape_4cyl_2WD.bin" model_type = "smartcore" speed_unit = "miles_per_hour" @@ -103,8 +111,9 @@ energy_rate_unit = "gallons_gasoline_per_mile" ideal_energy_rate = 0.02857143 real_world_energy_adjustment = 1.166 -[[traversal.energy_models]] +[[traversal.vehicles]] name = "2016_FORD_Explorer_4cyl_2WD" +type = "single_fuel" model_input_file = "models/2016_FORD_Explorer_4cyl_2WD.bin" model_type = "smartcore" speed_unit = "miles_per_hour" @@ -113,8 +122,9 @@ energy_rate_unit = "gallons_gasoline_per_mile" ideal_energy_rate = 0.02857143 real_world_energy_adjustment = 1.166 -[[traversal.energy_models]] +[[traversal.vehicles]] name = "2016_HYUNDAI_Elantra_4cyl_2WD" +type = "single_fuel" model_input_file = "models/2016_HYUNDAI_Elantra_4cyl_2WD.bin" model_type = "smartcore" speed_unit = "miles_per_hour" @@ -123,8 +133,9 @@ energy_rate_unit = "gallons_gasoline_per_mile" ideal_energy_rate = 0.02857143 real_world_energy_adjustment = 1.166 -[[traversal.energy_models]] +[[traversal.vehicles]] name = "2016_KIA_Optima_Hybrid" +type = "single_fuel" model_input_file = "models/2016_KIA_Optima_Hybrid.bin" model_type = "smartcore" speed_unit = "miles_per_hour" @@ -133,8 +144,9 @@ energy_rate_unit = "gallons_gasoline_per_mile" ideal_energy_rate = 0.02 real_world_energy_adjustment = 1.1252 -[[traversal.energy_models]] +[[traversal.vehicles]] name = "2016_Leaf_24_kWh" +type = "single_fuel" model_input_file = "models/2016_Leaf_24_kWh.bin" model_type = "smartcore" speed_unit = "miles_per_hour" @@ -143,8 +155,9 @@ energy_rate_unit = "kilowatt_hours_per_mile" ideal_energy_rate = 0.2 real_world_energy_adjustment = 1.3958 -[[traversal.energy_models]] +[[traversal.vehicles]] name = "2016_MITSUBISHI_i-MiEV" +type = "single_fuel" model_input_file = "models/2016_MITSUBISHI_i-MiEV.bin" model_type = "smartcore" speed_unit = "miles_per_hour" @@ -153,8 +166,9 @@ energy_rate_unit = "kilowatt_hours_per_mile" ideal_energy_rate = 0.2 real_world_energy_adjustment = 1.3958 -[[traversal.energy_models]] +[[traversal.vehicles]] name = "2016_Nissan_Leaf_30_kWh" +type = "single_fuel" model_input_file = "models/2016_Nissan_Leaf_30_kWh.bin" model_type = "smartcore" speed_unit = "miles_per_hour" @@ -163,8 +177,9 @@ energy_rate_unit = "kilowatt_hours_per_mile" ideal_energy_rate = 0.2 real_world_energy_adjustment = 1.3958 -[[traversal.energy_models]] +[[traversal.vehicles]] name = "2016_TESLA_Model_S60_2WD" +type = "single_fuel" model_input_file = "models/2016_TESLA_Model_S60_2WD.bin" model_type = "smartcore" speed_unit = "miles_per_hour" @@ -173,8 +188,9 @@ energy_rate_unit = "kilowatt_hours_per_mile" ideal_energy_rate = 0.2 real_world_energy_adjustment = 1.3958 -[[traversal.energy_models]] +[[traversal.vehicles]] name = "2016_TOYOTA_Camry_4cyl_2WD" +type = "single_fuel" model_input_file = "models/2016_TOYOTA_Camry_4cyl_2WD.bin" model_type = "smartcore" speed_unit = "miles_per_hour" @@ -183,8 +199,9 @@ energy_rate_unit = "gallons_gasoline_per_mile" ideal_energy_rate = 0.02857143 real_world_energy_adjustment = 1.166 -[[traversal.energy_models]] +[[traversal.vehicles]] name = "2016_TOYOTA_Corolla_4cyl_2WD" +type = "single_fuel" model_input_file = "models/2016_TOYOTA_Corolla_4cyl_2WD.bin" model_type = "smartcore" speed_unit = "miles_per_hour" @@ -193,8 +210,9 @@ energy_rate_unit = "gallons_gasoline_per_mile" ideal_energy_rate = 0.02857143 real_world_energy_adjustment = 1.166 -[[traversal.energy_models]] +[[traversal.vehicles]] name = "2016_TOYOTA_Highlander_Hybrid" +type = "single_fuel" model_input_file = "models/2016_TOYOTA_Highlander_Hybrid.bin" model_type = "smartcore" speed_unit = "miles_per_hour" @@ -203,8 +221,9 @@ energy_rate_unit = "gallons_gasoline_per_mile" ideal_energy_rate = 0.02 real_world_energy_adjustment = 1.1252 -[[traversal.energy_models]] +[[traversal.vehicles]] name = "2016_Toyota_Prius_Two_FWD" +type = "single_fuel" model_input_file = "models/2016_Toyota_Prius_Two_FWD.bin" model_type = "smartcore" speed_unit = "miles_per_hour" @@ -213,8 +232,9 @@ energy_rate_unit = "gallons_gasoline_per_mile" ideal_energy_rate = 0.02 real_world_energy_adjustment = 1.1252 -[[traversal.energy_models]] +[[traversal.vehicles]] name = "2017_CHEVROLET_Bolt" +type = "single_fuel" model_input_file = "models/2017_CHEVROLET_Bolt.bin" model_type = "smartcore" speed_unit = "miles_per_hour" @@ -223,8 +243,9 @@ energy_rate_unit = "kilowatt_hours_per_mile" ideal_energy_rate = 0.2 real_world_energy_adjustment = 1.3958 -[[traversal.energy_models]] +[[traversal.vehicles]] name = "2017_Maruti_Dzire_VDI" +type = "single_fuel" model_input_file = "models/2017_Maruti_Dzire_VDI.bin" model_type = "smartcore" speed_unit = "miles_per_hour" @@ -233,8 +254,9 @@ energy_rate_unit = "gallons_gasoline_per_mile" ideal_energy_rate = 0.02857143 real_world_energy_adjustment = 1.166 -[[traversal.energy_models]] +[[traversal.vehicles]] name = "2017_Toyota_Highlander_3.5_L" +type = "single_fuel" model_input_file = "models/2017_Toyota_Highlander_3.5_L.bin" model_type = "smartcore" speed_unit = "miles_per_hour" @@ -243,8 +265,9 @@ energy_rate_unit = "gallons_gasoline_per_mile" ideal_energy_rate = 0.02857143 real_world_energy_adjustment = 1.166 -[[traversal.energy_models]] +[[traversal.vehicles]] name = "2020_Chevrolet_Colorado_2WD_Diesel" +type = "single_fuel" model_input_file = "models/2020_Chevrolet_Colorado_2WD_Diesel.bin" model_type = "smartcore" speed_unit = "miles_per_hour" @@ -253,8 +276,9 @@ energy_rate_unit = "gallons_gasoline_per_mile" ideal_energy_rate = 0.02857143 real_world_energy_adjustment = 1.166 -[[traversal.energy_models]] +[[traversal.vehicles]] name = "2020_VW_Golf_1.5TSI" +type = "single_fuel" model_input_file = "models/2020_VW_Golf_1.5TSI.bin" model_type = "smartcore" speed_unit = "miles_per_hour" @@ -263,8 +287,9 @@ energy_rate_unit = "gallons_gasoline_per_mile" ideal_energy_rate = 0.02857143 real_world_energy_adjustment = 1.166 -[[traversal.energy_models]] +[[traversal.vehicles]] name = "2020_VW_Golf_2.0TDI" +type = "single_fuel" model_input_file = "models/2020_VW_Golf_2.0TDI.bin" model_type = "smartcore" speed_unit = "miles_per_hour" @@ -273,8 +298,9 @@ energy_rate_unit = "gallons_gasoline_per_mile" ideal_energy_rate = 0.02857143 real_world_energy_adjustment = 1.166 -[[traversal.energy_models]] +[[traversal.vehicles]] name = "2021_Fiat_Panda_Mild_Hybrid" +type = "single_fuel" model_input_file = "models/2021_Fiat_Panda_Mild_Hybrid.bin" model_type = "smartcore" speed_unit = "miles_per_hour" @@ -283,8 +309,9 @@ energy_rate_unit = "gallons_gasoline_per_mile" ideal_energy_rate = 0.02857143 real_world_energy_adjustment = 1.166 -[[traversal.energy_models]] +[[traversal.vehicles]] name = "2021_Peugot_3008" +type = "single_fuel" model_input_file = "models/2021_Peugot_3008.bin" model_type = "smartcore" speed_unit = "miles_per_hour" @@ -293,8 +320,9 @@ energy_rate_unit = "gallons_gasoline_per_mile" ideal_energy_rate = 0.02857143 real_world_energy_adjustment = 1.166 -[[traversal.energy_models]] +[[traversal.vehicles]] name = "2022_Ford_F-150_Lightning_4WD" +type = "single_fuel" model_input_file = "models/2022_Ford_F-150_Lightning_4WD.bin" model_type = "smartcore" speed_unit = "miles_per_hour" @@ -303,8 +331,9 @@ energy_rate_unit = "kilowatt_hours_per_mile" ideal_energy_rate = 0.2 real_world_energy_adjustment = 1.3958 -[[traversal.energy_models]] +[[traversal.vehicles]] name = "2022_Renault_Zoe_ZE50_R135" +type = "single_fuel" model_input_file = "models/2022_Renault_Zoe_ZE50_R135.bin" model_type = "smartcore" speed_unit = "miles_per_hour" @@ -313,8 +342,9 @@ energy_rate_unit = "kilowatt_hours_per_mile" ideal_energy_rate = 0.2 real_world_energy_adjustment = 1.3958 -[[traversal.energy_models]] +[[traversal.vehicles]] name = "2022_Tesla_Model_3_RWD" +type = "single_fuel" model_input_file = "models/2022_Tesla_Model_3_RWD.bin" model_type = "smartcore" speed_unit = "miles_per_hour" @@ -323,8 +353,9 @@ energy_rate_unit = "kilowatt_hours_per_mile" ideal_energy_rate = 0.2 real_world_energy_adjustment = 1.3958 -[[traversal.energy_models]] +[[traversal.vehicles]] name = "2022_Tesla_Model_Y_RWD" +type = "single_fuel" model_input_file = "models/2022_Tesla_Model_Y_RWD.bin" model_type = "smartcore" speed_unit = "miles_per_hour" @@ -333,8 +364,9 @@ energy_rate_unit = "kilowatt_hours_per_mile" ideal_energy_rate = 0.2 real_world_energy_adjustment = 1.3958 -[[traversal.energy_models]] +[[traversal.vehicles]] name = "2022_Toyota_Yaris_Hybrid_Mid" +type = "single_fuel" model_input_file = "models/2022_Toyota_Yaris_Hybrid_Mid.bin" model_type = "smartcore" speed_unit = "miles_per_hour" @@ -343,8 +375,9 @@ energy_rate_unit = "gallons_gasoline_per_mile" ideal_energy_rate = 0.02 real_world_energy_adjustment = 1.1252 -[[traversal.energy_models]] +[[traversal.vehicles]] name = "2022_Volvo_XC40_Recharge_twin" +type = "single_fuel" model_input_file = "models/2022_Volvo_XC40_Recharge_twin.bin" model_type = "smartcore" speed_unit = "miles_per_hour" @@ -353,8 +386,9 @@ energy_rate_unit = "kilowatt_hours_per_mile" ideal_energy_rate = 0.2 real_world_energy_adjustment = 1.3958 -[[traversal.energy_models]] +[[traversal.vehicles]] name = "2023_Mitsubishi_Pajero_Sport" +type = "single_fuel" model_input_file = "models/2023_Mitsubishi_Pajero_Sport.bin" model_type = "smartcore" speed_unit = "miles_per_hour" @@ -362,3 +396,123 @@ grade_unit = "decimal" energy_rate_unit = "gallons_gasoline_per_mile" ideal_energy_rate = 0.02857143 real_world_energy_adjustment = 1.166 + +[[traversal.vehicles]] +type = "dual_fuel" +name = "2016_BMW_i3_REx_PHEV" +battery_capacity = 12 +battery_capacity_unit = "kilowatt_hours" +[traversal.vehicles.charge_depleting] +name = "2016_BMW_i3_REx_PHEV_Charge_Depleting" +model_input_file = "models/2016_BMW_i3_REx_PHEV_Charge_Depleting.bin" +model_type = "smartcore" +speed_unit = "miles_per_hour" +grade_unit = "decimal" +energy_rate_unit = "kilowatt_hours_per_mile" +ideal_energy_rate = 0.2 +real_world_energy_adjustment = 1.3958 +[traversal.vehicles.charge_sustaining] +name = "2016_BMW_i3_REx_PHEV_Charge_Sustaining" +model_input_file = "models/2016_BMW_i3_REx_PHEV_Charge_Sustaining.bin" +model_type = "smartcore" +speed_unit = "miles_per_hour" +grade_unit = "decimal" +energy_rate_unit = "gallons_gasoline_per_mile" +ideal_energy_rate = 0.02 +real_world_energy_adjustment = 1.1252 + +[[traversal.vehicles]] +type = "dual_fuel" +name = "2016_CHEVROLET_Volt" +battery_capacity = 12 +battery_capacity_unit = "kilowatt_hours" +[traversal.vehicles.charge_depleting] +name = "2016_CHEVROLET_Volt_Charge_Depleting" +model_input_file = "models/2016_CHEVROLET_Volt_Charge_Depleting.bin" +model_type = "smartcore" +speed_unit = "miles_per_hour" +grade_unit = "decimal" +energy_rate_unit = "kilowatt_hours_per_mile" +ideal_energy_rate = 0.2 +real_world_energy_adjustment = 1.3958 +[traversal.vehicles.charge_sustaining] +name = "2016_CHEVROLET_Volt_Charge_Sustaining" +model_input_file = "models/2016_CHEVROLET_Volt_Charge_Sustaining.bin" +model_type = "smartcore" +speed_unit = "miles_per_hour" +grade_unit = "decimal" +energy_rate_unit = "gallons_gasoline_per_mile" +ideal_energy_rate = 0.02 +real_world_energy_adjustment = 1.1252 + +[[traversal.vehicles]] +type = "dual_fuel" +name = "2016_FORD_C-MAX_(PHEV)" +battery_capacity = 7.6 +battery_capacity_unit = "kilowatt_hours" +[traversal.vehicles.charge_depleting] +name = "2016_FORD_C-MAX_(PHEV)_Charge_Depleting" +model_input_file = "models/2016_FORD_C-MAX_(PHEV)_Charge_Depleting.bin" +model_type = "smartcore" +speed_unit = "miles_per_hour" +grade_unit = "decimal" +energy_rate_unit = "kilowatt_hours_per_mile" +ideal_energy_rate = 0.2 +real_world_energy_adjustment = 1.3958 +[traversal.vehicles.charge_sustaining] +name = "2016_FORD_C-MAX_(PHEV)_Charge_Sustaining" +model_input_file = "models/2016_FORD_C-MAX_(PHEV)_Charge_Sustaining.bin" +model_type = "smartcore" +speed_unit = "miles_per_hour" +grade_unit = "decimal" +energy_rate_unit = "gallons_gasoline_per_mile" +ideal_energy_rate = 0.02 +real_world_energy_adjustment = 1.1252 + +[[traversal.vehicles]] +type = "dual_fuel" +name = "2016_HYUNDAI_Sonata_PHEV" +battery_capacity = 9.8 +battery_capacity_unit = "kilowatt_hours" +[traversal.vehicles.charge_depleting] +name = "2016_HYUNDAI_Sonata_PHEV_Charge_Depleting" +model_input_file = "models/2016_HYUNDAI_Sonata_PHEV_Charge_Depleting.bin" +model_type = "smartcore" +speed_unit = "miles_per_hour" +grade_unit = "decimal" +energy_rate_unit = "kilowatt_hours_per_mile" +ideal_energy_rate = 0.2 +real_world_energy_adjustment = 1.3958 +[traversal.vehicles.charge_sustaining] +name = "2016_HYUNDAI_Sonata_PHEV_Charge_Sustaining" +model_input_file = "models/2016_HYUNDAI_Sonata_PHEV_Charge_Sustaining.bin" +model_type = "smartcore" +speed_unit = "miles_per_hour" +grade_unit = "decimal" +energy_rate_unit = "gallons_gasoline_per_mile" +ideal_energy_rate = 0.02 +real_world_energy_adjustment = 1.1252 + +[[traversal.vehicles]] +type = "dual_fuel" +name = "2017_Prius_Prime" +battery_capacity = 8 +battery_capacity_unit = "kilowatt_hours" +[traversal.vehicles.charge_depleting] +name = "2017_Prius_Prime_Charge_Depleting" +model_input_file = "models/2017_Prius_Prime_Charge_Depleting.bin" +model_type = "smartcore" +speed_unit = "miles_per_hour" +grade_unit = "decimal" +energy_rate_unit = "kilowatt_hours_per_mile" +ideal_energy_rate = 0.2 +real_world_energy_adjustment = 1.3958 +[traversal.vehicles.charge_sustaining] +name = "2017_Prius_Prime_Charge_Sustaining" +model_input_file = "models/2017_Prius_Prime_Charge_Sustaining.bin" +model_type = "smartcore" +speed_unit = "miles_per_hour" +grade_unit = "decimal" +energy_rate_unit = "gallons_gasoline_per_mile" +ideal_energy_rate = 0.02 +real_world_energy_adjustment = 1.1252 \ No newline at end of file diff --git a/rust/routee-compass-core/src/util/unit/energy_unit.rs b/rust/routee-compass-core/src/util/unit/energy_unit.rs index 3370a01e..1f718219 100644 --- a/rust/routee-compass-core/src/util/unit/energy_unit.rs +++ b/rust/routee-compass-core/src/util/unit/energy_unit.rs @@ -1,13 +1,25 @@ use serde::{Deserialize, Serialize}; -#[derive(Debug, Serialize, Deserialize, Clone, PartialEq)] +use super::Energy; + +#[derive(Debug, Serialize, Deserialize, Clone, PartialEq, Copy)] #[serde(rename_all = "snake_case")] pub enum EnergyUnit { GallonsGasoline, KilowattHours, } -impl EnergyUnit {} +impl EnergyUnit { + pub fn convert(&self, value: Energy, target: EnergyUnit) -> Energy { + use EnergyUnit as S; + match (self, target) { + (S::GallonsGasoline, S::GallonsGasoline) => value, + (S::GallonsGasoline, S::KilowattHours) => value * 33.41, + (S::KilowattHours, S::GallonsGasoline) => value * 0.0299, + (S::KilowattHours, S::KilowattHours) => value, + } + } +} impl std::fmt::Display for EnergyUnit { fn fmt(&self, f: &mut std::fmt::Formatter<'_>) -> std::fmt::Result { diff --git a/rust/routee-compass-powertrain/src/doc.md b/rust/routee-compass-powertrain/src/doc.md index 0ea5e76c..cbd066dc 100644 --- a/rust/routee-compass-powertrain/src/doc.md +++ b/rust/routee-compass-powertrain/src/doc.md @@ -11,7 +11,7 @@ RouteE Powertrain models are trained and exported via the RouteE Powertrain util ### Model Runtimes There are two underlying model runtimes available, [smartcore](https://smartcorelib.org/) and [ort](https://github.com/pykeio/ort) (for [ONNX](https://onnx.ai/) models). -By default, this crate is loaded with ONNX deactivated. +By default, this crate is loaded with ONNX deactivated. To activate the ONNX feature, pass the `onnx` feature flag during compilation. Runtime kernels for 3 common OSs have been provided in the onnx-runtime directory within this crate. For more information on cargo features, see The Cargo Book chapter on [Features](https://doc.rust-lang.org/cargo/reference/features.html). @@ -25,29 +25,32 @@ An example traversal model configuration that uses this crate may look like this ```toml [traversal] -type = "speed_grade_energy_model" -model_type = "smartcore" +type = "energy_model" speed_table_input_file = "edges-posted-speed-enumerated.txt.gz" -energy_model_input_file = "2016_TOYOTA_Camry_4cyl_2WD.bin" -ideal_energy_rate = 0.02857142857 speed_table_speed_unit = "kilometers_per_hour" -energy_model_speed_unit = "miles_per_hour" -energy_model_grade_unit = "decimal" -energy_model_energy_rate_unit = "gallons_gasoline_per_mile" output_time_unit = "minutes" output_distance_unit = "miles" -grade_table_input_file = "edges-grade-enumerated.txt.gz" -grade_table_grade_unit = "decimal" + +[[traversal.vehicles]] +name = "2012_Ford_Focus" +type = "single_fuel" +model_input_file = "models/2012_Ford_Focus.bin" +model_type = "smartcore" +speed_unit = "miles_per_hour" +grade_unit = "decimal" +energy_rate_unit = "gallons_gasoline_per_mile" +ideal_energy_rate = 0.02857143 +real_world_energy_adjustment = 1.166 ``` -This TOML section is deserialized into JSON and passed as arguments to the SpeedGradeEnergyModelBuilder which in turn loads the [SpeedGradeModelService] in this crate. -This in turn builds the [SpeedGradeModel]. +This TOML section is deserialized into JSON and passed as arguments to the EnergyModelBuilder which in turn loads the [EnergyModelService] in this crate. +This in turn builds the [EnergyTraversalModel]. ### Search -TraversalModels in this crate will add energy estimation to road network search, and will differ in their dependencies and evaluation procedures. +TraversalModels in this crate will add energy estimation to road network search, and will differ in their dependencies and evaluation procedures. -#### SpeedGradeModel +#### EnergyTraversalModel ##### Dependencies @@ -62,7 +65,7 @@ TraversalModels in this crate will add energy estimation to road network search, 2. compute travel time as `distance / speed` for this edge 3. lookup grade for this edge in table; if missing, use `0.0` 4. perform inference to retrieve energy rate from speed and grade values -5. compute energy as `energy_rate * distance * real_world_adjustment_factor` for this edge +5. compute energy as `energy_rate * distance * real_world_adjustment_factor` for this edge 6. compute link cost as `(energy * energy_cost_coefficient) + (time * (1 - energy_cost_coefficient))` ### Real-World Adjustment Factors @@ -70,13 +73,10 @@ TraversalModels in this crate will add energy estimation to road network search, In addition to calculating the energy based on a RouteE Powertrain output, an adjustment factor should be applied to capture real-world effects of running a powertrain in an environment. As a result of NREL research, some recommended values for this adjustment are: -powertrain type | factor ---- | --- -combustion vehicle (CV) | 1.1660 -hybrid vehicle (HV) | 1.1252 -electric vehicle (EV) | 1.3958 +| powertrain type | factor | +| ----------------------- | ------ | +| combustion vehicle (CV) | 1.1660 | +| hybrid vehicle (HV) | 1.1252 | +| electric vehicle (EV) | 1.3958 | A factor of 1.0 equates to 100% of the original energy value. - -[SpeedGradeModelService]: crate::routee::speed_grade_model_service::SpeedGradeModelService -[SpeedGradeModelService]: crate::routee::speed_grade_model::SpeedGradeModel \ No newline at end of file diff --git a/rust/routee-compass-powertrain/src/routee/energy_model_ops.rs b/rust/routee-compass-powertrain/src/routee/energy_model_ops.rs new file mode 100644 index 00000000..52ddc9b4 --- /dev/null +++ b/rust/routee-compass-powertrain/src/routee/energy_model_ops.rs @@ -0,0 +1,26 @@ +use routee_compass_core::{ + model::{road_network::edge_id::EdgeId, traversal::traversal_model_error::TraversalModelError}, + util::unit::Grade, +}; + +pub const ZERO_ENERGY: f64 = 1e-9; + +/// look up the grade from the grade table +pub fn get_grade( + grade_table: &Option<Box<[Grade]>>, + edge_id: EdgeId, +) -> Result<Grade, TraversalModelError> { + match grade_table { + None => Ok(Grade::ZERO), + Some(gt) => { + let grade: &Grade = gt.get(edge_id.as_usize()).ok_or( + TraversalModelError::MissingIdInTabularCostFunction( + format!("{}", edge_id), + String::from("EdgeId"), + String::from("grade table"), + ), + )?; + Ok(*grade) + } + } +} diff --git a/rust/routee-compass-powertrain/src/routee/speed_grade_energy_model_service.rs b/rust/routee-compass-powertrain/src/routee/energy_model_service.rs similarity index 80% rename from rust/routee-compass-powertrain/src/routee/speed_grade_energy_model_service.rs rename to rust/routee-compass-powertrain/src/routee/energy_model_service.rs index eccb29d2..fce8d601 100644 --- a/rust/routee-compass-powertrain/src/routee/speed_grade_energy_model_service.rs +++ b/rust/routee-compass-powertrain/src/routee/energy_model_service.rs @@ -1,5 +1,3 @@ -use super::prediction_model::SpeedGradePredictionModelRecord; -use super::speed_grade_energy_model::SpeedGradeEnergyModel; use routee_compass_core::model::traversal::default::speed_lookup_model::get_max_speed; use routee_compass_core::model::traversal::traversal_model::TraversalModel; use routee_compass_core::model::traversal::traversal_model_error::TraversalModelError; @@ -11,8 +9,11 @@ use std::collections::HashMap; use std::path::Path; use std::sync::Arc; +use super::energy_traversal_model::EnergyTraversalModel; +use super::vehicle::VehicleType; + #[derive(Clone)] -pub struct SpeedGradeEnergyModelService { +pub struct EnergyModelService { pub speed_table: Arc<Box<[Speed]>>, pub speeds_table_speed_unit: SpeedUnit, pub max_speed: Speed, @@ -20,21 +21,10 @@ pub struct SpeedGradeEnergyModelService { pub grade_table_grade_unit: GradeUnit, pub output_time_unit: TimeUnit, pub output_distance_unit: DistanceUnit, - pub energy_model_library: Arc<HashMap<String, Arc<SpeedGradePredictionModelRecord>>>, -} - -impl TraversalModelService for SpeedGradeEnergyModelService { - fn build( - &self, - query: &serde_json::Value, - ) -> Result<Arc<dyn TraversalModel>, TraversalModelError> { - let arc_self = Arc::new(self.clone()); - let m = SpeedGradeEnergyModel::try_from((arc_self, query))?; - Ok(Arc::new(m)) - } + pub vehicle_library: HashMap<String, Arc<dyn VehicleType>>, } -impl SpeedGradeEnergyModelService { +impl EnergyModelService { pub fn new<P: AsRef<Path>>( speed_table_path: &P, speeds_table_speed_unit: SpeedUnit, @@ -42,7 +32,7 @@ impl SpeedGradeEnergyModelService { grade_table_grade_unit_option: Option<GradeUnit>, output_time_unit_option: Option<TimeUnit>, output_distance_unit_option: Option<DistanceUnit>, - energy_model_library: HashMap<String, Arc<SpeedGradePredictionModelRecord>>, + vehicle_library: HashMap<String, Arc<dyn VehicleType>>, ) -> Result<Self, TraversalModelError> { let output_time_unit = output_time_unit_option.unwrap_or(BASE_TIME_UNIT); let output_distance_unit = output_distance_unit_option.unwrap_or(BASE_DISTANCE_UNIT); @@ -74,7 +64,7 @@ impl SpeedGradeEnergyModelService { let max_speed = get_max_speed(&speed_table)?; - Ok(SpeedGradeEnergyModelService { + Ok(EnergyModelService { speed_table, speeds_table_speed_unit, max_speed, @@ -82,7 +72,18 @@ impl SpeedGradeEnergyModelService { grade_table_grade_unit, output_time_unit, output_distance_unit, - energy_model_library: Arc::new(energy_model_library), + vehicle_library, }) } } + +impl TraversalModelService for EnergyModelService { + fn build( + &self, + parameters: &serde_json::Value, + ) -> Result<Arc<dyn TraversalModel>, TraversalModelError> { + let arc_self = Arc::new(self.clone()); + let model = EnergyTraversalModel::try_from((arc_self, parameters))?; + Ok(Arc::new(model)) + } +} diff --git a/rust/routee-compass-powertrain/src/routee/speed_grade_energy_model.rs b/rust/routee-compass-powertrain/src/routee/energy_traversal_model.rs similarity index 69% rename from rust/routee-compass-powertrain/src/routee/speed_grade_energy_model.rs rename to rust/routee-compass-powertrain/src/routee/energy_traversal_model.rs index 52d45757..a43dc69b 100644 --- a/rust/routee-compass-powertrain/src/routee/speed_grade_energy_model.rs +++ b/rust/routee-compass-powertrain/src/routee/energy_traversal_model.rs @@ -1,9 +1,12 @@ -use super::prediction_model::SpeedGradePredictionModelRecord; -use super::speed_grade_energy_model_service::SpeedGradeEnergyModelService; +use crate::routee::energy_model_ops::ZERO_ENERGY; + +use super::energy_model_ops::get_grade; +use super::energy_model_service::EnergyModelService; +use super::vehicle::vehicle_type::{VehicleState, VehicleType}; + use routee_compass_core::model::cost::Cost; use routee_compass_core::model::property::edge::Edge; use routee_compass_core::model::property::vertex::Vertex; -use routee_compass_core::model::road_network::edge_id::EdgeId; use routee_compass_core::model::traversal::default::speed_lookup_model::get_speed; use routee_compass_core::model::traversal::state::state_variable::StateVar; use routee_compass_core::model::traversal::state::traversal_state::TraversalState; @@ -15,18 +18,22 @@ use routee_compass_core::util::unit::as_f64::AsF64; use routee_compass_core::util::unit::*; use std::sync::Arc; -const ZERO_ENERGY: f64 = 1e-9; - -pub struct SpeedGradeEnergyModel { - pub service: Arc<SpeedGradeEnergyModelService>, - pub model_record: Arc<SpeedGradePredictionModelRecord>, +pub struct EnergyTraversalModel { + pub service: Arc<EnergyModelService>, + pub vehicle: Arc<dyn VehicleType>, pub energy_cost_coefficient: f64, } -impl TraversalModel for SpeedGradeEnergyModel { +impl TraversalModel for EnergyTraversalModel { fn initial_state(&self) -> TraversalState { - // distance, time, energy - vec![StateVar(0.0), StateVar(0.0), StateVar(0.0)] + // distance, time + let mut initial_state = vec![StateVar(0.0), StateVar(0.0)]; + + // vehicle state gets slots 2..n + let vehicle_state = self.vehicle.initial_state(); + initial_state.extend(vehicle_state); + + initial_state } /// estimate the cost of traveling between two vertices. /// given a distance estimate, @@ -47,12 +54,9 @@ impl TraversalModel for SpeedGradeEnergyModel { return Ok(Cost::ZERO); } - let (energy, _energy_unit) = Energy::create( - self.model_record.ideal_energy_rate, - self.model_record.energy_rate_unit, - distance, - self.service.output_distance_unit, - )?; + let (energy, _energy_unit) = self + .vehicle + .best_case_energy((distance, self.service.output_distance_unit))?; let time: Time = Time::create( self.service.max_speed, @@ -85,29 +89,24 @@ impl TraversalModel for SpeedGradeEnergyModel { self.service.output_time_unit.clone(), )?; - let (energy_rate, _energy_rate_unit) = self.model_record.prediction_model.predict( - speed, - self.service.speeds_table_speed_unit, - grade, - self.service.grade_table_grade_unit, + let energy_result = self.vehicle.consume_energy( + (speed, self.service.speeds_table_speed_unit), + (grade, self.service.grade_table_grade_unit), + (distance, self.service.output_distance_unit), + get_vehicle_state_from_state(state), )?; - let energy_rate_real_world = energy_rate * self.model_record.real_world_energy_adjustment; - - let (mut energy, _energy_unit) = Energy::create( - energy_rate_real_world, - self.model_record.energy_rate_unit, - distance, - self.service.output_distance_unit, - )?; + let mut energy = energy_result.energy; + // for now we need to truncate the energy at zero until we can handle these being negative if energy.as_f64() < 0.0 { energy = Energy::new(ZERO_ENERGY); log::debug!("negative energy encountered, setting to 1e-9"); } let total_cost = create_cost(energy, time, self.energy_cost_coefficient); - let updated_state = update_state(state, distance, time, energy); + + let updated_state = update_state(state, distance, time, energy_result.updated_state); let result = TraversalResult { total_cost, updated_state, @@ -118,30 +117,30 @@ impl TraversalModel for SpeedGradeEnergyModel { fn serialize_state(&self, state: &TraversalState) -> serde_json::Value { let distance = get_distance_from_state(state); let time = get_time_from_state(state); - let energy = get_energy_from_state(state); + let vehicle_state = get_vehicle_state_from_state(state); + let vehicle_state_summary = self.vehicle.serialize_state(vehicle_state); serde_json::json!({ "distance": distance, "time": time, - "energy": energy, + "vehicle": vehicle_state_summary, }) } - fn serialize_state_info(&self, _state: &TraversalState) -> serde_json::Value { - let energy_unit = self.model_record.energy_rate_unit.associated_energy_unit(); + fn serialize_state_info(&self, state: &TraversalState) -> serde_json::Value { + let vehicle_state = get_vehicle_state_from_state(state); + let vehicle_state_info = self.vehicle.serialize_state_info(vehicle_state); serde_json::json!({ "distance_unit": self.service.output_distance_unit, "time_unit": self.service.output_time_unit, - "energy_unit": energy_unit, + "vehicle_info": vehicle_state_info, }) } } -impl TryFrom<(Arc<SpeedGradeEnergyModelService>, &serde_json::Value)> for SpeedGradeEnergyModel { +impl TryFrom<(Arc<EnergyModelService>, &serde_json::Value)> for EnergyTraversalModel { type Error = TraversalModelError; - fn try_from( - input: (Arc<SpeedGradeEnergyModelService>, &serde_json::Value), - ) -> Result<Self, Self::Error> { + fn try_from(input: (Arc<EnergyModelService>, &serde_json::Value)) -> Result<Self, Self::Error> { let (service, conf) = input; let energy_cost_coefficient = match conf.get(String::from("energy_cost_coefficient")) { @@ -162,7 +161,6 @@ impl TryFrom<(Arc<SpeedGradeEnergyModelService>, &serde_json::Value)> for SpeedG } } }; - let prediction_model_name = conf .get("model_name".to_string()) .ok_or(TraversalModelError::BuildError( @@ -174,20 +172,21 @@ impl TryFrom<(Arc<SpeedGradeEnergyModelService>, &serde_json::Value)> for SpeedG ))? .to_string(); - let model_record = match service.energy_model_library.get(&prediction_model_name) { + let vehicle = match service.vehicle_library.get(&prediction_model_name) { None => { - let model_names: Vec<&String> = service.energy_model_library.keys().collect(); - return Err(TraversalModelError::BuildError(format!( - "No energy model found with model_name = '{}', try one of: {:?}", + let model_names: Vec<&String> = service.vehicle_library.keys().collect(); + Err(TraversalModelError::BuildError(format!( + "No vehicle found with model_name = '{}', try one of: {:?}", prediction_model_name, model_names - ))); + ))) } - Some(mr) => mr.clone(), - }; + Some(mr) => Ok(mr.clone()), + }? + .update_from_query(conf)?; - Ok(SpeedGradeEnergyModel { + Ok(EnergyTraversalModel { service, - model_record, + vehicle, energy_cost_coefficient, }) } @@ -206,12 +205,13 @@ fn update_state( state: &TraversalState, distance: Distance, time: Time, - energy: Energy, + vehicle_state: VehicleState, ) -> TraversalState { - let mut updated_state = state.clone(); - updated_state[0] = state[0] + distance.into(); - updated_state[1] = state[1] + time.into(); - updated_state[2] = state[2] + energy.into(); + let mut updated_state = Vec::new(); + + updated_state.push(state[0] + distance.into()); + updated_state.push(state[1] + time.into()); + updated_state.extend(vehicle_state); updated_state } @@ -223,33 +223,16 @@ fn get_time_from_state(state: &TraversalState) -> Time { Time::new(state[1].0) } -fn get_energy_from_state(state: &TraversalState) -> Energy { - Energy::new(state[2].0) -} - -/// look up the grade from the grade table -pub fn get_grade( - grade_table: &Option<Box<[Grade]>>, - edge_id: EdgeId, -) -> Result<Grade, TraversalModelError> { - match grade_table { - None => Ok(Grade::ZERO), - Some(gt) => { - let grade: &Grade = gt.get(edge_id.as_usize()).ok_or( - TraversalModelError::MissingIdInTabularCostFunction( - format!("{}", edge_id), - String::from("EdgeId"), - String::from("grade table"), - ), - )?; - Ok(*grade) - } - } +fn get_vehicle_state_from_state(state: &TraversalState) -> &[StateVar] { + &state[2..] } #[cfg(test)] mod tests { - use crate::routee::model_type::ModelType; + use crate::routee::{ + prediction::load_prediction_model, prediction::model_type::ModelType, + vehicle::default::single_fuel_vehicle::SingleFuelVehicle, + }; use super::*; use geo::coord; @@ -288,7 +271,7 @@ mod tests { distance: Distance::new(100.0), } } - let model_record = SpeedGradePredictionModelRecord::new( + let model_record = load_prediction_model( "Toyota_Camry".to_string(), &model_file_path, ModelType::Smartcore, @@ -299,10 +282,13 @@ mod tests { None, ) .unwrap(); - let mut model_library = HashMap::new(); - model_library.insert("Toyota_Camry".to_string(), Arc::new(model_record)); - let service = SpeedGradeEnergyModelService::new( + let camry = SingleFuelVehicle::new("Toyota_Camry".to_string(), model_record).unwrap(); + + let mut model_library: HashMap<String, Arc<dyn VehicleType>> = HashMap::new(); + model_library.insert("Toyota_Camry".to_string(), Arc::new(camry)); + + let service = EnergyModelService::new( &speed_file_path, SpeedUnit::KilometersPerHour, &Some(grade_file_path), @@ -317,7 +303,7 @@ mod tests { "model_name": "Toyota_Camry", "energy_cost_coefficient": 0.5, }); - let model = SpeedGradeEnergyModel::try_from((arc_service, &conf)).unwrap(); + let model = EnergyTraversalModel::try_from((arc_service, &conf)).unwrap(); let initial = model.initial_state(); let e1 = mock_edge(0); // 100 meters @ 10kph should take 36 seconds ((0.1/10) * 3600) diff --git a/rust/routee-compass-powertrain/src/routee/mod.rs b/rust/routee-compass-powertrain/src/routee/mod.rs index 3b546c57..e3eb32d0 100644 --- a/rust/routee-compass-powertrain/src/routee/mod.rs +++ b/rust/routee-compass-powertrain/src/routee/mod.rs @@ -1,8 +1,5 @@ -pub mod model_type; -pub mod prediction_model; -pub mod smartcore; -pub mod speed_grade_energy_model; -pub mod speed_grade_energy_model_service; - -#[cfg(feature = "onnx")] -pub mod onnx; +pub mod energy_model_ops; +pub mod energy_model_service; +pub mod energy_traversal_model; +pub mod prediction; +pub mod vehicle; diff --git a/rust/routee-compass-powertrain/src/routee/prediction/mod.rs b/rust/routee-compass-powertrain/src/routee/prediction/mod.rs new file mode 100644 index 00000000..05a8a2a4 --- /dev/null +++ b/rust/routee-compass-powertrain/src/routee/prediction/mod.rs @@ -0,0 +1,12 @@ +pub mod model_type; +pub mod prediction_model; +pub mod prediction_model_ops; +pub mod prediction_model_record; +pub mod smartcore; + +#[cfg(feature = "onnx")] +pub mod onnx; + +pub use prediction_model::PredictionModel; +pub use prediction_model_ops::load_prediction_model; +pub use prediction_model_record::PredictionModelRecord; diff --git a/rust/routee-compass-powertrain/src/routee/model_type.rs b/rust/routee-compass-powertrain/src/routee/prediction/model_type.rs similarity index 91% rename from rust/routee-compass-powertrain/src/routee/model_type.rs rename to rust/routee-compass-powertrain/src/routee/prediction/model_type.rs index 1799c9f8..6d2bc210 100644 --- a/rust/routee-compass-powertrain/src/routee/model_type.rs +++ b/rust/routee-compass-powertrain/src/routee/prediction/model_type.rs @@ -1,5 +1,5 @@ use super::{ - prediction_model::SpeedGradePredictionModel, + prediction_model::PredictionModel, smartcore::smartcore_speed_grade_model::SmartcoreSpeedGradeModel, }; use routee_compass_core::{ @@ -34,9 +34,9 @@ impl ModelType { energy_model_speed_unit: SpeedUnit, energy_model_grade_unit: GradeUnit, energy_model_energy_rate_unit: EnergyRateUnit, - ) -> Result<Arc<dyn SpeedGradePredictionModel>, TraversalModelError> { + ) -> Result<Arc<dyn PredictionModel>, TraversalModelError> { // Load random forest binary file - let model: Arc<dyn SpeedGradePredictionModel> = match self { + let model: Arc<dyn PredictionModel> = match self { ModelType::Smartcore => Arc::new(SmartcoreSpeedGradeModel::new( energy_model_path.clone(), energy_model_speed_unit, diff --git a/rust/routee-compass-powertrain/src/routee/onnx/mod.rs b/rust/routee-compass-powertrain/src/routee/prediction/onnx/mod.rs similarity index 100% rename from rust/routee-compass-powertrain/src/routee/onnx/mod.rs rename to rust/routee-compass-powertrain/src/routee/prediction/onnx/mod.rs diff --git a/rust/routee-compass-powertrain/src/routee/onnx/onnx_speed_grade_model.rs b/rust/routee-compass-powertrain/src/routee/prediction/onnx/onnx_speed_grade_model.rs similarity index 87% rename from rust/routee-compass-powertrain/src/routee/onnx/onnx_speed_grade_model.rs rename to rust/routee-compass-powertrain/src/routee/prediction/onnx/onnx_speed_grade_model.rs index c155d925..9025914d 100644 --- a/rust/routee-compass-powertrain/src/routee/onnx/onnx_speed_grade_model.rs +++ b/rust/routee-compass-powertrain/src/routee/prediction/onnx/onnx_speed_grade_model.rs @@ -1,6 +1,6 @@ use std::path::Path; -use crate::routee::prediction_model::SpeedGradePredictionModel; +use crate::routee::prediction::prediction_model::PredictionModel; use ndarray::CowArray; use ort::{ tensor::OrtOwnedTensor, Environment, GraphOptimizationLevel, Session, SessionBuilder, Value, @@ -17,14 +17,14 @@ pub struct OnnxSpeedGradeModel { energy_rate_unit: EnergyRateUnit, } -impl SpeedGradePredictionModel for OnnxSpeedGradeModel { +impl PredictionModel for OnnxSpeedGradeModel { fn predict( &self, - speed: Speed, - speed_unit: SpeedUnit, - grade: Grade, - grade_unit: GradeUnit, + speed: (Speed, SpeedUnit), + grade: (Grade, GradeUnit), ) -> Result<(EnergyRate, EnergyRateUnit), TraversalModelError> { + let (speed, speed_unit) = speed; + let (grade, grade_unit) = grade; let speed_value: f32 = speed_unit.convert(speed, self.speed_unit).as_f64() as f32; let grade_value: f32 = grade_unit.convert(grade, self.grade_unit).as_f64() as f32; let array = ndarray::Array1::from(vec![speed_value, grade_value]) @@ -89,8 +89,7 @@ mod test { use std::path::PathBuf; use crate::routee::{ - onnx::onnx_speed_grade_model::OnnxSpeedGradeModel, - prediction_model::SpeedGradePredictionModel, + prediction::onnx::onnx_speed_grade_model::OnnxSpeedGradeModel, prediction::PredictionModel, }; use rayon::prelude::*; use routee_compass_core::{ @@ -106,13 +105,10 @@ mod test { .join("src") .join("routee") .join("test") - .join("Toyota_Camry.onnx") - .to_str() - .unwrap() - .into(); - let model: Box<dyn SpeedGradePredictionModel> = Box::new( + .join("Toyota_Camry.onnx"); + let model: Box<dyn PredictionModel> = Box::new( OnnxSpeedGradeModel::new( - model_file_path, + &model_file_path, SpeedUnit::MilesPerHour, GradeUnit::Decimal, routee_compass_core::util::unit::EnergyRateUnit::GallonsGasolinePerMile, @@ -132,7 +128,7 @@ mod test { let results = inputs .par_iter() .map(|(speed, speed_unit, grade, grade_unit)| { - model.predict(*speed, *speed_unit, *grade, *grade_unit) + model.predict((*speed, *speed_unit), (*grade, *grade_unit)) }) .collect::<Vec<Result<(EnergyRate, EnergyRateUnit), TraversalModelError>>>(); @@ -141,7 +137,10 @@ mod test { // assert that all the results are the same let (expected_er, expected_eru) = model - .predict(input_speed, input_speed_unit, input_grade, input_grade_unit) + .predict( + (input_speed, input_speed_unit), + (input_grade, input_grade_unit), + ) .unwrap(); assert!(results.iter().all(|r| match r { Err(e) => panic!("{}", e), diff --git a/rust/routee-compass-powertrain/src/routee/prediction/prediction_model.rs b/rust/routee-compass-powertrain/src/routee/prediction/prediction_model.rs new file mode 100644 index 00000000..166365cc --- /dev/null +++ b/rust/routee-compass-powertrain/src/routee/prediction/prediction_model.rs @@ -0,0 +1,12 @@ +use routee_compass_core::{ + model::traversal::traversal_model_error::TraversalModelError, + util::unit::{EnergyRate, EnergyRateUnit, Grade, GradeUnit, Speed, SpeedUnit}, +}; + +pub trait PredictionModel: Send + Sync { + fn predict( + &self, + speed: (Speed, SpeedUnit), + grade: (Grade, GradeUnit), + ) -> Result<(EnergyRate, EnergyRateUnit), TraversalModelError>; +} diff --git a/rust/routee-compass-powertrain/src/routee/prediction/prediction_model_ops.rs b/rust/routee-compass-powertrain/src/routee/prediction/prediction_model_ops.rs new file mode 100644 index 00000000..f825f724 --- /dev/null +++ b/rust/routee-compass-powertrain/src/routee/prediction/prediction_model_ops.rs @@ -0,0 +1,106 @@ +use std::{path::Path, sync::Arc}; + +use routee_compass_core::{ + model::traversal::traversal_model_error::TraversalModelError, + util::unit::{EnergyRate, EnergyRateUnit, Grade, GradeUnit, Speed, SpeedUnit}, +}; + +use super::{ + model_type::ModelType, smartcore::smartcore_speed_grade_model::SmartcoreSpeedGradeModel, + PredictionModel, PredictionModelRecord, +}; + +#[cfg(feature = "onnx")] +use crate::routee::prediction::onnx::onnx_speed_grade_model::OnnxSpeedGradeModel; + +#[allow(clippy::too_many_arguments)] +pub fn load_prediction_model<P: AsRef<Path>>( + model_name: String, + model_path: &P, + model_type: ModelType, + speed_unit: SpeedUnit, + grade_unit: GradeUnit, + energy_rate_unit: EnergyRateUnit, + ideal_energy_rate_option: Option<EnergyRate>, + real_world_energy_adjustment_option: Option<f64>, +) -> Result<PredictionModelRecord, TraversalModelError> { + let prediction_model: Arc<dyn PredictionModel> = match model_type { + ModelType::Smartcore => { + let model = SmartcoreSpeedGradeModel::new( + model_path, + speed_unit, + grade_unit, + energy_rate_unit, + )?; + Arc::new(model) + } + ModelType::Onnx => { + #[cfg(feature = "onnx")] + { + let model = + OnnxSpeedGradeModel::new(model_path, speed_unit, grade_unit, energy_rate_unit)?; + Arc::new(model) + } + #[cfg(not(feature = "onnx"))] + { + return Err(TraversalModelError::BuildError( + "Cannot build Onnx model without `onnx` feature enabled for compass-powertrain" + .to_string(), + )); + } + } + }; + let ideal_energy_rate = match ideal_energy_rate_option { + None => find_min_energy_rate(&prediction_model, &energy_rate_unit)?, + Some(ier) => ier, + }; + + let real_world_energy_adjustment = real_world_energy_adjustment_option.unwrap_or(1.0); + + Ok(PredictionModelRecord { + name: model_name, + prediction_model, + model_type, + speed_unit, + grade_unit, + energy_rate_unit, + ideal_energy_rate, + real_world_energy_adjustment, + }) +} + +/// sweep a fixed set of speed and grade values to find the minimum energy per mile rate from the incoming rf model +pub fn find_min_energy_rate( + model: &Arc<dyn PredictionModel>, + energy_model_energy_rate_unit: &EnergyRateUnit, +) -> Result<EnergyRate, TraversalModelError> { + // sweep a fixed set of speed and grade values to find the minimum energy per mile rate from the incoming rf model + let mut minimum_energy_rate = EnergyRate::new(f64::MAX); + let start_time = std::time::Instant::now(); + + let grade = Grade::ZERO; + for speed_i32 in 20..80 { + let speed = Speed::new(speed_i32 as f64); + let (energy_rate, _) = model + .predict( + (speed, SpeedUnit::MilesPerHour), + (grade, GradeUnit::Percent), + ) + .map_err(|e| TraversalModelError::PredictionModel(e.to_string()))?; + if energy_rate < minimum_energy_rate { + minimum_energy_rate = energy_rate; + } + } + + let end_time = std::time::Instant::now(); + let search_time = end_time - start_time; + + log::debug!( + "found minimum energy: {}/{} in {} milliseconds", + minimum_energy_rate, + energy_model_energy_rate_unit, + search_time.as_millis() + ); + + Ok(minimum_energy_rate) +} diff --git a/rust/routee-compass-powertrain/src/routee/prediction/prediction_model_record.rs b/rust/routee-compass-powertrain/src/routee/prediction/prediction_model_record.rs new file mode 100644 index 00000000..1e97490b --- /dev/null +++ b/rust/routee-compass-powertrain/src/routee/prediction/prediction_model_record.rs @@ -0,0 +1,44 @@ +use std::sync::Arc; + +use routee_compass_core::{ + model::traversal::traversal_model_error::TraversalModelError, + util::unit::{ + Distance, DistanceUnit, Energy, EnergyRate, EnergyRateUnit, EnergyUnit, Grade, GradeUnit, + Speed, SpeedUnit, + }, +}; + +use super::{model_type::ModelType, PredictionModel}; +/// A struct to hold the prediction model and associated metadata +pub struct PredictionModelRecord { + pub name: String, + pub prediction_model: Arc<dyn PredictionModel>, + pub model_type: ModelType, + pub speed_unit: SpeedUnit, + pub grade_unit: GradeUnit, + pub energy_rate_unit: EnergyRateUnit, + pub ideal_energy_rate: EnergyRate, + pub real_world_energy_adjustment: f64, +} + +impl PredictionModelRecord { + pub fn predict( + &self, + speed: (Speed, SpeedUnit), + grade: (Grade, GradeUnit), + distance: (Distance, DistanceUnit), + ) -> Result<(Energy, EnergyUnit), TraversalModelError> { + let (distance, distance_unit) = distance; + let (energy_rate, _energy_rate_unit) = self.prediction_model.predict(speed, grade)?; + + let energy_rate_real_world = energy_rate * self.real_world_energy_adjustment; + + let (energy, energy_unit) = Energy::create( + energy_rate_real_world, + self.energy_rate_unit, + distance, + distance_unit, + )?; + Ok((energy, energy_unit)) + } +} diff --git a/rust/routee-compass-powertrain/src/routee/smartcore/mod.rs b/rust/routee-compass-powertrain/src/routee/prediction/smartcore/mod.rs similarity index 100% rename from rust/routee-compass-powertrain/src/routee/smartcore/mod.rs rename to rust/routee-compass-powertrain/src/routee/prediction/smartcore/mod.rs diff --git a/rust/routee-compass-powertrain/src/routee/smartcore/smartcore_speed_grade_model.rs b/rust/routee-compass-powertrain/src/routee/prediction/smartcore/smartcore_speed_grade_model.rs similarity index 88% rename from rust/routee-compass-powertrain/src/routee/smartcore/smartcore_speed_grade_model.rs rename to rust/routee-compass-powertrain/src/routee/prediction/smartcore/smartcore_speed_grade_model.rs index 6ce8d916..dede8ef4 100644 --- a/rust/routee-compass-powertrain/src/routee/smartcore/smartcore_speed_grade_model.rs +++ b/rust/routee-compass-powertrain/src/routee/prediction/smartcore/smartcore_speed_grade_model.rs @@ -1,6 +1,6 @@ use std::path::Path; -use crate::routee::prediction_model::SpeedGradePredictionModel; +use crate::routee::prediction::prediction_model::PredictionModel; use routee_compass_core::{ model::traversal::traversal_model_error::TraversalModelError, util::unit::{as_f64::AsF64, EnergyRate, EnergyRateUnit, Grade, GradeUnit, Speed, SpeedUnit}, @@ -16,14 +16,14 @@ pub struct SmartcoreSpeedGradeModel { energy_rate_unit: EnergyRateUnit, } -impl SpeedGradePredictionModel for SmartcoreSpeedGradeModel { +impl PredictionModel for SmartcoreSpeedGradeModel { fn predict( &self, - speed: Speed, - speed_unit: SpeedUnit, - grade: Grade, - grade_unit: GradeUnit, + speed: (Speed, SpeedUnit), + grade: (Grade, GradeUnit), ) -> Result<(EnergyRate, EnergyRateUnit), TraversalModelError> { + let (speed, speed_unit) = speed; + let (grade, grade_unit) = grade; let speed_value = speed_unit.convert(speed, self.speed_unit).as_f64(); let grade_value = grade_unit.convert(grade, self.grade_unit).as_f64(); let x = DenseMatrix::from_2d_vec(&vec![vec![speed_value, grade_value]]); diff --git a/rust/routee-compass-powertrain/src/routee/prediction_model.rs b/rust/routee-compass-powertrain/src/routee/prediction_model.rs deleted file mode 100644 index dcc7f18e..00000000 --- a/rust/routee-compass-powertrain/src/routee/prediction_model.rs +++ /dev/null @@ -1,132 +0,0 @@ -use std::{path::Path, sync::Arc}; - -use routee_compass_core::{ - model::traversal::traversal_model_error::TraversalModelError, - util::unit::{EnergyRate, EnergyRateUnit, Grade, GradeUnit, Speed, SpeedUnit}, -}; - -use super::{ - model_type::ModelType, smartcore::smartcore_speed_grade_model::SmartcoreSpeedGradeModel, -}; - -#[cfg(feature = "onnx")] -use crate::routee::onnx::onnx_speed_grade_model::OnnxSpeedGradeModel; - -pub struct SpeedGradePredictionModelRecord { - pub name: String, - pub prediction_model: Arc<dyn SpeedGradePredictionModel>, - pub model_type: ModelType, - pub speed_unit: SpeedUnit, - pub grade_unit: GradeUnit, - pub energy_rate_unit: EnergyRateUnit, - pub ideal_energy_rate: EnergyRate, - pub real_world_energy_adjustment: f64, -} - -impl SpeedGradePredictionModelRecord { - #[allow(clippy::too_many_arguments)] - pub fn new<P: AsRef<Path>>( - name: String, - model_path: &P, - model_type: ModelType, - speed_unit: SpeedUnit, - grade_unit: GradeUnit, - energy_rate_unit: EnergyRateUnit, - ideal_energy_rate_option: Option<EnergyRate>, - real_world_energy_adjustment_option: Option<f64>, - ) -> Result<Self, TraversalModelError> { - // Load random forest binary file - let prediction_model: Arc<dyn SpeedGradePredictionModel> = match model_type { - ModelType::Smartcore => { - let model = SmartcoreSpeedGradeModel::new( - model_path, - speed_unit, - grade_unit, - energy_rate_unit, - )?; - Arc::new(model) - } - ModelType::Onnx => { - #[cfg(feature = "onnx")] - { - let model = OnnxSpeedGradeModel::new( - model_path, - speed_unit, - grade_unit, - energy_rate_unit, - )?; - Arc::new(model) - } - #[cfg(not(feature = "onnx"))] - { - return Err( - TraversalModelError::BuildError( - "Cannot build Onnx model without `onnx` feature enabled for compass-powertrain" - .to_string(), - ) - ); - } - } - }; - let ideal_energy_rate = match ideal_energy_rate_option { - None => find_min_energy_rate(&prediction_model, &energy_rate_unit)?, - Some(ier) => ier, - }; - - let real_world_energy_adjustment = real_world_energy_adjustment_option.unwrap_or(1.0); - - Ok(Self { - name, - prediction_model, - model_type, - speed_unit, - grade_unit, - energy_rate_unit, - ideal_energy_rate, - real_world_energy_adjustment, - }) - } -} - -pub trait SpeedGradePredictionModel: Send + Sync { - fn predict( - &self, - speed: Speed, - speed_unit: SpeedUnit, - grade: Grade, - grade_unit: GradeUnit, - ) -> Result<(EnergyRate, EnergyRateUnit), TraversalModelError>; -} - -/// sweep a fixed set of speed and grade values to find the minimum energy per mile rate from the incoming rf model -pub fn find_min_energy_rate( - model: &Arc<dyn SpeedGradePredictionModel>, - energy_model_energy_rate_unit: &EnergyRateUnit, -) -> Result<EnergyRate, TraversalModelError> { - // sweep a fixed set of speed and grade values to find the minimum energy per mile rate from the incoming rf model - let mut minimum_energy_rate = EnergyRate::new(f64::MAX); - let start_time = std::time::Instant::now(); - - let grade = Grade::ZERO; - for speed_i32 in 20..80 { - let speed = Speed::new(speed_i32 as f64); - let (energy_rate, _) = model - .predict(speed, SpeedUnit::MilesPerHour, grade, GradeUnit::Percent) - .map_err(|e| TraversalModelError::PredictionModel(e.to_string()))?; - if energy_rate < minimum_energy_rate { - minimum_energy_rate = energy_rate; - } - } - - let end_time = std::time::Instant::now(); - let search_time = end_time - start_time; - - log::debug!( - "found minimum energy: {}/{} in {} milliseconds", - minimum_energy_rate, - energy_model_energy_rate_unit, - search_time.as_millis() - ); - - Ok(minimum_energy_rate) -} diff --git a/rust/routee-compass-powertrain/src/routee/vehicle/default/dual_fuel_vehicle.rs b/rust/routee-compass-powertrain/src/routee/vehicle/default/dual_fuel_vehicle.rs new file mode 100644 index 00000000..33b41057 --- /dev/null +++ b/rust/routee-compass-powertrain/src/routee/vehicle/default/dual_fuel_vehicle.rs @@ -0,0 +1,365 @@ +use std::sync::Arc; + +use routee_compass_core::{ + model::traversal::{ + state::{state_variable::StateVar, traversal_state::TraversalState}, + traversal_model_error::TraversalModelError, + }, + util::unit::{ + as_f64::AsF64, Distance, DistanceUnit, Energy, EnergyUnit, Grade, GradeUnit, Speed, + SpeedUnit, + }, +}; + +use crate::routee::{ + prediction::PredictionModelRecord, + vehicle::{vehicle_type::VehicleType, VehicleEnergyResult}, +}; + +pub struct DualFuelVehicle { + pub name: String, + pub charge_sustain_model: Arc<PredictionModelRecord>, + pub charge_depleting_model: Arc<PredictionModelRecord>, + pub battery_capacity: Energy, + pub starting_battery_energy: Energy, + pub battery_energy_unit: EnergyUnit, +} + +impl DualFuelVehicle { + pub fn new( + name: String, + charge_sustain_model: PredictionModelRecord, + charge_depleting_model: PredictionModelRecord, + battery_capacity: Energy, + starting_battery_energy: Energy, + battery_energy_unit: EnergyUnit, + ) -> Result<Self, TraversalModelError> { + Ok(Self { + name, + charge_sustain_model: Arc::new(charge_sustain_model), + charge_depleting_model: Arc::new(charge_depleting_model), + battery_capacity, + starting_battery_energy, + battery_energy_unit, + }) + } +} + +impl VehicleType for DualFuelVehicle { + fn name(&self) -> String { + self.name.clone() + } + fn initial_state(&self) -> TraversalState { + vec![ + StateVar(0.0), // accumulated electrical energy + StateVar(0.0), // accumulated gasoline energy + StateVar(self.starting_battery_energy.as_f64()), // battery energy remaining + ] + } + fn best_case_energy( + &self, + distance: (Distance, DistanceUnit), + ) -> Result<(Energy, EnergyUnit), TraversalModelError> { + let (distance, distance_unit) = distance; + + // assume lowest energy cost scenario for a PHEV is to just use the battery + let energy = Energy::create( + self.charge_depleting_model.ideal_energy_rate, + self.charge_depleting_model.energy_rate_unit, + distance, + distance_unit, + )?; + Ok(energy) + } + fn consume_energy( + &self, + speed: (Speed, SpeedUnit), + grade: (Grade, GradeUnit), + distance: (Distance, DistanceUnit), + state: &[StateVar], + ) -> Result<VehicleEnergyResult, TraversalModelError> { + let battery_soc_percentage = get_battery_soc_percent(self, state); + + let (electrical_energy, electrical_energy_unit, gasoline_energy, gasoline_energy_unit) = + get_phev_energy(self, battery_soc_percentage, speed, grade, distance)?; + + // convert both energy sources to kWh + let electrical_energy_kwh = + electrical_energy_unit.convert(electrical_energy, EnergyUnit::KilowattHours); + let gasoline_energy_kwh = + gasoline_energy_unit.convert(gasoline_energy, EnergyUnit::KilowattHours); + let total_energy_kwh = electrical_energy_kwh + gasoline_energy_kwh; + + let updated_state = update_state(state, electrical_energy, gasoline_energy); + + Ok(VehicleEnergyResult { + energy: total_energy_kwh, + energy_unit: EnergyUnit::KilowattHours, + updated_state, + }) + } + fn serialize_state(&self, state: &[StateVar]) -> serde_json::Value { + let battery_energy = get_electrical_energy_from_state(state); + let gasoline_energy = get_gasoline_energy_from_state(state); + let battery_soc_percent = get_battery_soc_percent(self, state); + serde_json::json!({ + "battery_energy": battery_energy.as_f64(), + "fuel_energy": gasoline_energy.as_f64(), + "battery_soc_percent": battery_soc_percent, + }) + } + + fn serialize_state_info(&self, _state: &[StateVar]) -> serde_json::Value { + let battery_energy_unit = self.battery_energy_unit; + let fuel_energy_unit = self + .charge_sustain_model + .energy_rate_unit + .associated_energy_unit(); + serde_json::json!({ + "battery_energy_unit": battery_energy_unit.to_string(), + "fuel_energy_unit": fuel_energy_unit.to_string(), + }) + } + + fn update_from_query( + &self, + query: &serde_json::Value, + ) -> Result<Arc<dyn VehicleType>, TraversalModelError> { + let starting_soc_percent = query + .get("starting_soc_percent".to_string()) + .ok_or(TraversalModelError::BuildError( + "No 'starting_soc_percent' key provided in query".to_string(), + ))? + .as_f64() + .ok_or(TraversalModelError::BuildError( + "Expected 'starting_soc_percent' value to be numeric".to_string(), + ))?; + if !(0.0..=100.0).contains(&starting_soc_percent) { + return Err(TraversalModelError::BuildError( + "Expected 'starting_soc_percent' value to be between 0 and 100".to_string(), + )); + } + let starting_battery_energy = self.battery_capacity * (starting_soc_percent / 100.0); + + let new_phev = DualFuelVehicle { + name: self.name.clone(), + charge_sustain_model: self.charge_sustain_model.clone(), + charge_depleting_model: self.charge_depleting_model.clone(), + battery_capacity: self.battery_capacity, + starting_battery_energy, + battery_energy_unit: self.battery_energy_unit, + }; + + Ok(Arc::new(new_phev)) + } +} + +fn update_state( + state: &[StateVar], + electrical_energy: Energy, + gasoline_energy: Energy, +) -> TraversalState { + let mut updated_state = Vec::with_capacity(state.len()); + + // accumulated electrical energy + updated_state.push(state[0] + electrical_energy.into()); + + // accumulated fuel energy + updated_state.push(state[1] + gasoline_energy.into()); + + // remaining battery energy + let current_battery_energy = get_remaining_battery_energy_from_state(state); + let new_battery_energy = (current_battery_energy - electrical_energy).max(Energy::new(0.0)); + updated_state.push(new_battery_energy.into()); + + updated_state +} +fn get_electrical_energy_from_state(state: &[StateVar]) -> Energy { + Energy::new(state[0].0) +} + +fn get_gasoline_energy_from_state(state: &[StateVar]) -> Energy { + Energy::new(state[1].0) +} + +fn get_remaining_battery_energy_from_state(state: &[StateVar]) -> Energy { + Energy::new(state[2].0) +} + +fn get_battery_soc_percent(vehicle: &DualFuelVehicle, state: &[StateVar]) -> f64 { + let battery_energy_unit = vehicle.battery_energy_unit; + + let battery_capacity_kwh = + battery_energy_unit.convert(vehicle.battery_capacity, EnergyUnit::KilowattHours); + + let remaining_battery_energy = get_remaining_battery_energy_from_state(state); + + let remaining_battery_energy_kwh = + battery_energy_unit.convert(remaining_battery_energy, EnergyUnit::KilowattHours); + + (remaining_battery_energy_kwh.as_f64() / battery_capacity_kwh.as_f64()) * 100.0 +} + +/// Compute the energy for the PHEV by converting gasoline to kWh. +/// This uses a simplified operation in which we assume that if the battery +/// SOC is greater than zero we can just operate on battery to traverse a link. +/// This is not entirely realistic as it's possible to arrive at a link with +/// 0.001% SOC and still need to use gasoline to traverse the link. +/// +/// In the future we could make this more sophisticated by calculating +/// the energy required to traverse the link using the battery and then +/// finding the point at which we would have to switch to gasoline +/// +/// Returns a tuple of (electrical_energy, electrical_energy_unit, gasoline_energy, gasoline_energy_unit) +fn get_phev_energy( + vehicle: &DualFuelVehicle, + battery_soc_percent: f64, + speed: (Speed, SpeedUnit), + grade: (Grade, GradeUnit), + distance: (Distance, DistanceUnit), +) -> Result<(Energy, EnergyUnit, Energy, EnergyUnit), TraversalModelError> { + let electrical_energy_unit = vehicle + .charge_depleting_model + .energy_rate_unit + .associated_energy_unit(); + let gasoline_energy_unit = vehicle + .charge_sustain_model + .energy_rate_unit + .associated_energy_unit(); + + if battery_soc_percent > 0.0 { + // assume we can just use the battery + let (electrical_energy, electrical_energy_unit) = vehicle + .charge_depleting_model + .predict(speed, grade, distance)?; + Ok(( + electrical_energy, + electrical_energy_unit, + Energy::new(0.0), + gasoline_energy_unit, + )) + } else { + // just use the gasoline engine + let (gasoline_energy, gasoline_energy_unit) = vehicle + .charge_sustain_model + .predict(speed, grade, distance)?; + Ok(( + Energy::new(0.0), + electrical_energy_unit, + gasoline_energy, + gasoline_energy_unit, + )) + } +} + +#[cfg(test)] +mod tests { + use routee_compass_core::util::unit::{EnergyRate, EnergyRateUnit}; + + use crate::routee::{prediction::load_prediction_model, prediction::model_type::ModelType}; + + use super::*; + + use std::path::PathBuf; + + fn mock_vehicle() -> DualFuelVehicle { + let charge_sustain_model_file_path = PathBuf::from(env!("CARGO_MANIFEST_DIR")) + .join("src") + .join("routee") + .join("test") + .join("2016_CHEVROLET_Volt_Charge_Sustaining.bin"); + let charge_depleting_model_file_path = PathBuf::from(env!("CARGO_MANIFEST_DIR")) + .join("src") + .join("routee") + .join("test") + .join("2016_CHEVROLET_Volt_Charge_Depleting.bin"); + + let charge_sustain_model_record = load_prediction_model( + "Chevy_Volt_Charge_Sustaining".to_string(), + &charge_sustain_model_file_path, + ModelType::Smartcore, + SpeedUnit::MilesPerHour, + GradeUnit::Decimal, + EnergyRateUnit::GallonsGasolinePerMile, + Some(EnergyRate::new(0.02)), + Some(1.1252), + ) + .unwrap(); + let charge_depleting_model_record = load_prediction_model( + "Chevy_Volt_Charge_Depleting".to_string(), + &charge_depleting_model_file_path, + ModelType::Smartcore, + SpeedUnit::MilesPerHour, + GradeUnit::Decimal, + EnergyRateUnit::KilowattHoursPerMile, + Some(EnergyRate::new(0.2)), + Some(1.3958), + ) + .unwrap(); + + DualFuelVehicle::new( + "Chevy_Volt".to_string(), + charge_sustain_model_record, + charge_depleting_model_record, + Energy::new(12.0), + Energy::new(12.0), + EnergyUnit::KilowattHours, + ) + .unwrap() + } + + #[test] + fn test_phev_energy_model_just_electric() { + let vehicle = mock_vehicle(); + let initial = vehicle.initial_state(); + + // starting at 100% SOC, we should be able to traverse 1000 meters + // without using any gasoline + let distance = (Distance::new(1000.0), DistanceUnit::Meters); + let speed = (Speed::new(60.0), SpeedUnit::MilesPerHour); + let grade = (Grade::new(0.0), GradeUnit::Decimal); + + let result = vehicle + .consume_energy(speed, grade, distance, &initial) + .unwrap(); + + let gasoline_energy = get_gasoline_energy_from_state(&result.updated_state); + assert!(gasoline_energy.as_f64() < 1e-9); + + let electrical_energy = get_electrical_energy_from_state(&result.updated_state); + assert!(electrical_energy.as_f64() > 0.0); + + let battery_percent_soc = get_battery_soc_percent(&vehicle, &result.updated_state); + assert!(battery_percent_soc < 100.0); + } + + #[test] + fn test_phev_energy_model_gas_and_electric() { + let vehicle = mock_vehicle(); + let initial = vehicle.initial_state(); + + // now let's traverse a really long link to deplete the battery + let distance = (Distance::new(100.0), DistanceUnit::Miles); + let speed = (Speed::new(60.0), SpeedUnit::MilesPerHour); + let grade = (Grade::new(0.0), GradeUnit::Decimal); + + let result = vehicle + .consume_energy(speed, grade, distance, &initial) + .unwrap(); + + let electrical_energy = get_electrical_energy_from_state(&result.updated_state); + let battery_percent_soc = get_battery_soc_percent(&vehicle, &result.updated_state); + + assert!(electrical_energy.as_f64() > 0.0); + assert!(battery_percent_soc < 1e-9); + + // and then traverse the same distance but this time we should only use gasoline energy + let result2 = vehicle + .consume_energy(speed, grade, distance, &result.updated_state) + .unwrap(); + + let gasoline_energy = get_gasoline_energy_from_state(&result2.updated_state); + + assert!(gasoline_energy.as_f64() > 0.0); + } +} diff --git a/rust/routee-compass-powertrain/src/routee/vehicle/default/mod.rs b/rust/routee-compass-powertrain/src/routee/vehicle/default/mod.rs new file mode 100644 index 00000000..daa93c3e --- /dev/null +++ b/rust/routee-compass-powertrain/src/routee/vehicle/default/mod.rs @@ -0,0 +1,2 @@ +pub mod dual_fuel_vehicle; +pub mod single_fuel_vehicle; diff --git a/rust/routee-compass-powertrain/src/routee/vehicle/default/single_fuel_vehicle.rs b/rust/routee-compass-powertrain/src/routee/vehicle/default/single_fuel_vehicle.rs new file mode 100644 index 00000000..d270e8a7 --- /dev/null +++ b/rust/routee-compass-powertrain/src/routee/vehicle/default/single_fuel_vehicle.rs @@ -0,0 +1,113 @@ +use std::sync::Arc; + +use routee_compass_core::{ + model::traversal::{ + state::state_variable::StateVar, traversal_model_error::TraversalModelError, + }, + util::unit::{ + as_f64::AsF64, Distance, DistanceUnit, Energy, EnergyUnit, Grade, GradeUnit, Speed, + SpeedUnit, + }, +}; + +use crate::routee::{ + prediction::PredictionModelRecord, + vehicle::{VehicleEnergyResult, VehicleState, VehicleType}, +}; + +pub struct SingleFuelVehicle { + pub name: String, + pub prediction_model_record: Arc<PredictionModelRecord>, +} + +impl SingleFuelVehicle { + pub fn new( + name: String, + prediction_model_record: PredictionModelRecord, + ) -> Result<Self, TraversalModelError> { + Ok(Self { + name, + prediction_model_record: Arc::new(prediction_model_record), + }) + } +} + +impl VehicleType for SingleFuelVehicle { + fn name(&self) -> String { + self.name.clone() + } + fn initial_state(&self) -> VehicleState { + // accumulated energy + vec![StateVar(0.0)] + } + fn best_case_energy( + &self, + distance: (Distance, DistanceUnit), + ) -> Result<(Energy, EnergyUnit), TraversalModelError> { + let (distance, distance_unit) = distance; + let energy = Energy::create( + self.prediction_model_record.ideal_energy_rate, + self.prediction_model_record.energy_rate_unit, + distance, + distance_unit, + )?; + Ok(energy) + } + fn consume_energy( + &self, + speed: (Speed, SpeedUnit), + grade: (Grade, GradeUnit), + distance: (Distance, DistanceUnit), + state: &[StateVar], + ) -> Result<VehicleEnergyResult, TraversalModelError> { + let (energy, energy_unit) = self + .prediction_model_record + .predict(speed, grade, distance)?; + + let updated_state = update_state(state, energy); + + Ok(VehicleEnergyResult { + energy, + energy_unit, + updated_state, + }) + } + fn serialize_state(&self, state: &[StateVar]) -> serde_json::Value { + let energy = get_energy_from_state(state); + serde_json::json!({ + "energy": energy.as_f64(), + }) + } + + fn serialize_state_info(&self, _state: &[StateVar]) -> serde_json::Value { + let energy_unit = self + .prediction_model_record + .energy_rate_unit + .associated_energy_unit(); + serde_json::json!({ + "energy_unit": energy_unit.to_string(), + }) + } + + fn update_from_query( + &self, + _query: &serde_json::Value, + ) -> Result<Arc<dyn VehicleType>, TraversalModelError> { + // just return a clone of self + Ok(Arc::new(SingleFuelVehicle { + name: self.name.clone(), + prediction_model_record: self.prediction_model_record.clone(), + })) + } +} + +fn update_state(state: &[StateVar], energy: Energy) -> VehicleState { + let mut new_state = Vec::with_capacity(state.len()); + new_state.push(state[0] + energy.into()); + new_state +} + +fn get_energy_from_state(state: &[StateVar]) -> Energy { + let energy = state[0].0; + Energy::new(energy) +} diff --git a/rust/routee-compass-powertrain/src/routee/vehicle/mod.rs b/rust/routee-compass-powertrain/src/routee/vehicle/mod.rs new file mode 100644 index 00000000..ef315eec --- /dev/null +++ b/rust/routee-compass-powertrain/src/routee/vehicle/mod.rs @@ -0,0 +1,6 @@ +pub mod default; +pub mod vehicle_energy_result; +pub mod vehicle_type; + +pub use vehicle_energy_result::VehicleEnergyResult; +pub use vehicle_type::{VehicleState, VehicleType}; diff --git a/rust/routee-compass-powertrain/src/routee/vehicle/vehicle_energy_result.rs b/rust/routee-compass-powertrain/src/routee/vehicle/vehicle_energy_result.rs new file mode 100644 index 00000000..f4fcb9f1 --- /dev/null +++ b/rust/routee-compass-powertrain/src/routee/vehicle/vehicle_energy_result.rs @@ -0,0 +1,9 @@ +use routee_compass_core::util::unit::{Energy, EnergyUnit}; + +use super::VehicleState; + +pub struct VehicleEnergyResult { + pub energy: Energy, + pub energy_unit: EnergyUnit, + pub updated_state: VehicleState, +} diff --git a/rust/routee-compass-powertrain/src/routee/vehicle/vehicle_type.rs b/rust/routee-compass-powertrain/src/routee/vehicle/vehicle_type.rs new file mode 100644 index 00000000..4d2ec107 --- /dev/null +++ b/rust/routee-compass-powertrain/src/routee/vehicle/vehicle_type.rs @@ -0,0 +1,82 @@ +use std::sync::Arc; + +use routee_compass_core::{ + model::traversal::{ + state::state_variable::StateVar, traversal_model_error::TraversalModelError, + }, + util::unit::{Distance, DistanceUnit, Energy, EnergyUnit, Grade, GradeUnit, Speed, SpeedUnit}, +}; + +use super::VehicleEnergyResult; + +pub type VehicleState = Vec<StateVar>; + +/// A Vehicle Type represents a class of vehicles with a specific operating model. +pub trait VehicleType: Send + Sync { + /// Return the name of the vehicle type + fn name(&self) -> String; + + /// Return the energy required to travel a certain distance at a certain speed and grade. + /// + /// Arguments: + /// * `speed` - The speed at which the vehicle is traveling + /// * `grade` - The grade of the road + /// * `distance` - The distance traveled + /// * `state` - The state of the vehicle + /// + /// Returns: + /// * `VehicleEnergyResult` - The energy required + fn consume_energy( + &self, + speed: (Speed, SpeedUnit), + grade: (Grade, GradeUnit), + distance: (Distance, DistanceUnit), + state: &[StateVar], + ) -> Result<VehicleEnergyResult, TraversalModelError>; + + /// Return the best case scenario for traveling a certain distance. + /// This is used in the a-star algorithm as a distance heuristic. + /// + /// Arguments: + /// * `distance` - The distance traveled + /// + /// Returns: + /// * `Energy` - The 'best case' energy required to travel the distance + fn best_case_energy( + &self, + distance: (Distance, DistanceUnit), + ) -> Result<(Energy, EnergyUnit), TraversalModelError>; + + /// Return the initial state of the vehicle + fn initial_state(&self) -> VehicleState; + + /// Serialize the state of the vehicle into JSON + /// + /// Arguments: + /// * `state` - The state of the vehicle + /// + /// Returns: + /// * `serde_json::Value` - The serialized state + fn serialize_state(&self, state: &[StateVar]) -> serde_json::Value; + + /// Serialize any supplemental state information (like units) into JSON + /// + /// Arguments: + /// * `state` - The state of the vehicle + /// + /// Returns: + /// * `serde_json::Value` - The serialized state information + fn serialize_state_info(&self, state: &[StateVar]) -> serde_json::Value; + + /// Give the vehicle a chance to update itself from the incoming query + /// + /// Arguments: + /// * `query` - The incoming query + /// + /// Returns: + /// * `Arc<dyn VehicleType>` - The updated vehicle type + fn update_from_query( + &self, + query: &serde_json::Value, + ) -> Result<Arc<dyn VehicleType>, TraversalModelError>; +} diff --git a/rust/routee-compass/src/app/compass/config/compass_app_builder.rs b/rust/routee-compass/src/app/compass/config/compass_app_builder.rs index bbe894c3..329f3a7c 100644 --- a/rust/routee-compass/src/app/compass/config/compass_app_builder.rs +++ b/rust/routee-compass/src/app/compass/config/compass_app_builder.rs @@ -7,8 +7,7 @@ use super::{ no_restriction_builder::NoRestrictionBuilder, road_class_builder::RoadClassBuilder, }, traversal_model::{ - distance_builder::DistanceBuilder, - speed_grade_energy_model_builder::SpeedGradeEnergyModelBuilder, + distance_builder::DistanceBuilder, energy_model_builder::EnergyModelBuilder, speed_lookup_builder::SpeedLookupBuilder, }, }; @@ -113,11 +112,11 @@ impl CompassAppBuilder { // Traversal model builders let dist: Box<dyn TraversalModelBuilder> = Box::new(DistanceBuilder {}); let velo: Box<dyn TraversalModelBuilder> = Box::new(SpeedLookupBuilder {}); - let smartcore: Box<dyn TraversalModelBuilder> = Box::new(SpeedGradeEnergyModelBuilder {}); + let energy_model: Box<dyn TraversalModelBuilder> = Box::new(EnergyModelBuilder {}); let tm_builders: HashMap<String, Box<dyn TraversalModelBuilder>> = HashMap::from([ (String::from("distance"), dist), (String::from("speed_table"), velo), - (String::from("speed_grade_energy_model"), smartcore), + (String::from("energy_model"), energy_model), ]); // Frontier model builders diff --git a/rust/routee-compass/src/app/compass/config/compass_configuration_field.rs b/rust/routee-compass/src/app/compass/config/compass_configuration_field.rs index e0d0a80d..09c248ec 100644 --- a/rust/routee-compass/src/app/compass/config/compass_configuration_field.rs +++ b/rust/routee-compass/src/app/compass/config/compass_configuration_field.rs @@ -13,6 +13,8 @@ pub enum CompassConfigurationField { Parallelism, QueryTimeoutMs, IncludeTree, + ChargeDepleting, + ChargeSustaining, } impl CompassConfigurationField { @@ -29,6 +31,8 @@ impl CompassConfigurationField { CompassConfigurationField::Plugins => "plugin", CompassConfigurationField::InputPlugins => "input_plugins", CompassConfigurationField::OutputPlugins => "output_plugins", + CompassConfigurationField::ChargeDepleting => "charge_depleting", + CompassConfigurationField::ChargeSustaining => "charge_sustaining", } } } diff --git a/rust/routee-compass/src/app/compass/config/traversal_model/energy_model_builder.rs b/rust/routee-compass/src/app/compass/config/traversal_model/energy_model_builder.rs new file mode 100644 index 00000000..c20ba155 --- /dev/null +++ b/rust/routee-compass/src/app/compass/config/traversal_model/energy_model_builder.rs @@ -0,0 +1,96 @@ +use std::collections::HashMap; +use std::sync::Arc; + +use crate::app::compass::config::compass_configuration_field::CompassConfigurationField; +use crate::app::compass::config::config_json_extension::ConfigJsonExtensions; +use routee_compass_core::model::traversal::traversal_model_builder::TraversalModelBuilder; +use routee_compass_core::model::traversal::traversal_model_error::TraversalModelError; +use routee_compass_core::model::traversal::traversal_model_service::TraversalModelService; +use routee_compass_core::util::unit::{DistanceUnit, GradeUnit, SpeedUnit, TimeUnit}; +use routee_compass_powertrain::routee::energy_model_service::EnergyModelService; + +use super::energy_model_vehicle_builders::VehicleBuilder; + +pub struct EnergyModelBuilder {} + +impl TraversalModelBuilder for EnergyModelBuilder { + fn build( + &self, + params: &serde_json::Value, + ) -> Result<Arc<dyn TraversalModelService>, TraversalModelError> { + let traversal_key = CompassConfigurationField::Traversal.to_string(); + + let speed_table_path = params + .get_config_path( + String::from("speed_table_input_file"), + traversal_key.clone(), + ) + .map_err(|e| TraversalModelError::BuildError(e.to_string()))?; + let speed_table_speed_unit = params + .get_config_serde::<SpeedUnit>( + String::from("speed_table_speed_unit"), + traversal_key.clone(), + ) + .map_err(|e| TraversalModelError::BuildError(e.to_string()))?; + + let grade_table_path = params + .get_config_path_optional( + String::from("grade_table_input_file"), + traversal_key.clone(), + ) + .map_err(|e| TraversalModelError::BuildError(e.to_string()))?; + let grade_table_grade_unit = params + .get_config_serde_optional::<GradeUnit>( + String::from("graph_grade_unit"), + traversal_key.clone(), + ) + .map_err(|e| TraversalModelError::BuildError(e.to_string()))?; + + let vehicle_configs = params + .get_config_array("vehicles".to_string(), traversal_key.clone()) + .map_err(|e| TraversalModelError::BuildError(e.to_string()))?; + + let mut vehicle_library = HashMap::new(); + + for vehicle_config in vehicle_configs { + let vehicle_type = vehicle_config + .get_config_string(String::from("type"), traversal_key.clone()) + .map_err(|e| TraversalModelError::BuildError(e.to_string()))?; + let vehicle_builder = VehicleBuilder::from_string(vehicle_type).map_err(|e| { + TraversalModelError::BuildError(format!( + "Error building vehicle builder: {}", + e.to_string() + )) + })?; + let vehicle = vehicle_builder + .build(&vehicle_config) + .map_err(|e| TraversalModelError::BuildError(e.to_string()))?; + vehicle_library.insert(vehicle.name(), vehicle); + } + + let output_time_unit_option = params + .get_config_serde_optional::<TimeUnit>( + String::from("output_time_unit"), + traversal_key.clone(), + ) + .map_err(|e| TraversalModelError::BuildError(e.to_string()))?; + let output_distance_unit_option = params + .get_config_serde_optional::<DistanceUnit>( + String::from("output_distance_unit"), + traversal_key.clone(), + ) + .map_err(|e| TraversalModelError::BuildError(e.to_string()))?; + + let service = EnergyModelService::new( + &speed_table_path, + speed_table_speed_unit, + &grade_table_path, + grade_table_grade_unit, + output_time_unit_option, + output_distance_unit_option, + vehicle_library, + )?; + + Ok(Arc::new(service)) + } +} diff --git a/rust/routee-compass/src/app/compass/config/traversal_model/energy_model_vehicle_builders.rs b/rust/routee-compass/src/app/compass/config/traversal_model/energy_model_vehicle_builders.rs new file mode 100644 index 00000000..d199502c --- /dev/null +++ b/rust/routee-compass/src/app/compass/config/traversal_model/energy_model_vehicle_builders.rs @@ -0,0 +1,136 @@ +use std::sync::Arc; + +use routee_compass_core::util::unit::{ + Energy, EnergyRate, EnergyRateUnit, EnergyUnit, GradeUnit, SpeedUnit, +}; +use routee_compass_powertrain::routee::{ + prediction::{load_prediction_model, model_type::ModelType, PredictionModelRecord}, + vehicle::{ + default::{dual_fuel_vehicle::DualFuelVehicle, single_fuel_vehicle::SingleFuelVehicle}, + VehicleType, + }, +}; + +use crate::app::compass::config::{ + compass_configuration_error::CompassConfigurationError, + compass_configuration_field::CompassConfigurationField, + config_json_extension::ConfigJsonExtensions, +}; + +pub enum VehicleBuilder { + SingleFuel, + DualFuel, +} + +impl VehicleBuilder { + pub fn from_string(vehicle_type: String) -> Result<VehicleBuilder, CompassConfigurationError> { + match vehicle_type.as_str() { + "single_fuel" => Ok(VehicleBuilder::SingleFuel), + "dual_fuel" => Ok(VehicleBuilder::DualFuel), + _ => Err(CompassConfigurationError::ExpectedFieldWithType( + "vehicle.type".to_string(), + "string".to_string(), + )), + } + } + pub fn build( + &self, + parameters: &serde_json::Value, + ) -> Result<Arc<dyn VehicleType>, CompassConfigurationError> { + match self { + VehicleBuilder::SingleFuel => build_conventional(parameters), + VehicleBuilder::DualFuel => build_plugin_hybrid(parameters), + } + } +} + +fn build_conventional( + parameters: &serde_json::Value, +) -> Result<Arc<dyn VehicleType>, CompassConfigurationError> { + let vehicle_key = String::from("single_fuel"); + let name = parameters.get_config_string(String::from("name"), vehicle_key.clone())?; + + let model_record = get_model_record_from_params(parameters, name.clone())?; + + let vehicle = SingleFuelVehicle::new(name, model_record)?; + + Ok(Arc::new(vehicle)) +} + +fn build_plugin_hybrid( + parameters: &serde_json::Value, +) -> Result<Arc<dyn VehicleType>, CompassConfigurationError> { + let vehicle_key = String::from("dual_fuel"); + let name = parameters.get_config_string(String::from("name"), vehicle_key.clone())?; + + let charge_depleting_params = + parameters.get_config_section(CompassConfigurationField::ChargeDepleting)?; + + let charge_depleting_record = get_model_record_from_params( + &charge_depleting_params, + format!("charge_depleting: {}", name.clone()), + )?; + let charge_sustain_params = + parameters.get_config_section(CompassConfigurationField::ChargeSustaining)?; + + let charge_sustain_record = get_model_record_from_params( + &charge_sustain_params, + format!("charge_sustain: {}", name.clone()), + )?; + + let battery_capacity = parameters + .get_config_serde::<Energy>(String::from("battery_capacity"), vehicle_key.clone())?; + let battery_energy_unit = parameters.get_config_serde::<EnergyUnit>( + String::from("battery_capacity_unit"), + vehicle_key.clone(), + )?; + let starting_battery_energy = battery_capacity; + let phev = DualFuelVehicle::new( + name, + charge_sustain_record, + charge_depleting_record, + battery_capacity, + starting_battery_energy, + battery_energy_unit, + )?; + Ok(Arc::new(phev)) +} + +fn get_model_record_from_params( + parameters: &serde_json::Value, + parent_key: String, +) -> Result<PredictionModelRecord, CompassConfigurationError> { + let name = parameters.get_config_string(String::from("name"), parent_key.clone())?; + let model_path = + parameters.get_config_path(String::from("model_input_file"), parent_key.clone())?; + let model_type = + parameters.get_config_serde::<ModelType>(String::from("model_type"), parent_key.clone())?; + let speed_unit = + parameters.get_config_serde::<SpeedUnit>(String::from("speed_unit"), parent_key.clone())?; + let ideal_energy_rate_option = parameters.get_config_serde_optional::<EnergyRate>( + String::from("ideal_energy_rate"), + parent_key.clone(), + )?; + let grade_unit = + parameters.get_config_serde::<GradeUnit>(String::from("grade_unit"), parent_key.clone())?; + + let energy_rate_unit = parameters + .get_config_serde::<EnergyRateUnit>(String::from("energy_rate_unit"), parent_key.clone())?; + let real_world_energy_adjustment_option = parameters.get_config_serde_optional::<f64>( + String::from("real_world_energy_adjustment"), + parent_key.clone(), + )?; + + let model_record = load_prediction_model( + name.clone(), + &model_path, + model_type, + speed_unit, + grade_unit, + energy_rate_unit, + ideal_energy_rate_option, + real_world_energy_adjustment_option, + )?; + + Ok(model_record) +} diff --git a/rust/routee-compass/src/app/compass/config/traversal_model/mod.rs b/rust/routee-compass/src/app/compass/config/traversal_model/mod.rs index 857fe071..4eace7aa 100644 --- a/rust/routee-compass/src/app/compass/config/traversal_model/mod.rs +++ b/rust/routee-compass/src/app/compass/config/traversal_model/mod.rs @@ -1,3 +1,4 @@ pub mod distance_builder; -pub mod speed_grade_energy_model_builder; +pub mod energy_model_builder; +pub mod energy_model_vehicle_builders; pub mod speed_lookup_builder; diff --git a/rust/routee-compass/src/app/compass/config/traversal_model/speed_grade_energy_model_builder.rs b/rust/routee-compass/src/app/compass/config/traversal_model/speed_grade_energy_model_builder.rs deleted file mode 100644 index d5b1c33b..00000000 --- a/rust/routee-compass/src/app/compass/config/traversal_model/speed_grade_energy_model_builder.rs +++ /dev/null @@ -1,138 +0,0 @@ -use std::collections::HashMap; -use std::sync::Arc; - -use crate::app::compass::config::compass_configuration_error::CompassConfigurationError; -use crate::app::compass::config::compass_configuration_field::CompassConfigurationField; -use crate::app::compass::config::config_json_extension::ConfigJsonExtensions; -use routee_compass_core::model::traversal::traversal_model_builder::TraversalModelBuilder; -use routee_compass_core::model::traversal::traversal_model_error::TraversalModelError; -use routee_compass_core::model::traversal::traversal_model_service::TraversalModelService; -use routee_compass_core::util::unit::{ - DistanceUnit, EnergyRate, EnergyRateUnit, GradeUnit, SpeedUnit, TimeUnit, -}; -use routee_compass_powertrain::routee::model_type::ModelType; -use routee_compass_powertrain::routee::prediction_model::SpeedGradePredictionModelRecord; -use routee_compass_powertrain::routee::speed_grade_energy_model_service::SpeedGradeEnergyModelService; - -pub struct SpeedGradeEnergyModelBuilder {} - -impl TraversalModelBuilder for SpeedGradeEnergyModelBuilder { - fn build( - &self, - params: &serde_json::Value, - ) -> Result<Arc<dyn TraversalModelService>, TraversalModelError> { - let traversal_key = CompassConfigurationField::Traversal.to_string(); - - let speed_table_path = params - .get_config_path( - String::from("speed_table_input_file"), - traversal_key.clone(), - ) - .map_err(|e| TraversalModelError::BuildError(e.to_string()))?; - let speed_table_speed_unit = params - .get_config_serde::<SpeedUnit>( - String::from("speed_table_speed_unit"), - traversal_key.clone(), - ) - .map_err(|e| TraversalModelError::BuildError(e.to_string()))?; - - let grade_table_path = params - .get_config_path_optional( - String::from("grade_table_input_file"), - traversal_key.clone(), - ) - .map_err(|e| TraversalModelError::BuildError(e.to_string()))?; - let grade_table_grade_unit = params - .get_config_serde_optional::<GradeUnit>( - String::from("graph_grade_unit"), - traversal_key.clone(), - ) - .map_err(|e| TraversalModelError::BuildError(e.to_string()))?; - - let energy_model_configs = params - .get_config_array("energy_models".to_string(), traversal_key.clone()) - .map_err(|e| TraversalModelError::BuildError(e.to_string()))?; - - let mut energy_model_library = HashMap::new(); - - for energy_model_config in energy_model_configs { - let name = energy_model_config - .get_config_string(String::from("name"), traversal_key.clone()) - .map_err(|e| TraversalModelError::BuildError(e.to_string()))?; - let model_path = energy_model_config - .get_config_path(String::from("model_input_file"), traversal_key.clone()) - .map_err(|e| TraversalModelError::BuildError(e.to_string()))?; - let model_type = energy_model_config - .get_config_serde::<ModelType>(String::from("model_type"), traversal_key.clone()) - .map_err(|e| TraversalModelError::BuildError(e.to_string()))?; - let speed_unit = energy_model_config - .get_config_serde::<SpeedUnit>(String::from("speed_unit"), traversal_key.clone()) - .map_err(|e| TraversalModelError::BuildError(e.to_string()))?; - let ideal_energy_rate_option = energy_model_config - .get_config_serde_optional::<EnergyRate>( - String::from("ideal_energy_rate"), - traversal_key.clone(), - ) - .map_err(|e| TraversalModelError::BuildError(e.to_string()))?; - let grade_unit = energy_model_config - .get_config_serde::<GradeUnit>(String::from("grade_unit"), traversal_key.clone()) - .map_err(|e| TraversalModelError::BuildError(e.to_string()))?; - - let energy_rate_unit = energy_model_config - .get_config_serde::<EnergyRateUnit>( - String::from("energy_rate_unit"), - traversal_key.clone(), - ) - .map_err(|e| TraversalModelError::BuildError(e.to_string()))?; - let real_world_energy_adjustment_option = params - .get_config_serde_optional::<f64>( - String::from("real_world_energy_adjustment"), - traversal_key.clone(), - ) - .map_err(|e| TraversalModelError::BuildError(e.to_string()))?; - - let model_record = SpeedGradePredictionModelRecord::new( - name.clone(), - &model_path, - model_type, - speed_unit, - grade_unit, - energy_rate_unit, - ideal_energy_rate_option, - real_world_energy_adjustment_option, - ) - .map_err(|e| TraversalModelError::BuildError(e.to_string()))?; - energy_model_library.insert(name, Arc::new(model_record)); - } - - let output_time_unit_option = params - .get_config_serde_optional::<TimeUnit>( - String::from("output_time_unit"), - traversal_key.clone(), - ) - .map_err(|e| TraversalModelError::BuildError(e.to_string()))?; - let output_distance_unit_option = params - .get_config_serde_optional::<DistanceUnit>( - String::from("output_distance_unit"), - traversal_key.clone(), - ) - .map_err(|e| TraversalModelError::BuildError(e.to_string()))?; - - let service = SpeedGradeEnergyModelService::new( - &speed_table_path, - speed_table_speed_unit, - &grade_table_path, - grade_table_grade_unit, - output_time_unit_option, - output_distance_unit_option, - energy_model_library, - ) - .map_err(CompassConfigurationError::TraversalModelError) - .map_err(|e| TraversalModelError::BuildError(e.to_string()))?; - // let service = SpeedGradeEnergyModelService { - // service: inner_service, - // }; - let result: Arc<dyn TraversalModelService> = Arc::new(service); - Ok(result) - } -}

PHEV implementation Our current energy traversal model supports conventional vehicles and battery electric vehicles but not plugin hybrid electric (PHEV) vehicles and would require a bit of a modification to do so. One primary limitation is that we're not keeping track of vehicle state of charge. This is important for a PHEV as it generally dictates what fuel type is going to be used. For example, if the state of charge is 100% and the route can be completed with battery energy only, we would report back 0 gasoline usage and X electrical energy usage. On the other hand, if the state of charge starts at 10% and the route cannot be completed using the battery energy, we might report back some electrical energy usage and some gasoline usage when the vehicle switched over to a different source of energy. Routee-Powertrain currently models PHEVs by creating two separate models, a charge sustaining (battery and gasoline) model and a charge depleting (just battery) model. To implement this as a traversal model we could build a simple PHEV that would use the charge depleting RouteE Powertrain model until the battery ran out of energy and then switch to the charge sustaining model for the remainder of the trip. This would require the user to pass in a state of charge parameter in the query and would require us to keep track of battery energy in our state variable. We would also need to define the battery capacity at the config level.

This also presents some difficulty with our existing implementation of applying an energy cost coefficient directly to the energy since the cost could be in variable units depending on the type of fuel being used. We could start by just converting the energy to a single unit (like converting gallons of gasoline to estimated equivalent kwh) but the best solution is probably to apply a $ value to each unit.

2023-11-16T22:08:28

0.0

[]

NREL/routee-compass

NREL__routee-compass-8

2d65ee947b27536b0f97fb7be813bb4f3dd9dfb2

diff --git a/.cargo/config.toml b/.cargo/config.toml index ca43966b..4d1815a7 100644 --- a/.cargo/config.toml +++ b/.cargo/config.toml @@ -3,7 +3,7 @@ check-revoke = false [env] ORT_STRATEGY = { value = "system" } -ORT_LIB_LOCATION = { value = "rust/compass-powertrain/onnx-runtime/v1.15.1/osx-x86_64", relative = true } +ORT_LIB_LOCATION = { value = "rust/routee-compass-powertrain/onnx-runtime/v1.15.1/osx-x86_64", relative = true } [target.x86_64-apple-darwin] rustflags = ["-C", "link-arg=-undefined", "-C", "link-arg=dynamic_lookup"] diff --git a/docs/config.md b/docs/config.md index 33310934..e325d29c 100644 --- a/docs/config.md +++ b/docs/config.md @@ -13,9 +13,9 @@ parallelism = 2 # the parameters for the underlying road network graph [graph] # a file containing all the graph edges and their adjacencies -edge_list_csv = "edges-compass.csv.gz" +edge_list_file = "edges-compass.csv.gz" # a file containing all the graph verticies -vertex_list_csv = "vertices-compass.csv.gz" +vertex_list_file = "vertices-compass.csv.gz" # if verbose is true, you'll see more information when loading the graph verbose = true @@ -26,9 +26,9 @@ type = "speed_grade_energy_model" # what underlying machine learn framework to use [smartcore | onnx] model_type = "smartcore" # speeds for each edge in the graph -speed_table_path = "edges-posted-speed-enumerated.txt.gz" +speed_table_file = "edges-posted-speed-enumerated.txt.gz" # the file for the routee-powertrain model -energy_model_path = "2016_TOYOTA_Camry_4cyl_2WD.bin" +energy_model_file = "2016_TOYOTA_Camry_4cyl_2WD.bin" # the "best case" energy rate for this particular vehicle (something like highway mpg) that's used in the a-star algorithm ideal_energy_rate = 0.02857142857 # the units of the above speed table @@ -103,7 +103,7 @@ The speed table traversal model uses a speed lookup table to compute the fastest ```toml [traversal] type = "speed_table" -speed_table_path = "edges-posted-speed-enumerated.txt.gz" +speed_table_file = "edges-posted-speed-enumerated.txt.gz" speed_unit = "kilometers_per_hour" output_distance_unit = "miles" output_time_unit = "minutes" @@ -117,10 +117,10 @@ The speed grade energy model computes energy (with a routee-powertrain vehicle m [traversal] type = "speed_grade_energy_model" model_type = "smartcore" -speed_table_path = "data/tomtom_metro_denver_network/edges-free-flow-speed-enumerated.txt.gz" -grade_table_path = "data/tomtom_metro_denver_network/edges-grade-enumerated.txt.gz" +speed_table_file = "data/tomtom_metro_denver_network/edges-free-flow-speed-enumerated.txt.gz" +grade_table_file = "data/tomtom_metro_denver_network/edges-grade-enumerated.txt.gz" grade_table_grade_unit = "decimal" -energy_model_path = "data/2016_TOYOTA_Camry_4cyl_2WD.bin" +energy_model_file = "data/2016_TOYOTA_Camry_4cyl_2WD.bin" ideal_energy_rate = 0.02857142857 speed_table_speed_unit = "kilometers_per_hour" energy_model_speed_unit = "miles_per_hour" diff --git a/docs/notebooks/open_street_maps_example.ipynb b/docs/notebooks/open_street_maps_example.ipynb index d19f81df..0641c4d7 100644 --- a/docs/notebooks/open_street_maps_example.ipynb +++ b/docs/notebooks/open_street_maps_example.ipynb @@ -184,17 +184,6 @@ { "cell_type": "code", "execution_count": 5, - "id": "30c0da5c-fd6d-44f2-873c-1ce8b863cda8", - "metadata": {}, - "outputs": [], - "source": [ - "import os\n", - "os.chdir(\"golden_co\")" - ] - }, - { - "cell_type": "code", - "execution_count": 6, "id": "edad6c15-c641-4010-b097-7d07aa86c1e7", "metadata": {}, "outputs": [ @@ -212,12 +201,12 @@ "name": "stderr", "output_type": "stream", "text": [ - "uuid file: 100%|██████████| 746/746 [00:00<00:00, 4245946.00it/s]it/s]" + "uuid file: 100%|██████████| 746/746 [00:00<00:00, 3849865.50it/s]it/s]" ] } ], "source": [ - "app = CompassApp.from_config_file(\"osm_default_energy.toml\") " + "app = CompassApp.from_config_file(\"golden_co/osm_default_energy.toml\") " ] }, { @@ -232,7 +221,7 @@ }, { "cell_type": "code", - "execution_count": 7, + "execution_count": 6, "id": "f01086d0-231f-4013-bc24-97f9c56670bd", "metadata": {}, "outputs": [], @@ -252,7 +241,7 @@ }, { "cell_type": "code", - "execution_count": 8, + "execution_count": 7, "id": "c6e9361b-5ffd-4ec5-a09a-d51e46752ebf", "metadata": {}, "outputs": [], @@ -270,7 +259,7 @@ }, { "cell_type": "code", - "execution_count": 9, + "execution_count": 8, "id": "2c51774c-acff-4569-a820-4be5ba5aa51b", "metadata": {}, "outputs": [ @@ -280,7 +269,7 @@ "2" ] }, - "execution_count": 9, + "execution_count": 8, "metadata": {}, "output_type": "execute_result" } @@ -299,7 +288,7 @@ }, { "cell_type": "code", - "execution_count": 10, + "execution_count": 9, "id": "d075cb16-b0ae-463a-a2ce-f94042bc13db", "metadata": {}, "outputs": [], @@ -309,7 +298,7 @@ }, { "cell_type": "code", - "execution_count": 11, + "execution_count": 10, "id": "961c910c-63ec-4c79-9522-19f24f151215", "metadata": {}, "outputs": [ @@ -318,13 +307,13 @@ "text/plain": [ "{'distance': 4.23738036063421,\n", " 'distance_unit': 'miles',\n", - " 'energy': 0.18821565778485838,\n", + " 'energy': 0.18816139103031826,\n", " 'energy_unit': 'gallons_gasoline',\n", - " 'time': 8.501927726263151,\n", + " 'time': 8.486183959367098,\n", " 'time_unit': 'minutes'}" ] }, - "execution_count": 11, + "execution_count": 10, "metadata": {}, "output_type": "execute_result" } @@ -335,7 +324,7 @@ }, { "cell_type": "code", - "execution_count": 12, + "execution_count": 11, "id": "e2c6cb39-49aa-4697-9785-fa839dc16a86", "metadata": {}, "outputs": [ @@ -344,13 +333,13 @@ "text/plain": [ "{'distance': 4.017396519730842,\n", " 'distance_unit': 'miles',\n", - " 'energy': 0.18722374836334246,\n", + " 'energy': 0.1871862788506351,\n", " 'energy_unit': 'gallons_gasoline',\n", - " 'time': 8.987039191276414,\n", + " 'time': 8.619006658572934,\n", " 'time_unit': 'minutes'}" ] }, - "execution_count": 12, + "execution_count": 11, "metadata": {}, "output_type": "execute_result" } @@ -362,7 +351,7 @@ }, { "cell_type": "code", - "execution_count": 13, + "execution_count": 12, "id": "30257102-07ee-4338-9c8c-41676e1995ed", "metadata": {}, "outputs": [], @@ -384,7 +373,7 @@ }, { "cell_type": "code", - "execution_count": 14, + "execution_count": 13, "id": "d3907fe2-14c8-4904-942a-d09fd7d42ab2", "metadata": {}, "outputs": [], @@ -402,7 +391,7 @@ }, { "cell_type": "code", - "execution_count": 15, + "execution_count": 14, "id": "1ac13aae-32fc-4c10-b888-6373b175655a", "metadata": {}, "outputs": [ @@ -436,7 +425,7 @@ " <meta name="viewport" content="width=device-width,\n", " initial-scale=1.0, maximum-scale=1.0, user-scalable=no" />\n", " <style>\n", - " #map_5e6fdd78aee60d0d6315972837ed641b {\n", + " #map_f0bafdc10b40d238de9da7b18732b217 {\n", " position: relative;\n", " width: 100.0%;\n", " height: 100.0%;\n", @@ -450,14 +439,14 @@ "<body>\n", " \n", " \n", - " <div class="folium-map" id="map_5e6fdd78aee60d0d6315972837ed641b" ></div>\n", + " <div class="folium-map" id="map_f0bafdc10b40d238de9da7b18732b217" ></div>\n", " \n", "</body>\n", "<script>\n", " \n", " \n", - " var map_5e6fdd78aee60d0d6315972837ed641b = L.map(\n", - " "map_5e6fdd78aee60d0d6315972837ed641b",\n", + " var map_f0bafdc10b40d238de9da7b18732b217 = L.map(\n", + " "map_f0bafdc10b40d238de9da7b18732b217",\n", " {\n", " center: [39.7465774, -105.2216276],\n", " crs: L.CRS.EPSG3857,\n", @@ -471,19 +460,19 @@ "\n", " \n", " \n", - " var tile_layer_ee2775096a1be4380f9bd39c42ca4e8e = L.tileLayer(\n", + " var tile_layer_41ead211ec41bcf73e91c5f85331e723 = L.tileLayer(\n", " "https://{s}.tile.openstreetmap.org/{z}/{x}/{y}.png",\n", " {"attribution": "Data by \\u0026copy; \\u003ca target=\\"_blank\\" href=\\"http://openstreetmap.org\\"\\u003eOpenStreetMap\\u003c/a\\u003e, under \\u003ca target=\\"_blank\\" href=\\"http://www.openstreetmap.org/copyright\\"\\u003eODbL\\u003c/a\\u003e.", "detectRetina": false, "maxNativeZoom": 18, "maxZoom": 18, "minZoom": 0, "noWrap": false, "opacity": 1, "subdomains": "abc", "tms": false}\n", - " ).addTo(map_5e6fdd78aee60d0d6315972837ed641b);\n", + " ).addTo(map_f0bafdc10b40d238de9da7b18732b217);\n", " \n", " \n", - " var poly_line_6406764e3f6bcb21b9ed2e3ae7f61509 = L.polyline(\n", + " var poly_line_51128cf08c47e0b255fa936899f6e0e4 = L.polyline(\n", " [[39.7273069, -105.1994169], [39.7272989, -105.1994535], [39.7272848, -105.1994868], [39.7272593, -105.1995218], [39.7272272, -105.1995461], [39.7271912, -105.1995576], [39.7271688, -105.199558], [39.7271461, -105.1995532], [39.7271142, -105.1995365], [39.7270868, -105.1995096], [39.7270657, -105.1994741], [39.7270519, -105.1994296], [39.7270483, -105.1993816], [39.727052, -105.1993539], [39.7270571, -105.1993247], [39.7270719, -105.1992916], [39.7270973, -105.1992566], [39.7271293, -105.1992323], [39.7271651, -105.1992207], [39.7271854, -105.1992226], [39.7272113, -105.199227], [39.7273496, -105.1991687], [39.7274185, -105.1991465], [39.7275326, -105.1990928], [39.7276828, -105.1990602], [39.7277455, -105.1990518], [39.7279702, -105.1990232], [39.7281558, -105.1990357], [39.7284384, -105.1990893], [39.7286639, -105.199177], [39.7288222, -105.1992773], [39.7289214, -105.199385], [39.7290292, -105.1995543], [39.7290506, -105.1995879], [39.7292157, -105.1998155], [39.7293094, -105.1998945], [39.7294079, -105.1999604], [39.7295437, -105.200014], [39.7296299, -105.2000721], [39.7297292, -105.2000757], [39.7297135, -105.2007081], [39.7297192, -105.2013106], [39.7298153, -105.2032041], [39.7298256, -105.2033784], [39.7298664, -105.2035708], [39.7298795, -105.203649], [39.7299267, -105.2038558], [39.7301082, -105.2042608], [39.7304911, -105.2048068], [39.7308741, -105.2053111], [39.7310676, -105.2055985], [39.7311933, -105.205824], [39.7312069, -105.2058543], [39.7313075, -105.2061158], [39.731348, -105.206248], [39.731386, -105.2064213], [39.7314237, -105.2067336], [39.731429, -105.2070565], [39.7314226, -105.207224], [39.7314076, -105.2074036], [39.7313611, -105.2077675], [39.7313265, -105.2079124], [39.7312831, -105.20806], [39.7312292, -105.2082086], [39.7311706, -105.208344], [39.7310965, -105.2084777], [39.7310049, -105.208616], [39.7309234, -105.2087235], [39.7308212, -105.2088303], [39.7306227, -105.2089819], [39.7304801, -105.2090736], [39.7303681, -105.2091988], [39.7303327, -105.2092597], [39.7303114, -105.2093181], [39.730299, -105.20939], [39.7302939, -105.2095126], [39.7303105, -105.2096804], [39.7304414, -105.2099267], [39.7305001, -105.2100344], [39.7305871, -105.2101842], [39.7307404, -105.2104481], [39.7309234, -105.2107229], [39.7310934, -105.2109639], [39.731238, -105.211145], [39.7314101, -105.2113464], [39.7315783, -105.2115336], [39.7317541, -105.2117064], [39.7319796, -105.2119089], [39.732359, -105.2122393], [39.7329187, -105.2127184], [39.7334557, -105.213188], [39.7338968, -105.2135676], [39.7342777, -105.2138968], [39.7347139, -105.2142786], [39.7354652, -105.2149301], [39.7356319, -105.2150733], [39.7358844, -105.2152902], [39.7363901, -105.2157275], [39.7369129, -105.2161859], [39.7377243, -105.2168831], [39.7390882, -105.2180682], [39.7397882, -105.2186718], [39.7403043, -105.2191078], [39.7405645, -105.2193336], [39.7406767, -105.2194225], [39.7415803, -105.2202142], [39.7416486, -105.2202727], [39.7417047, -105.2203151], [39.7421502, -105.2206932], [39.7425816, -105.221062], [39.7432687, -105.2215398], [39.7435797, -105.2217631], [39.7436721, -105.2218299], [39.7437366, -105.2218764], [39.7438794, -105.2219759], [39.7439546, -105.2220253], [39.7440082, -105.2220586], [39.7440819, -105.2221021], [39.7441377, -105.2221322], [39.7441808, -105.2221541], [39.7442796, -105.2222043], [39.7443357, -105.2222286], [39.7444062, -105.222259], [39.7444657, -105.2222871], [39.7444962, -105.2223029], [39.7445695, -105.2223431], [39.7446343, -105.2223808], [39.744683, -105.2224104], [39.7447215, -105.2224349], [39.7447658, -105.2224654], [39.744813, -105.222499], [39.7448547, -105.2225315], [39.7448881, -105.2225586], [39.7449328, -105.2225965], [39.7449815, -105.2226391], [39.7451542, -105.2227925], [39.7454784, -105.2230438], [39.7455034, -105.2230623], [39.7455159, -105.2230713], [39.7455403, -105.2230861], [39.7455548, -105.2230941], [39.7455766, -105.2231014], [39.7456018, -105.223106], [39.7456176, -105.2231071], [39.7456337, -105.223107], [39.7456638, -105.2231045], [39.7456814, -105.2231011], [39.7456983, -105.2230945], [39.7457224, -105.2230821], [39.7457338, -105.2230745], [39.7457588, -105.2230543], [39.7457709, -105.223043], [39.7457834, -105.2230284], [39.7458035, -105.2230027], [39.7458905, -105.2228787], [39.7460068, -105.2226624], [39.7461589, -105.2223894], [39.7463617, -105.2220126], [39.7464427, -105.2218672], [39.746482, -105.2217968], [39.7465202, -105.2217322], [39.7465523, -105.2216771], [39.7465774, -105.2216276], [39.7465928, -105.2215823], [39.7466004, -105.2215466], [39.7466042, -105.2215101], [39.7466044, -105.2214695], [39.746601, -105.2214368], [39.7465881, -105.2213862], [39.7465804, -105.2213567], [39.7465774, -105.2213321], [39.7465774, -105.2213048], [39.746581, -105.2212762], [39.7465874, -105.2212518], [39.7465978, -105.2212266], [39.7466125, -105.2212024], [39.7466266, -105.2211858], [39.7466485, -105.2211678], [39.7466708, -105.2211566], [39.746693, -105.2211511], [39.746715, -105.2211506], [39.7467203, -105.2211517], [39.7467798, -105.2211327], [39.7468101, -105.2211204], [39.7468347, -105.2211032], [39.7468673, -105.2210676], [39.7468863, -105.2210389], [39.7469079, -105.2209995], [39.7469752, -105.2208699], [39.7471073, -105.220655], [39.7471371, -105.2206129], [39.7471594, -105.2205845], [39.7472061, -105.2205376], [39.7473376, -105.2202962], [39.747357, -105.2202177], [39.7473682, -105.2201868], [39.7473848, -105.2201519], [39.7475089, -105.2199249], [39.7475861, -105.2197973], [39.7476254, -105.2197401], [39.7476917, -105.2196525], [39.7477724, -105.2195069], [39.7478875, -105.219299], [39.7479219, -105.2192372], [39.7479518, -105.219179], [39.7480957, -105.2189201], [39.7481337, -105.2188515], [39.7482567, -105.2186294], [39.7484851, -105.2182271], [39.7485167, -105.218164], [39.7485605, -105.2180881], [39.7490516, -105.217166], [39.7490955, -105.2170866], [39.7491263, -105.2170387], [39.7493736, -105.2165801], [39.7496841, -105.2160128], [39.7497482, -105.2160733], [39.7504593, -105.216753], [39.7505084, -105.2168007], [39.7505733, -105.2168567], [39.7509525, -105.2172063], [39.7512593, -105.2174884], [39.7513304, -105.2175506], [39.7513869, -105.217604], [39.7517458, -105.2179367], [39.7520945, -105.21826], [39.7521462, -105.2183047], [39.7522126, -105.218369], [39.7523012, -105.2184523], [39.7525052, -105.218642], [39.7527145, -105.2188366], [39.7528959, -105.2189993], [39.7529601, -105.2190649], [39.7530308, -105.219128], [39.7537033, -105.2197718], [39.753764, -105.2198302], [39.7538299, -105.2198915], [39.7542346, -105.2202638], [39.7545279, -105.2205433], [39.7545847, -105.2206031], [39.7546623, -105.2206728], [39.7553326, -105.2212752], [39.755407, -105.221342], [39.7554908, -105.2214211], [39.7561483, -105.2220415], [39.7562254, -105.2221143], [39.7563006, -105.2221853], [39.7566193, -105.222483], [39.7566315, -105.2224957], [39.7569475, -105.2227913], [39.7570561, -105.222895], [39.7574792, -105.2232693], [39.7575579, -105.2233409], [39.7576236, -105.2234019], [39.7578565, -105.2236199], [39.7579703, -105.2237258], [39.7580027, -105.2237561], [39.7583001, -105.2240544], [39.7583591, -105.2241084], [39.7587249, -105.2244446], [39.7588057, -105.22452], [39.7591609, -105.2248516], [39.7595089, -105.2251741], [39.7596887, -105.2253408], [39.7597554, -105.2254024], [39.7598802, -105.2255173], [39.7598939, -105.2255305], [39.7599658, -105.2255988], [39.7600186, -105.2256498], [39.7604008, -105.2259941], [39.7605066, -105.2260931], [39.7605924, -105.226174], [39.760667, -105.2262427], [39.7607946, -105.2263631], [39.7608753, -105.2264373], [39.7615829, -105.2271044], [39.7616266, -105.2271481], [39.7616788, -105.2272008], [39.762203, -105.2276943], [39.7623411, -105.2278299], [39.7630405, -105.2284565], [39.7632824, -105.2286685], [39.7637358, -105.2290679], [39.7640137, -105.229309], [39.7640954, -105.229381], [39.7641743, -105.229462], [39.76436, -105.2296495], [39.7645376, -105.2298658], [39.764824, -105.2301918], [39.7652051, -105.2307044], [39.7654711, -105.2310802], [39.7654801, -105.2310934], [39.7655308, -105.231168], [39.7655792, -105.231232], [39.7655918, -105.2312486], [39.7657719, -105.2315035], [39.7658973, -105.2316751], [39.7659962, -105.2318039], [39.7661018, -105.231924], [39.7662469, -105.2320614], [39.7664515, -105.2322588], [39.7671442, -105.2329283], [39.7673559, -105.2331244], [39.7674414, -105.2332053], [39.7675186, -105.2332845], [39.7675756, -105.2333522], [39.7676291, -105.2334243], [39.7676969, -105.2335369], [39.7677304, -105.233602], [39.767761, -105.2336649], [39.7677908, -105.233739], [39.7678172, -105.233812], [39.7678551, -105.2339422], [39.7678736, -105.2340294], [39.7678904, -105.2341257], [39.7679032, -105.2342329], [39.7679095, -105.2343359], [39.7679068, -105.2344027], [39.7679029, -105.234499], [39.7678875, -105.2346626], [39.7678651, -105.2347791], [39.7678384, -105.2348898], [39.7678006, -105.2350283], [39.7677517, -105.2352031], [39.7677119, -105.2353468]],\n", " {"bubblingMouseEvents": true, "color": "blue", "dashArray": null, "dashOffset": null, "fill": false, "fillColor": "blue", "fillOpacity": 0.2, "fillRule": "evenodd", "lineCap": "round", "lineJoin": "round", "noClip": false, "opacity": 0.8, "smoothFactor": 1.0, "stroke": true, "weight": 10}\n", - " ).addTo(map_5e6fdd78aee60d0d6315972837ed641b);\n", + " ).addTo(map_f0bafdc10b40d238de9da7b18732b217);\n", " \n", " \n", - " poly_line_6406764e3f6bcb21b9ed2e3ae7f61509.bindTooltip(\n", + " poly_line_51128cf08c47e0b255fa936899f6e0e4.bindTooltip(\n", " `<div>\n", " Shortest Time\n", " </div>`,\n", @@ -491,19 +480,19 @@ " );\n", " \n", " \n", - " var marker_6902ee7feb3c91f411bef91fffcafda8 = L.marker(\n", + " var marker_eb6728cbb45319dfc3741bebb57bec6e = L.marker(\n", " [39.7273069, -105.1994169],\n", " {}\n", - " ).addTo(map_5e6fdd78aee60d0d6315972837ed641b);\n", + " ).addTo(map_f0bafdc10b40d238de9da7b18732b217);\n", " \n", " \n", - " var icon_45c70091cf67f498c42e94b48f45100f = L.AwesomeMarkers.icon(\n", + " var icon_c34923765e4fff2b62cddf1bb4a18bea = L.AwesomeMarkers.icon(\n", " {"extraClasses": "fa-rotate-0", "icon": "circle", "iconColor": "white", "markerColor": "green", "prefix": "fa"}\n", " );\n", - " marker_6902ee7feb3c91f411bef91fffcafda8.setIcon(icon_45c70091cf67f498c42e94b48f45100f);\n", + " marker_eb6728cbb45319dfc3741bebb57bec6e.setIcon(icon_c34923765e4fff2b62cddf1bb4a18bea);\n", " \n", " \n", - " marker_6902ee7feb3c91f411bef91fffcafda8.bindTooltip(\n", + " marker_eb6728cbb45319dfc3741bebb57bec6e.bindTooltip(\n", " `<div>\n", " Origin\n", " </div>`,\n", @@ -511,19 +500,19 @@ " );\n", " \n", " \n", - " var marker_8561d2d910a8be3bc5d2cfa6a30c6063 = L.marker(\n", + " var marker_4ca91c0844bc6f79cdde370a45e8cf3c = L.marker(\n", " [39.7677119, -105.2353468],\n", " {}\n", - " ).addTo(map_5e6fdd78aee60d0d6315972837ed641b);\n", + " ).addTo(map_f0bafdc10b40d238de9da7b18732b217);\n", " \n", " \n", - " var icon_fea81095d90d74dad83b121c4e124834 = L.AwesomeMarkers.icon(\n", + " var icon_83eb5299c606253f41a628839749ec1f = L.AwesomeMarkers.icon(\n", " {"extraClasses": "fa-rotate-0", "icon": "circle", "iconColor": "white", "markerColor": "red", "prefix": "fa"}\n", " );\n", - " marker_8561d2d910a8be3bc5d2cfa6a30c6063.setIcon(icon_fea81095d90d74dad83b121c4e124834);\n", + " marker_4ca91c0844bc6f79cdde370a45e8cf3c.setIcon(icon_83eb5299c606253f41a628839749ec1f);\n", " \n", " \n", - " marker_8561d2d910a8be3bc5d2cfa6a30c6063.bindTooltip(\n", + " marker_4ca91c0844bc6f79cdde370a45e8cf3c.bindTooltip(\n", " `<div>\n", " Destination\n", " </div>`,\n", @@ -531,13 +520,13 @@ " );\n", " \n", " \n", - " var poly_line_3349ea569e76e0f0fe319322fd08afcb = L.polyline(\n", + " var poly_line_95e6364ccd9609756d15b7d7194b0d25 = L.polyline(\n", " [[39.7273069, -105.1994169], [39.7272989, -105.1994535], [39.7272848, -105.1994868], [39.7272593, -105.1995218], [39.7272272, -105.1995461], [39.7271912, -105.1995576], [39.7271688, -105.199558], [39.7271461, -105.1995532], [39.7271142, -105.1995365], [39.7270868, -105.1995096], [39.7270657, -105.1994741], [39.7270519, -105.1994296], [39.7270483, -105.1993816], [39.727052, -105.1993539], [39.7270571, -105.1993247], [39.7270719, -105.1992916], [39.7270973, -105.1992566], [39.7271293, -105.1992323], [39.7271651, -105.1992207], [39.7271854, -105.1992226], [39.7272113, -105.199227], [39.7273496, -105.1991687], [39.7274185, -105.1991465], [39.7275326, -105.1990928], [39.7276828, -105.1990602], [39.7277455, -105.1990518], [39.7279702, -105.1990232], [39.7281558, -105.1990357], [39.7284384, -105.1990893], [39.7286639, -105.199177], [39.7288222, -105.1992773], [39.7289214, -105.199385], [39.7290292, -105.1995543], [39.7290506, -105.1995879], [39.7292157, -105.1998155], [39.7293094, -105.1998945], [39.7294079, -105.1999604], [39.7295437, -105.200014], [39.7296299, -105.2000721], [39.7296469, -105.1996789], [39.7296738, -105.1987284], [39.7296906, -105.1986794], [39.7297244, -105.1985856], [39.7297272, -105.1984193], [39.7299723, -105.1984198], [39.7302117, -105.1984227], [39.7303235, -105.1984285], [39.7304434, -105.198443], [39.7305651, -105.1984732], [39.7307182, -105.1985167], [39.7308487, -105.1985623], [39.7309909, -105.1986311], [39.731127, -105.1987144], [39.7312292, -105.1987897], [39.7313292, -105.1988731], [39.7314292, -105.1989611], [39.731524, -105.1990554], [39.7316089, -105.1991466], [39.7317085, -105.1992639], [39.7318575, -105.1994488], [39.7319522, -105.1995697], [39.7320423, -105.1996716], [39.7321372, -105.1997578], [39.7322615, -105.1998506], [39.7323697, -105.1999163], [39.7324712, -105.1999637], [39.7325898, -105.2000081], [39.732689, -105.2000309], [39.7327987, -105.2000627], [39.732945, -105.2001049], [39.7330635, -105.2001374], [39.7331838, -105.2001914], [39.7332899, -105.2002528], [39.7334312, -105.200363], [39.7335257, -105.2004544], [39.7335523, -105.2004835], [39.7336132, -105.2005503], [39.7336902, -105.2006555], [39.733767, -105.2007557], [39.7338411, -105.2008491], [39.7339359, -105.2009412], [39.7340357, -105.2010154], [39.7340742, -105.201042], [39.7341117, -105.2010678], [39.7341542, -105.2010901], [39.734211, -105.2011198], [39.7343173, -105.2011712], [39.7344056, -105.2012021], [39.7345704, -105.2012376], [39.7347726, -105.2012753], [39.7351329, -105.2013311], [39.7352203, -105.2013441], [39.7353243, -105.201352], [39.7354415, -105.2013441], [39.7355491, -105.2013268], [39.7357022, -105.2012901], [39.7358395, -105.201259], [39.7362429, -105.2011675], [39.7367157, -105.2010639], [39.7367857, -105.2010539], [39.7368531, -105.2010442], [39.7369645, -105.2010356], [39.7370755, -105.201044], [39.7371895, -105.2010632], [39.7373179, -105.201099], [39.737714, -105.2012295], [39.7379567, -105.2013009], [39.7384451, -105.2014404], [39.7386872, -105.2015167], [39.7395717, -105.2017905], [39.7397153, -105.2018369], [39.7398269, -105.2018673], [39.7399356, -105.2018843], [39.7400618, -105.2018961], [39.740184, -105.2018932], [39.7403133, -105.2018823], [39.7404288, -105.2018645], [39.7405518, -105.2018387], [39.740665, -105.2018073], [39.7407597, -105.2017713], [39.7408728, -105.2017196], [39.740992, -105.2016551], [39.7411268, -105.2015625], [39.7412599, -105.2014489], [39.7413718, -105.2013453], [39.7414758, -105.2012326], [39.741608, -105.2010691], [39.7416901, -105.200964], [39.7417903, -105.2008396], [39.7417994, -105.200826], [39.7418968, -105.200686], [39.7418878, -105.200623], [39.7418917, -105.2005592], [39.7419081, -105.2004988], [39.7419393, -105.2004416], [39.7419818, -105.2003982], [39.7420319, -105.2003723], [39.7420864, -105.2003661], [39.7421399, -105.2003805], [39.742188, -105.2004144], [39.7422255, -105.2004628], [39.742251, -105.2005229], [39.7422628, -105.2005898], [39.7422597, -105.2006584], [39.7422421, -105.2007231], [39.7422712, -105.2008198], [39.7423227, -105.200901], [39.7428069, -105.2016549], [39.7429048, -105.2018114], [39.7429131, -105.2018245], [39.7431503, -105.2022035], [39.7434374, -105.2026477], [39.7435807, -105.2028659], [39.7437705, -105.2031843], [39.7441255, -105.2037193], [39.7444536, -105.2042289], [39.7445369, -105.204369], [39.7448087, -105.2048085], [39.7450389, -105.205197], [39.745164, -105.2054397], [39.7452617, -105.205642], [39.745359, -105.2058753], [39.7454413, -105.2060693], [39.7454698, -105.2061365], [39.7455586, -105.2063498], [39.7456394, -105.2066197], [39.7457343, -105.2068984], [39.7458096, -105.2071619], [39.7458432, -105.2072893], [39.7459057, -105.2075162], [39.7459547, -105.2077463], [39.7460104, -105.2080066], [39.7460312, -105.2081149], [39.7460508, -105.2082709], [39.7460836, -105.2084787], [39.7461092, -105.2086125], [39.7461165, -105.2087078], [39.7461128, -105.2088019], [39.7461078, -105.2089177], [39.7461163, -105.208995], [39.7461377, -105.2090672], [39.7461588, -105.2091147], [39.7462509, -105.2092159], [39.746281, -105.2092382], [39.7462978, -105.209256], [39.7463242, -105.209292], [39.7463458, -105.2093473], [39.746361, -105.2094251], [39.7463921, -105.2095768], [39.7464353, -105.2097133], [39.7465001, -105.2098285], [39.7467097, -105.21004], [39.7467814, -105.2101094], [39.7473357, -105.2106047], [39.7474554, -105.2107116], [39.7475615, -105.2108064], [39.7475958, -105.2108371], [39.747653, -105.2108907], [39.7481502, -105.2113567], [39.7483361, -105.2115309], [39.7483726, -105.2115651], [39.7484183, -105.2116079], [39.7484718, -105.2116584], [39.7487212, -105.2118951], [39.749238, -105.2123721], [39.7492814, -105.2124135], [39.7500059, -105.2130759], [39.7500454, -105.2131129], [39.7503812, -105.2134256], [39.7504627, -105.2135015], [39.7508219, -105.2138362], [39.7508792, -105.2138895], [39.7509348, -105.2139426], [39.7511093, -105.2141093], [39.7516271, -105.2146038], [39.7516716, -105.2146463], [39.751739, -105.2147085], [39.7524829, -105.2153953], [39.7525012, -105.2154123], [39.7532517, -105.2161101], [39.7533052, -105.2161598], [39.7540716, -105.2168674], [39.7541308, -105.2169221], [39.7544795, -105.2172508], [39.7546281, -105.2174522], [39.7546745, -105.2175037], [39.7548394, -105.2176869], [39.7549112, -105.2177871], [39.7548521, -105.2179433], [39.7546476, -105.2183047], [39.7544379, -105.2186805], [39.7544293, -105.218688], [39.7543723, -105.2187394], [39.7544376, -105.2188002], [39.7544872, -105.2188464], [39.7547934, -105.2191313], [39.7551071, -105.2194077], [39.7551368, -105.2194339], [39.755202, -105.2194913], [39.7552559, -105.2195411], [39.755615, -105.2198733], [39.755696, -105.2199479], [39.7559351, -105.2201693], [39.7559559, -105.220191], [39.7560109, -105.2202486], [39.7560703, -105.2203182], [39.756257, -105.2205368], [39.7562813, -105.2205598], [39.7564057, -105.2206774], [39.7565461, -105.2208058], [39.7567417, -105.2209442], [39.7566986, -105.2211001], [39.756638, -105.2213112], [39.7565732, -105.2214643], [39.7565209, -105.2215678], [39.7564595, -105.2216814], [39.7562914, -105.2219923], [39.7562254, -105.2221143], [39.7563006, -105.2221853], [39.7566193, -105.222483], [39.7566315, -105.2224957], [39.7569475, -105.2227913], [39.7570561, -105.222895], [39.7574792, -105.2232693], [39.7575579, -105.2233409], [39.7576236, -105.2234019], [39.7578565, -105.2236199], [39.7579703, -105.2237258], [39.7580027, -105.2237561], [39.7583001, -105.2240544], [39.7583591, -105.2241084], [39.7587249, -105.2244446], [39.7588057, -105.22452], [39.7591609, -105.2248516], [39.7595089, -105.2251741], [39.7596887, -105.2253408], [39.7597554, -105.2254024], [39.7598802, -105.2255173], [39.7598939, -105.2255305], [39.7599658, -105.2255988], [39.7600186, -105.2256498], [39.7604008, -105.2259941], [39.7605066, -105.2260931], [39.7605924, -105.226174], [39.760667, -105.2262427], [39.7607946, -105.2263631], [39.7608753, -105.2264373], [39.7615829, -105.2271044], [39.7616266, -105.2271481], [39.7616788, -105.2272008], [39.762203, -105.2276943], [39.7623411, -105.2278299], [39.7630405, -105.2284565], [39.7632824, -105.2286685], [39.7637358, -105.2290679], [39.7640137, -105.229309], [39.7640954, -105.229381], [39.7641743, -105.229462], [39.76436, -105.2296495], [39.7645376, -105.2298658], [39.764824, -105.2301918], [39.7652051, -105.2307044], [39.7654711, -105.2310802], [39.7654801, -105.2310934], [39.7655308, -105.231168], [39.7655792, -105.231232], [39.7655918, -105.2312486], [39.7657719, -105.2315035], [39.7658973, -105.2316751], [39.7659962, -105.2318039], [39.7661018, -105.231924], [39.7662469, -105.2320614], [39.7664515, -105.2322588], [39.7671442, -105.2329283], [39.7673559, -105.2331244], [39.7674414, -105.2332053], [39.7675186, -105.2332845], [39.7675756, -105.2333522], [39.7676291, -105.2334243], [39.7676969, -105.2335369], [39.7677304, -105.233602], [39.767761, -105.2336649], [39.7677908, -105.233739], [39.7678172, -105.233812], [39.7678551, -105.2339422], [39.7678736, -105.2340294], [39.7678904, -105.2341257], [39.7679032, -105.2342329], [39.7679095, -105.2343359], [39.7679068, -105.2344027], [39.7679029, -105.234499], [39.7678875, -105.2346626], [39.7678651, -105.2347791], [39.7678384, -105.2348898], [39.7678006, -105.2350283], [39.7677517, -105.2352031], [39.7677119, -105.2353468]],\n", " {"bubblingMouseEvents": true, "color": "green", "dashArray": null, "dashOffset": null, "fill": false, "fillColor": "green", "fillOpacity": 0.2, "fillRule": "evenodd", "lineCap": "round", "lineJoin": "round", "noClip": false, "opacity": 0.8, "smoothFactor": 1.0, "stroke": true, "weight": 10}\n", - " ).addTo(map_5e6fdd78aee60d0d6315972837ed641b);\n", + " ).addTo(map_f0bafdc10b40d238de9da7b18732b217);\n", " \n", " \n", - " poly_line_3349ea569e76e0f0fe319322fd08afcb.bindTooltip(\n", + " poly_line_95e6364ccd9609756d15b7d7194b0d25.bindTooltip(\n", " `<div>\n", " Least Energy\n", " </div>`,\n", @@ -545,19 +534,19 @@ " );\n", " \n", " \n", - " var marker_b9a3acafc0bb939f9f5fb5f2fa8ba7f7 = L.marker(\n", + " var marker_b7c7d5ee54034a20d3fee0fe3d1733e2 = L.marker(\n", " [39.7273069, -105.1994169],\n", " {}\n", - " ).addTo(map_5e6fdd78aee60d0d6315972837ed641b);\n", + " ).addTo(map_f0bafdc10b40d238de9da7b18732b217);\n", " \n", " \n", - " var icon_d6912a1b5e7e079949d11e32e00dea90 = L.AwesomeMarkers.icon(\n", + " var icon_a0c3146e5134d59292b0627e0c924a1f = L.AwesomeMarkers.icon(\n", " {"extraClasses": "fa-rotate-0", "icon": "circle", "iconColor": "white", "markerColor": "green", "prefix": "fa"}\n", " );\n", - " marker_b9a3acafc0bb939f9f5fb5f2fa8ba7f7.setIcon(icon_d6912a1b5e7e079949d11e32e00dea90);\n", + " marker_b7c7d5ee54034a20d3fee0fe3d1733e2.setIcon(icon_a0c3146e5134d59292b0627e0c924a1f);\n", " \n", " \n", - " marker_b9a3acafc0bb939f9f5fb5f2fa8ba7f7.bindTooltip(\n", + " marker_b7c7d5ee54034a20d3fee0fe3d1733e2.bindTooltip(\n", " `<div>\n", " Origin\n", " </div>`,\n", @@ -565,19 +554,19 @@ " );\n", " \n", " \n", - " var marker_d09760fd1b167457ff8d82085d575f1e = L.marker(\n", + " var marker_6fa879a02d34f1b588c3db5c0a5b6446 = L.marker(\n", " [39.7677119, -105.2353468],\n", " {}\n", - " ).addTo(map_5e6fdd78aee60d0d6315972837ed641b);\n", + " ).addTo(map_f0bafdc10b40d238de9da7b18732b217);\n", " \n", " \n", - " var icon_068d667c897f021e6c99fb3531ba48e3 = L.AwesomeMarkers.icon(\n", + " var icon_e58bcdf28679f4df3d0e2476c62b2048 = L.AwesomeMarkers.icon(\n", " {"extraClasses": "fa-rotate-0", "icon": "circle", "iconColor": "white", "markerColor": "red", "prefix": "fa"}\n", " );\n", - " marker_d09760fd1b167457ff8d82085d575f1e.setIcon(icon_068d667c897f021e6c99fb3531ba48e3);\n", + " marker_6fa879a02d34f1b588c3db5c0a5b6446.setIcon(icon_e58bcdf28679f4df3d0e2476c62b2048);\n", " \n", " \n", - " marker_d09760fd1b167457ff8d82085d575f1e.bindTooltip(\n", + " marker_6fa879a02d34f1b588c3db5c0a5b6446.bindTooltip(\n", " `<div>\n", " Destination\n", " </div>`,\n", @@ -588,10 +577,10 @@ "</html>\" style=\"position:absolute;width:100%;height:100%;left:0;top:0;border:none !important;\" allowfullscreen webkitallowfullscreen mozallowfullscreen></iframe></div></div>" ], "text/plain": [ - "<folium.folium.Map at 0x123135d10>" + "<folium.folium.Map at 0x1231d5ad0>" ] }, - "execution_count": 15, + "execution_count": 14, "metadata": {}, "output_type": "execute_result" } @@ -612,7 +601,7 @@ }, { "cell_type": "code", - "execution_count": 16, + "execution_count": 15, "id": "aa1e6b11", "metadata": {}, "outputs": [ @@ -646,7 +635,7 @@ " <meta name="viewport" content="width=device-width,\n", " initial-scale=1.0, maximum-scale=1.0, user-scalable=no" />\n", " <style>\n", - " #map_f5e9fdbd4de71b8ef0064aeb6f947a41 {\n", + " #map_ad8e071bae664cc7b1a973d4509a33a8 {\n", " position: relative;\n", " width: 100.0%;\n", " height: 100.0%;\n", @@ -660,14 +649,14 @@ "<body>\n", " \n", " \n", - " <div class="folium-map" id="map_f5e9fdbd4de71b8ef0064aeb6f947a41" ></div>\n", + " <div class="folium-map" id="map_ad8e071bae664cc7b1a973d4509a33a8" ></div>\n", " \n", "</body>\n", "<script>\n", " \n", " \n", - " var map_f5e9fdbd4de71b8ef0064aeb6f947a41 = L.map(\n", - " "map_f5e9fdbd4de71b8ef0064aeb6f947a41",\n", + " var map_ad8e071bae664cc7b1a973d4509a33a8 = L.map(\n", + " "map_ad8e071bae664cc7b1a973d4509a33a8",\n", " {\n", " center: [39.7465774, -105.2216276],\n", " crs: L.CRS.EPSG3857,\n", @@ -681,39 +670,39 @@ "\n", " \n", " \n", - " var tile_layer_b22ff300046905100f9b524af60a2514 = L.tileLayer(\n", + " var tile_layer_5afdab62ec2b889ffe53761579c5a24a = L.tileLayer(\n", " "https://{s}.tile.openstreetmap.org/{z}/{x}/{y}.png",\n", " {"attribution": "Data by \\u0026copy; \\u003ca target=\\"_blank\\" href=\\"http://openstreetmap.org\\"\\u003eOpenStreetMap\\u003c/a\\u003e, under \\u003ca target=\\"_blank\\" href=\\"http://www.openstreetmap.org/copyright\\"\\u003eODbL\\u003c/a\\u003e.", "detectRetina": false, "maxNativeZoom": 18, "maxZoom": 18, "minZoom": 0, "noWrap": false, "opacity": 1, "subdomains": "abc", "tms": false}\n", - " ).addTo(map_f5e9fdbd4de71b8ef0064aeb6f947a41);\n", + " ).addTo(map_ad8e071bae664cc7b1a973d4509a33a8);\n", " \n", " \n", - " var poly_line_8ee9a6fcbe409c7f61afa98464585f6a = L.polyline(\n", + " var poly_line_da49ac63cdfeea189953099ec23dbaf5 = L.polyline(\n", " [[39.7273069, -105.1994169], [39.7272989, -105.1994535], [39.7272848, -105.1994868], [39.7272593, -105.1995218], [39.7272272, -105.1995461], [39.7271912, -105.1995576], [39.7271688, -105.199558], [39.7271461, -105.1995532], [39.7271142, -105.1995365], [39.7270868, -105.1995096], [39.7270657, -105.1994741], [39.7270519, -105.1994296], [39.7270483, -105.1993816], [39.727052, -105.1993539], [39.7270571, -105.1993247], [39.7270719, -105.1992916], [39.7270973, -105.1992566], [39.7271293, -105.1992323], [39.7271651, -105.1992207], [39.7271854, -105.1992226], [39.7272113, -105.199227], [39.7273496, -105.1991687], [39.7274185, -105.1991465], [39.7275326, -105.1990928], [39.7276828, -105.1990602], [39.7277455, -105.1990518], [39.7279702, -105.1990232], [39.7281558, -105.1990357], [39.7284384, -105.1990893], [39.7286639, -105.199177], [39.7288222, -105.1992773], [39.7289214, -105.199385], [39.7290292, -105.1995543], [39.7290506, -105.1995879], [39.7292157, -105.1998155], [39.7293094, -105.1998945], [39.7294079, -105.1999604], [39.7295437, -105.200014], [39.7296299, -105.2000721], [39.7297292, -105.2000757], [39.7297135, -105.2007081], [39.7297192, -105.2013106], [39.7298153, -105.2032041], [39.7298256, -105.2033784], [39.7298664, -105.2035708], [39.7298795, -105.203649], [39.7299267, -105.2038558], [39.7301082, -105.2042608], [39.7304911, -105.2048068], [39.7308741, -105.2053111], [39.7310676, -105.2055985], [39.7311933, -105.205824], [39.7312069, -105.2058543], [39.7313075, -105.2061158], [39.731348, -105.206248], [39.731386, -105.2064213], [39.7314237, -105.2067336], [39.731429, -105.2070565], [39.7314226, -105.207224], [39.7314076, -105.2074036], [39.7313611, -105.2077675], [39.7313265, -105.2079124], [39.7312831, -105.20806], [39.7312292, -105.2082086], [39.7311706, -105.208344], [39.7310965, -105.2084777], [39.7310049, -105.208616], [39.7309234, -105.2087235], [39.7308212, -105.2088303], [39.7306227, -105.2089819], [39.7304801, -105.2090736], [39.7303681, -105.2091988], [39.7303327, -105.2092597], [39.7303114, -105.2093181], [39.730299, -105.20939], [39.7302939, -105.2095126], [39.7303105, -105.2096804], [39.7304414, -105.2099267], [39.7305001, -105.2100344], [39.7305871, -105.2101842], [39.7307404, -105.2104481], [39.7309234, -105.2107229], [39.7310934, -105.2109639], [39.731238, -105.211145], [39.7314101, -105.2113464], [39.7315783, -105.2115336], [39.7317541, -105.2117064], [39.7319796, -105.2119089], [39.732359, -105.2122393], [39.7329187, -105.2127184], [39.7334557, -105.213188], [39.7338968, -105.2135676], [39.7342777, -105.2138968], [39.7347139, -105.2142786], [39.7354652, -105.2149301], [39.7356319, -105.2150733], [39.7358844, -105.2152902], [39.7363901, -105.2157275], [39.7369129, -105.2161859], [39.7377243, -105.2168831], [39.7390882, -105.2180682], [39.7397882, -105.2186718], [39.7403043, -105.2191078], [39.7405645, -105.2193336], [39.7406767, -105.2194225], [39.7415803, -105.2202142], [39.7416486, -105.2202727], [39.7417047, -105.2203151], [39.7421502, -105.2206932], [39.7425816, -105.221062], [39.7432687, -105.2215398], [39.7435797, -105.2217631], [39.7436721, -105.2218299], [39.7437366, -105.2218764], [39.7438794, -105.2219759], [39.7439546, -105.2220253], [39.7440082, -105.2220586], [39.7440819, -105.2221021], [39.7441377, -105.2221322], [39.7441808, -105.2221541], [39.7442796, -105.2222043], [39.7443357, -105.2222286], [39.7444062, -105.222259], [39.7444657, -105.2222871], [39.7444962, -105.2223029], [39.7445695, -105.2223431], [39.7446343, -105.2223808], [39.744683, -105.2224104], [39.7447215, -105.2224349], [39.7447658, -105.2224654], [39.744813, -105.222499], [39.7448547, -105.2225315], [39.7448881, -105.2225586], [39.7449328, -105.2225965], [39.7449815, -105.2226391], [39.7451542, -105.2227925], [39.7454784, -105.2230438], [39.7455034, -105.2230623], [39.7455159, -105.2230713], [39.7455403, -105.2230861], [39.7455548, -105.2230941], [39.7455766, -105.2231014], [39.7456018, -105.223106], [39.7456176, -105.2231071], [39.7456337, -105.223107], [39.7456638, -105.2231045], [39.7456814, -105.2231011], [39.7456983, -105.2230945], [39.7457224, -105.2230821], [39.7457338, -105.2230745], [39.7457588, -105.2230543], [39.7457709, -105.223043], [39.7457834, -105.2230284], [39.7458035, -105.2230027], [39.7458905, -105.2228787], [39.7460068, -105.2226624], [39.7461589, -105.2223894], [39.7463617, -105.2220126], [39.7464427, -105.2218672], [39.746482, -105.2217968], [39.7465202, -105.2217322], [39.7465523, -105.2216771], [39.7465774, -105.2216276], [39.7465928, -105.2215823], [39.7466004, -105.2215466], [39.7466042, -105.2215101], [39.7466044, -105.2214695], [39.746601, -105.2214368], [39.7465881, -105.2213862], [39.7465804, -105.2213567], [39.7465774, -105.2213321], [39.7465774, -105.2213048], [39.746581, -105.2212762], [39.7465874, -105.2212518], [39.7465978, -105.2212266], [39.7466125, -105.2212024], [39.7466266, -105.2211858], [39.7466485, -105.2211678], [39.7466708, -105.2211566], [39.746693, -105.2211511], [39.746715, -105.2211506], [39.7467203, -105.2211517], [39.7467798, -105.2211327], [39.7468101, -105.2211204], [39.7468347, -105.2211032], [39.7468673, -105.2210676], [39.7468863, -105.2210389], [39.7469079, -105.2209995], [39.7469752, -105.2208699], [39.7471073, -105.220655], [39.7471371, -105.2206129], [39.7471594, -105.2205845], [39.7472061, -105.2205376], [39.7473376, -105.2202962], [39.747357, -105.2202177], [39.7473682, -105.2201868], [39.7473848, -105.2201519], [39.7475089, -105.2199249], [39.7475861, -105.2197973], [39.7476254, -105.2197401], [39.7476917, -105.2196525], [39.7477724, -105.2195069], [39.7478875, -105.219299], [39.7479219, -105.2192372], [39.7479518, -105.219179], [39.7480957, -105.2189201], [39.7481337, -105.2188515], [39.7482567, -105.2186294], [39.7484851, -105.2182271], [39.7485167, -105.218164], [39.7485605, -105.2180881], [39.7490516, -105.217166], [39.7490955, -105.2170866], [39.7491263, -105.2170387], [39.7493736, -105.2165801], [39.7496841, -105.2160128], [39.7497482, -105.2160733], [39.7504593, -105.216753], [39.7505084, -105.2168007], [39.7505733, -105.2168567], [39.7509525, -105.2172063], [39.7512593, -105.2174884], [39.7513304, -105.2175506], [39.7513869, -105.217604], [39.7517458, -105.2179367], [39.7520945, -105.21826], [39.7521462, -105.2183047], [39.7522126, -105.218369], [39.7523012, -105.2184523], [39.7525052, -105.218642], [39.7527145, -105.2188366], [39.7528959, -105.2189993], [39.7529601, -105.2190649], [39.7530308, -105.219128], [39.7537033, -105.2197718], [39.753764, -105.2198302], [39.7538299, -105.2198915], [39.7542346, -105.2202638], [39.7545279, -105.2205433], [39.7545847, -105.2206031], [39.7546623, -105.2206728], [39.7553326, -105.2212752], [39.755407, -105.221342], [39.7554908, -105.2214211], [39.7561483, -105.2220415], [39.7562254, -105.2221143], [39.7563006, -105.2221853], [39.7566193, -105.222483], [39.7566315, -105.2224957], [39.7569475, -105.2227913], [39.7570561, -105.222895], [39.7574792, -105.2232693], [39.7575579, -105.2233409], [39.7576236, -105.2234019], [39.7578565, -105.2236199], [39.7579703, -105.2237258], [39.7580027, -105.2237561], [39.7583001, -105.2240544], [39.7583591, -105.2241084], [39.7587249, -105.2244446], [39.7588057, -105.22452], [39.7591609, -105.2248516], [39.7595089, -105.2251741], [39.7596887, -105.2253408], [39.7597554, -105.2254024], [39.7598802, -105.2255173], [39.7598939, -105.2255305], [39.7599658, -105.2255988], [39.7600186, -105.2256498], [39.7604008, -105.2259941], [39.7605066, -105.2260931], [39.7605924, -105.226174], [39.760667, -105.2262427], [39.7607946, -105.2263631], [39.7608753, -105.2264373], [39.7615829, -105.2271044], [39.7616266, -105.2271481], [39.7616788, -105.2272008], [39.762203, -105.2276943], [39.7623411, -105.2278299], [39.7630405, -105.2284565], [39.7632824, -105.2286685], [39.7637358, -105.2290679], [39.7640137, -105.229309], [39.7640954, -105.229381], [39.7641743, -105.229462], [39.76436, -105.2296495], [39.7645376, -105.2298658], [39.764824, -105.2301918], [39.7652051, -105.2307044], [39.7654711, -105.2310802], [39.7654801, -105.2310934], [39.7655308, -105.231168], [39.7655792, -105.231232], [39.7655918, -105.2312486], [39.7657719, -105.2315035], [39.7658973, -105.2316751], [39.7659962, -105.2318039], [39.7661018, -105.231924], [39.7662469, -105.2320614], [39.7664515, -105.2322588], [39.7671442, -105.2329283], [39.7673559, -105.2331244], [39.7674414, -105.2332053], [39.7675186, -105.2332845], [39.7675756, -105.2333522], [39.7676291, -105.2334243], [39.7676969, -105.2335369], [39.7677304, -105.233602], [39.767761, -105.2336649], [39.7677908, -105.233739], [39.7678172, -105.233812], [39.7678551, -105.2339422], [39.7678736, -105.2340294], [39.7678904, -105.2341257], [39.7679032, -105.2342329], [39.7679095, -105.2343359], [39.7679068, -105.2344027], [39.7679029, -105.234499], [39.7678875, -105.2346626], [39.7678651, -105.2347791], [39.7678384, -105.2348898], [39.7678006, -105.2350283], [39.7677517, -105.2352031], [39.7677119, -105.2353468]],\n", " {"bubblingMouseEvents": true, "color": "#eff821", "dashArray": null, "dashOffset": null, "fill": false, "fillColor": "#eff821", "fillOpacity": 0.2, "fillRule": "evenodd", "lineCap": "round", "lineJoin": "round", "noClip": false, "opacity": 0.8, "smoothFactor": 1.0, "stroke": true, "weight": 10}\n", - " ).addTo(map_f5e9fdbd4de71b8ef0064aeb6f947a41);\n", + " ).addTo(map_ad8e071bae664cc7b1a973d4509a33a8);\n", " \n", " \n", - " poly_line_8ee9a6fcbe409c7f61afa98464585f6a.bindTooltip(\n", + " poly_line_da49ac63cdfeea189953099ec23dbaf5.bindTooltip(\n", " `<div>\n", - " 0.18821565778485838\n", + " 0.18816139103031826\n", " </div>`,\n", " {"sticky": true}\n", " );\n", " \n", " \n", - " var marker_f42f7bb5b3e19865927ef3411c0a0a48 = L.marker(\n", + " var marker_e384d848ce36e7fbaacb0ebe3c380791 = L.marker(\n", " [39.7273069, -105.1994169],\n", " {}\n", - " ).addTo(map_f5e9fdbd4de71b8ef0064aeb6f947a41);\n", + " ).addTo(map_ad8e071bae664cc7b1a973d4509a33a8);\n", " \n", " \n", - " var icon_15aeb0dc4825e560514b0f26c8c5cb20 = L.AwesomeMarkers.icon(\n", + " var icon_b55345fc80c9f7480e986bedc3b62179 = L.AwesomeMarkers.icon(\n", " {"extraClasses": "fa-rotate-0", "icon": "circle", "iconColor": "white", "markerColor": "green", "prefix": "fa"}\n", " );\n", - " marker_f42f7bb5b3e19865927ef3411c0a0a48.setIcon(icon_15aeb0dc4825e560514b0f26c8c5cb20);\n", + " marker_e384d848ce36e7fbaacb0ebe3c380791.setIcon(icon_b55345fc80c9f7480e986bedc3b62179);\n", " \n", " \n", - " marker_f42f7bb5b3e19865927ef3411c0a0a48.bindTooltip(\n", + " marker_e384d848ce36e7fbaacb0ebe3c380791.bindTooltip(\n", " `<div>\n", " Origin\n", " </div>`,\n", @@ -721,19 +710,19 @@ " );\n", " \n", " \n", - " var marker_bd64e10e467bfaa6054456bc65c65dcf = L.marker(\n", + " var marker_0e442f1cd248b85e7b27087a06524f49 = L.marker(\n", " [39.7677119, -105.2353468],\n", " {}\n", - " ).addTo(map_f5e9fdbd4de71b8ef0064aeb6f947a41);\n", + " ).addTo(map_ad8e071bae664cc7b1a973d4509a33a8);\n", " \n", " \n", - " var icon_3e1f9de6460d5c0512d0986e97601baa = L.AwesomeMarkers.icon(\n", + " var icon_dcd2bb1e268f16dded3a453f250931ef = L.AwesomeMarkers.icon(\n", " {"extraClasses": "fa-rotate-0", "icon": "circle", "iconColor": "white", "markerColor": "red", "prefix": "fa"}\n", " );\n", - " marker_bd64e10e467bfaa6054456bc65c65dcf.setIcon(icon_3e1f9de6460d5c0512d0986e97601baa);\n", + " marker_0e442f1cd248b85e7b27087a06524f49.setIcon(icon_dcd2bb1e268f16dded3a453f250931ef);\n", " \n", " \n", - " marker_bd64e10e467bfaa6054456bc65c65dcf.bindTooltip(\n", + " marker_0e442f1cd248b85e7b27087a06524f49.bindTooltip(\n", " `<div>\n", " Destination\n", " </div>`,\n", @@ -741,33 +730,33 @@ " );\n", " \n", " \n", - " var poly_line_63ad4d4e5561017b7e7fd40a459bf578 = L.polyline(\n", + " var poly_line_453a4ea787ffe3c7d6510177b14e1d10 = L.polyline(\n", " [[39.7273069, -105.1994169], [39.7272989, -105.1994535], [39.7272848, -105.1994868], [39.7272593, -105.1995218], [39.7272272, -105.1995461], [39.7271912, -105.1995576], [39.7271688, -105.199558], [39.7271461, -105.1995532], [39.7271142, -105.1995365], [39.7270868, -105.1995096], [39.7270657, -105.1994741], [39.7270519, -105.1994296], [39.7270483, -105.1993816], [39.727052, -105.1993539], [39.7270571, -105.1993247], [39.7270719, -105.1992916], [39.7270973, -105.1992566], [39.7271293, -105.1992323], [39.7271651, -105.1992207], [39.7271854, -105.1992226], [39.7272113, -105.199227], [39.7273496, -105.1991687], [39.7274185, -105.1991465], [39.7275326, -105.1990928], [39.7276828, -105.1990602], [39.7277455, -105.1990518], [39.7279702, -105.1990232], [39.7281558, -105.1990357], [39.7284384, -105.1990893], [39.7286639, -105.199177], [39.7288222, -105.1992773], [39.7289214, -105.199385], [39.7290292, -105.1995543], [39.7290506, -105.1995879], [39.7292157, -105.1998155], [39.7293094, -105.1998945], [39.7294079, -105.1999604], [39.7295437, -105.200014], [39.7296299, -105.2000721], [39.7296469, -105.1996789], [39.7296738, -105.1987284], [39.7296906, -105.1986794], [39.7297244, -105.1985856], [39.7297272, -105.1984193], [39.7299723, -105.1984198], [39.7302117, -105.1984227], [39.7303235, -105.1984285], [39.7304434, -105.198443], [39.7305651, -105.1984732], [39.7307182, -105.1985167], [39.7308487, -105.1985623], [39.7309909, -105.1986311], [39.731127, -105.1987144], [39.7312292, -105.1987897], [39.7313292, -105.1988731], [39.7314292, -105.1989611], [39.731524, -105.1990554], [39.7316089, -105.1991466], [39.7317085, -105.1992639], [39.7318575, -105.1994488], [39.7319522, -105.1995697], [39.7320423, -105.1996716], [39.7321372, -105.1997578], [39.7322615, -105.1998506], [39.7323697, -105.1999163], [39.7324712, -105.1999637], [39.7325898, -105.2000081], [39.732689, -105.2000309], [39.7327987, -105.2000627], [39.732945, -105.2001049], [39.7330635, -105.2001374], [39.7331838, -105.2001914], [39.7332899, -105.2002528], [39.7334312, -105.200363], [39.7335257, -105.2004544], [39.7335523, -105.2004835], [39.7336132, -105.2005503], [39.7336902, -105.2006555], [39.733767, -105.2007557], [39.7338411, -105.2008491], [39.7339359, -105.2009412], [39.7340357, -105.2010154], [39.7340742, -105.201042], [39.7341117, -105.2010678], [39.7341542, -105.2010901], [39.734211, -105.2011198], [39.7343173, -105.2011712], [39.7344056, -105.2012021], [39.7345704, -105.2012376], [39.7347726, -105.2012753], [39.7351329, -105.2013311], [39.7352203, -105.2013441], [39.7353243, -105.201352], [39.7354415, -105.2013441], [39.7355491, -105.2013268], [39.7357022, -105.2012901], [39.7358395, -105.201259], [39.7362429, -105.2011675], [39.7367157, -105.2010639], [39.7367857, -105.2010539], [39.7368531, -105.2010442], [39.7369645, -105.2010356], [39.7370755, -105.201044], [39.7371895, -105.2010632], [39.7373179, -105.201099], [39.737714, -105.2012295], [39.7379567, -105.2013009], [39.7384451, -105.2014404], [39.7386872, -105.2015167], [39.7395717, -105.2017905], [39.7397153, -105.2018369], [39.7398269, -105.2018673], [39.7399356, -105.2018843], [39.7400618, -105.2018961], [39.740184, -105.2018932], [39.7403133, -105.2018823], [39.7404288, -105.2018645], [39.7405518, -105.2018387], [39.740665, -105.2018073], [39.7407597, -105.2017713], [39.7408728, -105.2017196], [39.740992, -105.2016551], [39.7411268, -105.2015625], [39.7412599, -105.2014489], [39.7413718, -105.2013453], [39.7414758, -105.2012326], [39.741608, -105.2010691], [39.7416901, -105.200964], [39.7417903, -105.2008396], [39.7417994, -105.200826], [39.7418968, -105.200686], [39.7418878, -105.200623], [39.7418917, -105.2005592], [39.7419081, -105.2004988], [39.7419393, -105.2004416], [39.7419818, -105.2003982], [39.7420319, -105.2003723], [39.7420864, -105.2003661], [39.7421399, -105.2003805], [39.742188, -105.2004144], [39.7422255, -105.2004628], [39.742251, -105.2005229], [39.7422628, -105.2005898], [39.7422597, -105.2006584], [39.7422421, -105.2007231], [39.7422712, -105.2008198], [39.7423227, -105.200901], [39.7428069, -105.2016549], [39.7429048, -105.2018114], [39.7429131, -105.2018245], [39.7431503, -105.2022035], [39.7434374, -105.2026477], [39.7435807, -105.2028659], [39.7437705, -105.2031843], [39.7441255, -105.2037193], [39.7444536, -105.2042289], [39.7445369, -105.204369], [39.7448087, -105.2048085], [39.7450389, -105.205197], [39.745164, -105.2054397], [39.7452617, -105.205642], [39.745359, -105.2058753], [39.7454413, -105.2060693], [39.7454698, -105.2061365], [39.7455586, -105.2063498], [39.7456394, -105.2066197], [39.7457343, -105.2068984], [39.7458096, -105.2071619], [39.7458432, -105.2072893], [39.7459057, -105.2075162], [39.7459547, -105.2077463], [39.7460104, -105.2080066], [39.7460312, -105.2081149], [39.7460508, -105.2082709], [39.7460836, -105.2084787], [39.7461092, -105.2086125], [39.7461165, -105.2087078], [39.7461128, -105.2088019], [39.7461078, -105.2089177], [39.7461163, -105.208995], [39.7461377, -105.2090672], [39.7461588, -105.2091147], [39.7462509, -105.2092159], [39.746281, -105.2092382], [39.7462978, -105.209256], [39.7463242, -105.209292], [39.7463458, -105.2093473], [39.746361, -105.2094251], [39.7463921, -105.2095768], [39.7464353, -105.2097133], [39.7465001, -105.2098285], [39.7467097, -105.21004], [39.7467814, -105.2101094], [39.7473357, -105.2106047], [39.7474554, -105.2107116], [39.7475615, -105.2108064], [39.7475958, -105.2108371], [39.747653, -105.2108907], [39.7481502, -105.2113567], [39.7483361, -105.2115309], [39.7483726, -105.2115651], [39.7484183, -105.2116079], [39.7484718, -105.2116584], [39.7487212, -105.2118951], [39.749238, -105.2123721], [39.7492814, -105.2124135], [39.7500059, -105.2130759], [39.7500454, -105.2131129], [39.7503812, -105.2134256], [39.7504627, -105.2135015], [39.7508219, -105.2138362], [39.7508792, -105.2138895], [39.7509348, -105.2139426], [39.7511093, -105.2141093], [39.7516271, -105.2146038], [39.7516716, -105.2146463], [39.751739, -105.2147085], [39.7524829, -105.2153953], [39.7525012, -105.2154123], [39.7532517, -105.2161101], [39.7533052, -105.2161598], [39.7540716, -105.2168674], [39.7541308, -105.2169221], [39.7544795, -105.2172508], [39.7546281, -105.2174522], [39.7546745, -105.2175037], [39.7548394, -105.2176869], [39.7549112, -105.2177871], [39.7548521, -105.2179433], [39.7546476, -105.2183047], [39.7544379, -105.2186805], [39.7544293, -105.218688], [39.7543723, -105.2187394], [39.7544376, -105.2188002], [39.7544872, -105.2188464], [39.7547934, -105.2191313], [39.7551071, -105.2194077], [39.7551368, -105.2194339], [39.755202, -105.2194913], [39.7552559, -105.2195411], [39.755615, -105.2198733], [39.755696, -105.2199479], [39.7559351, -105.2201693], [39.7559559, -105.220191], [39.7560109, -105.2202486], [39.7560703, -105.2203182], [39.756257, -105.2205368], [39.7562813, -105.2205598], [39.7564057, -105.2206774], [39.7565461, -105.2208058], [39.7567417, -105.2209442], [39.7566986, -105.2211001], [39.756638, -105.2213112], [39.7565732, -105.2214643], [39.7565209, -105.2215678], [39.7564595, -105.2216814], [39.7562914, -105.2219923], [39.7562254, -105.2221143], [39.7563006, -105.2221853], [39.7566193, -105.222483], [39.7566315, -105.2224957], [39.7569475, -105.2227913], [39.7570561, -105.222895], [39.7574792, -105.2232693], [39.7575579, -105.2233409], [39.7576236, -105.2234019], [39.7578565, -105.2236199], [39.7579703, -105.2237258], [39.7580027, -105.2237561], [39.7583001, -105.2240544], [39.7583591, -105.2241084], [39.7587249, -105.2244446], [39.7588057, -105.22452], [39.7591609, -105.2248516], [39.7595089, -105.2251741], [39.7596887, -105.2253408], [39.7597554, -105.2254024], [39.7598802, -105.2255173], [39.7598939, -105.2255305], [39.7599658, -105.2255988], [39.7600186, -105.2256498], [39.7604008, -105.2259941], [39.7605066, -105.2260931], [39.7605924, -105.226174], [39.760667, -105.2262427], [39.7607946, -105.2263631], [39.7608753, -105.2264373], [39.7615829, -105.2271044], [39.7616266, -105.2271481], [39.7616788, -105.2272008], [39.762203, -105.2276943], [39.7623411, -105.2278299], [39.7630405, -105.2284565], [39.7632824, -105.2286685], [39.7637358, -105.2290679], [39.7640137, -105.229309], [39.7640954, -105.229381], [39.7641743, -105.229462], [39.76436, -105.2296495], [39.7645376, -105.2298658], [39.764824, -105.2301918], [39.7652051, -105.2307044], [39.7654711, -105.2310802], [39.7654801, -105.2310934], [39.7655308, -105.231168], [39.7655792, -105.231232], [39.7655918, -105.2312486], [39.7657719, -105.2315035], [39.7658973, -105.2316751], [39.7659962, -105.2318039], [39.7661018, -105.231924], [39.7662469, -105.2320614], [39.7664515, -105.2322588], [39.7671442, -105.2329283], [39.7673559, -105.2331244], [39.7674414, -105.2332053], [39.7675186, -105.2332845], [39.7675756, -105.2333522], [39.7676291, -105.2334243], [39.7676969, -105.2335369], [39.7677304, -105.233602], [39.767761, -105.2336649], [39.7677908, -105.233739], [39.7678172, -105.233812], [39.7678551, -105.2339422], [39.7678736, -105.2340294], [39.7678904, -105.2341257], [39.7679032, -105.2342329], [39.7679095, -105.2343359], [39.7679068, -105.2344027], [39.7679029, -105.234499], [39.7678875, -105.2346626], [39.7678651, -105.2347791], [39.7678384, -105.2348898], [39.7678006, -105.2350283], [39.7677517, -105.2352031], [39.7677119, -105.2353468]],\n", " {"bubblingMouseEvents": true, "color": "#0c0786", "dashArray": null, "dashOffset": null, "fill": false, "fillColor": "#0c0786", "fillOpacity": 0.2, "fillRule": "evenodd", "lineCap": "round", "lineJoin": "round", "noClip": false, "opacity": 0.8, "smoothFactor": 1.0, "stroke": true, "weight": 10}\n", - " ).addTo(map_f5e9fdbd4de71b8ef0064aeb6f947a41);\n", + " ).addTo(map_ad8e071bae664cc7b1a973d4509a33a8);\n", " \n", " \n", - " poly_line_63ad4d4e5561017b7e7fd40a459bf578.bindTooltip(\n", + " poly_line_453a4ea787ffe3c7d6510177b14e1d10.bindTooltip(\n", " `<div>\n", - " 0.18722374836334246\n", + " 0.1871862788506351\n", " </div>`,\n", " {"sticky": true}\n", " );\n", " \n", " \n", - " var marker_57adc3ad6e7d8c7bd278d30d1df431de = L.marker(\n", + " var marker_e0b5ef061c2c427ae01d059e6d72c777 = L.marker(\n", " [39.7273069, -105.1994169],\n", " {}\n", - " ).addTo(map_f5e9fdbd4de71b8ef0064aeb6f947a41);\n", + " ).addTo(map_ad8e071bae664cc7b1a973d4509a33a8);\n", " \n", " \n", - " var icon_917c456f73b69b0b53fa8d8048d5f344 = L.AwesomeMarkers.icon(\n", + " var icon_5b733c3811ecc918d30b6cdbb7c63b27 = L.AwesomeMarkers.icon(\n", " {"extraClasses": "fa-rotate-0", "icon": "circle", "iconColor": "white", "markerColor": "green", "prefix": "fa"}\n", " );\n", - " marker_57adc3ad6e7d8c7bd278d30d1df431de.setIcon(icon_917c456f73b69b0b53fa8d8048d5f344);\n", + " marker_e0b5ef061c2c427ae01d059e6d72c777.setIcon(icon_5b733c3811ecc918d30b6cdbb7c63b27);\n", " \n", " \n", - " marker_57adc3ad6e7d8c7bd278d30d1df431de.bindTooltip(\n", + " marker_e0b5ef061c2c427ae01d059e6d72c777.bindTooltip(\n", " `<div>\n", " Origin\n", " </div>`,\n", @@ -775,19 +764,19 @@ " );\n", " \n", " \n", - " var marker_9f05f804eab779baf28827016a73a4e7 = L.marker(\n", + " var marker_ea7244601f3a5c7f675da2105a2d5354 = L.marker(\n", " [39.7677119, -105.2353468],\n", " {}\n", - " ).addTo(map_f5e9fdbd4de71b8ef0064aeb6f947a41);\n", + " ).addTo(map_ad8e071bae664cc7b1a973d4509a33a8);\n", " \n", " \n", - " var icon_9c0d129fcb8005579733c0f5eae70f74 = L.AwesomeMarkers.icon(\n", + " var icon_2266673043479453c0c2882dd6b522ed = L.AwesomeMarkers.icon(\n", " {"extraClasses": "fa-rotate-0", "icon": "circle", "iconColor": "white", "markerColor": "red", "prefix": "fa"}\n", " );\n", - " marker_9f05f804eab779baf28827016a73a4e7.setIcon(icon_9c0d129fcb8005579733c0f5eae70f74);\n", + " marker_ea7244601f3a5c7f675da2105a2d5354.setIcon(icon_2266673043479453c0c2882dd6b522ed);\n", " \n", " \n", - " marker_9f05f804eab779baf28827016a73a4e7.bindTooltip(\n", + " marker_ea7244601f3a5c7f675da2105a2d5354.bindTooltip(\n", " `<div>\n", " Destination\n", " </div>`,\n", @@ -798,10 +787,10 @@ "</html>\" style=\"position:absolute;width:100%;height:100%;left:0;top:0;border:none !important;\" allowfullscreen webkitallowfullscreen mozallowfullscreen></iframe></div></div>" ], "text/plain": [ - "<folium.folium.Map at 0x124205e50>" + "<folium.folium.Map at 0x1231d71d0>" ] }, - "execution_count": 16, + "execution_count": 15, "metadata": {}, "output_type": "execute_result" } diff --git a/python/nrel/routee/compass/io/generate_dataset.py b/python/nrel/routee/compass/io/generate_dataset.py index f3047028..cc58bfac 100644 --- a/python/nrel/routee/compass/io/generate_dataset.py +++ b/python/nrel/routee/compass/io/generate_dataset.py @@ -179,7 +179,7 @@ def replace_id(vertex_uuid): if filename == "osm_default_energy.toml": if add_grade: init_toml["traversal"][ - "grade_table_path" + "grade_table_file" ] = "edges-grade-enumerated.txt.gz" init_toml["traversal"]["grade_table_grade_unit"] = "decimal" @@ -196,7 +196,7 @@ def replace_id(vertex_uuid): init_toml["traversal"][ "real_world_energy_adjustment" ] = adjustment_factor - init_toml["traversal"]["energy_model_path"] = f"{energy_model}.bin" + init_toml["traversal"]["energy_model_file"] = f"{energy_model}.bin" energy_model_energy_rate_unit = ENERGY_OUTPUT_UNITS.get( powertrain_type diff --git a/python/nrel/routee/compass/resources/osm_default_distance.toml b/python/nrel/routee/compass/resources/osm_default_distance.toml index ee84bb3f..c9d6d012 100644 --- a/python/nrel/routee/compass/resources/osm_default_distance.toml +++ b/python/nrel/routee/compass/resources/osm_default_distance.toml @@ -1,8 +1,8 @@ parallelism = 2 [graph] -edge_list_csv = "edges-compass.csv.gz" -vertex_list_csv = "vertices-compass.csv.gz" +edge_list_file = "edges-compass.csv.gz" +vertex_list_file = "vertices-compass.csv.gz" verbose = true [traversal] diff --git a/python/nrel/routee/compass/resources/osm_default_energy.toml b/python/nrel/routee/compass/resources/osm_default_energy.toml index ba67777b..47e90ca2 100644 --- a/python/nrel/routee/compass/resources/osm_default_energy.toml +++ b/python/nrel/routee/compass/resources/osm_default_energy.toml @@ -1,14 +1,14 @@ parallelism = 2 [graph] -edge_list_csv = "edges-compass.csv.gz" -vertex_list_csv = "vertices-compass.csv.gz" +edge_list_file = "edges-compass.csv.gz" +vertex_list_file = "vertices-compass.csv.gz" verbose = true [traversal] type = "speed_grade_energy_model" model_type = "smartcore" -speed_table_path = "edges-posted-speed-enumerated.txt.gz" +speed_table_file = "edges-posted-speed-enumerated.txt.gz" speed_table_speed_unit = "kilometers_per_hour" energy_model_speed_unit = "miles_per_hour" energy_model_grade_unit = "decimal" diff --git a/python/nrel/routee/compass/resources/osm_default_speed.toml b/python/nrel/routee/compass/resources/osm_default_speed.toml index 245465dd..8a5279ca 100644 --- a/python/nrel/routee/compass/resources/osm_default_speed.toml +++ b/python/nrel/routee/compass/resources/osm_default_speed.toml @@ -1,13 +1,13 @@ parallelism = 2 [graph] -edge_list_csv = "edges-compass.csv.gz" -vertex_list_csv = "vertices-compass.csv.gz" +edge_list_file = "edges-compass.csv.gz" +vertex_list_file = "vertices-compass.csv.gz" verbose = true [traversal] type = "speed_table" -speed_table_path = "edges-posted-speed-enumerated.txt.gz" +speed_table_file = "edges-posted-speed-enumerated.txt.gz" speed_unit = "kilometers_per_hour" output_distance_unit = "miles" output_time_unit = "minutes" diff --git a/rust/routee-compass-core/src/algorithm/component/scc.rs b/rust/routee-compass-core/src/algorithm/component/scc.rs index 9a1fa8c8..d657c0b8 100644 --- a/rust/routee-compass-core/src/algorithm/component/scc.rs +++ b/rust/routee-compass-core/src/algorithm/component/scc.rs @@ -148,26 +148,50 @@ fn largest_strongly_connected_component(graph: &Graph) -> Result<Vec<VertexId>, #[cfg(test)] mod tests { use super::*; - use crate::model::graph::graph_config::GraphConfig; - use std::path::PathBuf; + use crate::model::property::{edge::Edge, vertex::Vertex}; + use std::collections::HashMap; fn build_mock_graph() -> Graph { - // A test graph with 2 strongly connected components - let test_path = PathBuf::from(env!("CARGO_MANIFEST_DIR")) - .join("src") - .join("algorithm") - .join("component") - .join("test"); - let edge_list_csv = test_path.join("edges.csv").to_str().unwrap().to_string(); - let vertex_list_csv = test_path.join("vertices.csv").to_str().unwrap().to_string(); - let graph_conf = GraphConfig { - edge_list_csv, - vertex_list_csv, - n_edges: None, - n_vertices: None, - verbose: false, + let vertices = vec![ + Vertex::new(0, 0.0, 0.0), + Vertex::new(1, 1.0, 1.0), + Vertex::new(2, 2.0, 2.0), + Vertex::new(3, 3.0, 3.0), + Vertex::new(4, 4.0, 4.0), + ]; + + let edges = vec![ + Edge::new(0, 0, 1, 10.0), + Edge::new(1, 1, 0, 10.0), + Edge::new(2, 1, 2, 10.0), + Edge::new(3, 2, 1, 10.0), + Edge::new(4, 2, 3, 10.0), + Edge::new(5, 3, 2, 10.0), + Edge::new(6, 3, 0, 10.0), + Edge::new(7, 0, 3, 10.0), + Edge::new(8, 0, 2, 10.0), + Edge::new(9, 1, 3, 10.0), + Edge::new(10, 2, 0, 10.0), + Edge::new(11, 3, 1, 10.0), + Edge::new(12, 4, 4, 10.0), + ]; + + // Create the adjacency and reverse adjacency lists. + let mut adj = vec![HashMap::new(); vertices.len()]; + let mut rev = vec![HashMap::new(); vertices.len()]; + + for edge in &edges { + adj[edge.src_vertex_id.0].insert(edge.edge_id, edge.dst_vertex_id); + rev[edge.dst_vertex_id.0].insert(edge.edge_id, edge.src_vertex_id); + } + + // Construct the Graph instance. + let graph = Graph { + adj, + rev, + edges, + vertices, }; - let graph = Graph::try_from(&graph_conf).unwrap(); graph } diff --git a/rust/routee-compass-core/src/algorithm/search/a_star/a_star.rs b/rust/routee-compass-core/src/algorithm/search/a_star/a_star.rs index c57b2082..7835875c 100644 --- a/rust/routee-compass-core/src/algorithm/search/a_star/a_star.rs +++ b/rust/routee-compass-core/src/algorithm/search/a_star/a_star.rs @@ -330,13 +330,50 @@ mod tests { use crate::algorithm::search::backtrack::vertex_oriented_route; use crate::model::frontier::default::no_restriction; use crate::model::graph::graph::Graph; - use crate::model::graph::graph_config::GraphConfig; + use crate::model::property::edge::Edge; + use crate::model::property::vertex::Vertex; use crate::model::traversal::default::distance::DistanceModel; use crate::model::traversal::traversal_model::TraversalModel; use crate::util::unit::DistanceUnit; use crate::{model::graph::edge_id::EdgeId, util::read_only_lock::DriverReadOnlyLock}; use rayon::prelude::*; + fn build_mock_graph() -> Graph { + let vertices = vec![ + Vertex::new(0, 0.0, 0.0), + Vertex::new(1, 0.0, 0.0), + Vertex::new(2, 0.0, 0.0), + Vertex::new(3, 0.0, 0.0), + ]; + + let edges = vec![ + Edge::new(0, 0, 1, 10.0), + Edge::new(1, 1, 0, 10.0), + Edge::new(2, 1, 2, 2.0), + Edge::new(3, 2, 1, 2.0), + Edge::new(4, 2, 3, 1.0), + Edge::new(5, 3, 2, 1.0), + Edge::new(6, 3, 0, 2.0), + Edge::new(7, 0, 3, 2.0), + ]; + + let mut adj = vec![HashMap::new(); vertices.len()]; + let mut rev = vec![HashMap::new(); vertices.len()]; + + for edge in &edges { + adj[edge.src_vertex_id.0].insert(edge.edge_id, edge.dst_vertex_id); + rev[edge.dst_vertex_id.0].insert(edge.edge_id, edge.src_vertex_id); + } + + let graph = Graph { + adj, + rev, + edges, + vertices, + }; + graph + } + #[test] fn test_e2e_queries() { // simple box world that exists in a non-euclidean plane that stretches @@ -386,22 +423,7 @@ mod tests { // setup the graph, traversal model, and a* heuristic to be shared across the queries in parallel // these live in the "driver" process and are passed as read-only memory to each executor process - let test_path = PathBuf::from(env!("CARGO_MANIFEST_DIR")) - .join("src") - .join("algorithm") - .join("search") - .join("a_star") - .join("test"); - let edge_list_csv = test_path.join("edges.csv").to_str().unwrap().to_string(); - let vertex_list_csv = test_path.join("vertices.csv").to_str().unwrap().to_string(); - let graph_conf = GraphConfig { - edge_list_csv, - vertex_list_csv, - n_edges: None, - n_vertices: None, - verbose: false, - }; - let graph = Graph::try_from(&graph_conf).unwrap(); + let graph = build_mock_graph(); let driver_dg = Arc::new(DriverReadOnlyLock::new(graph)); let no_restriction: Box<dyn FrontierModel> = Box::new(no_restriction::NoRestriction {}); diff --git a/rust/routee-compass-core/src/model/graph/edge_loader.rs b/rust/routee-compass-core/src/model/graph/edge_loader.rs index 7f8bbc15..164818e4 100644 --- a/rust/routee-compass-core/src/model/graph/edge_loader.rs +++ b/rust/routee-compass-core/src/model/graph/edge_loader.rs @@ -1,4 +1,3 @@ -use super::graph_config::GraphConfig; use crate::{ model::{ graph::graph_error::GraphError, @@ -9,7 +8,10 @@ use crate::{ }; use kdam::Bar; use kdam::BarExt; -use std::collections::{HashMap, HashSet}; +use std::{ + collections::{HashMap, HashSet}, + path::PathBuf, +}; pub struct EdgeLoader { pub edges: Vec<Edge>, @@ -17,13 +19,13 @@ pub struct EdgeLoader { pub rev: Vec<HashMap<EdgeId, VertexId>>, } -pub struct EdgeLoaderConfig<'a> { - pub config: &'a GraphConfig, +pub struct EdgeLoaderConfig { + pub edge_list_csv: PathBuf, pub n_edges: usize, pub n_vertices: usize, } -impl<'a> TryFrom<EdgeLoaderConfig<'a>> for EdgeLoader { +impl TryFrom<EdgeLoaderConfig> for EdgeLoader { type Error = GraphError; fn try_from(c: EdgeLoaderConfig) -> Result<Self, Self::Error> { @@ -62,8 +64,7 @@ impl<'a> TryFrom<EdgeLoaderConfig<'a>> for EdgeLoader { pb.update(1); }); - let edges = - read_utils::vec_from_csv(&c.config.edge_list_csv, true, Some(c.n_edges), Some(cb))?; + let edges = read_utils::vec_from_csv(&c.edge_list_csv, true, Some(c.n_edges), Some(cb))?; print!("\n"); let result = EdgeLoader { edges, adj, rev }; diff --git a/rust/routee-compass-core/src/model/graph/graph.rs b/rust/routee-compass-core/src/model/graph/graph.rs index a4162183..75508359 100644 --- a/rust/routee-compass-core/src/model/graph/graph.rs +++ b/rust/routee-compass-core/src/model/graph/graph.rs @@ -1,14 +1,12 @@ use crate::algorithm::search::direction::Direction; -use crate::model::graph::edge_loader::EdgeLoader; -use crate::model::graph::edge_loader::EdgeLoaderConfig; -use crate::model::graph::graph_config::GraphConfig; use crate::model::graph::graph_error::GraphError; -use crate::model::graph::vertex_loader::VertexLoaderConfig; use crate::model::graph::{edge_id::EdgeId, vertex_id::VertexId}; use crate::model::property::edge::Edge; use crate::model::property::vertex::Vertex; -use log::info; use std::collections::HashMap; +use std::path::Path; + +use super::graph_loader::graph_from_files; /// Road network topology represented as an adjacency list. /// The `EdgeId` and `VertexId` values correspond to edge and @@ -28,6 +26,8 @@ use std::collections::HashMap; /// collections will prefer chained iterators. A few will collect /// into Vecs because of error handling or lifetimes, but those cases will only produce a /// smaller subset of the source data. + +#[derive(Debug)] pub struct Graph { pub adj: Vec<HashMap<EdgeId, VertexId>>, pub rev: Vec<HashMap<EdgeId, VertexId>>, @@ -36,6 +36,32 @@ pub struct Graph { } impl Graph { + /// Build a `Graph` from a pair of CSV files. + /// You can also pass in the number of edges and vertices to avoid + /// scanning the input files and potentially building vectors that + /// have more memory than needed. + /// + /// # Arguments + /// + /// * `edge_list_csv` - path to the CSV file containing edge attributes + /// * `vertex_list_csv` - path to the CSV file containing vertex attributes + /// * `n_edges` - number of edges in the graph + /// * `n_vertices` - number of vertices in the graph + /// * `verbose` - whether to print progress information to the console + /// + /// # Returns + /// + /// A graph instance, or an error if an IO error occurred. + /// + pub fn from_files<P: AsRef<Path>>( + edge_list_csv: &P, + vertex_list_csv: &P, + n_edges: Option<usize>, + n_vertices: Option<usize>, + verbose: Option<bool>, + ) -> Result<Graph, GraphError> { + graph_from_files(edge_list_csv, vertex_list_csv, n_edges, n_vertices, verbose) + } /// number of edges in the Graph pub fn n_edges(&self) -> usize { self.edges.len() @@ -285,56 +311,3 @@ impl Graph { .collect() } } - -impl TryFrom<&GraphConfig> for Graph { - type Error = GraphError; - - /// tries to build a Graph from a GraphConfig. - /// - /// for both edge and vertex lists, we assume all ids can be used as indices - /// to an array data structure. to find the size of each array, we pass once - /// through each file to count the number of rows (minus header) of the CSV. - /// then we can build a Vec *once* and insert rows as we decode them without - /// a sort. - /// - /// # Arguments - /// - /// * `config` - configuration object for building a `Graph` instance - /// - /// # Returns - /// - /// A graph instance, or an error if an IO error occurred. - fn try_from(config: &GraphConfig) -> Result<Self, Self::Error> { - info!("checking file length of edge and vertex input files"); - let (n_edges, n_vertices) = config.read_file_sizes()?; - info!( - "creating data structures to hold {} edges, {} vertices", - n_edges, n_vertices - ); - - info!("reading edge list"); - - let e_conf = EdgeLoaderConfig { - config: &config, - n_edges, - n_vertices, - }; - let e_result = EdgeLoader::try_from(e_conf)?; - - info!("reading vertex list"); - let v_conf = VertexLoaderConfig { - config: &config, - n_vertices, - }; - let vertices: Vec<Vertex> = v_conf.try_into()?; - - let graph = Graph { - adj: e_result.adj, - rev: e_result.rev, - edges: e_result.edges, - vertices, - }; - - Ok(graph) - } -} diff --git a/rust/routee-compass-core/src/model/graph/graph_config.rs b/rust/routee-compass-core/src/model/graph/graph_config.rs deleted file mode 100644 index 559fd25c..00000000 --- a/rust/routee-compass-core/src/model/graph/graph_config.rs +++ /dev/null @@ -1,59 +0,0 @@ -use crate::model::graph::graph_error::GraphError; -use crate::util::fs::fs_utils::line_count; -use log::warn; -use serde::{Deserialize, Serialize}; - -#[derive(Debug, Deserialize, Serialize)] -pub struct GraphConfig { - pub edge_list_csv: String, - pub vertex_list_csv: String, - pub n_edges: Option<usize>, - pub n_vertices: Option<usize>, - pub verbose: bool, -} - -impl GraphConfig { - pub fn read_file_sizes(&self) -> Result<(usize, usize), GraphError> { - let n_edges = self - .get_n_edges() - .map_err(|e| GraphError::IOError { source: e })?; - let n_vertices = self - .get_n_vertices() - .map_err(|e| GraphError::IOError { source: e })?; - if n_edges < 1 { - return Err(GraphError::EmptyFileSource { - filename: self.edge_list_csv.clone(), - }); - } - if n_vertices < 1 { - return Err(GraphError::EmptyFileSource { - filename: self.vertex_list_csv.clone(), - }); - } - Ok((n_edges, n_vertices)) - } - - pub fn get_n_edges(&self) -> std::io::Result<usize> { - match self.n_edges { - Some(n) => Ok(n), - None => { - warn!("edge list size not provided, scanning input to determine size"); - let is_gzip = self.edge_list_csv.ends_with(".gz"); - let n = line_count(self.edge_list_csv.clone(), is_gzip)?; - Ok(n - 1) // drop count of header line - } - } - } - - pub fn get_n_vertices(&self) -> std::io::Result<usize> { - match self.n_vertices { - Some(n) => Ok(n), - None => { - warn!("vertex list size not provided, scanning input to determine size"); - let is_gzip = self.vertex_list_csv.ends_with(".gz"); - let n = line_count(self.vertex_list_csv.clone(), is_gzip)?; - Ok(n - 1) // drop count of header line - } - } - } -} diff --git a/rust/routee-compass-core/src/model/graph/graph_error.rs b/rust/routee-compass-core/src/model/graph/graph_error.rs index fa161884..70652ab4 100644 --- a/rust/routee-compass-core/src/model/graph/graph_error.rs +++ b/rust/routee-compass-core/src/model/graph/graph_error.rs @@ -1,3 +1,5 @@ +use std::path::PathBuf; + use crate::model::graph::{edge_id::EdgeId, vertex_id::VertexId}; #[derive(thiserror::Error, Debug)] @@ -17,7 +19,7 @@ pub enum GraphError { #[error("error in test setup")] TestError, #[error("{filename} file source was empty")] - EmptyFileSource { filename: String }, + EmptyFileSource { filename: PathBuf }, #[error("failure reading TomTom graph: {source}")] IOError { #[from] diff --git a/rust/routee-compass-core/src/model/graph/graph_loader.rs b/rust/routee-compass-core/src/model/graph/graph_loader.rs new file mode 100644 index 00000000..5bd0fa95 --- /dev/null +++ b/rust/routee-compass-core/src/model/graph/graph_loader.rs @@ -0,0 +1,97 @@ +use std::path::Path; + +use log::warn; + +use crate::{model::property::vertex::Vertex, util::fs::fs_utils::line_count}; + +use super::{ + edge_loader::{EdgeLoader, EdgeLoaderConfig}, + graph::Graph, + graph_error::GraphError, + vertex_loader::VertexLoaderConfig, +}; + +pub fn graph_from_files<P: AsRef<Path>>( + edge_list_csv: &P, + vertex_list_csv: &P, + n_edges: Option<usize>, + n_vertices: Option<usize>, + verbose: Option<bool>, +) -> Result<Graph, GraphError> { + let verbose = verbose.unwrap_or(false); + let n_edges = match n_edges { + Some(n) => n, + None => { + if verbose { + warn!("edge list size not provided, scanning input to determine size"); + } + get_n_edges(&edge_list_csv)? + } + }; + + let n_vertices = match n_vertices { + Some(n) => n, + None => { + if verbose { + warn!("vertex list size not provided, scanning input to determine size"); + } + get_n_vertices(&vertex_list_csv)? + } + }; + let e_conf = EdgeLoaderConfig { + edge_list_csv: edge_list_csv.as_ref().to_path_buf(), + n_edges, + n_vertices, + }; + + let e_result = EdgeLoader::try_from(e_conf)?; + + let v_conf = VertexLoaderConfig { + vertex_list_csv: vertex_list_csv.as_ref().to_path_buf(), + n_vertices, + }; + + let vertices: Vec<Vertex> = v_conf.try_into()?; + + let graph = Graph { + adj: e_result.adj, + rev: e_result.rev, + edges: e_result.edges, + vertices, + }; + + Ok(graph) +} + +fn get_n_edges<P: AsRef<Path>>(edge_list_csv: &P) -> Result<usize, GraphError> { + // check if the extension is .gz + let is_gzip = edge_list_csv + .as_ref() + .to_path_buf() + .extension() + .map(|ext| ext.to_str() == Some("gz")) + .unwrap_or(false); + let n = line_count(edge_list_csv.clone(), is_gzip)?; + if n < 1 { + return Err(GraphError::EmptyFileSource { + filename: edge_list_csv.as_ref().to_path_buf(), + }); + } + Ok(n - 1) // drop count of header line +} + +fn get_n_vertices<P: AsRef<Path>>(vertex_list_csv: &P) -> Result<usize, GraphError> { + let is_gzip = vertex_list_csv + .as_ref() + .to_path_buf() + .extension() + .map(|ext| ext.to_str() == Some("gz")) + .unwrap_or(false); + let n = line_count(vertex_list_csv.clone(), is_gzip)?; + if n < 1 { + return Err(GraphError::EmptyFileSource { + filename: vertex_list_csv.as_ref().to_path_buf(), + }); + } + Ok(n - 1) // drop count of header line +} diff --git a/rust/routee-compass-core/src/model/graph/mod.rs b/rust/routee-compass-core/src/model/graph/mod.rs index 50fd24da..2e41655a 100644 --- a/rust/routee-compass-core/src/model/graph/mod.rs +++ b/rust/routee-compass-core/src/model/graph/mod.rs @@ -1,7 +1,7 @@ pub mod edge_id; pub mod edge_loader; pub mod graph; -pub mod graph_config; pub mod graph_error; +pub mod graph_loader; pub mod vertex_id; pub mod vertex_loader; diff --git a/rust/routee-compass-core/src/model/graph/vertex_loader.rs b/rust/routee-compass-core/src/model/graph/vertex_loader.rs index a181743d..83734cee 100644 --- a/rust/routee-compass-core/src/model/graph/vertex_loader.rs +++ b/rust/routee-compass-core/src/model/graph/vertex_loader.rs @@ -1,18 +1,19 @@ -use crate::model::graph::graph_config::GraphConfig; +use std::path::PathBuf; + use crate::model::graph::graph_error::GraphError; use crate::model::property::vertex::Vertex; use crate::util::fs::read_utils; use kdam::{Bar, BarExt}; -pub struct VertexLoaderConfig<'a> { - pub config: &'a GraphConfig, +pub struct VertexLoaderConfig { + pub vertex_list_csv: PathBuf, pub n_vertices: usize, } -impl<'a> TryFrom<VertexLoaderConfig<'a>> for Vec<Vertex> { +impl TryFrom<VertexLoaderConfig> for Vec<Vertex> { type Error = GraphError; - fn try_from(conf: VertexLoaderConfig<'a>) -> Result<Self, Self::Error> { + fn try_from(conf: VertexLoaderConfig) -> Result<Self, Self::Error> { let mut processed: usize = 0; let mut pb = Bar::builder() .total(conf.n_vertices) @@ -25,12 +26,8 @@ impl<'a> TryFrom<VertexLoaderConfig<'a>> for Vec<Vertex> { pb.update(1); processed = processed + 1; }); - let result: Vec<Vertex> = read_utils::vec_from_csv( - &conf.config.vertex_list_csv, - true, - Some(conf.n_vertices), - Some(cb), - )?; + let result: Vec<Vertex> = + read_utils::vec_from_csv(&conf.vertex_list_csv, true, Some(conf.n_vertices), Some(cb))?; print!("\n"); Ok(result) diff --git a/rust/routee-compass-core/src/model/property/edge.rs b/rust/routee-compass-core/src/model/property/edge.rs index d02a78b1..cbc5a2a4 100644 --- a/rust/routee-compass-core/src/model/property/edge.rs +++ b/rust/routee-compass-core/src/model/property/edge.rs @@ -13,6 +13,17 @@ pub struct Edge { pub distance: Distance, } +impl Edge { + pub fn new(edge_id: usize, src_vertex_id: usize, dst_vertex_id: usize, distance: f64) -> Self { + Self { + edge_id: EdgeId(edge_id), + src_vertex_id: VertexId(src_vertex_id), + dst_vertex_id: VertexId(dst_vertex_id), + distance: Distance::new(distance), + } + } +} + impl Default for Edge { fn default() -> Self { Edge { diff --git a/rust/routee-compass-core/src/model/traversal/default/speed_lookup_model.rs b/rust/routee-compass-core/src/model/traversal/default/speed_lookup_model.rs index 03eb0bc7..6b547753 100644 --- a/rust/routee-compass-core/src/model/traversal/default/speed_lookup_model.rs +++ b/rust/routee-compass-core/src/model/traversal/default/speed_lookup_model.rs @@ -1,7 +1,9 @@ +use std::path::{Path, PathBuf}; + use crate::model::graph::edge_id::EdgeId; use crate::util::fs::read_decoders; use crate::util::geo::haversine; -use crate::util::unit::{Distance, DistanceUnit, Grade, BASE_SPEED_UNIT}; +use crate::util::unit::{Distance, DistanceUnit}; use crate::util::unit::{SpeedUnit, Time, TimeUnit, BASE_DISTANCE_UNIT, BASE_TIME_UNIT}; use crate::{ model::{ @@ -26,15 +28,20 @@ pub struct SpeedLookupModel { } impl SpeedLookupModel { - pub fn new( - speed_table_path: &String, + pub fn new<P: AsRef<Path>>( + speed_table_path: &P, speed_unit: SpeedUnit, output_distance_unit_opt: Option<DistanceUnit>, output_time_unit_opt: Option<TimeUnit>, ) -> Result<SpeedLookupModel, TraversalModelError> { let speed_table: Vec<Speed> = - read_utils::read_raw_file(speed_table_path, read_decoders::default, None).map_err( - |e| TraversalModelError::FileReadError(speed_table_path.clone(), e.to_string()), + read_utils::read_raw_file(&speed_table_path, read_decoders::default, None).map_err( + |e| { + TraversalModelError::FileReadError( + speed_table_path.as_ref().to_path_buf(), + e.to_string(), + ) + }, )?; let max_speed = get_max_speed(&speed_table)?; let output_time_unit = output_time_unit_opt.unwrap_or(BASE_TIME_UNIT); @@ -200,7 +207,7 @@ mod tests { distance: Distance::new(100.0), }; } - fn filepath() -> String { + fn filepath() -> PathBuf { let filepath = PathBuf::from(env!("CARGO_MANIFEST_DIR")) .join("src") .join("model") @@ -208,8 +215,7 @@ mod tests { .join("default") .join("test") .join("velocities.txt"); - let filename = filepath.to_str().unwrap().to_owned(); - return filename; + filepath } fn approx_eq(a: f64, b: f64, error: f64) { diff --git a/rust/routee-compass-core/src/model/traversal/traversal_model_error.rs b/rust/routee-compass-core/src/model/traversal/traversal_model_error.rs index 15ff4e9d..4f3d043e 100644 --- a/rust/routee-compass-core/src/model/traversal/traversal_model_error.rs +++ b/rust/routee-compass-core/src/model/traversal/traversal_model_error.rs @@ -1,10 +1,12 @@ +use std::path::PathBuf; + use super::state::traversal_state::TraversalState; use crate::util::unit::UnitError; #[derive(thiserror::Error, Debug)] pub enum TraversalModelError { #[error("failure building traversal model from file {0}: {1}")] - FileReadError(String, String), + FileReadError(PathBuf, String), #[error("failure building traversal model: {0}")] BuildError(String), #[error("numeric error during calculation: {0}")] diff --git a/rust/routee-compass-powertrain/README.md b/rust/routee-compass-powertrain/README.md index 6c6ccac2..5f87f34f 100644 --- a/rust/routee-compass-powertrain/README.md +++ b/rust/routee-compass-powertrain/README.md @@ -27,8 +27,8 @@ An example traversal model configuration that uses this crate may look like this [traversal] type = "speed_grade_energy_model" model_type = "smartcore" -speed_table_path = "edges-posted-speed-enumerated.txt.gz" -energy_model_path = "2016_TOYOTA_Camry_4cyl_2WD.bin" +speed_table_file = "edges-posted-speed-enumerated.txt.gz" +energy_model_file = "2016_TOYOTA_Camry_4cyl_2WD.bin" ideal_energy_rate = 0.02857142857 speed_table_speed_unit = "kilometers_per_hour" energy_model_speed_unit = "miles_per_hour" @@ -36,7 +36,7 @@ energy_model_grade_unit = "decimal" energy_model_energy_rate_unit = "gallons_gasoline_per_mile" output_time_unit = "minutes" output_distance_unit = "miles" -grade_table_path = "edges-grade-enumerated.txt.gz" +grade_table_file = "edges-grade-enumerated.txt.gz" grade_table_grade_unit = "decimal" ``` diff --git a/rust/routee-compass-powertrain/src/routee/model_type.rs b/rust/routee-compass-powertrain/src/routee/model_type.rs index 31da2782..c2d69236 100644 --- a/rust/routee-compass-powertrain/src/routee/model_type.rs +++ b/rust/routee-compass-powertrain/src/routee/model_type.rs @@ -1,4 +1,7 @@ -use std::sync::Arc; +use std::{ + path::{Path, PathBuf}, + sync::Arc, +}; use routee_compass_core::{ model::traversal::traversal_model_error::TraversalModelError, @@ -30,9 +33,9 @@ impl std::fmt::Display for ModelType { impl ModelType { /// builds a speed grade energy prediction model - pub fn build( + pub fn build<P: AsRef<Path>>( &self, - energy_model_path: String, + energy_model_path: &P, energy_model_speed_unit: SpeedUnit, energy_model_grade_unit: GradeUnit, energy_model_energy_rate_unit: EnergyRateUnit, diff --git a/rust/routee-compass-powertrain/src/routee/smartcore/smartcore_speed_grade_model.rs b/rust/routee-compass-powertrain/src/routee/smartcore/smartcore_speed_grade_model.rs index 42e6a8b1..58a0dd2d 100644 --- a/rust/routee-compass-powertrain/src/routee/smartcore/smartcore_speed_grade_model.rs +++ b/rust/routee-compass-powertrain/src/routee/smartcore/smartcore_speed_grade_model.rs @@ -1,3 +1,5 @@ +use std::path::Path; + use crate::routee::prediction_model::SpeedGradePredictionModel; use routee_compass_core::{ model::traversal::traversal_model_error::TraversalModelError, @@ -36,19 +38,25 @@ impl SpeedGradePredictionModel for SmartcoreSpeedGradeModel { } impl SmartcoreSpeedGradeModel { - pub fn new( - routee_model_path: String, + pub fn new<P: AsRef<Path>>( + routee_model_path: &P, speed_unit: SpeedUnit, grade_unit: GradeUnit, energy_rate_unit: EnergyRateUnit, ) -> Result<Self, TraversalModelError> { // Load random forest binary file let rf_binary = std::fs::read(routee_model_path.clone()).map_err(|e| { - TraversalModelError::FileReadError(routee_model_path.clone(), e.to_string()) + TraversalModelError::FileReadError( + routee_model_path.as_ref().to_path_buf(), + e.to_string(), + ) })?; let rf: RandomForestRegressor<f64, f64, DenseMatrix<f64>, Vec<f64>> = bincode::deserialize(&rf_binary).map_err(|e| { - TraversalModelError::FileReadError(routee_model_path.clone(), e.to_string()) + TraversalModelError::FileReadError( + routee_model_path.as_ref().to_path_buf(), + e.to_string(), + ) })?; Ok(SmartcoreSpeedGradeModel { rf, diff --git a/rust/routee-compass-powertrain/src/routee/speed_grade_model.rs b/rust/routee-compass-powertrain/src/routee/speed_grade_model.rs index 2870e8ef..8930c5cb 100644 --- a/rust/routee-compass-powertrain/src/routee/speed_grade_model.rs +++ b/rust/routee-compass-powertrain/src/routee/speed_grade_model.rs @@ -255,9 +255,6 @@ mod tests { .join("routee") .join("test") .join("Toyota_Camry.bin"); - let speed_file_name = speed_file_path.to_str().unwrap(); - let grade_file_name = grade_file_path.to_str().unwrap(); - let model_file_name = model_file_path.to_str().unwrap(); let v = Vertex { vertex_id: VertexId(0), coordinate: coord! {x: -86.67, y: 36.12}, @@ -270,15 +267,12 @@ mod tests { distance: Distance::new(100.0), }; } - let speed_file = String::from(speed_file_name); - let grade_file = String::from(grade_file_name); - let routee_model_path = String::from(model_file_name); let service = SpeedGradeModelService::new( - speed_file, + &speed_file_path, SpeedUnit::KilometersPerHour, + &Some(grade_file_path), None, - None, - routee_model_path, + &model_file_path, ModelType::Smartcore, None, SpeedUnit::MilesPerHour, diff --git a/rust/routee-compass-powertrain/src/routee/speed_grade_model_service.rs b/rust/routee-compass-powertrain/src/routee/speed_grade_model_service.rs index bfa890ef..266dc51d 100644 --- a/rust/routee-compass-powertrain/src/routee/speed_grade_model_service.rs +++ b/rust/routee-compass-powertrain/src/routee/speed_grade_model_service.rs @@ -7,6 +7,8 @@ use routee_compass_core::model::traversal::traversal_model_error::TraversalModel use routee_compass_core::util::fs::read_decoders; use routee_compass_core::util::fs::read_utils; use routee_compass_core::util::unit::*; +use std::path::Path; +use std::path::PathBuf; use std::sync::Arc; #[cfg(feature = "onnx")] @@ -30,12 +32,12 @@ pub struct SpeedGradeModelService { } impl SpeedGradeModelService { - pub fn new( - speed_table_path: String, + pub fn new<P: AsRef<Path>>( + speed_table_path: &P, speeds_table_speed_unit: SpeedUnit, - grade_table_path_option: Option<String>, + grade_table_path_option: &Option, grade_table_grade_unit_option: Option<GradeUnit>, - energy_model_path: String, + energy_model_path: &P, model_type: ModelType, ideal_energy_rate_option: Option<EnergyRate>, energy_model_speed_unit: SpeedUnit, @@ -63,14 +65,22 @@ impl SpeedGradeModelService { // load speeds table let speed_table: Arc<Vec<Speed>> = Arc::new( read_utils::read_raw_file(&speed_table_path, read_decoders::default, None).map_err( - |e| TraversalModelError::FileReadError(speed_table_path.clone(), e.to_string()), + |e| { + TraversalModelError::FileReadError( + speed_table_path.as_ref().to_path_buf(), + e.to_string(), + ) + }, )?, ); let grade_table: Arc<Option<Vec<Grade>>> = match grade_table_path_option { Some(gtp) => Arc::new(Some( read_utils::read_raw_file(&gtp, read_decoders::default, None).map_err(|e| { - TraversalModelError::FileReadError(speed_table_path.clone(), e.to_string()) + TraversalModelError::FileReadError( + speed_table_path.as_ref().to_path_buf(), + e.to_string(), + ) })?, )), None => Arc::new(None), diff --git a/rust/routee-compass-py/src/app_wrapper.rs b/rust/routee-compass-py/src/app_wrapper.rs index 5d4a1663..c365c44b 100644 --- a/rust/routee-compass-py/src/app_wrapper.rs +++ b/rust/routee-compass-py/src/app_wrapper.rs @@ -1,4 +1,4 @@ -use std::path::PathBuf; +use std::path::Path; use pyo3::{exceptions::PyException, prelude::*, types::PyType}; use routee_compass::app::compass::compass_app::CompassApp; @@ -12,7 +12,7 @@ pub struct CompassAppWrapper { impl CompassAppWrapper { #[classmethod] pub fn _from_config_file(_cls: &PyType, config_file: String) -> PyResult<Self> { - let config_path = PathBuf::from(config_file.clone()); + let config_path = Path::new(&config_file); let routee_compass = CompassApp::try_from(config_path).map_err(|e| { PyException::new_err(format!( "Could not create CompassApp from config file {}: {}", diff --git a/rust/routee-compass/src/app/compass/compass_app.rs b/rust/routee-compass/src/app/compass/compass_app.rs index 84391770..b1941729 100644 --- a/rust/routee-compass/src/app/compass/compass_app.rs +++ b/rust/routee-compass/src/app/compass/compass_app.rs @@ -2,10 +2,13 @@ use super::config::compass_app_builder::CompassAppBuilder; use crate::{ app::{ compass::{ - compass_configuration_field::CompassConfigurationField, - config::termination_model_builder::TerminationModelBuilder, + compass_app_error::CompassAppError, + config::{ + compass_configuration_field::CompassConfigurationField, + config_json_extension::ConfigJsonExtensions, graph_builder::DefaultGraphBuilder, + termination_model_builder::TerminationModelBuilder, + }, }, - compass_app_error::CompassAppError, search::{search_app::SearchApp, search_app_result::SearchAppResult}, }, plugin::{ @@ -18,14 +21,12 @@ use config::Config; use itertools::{Either, Itertools}; use rayon::{current_num_threads, prelude::*}; use routee_compass_core::{ - algorithm::search::search_algorithm::SearchAlgorithm, - model::{ - cost::cost::Cost, - graph::{graph::Graph, graph_config::GraphConfig}, - }, + algorithm::search::search_algorithm::SearchAlgorithm, model::cost::cost::Cost, util::duration_extension::DurationExtension, }; -use std::path::PathBuf; +use std::path::{Path, PathBuf}; + +pub const CONFIG_FILE_KEY: &str = "config_file"; /// Instance of RouteE Compass as an application. /// When constructed, it holds @@ -42,7 +43,7 @@ pub struct CompassApp { pub parallelism: usize, } -impl TryFrom<PathBuf> for CompassApp { +impl TryFrom<&Path> for CompassApp { type Error = CompassAppError; /// Builds a CompassApp from a configuration filepath, using the default CompassAppBuilder. @@ -57,18 +58,30 @@ impl TryFrom<PathBuf> for CompassApp { /// # Returns /// /// * an instance of [`CompassApp`], or an error if load failed. - fn try_from(conf_file: PathBuf) -> Result<Self, Self::Error> { + fn try_from(conf_file: &Path) -> Result<Self, Self::Error> { let default_config = config::File::from_str( include_str!("config.default.toml"), config::FileFormat::Toml, ); + // We want to store the location of where the config file + // was found so we can use it later to resolve relative paths + let conf_file_string = conf_file + .to_str() + .ok_or(CompassAppError::InternalError( + "Could not parse incoming config file path".to_string(), + ))? + .to_string(); + let config = Config::builder() .add_source(default_config) .add_source(config::File::from(conf_file)) + .set_override(CONFIG_FILE_KEY, conf_file_string)? .build() .map_err(CompassAppError::ConfigError)?; - log::info!("Config: {:?}", config); + + log::debug!("Loaded config: {:?}", config); + let builder = CompassAppBuilder::default(); let compass_app = CompassApp::try_from((&config, &builder))?; return Ok(compass_app); @@ -98,14 +111,22 @@ impl TryFrom<(&Config, &CompassAppBuilder)> for CompassApp { fn try_from(pair: (&Config, &CompassAppBuilder)) -> Result<Self, Self::Error> { let (config, builder) = pair; - let alg_params = - config.get::<serde_json::Value>(CompassConfigurationField::Algorithm.to_str())?; + // Get the root config path so we can resolve paths relative + // to where the config file is located. + let root_config_path = config.get::<PathBuf>(CONFIG_FILE_KEY)?; + + let config_json = config + .clone() + .try_deserialize::<serde_json::Value>()? + .normalize_file_paths(&root_config_path)?; + + let alg_params = config_json.get_config_section(CompassConfigurationField::Algorithm)?; let search_algorithm = SearchAlgorithm::try_from(&alg_params)?; // build traversal model let traversal_start = Local::now(); let traversal_params = - config.get::<serde_json::Value>(CompassConfigurationField::Traversal.to_str())?; + config_json.get_config_section(CompassConfigurationField::Traversal)?; let traversal_model_service = builder.build_traversal_model_service(&traversal_params)?; let traversal_duration = (Local::now() - traversal_start) .to_std() @@ -118,7 +139,7 @@ impl TryFrom<(&Config, &CompassAppBuilder)> for CompassApp { // build frontier model let frontier_start = Local::now(); let frontier_params = - config.get::<serde_json::Value>(CompassConfigurationField::Frontier.to_str())?; + config_json.get_config_section(CompassConfigurationField::Frontier)?; let frontier_model = builder.build_frontier_model(frontier_params)?; let frontier_duration = (Local::now() - frontier_start) .to_std() @@ -130,15 +151,13 @@ impl TryFrom<(&Config, &CompassAppBuilder)> for CompassApp { // build termination model let termination_model_json = - config.get::<serde_json::Value>(CompassConfigurationField::Termination.to_str())?; + config_json.get_config_section(CompassConfigurationField::Termination)?; let termination_model = TerminationModelBuilder::build(&termination_model_json)?; // build graph let graph_start = Local::now(); - let graph_conf = &config - .get::<GraphConfig>(CompassConfigurationField::Graph.to_str()) - .map_err(CompassAppError::ConfigError)?; - let graph = Graph::try_from(graph_conf)?; + let graph_params = config_json.get_config_section(CompassConfigurationField::Graph)?; + let graph = DefaultGraphBuilder::build(&graph_params)?; let graph_duration = (Local::now() - graph_start) .to_std() .map_err(|e| CompassAppError::InternalError(e.to_string()))?; @@ -165,11 +184,10 @@ impl TryFrom<(&Config, &CompassAppBuilder)> for CompassApp { // build plugins let plugins_start = Local::now(); - let plugins_config = - config.get::<serde_json::Value>(CompassConfigurationField::Plugins.to_str())?; + let plugins_config = config_json.get_config_section(CompassConfigurationField::Plugins)?; - let input_plugins = builder.build_input_plugins(plugins_config.clone())?; - let output_plugins = builder.build_output_plugins(plugins_config.clone())?; + let input_plugins = builder.build_input_plugins(&plugins_config)?; + let output_plugins = builder.build_output_plugins(&plugins_config)?; let plugins_duration = to_std(Local::now() - plugins_start)?; log::info!( @@ -423,8 +441,8 @@ mod tests { .join("speeds_test") .join("speeds_debug.toml"); - let app = CompassApp::try_from(conf_file_test) - .or(CompassApp::try_from(conf_file_debug)) + let app = CompassApp::try_from(conf_file_test.as_path()) + .or(CompassApp::try_from(conf_file_debug.as_path())) .unwrap(); let query = serde_json::json!({ "origin_vertex": 0, diff --git a/rust/routee-compass/src/app/compass_app_error.rs b/rust/routee-compass/src/app/compass/compass_app_error.rs similarity index 98% rename from rust/routee-compass/src/app/compass_app_error.rs rename to rust/routee-compass/src/app/compass/compass_app_error.rs index ed0ae0ba..ba73aa2d 100644 --- a/rust/routee-compass/src/app/compass_app_error.rs +++ b/rust/routee-compass/src/app/compass/compass_app_error.rs @@ -1,4 +1,4 @@ -use super::compass::{ +use super::{ compass_input_field::CompassInputField, config::compass_configuration_error::CompassConfigurationError, }; diff --git a/rust/routee-compass/src/app/compass/compass_json_extensions.rs b/rust/routee-compass/src/app/compass/compass_json_extensions.rs index 1b93fa99..3c0ad2a1 100644 --- a/rust/routee-compass/src/app/compass/compass_json_extensions.rs +++ b/rust/routee-compass/src/app/compass/compass_json_extensions.rs @@ -1,5 +1,5 @@ use super::compass_input_field::CompassInputField; -use crate::app::compass_app_error::CompassAppError; +use crate::app::compass::compass_app_error::CompassAppError; pub trait CompassJsonExtensions { fn get_queries(&self) -> Result<Vec<serde_json::Value>, CompassAppError>; diff --git a/rust/routee-compass/src/app/compass/config.default.toml b/rust/routee-compass/src/app/compass/config.default.toml index 538edbf0..d1f4ab37 100644 --- a/rust/routee-compass/src/app/compass/config.default.toml +++ b/rust/routee-compass/src/app/compass/config.default.toml @@ -1,8 +1,6 @@ parallelism = 2 [graph] -edge_list_csv = "edges.csv.gz" -vertex_list_csv = "vertices.csv.gz" verbose = true [algorithm] @@ -20,5 +18,5 @@ limit = "00:02:00" frequency = 100_000 [plugin] -input_plugins = [{ type = "vertex_rtree", vertices_file = "vertices.csv.gz" }] +input_plugins = [] output_plugins = [] diff --git a/rust/routee-compass/src/app/compass/config/builders.rs b/rust/routee-compass/src/app/compass/config/builders.rs index 4c7c06d3..c5d24896 100644 --- a/rust/routee-compass/src/app/compass/config/builders.rs +++ b/rust/routee-compass/src/app/compass/config/builders.rs @@ -1,7 +1,8 @@ use super::compass_configuration_error::CompassConfigurationError; use crate::plugin::{input::input_plugin::InputPlugin, output::output_plugin::OutputPlugin}; use routee_compass_core::model::{ - frontier::frontier_model::FrontierModel, traversal::traversal_model::TraversalModel, + frontier::frontier_model::FrontierModel, graph::graph::Graph, + traversal::traversal_model::TraversalModel, }; use std::sync::Arc; @@ -11,7 +12,7 @@ use std::sync::Arc; /// A [`TraversalModelBuilder`] instance should be an empty struct that implements /// this trait. pub trait TraversalModelBuilder { - /// Builds a [`TraversalModelService`] from JSON configuration. + /// Builds a [`TraversalModelService`] from configuration. /// /// # Arguments /// diff --git a/rust/routee-compass/src/app/compass/config/compass_app_builder.rs b/rust/routee-compass/src/app/compass/config/compass_app_builder.rs index 76fc91eb..ecc7f08b 100644 --- a/rust/routee-compass/src/app/compass/config/compass_app_builder.rs +++ b/rust/routee-compass/src/app/compass/config/compass_app_builder.rs @@ -4,6 +4,7 @@ use super::{ TraversalModelService, }, compass_configuration_error::CompassConfigurationError, + compass_configuration_field::CompassConfigurationField, frontier_model::{ no_restriction_builder::NoRestrictionBuilder, road_class_builder::RoadClassBuilder, }, @@ -13,23 +14,18 @@ use super::{ speed_lookup_builder::SpeedLookupBuilder, }, }; -use crate::{ - app::compass::compass_configuration_field::CompassConfigurationField, - plugin::{ - input::{ - default::{ - grid_search::builder::GridSearchBuilder, rtree::builder::VertexRTreeBuilder, - }, - input_plugin::InputPlugin, - }, - output::{ - default::{ - edgeidlist::builder::EdgeIdListOutputPluginBuilder, - summary::builder::SummaryOutputPluginBuilder, - traversal::builder::TraversalPluginBuilder, uuid::builder::UUIDOutputPluginBuilder, - }, - output_plugin::OutputPlugin, +use crate::plugin::{ + input::{ + default::{grid_search::builder::GridSearchBuilder, rtree::builder::VertexRTreeBuilder}, + input_plugin::InputPlugin, + }, + output::{ + default::{ + edgeidlist::builder::EdgeIdListOutputPluginBuilder, + summary::builder::SummaryOutputPluginBuilder, + traversal::builder::TraversalPluginBuilder, uuid::builder::UUIDOutputPluginBuilder, }, + output_plugin::OutputPlugin, }, }; use routee_compass_core::model::frontier::frontier_model::FrontierModel; @@ -198,14 +194,14 @@ impl CompassAppBuilder { pub fn build_input_plugins( &self, - config: serde_json::Value, + config: &serde_json::Value, ) -> Result<Vec<Box<dyn InputPlugin>>, CompassConfigurationError> { - let input_plugins_obj = config.get("input_plugins").ok_or( - CompassConfigurationError::ExpectedFieldForComponent( + let input_plugins_obj = config + .get(CompassConfigurationField::InputPlugins.to_str()) + .ok_or(CompassConfigurationError::ExpectedFieldForComponent( CompassConfigurationField::InputPlugins.to_string(), - String::from("input_plugins"), - ), - )?; + CompassConfigurationField::Plugins.to_string(), + ))?; let input_plugins = input_plugins_obj.as_array().ok_or( CompassConfigurationError::ExpectedFieldWithType( String::from("input_plugins"), @@ -247,14 +243,14 @@ impl CompassAppBuilder { pub fn build_output_plugins( &self, - config: serde_json::Value, + config: &serde_json::Value, ) -> Result<Vec<Box<dyn OutputPlugin>>, CompassConfigurationError> { - let output_plugins_obj = config.get("output_plugins").ok_or( - CompassConfigurationError::ExpectedFieldForComponent( + let output_plugins_obj = config + .get(CompassConfigurationField::OutputPlugins.to_str()) + .ok_or(CompassConfigurationError::ExpectedFieldForComponent( CompassConfigurationField::OutputPlugins.to_string(), - String::from("output_plugins"), - ), - )?; + CompassConfigurationField::Plugins.to_string(), + ))?; let output_plugins = output_plugins_obj.as_array().ok_or( CompassConfigurationError::ExpectedFieldWithType( String::from("output_plugins"), diff --git a/rust/routee-compass/src/app/compass/config/compass_configuration_error.rs b/rust/routee-compass/src/app/compass/config/compass_configuration_error.rs index aaa627c3..0f46547d 100644 --- a/rust/routee-compass/src/app/compass/config/compass_configuration_error.rs +++ b/rust/routee-compass/src/app/compass/config/compass_configuration_error.rs @@ -1,5 +1,8 @@ +use config::ConfigError; use routee_compass_core::{ - model::traversal::traversal_model_error::TraversalModelError, + model::{ + graph::graph_error::GraphError, traversal::traversal_model_error::TraversalModelError, + }, util::conversion::conversion_error::ConversionError, }; @@ -15,6 +18,31 @@ pub enum CompassConfigurationError { ExpectedFieldWithTypeUnrecognized(String, String, String), #[error("unknown module {0} for component {1} provided by configuration")] UnknownModelNameForComponent(String, String), + #[error( + r#" + File '{0}' was not found. + This file came from field '{1}' for component '{2}'. + + First, make sure this file path is either relative to your config file, + or, is provided as an absolute path. + + Second, make sure the file exists. + + Third, make sure the config key ends with '_file' which is a schema requirement + for the CompassApp config. + "# + )] + FileNotFoundForComponent(String, String, String), + #[error("could not normalize incoming file {0}")] + FileNormalizationError(String), + #[error("Could not find incoming configuration file, tried {0} and {1}. Make sure the file exists and that the config key ends with '_file'")] + FileNormalizationNotFound(String, String), + #[error("{0}")] + InsertError(String), + #[error(transparent)] + GraphError(#[from] GraphError), + #[error(transparent)] + ConfigError(#[from] ConfigError), #[error(transparent)] IoError(#[from] std::io::Error), #[error(transparent)] diff --git a/rust/routee-compass/src/app/compass/compass_configuration_field.rs b/rust/routee-compass/src/app/compass/config/compass_configuration_field.rs similarity index 100% rename from rust/routee-compass/src/app/compass/compass_configuration_field.rs rename to rust/routee-compass/src/app/compass/config/compass_configuration_field.rs diff --git a/rust/routee-compass/src/app/compass/config/config_json_extension.rs b/rust/routee-compass/src/app/compass/config/config_json_extension.rs index 56b09948..9668f671 100644 --- a/rust/routee-compass/src/app/compass/config/config_json_extension.rs +++ b/rust/routee-compass/src/app/compass/config/config_json_extension.rs @@ -1,8 +1,26 @@ use super::compass_configuration_error::CompassConfigurationError; +use super::compass_configuration_field::CompassConfigurationField; use serde::de; -use std::str::FromStr; +use std::{ + path::{Path, PathBuf}, + str::FromStr, +}; pub trait ConfigJsonExtensions { + fn get_config_section( + &self, + section: CompassConfigurationField, + ) -> Result<serde_json::Value, CompassConfigurationError>; + fn get_config_path( + &self, + key: String, + parent_key: String, + ) -> Result<PathBuf, CompassConfigurationError>; + fn get_config_path_optional( + &self, + key: String, + parent_key: String, + ) -> Result<Option<PathBuf>, CompassConfigurationError>; fn get_config_string( &self, key: String, @@ -33,9 +51,60 @@ pub trait ConfigJsonExtensions { key: String, parent_key: String, ) -> Result<Option<T>, CompassConfigurationError>; + fn normalize_file_paths( + &self, + root_config_path: &Path, + ) -> Result<serde_json::Value, CompassConfigurationError>; } impl ConfigJsonExtensions for serde_json::Value { + fn get_config_section( + &self, + section: CompassConfigurationField, + ) -> Result<serde_json::Value, CompassConfigurationError> { + let section = self + .get(section.to_str()) + .ok_or(CompassConfigurationError::ExpectedFieldForComponent( + section.to_string(), + String::from(""), + ))? + .clone(); + + Ok(section) + } + fn get_config_path_optional( + &self, + key: String, + parent_key: String, + ) -> Result<Option<PathBuf>, CompassConfigurationError> { + match self.get(key.clone()) { + None => Ok(None), + Some(_) => { + let config_path = self.get_config_path(key, parent_key)?; + Ok(Some(config_path)) + } + } + } + fn get_config_path( + &self, + key: String, + parent_key: String, + ) -> Result<PathBuf, CompassConfigurationError> { + let path_string = self.get_config_string(key.clone(), parent_key.clone())?; + let path = PathBuf::from(&path_string); + + // if file can be found, just return it + if path.is_file() { + return Ok(path); + } else { + // can't find the file + return Err(CompassConfigurationError::FileNotFoundForComponent( + path_string, + key, + parent_key, + )); + } + } fn get_config_string( &self, key: String, @@ -154,4 +223,79 @@ impl ConfigJsonExtensions for serde_json::Value { } } } + + /// This function is used to normalize file paths in the configuration JSON object. + /// Incoming file paths can be in one of three locations: + /// + /// 1. Absolute path + /// 2. Relative path to the config file itself + /// 3. Relative path to where the application is being run + /// + /// This function scans each key value pair in the config and for any key that + /// ends with `_file`, it will attempt to normalize the path such that the application + /// can find the file regardless of where it is being executed. + /// + /// Arguments: + /// + /// * `root_config_path` - The path to the where the config file is located. + /// + /// Returns: + /// + /// * `Result<serde_json::Value, CompassConfigurationError>` - The JSON object with normalized paths. + fn normalize_file_paths( + &self, + root_config_path: &Path, + ) -> Result<serde_json::Value, CompassConfigurationError> { + match self { + serde_json::Value::String(path_string) => { + let path = Path::new(path_string); + + // no need to modify if the file exists + if path.is_file() { + return Ok(serde_json::Value::String(path_string.clone())); + } + + // next we try adding the root config path and see if that exists + let root_config_parent = match root_config_path.parent() { + Some(parent) => parent, + None => Path::new(""), + }; + let new_path = root_config_parent.join(&path); + let new_path_string = new_path + .to_str() + .ok_or(CompassConfigurationError::FileNormalizationError( + path_string.clone(), + ))? + .to_string(); + if new_path.is_file() { + Ok(serde_json::Value::String(new_path_string)) + } else { + // if we can't find the file in either location, we throw an error + Err(CompassConfigurationError::FileNormalizationNotFound( + path_string.clone(), + new_path_string, + )) + } + } + serde_json::Value::Object(obj) => { + let mut new_obj = serde_json::map::Map::new(); + for (key, value) in obj.iter() { + if key.ends_with("_file") || value.is_object() || value.is_array() { + new_obj.insert(key.clone(), value.normalize_file_paths(root_config_path)?); + } else { + new_obj.insert(key.clone(), value.clone()); + } + } + Ok(serde_json::Value::Object(new_obj)) + } + serde_json::Value::Array(arr) => { + let mut new_arr = Vec::new(); + for value in arr.iter() { + new_arr.push(value.normalize_file_paths(root_config_path)?); + } + Ok(serde_json::Value::Array(new_arr)) + } + _ => Ok(self.clone()), + } + } } diff --git a/rust/routee-compass/src/app/compass/config/frontier_model/road_class_builder.rs b/rust/routee-compass/src/app/compass/config/frontier_model/road_class_builder.rs index 6a018e0d..4cfe56b3 100644 --- a/rust/routee-compass/src/app/compass/config/frontier_model/road_class_builder.rs +++ b/rust/routee-compass/src/app/compass/config/frontier_model/road_class_builder.rs @@ -1,9 +1,7 @@ -use crate::app::compass::{ +use crate::app::compass::config::{ + builders::FrontierModelBuilder, compass_configuration_error::CompassConfigurationError, compass_configuration_field::CompassConfigurationField, - config::{ - builders::FrontierModelBuilder, compass_configuration_error::CompassConfigurationError, - config_json_extension::ConfigJsonExtensions, - }, + config_json_extension::ConfigJsonExtensions, }; use routee_compass_core::{ model::frontier::{default::road_class::RoadClassFrontierModel, frontier_model::FrontierModel}, @@ -23,7 +21,7 @@ impl FrontierModelBuilder for RoadClassBuilder { let valid_road_class_key = String::from("valid_road_classes"); let road_class_file = - parameters.get_config_string(road_class_file_key.clone(), frontier_key.clone())?; + parameters.get_config_path(road_class_file_key.clone(), frontier_key.clone())?; let road_classes_json = parameters .get(valid_road_class_key.clone()) diff --git a/rust/routee-compass/src/app/compass/config/graph_builder.rs b/rust/routee-compass/src/app/compass/config/graph_builder.rs new file mode 100644 index 00000000..1ddc01a3 --- /dev/null +++ b/rust/routee-compass/src/app/compass/config/graph_builder.rs @@ -0,0 +1,51 @@ +use routee_compass_core::model::graph::graph::Graph; + +use crate::app::compass::config::compass_configuration_field::CompassConfigurationField; + +use super::{ + compass_configuration_error::CompassConfigurationError, + config_json_extension::ConfigJsonExtensions, +}; + +pub struct DefaultGraphBuilder {} + +impl DefaultGraphBuilder { + /// tries to build a Graph from a JSON object. + /// + /// for both edge and vertex lists, we assume all ids can be used as indices + /// to an array data structure. to find the size of each array, we pass once + /// through each file to count the number of rows (minus header) of the CSV. + /// then we can build a Vec *once* and insert rows as we decode them without + /// a sort. + /// + /// # Arguments + /// + /// * `params` - configuration JSON object for building a `Graph` instance + /// + /// # Returns + /// + /// A graph instance, or an error if an IO error occurred. + pub fn build(params: &serde_json::Value) -> Result<Graph, CompassConfigurationError> { + let graph_key = CompassConfigurationField::Graph.to_string(); + let edge_list_csv = + params.get_config_path(String::from("edge_list_file"), graph_key.clone())?; + let vertex_list_csv = + params.get_config_path(String::from("vertex_list_file"), graph_key.clone())?; + let n_edges = + params.get_config_serde_optional(String::from("n_edges"), graph_key.clone())?; + let n_vertices = + params.get_config_serde_optional(String::from("n_vertices"), graph_key.clone())?; + let verbose: Option<bool> = + params.get_config_serde_optional(String::from("verbose"), graph_key.clone())?; + + let graph = Graph::from_files( + &edge_list_csv, + &vertex_list_csv, + n_edges, + n_vertices, + verbose, + )?; + + Ok(graph) + } +} diff --git a/rust/routee-compass/src/app/compass/config/mod.rs b/rust/routee-compass/src/app/compass/config/mod.rs index 6142bc6b..c3b300ef 100644 --- a/rust/routee-compass/src/app/compass/config/mod.rs +++ b/rust/routee-compass/src/app/compass/config/mod.rs @@ -1,7 +1,9 @@ pub mod builders; pub mod compass_app_builder; pub mod compass_configuration_error; +pub mod compass_configuration_field; pub mod config_json_extension; pub mod frontier_model; +pub mod graph_builder; pub mod termination_model_builder; pub mod traversal_model; diff --git a/rust/routee-compass/src/app/compass/config/termination_model_builder.rs b/rust/routee-compass/src/app/compass/config/termination_model_builder.rs index 9f392a8d..73415295 100644 --- a/rust/routee-compass/src/app/compass/config/termination_model_builder.rs +++ b/rust/routee-compass/src/app/compass/config/termination_model_builder.rs @@ -1,5 +1,5 @@ use super::compass_configuration_error::CompassConfigurationError; -use crate::app::compass::compass_configuration_field::CompassConfigurationField; +use crate::app::compass::config::compass_configuration_field::CompassConfigurationField; use log; use routee_compass_core::model::termination::termination_model::TerminationModel; use routee_compass_core::util::conversion::duration_extension::DurationExtension; diff --git a/rust/routee-compass/src/app/compass/config/traversal_model/distance_builder.rs b/rust/routee-compass/src/app/compass/config/traversal_model/distance_builder.rs index e6b31906..0ca1f9a8 100644 --- a/rust/routee-compass/src/app/compass/config/traversal_model/distance_builder.rs +++ b/rust/routee-compass/src/app/compass/config/traversal_model/distance_builder.rs @@ -1,10 +1,8 @@ use crate::app::compass::config::config_json_extension::ConfigJsonExtensions; -use crate::app::compass::{ +use crate::app::compass::config::{ + builders::{TraversalModelBuilder, TraversalModelService}, + compass_configuration_error::CompassConfigurationError, compass_configuration_field::CompassConfigurationField, - config::{ - builders::{TraversalModelBuilder, TraversalModelService}, - compass_configuration_error::CompassConfigurationError, - }, }; use routee_compass_core::model::traversal::traversal_model::TraversalModel; use routee_compass_core::util::unit::BASE_DISTANCE_UNIT; diff --git a/rust/routee-compass/src/app/compass/config/traversal_model/speed_grade_energy_model_builder.rs b/rust/routee-compass/src/app/compass/config/traversal_model/speed_grade_energy_model_builder.rs index 23393eb1..f7f17127 100644 --- a/rust/routee-compass/src/app/compass/config/traversal_model/speed_grade_energy_model_builder.rs +++ b/rust/routee-compass/src/app/compass/config/traversal_model/speed_grade_energy_model_builder.rs @@ -1,7 +1,7 @@ use std::sync::Arc; -use crate::app::compass::compass_configuration_field::CompassConfigurationField; use crate::app::compass::config::builders::TraversalModelService; +use crate::app::compass::config::compass_configuration_field::CompassConfigurationField; use crate::app::compass::config::config_json_extension::ConfigJsonExtensions; use crate::app::compass::config::{ builders::TraversalModelBuilder, compass_configuration_error::CompassConfigurationError, @@ -28,23 +28,21 @@ impl TraversalModelBuilder for SpeedGradeEnergyModelBuilder { let traversal_key = CompassConfigurationField::Traversal.to_string(); let speed_table_path = - params.get_config_string(String::from("speed_table_path"), traversal_key.clone())?; + params.get_config_path(String::from("speed_table_file"), traversal_key.clone())?; let speed_table_speed_unit = params.get_config_serde::<SpeedUnit>( String::from("speed_table_speed_unit"), traversal_key.clone(), )?; - let grade_table_path = params.get_config_serde_optional::<String>( - String::from("grade_table_path"), - traversal_key.clone(), - )?; + let grade_table_path = params + .get_config_path_optional(String::from("grade_table_file"), traversal_key.clone())?; let grade_table_grade_unit = params.get_config_serde_optional::<GradeUnit>( String::from("graph_grade_unit"), traversal_key.clone(), )?; let energy_model_path = - params.get_config_string(String::from("energy_model_path"), traversal_key.clone())?; + params.get_config_path(String::from("energy_model_file"), traversal_key.clone())?; let model_type = params .get_config_serde::<ModelType>(String::from("model_type"), traversal_key.clone())?; let energy_model_speed_unit = params.get_config_serde::<SpeedUnit>( @@ -80,11 +78,11 @@ impl TraversalModelBuilder for SpeedGradeEnergyModelBuilder { )?; let inner_service = SpeedGradeModelService::new( - speed_table_path, + &speed_table_path, speed_table_speed_unit, - grade_table_path, + &grade_table_path, grade_table_grade_unit, - energy_model_path, + &energy_model_path, model_type, ideal_energy_rate_option, energy_model_speed_unit, diff --git a/rust/routee-compass/src/app/compass/config/traversal_model/speed_lookup_builder.rs b/rust/routee-compass/src/app/compass/config/traversal_model/speed_lookup_builder.rs index 18fa13ce..eaf30d18 100644 --- a/rust/routee-compass/src/app/compass/config/traversal_model/speed_lookup_builder.rs +++ b/rust/routee-compass/src/app/compass/config/traversal_model/speed_lookup_builder.rs @@ -1,5 +1,5 @@ -use crate::app::compass::compass_configuration_field::CompassConfigurationField; use crate::app::compass::config::builders::TraversalModelService; +use crate::app::compass::config::compass_configuration_field::CompassConfigurationField; use crate::app::compass::config::config_json_extension::ConfigJsonExtensions; use crate::app::compass::config::{ builders::TraversalModelBuilder, compass_configuration_error::CompassConfigurationError, @@ -23,7 +23,7 @@ impl TraversalModelBuilder for SpeedLookupBuilder { let traversal_key = CompassConfigurationField::Traversal.to_string(); // todo: optional output time unit let filename = - params.get_config_string(String::from("speed_table_path"), traversal_key.clone())?; + params.get_config_path(String::from("speed_table_file"), traversal_key.clone())?; let speed_unit = params .get_config_serde::<SpeedUnit>(String::from("speed_unit"), traversal_key.clone())?; let distance_unit = params.get_config_serde_optional::<DistanceUnit>( diff --git a/rust/routee-compass/src/app/compass/mod.rs b/rust/routee-compass/src/app/compass/mod.rs index e3b2835f..cdb5ee53 100644 --- a/rust/routee-compass/src/app/compass/mod.rs +++ b/rust/routee-compass/src/app/compass/mod.rs @@ -1,6 +1,6 @@ pub mod compass_app; pub mod compass_app_args; -pub mod compass_configuration_field; +pub mod compass_app_error; pub mod compass_input_field; pub mod compass_json_extensions; pub mod config; diff --git a/rust/routee-compass/src/app/geom/geom_app.rs b/rust/routee-compass/src/app/geom/geom_app.rs index 731d9eae..37b3cd83 100644 --- a/rust/routee-compass/src/app/geom/geom_app.rs +++ b/rust/routee-compass/src/app/geom/geom_app.rs @@ -1,4 +1,4 @@ -use crate::app::compass_app_error::CompassAppError; +use crate::app::compass::compass_app_error::CompassAppError; use crate::plugin::output::default::traversal::utils::parse_linestring; use geo::LineString; use kdam::Bar; diff --git a/rust/routee-compass/src/app/mod.rs b/rust/routee-compass/src/app/mod.rs index b7d62691..ae855b0d 100644 --- a/rust/routee-compass/src/app/mod.rs +++ b/rust/routee-compass/src/app/mod.rs @@ -1,4 +1,3 @@ pub mod compass; -pub mod compass_app_error; pub mod geom; pub mod search; diff --git a/rust/routee-compass/src/app/search/search_app.rs b/rust/routee-compass/src/app/search/search_app.rs index d37019f2..d2d5780a 100644 --- a/rust/routee-compass/src/app/search/search_app.rs +++ b/rust/routee-compass/src/app/search/search_app.rs @@ -1,6 +1,9 @@ use super::search_app_result::SearchAppResult; use crate::{ - app::{compass::config::builders::TraversalModelService, compass_app_error::CompassAppError}, + app::{ + compass::compass_app_error::CompassAppError, + compass::config::builders::TraversalModelService, + }, plugin::input::input_json_extensions::InputJsonExtensions, }; use chrono::Local; diff --git a/rust/routee-compass/src/main.rs b/rust/routee-compass/src/main.rs index 9f9b5c47..899a1d6f 100644 --- a/rust/routee-compass/src/main.rs +++ b/rust/routee-compass/src/main.rs @@ -1,9 +1,9 @@ use clap::Parser; -use log::info; +use log::{error, info}; use routee_compass::app::compass::compass_app::CompassApp; use routee_compass::app::compass::compass_app_args::CompassAppArgs; +use routee_compass::app::compass::compass_app_error::CompassAppError; use routee_compass::app::compass::compass_json_extensions::CompassJsonExtensions; -use routee_compass::app::compass_app_error::CompassAppError; use std::error::Error; use std::fs::File; use std::io::BufReader; @@ -18,7 +18,13 @@ fn main() -> Result<(), Box<dyn Error>> { "No configuration file specified".to_string(), ))?; - let compass_app = CompassApp::try_from(conf_file)?; + let compass_app = match CompassApp::try_from(conf_file.as_path()) { + Ok(app) => app, + Err(e) => { + error!("Could not build CompassApp from config file: {}", e); + return Err(Box::new(e)); + } + }; // read user file containing JSON query/queries let query_file = File::open(args.query_file.clone()).map_err(|_e| { diff --git a/rust/routee-compass/src/plugin/input/default/rtree/builder.rs b/rust/routee-compass/src/plugin/input/default/rtree/builder.rs index 6c110850..491a5c43 100644 --- a/rust/routee-compass/src/plugin/input/default/rtree/builder.rs +++ b/rust/routee-compass/src/plugin/input/default/rtree/builder.rs @@ -1,8 +1,7 @@ -use std::path::PathBuf; - use crate::{ app::compass::config::{ builders::InputPluginBuilder, compass_configuration_error::CompassConfigurationError, + config_json_extension::ConfigJsonExtensions, }, plugin::input::input_plugin::InputPlugin, }; @@ -17,19 +16,10 @@ impl InputPluginBuilder for VertexRTreeBuilder { parameters: &serde_json::Value, ) -> Result<Box<dyn InputPlugin>, CompassConfigurationError> { let vertex_filename_key = String::from("vertices_file"); - let vertex_filename = parameters - .get(&vertex_filename_key) - .ok_or(CompassConfigurationError::ExpectedFieldForComponent( - vertex_filename_key.clone(), - String::from("Vertex RTree Input Plugin"), - ))? - .as_str() - .map(String::from) - .ok_or(CompassConfigurationError::ExpectedFieldWithType( - vertex_filename_key.clone(), - String::from("String"), - ))?; - let vertex_path = PathBuf::from(vertex_filename); + let vertex_path = parameters.get_config_path( + vertex_filename_key, + String::from("Vertex RTree Input Plugin"), + )?; let rtree = RTreePlugin::from_file(&vertex_path).map_err(CompassConfigurationError::PluginError)?; let m: Box<dyn InputPlugin> = Box::new(rtree); diff --git a/rust/routee-compass/src/plugin/input/default/rtree/plugin.rs b/rust/routee-compass/src/plugin/input/default/rtree/plugin.rs index 499d2c8c..c37bc082 100644 --- a/rust/routee-compass/src/plugin/input/default/rtree/plugin.rs +++ b/rust/routee-compass/src/plugin/input/default/rtree/plugin.rs @@ -1,4 +1,4 @@ -use std::path::Path; +use std::path::{Path, PathBuf}; use crate::plugin::input::input_json_extensions::InputJsonExtensions; use crate::plugin::input::input_plugin::InputPlugin; diff --git a/rust/routee-compass/src/plugin/output/default/traversal/builder.rs b/rust/routee-compass/src/plugin/output/default/traversal/builder.rs index d2dd2596..be82bb4d 100644 --- a/rust/routee-compass/src/plugin/output/default/traversal/builder.rs +++ b/rust/routee-compass/src/plugin/output/default/traversal/builder.rs @@ -45,7 +45,7 @@ impl OutputPluginBuilder for TraversalPluginBuilder { let tree_geometry_key = String::from("tree"); let geometry_filename = - parameters.get_config_string(geometry_filename_key, parent_key.clone())?; + parameters.get_config_path(geometry_filename_key, parent_key.clone())?; let route: Option<TraversalOutputFormat> = parameters.get_config_serde_optional(route_geometry_key, parent_key.clone())?; let tree: Option<TraversalOutputFormat> = diff --git a/rust/routee-compass/src/plugin/output/default/traversal/plugin.rs b/rust/routee-compass/src/plugin/output/default/traversal/plugin.rs index af6d4d31..4a784919 100644 --- a/rust/routee-compass/src/plugin/output/default/traversal/plugin.rs +++ b/rust/routee-compass/src/plugin/output/default/traversal/plugin.rs @@ -1,3 +1,6 @@ +use std::path::Path; +use std::path::PathBuf; + use super::json_extensions::TraversalJsonField; use super::traversal_output_format::TraversalOutputFormat; use super::utils::parse_linestring; @@ -18,15 +21,15 @@ pub struct TraversalPlugin { } impl TraversalPlugin { - pub fn from_file( - filename: &String, + pub fn from_file<P: AsRef<Path>>( + filename: &P, route: Option<TraversalOutputFormat>, tree: Option<TraversalOutputFormat>, ) -> Result<TraversalPlugin, PluginError> { let count = fs_utils::line_count(filename.clone(), fs_utils::is_gzip(&filename)).map_err(|e| { PluginError::FileReadError { - filename: filename.clone(), + filename: filename.as_ref().to_path_buf(), message: e.to_string(), } })?; @@ -43,7 +46,7 @@ impl TraversalPlugin { }); let geoms = read_raw_file(&filename, parse_linestring, Some(cb)).map_err(|e| { PluginError::FileReadError { - filename: filename.clone(), + filename: filename.as_ref().to_path_buf(), message: e.to_string(), } })?; @@ -171,7 +174,7 @@ mod tests { route_runtime: Duration::ZERO, total_runtime: Duration::ZERO, }; - let filename = mock_geometry_file().to_str().unwrap().to_string(); + let filename = mock_geometry_file(); let route_geometry = true; let tree_geometry = false; let geom_plugin = diff --git a/rust/routee-compass/src/plugin/output/default/uuid/builder.rs b/rust/routee-compass/src/plugin/output/default/uuid/builder.rs index fc6ed2a1..8281be3c 100644 --- a/rust/routee-compass/src/plugin/output/default/uuid/builder.rs +++ b/rust/routee-compass/src/plugin/output/default/uuid/builder.rs @@ -1,6 +1,7 @@ use crate::{ app::compass::config::{ builders::OutputPluginBuilder, compass_configuration_error::CompassConfigurationError, + config_json_extension::ConfigJsonExtensions, }, plugin::output::output_plugin::OutputPlugin, }; @@ -15,18 +16,7 @@ impl OutputPluginBuilder for UUIDOutputPluginBuilder { parameters: &serde_json::Value, ) -> Result<Box<dyn OutputPlugin>, CompassConfigurationError> { let uuid_filename_key = String::from("uuid_file"); - let uuid_filename = parameters - .get(&uuid_filename_key) - .ok_or(CompassConfigurationError::ExpectedFieldForComponent( - uuid_filename_key.clone(), - String::from("UUID Output Plugin"), - ))? - .as_str() - .map(String::from) - .ok_or(CompassConfigurationError::ExpectedFieldWithType( - uuid_filename_key.clone(), - String::from("String"), - ))?; + let uuid_filename = parameters.get_config_path(uuid_filename_key, String::from("uuid"))?; let uuid_plugin = UUIDOutputPlugin::from_file(&uuid_filename) .map_err(CompassConfigurationError::PluginError)?; diff --git a/rust/routee-compass/src/plugin/output/default/uuid/plugin.rs b/rust/routee-compass/src/plugin/output/default/uuid/plugin.rs index 40a7341d..9954e999 100644 --- a/rust/routee-compass/src/plugin/output/default/uuid/plugin.rs +++ b/rust/routee-compass/src/plugin/output/default/uuid/plugin.rs @@ -1,3 +1,5 @@ +use std::path::{Path, PathBuf}; + use super::json_extensions::UUIDJsonExtensions; use crate::app::search::search_app_result::SearchAppResult; @@ -12,11 +14,11 @@ pub struct UUIDOutputPlugin { } impl UUIDOutputPlugin { - pub fn from_file(filename: &String) -> Result<UUIDOutputPlugin, PluginError> { + pub fn from_file<P: AsRef<Path>>(filename: &P) -> Result<UUIDOutputPlugin, PluginError> { let count = fs_utils::line_count(filename.clone(), fs_utils::is_gzip(&filename)).map_err(|e| { PluginError::FileReadError { - filename: filename.clone(), + filename: filename.as_ref().to_path_buf(), message: e.to_string(), } })?; @@ -34,7 +36,7 @@ impl UUIDOutputPlugin { let uuids = read_raw_file(&filename, |_idx, row| Ok(row), Some(cb)).map_err(|e| { PluginError::FileReadError { - filename: filename.clone(), + filename: filename.as_ref().to_path_buf(), message: e.to_string(), } })?; diff --git a/rust/routee-compass/src/plugin/plugin_error.rs b/rust/routee-compass/src/plugin/plugin_error.rs index 62200727..62a89fd7 100644 --- a/rust/routee-compass/src/plugin/plugin_error.rs +++ b/rust/routee-compass/src/plugin/plugin_error.rs @@ -1,3 +1,5 @@ +use std::path::PathBuf; + use geo::Coord; use routee_compass_core::{ algorithm::search::search_error::SearchError, model::graph::edge_id::EdgeId, @@ -16,7 +18,7 @@ pub enum PluginError { #[error("nearest vertex not found for coord {0:?}")] NearestVertexNotFound(Coord), #[error("unable to read file {filename} due to {message}")] - FileReadError { filename: String, message: String }, + FileReadError { filename: PathBuf, message: String }, #[error(transparent)] JsonError(#[from] serde_json::Error), #[error(transparent)]

Standardize paths types Right now we have a mix of different ways to represent our paths in the application, for example: - `PathBuf` - `&Path` - `P: AsRef<Path>` We should spend some time to review and attempt to standardize these where it makes sense to do so.

2023-10-27T20:56:54

0.0

[]

aerleon/aerleon

aerleon__aerleon-335

fb514ba28a0d535fe914d4dd602a180b69b041bb

diff --git a/aerleon/lib/speedway.py b/aerleon/lib/speedway.py index 4a0c1835..632b7d0b 100644 --- a/aerleon/lib/speedway.py +++ b/aerleon/lib/speedway.py @@ -43,6 +43,6 @@ class Speedway(iptables.Iptables): _RENDER_PREFIX = '*filter' _RENDER_SUFFIX = 'COMMIT' - _DEFAULTACTION_FORMAT = ':%s %s' + _DEFAULTACTION_FORMAT = ':%s %s [0:0]' _TERM = Term

Speedway does not include packet/byte count in its output Normally the iptables-save command will provide details on how many packets and bytes have gone through the specific filter. The syntax is like ``` :OUTPUT DROP [0:0] -N O_deny-all -A O_deny-all -p tcp -j DROP -A O_deny-all -p udp -j DROP COMMIT ``` where the counters are specifically `[0:0]` This is not included in our output. This may cause some unknown incompatibility we have not heard of or it might be fine. In either case any parser that is working off outputs from `iptables-save` will expect to see this in our output.

2023-08-07T03:07:22

0.0

[]

aerleon/aerleon

aerleon__aerleon-334

7ba9f2b08f3fea764defe38e3c2b6fb71cfc0e55

diff --git a/aerleon/lib/juniper.py b/aerleon/lib/juniper.py index 13f00d23..1132946d 100644 --- a/aerleon/lib/juniper.py +++ b/aerleon/lib/juniper.py @@ -16,7 +16,6 @@ """Juniper JCL generator.""" - from typing import Dict, List, Set, Tuple, Union from absl import logging @@ -486,17 +485,25 @@ def __str__(self): # protocol if self.term.protocol: + # Juniper has deprecated the use of icmpv6 in favor of icmp6. + # https://github.com/aerleon/aerleon/issues/323 + protocol = self.term.protocol[:] + if 'icmpv6' in self.term.protocol: + loc = protocol.index('icmpv6') + protocol[loc] = 'icmp6' # both are supported on JunOS, but only icmp6 is supported # on SRX loopback stateless filter - config.Append(family_keywords['protocol'] + ' ' + self._Group(self.term.protocol)) + config.Append(family_keywords['protocol'] + ' ' + self._Group(protocol)) # protocol if self.term.protocol_except: # same as above + protocol_except = self.term.protocol_except[:] + if 'icmpv6' in self.term.protocol_except: + loc = protocol_except.index('icmpv6') + protocol_except[loc] = 'icmp6' config.Append( - family_keywords['protocol-except'] - + ' ' - + self._Group(self.term.protocol_except) + family_keywords['protocol-except'] + ' ' + self._Group(protocol_except) ) # port

Minor regression on juniper generator After upgrading from Capirca to Aerleon, the generated `juniper` policies use `icmpv6` instead of `icmp6` ```diff [edit firewall family inet6 filter loopback6 term allow_ok_icmp6 from] - next-header icmp6; + next-header icmpv6; ``` Both work but according to Juniper's [doc](https://www.juniper.net/documentation/us/en/software/junos/routing-policy/topics/concept/firewall-filter-match-conditions-for-ipv6-traffic.html): > Note: next-header icmp6 and next-header icmpv6 match conditions perform the same function. next-header icmp6 is the preferred option. next-header icmpv6 is hidden in the Junos OS CLI. So there is a risk that `icmpv6` becomes decommissioned at some point and breaks.

Does your policy look like ``` - name: foo protocol-except: icmp6 action: accept ``` or ``` - name: bar protocol-except: icmpv6 action: accept ``` The code as it exists now should be outputting verbatim what ever the protocol-except is. I believe this was the same in Capirca too. Thanks, I tested it with https://gerrit.wikimedia.org/r/c/operations/homer/public/+/945550 which is the only location in our policies where "icmpv6" is present, but unfortunately this fails with: > aerleon.lib.aclgenerator.UnsupportedFilterError: Protocol(s) ['icmp6'] are not supported.

2023-08-05T05:55:02

0.0

[]

aerleon/aerleon

aerleon__aerleon-235

71d1b70b6301283c66adada532410abef21f508a

diff --git a/policies/includes/arista_terms_include.yaml b/policies/includes/arista_terms_include.yaml index af9358b7..5eede3ca 100644 --- a/policies/includes/arista_terms_include.yaml +++ b/policies/includes/arista_terms_include.yaml @@ -10,6 +10,14 @@ terms: counter: wonky-prots-loopback action: accept + - name: accept-traceroute-exc + comment: "allow inbound traceroute from any source." + destination-port: TRACEROUTE + protocol: udp + counter: inbound-traceroute + action: accept + platform-exclude: arista_tp + - name: accept-traceroute comment: "allow inbound traceroute from any source." destination-port: TRACEROUTE diff --git a/policies/pol/sample_arista_tp.pol b/policies/pol/sample_arista_tp.pol index 04cabf27..b39a5c55 100644 --- a/policies/pol/sample_arista_tp.pol +++ b/policies/pol/sample_arista_tp.pol @@ -29,7 +29,6 @@ term accept-traceroute-exc { protocol:: udp counter:: inbound-traceroute action:: accept - expiration:: 2001-12-31 platform-exclude:: arista_tp }

Move term platform check to ACLGenerator Move term platform check to ACLGenerator

# [Codecov](https://codecov.io/gh/aerleon/aerleon/pull/218?src=pr&el=h1&utm_medium=referral&utm_source=github&utm_content=comment&utm_campaign=pr+comments&utm_term=aerleon) Report Base: **90.61**% // Head: **91.14**% // Increases project coverage by **`+0.52%`** :tada: > Coverage data is based on head [(`85ae7b2`)](https://codecov.io/gh/aerleon/aerleon/pull/218?src=pr&el=desc&utm_medium=referral&utm_source=github&utm_content=comment&utm_campaign=pr+comments&utm_term=aerleon) compared to base [(`6b3be9d`)](https://codecov.io/gh/aerleon/aerleon/commit/6b3be9dc015f0def70a953ecc396367532c23b2d?el=desc&utm_medium=referral&utm_source=github&utm_content=comment&utm_campaign=pr+comments&utm_term=aerleon). > Patch coverage: 98.43% of modified lines in pull request are covered. <details><summary>Additional details and impacted files</summary> ```diff @@ Coverage Diff @@ ## main #218 +/- ## ========================================== + Coverage 90.61% 91.14% +0.52% ========================================== Files 95 95 Lines 21960 21742 -218 Branches 3382 3265 -117 ========================================== - Hits 19900 19816 -84 + Misses 1340 1256 -84 + Partials 720 670 -50 ``` | Flag | Coverage Δ | | |---|---|---| | tests | `91.12% <98.43%> (+0.52%)` | :arrow_up: | Flags with carried forward coverage won't be shown. [Click here](https://docs.codecov.io/docs/carryforward-flags?utm_medium=referral&utm_source=github&utm_content=comment&utm_campaign=pr+comments&utm_term=aerleon#carryforward-flags-in-the-pull-request-comment) to find out more. | [Impacted Files](https://codecov.io/gh/aerleon/aerleon/pull/218?src=pr&el=tree&utm_medium=referral&utm_source=github&utm_content=comment&utm_campaign=pr+comments&utm_term=aerleon) | Coverage Δ | | |---|---|---| | [aerleon/lib/arista\_tp.py](https://codecov.io/gh/aerleon/aerleon/pull/218?src=pr&el=tree&utm_medium=referral&utm_source=github&utm_content=comment&utm_campaign=pr+comments&utm_term=aerleon#diff-YWVybGVvbi9saWIvYXJpc3RhX3RwLnB5) | `93.66% <ø> (+1.06%)` | :arrow_up: | | [aerleon/lib/cisco.py](https://codecov.io/gh/aerleon/aerleon/pull/218?src=pr&el=tree&utm_medium=referral&utm_source=github&utm_content=comment&utm_campaign=pr+comments&utm_term=aerleon#diff-YWVybGVvbi9saWIvY2lzY28ucHk=) | `89.07% <ø> (+3.13%)` | :arrow_up: | | [aerleon/lib/gcp\_hf.py](https://codecov.io/gh/aerleon/aerleon/pull/218?src=pr&el=tree&utm_medium=referral&utm_source=github&utm_content=comment&utm_campaign=pr+comments&utm_term=aerleon#diff-YWVybGVvbi9saWIvZ2NwX2hmLnB5) | `95.60% <ø> (+1.06%)` | :arrow_up: | | [aerleon/lib/iptables.py](https://codecov.io/gh/aerleon/aerleon/pull/218?src=pr&el=tree&utm_medium=referral&utm_source=github&utm_content=comment&utm_campaign=pr+comments&utm_term=aerleon#diff-YWVybGVvbi9saWIvaXB0YWJsZXMucHk=) | `90.52% <ø> (+0.86%)` | :arrow_up: | | [aerleon/lib/juniper.py](https://codecov.io/gh/aerleon/aerleon/pull/218?src=pr&el=tree&utm_medium=referral&utm_source=github&utm_content=comment&utm_campaign=pr+comments&utm_term=aerleon#diff-YWVybGVvbi9saWIvanVuaXBlci5weQ==) | `86.80% <ø> (+0.97%)` | :arrow_up: | | [aerleon/lib/junipersrx.py](https://codecov.io/gh/aerleon/aerleon/pull/218?src=pr&el=tree&utm_medium=referral&utm_source=github&utm_content=comment&utm_campaign=pr+comments&utm_term=aerleon#diff-YWVybGVvbi9saWIvanVuaXBlcnNyeC5weQ==) | `90.10% <ø> (+0.68%)` | :arrow_up: | | [aerleon/lib/nsxv.py](https://codecov.io/gh/aerleon/aerleon/pull/218?src=pr&el=tree&utm_medium=referral&utm_source=github&utm_content=comment&utm_campaign=pr+comments&utm_term=aerleon#diff-YWVybGVvbi9saWIvbnN4di5weQ==) | `88.35% <ø> (+3.33%)` | :arrow_up: | | [aerleon/lib/packetfilter.py](https://codecov.io/gh/aerleon/aerleon/pull/218?src=pr&el=tree&utm_medium=referral&utm_source=github&utm_content=comment&utm_campaign=pr+comments&utm_term=aerleon#diff-YWVybGVvbi9saWIvcGFja2V0ZmlsdGVyLnB5) | `83.09% <ø> (+1.49%)` | :arrow_up: | | [aerleon/lib/pcap.py](https://codecov.io/gh/aerleon/aerleon/pull/218?src=pr&el=tree&utm_medium=referral&utm_source=github&utm_content=comment&utm_campaign=pr+comments&utm_term=aerleon#diff-YWVybGVvbi9saWIvcGNhcC5weQ==) | `71.93% <ø> (+1.32%)` | :arrow_up: | | [aerleon/lib/windows.py](https://codecov.io/gh/aerleon/aerleon/pull/218?src=pr&el=tree&utm_medium=referral&utm_source=github&utm_content=comment&utm_campaign=pr+comments&utm_term=aerleon#diff-YWVybGVvbi9saWIvd2luZG93cy5weQ==) | `74.64% <ø> (+2.03%)` | :arrow_up: | | ... and [16 more](https://codecov.io/gh/aerleon/aerleon/pull/218?src=pr&el=tree-more&utm_medium=referral&utm_source=github&utm_content=comment&utm_campaign=pr+comments&utm_term=aerleon) | | Help us with your feedback. Take ten seconds to tell us [how you rate us](https://about.codecov.io/nps?utm_medium=referral&utm_source=github&utm_content=comment&utm_campaign=pr+comments&utm_term=aerleon). Have a feature suggestion? [Share it here.](https://app.codecov.io/gh/feedback/?utm_medium=referral&utm_source=github&utm_content=comment&utm_campaign=pr+comments&utm_term=aerleon) </details> [:umbrella: View full report at Codecov](https://codecov.io/gh/aerleon/aerleon/pull/218?src=pr&el=continue&utm_medium=referral&utm_source=github&utm_content=comment&utm_campaign=pr+comments&utm_term=aerleon). :loudspeaker: Do you have feedback about the report comment? [Let us know in this issue](https://about.codecov.io/codecov-pr-comment-feedback/?utm_medium=referral&utm_source=github&utm_content=comment&utm_campaign=pr+comments&utm_term=aerleon). # [Codecov](https://codecov.io/gh/aerleon/aerleon/pull/218?src=pr&el=h1&utm_medium=referral&utm_source=github&utm_content=comment&utm_campaign=pr+comments&utm_term=aerleon) Report Base: **90.61**% // Head: **91.14**% // Increases project coverage by **`+0.52%`** :tada: > Coverage data is based on head [(`85ae7b2`)](https://codecov.io/gh/aerleon/aerleon/pull/218?src=pr&el=desc&utm_medium=referral&utm_source=github&utm_content=comment&utm_campaign=pr+comments&utm_term=aerleon) compared to base [(`6b3be9d`)](https://codecov.io/gh/aerleon/aerleon/commit/6b3be9dc015f0def70a953ecc396367532c23b2d?el=desc&utm_medium=referral&utm_source=github&utm_content=comment&utm_campaign=pr+comments&utm_term=aerleon). > Patch coverage: 98.43% of modified lines in pull request are covered. <details><summary>Additional details and impacted files</summary> ```diff @@ Coverage Diff @@ ## main #218 +/- ## ========================================== + Coverage 90.61% 91.14% +0.52% ========================================== Files 95 95 Lines 21960 21742 -218 Branches 3382 3265 -117 ========================================== - Hits 19900 19816 -84 + Misses 1340 1256 -84 + Partials 720 670 -50 ``` | Flag | Coverage Δ | | |---|---|---| | tests | `91.12% <98.43%> (+0.52%)` | :arrow_up: | Flags with carried forward coverage won't be shown. [Click here](https://docs.codecov.io/docs/carryforward-flags?utm_medium=referral&utm_source=github&utm_content=comment&utm_campaign=pr+comments&utm_term=aerleon#carryforward-flags-in-the-pull-request-comment) to find out more. | [Impacted Files](https://codecov.io/gh/aerleon/aerleon/pull/218?src=pr&el=tree&utm_medium=referral&utm_source=github&utm_content=comment&utm_campaign=pr+comments&utm_term=aerleon) | Coverage Δ | | |---|---|---| | [aerleon/lib/arista\_tp.py](https://codecov.io/gh/aerleon/aerleon/pull/218?src=pr&el=tree&utm_medium=referral&utm_source=github&utm_content=comment&utm_campaign=pr+comments&utm_term=aerleon#diff-YWVybGVvbi9saWIvYXJpc3RhX3RwLnB5) | `93.66% <ø> (+1.06%)` | :arrow_up: | | [aerleon/lib/cisco.py](https://codecov.io/gh/aerleon/aerleon/pull/218?src=pr&el=tree&utm_medium=referral&utm_source=github&utm_content=comment&utm_campaign=pr+comments&utm_term=aerleon#diff-YWVybGVvbi9saWIvY2lzY28ucHk=) | `89.07% <ø> (+3.13%)` | :arrow_up: | | [aerleon/lib/gcp\_hf.py](https://codecov.io/gh/aerleon/aerleon/pull/218?src=pr&el=tree&utm_medium=referral&utm_source=github&utm_content=comment&utm_campaign=pr+comments&utm_term=aerleon#diff-YWVybGVvbi9saWIvZ2NwX2hmLnB5) | `95.60% <ø> (+1.06%)` | :arrow_up: | | [aerleon/lib/iptables.py](https://codecov.io/gh/aerleon/aerleon/pull/218?src=pr&el=tree&utm_medium=referral&utm_source=github&utm_content=comment&utm_campaign=pr+comments&utm_term=aerleon#diff-YWVybGVvbi9saWIvaXB0YWJsZXMucHk=) | `90.52% <ø> (+0.86%)` | :arrow_up: | | [aerleon/lib/juniper.py](https://codecov.io/gh/aerleon/aerleon/pull/218?src=pr&el=tree&utm_medium=referral&utm_source=github&utm_content=comment&utm_campaign=pr+comments&utm_term=aerleon#diff-YWVybGVvbi9saWIvanVuaXBlci5weQ==) | `86.80% <ø> (+0.97%)` | :arrow_up: | | [aerleon/lib/junipersrx.py](https://codecov.io/gh/aerleon/aerleon/pull/218?src=pr&el=tree&utm_medium=referral&utm_source=github&utm_content=comment&utm_campaign=pr+comments&utm_term=aerleon#diff-YWVybGVvbi9saWIvanVuaXBlcnNyeC5weQ==) | `90.10% <ø> (+0.68%)` | :arrow_up: | | [aerleon/lib/nsxv.py](https://codecov.io/gh/aerleon/aerleon/pull/218?src=pr&el=tree&utm_medium=referral&utm_source=github&utm_content=comment&utm_campaign=pr+comments&utm_term=aerleon#diff-YWVybGVvbi9saWIvbnN4di5weQ==) | `88.35% <ø> (+3.33%)` | :arrow_up: | | [aerleon/lib/packetfilter.py](https://codecov.io/gh/aerleon/aerleon/pull/218?src=pr&el=tree&utm_medium=referral&utm_source=github&utm_content=comment&utm_campaign=pr+comments&utm_term=aerleon#diff-YWVybGVvbi9saWIvcGFja2V0ZmlsdGVyLnB5) | `83.09% <ø> (+1.49%)` | :arrow_up: | | [aerleon/lib/pcap.py](https://codecov.io/gh/aerleon/aerleon/pull/218?src=pr&el=tree&utm_medium=referral&utm_source=github&utm_content=comment&utm_campaign=pr+comments&utm_term=aerleon#diff-YWVybGVvbi9saWIvcGNhcC5weQ==) | `71.93% <ø> (+1.32%)` | :arrow_up: | | [aerleon/lib/windows.py](https://codecov.io/gh/aerleon/aerleon/pull/218?src=pr&el=tree&utm_medium=referral&utm_source=github&utm_content=comment&utm_campaign=pr+comments&utm_term=aerleon#diff-YWVybGVvbi9saWIvd2luZG93cy5weQ==) | `74.64% <ø> (+2.03%)` | :arrow_up: | | ... and [16 more](https://codecov.io/gh/aerleon/aerleon/pull/218?src=pr&el=tree-more&utm_medium=referral&utm_source=github&utm_content=comment&utm_campaign=pr+comments&utm_term=aerleon) | | Help us with your feedback. Take ten seconds to tell us [how you rate us](https://about.codecov.io/nps?utm_medium=referral&utm_source=github&utm_content=comment&utm_campaign=pr+comments&utm_term=aerleon). Have a feature suggestion? [Share it here.](https://app.codecov.io/gh/feedback/?utm_medium=referral&utm_source=github&utm_content=comment&utm_campaign=pr+comments&utm_term=aerleon) </details> [:umbrella: View full report at Codecov](https://codecov.io/gh/aerleon/aerleon/pull/218?src=pr&el=continue&utm_medium=referral&utm_source=github&utm_content=comment&utm_campaign=pr+comments&utm_term=aerleon). :loudspeaker: Do you have feedback about the report comment? [Let us know in this issue](https://about.codecov.io/codecov-pr-comment-feedback/?utm_medium=referral&utm_source=github&utm_content=comment&utm_campaign=pr+comments&utm_term=aerleon).

2023-02-03T09:10:15

0.0

[]

aerleon/aerleon

aerleon__aerleon-205

0c764e8e94f078b604da1d831ee2312807e8264b

diff --git a/aerleon/lib/aclgenerator.py b/aerleon/lib/aclgenerator.py index b1c95a7c..04573f5e 100644 --- a/aerleon/lib/aclgenerator.py +++ b/aerleon/lib/aclgenerator.py @@ -17,11 +17,13 @@ """ACL Generator base class.""" import copy +import datetime import hashlib -import logging import re import string +from absl import logging + from aerleon.lib import policy @@ -303,6 +305,8 @@ def __init__(self, pol, exp_info): all_err = [] all_warn = [] for header, terms in pol.filters: + terms[:] = self._FilteredTerms(header, terms, exp_info) + # Verify valid keywords # error on unsupported optional keywords that could result # in dangerous or unexpected results @@ -541,6 +545,31 @@ def HexDigest(self, name, truncation_length=None): name_bytes = name.encode('UTF-8') return hashlib.sha256(name_bytes).hexdigest()[:truncation_length] + def _FilteredTerms(self, header, terms, exp_info): + new_terms = [] + filter_name = header.FilterName(self._PLATFORM) + current_date = datetime.datetime.utcnow().date() + exp_info_date = current_date + datetime.timedelta(weeks=exp_info) + + for term in terms: + if term.expiration: + if term.expiration <= current_date: + logging.warning( + 'WARNING: Term %s in policy %s is expired and will not be rendered.', + term.name, + filter_name, + ) + continue + elif term.expiration <= exp_info_date: + logging.info( + "INFO: Term %s in policy %s expires in less than two weeks.", + term.name, + filter_name, + ) + + new_terms.append(term) + + return new_terms def ProtocolNameToNumber(protocols, proto_to_num, name_to_num_map): """Convert a protocol name to a numeric value. diff --git a/aerleon/lib/arista_tp.py b/aerleon/lib/arista_tp.py index 27988fbb..bcf35d97 100644 --- a/aerleon/lib/arista_tp.py +++ b/aerleon/lib/arista_tp.py @@ -16,7 +16,6 @@ """arista traffic-policy generator.""" import copy -import datetime import re from absl import logging @@ -244,7 +243,7 @@ def __str__(self): if not has_match_criteria and not is_default_term: # this term doesn't match on anything and isn't a default-term logging.warning( - "WARNING: term %s has no valid match criteria and " "will not be rendered.", + "WARNING: term %s has no valid match criteria and will not be rendered.", self.term.name, ) return "" @@ -718,9 +717,6 @@ def _TranslatePolicy(self, pol, exp_info): self.arista_traffic_policies = [] af_map_txt = {"inet": "ipv4", "inet6": "ipv6"} - current_date = datetime.datetime.utcnow().date() - exp_info_date = current_date + datetime.timedelta(weeks=exp_info) - for header, terms in pol.filters: filter_options = header.FilterOptions(self._PLATFORM) filter_name = header.FilterName(self._PLATFORM) @@ -778,22 +774,6 @@ def _TranslatePolicy(self, pol, exp_info): if not term: continue - if term.expiration: - if term.expiration <= exp_info_date: - logging.info( - "INFO: term %s in policy %s expires " "in less than two weeks.", - term.name, - filter_name, - ) - if term.expiration <= current_date: - logging.warning( - "WARNING: term %s in policy %s is expired and " - "will not be rendered.", - term.name, - filter_name, - ) - continue - # emit warnings for unsupported options / terms if term.option: unsupported_opts = [] diff --git a/aerleon/lib/aruba.py b/aerleon/lib/aruba.py index 49b2edb6..1eec765d 100644 --- a/aerleon/lib/aruba.py +++ b/aerleon/lib/aruba.py @@ -16,10 +16,10 @@ """Aruba generator.""" -import datetime -import logging from typing import List, Tuple +from absl import logging + from aerleon.lib import aclgenerator _COMMENT_MARKER = '#' @@ -271,9 +271,6 @@ def _BuildTokens(self): def _TranslatePolicy(self, pol, exp_info): self.aruba_policies = [] - current_date = datetime.datetime.utcnow().date() - exp_info_date = current_date + datetime.timedelta(weeks=exp_info) - for header, terms in pol.filters: filter_name = header.FilterName(self._PLATFORM) filter_options = header.FilterOptions(self._PLATFORM) @@ -288,23 +285,6 @@ def _TranslatePolicy(self, pol, exp_info): new_terms = [] for term in terms: - if term.expiration: - if term.expiration <= exp_info_date: - logging.info( - 'INFO: Term %s in policy %s expires ' 'in less than two weeks.', - term.name, - filter_name, - ) - - if term.expiration <= current_date: - logging.warning( - 'WARNING: Term %s in policy %s is expired and ' - 'will not be rendered.', - term.name, - filter_name, - ) - continue - new_terms.append(Term(term, filter_type, verbose)) self.aruba_policies.append((filter_name, new_terms, filter_type)) diff --git a/aerleon/lib/cisco.py b/aerleon/lib/cisco.py index 7d049c2c..26cefbb8 100644 --- a/aerleon/lib/cisco.py +++ b/aerleon/lib/cisco.py @@ -16,7 +16,6 @@ """Cisco generator.""" -import datetime import ipaddress from typing import Union, cast @@ -991,9 +990,6 @@ def _BuildTokens(self): def _TranslatePolicy(self, pol, exp_info): self.cisco_policies = [] - current_date = datetime.datetime.utcnow().date() - exp_info_date = current_date + datetime.timedelta(weeks=exp_info) - # a mixed filter outputs both ipv4 and ipv6 acls in the same output file good_filters = ['extended', 'standard', 'object-group', 'inet6', 'mixed', 'enable_dsmo'] @@ -1060,22 +1056,6 @@ def _TranslatePolicy(self, pol, exp_info): if term.restrict_address_family and term.restrict_address_family != af: continue - if term.expiration: - if term.expiration <= exp_info_date: - logging.info( - 'INFO: Term %s in policy %s expires ' 'in less than two weeks.', - term.name, - filter_name, - ) - if term.expiration <= current_date: - logging.warning( - 'WARNING: Term %s in policy %s is expired and ' - 'will not be rendered.', - term.name, - filter_name, - ) - continue - # render terms based on filter type if next_filter == 'standard': # keep track of sequence numbers across terms diff --git a/aerleon/lib/ciscoasa.py b/aerleon/lib/ciscoasa.py index b3b75d8c..8f7651e5 100644 --- a/aerleon/lib/ciscoasa.py +++ b/aerleon/lib/ciscoasa.py @@ -16,12 +16,12 @@ """Cisco ASA renderer.""" -import datetime import ipaddress -import logging import re from typing import cast +from absl import logging + from aerleon.lib import aclgenerator, cisco, nacaddr, summarizer _ACTION_TABLE = { @@ -335,9 +335,6 @@ def _BuildTokens(self): def _TranslatePolicy(self, pol, exp_info): self.ciscoasa_policies = [] - current_date = datetime.date.today() - exp_info_date = current_date + datetime.timedelta(weeks=exp_info) - for header, terms in self.policy.filters: filter_name = header.FilterName(self._PLATFORM) filter_options = header.FilterOptions(self._PLATFORM) @@ -346,22 +343,6 @@ def _TranslatePolicy(self, pol, exp_info): new_terms = [] # now add the terms for term in terms: - if term.expiration: - if term.expiration <= exp_info_date: - logging.info( - 'INFO: Term %s in policy %s expires ' 'in less than two weeks.', - term.name, - filter_name, - ) - if term.expiration <= current_date: - logging.warning( - 'WARNING: Term %s in policy %s is expired and ' - 'will not be rendered.', - term.name, - filter_name, - ) - continue - new_terms.append(str(Term(term, filter_name, enable_dsmo=enable_dsmo))) self.ciscoasa_policies.append((header, filter_name, new_terms)) diff --git a/aerleon/lib/cloudarmor.py b/aerleon/lib/cloudarmor.py index 56124009..96e3d840 100644 --- a/aerleon/lib/cloudarmor.py +++ b/aerleon/lib/cloudarmor.py @@ -9,7 +9,8 @@ import copy import json -import logging + +from absl import logging from aerleon.lib import aclgenerator diff --git a/aerleon/lib/demo.py b/aerleon/lib/demo.py index 914fb347..0a1a6195 100644 --- a/aerleon/lib/demo.py +++ b/aerleon/lib/demo.py @@ -15,8 +15,7 @@ # """Demo generator for Aerleon.""" -import datetime -import logging +from absl import logging from aerleon.lib import aclgenerator @@ -192,8 +191,6 @@ class Demo(aclgenerator.ACLGenerator): ) def _TranslatePolicy(self, pol, exp_info): - current_date = datetime.date.today() - exp_info_date = current_date + datetime.timedelta(weeks=exp_info) self.demo_policies = [] for header, terms in pol.filters: filter_options = header.FilterOptions('demo') @@ -209,21 +206,6 @@ def _TranslatePolicy(self, pol, exp_info): if term.name in term_names: raise DemoFilterError('Duplicate term name') term_names.add(term.name) - if term.expiration: - if term.expiration <= exp_info_date: - logging.info( - 'INFO: Term %s in policy %s expires ' 'in less than two weeks.', - term.name, - filter_name, - ) - if term.expiration <= current_date: - logging.warning( - 'WARNING: Term %s in policy %s is expired and ' - 'will not be rendered.', - term.name, - filter_name, - ) - continue new_terms.append(Term(term, filter_type)) self.demo_policies.append( (header, filter_name, filter_type, interface_specific, new_terms) diff --git a/aerleon/lib/gce.py b/aerleon/lib/gce.py index 3243dda8..5ea1f4ed 100644 --- a/aerleon/lib/gce.py +++ b/aerleon/lib/gce.py @@ -22,13 +22,13 @@ """ import copy -import datetime import ipaddress import json -import logging import re from typing import Any, Dict +from absl import logging + from aerleon.lib import gcp, nacaddr @@ -444,9 +444,6 @@ def _TranslatePolicy(self, pol, exp_info): total_attribute_count = 0 total_rule_count = 0 - current_date = datetime.datetime.utcnow().date() - exp_info_date = current_date + datetime.timedelta(weeks=exp_info) - for header, terms in pol.filters: filter_options = header.FilterOptions(self._PLATFORM) filter_name = header.FilterName(self._PLATFORM) @@ -479,6 +476,9 @@ def _TranslatePolicy(self, pol, exp_info): else: logging.warning('GCE filter does not specify a network.') + if not terms: + continue + term_names = set() if IsDefaultDeny(terms[-1]): terms[-1].protocol = ['all'] @@ -519,21 +519,6 @@ def _TranslatePolicy(self, pol, exp_info): term_names.add(term.name) term.direction = direction - if term.expiration: - if term.expiration <= exp_info_date: - logging.info( - 'INFO: Term %s in policy %s expires ' 'in less than two weeks.', - term.name, - filter_name, - ) - if term.expiration <= current_date: - logging.warning( - 'WARNING: Term %s in policy %s is expired and ' - 'will not be rendered.', - term.name, - filter_name, - ) - continue if term.option: raise GceFirewallError('GCE firewall does not support term options.') diff --git a/aerleon/lib/iptables.py b/aerleon/lib/iptables.py index 0a811ad6..3a189a1f 100644 --- a/aerleon/lib/iptables.py +++ b/aerleon/lib/iptables.py @@ -16,7 +16,6 @@ """Iptables generator.""" -import datetime import re from string import Template # pylint: disable=g-importing-member @@ -769,9 +768,6 @@ def _WarnIfCustomTarget(self, target): def _TranslatePolicy(self, pol, exp_info): """Translate a policy from objects into strings.""" - current_date = datetime.datetime.utcnow().date() - exp_info_date = current_date + datetime.timedelta(weeks=exp_info) - default_action = None good_default_actions = ['ACCEPT', 'DROP'] good_afs = ['inet', 'inet6'] @@ -869,22 +865,6 @@ def _TranslatePolicy(self, pol, exp_info): if not term: continue - if term.expiration: - if term.expiration <= exp_info_date: - logging.info( - 'INFO: Term %s in policy %s expires ' 'in less than two weeks.', - term.name, - filter_name, - ) - if term.expiration <= current_date: - logging.warning( - 'WARNING: Term %s in policy %s is expired and ' - 'will not be rendered.', - term.name, - filter_name, - ) - continue - new_terms.append( self._TERM( term, diff --git a/aerleon/lib/juniper.py b/aerleon/lib/juniper.py index 61f2f897..3fd85de1 100644 --- a/aerleon/lib/juniper.py +++ b/aerleon/lib/juniper.py @@ -16,7 +16,6 @@ """Juniper JCL generator.""" -import datetime from absl import logging @@ -981,9 +980,6 @@ def _BuildTokens(self): def _TranslatePolicy(self, pol, exp_info): self.juniper_policies = [] - current_date = datetime.datetime.utcnow().date() - exp_info_date = current_date + datetime.timedelta(weeks=exp_info) - for header, terms in pol.filters: filter_options = header.FilterOptions(self._PLATFORM) filter_name = header.FilterName(self._PLATFORM) @@ -1061,21 +1057,6 @@ def _TranslatePolicy(self, pol, exp_info): if not term: continue - if term.expiration: - if term.expiration <= exp_info_date: - logging.info( - 'INFO: Term %s in policy %s expires ' 'in less than two weeks.', - term.name, - filter_name, - ) - if term.expiration <= current_date: - logging.warning( - 'WARNING: Term %s in policy %s is expired and ' - 'will not be rendered.', - term.name, - filter_name, - ) - continue if 'is-fragment' in term.option and filter_type == 'inet6': raise JuniperFragmentInV6Error( 'The term %s uses "is-fragment" but ' 'is a v6 policy.' % term.name diff --git a/aerleon/lib/junipermsmpc.py b/aerleon/lib/junipermsmpc.py index d697643c..006e2fa2 100644 --- a/aerleon/lib/junipermsmpc.py +++ b/aerleon/lib/junipermsmpc.py @@ -15,8 +15,7 @@ # """Juniper MS-MPC generator for Aerleon.""" -import datetime -import logging +from absl import logging from aerleon.lib import aclgenerator, juniper, nacaddr @@ -513,8 +512,6 @@ def _GenerateApplications(self, filter_name): return [] def _TranslatePolicy(self, pol, exp_info): - current_date = datetime.date.today() - exp_info_date = current_date + datetime.timedelta(weeks=exp_info) self.junipermsmpc_policies = [] for header, terms in pol.filters: filter_options = header.FilterOptions(self._PLATFORM) @@ -597,21 +594,6 @@ def _TranslatePolicy(self, pol, exp_info): if term.name in term_names: raise JuniperMSMPCFilterError('Duplicate term name') term_names.add(term.name) - if term.expiration: - if term.expiration <= exp_info_date: - logging.info( - 'INFO: Term %s in policy %s expires ' 'in less than two weeks.', - term.name, - filter_name, - ) - if term.expiration <= current_date: - logging.warning( - 'WARNING: Term %s in policy %s is expired and ' - 'will not be rendered.', - term.name, - filter_name, - ) - continue new_term = Term(term, filter_type, noverbose, filter_name, apply_groups) new_terms.append(new_term) diff --git a/aerleon/lib/junipersrx.py b/aerleon/lib/junipersrx.py index faa39db9..c4a02a9a 100644 --- a/aerleon/lib/junipersrx.py +++ b/aerleon/lib/junipersrx.py @@ -20,7 +20,6 @@ import collections import copy -import datetime import heapq import ipaddress import itertools @@ -361,9 +360,6 @@ def _TranslatePolicy(self, pol, exp_info): ConflictingApplicationSetsError: When two duplicate named terms have conflicting application entries """ - current_date = datetime.datetime.utcnow().date() - exp_info_date = current_date + datetime.timedelta(weeks=exp_info) - for header, terms in pol.filters: filter_options = header.FilterOptions(self._PLATFORM) @@ -499,23 +495,6 @@ def _TranslatePolicy(self, pol, exp_info): raise SRXDuplicateTermError('You have a duplicate term: %s' % term.name) term_dup_check.add(term.name) - if term.expiration: - if term.expiration <= exp_info_date: - logging.info( - 'INFO: Term %s in policy %s>%s expires ' 'in less than two weeks.', - term.name, - self.from_zone, - self.to_zone, - ) - if term.expiration <= current_date: - logging.warning( - 'WARNING: Term %s in policy %s>%s is expired.', - term.name, - self.from_zone, - self.to_zone, - ) - continue - # SRX address books leverage network token names for IPs. # When excluding addresses, we lose those distinct names so we need # to create a new unique name based off the term name before excluding. diff --git a/aerleon/lib/k8s.py b/aerleon/lib/k8s.py index 932f7cca..df7f04ad 100644 --- a/aerleon/lib/k8s.py +++ b/aerleon/lib/k8s.py @@ -21,11 +21,10 @@ """ import copy -import datetime -import logging import re import yaml +from absl import logging from aerleon.lib import aclgenerator @@ -307,9 +306,6 @@ def _TranslatePolicy(self, pol, exp_info): self.network_policies = [] total_rule_count = 0 - current_date = datetime.datetime.utcnow().date() - exp_info_date = current_date + datetime.timedelta(weeks=exp_info) - for header, terms in pol.filters: filter_options = header.FilterOptions(self._PLATFORM) filter_name = header.FilterName(self._PLATFORM) @@ -341,21 +337,6 @@ def _TranslatePolicy(self, pol, exp_info): term_names.add(term.name) term.direction = direction - if term.expiration: - if term.expiration <= current_date: - logging.warning( - 'WARNING: Term %s in policy %s is expired and ' - 'will not be rendered.', - term.name, - filter_name, - ) - continue - if term.expiration <= exp_info_date: - logging.info( - 'INFO: Term %s in policy %s expires ' 'in less than two weeks.', - term.name, - filter_name, - ) if term.option: raise K8sNetworkPolicyError( 'Kubernetes NetworkPolicy does not support term options.' diff --git a/aerleon/lib/nftables.py b/aerleon/lib/nftables.py index 01dfb6ac..865958c9 100644 --- a/aerleon/lib/nftables.py +++ b/aerleon/lib/nftables.py @@ -17,8 +17,8 @@ import collections import copy -import datetime -import logging + +from absl import logging from aerleon.lib import aclgenerator, nacaddr @@ -641,8 +641,6 @@ def _TranslatePolicy(self, pol, exp_info): pol_counter = 0 - current_date = datetime.date.today() - exp_info_date = current_date + datetime.timedelta(weeks=exp_info) for header, terms in pol.filters: filter_options = header.FilterOptions('nftables') ( @@ -683,21 +681,6 @@ def _TranslatePolicy(self, pol, exp_info): term.name, ) continue - if term.expiration: - if term.expiration <= exp_info_date: - logging.info( - 'INFO: Term %s in policy %s expires ' 'in less than two weeks.', - term.name, - nf_af, - ) - if term.expiration <= current_date: - logging.warning( - 'WARNING: Term %s in policy %s is expired and ' - 'will not be rendered.', - term.name, - nf_af, - ) - continue # Handle address excludes before building nft address book dict. for i in term.source_address_exclude: term.source_address = nacaddr.RemoveAddressFromList(term.source_address, i) diff --git a/aerleon/lib/nsxv.py b/aerleon/lib/nsxv.py index 7f252feb..587b735f 100644 --- a/aerleon/lib/nsxv.py +++ b/aerleon/lib/nsxv.py @@ -16,7 +16,6 @@ """Nsxv generator.""" -import datetime import re import xml @@ -522,9 +521,6 @@ def _BuildTokens(self): def _TranslatePolicy(self, pol, exp_info): self.nsxv_policies = [] - current_date = datetime.datetime.utcnow().date() - exp_info_date = current_date + datetime.timedelta(weeks=exp_info) - for header, terms in pol.filters: filter_options = header.FilterOptions(self._PLATFORM) if len(filter_options) >= 2: @@ -544,21 +540,6 @@ def _TranslatePolicy(self, pol, exp_info): raise NsxvDuplicateTermError('There are multiple terms named: %s' % term.name) term_names.add(term.name) - if term.expiration: - if term.expiration <= exp_info_date: - logging.info( - 'INFO: Term %s in policy %s expires ' 'in less than two weeks.', - term.name, - filter_name, - ) - if term.expiration <= current_date: - logging.warning( - 'WARNING: Term %s in policy %s is expired and ' - 'will not be rendered.', - term.name, - filter_name, - ) - continue # Get the mapped action value # If there is no mapped action value term is not rendered mapped_action = _ACTION_TABLE.get(str(term.action[0])) diff --git a/aerleon/lib/openconfig.py b/aerleon/lib/openconfig.py index 54e661ab..e4616fdc 100644 --- a/aerleon/lib/openconfig.py +++ b/aerleon/lib/openconfig.py @@ -21,14 +21,14 @@ """ import copy -import datetime import ipaddress import json -import logging import re from collections import defaultdict from typing import Any, Dict +from absl import logging + from aerleon.lib import aclgenerator @@ -208,9 +208,6 @@ def _TranslatePolicy(self, pol, exp_info): self.oc_policies = [] total_rule_count = 0 - current_date = datetime.datetime.utcnow().date() - exp_info_date = current_date + datetime.timedelta(weeks=exp_info) - for header, terms in pol.filters: filter_options = header.FilterOptions(self._PLATFORM) filter_name = header.FilterName(self._PLATFORM) @@ -227,21 +224,6 @@ def _TranslatePolicy(self, pol, exp_info): for term in terms: - if term.expiration: - if term.expiration <= exp_info_date: - logging.info( - 'INFO: Term %s in policy %s expires ' 'in less than two weeks.', - term.name, - filter_name, - ) - if term.expiration <= current_date: - logging.warning( - 'WARNING: Term %s in policy %s is expired and ' - 'will not be rendered.', - term.name, - filter_name, - ) - continue # TODO(b/196430344): Add support for options such as # established/rst/first-fragment if term.option: diff --git a/aerleon/lib/packetfilter.py b/aerleon/lib/packetfilter.py index 6737c291..f0316539 100644 --- a/aerleon/lib/packetfilter.py +++ b/aerleon/lib/packetfilter.py @@ -18,7 +18,6 @@ import collections import copy -import datetime from typing import cast from absl import logging @@ -458,9 +457,6 @@ def _TranslatePolicy(self, pol, exp_info): self.pf_policies = [] self.address_book = {} self.def_short_to_long = {} - current_date = datetime.datetime.utcnow().date() - exp_info_date = current_date + datetime.timedelta(weeks=exp_info) - good_afs = ['inet', 'inet6', 'mixed'] good_options = ['in', 'out', 'nostate'] all_protocols_stateful = True @@ -573,22 +569,6 @@ def _TranslatePolicy(self, pol, exp_info): if not term: continue - if term.expiration: - if term.expiration <= exp_info_date: - logging.info( - 'INFO: Term %s in policy %s expires ' 'in less than two weeks.', - term.name, - filter_name, - ) - if term.expiration <= current_date: - logging.warning( - 'WARNING: Term %s in policy %s is expired and ' - 'will not be rendered.', - term.name, - filter_name, - ) - continue - if term.destination_interface and term.source_interface: raise Error( "packetfilter only supports destination interface XOR source interface per term" diff --git a/aerleon/lib/paloaltofw.py b/aerleon/lib/paloaltofw.py index 3c3c14e6..e89917f6 100644 --- a/aerleon/lib/paloaltofw.py +++ b/aerleon/lib/paloaltofw.py @@ -16,12 +16,12 @@ import collections import copy -import datetime -import logging import re import xml.etree.ElementTree as etree from xml.dom import minidom +from absl import logging + from aerleon.lib import aclgenerator, addressbook, nacaddr, policy @@ -390,8 +390,6 @@ def _TranslatePolicy(self, pol, exp_info): PaloAltoFWUnsupportedProtocolError: The term contains unsupporter protocol name. """ - current_date = datetime.date.today() - exp_info_date = current_date + datetime.timedelta(weeks=exp_info) first_addr_obj = None address_book_dup_check = set() @@ -514,24 +512,6 @@ def _TranslatePolicy(self, pol, exp_info): % (term.name, ', '.join(term.pan_application), ', '.join(term.protocol)) ) - if term.expiration: - if term.expiration <= exp_info_date: - logging.info( - "INFO: Term %s in policy %s>%s expires " "in less than two weeks.", - term.name, - self.from_zone, - self.to_zone, - ) - if term.expiration <= current_date: - logging.warning( - "WARNING: Term %s in policy %s>%s is expired and " - "will not be rendered.", - term.name, - self.from_zone, - self.to_zone, - ) - continue - for i in term.source_address_exclude: term.source_address = nacaddr.RemoveAddressFromList(term.source_address, i) for i in term.destination_address_exclude: diff --git a/aerleon/lib/pcap.py b/aerleon/lib/pcap.py index 1f7c00a8..01fea9d4 100644 --- a/aerleon/lib/pcap.py +++ b/aerleon/lib/pcap.py @@ -28,7 +28,6 @@ Stolen liberally from packetfilter.py. """ -import datetime from absl import logging @@ -375,9 +374,6 @@ def _BuildTokens(self): def _TranslatePolicy(self, pol, exp_info): self.pcap_policies = [] - current_date = datetime.datetime.utcnow().date() - exp_info_date = current_date + datetime.timedelta(weeks=exp_info) - good_afs = ['inet', 'inet6', 'mixed'] good_options = ['in', 'out'] direction = '' @@ -433,22 +429,6 @@ def _TranslatePolicy(self, pol, exp_info): 'You have a duplicate term: %s' % term.name ) - if term.expiration: - if term.expiration <= exp_info_date: - logging.info( - 'INFO: Term %s in policy %s expires ' 'in less than two weeks.', - term.name, - filter_name, - ) - if term.expiration <= current_date: - logging.warning( - 'WARNING: Term %s in policy %s is expired and ' - 'will not be rendered.', - term.name, - filter_name, - ) - continue - if not term: continue diff --git a/aerleon/lib/policy.py b/aerleon/lib/policy.py index 3e133005..097c7757 100644 --- a/aerleon/lib/policy.py +++ b/aerleon/lib/policy.py @@ -1690,7 +1690,13 @@ def FilterName(self, platform): for target in self.target: if target.platform == platform: if target.options: - return target.options[0] + if platform in ['srx', 'paloalto']: + if len(target.options) >= 3: + return '%s>%s' % (target.options[1], target.options[3]) + else: + return None + else: + return target.options[0] return None def __str__(self): diff --git a/aerleon/lib/windows.py b/aerleon/lib/windows.py index 55dcb932..89172d63 100644 --- a/aerleon/lib/windows.py +++ b/aerleon/lib/windows.py @@ -16,7 +16,6 @@ """Generic Windows security policy generator; requires subclassing.""" -import datetime import string from absl import logging @@ -267,9 +266,6 @@ def _BuildTokens(self): def _TranslatePolicy(self, pol, exp_info): """Translate a policy from objects into strings.""" self.windows_policies = [] - current_date = datetime.datetime.utcnow().date() - exp_info_date = current_date + datetime.timedelta(weeks=exp_info) - default_action = None good_default_actions = ['permit', 'block'] good_options = [] @@ -339,21 +335,6 @@ def _TranslatePolicy(self, pol, exp_info): ) term_names.add(term.name) - if term.expiration: - if term.expiration <= exp_info_date: - logging.info( - 'INFO: Term %s in policy %s expires ' 'in less than two weeks.', - term.name, - filter_name, - ) - if term.expiration <= current_date: - logging.warning( - 'WARNING: Term %s in policy %s is expired and ' - 'will not be rendered.', - term.name, - filter_name, - ) - continue if 'established' in term.option or 'tcp-established' in term.option: continue new_terms.append(self._TERM(term, filter_name, default_action, filter_type))

Hoist "expiration" impl into core Currently the implementation of the 'expiration' property is handled through a block of copy-paste code in every generator. Should a generator author forget to paste in this code, expiration won't work. The core application should handle the expiration implementation and not even pass it through to the generator.

2023-01-26T07:38:34

0.0

[]

aerleon/aerleon

aerleon__aerleon-192

44fba2be640eac226dc98a4af0996ce8d0ed2093

diff --git a/aerleon/lib/junipermsmpc.py b/aerleon/lib/junipermsmpc.py index eb338787..b6fbc1ee 100644 --- a/aerleon/lib/junipermsmpc.py +++ b/aerleon/lib/junipermsmpc.py @@ -34,6 +34,7 @@ class Term(juniper.Term): _PLATFORM = 'msmpc' _DEFAULT_INDENT = 20 + _NO_APPLY_GROUPS_INDENT = 12 _ACTIONS = {'accept': 'accept', 'deny': 'discard', 'reject': 'reject'} # msmpc supports a limited number of protocol names # https://www.juniper.net/documentation/us/en/software/junos/security-policies/topics/ref/statement/applications-edit-protocol.html @@ -55,12 +56,13 @@ class Term(juniper.Term): 'udp', ) - def __init__(self, term, term_type, noverbose, filter_name): + def __init__(self, term, term_type, noverbose, filter_name, apply_groups): enable_dsmo = False super().__init__(term, term_type, enable_dsmo, noverbose) self.term = term self.term_type = term_type self.noverbose = noverbose + self.apply_groups = apply_groups self.filter_name = filter_name for prot in self.term.protocol: @@ -81,7 +83,11 @@ def __str__(self): if self.enable_dsmo: raise NotImplementedError('enable_dsmo not implemented for msmpc') - ret_str = juniper.Config(indent=self._DEFAULT_INDENT) + if self.apply_groups: + indent = self._DEFAULT_INDENT + else: + indent = self._NO_APPLY_GROUPS_INDENT + ret_str = juniper.Config(indent=indent) # COMMENTS # this deals just fine with multi line comments, but we could probably @@ -516,11 +522,15 @@ def _TranslatePolicy(self, pol, exp_info): filter_options = header.FilterOptions(self._PLATFORM) filter_name = header.FilterName(self._PLATFORM) filter_options.remove(filter_name) + apply_groups = 'no-apply-groups' not in filter_options filter_direction = None filter_type = None noverbose = 'noverbose' in filter_options self.applications[filter_name] = [] + if not apply_groups: + filter_options.remove('no-apply-groups') + if noverbose: # noverbose is a strict boolean, remove it # prior to iterating through the other options @@ -604,7 +614,7 @@ def _TranslatePolicy(self, pol, exp_info): filter_name, ) continue - new_term = Term(term, filter_type, noverbose, filter_name) + new_term = Term(term, filter_type, noverbose, filter_name, apply_groups) new_terms.append(new_term) # Because MSMPC terms can contain inet and inet6 addresses. We have to @@ -660,7 +670,9 @@ def _TranslatePolicy(self, pol, exp_info): new_application_set['name'] = modified_term_name self.applications[filter_name].append(new_application_set) - self.junipermsmpc_policies.append((header, filter_name, filter_direction, new_terms)) + self.junipermsmpc_policies.append( + (header, filter_name, filter_direction, new_terms, apply_groups) + ) def _Group(self, group, lc=True): """If 1 item return it, else return [ item1 item2 ]. @@ -707,9 +719,16 @@ def _FormattedGroup(el, lc=True): def __str__(self): target = juniper.Config() - for (header, filter_name, filter_direction, terms) in self.junipermsmpc_policies: - target.Append('groups {') - target.Append('replace:') + for ( + header, + filter_name, + filter_direction, + terms, + apply_groups, + ) in self.junipermsmpc_policies: + if apply_groups: + target.Append('groups {') + target.Append('replace:') target.Append('/*') # we want the acl to contain id and date tags, but p4 will expand @@ -725,7 +744,8 @@ def __str__(self): target.Append('** ' + line) target.Append('*/') - target.Append('%s {' % filter_name) + if apply_groups: + target.Append('%s {' % filter_name) target.Append('services {') target.Append('stateful-firewall {') target.Append('rule %s {' % filter_name) @@ -739,9 +759,10 @@ def __str__(self): target.Append('}') # services { ... } for line in self._GenerateApplications(filter_name): target.Append(line) - target.Append('}') # filter_name { ... } - target.Append('}') # groups { ... } - target.Append('apply-groups %s;' % filter_name) + if apply_groups: + target.Append('}') # filter_name { ... } + target.Append('}') # groups { ... } + target.Append('apply-groups %s;' % filter_name) return str(target) + '\n' diff --git a/docs/reference/generators.md b/docs/reference/generators.md index bb9bf836..d27b4db6 100644 --- a/docs/reference/generators.md +++ b/docs/reference/generators.md @@ -820,7 +820,7 @@ The Juniper MSMPC header designation has the following format: ```yaml targets: - msmpc: [filter name] {inet|inet6|mixed} {noverbose} {ingress|egress} + msmpc: [filter name] {inet|inet6|mixed} {noverbose} {ingress|egress} [no-apply-groups] ``` * _filter name_: defines the name of the Juniper msmpc filter. @@ -829,6 +829,7 @@ targets: * _noverbose_: omit additional term and address comments. * _ingress_: filter will be applied in the input direction. * _egress_: filter will be appliced in the output direction. +* _no-apply-groups_: generate configuration without `apply-groups` (optional) When inet4 or inet6 is specified, naming tokens with both IPv4 and IPv6 filters will be rendered using only the specified addresses.

[LEGACY ISSUE PR307] New flag: --[no]apply_groups to enable/disable apply-groups creation for Juniper MS-MPC

There is a PR for this pending. We should invite them to contribute here. My comments for that PR is that these are not appropriate as flags and should be options for the platform itself.

2023-01-23T10:18:58

0.0

[]

ndif-team/nnsight

ndif-team__nnsight-179

da36d3cff8c6c4b490daf1e7451b3acb6010b3fd

diff --git a/src/nnsight/__init__.py b/src/nnsight/__init__.py index bde37c2e..a0dff340 100644 --- a/src/nnsight/__init__.py +++ b/src/nnsight/__init__.py @@ -33,10 +33,9 @@ for key, value in getmembers(math, isbuiltin): DEFAULT_PATCHER.add(Patch(math, proxy_wrapper(value), key)) -# TODO THis does not work. Because of accelerate also patching? because they are overloaded? -#DEFAULT_PATCHER.add(Patch(torch, proxy_wrapper(torch.zeros), "zeros")) -# DEFAULT_PATCHER.add(Patch(torch, proxy_wrapper(torch.ones), "ones")) -# DEFAULT_PATCHER.add(Patch(torch, proxy_wrapper(torch.rand), "rand")) +DEFAULT_PATCHER.add(Patch(torch, proxy_wrapper(torch.zeros), "zeros")) +DEFAULT_PATCHER.add(Patch(torch, proxy_wrapper(torch.ones), "ones")) +DEFAULT_PATCHER.add(Patch(torch, proxy_wrapper(torch.rand), "rand")) from torch._subclasses.fake_tensor import FakeTensor diff --git a/src/nnsight/intervention.py b/src/nnsight/intervention.py index e4655574..801930d9 100644 --- a/src/nnsight/intervention.py +++ b/src/nnsight/intervention.py @@ -107,6 +107,8 @@ def grad(self, value: Union[InterventionProxy, Any]) -> None: """ self.node.add(target="swap", args=[self.grad.node, value], value=True) + self.__dict__["_grad"] = None + def __call__(self, *args, **kwargs) -> Self: # We don't want to call backward on fake tensors @@ -148,7 +150,7 @@ def __setattr__( ) -> None: if key == "grad": - getattr(self.__class__, key).fset(self, value) + return getattr(self.__class__, key).fset(self, value) return super().__setattr__(key, value) diff --git a/src/nnsight/models/NNsightModel.py b/src/nnsight/models/NNsightModel.py index 32d32204..058772c4 100644 --- a/src/nnsight/models/NNsightModel.py +++ b/src/nnsight/models/NNsightModel.py @@ -66,9 +66,7 @@ def __init__( # Otherwise load from _load(...). if not self._custom_model: - # accelerate.init_empty_weights makes all parameters loaded on the 'meta' device. - # Also do .to('meta') because why not. - with accelerate.init_empty_weights(include_buffers=True): + with torch.device('meta'): self._model = self._load(self._model_key, *args, **kwargs).to("meta") self._envoy = Envoy(self._model) diff --git a/src/nnsight/pydantics/format/types.py b/src/nnsight/pydantics/format/types.py index 6692006e..55ffa724 100644 --- a/src/nnsight/pydantics/format/types.py +++ b/src/nnsight/pydantics/format/types.py @@ -86,6 +86,14 @@ def compile(self, graph: Graph, nodes: Dict[str, NodeModel]) -> slice: self.step.compile(graph, nodes), ) +class EllipsisModel(BaseModel): + model_config = ConfigDict(arbitrary_types_allowed=True) + type_name: Literal["ELLIPSIS"] = "ELLIPSIS" + + def compile(self, graph: Graph, nodes: Dict[str, NodeModel]) -> type(...): # It will be better to use EllipsisType, but it requires python>=3.10 + return ... + + class ListModel(BaseModel): model_config = ConfigDict(arbitrary_types_allowed=True) @@ -165,6 +173,14 @@ def compile(self, graph: Graph, nodes: Dict[str, NodeModel]) -> FUNCTION: ), ] +EllipsisType = Annotated[ + type(...), # It will be better to use EllipsisType, but it requires python>=3.10 + AfterValidator( + lambda value: EllipsisModel() + ), +] + + ListType = Annotated[list, AfterValidator(lambda value: ListModel(values=value))] TupleType = Annotated[ @@ -201,6 +217,7 @@ def compile(self, graph: Graph, nodes: Dict[str, NodeModel]) -> FUNCTION: ListModel, TupleModel, DictModel, + EllipsisModel ], Field(discriminator="type_name"), ], @@ -212,5 +229,6 @@ def compile(self, graph: Graph, nodes: Dict[str, NodeModel]) -> FUNCTION: ListType, TupleType, DictType, + EllipsisType ], ] diff --git a/src/nnsight/tracing/Node.py b/src/nnsight/tracing/Node.py index f1df651c..d70afc6a 100644 --- a/src/nnsight/tracing/Node.py +++ b/src/nnsight/tracing/Node.py @@ -2,7 +2,8 @@ import inspect import weakref -from typing import TYPE_CHECKING, Any, Callable, Dict, List, Optional, Tuple, Union +from typing import (TYPE_CHECKING, Any, Callable, Dict, List, Optional, Tuple, + Union) import torch from torch._subclasses.fake_tensor import FakeTensor @@ -45,8 +46,6 @@ def __init__( args = list() if kwargs is None: kwargs = dict() - - args = list(args) self.name = name self.graph: "Graph" = graph @@ -64,7 +63,9 @@ def __init__( # Resolve values from completed tracer/runner contexts self.args = util.apply(self.args, lambda x: x.value if x.done() else x, Node) - self.kwargs = util.apply(self.kwargs, lambda x: x.value if x.done() else x, Node) + self.kwargs = util.apply( + self.kwargs, lambda x: x.value if x.done() else x, Node + ) # Add all arguments that are nodes to nodes dependencies # (unless the arg is already .done(), for when you want to apply things to proxies after model execution?) @@ -187,8 +188,8 @@ def prepare_inputs( Returns: Any: Prepared inputs. """ - - inputs = util.apply(inputs, lambda x : x, object) + + inputs = util.apply(inputs, lambda x: x, inspect._empty) def _value(node: Proxy | Node): diff --git a/src/nnsight/tracing/Proxy.py b/src/nnsight/tracing/Proxy.py index bb0b4c08..3476755e 100755 --- a/src/nnsight/tracing/Proxy.py +++ b/src/nnsight/tracing/Proxy.py @@ -221,13 +221,19 @@ def __rtruediv__(self, other: Union[Proxy, Any]) -> Self: target=operator.truediv, args=[other, self.node], ) - + def __floordiv__(self, other: Union[Proxy, Any]) -> Self: return self.node.add( target=operator.floordiv, args=[self.node, other], ) + def __rfloordiv__(self, other: Union[Proxy, Any]) -> Self: + return self.node.add( + target=operator.floordiv, + args=[other, self.node], + ) + def __eq__(self, other: Union[Proxy, Any]) -> Self: return self.node.add(target=operator.eq, args=[self.node, other])

Meta device Loading parameters to the `'meta'` device & `torch.zeros()`, `torch.ones()`, and `torch.rand()` are now supported by nnsight via patching.

2024-07-23T20:11:40

0.0

[]

ndif-team/nnsight

ndif-team__nnsight-81

24253c42dbf19cc6a13235adfced6ba70cc97073

diff --git a/.gitignore b/.gitignore index 78373ef0..1e311508 100755 --- a/.gitignore +++ b/.gitignore @@ -3,4 +3,5 @@ dist/ public/ *.egg-info/ .installed.cfg +.vscode __pycache__/ diff --git a/.vscode/settings.json b/.vscode/settings.json deleted file mode 100644 index 45268e01..00000000 --- a/.vscode/settings.json +++ /dev/null @@ -1,5 +0,0 @@ -{ - "restructuredtext.syntaxHighlighting.disabled": true, - "restructuredtext.preview.docutils.disabled": true, - "esbonio.sphinx.confDir": "" -} \ No newline at end of file diff --git a/src/nnsight/intervention.py b/src/nnsight/intervention.py index 5977e68a..461e7908 100644 --- a/src/nnsight/intervention.py +++ b/src/nnsight/intervention.py @@ -162,6 +162,11 @@ def shape(self) -> Collection[torch.Size]: if not self.node.is_tracing(): return util.apply(self.value, lambda x: x.shape, torch.Tensor) + + # If we haven't scanned in a proxy_value, just return a proxy to get the attribute. + if self.node.proxy_value is None: + + return super().__getattr__('shape') return util.apply(self.node.proxy_value, lambda x: x.shape, torch.Tensor) @@ -176,6 +181,11 @@ def device(self) -> Collection[torch.device]: if not self.node.is_tracing(): return util.apply(self.value, lambda x: x.device, torch.Tensor) + + # If we haven't scanned in a proxy_value, just return a proxy to get the attribute. + if self.node.proxy_value is None: + + return super().__getattr__('device') return util.apply(self.node.proxy_value, lambda x: x.device, torch.Tensor) @@ -190,6 +200,11 @@ def dtype(self) -> Collection[torch.device]: if not self.node.is_tracing(): return util.apply(self.value, lambda x: x.dtype, torch.Tensor) + + # If we haven't scanned in a proxy_value, just return a proxy to get the attribute. + if self.node.proxy_value is None: + + return super().__getattr__('dtype') return util.apply(self.node.proxy_value, lambda x: x.dtype, torch.Tensor) diff --git a/src/nnsight/tracing/Graph.py b/src/nnsight/tracing/Graph.py index d8647434..184725b7 100644 --- a/src/nnsight/tracing/Graph.py +++ b/src/nnsight/tracing/Graph.py @@ -5,7 +5,7 @@ from typing import Any, Callable, Dict, List, Type, Union import torch -from torch._subclasses.fake_tensor import FakeTensorMode +from torch._subclasses.fake_tensor import FakeCopyMode, FakeTensorMode from torch.fx.experimental.symbolic_shapes import ShapeEnv from .. import util @@ -141,18 +141,13 @@ def add( allow_non_fake_inputs=True, shape_env=ShapeEnv(assume_static_by_default=True), ) as fake_mode: - - try: - + with FakeCopyMode(fake_mode): + value = target( *Node.prepare_proxy_values(_args), **Node.prepare_proxy_values(_kwargs), ) - except RuntimeError: - - value = None - target_name = target if isinstance(target, str) else target.__name__ if target_name not in self.name_idx: diff --git a/src/nnsight/tracing/Node.py b/src/nnsight/tracing/Node.py index aff0dd77..97530e6e 100644 --- a/src/nnsight/tracing/Node.py +++ b/src/nnsight/tracing/Node.py @@ -4,6 +4,7 @@ from typing import TYPE_CHECKING, Any, Callable, Dict, List, Optional, Tuple, Union import torch +from torch._subclasses.fake_tensor import FakeTensor from .. import util from ..logger import logger @@ -59,6 +60,22 @@ def slice_to_value(arg: slice): if torch._GLOBAL_DEVICE_CONTEXT is not None else None ) + + if device is None: + + # Arguments might be tensors created outside of scanning. Also the model might be a 'meta' pre-dispatched version of the model. + # That means the tensors as args and the model are different devices but we dont want to have to have the users move tensors to 'meta' + # So only when theres a FakeTensor with device meta, we move other tensors also to meta. + + def get_device(tensor:torch.Tensor): + + nonlocal device + + if device is None and isinstance(tensor, FakeTensor) and tensor.device.type == 'meta': + + device = tensor.device.type + + util.apply(values, get_device, torch.Tensor) if device is not None:

If theres no scanned in value for a Proxy and .shape, .dtype, or .de… …vice is called on it, return a Proxy of the operation, not the actual attribute on the proxy_value

2024-03-01T20:06:29

0.0

[]

ndif-team/nnsight

ndif-team__nnsight-77

0513025c6112b8b1512b5637a1cc7592f5730136

diff --git a/.DS_Store b/.DS_Store new file mode 100644 index 00000000..ecf0fd2f Binary files /dev/null and b/.DS_Store differ diff --git a/NNsight_v0_2.ipynb b/NNsight_v0_2.ipynb index 675ea875..b95e9a11 100644 --- a/NNsight_v0_2.ipynb +++ b/NNsight_v0_2.ipynb @@ -1,1191 +1,1627 @@ { - "nbformat": 4, - "nbformat_minor": 0, - "metadata": { - "colab": { - "provenance": [], - "toc_visible": true, - "include_colab_link": true + "cells": [ + { + "cell_type": "markdown", + "metadata": { + "colab_type": "text", + "id": "view-in-github" + }, + "source": [ + "<a href=\"https://colab.research.google.com/github/JadenFiotto-Kaufman/nnsight/blob/dev/NNsight_v0_2.ipynb\" target=\"_parent\"><img src=\"https://colab.research.google.com/assets/colab-badge.svg\" alt=\"Open In Colab\"/></a>" + ] }, - "kernelspec": { - "name": "python3", - "display_name": "Python 3" + { + "cell_type": "markdown", + "metadata": { + "id": "-Scfr942GwO4" + }, + "source": [ + "<img src=\"https://nnsight.net/_static/images/nnsight_logo.svg\" alt=\"drawing\" width=\"200\"/>\n", + "\n", + "# **NNsight**\n", + "## The API for a transparent science on black-box AI\n", + "\n", + "In this era of large-scale deep learning, the most interesting AI models are massive black boxes that are hard to run. Ordinary commercial inference service APIs let you interact with huge models, but they do not let you access model internals.\n", + "\n", + "The nnsight library is different: it gives you full access to all the neural network internals. When used together with a remote service like the [National Deep Inference Facility](https://thevisible.net/docs/NDIF-proposal.pdf) (NDIF), it lets you run complex experiments on huge open source models easily, with fully transparent access.\n", + "\n", + "Our team wants to enable entire labs and independent researchers alike, as we believe a large, passionate, and collaborative community will produce the next big insights on a profoundly important field.\n", + "\n", + "\n" + ] }, - "language_info": { - "name": "python" + { + "cell_type": "markdown", + "metadata": { + "id": "N1OemD2VGyZx" + }, + "source": [ + "# 1 First, let's start small\n", + "\n", + "\n", + "\n" + ] }, - "widgets": { - "application/vnd.jupyter.widget-state+json": { - "0a676b8f2dc34fed9731f19a03e4bfa1": { - "model_module": "@jupyter-widgets/controls", - "model_name": "HBoxModel", - "model_module_version": "1.5.0", - "state": { - "_dom_classes": [], - "_model_module": "@jupyter-widgets/controls", - "_model_module_version": "1.5.0", - "_model_name": "HBoxModel", - "_view_count": null, - "_view_module": "@jupyter-widgets/controls", - "_view_module_version": "1.5.0", - "_view_name": "HBoxView", - "box_style": "", - "children": [ - "IPY_MODEL_74358c341daa44faa41a92b0eb073615", - "IPY_MODEL_c40e17a994ed44d79ef54421f4772858", - "IPY_MODEL_431ffcb0dbb1419dbbb237a29842b905" - ], - "layout": "IPY_MODEL_c28c185a508e459b95edb29afedd9c79" - } - }, - "74358c341daa44faa41a92b0eb073615": { - "model_module": "@jupyter-widgets/controls", - "model_name": "HTMLModel", - "model_module_version": "1.5.0", - "state": { - "_dom_classes": [], - "_model_module": "@jupyter-widgets/controls", - "_model_module_version": "1.5.0", - "_model_name": "HTMLModel", - "_view_count": null, - "_view_module": "@jupyter-widgets/controls", - "_view_module_version": "1.5.0", - "_view_name": "HTMLView", - "description": "", - "description_tooltip": null, - "layout": "IPY_MODEL_7a0f3d2deeb14bbba7f0df208133a7e2", - "placeholder": "", - "style": "IPY_MODEL_94cdcf1fb0914dfe99c3c615e68b7ec8", - "value": "config.json: 100%" - } + { + "cell_type": "markdown", + "metadata": { + "id": "WyLMmhrAKTNM" + }, + "source": [ + "## The Tracing Context\n", + "\n", + "To demonstrate the core functionality and syntax of nnsight, we'll define and use a tiny two layer neural network." + ] + }, + { + "cell_type": "code", + "execution_count": 1, + "metadata": { + "id": "R0RJijD0eXwf" + }, + "outputs": [], + "source": [ + "# Install nnsight\n", + "!pip install nnsight\n", + "\n", + "from IPython.display import clear_output\n", + "\n", + "clear_output()" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "bgydw7i3HmIH" + }, + "source": [ + "Our little model here is composed of four sub-modules, two linear layers ('layer1', 'layer2'). We specify the sizes of each of these modules, and create some complementary example input." + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "id": "2pX2Wg8Ceo6N" + }, + "outputs": [], + "source": [ + "from collections import OrderedDict\n", + "import torch\n", + "\n", + "input_size = 5\n", + "hidden_dims = 10\n", + "output_size = 2\n", + "\n", + "net = torch.nn.Sequential(OrderedDict([\n", + " ('layer1', torch.nn.Linear(input_size, hidden_dims)),\n", + " ('layer2', torch.nn.Linear(hidden_dims, output_size)),\n", + "])).requires_grad_(False)\n", + "\n", + "input = torch.rand((1, input_size))" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "tIPa2h2pJIwl" + }, + "source": [ + "The core object of the nnsight package is `NNsight`. This wraps around a given pytorch model to enable the capabilites nnsight provides." + ] + }, + { + "cell_type": "code", + "execution_count": 2, + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/", + "height": 426, + "referenced_widgets": [ + "b3078a7a41c642839748d9864220cbde", + "3aa1aef302474a0cba55152e311acd40", + "dda1704b991a44dfb2b28191f307e081", + "5c0847c658d3489ca529d1afc8c96ed5", + "9fbbb0bf7d8e4c12939193791efc590f", + "ded9b4b26ce14ba0aae0b19e0274fcd0", + "47b21a1460f446d9911f2d06c9dd5bbd", + "21f5a81dd29549e69b5cfa2661371e35", + "5f608f4313d64e458314ace8237708b1", + "5266eb8eb83946e0acbda1362f3bd48a", + "8d7ef198c08b4a4ab49501e2204bdbb6" + ] }, - "c40e17a994ed44d79ef54421f4772858": { - "model_module": "@jupyter-widgets/controls", - "model_name": "FloatProgressModel", - "model_module_version": "1.5.0", - "state": { - "_dom_classes": [], - "_model_module": "@jupyter-widgets/controls", - "_model_module_version": "1.5.0", - "_model_name": "FloatProgressModel", - "_view_count": null, - "_view_module": "@jupyter-widgets/controls", - "_view_module_version": "1.5.0", - "_view_name": "ProgressView", - "bar_style": "success", - "description": "", - "description_tooltip": null, - "layout": "IPY_MODEL_f3e24aa6619e4185a3e2f53f2e473e45", - "max": 665, - "min": 0, - "orientation": "horizontal", - "style": "IPY_MODEL_b7be20bd738940658f710309729d8ccd", - "value": 665 - } + "id": "8H9R_ynTJI5y", + "outputId": "5d6b545d-750e-441a-fde9-2dc53e2f1f5e" + }, + "outputs": [ + { + "name": "stderr", + "output_type": "stream", + "text": [ + "The cache for model files in Transformers v4.22.0 has been updated. Migrating your old cache. This is a one-time only operation. You can interrupt this and resume the migration later on by calling `transformers.utils.move_cache()`.\n" + ] }, - "431ffcb0dbb1419dbbb237a29842b905": { - "model_module": "@jupyter-widgets/controls", - "model_name": "HTMLModel", - "model_module_version": "1.5.0", - "state": { - "_dom_classes": [], - "_model_module": "@jupyter-widgets/controls", - "_model_module_version": "1.5.0", - "_model_name": "HTMLModel", - "_view_count": null, - "_view_module": "@jupyter-widgets/controls", - "_view_module_version": "1.5.0", - "_view_name": "HTMLView", - "description": "", - "description_tooltip": null, - "layout": "IPY_MODEL_562a287082eb4882af864062ae13c91b", - "placeholder": "", - "style": "IPY_MODEL_11e99924bdf84739b3cfc7d676b797e5", - "value": " 665/665 [00:00<00:00, 12.8kB/s]" - } + { + "data": { + "application/vnd.jupyter.widget-view+json": { + "model_id": "b3078a7a41c642839748d9864220cbde", + "version_major": 2, + "version_minor": 0 + }, + "text/plain": [ + "0it [00:00, ?it/s]" + ] + }, + "metadata": {}, + "output_type": "display_data" }, - "c28c185a508e459b95edb29afedd9c79": { - "model_module": "@jupyter-widgets/base", - "model_name": "LayoutModel", - "model_module_version": "1.2.0", - "state": { - "_model_module": "@jupyter-widgets/base", - "_model_module_version": "1.2.0", - "_model_name": "LayoutModel", - "_view_count": null, - "_view_module": "@jupyter-widgets/base", - "_view_module_version": "1.2.0", - "_view_name": "LayoutView", - "align_content": null, - "align_items": null, - "align_self": null, - "border": null, - "bottom": null, - "display": null, - "flex": null, - "flex_flow": null, - "grid_area": null, - "grid_auto_columns": null, - "grid_auto_flow": null, - "grid_auto_rows": null, - "grid_column": null, - "grid_gap": null, - "grid_row": null, - "grid_template_areas": null, - "grid_template_columns": null, - "grid_template_rows": null, - "height": null, - "justify_content": null, - "justify_items": null, - "left": null, - "margin": null, - "max_height": null, - "max_width": null, - "min_height": null, - "min_width": null, - "object_fit": null, - "object_position": null, - "order": null, - "overflow": null, - "overflow_x": null, - "overflow_y": null, - "padding": null, - "right": null, - "top": null, - "visibility": null, - "width": null - } - }, - "7a0f3d2deeb14bbba7f0df208133a7e2": { - "model_module": "@jupyter-widgets/base", - "model_name": "LayoutModel", - "model_module_version": "1.2.0", - "state": { - "_model_module": "@jupyter-widgets/base", - "_model_module_version": "1.2.0", - "_model_name": "LayoutModel", - "_view_count": null, - "_view_module": "@jupyter-widgets/base", - "_view_module_version": "1.2.0", - "_view_name": "LayoutView", - "align_content": null, - "align_items": null, - "align_self": null, - "border": null, - "bottom": null, - "display": null, - "flex": null, - "flex_flow": null, - "grid_area": null, - "grid_auto_columns": null, - "grid_auto_flow": null, - "grid_auto_rows": null, - "grid_column": null, - "grid_gap": null, - "grid_row": null, - "grid_template_areas": null, - "grid_template_columns": null, - "grid_template_rows": null, - "height": null, - "justify_content": null, - "justify_items": null, - "left": null, - "margin": null, - "max_height": null, - "max_width": null, - "min_height": null, - "min_width": null, - "object_fit": null, - "object_position": null, - "order": null, - "overflow": null, - "overflow_x": null, - "overflow_y": null, - "padding": null, - "right": null, - "top": null, - "visibility": null, - "width": null - } - }, - "94cdcf1fb0914dfe99c3c615e68b7ec8": { - "model_module": "@jupyter-widgets/controls", - "model_name": "DescriptionStyleModel", - "model_module_version": "1.5.0", - "state": { - "_model_module": "@jupyter-widgets/controls", - "_model_module_version": "1.5.0", - "_model_name": "DescriptionStyleModel", - "_view_count": null, - "_view_module": "@jupyter-widgets/base", - "_view_module_version": "1.2.0", - "_view_name": "StyleView", - "description_width": "" - } - }, - "f3e24aa6619e4185a3e2f53f2e473e45": { - "model_module": "@jupyter-widgets/base", - "model_name": "LayoutModel", - "model_module_version": "1.2.0", - "state": { - "_model_module": "@jupyter-widgets/base", - "_model_module_version": "1.2.0", - "_model_name": "LayoutModel", - "_view_count": null, - "_view_module": "@jupyter-widgets/base", - "_view_module_version": "1.2.0", - "_view_name": "LayoutView", - "align_content": null, - "align_items": null, - "align_self": null, - "border": null, - "bottom": null, - "display": null, - "flex": null, - "flex_flow": null, - "grid_area": null, - "grid_auto_columns": null, - "grid_auto_flow": null, - "grid_auto_rows": null, - "grid_column": null, - "grid_gap": null, - "grid_row": null, - "grid_template_areas": null, - "grid_template_columns": null, - "grid_template_rows": null, - "height": null, - "justify_content": null, - "justify_items": null, - "left": null, - "margin": null, - "max_height": null, - "max_width": null, - "min_height": null, - "min_width": null, - "object_fit": null, - "object_position": null, - "order": null, - "overflow": null, - "overflow_x": null, - "overflow_y": null, - "padding": null, - "right": null, - "top": null, - "visibility": null, - "width": null - } - }, - "b7be20bd738940658f710309729d8ccd": { - "model_module": "@jupyter-widgets/controls", - "model_name": "ProgressStyleModel", - "model_module_version": "1.5.0", - "state": { - "_model_module": "@jupyter-widgets/controls", - "_model_module_version": "1.5.0", - "_model_name": "ProgressStyleModel", - "_view_count": null, - "_view_module": "@jupyter-widgets/base", - "_view_module_version": "1.2.0", - "_view_name": "StyleView", - "bar_color": null, - "description_width": "" - } - }, - "562a287082eb4882af864062ae13c91b": { - "model_module": "@jupyter-widgets/base", - "model_name": "LayoutModel", - "model_module_version": "1.2.0", - "state": { - "_model_module": "@jupyter-widgets/base", - "_model_module_version": "1.2.0", - "_model_name": "LayoutModel", - "_view_count": null, - "_view_module": "@jupyter-widgets/base", - "_view_module_version": "1.2.0", - "_view_name": "LayoutView", - "align_content": null, - "align_items": null, - "align_self": null, - "border": null, - "bottom": null, - "display": null, - "flex": null, - "flex_flow": null, - "grid_area": null, - "grid_auto_columns": null, - "grid_auto_flow": null, - "grid_auto_rows": null, - "grid_column": null, - "grid_gap": null, - "grid_row": null, - "grid_template_areas": null, - "grid_template_columns": null, - "grid_template_rows": null, - "height": null, - "justify_content": null, - "justify_items": null, - "left": null, - "margin": null, - "max_height": null, - "max_width": null, - "min_height": null, - "min_width": null, - "object_fit": null, - "object_position": null, - "order": null, - "overflow": null, - "overflow_x": null, - "overflow_y": null, - "padding": null, - "right": null, - "top": null, - "visibility": null, - "width": null - } - }, - "11e99924bdf84739b3cfc7d676b797e5": { - "model_module": "@jupyter-widgets/controls", - "model_name": "DescriptionStyleModel", - "model_module_version": "1.5.0", - "state": { - "_model_module": "@jupyter-widgets/controls", - "_model_module_version": "1.5.0", - "_model_name": "DescriptionStyleModel", - "_view_count": null, - "_view_module": "@jupyter-widgets/base", - "_view_module_version": "1.2.0", - "_view_name": "StyleView", - "description_width": "" - } - }, - "60fe026e908741209637604edfeb0563": { - "model_module": "@jupyter-widgets/controls", - "model_name": "HBoxModel", - "model_module_version": "1.5.0", - "state": { - "_dom_classes": [], - "_model_module": "@jupyter-widgets/controls", - "_model_module_version": "1.5.0", - "_model_name": "HBoxModel", - "_view_count": null, - "_view_module": "@jupyter-widgets/controls", - "_view_module_version": "1.5.0", - "_view_name": "HBoxView", - "box_style": "", - "children": [ - "IPY_MODEL_e7a73cfc3fa04425afc08368532ef0be", - "IPY_MODEL_d839da1c31a34659988e3420ba520725", - "IPY_MODEL_96a3b8c765374a5e98eaf713dd5e85d8" - ], - "layout": "IPY_MODEL_b49ce0d00e5b4df097b3a9fc4c0bd8e7" - } - }, - "e7a73cfc3fa04425afc08368532ef0be": { - "model_module": "@jupyter-widgets/controls", - "model_name": "HTMLModel", - "model_module_version": "1.5.0", - "state": { - "_dom_classes": [], - "_model_module": "@jupyter-widgets/controls", - "_model_module_version": "1.5.0", - "_model_name": "HTMLModel", - "_view_count": null, - "_view_module": "@jupyter-widgets/controls", - "_view_module_version": "1.5.0", - "_view_name": "HTMLView", - "description": "", - "description_tooltip": null, - "layout": "IPY_MODEL_b2748df9aff540df954f002310606a56", - "placeholder": "", - "style": "IPY_MODEL_f4caf2bc58704464b6ac7bf6da31b71d", - "value": "vocab.json: 100%" - } - }, - "d839da1c31a34659988e3420ba520725": { - "model_module": "@jupyter-widgets/controls", - "model_name": "FloatProgressModel", - "model_module_version": "1.5.0", - "state": { - "_dom_classes": [], - "_model_module": "@jupyter-widgets/controls", - "_model_module_version": "1.5.0", - "_model_name": "FloatProgressModel", - "_view_count": null, - "_view_module": "@jupyter-widgets/controls", - "_view_module_version": "1.5.0", - "_view_name": "ProgressView", - "bar_style": "success", - "description": "", - "description_tooltip": null, - "layout": "IPY_MODEL_3731a5747724403e86ef1abaf4fb389b", - "max": 1042301, - "min": 0, - "orientation": "horizontal", - "style": "IPY_MODEL_688bf676958a4050b73cc9597f5110de", - "value": 1042301 - } - }, - "96a3b8c765374a5e98eaf713dd5e85d8": { - "model_module": "@jupyter-widgets/controls", - "model_name": "HTMLModel", - "model_module_version": "1.5.0", - "state": { - "_dom_classes": [], - "_model_module": "@jupyter-widgets/controls", - "_model_module_version": "1.5.0", - "_model_name": "HTMLModel", - "_view_count": null, - "_view_module": "@jupyter-widgets/controls", - "_view_module_version": "1.5.0", - "_view_name": "HTMLView", - "description": "", - "description_tooltip": null, - "layout": "IPY_MODEL_8776d04bfd3748b69e10708466e9c132", - "placeholder": "", - "style": "IPY_MODEL_56490c20aef04fe4a63ba7c990e71e77", - "value": " 1.04M/1.04M [00:00<00:00, 3.16MB/s]" - } + { + "ename": "ImportError", + "evalue": "cannot import name 'NNsight' from 'nnsight' (/usr/local/lib/python3.10/dist-packages/nnsight/__init__.py)", + "output_type": "error", + "traceback": [ + "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m", + "\u001b[0;31mImportError\u001b[0m Traceback (most recent call last)", + "\u001b[0;32m<ipython-input-2-801ad166c2ee>\u001b[0m in \u001b[0;36m<cell line: 1>\u001b[0;34m()\u001b[0m\n\u001b[0;32m----> 1\u001b[0;31m \u001b[0;32mfrom\u001b[0m \u001b[0mnnsight\u001b[0m \u001b[0;32mimport\u001b[0m \u001b[0mNNsight\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m 2\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 3\u001b[0m \u001b[0mmodel\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mNNsight\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mnet\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n", + "\u001b[0;31mImportError\u001b[0m: cannot import name 'NNsight' from 'nnsight' (/usr/local/lib/python3.10/dist-packages/nnsight/__init__.py)", + "", + "\u001b[0;31m---------------------------------------------------------------------------\u001b[0;32m\nNOTE: If your import is failing due to a missing package, you can\nmanually install dependencies using either !pip or !apt.\n\nTo view examples of installing some common dependencies, click the\n\"Open Examples\" button below.\n\u001b[0;31m---------------------------------------------------------------------------\u001b[0m\n" + ] + } + ], + "source": [ + "from nnsight import NNsight\n", + "\n", + "model = NNsight(net)" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "2NfISlQ_Ilvp" + }, + "source": [ + "Printing a Pytorch model shows a named hierarchy of modules which is very useful when accessing sub-components directly. NNsight models work the same." + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/" }, - "b49ce0d00e5b4df097b3a9fc4c0bd8e7": { - "model_module": "@jupyter-widgets/base", - "model_name": "LayoutModel", - "model_module_version": "1.2.0", - "state": { - "_model_module": "@jupyter-widgets/base", - "_model_module_version": "1.2.0", - "_model_name": "LayoutModel", - "_view_count": null, - "_view_module": "@jupyter-widgets/base", - "_view_module_version": "1.2.0", - "_view_name": "LayoutView", - "align_content": null, - "align_items": null, - "align_self": null, - "border": null, - "bottom": null, - "display": null, - "flex": null, - "flex_flow": null, - "grid_area": null, - "grid_auto_columns": null, - "grid_auto_flow": null, - "grid_auto_rows": null, - "grid_column": null, - "grid_gap": null, - "grid_row": null, - "grid_template_areas": null, - "grid_template_columns": null, - "grid_template_rows": null, - "height": null, - "justify_content": null, - "justify_items": null, - "left": null, - "margin": null, - "max_height": null, - "max_width": null, - "min_height": null, - "min_width": null, - "object_fit": null, - "object_position": null, - "order": null, - "overflow": null, - "overflow_x": null, - "overflow_y": null, - "padding": null, - "right": null, - "top": null, - "visibility": null, - "width": null - } + "id": "bYtnbJHvlGZV", + "outputId": "5eb7c572-451c-43c6-956f-af224ae1867c" + }, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "Sequential(\n", + " (layer1): Linear(in_features=5, out_features=10, bias=True)\n", + " (layer2): Linear(in_features=10, out_features=2, bias=True)\n", + ")\n" + ] + } + ], + "source": [ + "print(model)" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "djC5kyJWLuUH" + }, + "source": [ + "Before we actually get to using the model we just created, let's talk about Python contexts.\n", + "\n", + "Python contexts define a scope using the `with` statement and are often used to create some object, or initiate some logic, that you later want to destroy or conclude.\n", + "\n", + "The most common application is opening files like the following example:\n", + "\n", + "```python\n", + "with open('myfile.txt', 'r') as file:\n", + " text = file.read()\n", + "```\n", + "\n", + "Python uses the `with` keyword to enter a context-like object. This object defines logic to be run at the start of the `with` block, as well as logic to be run when exiting. When using `with` for a file, entering the context opens the file and exiting the context closes it. Being within the context means we can read from the file. Simple enough! Now we can discuss how `nnsight` uses contexts to enable intuitive access into the internals of a neural network.\n" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "iNvuCOeyojcA" + }, + "source": [ + "The main tool with `nnsight` is a context for tracing.\n", + "\n", + "We enter the tracing context by calling `model.trace(<input>)` on an `NNsight` model, which defines how we want to run the model. Inside the context, we will be able to customize how the neural network runs. The model is actually run upon exiting the tracing context." + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "id": "qEXQ4auPSL-m" + }, + "outputs": [], + "source": [ + "with model.trace(input) as tracer:\n", + " pass" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "UZQsHinjqicJ" + }, + "source": [ + "But where's the output? To get that, we'll have to learn how to request it from within the tracing conext." + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "xMfBpYzDPMoB" + }, + "source": [ + "## Getting" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "B5_aFwFRv0ax" + }, + "source": [ + "Earlier, when we wrapped our little neural net with the `NNsight` class. This added a couple properties to each module in the model (including the root model itself). The two most important ones are `.input` and `.output`.\n", + "\n", + "```python\n", + "model.input\n", + "model.output\n", + "```\n", + "\n", + "The names are self explainatory. They correspond to the inputs and outputs of their respective modules during a forward pass of the model. We can use these attributes inside the `with` block.\n", + "\n", + "However, it is important to understand that the model is not executed until the end of the tracing context. How can we access inputs and outputs before the model is run? The trick is deferred execution.\n", + "\n", + "`.input` and `.output` are Proxies for the eventual inputs and outputs of a module. In other words, when you access `model.output` what you are communicating to `nnsight` is, \"When you compute the output of `model`, please grab it for me and put the value into its corresponding Proxy object's `.value` attribute.\" Let's try it:" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/", + "height": 303 }, - "b2748df9aff540df954f002310606a56": { - "model_module": "@jupyter-widgets/base", - "model_name": "LayoutModel", - "model_module_version": "1.2.0", - "state": { - "_model_module": "@jupyter-widgets/base", - "_model_module_version": "1.2.0", - "_model_name": "LayoutModel", - "_view_count": null, - "_view_module": "@jupyter-widgets/base", - "_view_module_version": "1.2.0", - "_view_name": "LayoutView", - "align_content": null, - "align_items": null, - "align_self": null, - "border": null, - "bottom": null, - "display": null, - "flex": null, - "flex_flow": null, - "grid_area": null, - "grid_auto_columns": null, - "grid_auto_flow": null, - "grid_auto_rows": null, - "grid_column": null, - "grid_gap": null, - "grid_row": null, - "grid_template_areas": null, - "grid_template_columns": null, - "grid_template_rows": null, - "height": null, - "justify_content": null, - "justify_items": null, - "left": null, - "margin": null, - "max_height": null, - "max_width": null, - "min_height": null, - "min_width": null, - "object_fit": null, - "object_position": null, - "order": null, - "overflow": null, - "overflow_x": null, - "overflow_y": null, - "padding": null, - "right": null, - "top": null, - "visibility": null, - "width": null - } + "id": "cYe8-r9ptGaG", + "outputId": "adf25efe-d2cf-4f65-9eaf-15b155762f4b" + }, + "outputs": [ + { + "ename": "ValueError", + "evalue": "Accessing Proxy value before it's been set.", + "output_type": "error", + "traceback": [ + "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m", + "\u001b[0;31mValueError\u001b[0m Traceback (most recent call last)", + "\u001b[0;32m<ipython-input-34-c7e0c74b12fa>\u001b[0m in \u001b[0;36m<cell line: 5>\u001b[0;34m()\u001b[0m\n\u001b[1;32m 3\u001b[0m \u001b[0moutput\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mmodel\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0moutput\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 4\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m----> 5\u001b[0;31m \u001b[0mprint\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0moutput\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mvalue\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m", + "\u001b[0;32m/usr/local/lib/python3.10/dist-packages/nnsight/tracing/Proxy.py\u001b[0m in \u001b[0;36mvalue\u001b[0;34m(self)\u001b[0m\n\u001b[1;32m 47\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 48\u001b[0m \u001b[0;32mif\u001b[0m \u001b[0;32mnot\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mnode\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mdone\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m---> 49\u001b[0;31m \u001b[0;32mraise\u001b[0m \u001b[0mValueError\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m\"Accessing Proxy value before it's been set.\"\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m 50\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 51\u001b[0m \u001b[0;32mreturn\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mnode\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mvalue\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n", + "\u001b[0;31mValueError\u001b[0m: Accessing Proxy value before it's been set." + ] + } + ], + "source": [ + "with model.trace(input) as tracer:\n", + "\n", + " output = model.output\n", + "\n", + "print(output.value)" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "sBz5qfwuR-6F" + }, + "source": [ + "Oh no an error! \"Accessing Proxy value before it's been set.\"\n", + "\n", + "Why doesn't our `output` have a `value`?\n", + "\n", + "Proxy objects will only have their value at the end of a context if we call `.save()` on them. This helps to reduce memory costs. Adding `.save()` fixes the error:" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/" }, - "f4caf2bc58704464b6ac7bf6da31b71d": { - "model_module": "@jupyter-widgets/controls", - "model_name": "DescriptionStyleModel", - "model_module_version": "1.5.0", - "state": { - "_model_module": "@jupyter-widgets/controls", - "_model_module_version": "1.5.0", - "_model_name": "DescriptionStyleModel", - "_view_count": null, - "_view_module": "@jupyter-widgets/base", - "_view_module_version": "1.2.0", - "_view_name": "StyleView", - "description_width": "" - } + "id": "y_bXRd5dvsBu", + "outputId": "a8e8985c-2661-46a8-975b-d43f67cfc162" + }, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "tensor([[ 0.1473, -0.1518]])\n" + ] + } + ], + "source": [ + "with model.trace(input) as tracer:\n", + "\n", + " output = model.output.save()\n", + "\n", + "print(output.value)" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "t5C0UZCwvrYn" + }, + "source": [ + "Success! We now have the model output. You just completed your first intervention using `nnsight`.\n", + "\n", + "\n", + "Each time you access a module's input or output, you create an *intervention* in the neural network's forward pass. Collectively these requests form the *intervention graph*. We call the process of executing it alongside the model's normal computation graph, *interleaving*.\n", + "\n", + "<details>\n", + "<summary>On Model output</summary>\n", + "\n", + "---\n", + "\n", + "If you don't need to access anything other than the final model output, you can call the tracing context with `trace=False` and not use it as a context:\n", + "\n", + "```python\n", + "output = model.trace(<inputs>, trace=False)\n", + "```\n", + "\n", + "---\n", + "\n", + "</details>\n", + "\n", + "Just like we saved the output of the model as a whole, we can save the output of any of its submodules. We use normal Python attribute syntax. We can discover how to access them by name by printing out the model:" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/" }, - "3731a5747724403e86ef1abaf4fb389b": { - "model_module": "@jupyter-widgets/base", - "model_name": "LayoutModel", - "model_module_version": "1.2.0", - "state": { - "_model_module": "@jupyter-widgets/base", - "_model_module_version": "1.2.0", - "_model_name": "LayoutModel", - "_view_count": null, - "_view_module": "@jupyter-widgets/base", - "_view_module_version": "1.2.0", - "_view_name": "LayoutView", - "align_content": null, - "align_items": null, - "align_self": null, - "border": null, - "bottom": null, - "display": null, - "flex": null, - "flex_flow": null, - "grid_area": null, - "grid_auto_columns": null, - "grid_auto_flow": null, - "grid_auto_rows": null, - "grid_column": null, - "grid_gap": null, - "grid_row": null, - "grid_template_areas": null, - "grid_template_columns": null, - "grid_template_rows": null, - "height": null, - "justify_content": null, - "justify_items": null, - "left": null, - "margin": null, - "max_height": null, - "max_width": null, - "min_height": null, - "min_width": null, - "object_fit": null, - "object_position": null, - "order": null, - "overflow": null, - "overflow_x": null, - "overflow_y": null, - "padding": null, - "right": null, - "top": null, - "visibility": null, - "width": null - } + "id": "0WcVUSP-0CJi", + "outputId": "7fd9d6ac-152d-4431-9a4d-0b60321638eb" + }, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "Sequential(\n", + " (layer1): Linear(in_features=5, out_features=10, bias=True)\n", + " (layer2): Linear(in_features=10, out_features=2, bias=True)\n", + ")\n" + ] + } + ], + "source": [ + "print(model)" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/" }, - "688bf676958a4050b73cc9597f5110de": { - "model_module": "@jupyter-widgets/controls", - "model_name": "ProgressStyleModel", - "model_module_version": "1.5.0", - "state": { - "_model_module": "@jupyter-widgets/controls", - "_model_module_version": "1.5.0", - "_model_name": "ProgressStyleModel", - "_view_count": null, - "_view_module": "@jupyter-widgets/base", - "_view_module_version": "1.2.0", - "_view_name": "StyleView", - "bar_color": null, - "description_width": "" - } + "id": "akWr-cNqy-9O", + "outputId": "68b9e071-c609-4b70-906d-2f2cd93b4495" + }, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "tensor([[ 0.0458, 0.5267, 0.7119, 0.4046, 0.2460, 0.7998, 0.4485, -0.2506,\n", + " 0.2968, -0.8834]])\n" + ] + } + ], + "source": [ + "with model.trace(input) as tracer:\n", + "\n", + " l1_output = model.layer1.output.save()\n", + "\n", + "print(l1_output.value)" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "85A-aP_03ht6" + }, + "source": [ + "Let's do the same for the input of layer2. While we're at it, let's also drop the `as tracer`, as we won't be needing the tracer object itself for a few sections:" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/" + }, + "id": "3EHEN38N3nXR", + "outputId": "5afd17a7-0bcc-4d1a-c131-7038bde9455d" + }, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "((tensor([[ 0.0458, 0.5267, 0.7119, 0.4046, 0.2460, 0.7998, 0.4485, -0.2506,\n", + " 0.2968, -0.8834]]),), {})\n" + ] + } + ], + "source": [ + "with model.trace(input):\n", + "\n", + " l2_input = model.layer2.input.save()\n", + "\n", + "print(l2_input.value)" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "jk-U8zi33Gi-" + }, + "source": [ + "<details>\n", + " <summary>On module inputs</summary>\n", + "\n", + " ---\n", + "\n", + " Notice how the value for `l2_input`, was not just a single tensor.\n", + " The type/shape of values from `.input` is in the form of:\n", + "\n", + " tuple(tuple(args), dictionary(kwargs))\n", + "\n", + " Where the first index of the tuple is itself a tuple of all positional arguments, and the second index is a dictionary of the keyword arguments.\n", + "\n", + " ---\n", + "\n", + "</details>\n", + "\n", + "\n", + "Now that we can access activations, we also want to do some post-processing on it. Let's find out which dimension of layer1's output has the highest value." + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "p7Xo_PHyPr4p" + }, + "source": [ + "## Functions, Methods, and Operations" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "ehSofWbx5DSx" + }, + "source": [ + "We could do this by calling `torch.argmax(...)` after the tracing context or we can just leverage the fact that `nnsight` handles functions and methods within the tracing context, by creating a Proxy request for it:" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/" }, - "8776d04bfd3748b69e10708466e9c132": { - "model_module": "@jupyter-widgets/base", - "model_name": "LayoutModel", - "model_module_version": "1.2.0", - "state": { - "_model_module": "@jupyter-widgets/base", - "_model_module_version": "1.2.0", - "_model_name": "LayoutModel", - "_view_count": null, - "_view_module": "@jupyter-widgets/base", - "_view_module_version": "1.2.0", - "_view_name": "LayoutView", - "align_content": null, - "align_items": null, - "align_self": null, - "border": null, - "bottom": null, - "display": null, - "flex": null, - "flex_flow": null, - "grid_area": null, - "grid_auto_columns": null, - "grid_auto_flow": null, - "grid_auto_rows": null, - "grid_column": null, - "grid_gap": null, - "grid_row": null, - "grid_template_areas": null, - "grid_template_columns": null, - "grid_template_rows": null, - "height": null, - "justify_content": null, - "justify_items": null, - "left": null, - "margin": null, - "max_height": null, - "max_width": null, - "min_height": null, - "min_width": null, - "object_fit": null, - "object_position": null, - "order": null, - "overflow": null, - "overflow_x": null, - "overflow_y": null, - "padding": null, - "right": null, - "top": null, - "visibility": null, - "width": null - } + "id": "c5XCiSZn2p3k", + "outputId": "97d34dee-7bb5-4735-8c53-6aac6377af9e" + }, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "tensor(5)\n" + ] + } + ], + "source": [ + "with model.trace(input):\n", + "\n", + " # Note we don't need to call .save() on the output,\n", + " # as we're only using its value within the tracing context.\n", + " l1_output = model.layer1.output\n", + "\n", + " l1_amax = torch.argmax(l1_output, dim=1).save()\n", + "\n", + "print(l1_amax[0])" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "BGPUJvWq_bOp" + }, + "source": [ + "Nice! That worked seamlessly, but hold on, how come we didn't need to call `.value[0]` on the result? In previous sections, we were just being explicit to get an understanding of Proxies and their value. In practice, however, `nnsight` knows that when outside of the tracing context we only care about the actual value, and so printing, indexing, and applying functions all immediately return and reflect the data in `.value`. So for the rest of the tutorial we won't use it.\n", + "\n", + "The same principles work for methods and operations as well:" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/" }, - "56490c20aef04fe4a63ba7c990e71e77": { - "model_module": "@jupyter-widgets/controls", - "model_name": "DescriptionStyleModel", - "model_module_version": "1.5.0", - "state": { - "_model_module": "@jupyter-widgets/controls", - "_model_module_version": "1.5.0", - "_model_name": "DescriptionStyleModel", - "_view_count": null, - "_view_module": "@jupyter-widgets/base", - "_view_module_version": "1.2.0", - "_view_name": "StyleView", - "description_width": "" - } + "id": "FIcOYeEjFuln", + "outputId": "d836abd3-b713-44c5-aedc-95bf85abf11a" + }, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "tensor(2.3416)\n" + ] + } + ], + "source": [ + "with model.trace(input):\n", + "\n", + " value = (model.layer1.output.sum() + model.layer2.output.sum()).save()\n", + "\n", + "print(value)" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "V0vNOJtJ2oFR" + }, + "source": [ + "By default, torch functiona, methods and all operators work with `nnsight`. We also enable the use of the `einops` library.\n", + "\n", + "So to recap, the above code block is saying to `nnsight`, \"Run the model with the given `input`. When the output of layer1 is computed, take its sum. Then do the same for layer2. Now that both of those are computed, add them and make sure not to delete this value as I wish to use it outside of the tracing context.\"\n", + "\n", + "Getting and analyzing the activations from various points in a model can be really insightful, and a number of ML techniques do exactly that. However, often times we not only want to view the computation of a model, but influence it as well." + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "m_5qH5gHPOT_" + }, + "source": [ + "## Setting" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "rgju-b_IOLlq" + }, + "source": [ + "To demonstrate the effect of editing the flow of information through the model, let's set the first dimension of the first layer's output to 0. `NNsight` makes this really easy using '=' operator:" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/" }, - "b8ad433cc3dd4a8689cb06655158d545": { - "model_module": "@jupyter-widgets/controls", - "model_name": "HBoxModel", - "model_module_version": "1.5.0", - "state": { - "_dom_classes": [], - "_model_module": "@jupyter-widgets/controls", - "_model_module_version": "1.5.0", - "_model_name": "HBoxModel", - "_view_count": null, - "_view_module": "@jupyter-widgets/controls", - "_view_module_version": "1.5.0", - "_view_name": "HBoxView", - "box_style": "", - "children": [ - "IPY_MODEL_e51620404d244a4e806eeaf3b96a474f", - "IPY_MODEL_e4568f72971749ac92cc5577cb9c4f9c", - "IPY_MODEL_1a1ada8b122d4502949c4c75861a82bf" - ], - "layout": "IPY_MODEL_18d4c2302f424bc18f9215f2d5b83393" - } + "id": "-6y2wzJqOz3a", + "outputId": "6274e5c5-c5fe-4e07-87c4-a119cfd09742" + }, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "Before: tensor([[ 0.0458, 0.5267, 0.7119, 0.4046, 0.2460, 0.7998, 0.4485, -0.2506,\n", + " 0.2968, -0.8834]])\n", + "After: tensor([[ 0.0000, 0.5267, 0.7119, 0.4046, 0.2460, 0.7998, 0.4485, -0.2506,\n", + " 0.2968, -0.8834]])\n" + ] + } + ], + "source": [ + "with model.trace(input):\n", + "\n", + " # Save the output before the edit to compare.\n", + " # Notice we apply .clone() before saving as the setting operation is in-place.\n", + " l1_output_before = model.layer1.output.clone().save()\n", + "\n", + " # Access the 0th index of the hidden state dimension and set it to 0.\n", + " model.layer1.output[:, 0] = 0\n", + "\n", + " # Save the output after to see our edit.\n", + " l1_output_after = model.layer1.output.save()\n", + "\n", + "print(\"Before:\", l1_output_before)\n", + "print(\"After:\", l1_output_after)" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "oZz9SMs3Y_iS" + }, + "source": [ + "Seems our change was reflected. Now the same for the last dimension:" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/", + "height": 363 }, - "e51620404d244a4e806eeaf3b96a474f": { - "model_module": "@jupyter-widgets/controls", - "model_name": "HTMLModel", - "model_module_version": "1.5.0", - "state": { - "_dom_classes": [], - "_model_module": "@jupyter-widgets/controls", - "_model_module_version": "1.5.0", - "_model_name": "HTMLModel", - "_view_count": null, - "_view_module": "@jupyter-widgets/controls", - "_view_module_version": "1.5.0", - "_view_name": "HTMLView", - "description": "", - "description_tooltip": null, - "layout": "IPY_MODEL_f28f557210b7402b8b2d0fd83f038dc1", - "placeholder": "", - "style": "IPY_MODEL_bc8190c700e949ebb6f874c8f8055161", - "value": "merges.txt: 100%" - } + "id": "qwlqvHFcxld2", + "outputId": "28ae7175-5af0-4bd0-f9f0-21680a4fd748" + }, + "outputs": [ + { + "ename": "IndexError", + "evalue": "index 10 is out of bounds for dimension 1 with size 10", + "output_type": "error", + "traceback": [ + "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m", + "\u001b[0;31mIndexError\u001b[0m Traceback (most recent call last)", + "\u001b[0;32m<ipython-input-42-a1e18ebd4137>\u001b[0m in \u001b[0;36m<cell line: 1>\u001b[0;34m()\u001b[0m\n\u001b[0;32m----> 1\u001b[0;31m \u001b[0;32mwith\u001b[0m \u001b[0mmodel\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mtrace\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0minput\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m 2\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 3\u001b[0m \u001b[0;31m# Save the output before the edit to compare.\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 4\u001b[0m \u001b[0;31m# Notice we apply .clone() before saving as the setting operation is in-place.\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 5\u001b[0m \u001b[0ml1_output_before\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mmodel\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mlayer1\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0moutput\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mclone\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0msave\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n", + "\u001b[0;32m/usr/local/lib/python3.10/dist-packages/nnsight/contexts/Runner.py\u001b[0m in \u001b[0;36m__exit__\u001b[0;34m(self, exc_type, exc_val, exc_tb)\u001b[0m\n\u001b[1;32m 40\u001b[0m \u001b[0;32mraise\u001b[0m \u001b[0mexc_val\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 41\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m---> 42\u001b[0;31m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_graph\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mtracing\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0;32mFalse\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m 43\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 44\u001b[0m \u001b[0;32mif\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mremote\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n", + "\u001b[0;32m<ipython-input-42-a1e18ebd4137>\u001b[0m in \u001b[0;36m<cell line: 1>\u001b[0;34m()\u001b[0m\n\u001b[1;32m 6\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 7\u001b[0m \u001b[0;31m# Access the last index of the hidden state dimension and set it to 0.\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m----> 8\u001b[0;31m \u001b[0mmodel\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mlayer1\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0moutput\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mhidden_dims\u001b[0m\u001b[0;34m]\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0;36m0\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m 9\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 10\u001b[0m \u001b[0;31m# Save the output after to see our edit.\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n", + "\u001b[0;32m/usr/local/lib/python3.10/dist-packages/nnsight/tracing/Proxy.py\u001b[0m in \u001b[0;36m__setitem__\u001b[0;34m(self, key, value)\u001b[0m\n\u001b[1;32m 90\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 91\u001b[0m \u001b[0;32mdef\u001b[0m \u001b[0m__setitem__\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mkey\u001b[0m\u001b[0;34m:\u001b[0m \u001b[0mUnion\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0mProxy\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mAny\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mvalue\u001b[0m\u001b[0;34m:\u001b[0m \u001b[0mUnion\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0mSelf\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mAny\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m)\u001b[0m \u001b[0;34m->\u001b[0m \u001b[0;32mNone\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m---> 92\u001b[0;31m self.node.graph.add(\n\u001b[0m\u001b[1;32m 93\u001b[0m \u001b[0mtarget\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0moperator\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0msetitem\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 94\u001b[0m \u001b[0margs\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mnode\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mkey\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mvalue\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n", + "\u001b[0;32m/usr/local/lib/python3.10/dist-packages/nnsight/tracing/Graph.py\u001b[0m in \u001b[0;36madd\u001b[0;34m(self, target, value, args, kwargs, name)\u001b[0m\n\u001b[1;32m 144\u001b[0m \u001b[0;32mtry\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 145\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 146\u001b[0;31m value = target(\n\u001b[0m\u001b[1;32m 147\u001b[0m \u001b[0;34m*\u001b[0m\u001b[0mNode\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mprepare_proxy_values\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0m_args\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 148\u001b[0m \u001b[0;34m**\u001b[0m\u001b[0mNode\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mprepare_proxy_values\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0m_kwargs\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n", + "\u001b[0;31mIndexError\u001b[0m: index 10 is out of bounds for dimension 1 with size 10" + ] + } + ], + "source": [ + "with model.trace(input):\n", + "\n", + " # Save the output before the edit to compare.\n", + " # Notice we apply .clone() before saving as the setting operation is in-place.\n", + " l1_output_before = model.layer1.output.clone().save()\n", + "\n", + " # Access the last index of the hidden state dimension and set it to 0.\n", + " model.layer1.output[:, hidden_dims] = 0\n", + "\n", + " # Save the output after to see our edit.\n", + " l1_output_after = model.layer1.output.save()\n", + "\n", + "print(\"Before:\", l1_output_before)\n", + "print(\"After:\", l1_output_after)" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "JCNR_Jkpxr8l" + }, + "source": [ + "Ah of course, we needed to index at `hidden_dims - 1` not `hidden_dims`. How did `nnsight` know there was this indexing error before leaving the tracing context?\n", + "\n", + "Earlier when discussing contexts in Python, we learned some logic happens upon entering, and some logic happens upon exiting. We know the model is actually run on exit, but what happens on enter? Our input IS actually run though the model, however under its own \"fake\" context. This means the input makes its way through all of the model operations, allowing `nnsight` to record the shapes and data types of module inputs and outputs! The operations are never executed using tensors with real values so it doesn't incur any memory costs. Then, when creating proxy requests like the setting one above, `nnsight` also attempts to execute the request on the \"fake\" values we recorded. Hence, it lets us know if our request is feasible before even running the model.\n", + "\n", + "<details>\n", + "<summary>On scanning</summary>\n", + "\n", + "---\n", + "\n", + "\"Scanning\" is what we call running \"fake\" inputs throught the model to collect information like shapes and types. \"Validating\" is what we call trying to execute your intervention proxies with \"fake\" inputs to see if they work. If you are doing anything in a loop where efficiency is important, you should turn off scanning and validating. You can turn off validating in `.trace(...)` like `.trace(..., validate=False)`. You can turn off scanning in `Tracer.invoke(...)` ([see the Batching section](#batching-id)) like `Tracer.invoke(..., scan=False)`\n", + "\n", + "---\n", + "\n", + "</details>\n", + "\n", + "Let's try again with the correct indexing, and view the shape of the output before leaving the tracing context:" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/" }, - "e4568f72971749ac92cc5577cb9c4f9c": { - "model_module": "@jupyter-widgets/controls", - "model_name": "FloatProgressModel", - "model_module_version": "1.5.0", - "state": { - "_dom_classes": [], - "_model_module": "@jupyter-widgets/controls", - "_model_module_version": "1.5.0", - "_model_name": "FloatProgressModel", - "_view_count": null, - "_view_module": "@jupyter-widgets/controls", - "_view_module_version": "1.5.0", - "_view_name": "ProgressView", - "bar_style": "success", - "description": "", - "description_tooltip": null, - "layout": "IPY_MODEL_ceebda1d07b545f5b605551c08e73be8", - "max": 456318, - "min": 0, - "orientation": "horizontal", - "style": "IPY_MODEL_6aa96b80e3b746b48f5c13bcafe1134e", - "value": 456318 - } + "id": "Wf8ugJKB2mru", + "outputId": "4eed6c19-cb8c-4deb-e0e1-25aa80959c64" + }, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "layer1 output shape: torch.Size([1, 10])\n", + "Before: tensor([[ 0.0458, 0.5267, 0.7119, 0.4046, 0.2460, 0.7998, 0.4485, -0.2506,\n", + " 0.2968, -0.8834]])\n", + "After: tensor([[ 0.0458, 0.5267, 0.7119, 0.4046, 0.2460, 0.7998, 0.4485, -0.2506,\n", + " 0.2968, 0.0000]])\n" + ] + } + ], + "source": [ + "with model.trace(input):\n", + "\n", + " # Save the output before the edit to compare.\n", + " # Notice we apply .clone() before saving as the setting operation is in-place.\n", + " l1_output_before = model.layer1.output.clone().save()\n", + "\n", + " print(f\"layer1 output shape: {model.layer1.output.shape}\")\n", + "\n", + " # Access the last index of the hidden state dimension and set it to 0.\n", + " model.layer1.output[:, hidden_dims - 1] = 0\n", + "\n", + " # Save the output after to see our edit.\n", + " l1_output_after = model.layer1.output.save()\n", + "\n", + "print(\"Before:\", l1_output_before)\n", + "print(\"After:\", l1_output_after)" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "9DH7PeL13-WU" + }, + "source": [ + "We can also just replace proxy inputs and outputs with tensors of the same shape and type. Let's use the shape information we have at our disposal to add noise to the output, and replace it with this new noised tensor:" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/" }, - "1a1ada8b122d4502949c4c75861a82bf": { - "model_module": "@jupyter-widgets/controls", - "model_name": "HTMLModel", - "model_module_version": "1.5.0", - "state": { - "_dom_classes": [], - "_model_module": "@jupyter-widgets/controls", - "_model_module_version": "1.5.0", - "_model_name": "HTMLModel", - "_view_count": null, - "_view_module": "@jupyter-widgets/controls", - "_view_module_version": "1.5.0", - "_view_name": "HTMLView", - "description": "", - "description_tooltip": null, - "layout": "IPY_MODEL_8de63414803f41c29e75242756cfd67f", - "placeholder": "", - "style": "IPY_MODEL_f693dfed91df492c986afbd4ff925f89", - "value": " 456k/456k [00:00<00:00, 1.87MB/s]" - } + "id": "AP5wRmz_4YE7", + "outputId": "4c30dc1d-232d-4900-bfdd-0c93a9b23ad7" + }, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "Before: tensor([[ 0.0458, 0.5267, 0.7119, 0.4046, 0.2460, 0.7998, 0.4485, -0.2506,\n", + " 0.2968, -0.8834]])\n", + "After: tensor([[ 0.0581, 0.5168, 0.6561, 0.4083, 0.2617, 0.7800, 0.4080, -0.2213,\n", + " 0.3394, -0.9187]])\n" + ] + } + ], + "source": [ + "with model.trace(input):\n", + "\n", + " # Save the output before the edit to compare.\n", + " # Notice we apply .clone() before saving as the setting operation is in-place.\n", + " l1_output_before = model.layer1.output.clone().save()\n", + "\n", + " # Create random noise with variance of .001\n", + " noise = (0.001**0.5)*torch.randn(l1_output_before.shape)\n", + "\n", + " # Add to original value and replace.\n", + " model.layer1.output = l1_output_before + noise\n", + "\n", + " # Save the output after to see our edit.\n", + " l1_output_after = model.layer1.output.save()\n", + "\n", + "print(\"Before:\", l1_output_before)\n", + "print(\"After:\", l1_output_after)" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "s016CelFP8sx" + }, + "source": [ + "## Gradients\n", + "\n", + "`NNsight` can also let you apply backprop and access gradients with respect to a loss. Like `.input` and `.output` on modules, `nnsight` also exposes `.grad` on Proxies themselves (assuming they are proxies of tensors):" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/" }, - "18d4c2302f424bc18f9215f2d5b83393": { - "model_module": "@jupyter-widgets/base", - "model_name": "LayoutModel", - "model_module_version": "1.2.0", - "state": { - "_model_module": "@jupyter-widgets/base", - "_model_module_version": "1.2.0", - "_model_name": "LayoutModel", - "_view_count": null, - "_view_module": "@jupyter-widgets/base", - "_view_module_version": "1.2.0", - "_view_name": "LayoutView", - "align_content": null, - "align_items": null, - "align_self": null, - "border": null, - "bottom": null, - "display": null, - "flex": null, - "flex_flow": null, - "grid_area": null, - "grid_auto_columns": null, - "grid_auto_flow": null, - "grid_auto_rows": null, - "grid_column": null, - "grid_gap": null, - "grid_row": null, - "grid_template_areas": null, - "grid_template_columns": null, - "grid_template_rows": null, - "height": null, - "justify_content": null, - "justify_items": null, - "left": null, - "margin": null, - "max_height": null, - "max_width": null, - "min_height": null, - "min_width": null, - "object_fit": null, - "object_position": null, - "order": null, - "overflow": null, - "overflow_x": null, - "overflow_y": null, - "padding": null, - "right": null, - "top": null, - "visibility": null, - "width": null - } + "id": "L-cYUNNyRDn1", + "outputId": "d03dafa1-ef1e-40f1-de28-4dc2f77fe4db" + }, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "Layer 1 output gradient: tensor([[ 0.4545, -0.0596, -0.2059, 0.4643, -0.4211, -0.2813, 0.2126, 0.5016,\n", + " -0.0126, -0.1564]])\n", + "Layer 2 output gradient: tensor([[1., 1.]])\n" + ] + } + ], + "source": [ + "with model.trace(input):\n", + "\n", + " # We need to explicitly have the tensor require grad\n", + " # as the model we defined earlier turned off requiring grad.\n", + " model.layer1.output.requires_grad = True\n", + "\n", + " # We call .grad on a tensor Proxy to communicate we want to store its gradient.\n", + " # We need to call .save() of course as .grad is its own Proxy.\n", + " layer1_output_grad = model.layer1.output.grad.save()\n", + " layer2_output_grad = model.layer2.output.grad.save()\n", + "\n", + " # Need a loss to propagate through the later modules in order to have a grad.\n", + " loss = model.output.sum()\n", + " loss.backward()\n", + "\n", + "print(\"Layer 1 output gradient:\", layer1_output_grad)\n", + "print(\"Layer 2 output gradient:\", layer2_output_grad)\n" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "1o0JaaYvWlHG" + }, + "source": [ + "All of the features we learned previously, also apply to `.grad`. In other words, we can apply operations to and edit the gradients. Let's zero the grad of `layer1` and double the grad of `layer2`." + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/" }, - "f28f557210b7402b8b2d0fd83f038dc1": { - "model_module": "@jupyter-widgets/base", - "model_name": "LayoutModel", - "model_module_version": "1.2.0", - "state": { - "_model_module": "@jupyter-widgets/base", - "_model_module_version": "1.2.0", - "_model_name": "LayoutModel", - "_view_count": null, - "_view_module": "@jupyter-widgets/base", - "_view_module_version": "1.2.0", - "_view_name": "LayoutView", - "align_content": null, - "align_items": null, - "align_self": null, - "border": null, - "bottom": null, - "display": null, - "flex": null, - "flex_flow": null, - "grid_area": null, - "grid_auto_columns": null, - "grid_auto_flow": null, - "grid_auto_rows": null, - "grid_column": null, - "grid_gap": null, - "grid_row": null, - "grid_template_areas": null, - "grid_template_columns": null, - "grid_template_rows": null, - "height": null, - "justify_content": null, - "justify_items": null, - "left": null, - "margin": null, - "max_height": null, - "max_width": null, - "min_height": null, - "min_width": null, - "object_fit": null, - "object_position": null, - "order": null, - "overflow": null, - "overflow_x": null, - "overflow_y": null, - "padding": null, - "right": null, - "top": null, - "visibility": null, - "width": null - } + "id": "9bFoaJpOWlRb", + "outputId": "1c026534-9ffe-40e7-d016-e4565ddde8ac" + }, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "Layer 1 output gradient: tensor([[0., 0., 0., 0., 0., 0., 0., 0., 0., 0.]])\n", + "Layer 2 output gradient: tensor([[2., 2.]])\n" + ] + } + ], + "source": [ + "with model.trace(input):\n", + "\n", + " # We need to explicitly have the tensor require grad\n", + " # as the model we defined earlier turned off requiring grad.\n", + " model.layer1.output.requires_grad = True\n", + "\n", + " model.layer1.output.grad[:] = 0\n", + " model.layer2.output.grad = model.layer2.output.grad.clone() * 2\n", + "\n", + " layer1_output_grad = model.layer1.output.grad.save()\n", + " layer2_output_grad = model.layer2.output.grad.save()\n", + "\n", + " # Need a loss to propagate through the later modules in order to have a grad.\n", + " loss = model.output.sum()\n", + " loss.backward()\n", + "\n", + "print(\"Layer 1 output gradient:\", layer1_output_grad)\n", + "print(\"Layer 2 output gradient:\", layer2_output_grad)" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "gvQ1nZgYQDG3" + }, + "source": [ + "# 2 Bigger" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "miQgUY4NAmDQ" + }, + "source": [ + "Now that we have the basics of `nnsight` under our belt, we can scale our model up and combine the techniques we've learned into more interesting experiments.\n", + "\n", + "The `NNsight` class is very bare bones. It wraps a pre-defined model and does no pre-processing on the inputs we enter. It's designed to be extended with more complex and powerful types of models and we're excited to see what can be done to leverage its features." + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "5TJDblHiQpp1" + }, + "source": [ + "## LanguageModel" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "9l9mZOY5HFH2" + }, + "source": [ + "`LanguageModel` is a subclass of `NNsight`. While we could define and create a model to pass in directly, `LanguageModel` includes special support for Huggingface language models, including automatically loading models from a Huggingface ID, and loading the model together with the appropriate tokenizer.\n", + "\n", + "Here is how you can use `LanguageModel` to load `GPT-2`:" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/", + "height": 769, + "referenced_widgets": [ + "0a676b8f2dc34fed9731f19a03e4bfa1", + "74358c341daa44faa41a92b0eb073615", + "c40e17a994ed44d79ef54421f4772858", + "431ffcb0dbb1419dbbb237a29842b905", + "c28c185a508e459b95edb29afedd9c79", + "7a0f3d2deeb14bbba7f0df208133a7e2", + "94cdcf1fb0914dfe99c3c615e68b7ec8", + "f3e24aa6619e4185a3e2f53f2e473e45", + "b7be20bd738940658f710309729d8ccd", + "562a287082eb4882af864062ae13c91b", + "11e99924bdf84739b3cfc7d676b797e5", + "60fe026e908741209637604edfeb0563", + "e7a73cfc3fa04425afc08368532ef0be", + "d839da1c31a34659988e3420ba520725", + "96a3b8c765374a5e98eaf713dd5e85d8", + "b49ce0d00e5b4df097b3a9fc4c0bd8e7", + "b2748df9aff540df954f002310606a56", + "f4caf2bc58704464b6ac7bf6da31b71d", + "3731a5747724403e86ef1abaf4fb389b", + "688bf676958a4050b73cc9597f5110de", + "8776d04bfd3748b69e10708466e9c132", + "56490c20aef04fe4a63ba7c990e71e77", + "b8ad433cc3dd4a8689cb06655158d545", + "e51620404d244a4e806eeaf3b96a474f", + "e4568f72971749ac92cc5577cb9c4f9c", + "1a1ada8b122d4502949c4c75861a82bf", + "18d4c2302f424bc18f9215f2d5b83393", + "f28f557210b7402b8b2d0fd83f038dc1", + "bc8190c700e949ebb6f874c8f8055161", + "ceebda1d07b545f5b605551c08e73be8", + "6aa96b80e3b746b48f5c13bcafe1134e", + "8de63414803f41c29e75242756cfd67f", + "f693dfed91df492c986afbd4ff925f89", + "cef2516aabc24f6d852079984afa8090", + "95d303d83d3348a0adfdf6b5cb94c41c", + "9373fcd054014760b23e3acc928da0a5", + "7e6b41ec535648b190bfc655d9e430dd", + "d0198fa98e5f412380f2ba3e9448aef3", + "eb0c489b913744a28304e5334bb66532", + "c80603555c634e338f16c0021f752420", + "0fddebfb42e847e08890d09f30161984", + "d07a0a49a2924092affc0e07bf0b5ef0", + "beb4231b49024d8a9bb7ead19647d99d", + "26d09aa072d64af0bf4df9a89309093a" + ] }, - "bc8190c700e949ebb6f874c8f8055161": { - "model_module": "@jupyter-widgets/controls", - "model_name": "DescriptionStyleModel", - "model_module_version": "1.5.0", - "state": { - "_model_module": "@jupyter-widgets/controls", - "_model_module_version": "1.5.0", - "_model_name": "DescriptionStyleModel", - "_view_count": null, - "_view_module": "@jupyter-widgets/base", - "_view_module_version": "1.2.0", - "_view_name": "StyleView", - "description_width": "" - } + "id": "1OD2z7d3HQJU", + "outputId": "7a53864c-6965-4c25-9668-4b0d56db9f6b" + }, + "outputs": [ + { + "name": "stderr", + "output_type": "stream", + "text": [ + "/usr/local/lib/python3.10/dist-packages/huggingface_hub/utils/_token.py:88: UserWarning: \n", + "The secret `HF_TOKEN` does not exist in your Colab secrets.\n", + "To authenticate with the Hugging Face Hub, create a token in your settings tab (https://huggingface.co/settings/tokens), set it as secret in your Google Colab and restart your session.\n", + "You will be able to reuse this secret in all of your notebooks.\n", + "Please note that authentication is recommended but still optional to access public models or datasets.\n", + " warnings.warn(\n" + ] }, - "ceebda1d07b545f5b605551c08e73be8": { - "model_module": "@jupyter-widgets/base", - "model_name": "LayoutModel", - "model_module_version": "1.2.0", - "state": { - "_model_module": "@jupyter-widgets/base", - "_model_module_version": "1.2.0", - "_model_name": "LayoutModel", - "_view_count": null, - "_view_module": "@jupyter-widgets/base", - "_view_module_version": "1.2.0", - "_view_name": "LayoutView", - "align_content": null, - "align_items": null, - "align_self": null, - "border": null, - "bottom": null, - "display": null, - "flex": null, - "flex_flow": null, - "grid_area": null, - "grid_auto_columns": null, - "grid_auto_flow": null, - "grid_auto_rows": null, - "grid_column": null, - "grid_gap": null, - "grid_row": null, - "grid_template_areas": null, - "grid_template_columns": null, - "grid_template_rows": null, - "height": null, - "justify_content": null, - "justify_items": null, - "left": null, - "margin": null, - "max_height": null, - "max_width": null, - "min_height": null, - "min_width": null, - "object_fit": null, - "object_position": null, - "order": null, - "overflow": null, - "overflow_x": null, - "overflow_y": null, - "padding": null, - "right": null, - "top": null, - "visibility": null, - "width": null - } + { + "data": { + "application/vnd.jupyter.widget-view+json": { + "model_id": "0a676b8f2dc34fed9731f19a03e4bfa1", + "version_major": 2, + "version_minor": 0 + }, + "text/plain": [ + "config.json: 0%| | 0.00/665 [00:00<?, ?B/s]" + ] + }, + "metadata": {}, + "output_type": "display_data" }, - "6aa96b80e3b746b48f5c13bcafe1134e": { - "model_module": "@jupyter-widgets/controls", - "model_name": "ProgressStyleModel", - "model_module_version": "1.5.0", - "state": { - "_model_module": "@jupyter-widgets/controls", - "_model_module_version": "1.5.0", - "_model_name": "ProgressStyleModel", - "_view_count": null, - "_view_module": "@jupyter-widgets/base", - "_view_module_version": "1.2.0", - "_view_name": "StyleView", - "bar_color": null, - "description_width": "" - } + { + "data": { + "application/vnd.jupyter.widget-view+json": { + "model_id": "60fe026e908741209637604edfeb0563", + "version_major": 2, + "version_minor": 0 + }, + "text/plain": [ + "vocab.json: 0%| | 0.00/1.04M [00:00<?, ?B/s]" + ] + }, + "metadata": {}, + "output_type": "display_data" }, - "8de63414803f41c29e75242756cfd67f": { - "model_module": "@jupyter-widgets/base", - "model_name": "LayoutModel", - "model_module_version": "1.2.0", - "state": { - "_model_module": "@jupyter-widgets/base", - "_model_module_version": "1.2.0", - "_model_name": "LayoutModel", - "_view_count": null, - "_view_module": "@jupyter-widgets/base", - "_view_module_version": "1.2.0", - "_view_name": "LayoutView", - "align_content": null, - "align_items": null, - "align_self": null, - "border": null, - "bottom": null, - "display": null, - "flex": null, - "flex_flow": null, - "grid_area": null, - "grid_auto_columns": null, - "grid_auto_flow": null, - "grid_auto_rows": null, - "grid_column": null, - "grid_gap": null, - "grid_row": null, - "grid_template_areas": null, - "grid_template_columns": null, - "grid_template_rows": null, - "height": null, - "justify_content": null, - "justify_items": null, - "left": null, - "margin": null, - "max_height": null, - "max_width": null, - "min_height": null, - "min_width": null, - "object_fit": null, - "object_position": null, - "order": null, - "overflow": null, - "overflow_x": null, - "overflow_y": null, - "padding": null, - "right": null, - "top": null, - "visibility": null, - "width": null - } + { + "data": { + "application/vnd.jupyter.widget-view+json": { + "model_id": "b8ad433cc3dd4a8689cb06655158d545", + "version_major": 2, + "version_minor": 0 + }, + "text/plain": [ + "merges.txt: 0%| | 0.00/456k [00:00<?, ?B/s]" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "application/vnd.jupyter.widget-view+json": { + "model_id": "cef2516aabc24f6d852079984afa8090", + "version_major": 2, + "version_minor": 0 + }, + "text/plain": [ + "tokenizer.json: 0%| | 0.00/1.36M [00:00<?, ?B/s]" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "name": "stdout", + "output_type": "stream", + "text": [ + "GPT2LMHeadModel(\n", + " (transformer): GPT2Model(\n", + " (wte): Embedding(50257, 768)\n", + " (wpe): Embedding(1024, 768)\n", + " (drop): Dropout(p=0.1, inplace=False)\n", + " (h): ModuleList(\n", + " (0-11): 12 x GPT2Block(\n", + " (ln_1): LayerNorm((768,), eps=1e-05, elementwise_affine=True)\n", + " (attn): GPT2Attention(\n", + " (c_attn): Conv1D()\n", + " (c_proj): Conv1D()\n", + " (attn_dropout): Dropout(p=0.1, inplace=False)\n", + " (resid_dropout): Dropout(p=0.1, inplace=False)\n", + " )\n", + " (ln_2): LayerNorm((768,), eps=1e-05, elementwise_affine=True)\n", + " (mlp): GPT2MLP(\n", + " (c_fc): Conv1D()\n", + " (c_proj): Conv1D()\n", + " (act): NewGELUActivation()\n", + " (dropout): Dropout(p=0.1, inplace=False)\n", + " )\n", + " )\n", + " )\n", + " (ln_f): LayerNorm((768,), eps=1e-05, elementwise_affine=True)\n", + " )\n", + " (lm_head): Linear(in_features=768, out_features=50257, bias=False)\n", + " (generator): WrapperModule()\n", + ")\n" + ] + } + ], + "source": [ + "from nnsight import LanguageModel\n", + "\n", + "model = LanguageModel('openai-community/gpt2', device_map=\"auto\")\n", + "\n", + "print(model)" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "Hw2VaAr_ezkj" + }, + "source": [ + "<details>\n", + "<summary>On Model Initialization</summary>\n", + "\n", + "---\n", + "\n", + "A few important things to note:\n", + "\n", + " Keyword arguments passed to the initialization of `LanguageModel` is forwarded to HuggingFace specific loading logic. In this case, `device_map` specifies which devices to use and its value `auto` indicates to evenly distribute it to all available GPUs (and cpu if no GPUs available). Other arguments can be found here: https://huggingface.co/docs/transformers/model_doc/auto#transformers.AutoModelForCausalLM\n", + "\n", + "\n", + " When we initialize `LanguageModel`, we aren't yet loading the parameters of the model into memory. We are actually loading a 'meta' version of the model which doesn't take up any memory, but still allows us to view and trace actions on it. After exiting the first tracing context, the model is then fully loaded into memory. To load into memory on initialization, you can pass `dispatch=True` into `LanguageModel` like `LanguageModel('openai-community/gpt2', device_map=\"auto\", dispatch=True)`.\n", + "\n", + "---\n", + "\n", + "</details>\n", + "\n", + "\n", + "Let's put together some of the features we applied to the small model, but now on `GPT-2`. Unlike `NNsight`, `LanguageModel` does define logic to pre-process inputs upon entering the tracing context. This makes interacting with the model simpler without having to directly access the tokenizer.\n", + "\n", + "In the following example, we ablate the value coming from the last layer's MLP module and decode the logits to see what token the model predicts without influence from that particular module:" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/" }, - "f693dfed91df492c986afbd4ff925f89": { - "model_module": "@jupyter-widgets/controls", - "model_name": "DescriptionStyleModel", - "model_module_version": "1.5.0", - "state": { - "_model_module": "@jupyter-widgets/controls", - "_model_module_version": "1.5.0", - "_model_name": "DescriptionStyleModel", - "_view_count": null, - "_view_module": "@jupyter-widgets/base", - "_view_module_version": "1.2.0", - "_view_name": "StyleView", - "description_width": "" - } + "id": "mLSapaMCgLNU", + "outputId": "192b8998-ecca-4f25-ecf2-5196c0d2d672" + }, + "outputs": [ + { + "name": "stderr", + "output_type": "stream", + "text": [ + "You're using a GPT2TokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n" + ] }, - "cef2516aabc24f6d852079984afa8090": { - "model_module": "@jupyter-widgets/controls", - "model_name": "HBoxModel", - "model_module_version": "1.5.0", - "state": { - "_dom_classes": [], - "_model_module": "@jupyter-widgets/controls", - "_model_module_version": "1.5.0", - "_model_name": "HBoxModel", - "_view_count": null, - "_view_module": "@jupyter-widgets/controls", - "_view_module_version": "1.5.0", - "_view_name": "HBoxView", - "box_style": "", - "children": [ - "IPY_MODEL_95d303d83d3348a0adfdf6b5cb94c41c", - "IPY_MODEL_9373fcd054014760b23e3acc928da0a5", - "IPY_MODEL_7e6b41ec535648b190bfc655d9e430dd" - ], - "layout": "IPY_MODEL_d0198fa98e5f412380f2ba3e9448aef3" - } + { + "name": "stdout", + "output_type": "stream", + "text": [ + "Token IDs: tensor([[ 262, 12, 417, 8765, 11, 257, 262, 3504, 338, 3576]])\n", + "Prediction: London\n" + ] + } + ], + "source": [ + "with model.trace('The Eiffel Tower is in the city of'):\n", + "\n", + " # Access the last layer using h[-1] as it's a ModuleList\n", + " # Access the first index of .output as that's where the hidden states are.\n", + " model.transformer.h[-1].mlp.output[0][:] = 0\n", + "\n", + " # Logits come out of model.lm_head and we apply argmax to get the predicted token ids.\n", + " token_ids = model.lm_head.output.argmax(dim=-1).save()\n", + "\n", + "print(\"Token IDs:\", token_ids)\n", + "\n", + "# Apply the tokenizer to decode the ids into words after the tracing context.\n", + "print(\"Prediction:\", model.tokenizer.decode(token_ids[0][-1]))\n" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "X2-HiCQhkv-B" + }, + "source": [ + "You just ran a little intervention on a much more complex model with a lot more parameters! An important piece of information we're missing though is what the prediction would look like without our ablation.\n", + "\n", + "Of course we could just run two tracing contexts and compare the outputs. This, however, would require two forward passes through the model. `NNsight` can do better than that." + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "vLjHnRyRPXjp" + }, + "source": [ + "<a name=\"batching-id\"></a>\n", + "## Batching" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "_Gc-Zr6NmEOn" + }, + "source": [ + "It's time to bring back the `Tracer` object we dropped before. See, when you call `.trace(...)` with some input, it's actually creating two different contexts behind the scenes. The second one is the invoker context. Being within this context just means that `.input` and `.output` should refer only to the input you've given invoke. Calling `.trace(...)` with some input just means there's only one input and therefore only one invoker context.\n", + "\n", + "We can call `.trace()` without input and call `Tracer.invoke(...)` to manually create the invoker context with our input. Now every subsequent time we call `.invoke(...)`, new interventions will only refer to the input in that particular invoke. When exiting the tracing context, the inputs from all of the invokers will be batched together, and they will be executed in one forward pass! So let's do the ablation experiment, and compute a 'control' output to compare to:\n", + "\n", + "<details>\n", + "<summary>On the invoker context</summary>\n", + "\n", + "---\n", + "\n", + "Note that when injecting data to only the relevant invoker interventions, `nnsight` tries, but can't guarantee, that it can narrow the data into the right batch idxs (in the case of an object as input or output). So there are cases where all invokes will get all of the data.\n", + "\n", + "Just like `.trace(...)` created a `Tracer` object, `.invoke(...)` creates an `Invoker` object. The `Invoker` object has post-processed inputs at `invoker.inputs`, which can be useful for seeing information about your input. If you are using `.trace(...)` with inputs, you can still access the invoker object at `tracer._invoker`.\n", + "\n", + "Keyword arguments given to `.invoke(..)` make its way to the input pre-processing. For example in `LanguageModel`, the keyword arguments are used to tokenize like `max_length` and `truncation`. If you need to pass in keyword arguments directly to one input `.trace(...)`, you can pass an `invoker_args` keyword argument that should be a dictionary of keyword arguments for the invoker. `.trace(..., invoker_args={...})`\n", + "\n", + "---\n", + "\n", + "</details>" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/" }, - "95d303d83d3348a0adfdf6b5cb94c41c": { - "model_module": "@jupyter-widgets/controls", - "model_name": "HTMLModel", - "model_module_version": "1.5.0", - "state": { - "_dom_classes": [], - "_model_module": "@jupyter-widgets/controls", - "_model_module_version": "1.5.0", - "_model_name": "HTMLModel", - "_view_count": null, - "_view_module": "@jupyter-widgets/controls", - "_view_module_version": "1.5.0", - "_view_name": "HTMLView", - "description": "", - "description_tooltip": null, - "layout": "IPY_MODEL_eb0c489b913744a28304e5334bb66532", - "placeholder": "", - "style": "IPY_MODEL_c80603555c634e338f16c0021f752420", - "value": "tokenizer.json: 100%" - } + "id": "Kdcq4oCNmEua", + "outputId": "e6d33c07-d22a-41af-b84a-91c9681c7524" + }, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "Original token IDs: tensor([[ 198, 12, 417, 8765, 318, 257, 262, 3504, 7372, 6342]])\n", + "Intervention token IDs: tensor([[ 262, 12, 417, 8765, 11, 257, 262, 3504, 338, 3576]])\n", + "Original prediction: Paris\n", + "Intervention prediction: London\n" + ] + } + ], + "source": [ + "with model.trace() as tracer:\n", + "\n", + " with tracer.invoke('The Eiffel Tower is in the city of'):\n", + "\n", + " # Ablate the last MLP for only this batch.\n", + " model.transformer.h[-1].mlp.output[0][:] = 0\n", + "\n", + " # Get the output for only the intervened on batch.\n", + " token_ids_intervention = model.lm_head.output.argmax(dim=-1).save()\n", + "\n", + " with tracer.invoke('The Eiffel Tower is in the city of'):\n", + "\n", + " # Get the output for only the original batch.\n", + " token_ids_original = model.lm_head.output.argmax(dim=-1).save()\n", + "\n", + "print(\"Original token IDs:\", token_ids_original)\n", + "print(\"Intervention token IDs:\", token_ids_intervention)\n", + "\n", + "print(\"Original prediction:\", model.tokenizer.decode(token_ids_original[0][-1]))\n", + "print(\"Intervention prediction:\", model.tokenizer.decode(token_ids_intervention[0][-1]))" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "8BsybgsK2Bbr" + }, + "source": [ + "So it did end up affecting what the model predicted. That's pretty neat!\n", + "\n", + "Another cool thing with multiple invokes is that the Proxies can interact between them. Here we transfer the word token embeddings from a real prompt into another placeholder prompt. Therefore the latter prompt produces the output of the former prompt:" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/" + }, + "id": "syKTI_KhpvCY", + "outputId": "b2238c5c-d2d8-47bf-b73a-730f1e03cf99" + }, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "Original prediction: _\n", + "Intervention prediction: Paris\n" + ] + } + ], + "source": [ + "with model.trace() as tracer:\n", + "\n", + " with tracer.invoke(\"The Eiffel Tower is in the city of\"):\n", + "\n", + " embeddings = model.transformer.wte.output\n", + "\n", + " with tracer.invoke(\"_ _ _ _ _ _ _ _ _ _\"):\n", + "\n", + " model.transformer.wte.output = embeddings\n", + "\n", + " token_ids_intervention = model.lm_head.output.argmax(dim=-1).save()\n", + "\n", + " with tracer.invoke(\"_ _ _ _ _ _ _ _ _ _\"):\n", + "\n", + " token_ids_original = model.lm_head.output.argmax(dim=-1).save()\n", + "\n", + "print(\"Original prediction:\", model.tokenizer.decode(token_ids_original[0][-1]))\n", + "print(\"Intervention prediction:\", model.tokenizer.decode(token_ids_intervention[0][-1]))" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "vvWA-CWqQtah" + }, + "source": [ + "## .next()\n", + "\n", + "Some HuggingFace models define methods to generate multiple outputs at a time. `LanguageModel` wraps that functionality to provide the same tracing features by using `.generate(...)` instead of `.trace(...)`. This calls the underlying model's `.generate` method. It passes the output through a `model.generator` module that we've added onto the model, allowing you to get the generate output at `model.generator.output`.\n", + "\n", + "In a case like this, the underlying model is called more than once; the modules of said model produce more than one output. Which iteration should a given `module.output` refer to? That's where `Module.next()` comes in.\n", + "\n", + "Each module has a call idx associated with it and `.next()` simply increments that attribute. At the time of execution, data is injected into the intervention graph only at the iteration that matches the call idx." + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/" }, - "9373fcd054014760b23e3acc928da0a5": { - "model_module": "@jupyter-widgets/controls", - "model_name": "FloatProgressModel", - "model_module_version": "1.5.0", - "state": { - "_dom_classes": [], - "_model_module": "@jupyter-widgets/controls", - "_model_module_version": "1.5.0", - "_model_name": "FloatProgressModel", - "_view_count": null, - "_view_module": "@jupyter-widgets/controls", - "_view_module_version": "1.5.0", - "_view_name": "ProgressView", - "bar_style": "success", - "description": "", - "description_tooltip": null, - "layout": "IPY_MODEL_0fddebfb42e847e08890d09f30161984", - "max": 1355256, - "min": 0, - "orientation": "horizontal", - "style": "IPY_MODEL_d07a0a49a2924092affc0e07bf0b5ef0", - "value": 1355256 - } + "id": "Yy5Z9NE1GkaN", + "outputId": "0b4c1bd3-a6f6-4ba1-9cb8-6019c222311f" + }, + "outputs": [ + { + "name": "stderr", + "output_type": "stream", + "text": [ + "Setting `pad_token_id` to `eos_token_id`:50256 for open-end generation.\n", + "A decoder-only architecture is being used, but right-padding was detected! For correct generation results, please set `padding_side='left'` when initializing the tokenizer.\n", + "Setting `pad_token_id` to `eos_token_id`:50256 for open-end generation.\n" + ] }, - "7e6b41ec535648b190bfc655d9e430dd": { + { + "name": "stdout", + "output_type": "stream", + "text": [ + "Prediction 1: Paris\n", + "Prediction 2: ,\n", + "Prediction 3: and\n", + "All token ids: tensor([[ 464, 412, 733, 417, 8765, 318, 287, 262, 1748, 286, 6342, 11,\n", + " 290]])\n", + "All prediction: ['The Eiffel Tower is in the city of Paris, and']\n" + ] + } + ], + "source": [ + "with model.generate(\"The Eiffel Tower is in the city of\", max_new_tokens=3):\n", + "\n", + " token_ids_1 = model.lm_head.output.argmax(dim=-1).save()\n", + "\n", + " token_ids_2 = model.lm_head.next().output.argmax(dim=-1).save()\n", + "\n", + " token_ids_3 = model.lm_head.next().output.argmax(dim=-1).save()\n", + "\n", + " output = model.generator.output.save()\n", + "\n", + "print(\"Prediction 1: \", model.tokenizer.decode(token_ids_1[0][-1]))\n", + "print(\"Prediction 2: \", model.tokenizer.decode(token_ids_2[0][-1]))\n", + "print(\"Prediction 3: \", model.tokenizer.decode(token_ids_3[0][-1]))\n", + "\n", + "print(\"All token ids: \", output)\n", + "\n", + "print(\"All prediction: \", model.tokenizer.batch_decode(output))" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "rpkX-LwBQZHo" + }, + "source": [ + "# 3 I thought you said huge models?" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "uCZR65VmILEb" + }, + "source": [ + "`NNsight` is only one part of our project to democratize access to AI internals. The other half is `NDIF` (National Deep Inference Facility).\n", + "\n", + "The interaction between the two is fairly straightforward. The `intervention graph` we create via the tracing context can be encoded into a custom json format and sent via an http request to the `NDIF` servers. `NDIF` then decodes the `intervention graph` and `interleaves` it alongside the specified model.\n", + "\n", + "To see which models are currently being hosted, check out the following status page: https://nnsight.net/status/" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "65Ks1LUvQaER" + }, + "source": [ + "## Remote execution" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "Aa-ZuVOFKS5k" + }, + "source": [ + "In its current state, `NDIF` requires you to recieve an API key. Therefore, to run the rest of this colab, you would need one of your own. To get one, simply join the [NDIF discord](https://discord.gg/6uFJmCSwW7) and introduce yourself on the `#introductions` channel. Then DM either @JadenFK or @caden and we'll create one for you.\n", + "\n", + "Once you have one, to register your api key with `nnsight`, do the following:" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "id": "ax1NWoS9MJeZ" + }, + "outputs": [], + "source": [ + "from nnsight import CONFIG\n", + "\n", + "CONFIG.set_default_api_key(\"<your api key here>\")" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "mvXLOh65MXmF" + }, + "source": [ + "This only needs to be run once as it will save this api key as the default in a config file along with the `nnsight` installation.\n", + "\n", + "To amp things up a few levels, let's demonstrate using `nnsight`'s tracing context with one of the larger open source language models, `Llama-2-70b`!" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "id": "Q1N04sJPZnJt" + }, + "outputs": [], + "source": [ + "import os\n", + "\n", + "# llama2 70b is a gated model and you need access via your huggingface token\n", + "os.environ['HF_TOKEN'] = \"<your huggingface token>\"\n", + "\n", + "# llama response object requires the version of transformers from github\n", + "!pip uninstall -y transformers\n", + "!pip install git+https://github.com/huggingface/transformers\n", + "\n", + "clear_output()" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "id": "oqVzjNoyNGc7" + }, + "outputs": [], + "source": [ + "# We'll never actually load the parameters so no need to specify a device_map.\n", + "model = LanguageModel(\"meta-llama/Llama-2-70b-hf\")\n", + "\n", + "# All we need to specify using NDIF vs executing locally is remote=True.\n", + "with model.trace('The Eiffel Tower is in the city of', remote=True) as runner:\n", + "\n", + " hidden_states = model.model.layers[-1].output.save()\n", + "\n", + " output = model.output.save()\n", + "\n", + "print(hidden_states)\n", + "\n", + "print(output['logits'])" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "IceVpnZcZ9F0" + }, + "source": [ + "It really is as simple as `remote=True`. All of the techniques we went through in earlier sections work just the same when running locally and remotely.\n", + "\n", + "Note that both `nnsight`, but especially `NDIF`, is in active development and therefore there may be caveats, changes, and errors to work through." + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "P3zRm-7VRRov" + }, + "source": [ + "# Getting Involved!" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "BnCc9xgjvxEP" + }, + "source": [ + "If you're interested in following updates to `nnsight`, contributing, giving feedback, or finding collaborators, please join the [NDIF discord](https://discord.gg/6uFJmCSwW7)!\n", + "\n", + "The [Mech Interp discord](https://discord.gg/km2RQBzaUn) is also a fantastic place to discuss all things mech interp with a really cool community.\n", + "\n", + "Our website [nnsight.net](https://nnsight.net/), has a bunch more tutorials detailing more complex interpretability techniques using `nnsight`. If you want to share any of the work you do using `nnsight`, let others know on either of the discords above and we might turn it into a tutorial on our website.\n", + "\n", + "💟\n", + "\n", + "\n", + "\n" + ] + } + ], + "metadata": { + "colab": { + "include_colab_link": true, + "provenance": [], + "toc_visible": true + }, + "kernelspec": { + "display_name": "Python 3", + "name": "python3" + }, + "language_info": { + "name": "python" + }, + "widgets": { + "application/vnd.jupyter.widget-state+json": { + "0a676b8f2dc34fed9731f19a03e4bfa1": { "model_module": "@jupyter-widgets/controls", - "model_name": "HTMLModel", "model_module_version": "1.5.0", + "model_name": "HBoxModel", "state": { "_dom_classes": [], "_model_module": "@jupyter-widgets/controls", "_model_module_version": "1.5.0", - "_model_name": "HTMLModel", + "_model_name": "HBoxModel", "_view_count": null, "_view_module": "@jupyter-widgets/controls", "_view_module_version": "1.5.0", - "_view_name": "HTMLView", - "description": "", - "description_tooltip": null, - "layout": "IPY_MODEL_beb4231b49024d8a9bb7ead19647d99d", - "placeholder": "", - "style": "IPY_MODEL_26d09aa072d64af0bf4df9a89309093a", - "value": " 1.36M/1.36M [00:00<00:00, 4.16MB/s]" + "_view_name": "HBoxView", + "box_style": "", + "children": [ + "IPY_MODEL_74358c341daa44faa41a92b0eb073615", + "IPY_MODEL_c40e17a994ed44d79ef54421f4772858", + "IPY_MODEL_431ffcb0dbb1419dbbb237a29842b905" + ], + "layout": "IPY_MODEL_c28c185a508e459b95edb29afedd9c79" } }, - "d0198fa98e5f412380f2ba3e9448aef3": { + "0fddebfb42e847e08890d09f30161984": { "model_module": "@jupyter-widgets/base", - "model_name": "LayoutModel", "model_module_version": "1.2.0", - "state": { - "_model_module": "@jupyter-widgets/base", - "_model_module_version": "1.2.0", - "_model_name": "LayoutModel", - "_view_count": null, - "_view_module": "@jupyter-widgets/base", - "_view_module_version": "1.2.0", - "_view_name": "LayoutView", - "align_content": null, - "align_items": null, - "align_self": null, - "border": null, - "bottom": null, - "display": null, - "flex": null, - "flex_flow": null, - "grid_area": null, - "grid_auto_columns": null, - "grid_auto_flow": null, - "grid_auto_rows": null, - "grid_column": null, - "grid_gap": null, - "grid_row": null, - "grid_template_areas": null, - "grid_template_columns": null, - "grid_template_rows": null, - "height": null, - "justify_content": null, - "justify_items": null, - "left": null, - "margin": null, - "max_height": null, - "max_width": null, - "min_height": null, - "min_width": null, - "object_fit": null, - "object_position": null, - "order": null, - "overflow": null, - "overflow_x": null, - "overflow_y": null, - "padding": null, - "right": null, - "top": null, - "visibility": null, - "width": null - } - }, - "eb0c489b913744a28304e5334bb66532": { - "model_module": "@jupyter-widgets/base", "model_name": "LayoutModel", - "model_module_version": "1.2.0", "state": { "_model_module": "@jupyter-widgets/base", "_model_module_version": "1.2.0", @@ -1234,10 +1670,10 @@ "width": null } }, - "c80603555c634e338f16c0021f752420": { + "11e99924bdf84739b3cfc7d676b797e5": { "model_module": "@jupyter-widgets/controls", - "model_name": "DescriptionStyleModel", "model_module_version": "1.5.0", + "model_name": "DescriptionStyleModel", "state": { "_model_module": "@jupyter-widgets/controls", "_model_module_version": "1.5.0", @@ -1249,10 +1685,10 @@ "description_width": "" } }, - "0fddebfb42e847e08890d09f30161984": { + "18d4c2302f424bc18f9215f2d5b83393": { "model_module": "@jupyter-widgets/base", - "model_name": "LayoutModel", "model_module_version": "1.2.0", + "model_name": "LayoutModel", "state": { "_model_module": "@jupyter-widgets/base", "_model_module_version": "1.2.0", @@ -1301,26 +1737,31 @@ "width": null } }, - "d07a0a49a2924092affc0e07bf0b5ef0": { + "1a1ada8b122d4502949c4c75861a82bf": { "model_module": "@jupyter-widgets/controls", - "model_name": "ProgressStyleModel", "model_module_version": "1.5.0", + "model_name": "HTMLModel", "state": { + "_dom_classes": [], "_model_module": "@jupyter-widgets/controls", "_model_module_version": "1.5.0", - "_model_name": "ProgressStyleModel", + "_model_name": "HTMLModel", "_view_count": null, - "_view_module": "@jupyter-widgets/base", - "_view_module_version": "1.2.0", - "_view_name": "StyleView", - "bar_color": null, - "description_width": "" + "_view_module": "@jupyter-widgets/controls", + "_view_module_version": "1.5.0", + "_view_name": "HTMLView", + "description": "", + "description_tooltip": null, + "layout": "IPY_MODEL_8de63414803f41c29e75242756cfd67f", + "placeholder": "", + "style": "IPY_MODEL_f693dfed91df492c986afbd4ff925f89", + "value": " 456k/456k [00:00<00:00, 1.87MB/s]" } }, - "beb4231b49024d8a9bb7ead19647d99d": { + "21f5a81dd29549e69b5cfa2661371e35": { "model_module": "@jupyter-widgets/base", - "model_name": "LayoutModel", "model_module_version": "1.2.0", + "model_name": "LayoutModel", "state": { "_model_module": "@jupyter-widgets/base", "_model_module_version": "1.2.0", @@ -1366,13 +1807,13 @@ "right": null, "top": null, "visibility": null, - "width": null + "width": "20px" } }, "26d09aa072d64af0bf4df9a89309093a": { "model_module": "@jupyter-widgets/controls", - "model_name": "DescriptionStyleModel", "model_module_version": "1.5.0", + "model_name": "DescriptionStyleModel", "state": { "_model_module": "@jupyter-widgets/controls", "_model_module_version": "1.5.0", @@ -1384,98 +1825,10 @@ "description_width": "" } }, - "b3078a7a41c642839748d9864220cbde": { - "model_module": "@jupyter-widgets/controls", - "model_name": "HBoxModel", - "model_module_version": "1.5.0", - "state": { - "_dom_classes": [], - "_model_module": "@jupyter-widgets/controls", - "_model_module_version": "1.5.0", - "_model_name": "HBoxModel", - "_view_count": null, - "_view_module": "@jupyter-widgets/controls", - "_view_module_version": "1.5.0", - "_view_name": "HBoxView", - "box_style": "", - "children": [ - "IPY_MODEL_3aa1aef302474a0cba55152e311acd40", - "IPY_MODEL_dda1704b991a44dfb2b28191f307e081", - "IPY_MODEL_5c0847c658d3489ca529d1afc8c96ed5" - ], - "layout": "IPY_MODEL_9fbbb0bf7d8e4c12939193791efc590f" - } - }, - "3aa1aef302474a0cba55152e311acd40": { - "model_module": "@jupyter-widgets/controls", - "model_name": "HTMLModel", - "model_module_version": "1.5.0", - "state": { - "_dom_classes": [], - "_model_module": "@jupyter-widgets/controls", - "_model_module_version": "1.5.0", - "_model_name": "HTMLModel", - "_view_count": null, - "_view_module": "@jupyter-widgets/controls", - "_view_module_version": "1.5.0", - "_view_name": "HTMLView", - "description": "", - "description_tooltip": null, - "layout": "IPY_MODEL_ded9b4b26ce14ba0aae0b19e0274fcd0", - "placeholder": "", - "style": "IPY_MODEL_47b21a1460f446d9911f2d06c9dd5bbd", - "value": "" - } - }, - "dda1704b991a44dfb2b28191f307e081": { - "model_module": "@jupyter-widgets/controls", - "model_name": "FloatProgressModel", - "model_module_version": "1.5.0", - "state": { - "_dom_classes": [], - "_model_module": "@jupyter-widgets/controls", - "_model_module_version": "1.5.0", - "_model_name": "FloatProgressModel", - "_view_count": null, - "_view_module": "@jupyter-widgets/controls", - "_view_module_version": "1.5.0", - "_view_name": "ProgressView", - "bar_style": "success", - "description": "", - "description_tooltip": null, - "layout": "IPY_MODEL_21f5a81dd29549e69b5cfa2661371e35", - "max": 1, - "min": 0, - "orientation": "horizontal", - "style": "IPY_MODEL_5f608f4313d64e458314ace8237708b1", - "value": 0 - } - }, - "5c0847c658d3489ca529d1afc8c96ed5": { - "model_module": "@jupyter-widgets/controls", - "model_name": "HTMLModel", - "model_module_version": "1.5.0", - "state": { - "_dom_classes": [], - "_model_module": "@jupyter-widgets/controls", - "_model_module_version": "1.5.0", - "_model_name": "HTMLModel", - "_view_count": null, - "_view_module": "@jupyter-widgets/controls", - "_view_module_version": "1.5.0", - "_view_name": "HTMLView", - "description": "", - "description_tooltip": null, - "layout": "IPY_MODEL_5266eb8eb83946e0acbda1362f3bd48a", - "placeholder": "", - "style": "IPY_MODEL_8d7ef198c08b4a4ab49501e2204bdbb6", - "value": " 0/0 [00:00<?, ?it/s]" - } - }, - "9fbbb0bf7d8e4c12939193791efc590f": { + "3731a5747724403e86ef1abaf4fb389b": { "model_module": "@jupyter-widgets/base", - "model_name": "LayoutModel", "model_module_version": "1.2.0", + "model_name": "LayoutModel", "state": { "_model_module": "@jupyter-widgets/base", "_model_module_version": "1.2.0", @@ -1524,62 +1877,52 @@ "width": null } }, - "ded9b4b26ce14ba0aae0b19e0274fcd0": { - "model_module": "@jupyter-widgets/base", - "model_name": "LayoutModel", - "model_module_version": "1.2.0", + "3aa1aef302474a0cba55152e311acd40": { + "model_module": "@jupyter-widgets/controls", + "model_module_version": "1.5.0", + "model_name": "HTMLModel", "state": { - "_model_module": "@jupyter-widgets/base", - "_model_module_version": "1.2.0", - "_model_name": "LayoutModel", - "_view_count": null, - "_view_module": "@jupyter-widgets/base", - "_view_module_version": "1.2.0", - "_view_name": "LayoutView", - "align_content": null, - "align_items": null, - "align_self": null, - "border": null, - "bottom": null, - "display": null, - "flex": null, - "flex_flow": null, - "grid_area": null, - "grid_auto_columns": null, - "grid_auto_flow": null, - "grid_auto_rows": null, - "grid_column": null, - "grid_gap": null, - "grid_row": null, - "grid_template_areas": null, - "grid_template_columns": null, - "grid_template_rows": null, - "height": null, - "justify_content": null, - "justify_items": null, - "left": null, - "margin": null, - "max_height": null, - "max_width": null, - "min_height": null, - "min_width": null, - "object_fit": null, - "object_position": null, - "order": null, - "overflow": null, - "overflow_x": null, - "overflow_y": null, - "padding": null, - "right": null, - "top": null, - "visibility": null, - "width": null + "_dom_classes": [], + "_model_module": "@jupyter-widgets/controls", + "_model_module_version": "1.5.0", + "_model_name": "HTMLModel", + "_view_count": null, + "_view_module": "@jupyter-widgets/controls", + "_view_module_version": "1.5.0", + "_view_name": "HTMLView", + "description": "", + "description_tooltip": null, + "layout": "IPY_MODEL_ded9b4b26ce14ba0aae0b19e0274fcd0", + "placeholder": "", + "style": "IPY_MODEL_47b21a1460f446d9911f2d06c9dd5bbd", + "value": "" + } + }, + "431ffcb0dbb1419dbbb237a29842b905": { + "model_module": "@jupyter-widgets/controls", + "model_module_version": "1.5.0", + "model_name": "HTMLModel", + "state": { + "_dom_classes": [], + "_model_module": "@jupyter-widgets/controls", + "_model_module_version": "1.5.0", + "_model_name": "HTMLModel", + "_view_count": null, + "_view_module": "@jupyter-widgets/controls", + "_view_module_version": "1.5.0", + "_view_name": "HTMLView", + "description": "", + "description_tooltip": null, + "layout": "IPY_MODEL_562a287082eb4882af864062ae13c91b", + "placeholder": "", + "style": "IPY_MODEL_11e99924bdf84739b3cfc7d676b797e5", + "value": " 665/665 [00:00<00:00, 12.8kB/s]" } }, "47b21a1460f446d9911f2d06c9dd5bbd": { "model_module": "@jupyter-widgets/controls", - "model_name": "DescriptionStyleModel", "model_module_version": "1.5.0", + "model_name": "DescriptionStyleModel", "state": { "_model_module": "@jupyter-widgets/controls", "_model_module_version": "1.5.0", @@ -1591,10 +1934,10 @@ "description_width": "" } }, - "21f5a81dd29549e69b5cfa2661371e35": { + "5266eb8eb83946e0acbda1362f3bd48a": { "model_module": "@jupyter-widgets/base", - "model_name": "LayoutModel", "model_module_version": "1.2.0", + "model_name": "LayoutModel", "state": { "_model_module": "@jupyter-widgets/base", "_model_module_version": "1.2.0", @@ -1640,29 +1983,13 @@ "right": null, "top": null, "visibility": null, - "width": "20px" - } - }, - "5f608f4313d64e458314ace8237708b1": { - "model_module": "@jupyter-widgets/controls", - "model_name": "ProgressStyleModel", - "model_module_version": "1.5.0", - "state": { - "_model_module": "@jupyter-widgets/controls", - "_model_module_version": "1.5.0", - "_model_name": "ProgressStyleModel", - "_view_count": null, - "_view_module": "@jupyter-widgets/base", - "_view_module_version": "1.2.0", - "_view_name": "StyleView", - "bar_color": null, - "description_width": "" + "width": null } }, - "5266eb8eb83946e0acbda1362f3bd48a": { + "562a287082eb4882af864062ae13c91b": { "model_module": "@jupyter-widgets/base", - "model_name": "LayoutModel", "model_module_version": "1.2.0", + "model_name": "LayoutModel", "state": { "_model_module": "@jupyter-widgets/base", "_model_module_version": "1.2.0", @@ -1711,10 +2038,10 @@ "width": null } }, - "8d7ef198c08b4a4ab49501e2204bdbb6": { + "56490c20aef04fe4a63ba7c990e71e77": { "model_module": "@jupyter-widgets/controls", - "model_name": "DescriptionStyleModel", "model_module_version": "1.5.0", + "model_name": "DescriptionStyleModel", "state": { "_model_module": "@jupyter-widgets/controls", "_model_module_version": "1.5.0", @@ -1725,1599 +2052,1263 @@ "_view_name": "StyleView", "description_width": "" } - } - } - } - }, - "cells": [ - { - "cell_type": "markdown", - "metadata": { - "id": "view-in-github", - "colab_type": "text" - }, - "source": [ - "<a href=\"https://colab.research.google.com/github/JadenFiotto-Kaufman/nnsight/blob/dev/NNsight_v0_2.ipynb\" target=\"_parent\"><img src=\"https://colab.research.google.com/assets/colab-badge.svg\" alt=\"Open In Colab\"/></a>" - ] - }, - { - "cell_type": "markdown", - "source": [ - "<img src=\"https://nnsight.net/_static/images/nnsight_logo.svg\" alt=\"drawing\" width=\"200\"/>\n", - "\n", - "# **NNsight**\n", - "## The API for a transparent science on black-box AI\n", - "\n", - "In this era of large-scale deep learning, the most interesting AI models are massive black boxes that are hard to run. Ordinary commercial inference service APIs let you interact with huge models, but they do not let you access model internals.\n", - "\n", - "The nnsight library is different: it gives you full access to all the neural network internals. When used together with a remote service like the [National Deep Inference Facility](https://thevisible.net/docs/NDIF-proposal.pdf) (NDIF), it lets you run complex experiments on huge open source models easily, with fully transparent access.\n", - "\n", - "Our team wants to enable entire labs and independent researchers alike, as we believe a large, passionate, and collaborative community will produce the next big insights on a profoundly important field.\n", - "\n", - "\n" - ], - "metadata": { - "id": "-Scfr942GwO4" - } - }, - { - "cell_type": "markdown", - "source": [ - "# 1 First, let's start small\n", - "\n", - "\n", - "\n" - ], - "metadata": { - "id": "N1OemD2VGyZx" - } - }, - { - "cell_type": "markdown", - "source": [ - "## The Tracing Context\n", - "\n", - "To demonstrate the core functionality and syntax of nnsight, we'll define and use a tiny two layer neural network." - ], - "metadata": { - "id": "WyLMmhrAKTNM" - } - }, - { - "cell_type": "code", - "execution_count": 1, - "metadata": { - "id": "R0RJijD0eXwf" - }, - "outputs": [], - "source": [ - "# Install nnsight\n", - "!pip install nnsight==0.2.3.dev0\n", - "\n", - "from IPython.display import clear_output\n", - "\n", - "clear_output()" - ] - }, - { - "cell_type": "markdown", - "source": [ - "Our little model here is composed of four sub-modules, two linear layers ('layer1', 'layer2'). We specify the sizes of each of these modules, and create some complementary example input." - ], - "metadata": { - "id": "bgydw7i3HmIH" - } - }, - { - "cell_type": "code", - "source": [ - "from collections import OrderedDict\n", - "import torch\n", - "\n", - "input_size = 5\n", - "hidden_dims = 10\n", - "output_size = 2\n", - "\n", - "net = torch.nn.Sequential(OrderedDict([\n", - " ('layer1', torch.nn.Linear(input_size, hidden_dims)),\n", - " ('layer2', torch.nn.Linear(hidden_dims, output_size)),\n", - "])).requires_grad_(False)\n", - "\n", - "input = torch.rand((1, input_size))" - ], - "metadata": { - "id": "2pX2Wg8Ceo6N" - }, - "execution_count": null, - "outputs": [] - }, - { - "cell_type": "markdown", - "source": [ - "The core object of the nnsight package is `NNsight`. This wraps around a given pytorch model to enable the capabilites nnsight provides." - ], - "metadata": { - "id": "tIPa2h2pJIwl" - } - }, - { - "cell_type": "code", - "source": [ - "from nnsight import NNsight\n", - "\n", - "model = NNsight(net)" - ], - "metadata": { - "id": "8H9R_ynTJI5y", - "outputId": "5d6b545d-750e-441a-fde9-2dc53e2f1f5e", - "colab": { - "base_uri": "https://localhost:8080/", - "height": 426, - "referenced_widgets": [ - "b3078a7a41c642839748d9864220cbde", - "3aa1aef302474a0cba55152e311acd40", - "dda1704b991a44dfb2b28191f307e081", - "5c0847c658d3489ca529d1afc8c96ed5", - "9fbbb0bf7d8e4c12939193791efc590f", - "ded9b4b26ce14ba0aae0b19e0274fcd0", - "47b21a1460f446d9911f2d06c9dd5bbd", - "21f5a81dd29549e69b5cfa2661371e35", - "5f608f4313d64e458314ace8237708b1", - "5266eb8eb83946e0acbda1362f3bd48a", - "8d7ef198c08b4a4ab49501e2204bdbb6" - ] - } - }, - "execution_count": 2, - "outputs": [ - { - "output_type": "stream", - "name": "stderr", - "text": [ - "The cache for model files in Transformers v4.22.0 has been updated. Migrating your old cache. This is a one-time only operation. You can interrupt this and resume the migration later on by calling `transformers.utils.move_cache()`.\n" - ] - }, - { - "output_type": "display_data", - "data": { - "text/plain": [ - "0it [00:00, ?it/s]" - ], - "application/vnd.jupyter.widget-view+json": { - "version_major": 2, - "version_minor": 0, - "model_id": "b3078a7a41c642839748d9864220cbde" - } - }, - "metadata": {} - }, - { - "output_type": "error", - "ename": "ImportError", - "evalue": "cannot import name 'NNsight' from 'nnsight' (/usr/local/lib/python3.10/dist-packages/nnsight/__init__.py)", - "traceback": [ - "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m", - "\u001b[0;31mImportError\u001b[0m Traceback (most recent call last)", - "\u001b[0;32m<ipython-input-2-801ad166c2ee>\u001b[0m in \u001b[0;36m<cell line: 1>\u001b[0;34m()\u001b[0m\n\u001b[0;32m----> 1\u001b[0;31m \u001b[0;32mfrom\u001b[0m \u001b[0mnnsight\u001b[0m \u001b[0;32mimport\u001b[0m \u001b[0mNNsight\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m 2\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 3\u001b[0m \u001b[0mmodel\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mNNsight\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mnet\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n", - "\u001b[0;31mImportError\u001b[0m: cannot import name 'NNsight' from 'nnsight' (/usr/local/lib/python3.10/dist-packages/nnsight/__init__.py)", - "", - "\u001b[0;31m---------------------------------------------------------------------------\u001b[0;32m\nNOTE: If your import is failing due to a missing package, you can\nmanually install dependencies using either !pip or !apt.\n\nTo view examples of installing some common dependencies, click the\n\"Open Examples\" button below.\n\u001b[0;31m---------------------------------------------------------------------------\u001b[0m\n" - ], - "errorDetails": { - "actions": [ - { - "action": "open_url", - "actionText": "Open Examples", - "url": "/notebooks/snippets/importing_libraries.ipynb" - } - ] - } - } - ] - }, - { - "cell_type": "markdown", - "source": [ - "Printing a Pytorch model shows a named hierarchy of modules which is very useful when accessing sub-components directly. NNsight models work the same." - ], - "metadata": { - "id": "2NfISlQ_Ilvp" - } - }, - { - "cell_type": "code", - "source": [ - "print(model)" - ], - "metadata": { - "id": "bYtnbJHvlGZV", - "colab": { - "base_uri": "https://localhost:8080/" }, - "outputId": "5eb7c572-451c-43c6-956f-af224ae1867c" - }, - "execution_count": null, - "outputs": [ - { - "output_type": "stream", - "name": "stdout", - "text": [ - "Sequential(\n", - " (layer1): Linear(in_features=5, out_features=10, bias=True)\n", - " (layer2): Linear(in_features=10, out_features=2, bias=True)\n", - ")\n" - ] - } - ] - }, - { - "cell_type": "markdown", - "source": [ - "Before we actually get to using the model we just created, let's talk about Python contexts.\n", - "\n", - "Python contexts define a scope using the `with` statement and are often used to create some object, or initiate some logic, that you later want to destroy or conclude.\n", - "\n", - "The most common application is opening files like the following example:\n", - "\n", - "```python\n", - "with open('myfile.txt', 'r') as file:\n", - " text = file.read()\n", - "```\n", - "\n", - "Python uses the `with` keyword to enter a context-like object. This object defines logic to be run at the start of the `with` block, as well as logic to be run when exiting. When using `with` for a file, entering the context opens the file and exiting the context closes it. Being within the context means we can read from the file. Simple enough! Now we can discuss how `nnsight` uses contexts to enable intuitive access into the internals of a neural network.\n" - ], - "metadata": { - "id": "djC5kyJWLuUH" - } - }, - { - "cell_type": "markdown", - "source": [ - "The main tool with `nnsight` is a context for tracing.\n", - "\n", - "We enter the tracing context by calling `model.trace(<input>)` on an `NNsight` model, which defines how we want to run the model. Inside the context, we will be able to customize how the neural network runs. The model is actually run upon exiting the tracing context." - ], - "metadata": { - "id": "iNvuCOeyojcA" - } - }, - { - "cell_type": "code", - "source": [ - "with model.trace(input) as tracer:\n", - " pass" - ], - "metadata": { - "id": "qEXQ4auPSL-m" - }, - "execution_count": null, - "outputs": [] - }, - { - "cell_type": "markdown", - "source": [ - "But where's the output? To get that, we'll have to learn how to request it from within the tracing conext." - ], - "metadata": { - "id": "UZQsHinjqicJ" - } - }, - { - "cell_type": "markdown", - "source": [ - "## Getting" - ], - "metadata": { - "id": "xMfBpYzDPMoB" - } - }, - { - "cell_type": "markdown", - "source": [ - "Earlier, when we wrapped our little neural net with the `NNsight` class. This added a couple properties to each module in the model (including the root model itself). The two most important ones are `.input` and `.output`.\n", - "\n", - "```python\n", - "model.input\n", - "model.output\n", - "```\n", - "\n", - "The names are self explainatory. They correspond to the inputs and outputs of their respective modules during a forward pass of the model. We can use these attributes inside the `with` block.\n", - "\n", - "However, it is important to understand that the model is not executed until the end of the tracing context. How can we access inputs and outputs before the model is run? The trick is deferred execution.\n", - "\n", - "`.input` and `.output` are Proxies for the eventual inputs and outputs of a module. In other words, when you access `model.output` what you are communicating to `nnsight` is, \"When you compute the output of `model`, please grab it for me and put the value into its corresponding Proxy object's `.value` attribute.\" Let's try it:" - ], - "metadata": { - "id": "B5_aFwFRv0ax" - } - }, - { - "cell_type": "code", - "source": [ - "with model.trace(input) as tracer:\n", - "\n", - " output = model.output\n", - "\n", - "print(output.value)" - ], - "metadata": { - "id": "cYe8-r9ptGaG", - "colab": { - "base_uri": "https://localhost:8080/", - "height": 303 + "5c0847c658d3489ca529d1afc8c96ed5": { + "model_module": "@jupyter-widgets/controls", + "model_module_version": "1.5.0", + "model_name": "HTMLModel", + "state": { + "_dom_classes": [], + "_model_module": "@jupyter-widgets/controls", + "_model_module_version": "1.5.0", + "_model_name": "HTMLModel", + "_view_count": null, + "_view_module": "@jupyter-widgets/controls", + "_view_module_version": "1.5.0", + "_view_name": "HTMLView", + "description": "", + "description_tooltip": null, + "layout": "IPY_MODEL_5266eb8eb83946e0acbda1362f3bd48a", + "placeholder": "", + "style": "IPY_MODEL_8d7ef198c08b4a4ab49501e2204bdbb6", + "value": " 0/0 [00:00<?, ?it/s]" + } }, - "outputId": "adf25efe-d2cf-4f65-9eaf-15b155762f4b" - }, - "execution_count": null, - "outputs": [ - { - "output_type": "error", - "ename": "ValueError", - "evalue": "Accessing Proxy value before it's been set.", - "traceback": [ - "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m", - "\u001b[0;31mValueError\u001b[0m Traceback (most recent call last)", - "\u001b[0;32m<ipython-input-34-c7e0c74b12fa>\u001b[0m in \u001b[0;36m<cell line: 5>\u001b[0;34m()\u001b[0m\n\u001b[1;32m 3\u001b[0m \u001b[0moutput\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mmodel\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0moutput\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 4\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m----> 5\u001b[0;31m \u001b[0mprint\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0moutput\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mvalue\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m", - "\u001b[0;32m/usr/local/lib/python3.10/dist-packages/nnsight/tracing/Proxy.py\u001b[0m in \u001b[0;36mvalue\u001b[0;34m(self)\u001b[0m\n\u001b[1;32m 47\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 48\u001b[0m \u001b[0;32mif\u001b[0m \u001b[0;32mnot\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mnode\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mdone\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m---> 49\u001b[0;31m \u001b[0;32mraise\u001b[0m \u001b[0mValueError\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m\"Accessing Proxy value before it's been set.\"\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m 50\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 51\u001b[0m \u001b[0;32mreturn\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mnode\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mvalue\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n", - "\u001b[0;31mValueError\u001b[0m: Accessing Proxy value before it's been set." - ] - } - ] - }, - { - "cell_type": "markdown", - "source": [ - "Oh no an error! \"Accessing Proxy value before it's been set.\"\n", - "\n", - "Why doesn't our `output` have a `value`?\n", - "\n", - "Proxy objects will only have their value at the end of a context if we call `.save()` on them. This helps to reduce memory costs. Adding `.save()` fixes the error:" - ], - "metadata": { - "id": "sBz5qfwuR-6F" - } - }, - { - "cell_type": "code", - "source": [ - "with model.trace(input) as tracer:\n", - "\n", - " output = model.output.save()\n", - "\n", - "print(output.value)" - ], - "metadata": { - "id": "y_bXRd5dvsBu", - "colab": { - "base_uri": "https://localhost:8080/" + "5f608f4313d64e458314ace8237708b1": { + "model_module": "@jupyter-widgets/controls", + "model_module_version": "1.5.0", + "model_name": "ProgressStyleModel", + "state": { + "_model_module": "@jupyter-widgets/controls", + "_model_module_version": "1.5.0", + "_model_name": "ProgressStyleModel", + "_view_count": null, + "_view_module": "@jupyter-widgets/base", + "_view_module_version": "1.2.0", + "_view_name": "StyleView", + "bar_color": null, + "description_width": "" + } }, - "outputId": "a8e8985c-2661-46a8-975b-d43f67cfc162" - }, - "execution_count": null, - "outputs": [ - { - "output_type": "stream", - "name": "stdout", - "text": [ - "tensor([[ 0.1473, -0.1518]])\n" - ] - } - ] - }, - { - "cell_type": "markdown", - "source": [ - "Success! We now have the model output. You just completed your first intervention using `nnsight`.\n", - "\n", - "\n", - "Each time you access a module's input or output, you create an *intervention* in the neural network's forward pass. Collectively these requests form the *intervention graph*. We call the process of executing it alongside the model's normal computation graph, *interleaving*.\n", - "\n", - "<details>\n", - "<summary>On Model output</summary>\n", - "\n", - "---\n", - "\n", - "If you don't need to access anything other than the final model output, you can call the tracing context with `trace=False` and not use it as a context:\n", - "\n", - "```python\n", - "output = model.trace(<inputs>, trace=False)\n", - "```\n", - "\n", - "---\n", - "\n", - "</details>\n", - "\n", - "Just like we saved the output of the model as a whole, we can save the output of any of its submodules. We use normal Python attribute syntax. We can discover how to access them by name by printing out the model:" - ], - "metadata": { - "id": "t5C0UZCwvrYn" - } - }, - { - "cell_type": "code", - "source": [ - "print(model)" - ], - "metadata": { - "colab": { - "base_uri": "https://localhost:8080/" + "60fe026e908741209637604edfeb0563": { + "model_module": "@jupyter-widgets/controls", + "model_module_version": "1.5.0", + "model_name": "HBoxModel", + "state": { + "_dom_classes": [], + "_model_module": "@jupyter-widgets/controls", + "_model_module_version": "1.5.0", + "_model_name": "HBoxModel", + "_view_count": null, + "_view_module": "@jupyter-widgets/controls", + "_view_module_version": "1.5.0", + "_view_name": "HBoxView", + "box_style": "", + "children": [ + "IPY_MODEL_e7a73cfc3fa04425afc08368532ef0be", + "IPY_MODEL_d839da1c31a34659988e3420ba520725", + "IPY_MODEL_96a3b8c765374a5e98eaf713dd5e85d8" + ], + "layout": "IPY_MODEL_b49ce0d00e5b4df097b3a9fc4c0bd8e7" + } }, - "id": "0WcVUSP-0CJi", - "outputId": "7fd9d6ac-152d-4431-9a4d-0b60321638eb" - }, - "execution_count": null, - "outputs": [ - { - "output_type": "stream", - "name": "stdout", - "text": [ - "Sequential(\n", - " (layer1): Linear(in_features=5, out_features=10, bias=True)\n", - " (layer2): Linear(in_features=10, out_features=2, bias=True)\n", - ")\n" - ] - } - ] - }, - { - "cell_type": "code", - "source": [ - "with model.trace(input) as tracer:\n", - "\n", - " l1_output = model.layer1.output.save()\n", - "\n", - "print(l1_output.value)" - ], - "metadata": { - "colab": { - "base_uri": "https://localhost:8080/" + "688bf676958a4050b73cc9597f5110de": { + "model_module": "@jupyter-widgets/controls", + "model_module_version": "1.5.0", + "model_name": "ProgressStyleModel", + "state": { + "_model_module": "@jupyter-widgets/controls", + "_model_module_version": "1.5.0", + "_model_name": "ProgressStyleModel", + "_view_count": null, + "_view_module": "@jupyter-widgets/base", + "_view_module_version": "1.2.0", + "_view_name": "StyleView", + "bar_color": null, + "description_width": "" + } }, - "id": "akWr-cNqy-9O", - "outputId": "68b9e071-c609-4b70-906d-2f2cd93b4495" - }, - "execution_count": null, - "outputs": [ - { - "output_type": "stream", - "name": "stdout", - "text": [ - "tensor([[ 0.0458, 0.5267, 0.7119, 0.4046, 0.2460, 0.7998, 0.4485, -0.2506,\n", - " 0.2968, -0.8834]])\n" - ] - } - ] - }, - { - "cell_type": "markdown", - "source": [ - "Let's do the same for the input of layer2. While we're at it, let's also drop the `as tracer`, as we won't be needing the tracer object itself for a few sections:" - ], - "metadata": { - "id": "85A-aP_03ht6" - } - }, - { - "cell_type": "code", - "source": [ - "with model.trace(input):\n", - "\n", - " l2_input = model.layer2.input.save()\n", - "\n", - "print(l2_input.value)" - ], - "metadata": { - "id": "3EHEN38N3nXR", - "colab": { - "base_uri": "https://localhost:8080/" + "6aa96b80e3b746b48f5c13bcafe1134e": { + "model_module": "@jupyter-widgets/controls", + "model_module_version": "1.5.0", + "model_name": "ProgressStyleModel", + "state": { + "_model_module": "@jupyter-widgets/controls", + "_model_module_version": "1.5.0", + "_model_name": "ProgressStyleModel", + "_view_count": null, + "_view_module": "@jupyter-widgets/base", + "_view_module_version": "1.2.0", + "_view_name": "StyleView", + "bar_color": null, + "description_width": "" + } }, - "outputId": "5afd17a7-0bcc-4d1a-c131-7038bde9455d" - }, - "execution_count": null, - "outputs": [ - { - "output_type": "stream", - "name": "stdout", - "text": [ - "((tensor([[ 0.0458, 0.5267, 0.7119, 0.4046, 0.2460, 0.7998, 0.4485, -0.2506,\n", - " 0.2968, -0.8834]]),), {})\n" - ] - } - ] - }, - { - "cell_type": "markdown", - "source": [ - "<details>\n", - " <summary>On module inputs</summary>\n", - "\n", - " ---\n", - "\n", - " Notice how the value for `l2_input`, was not just a single tensor.\n", - " The type/shape of values from `.input` is in the form of:\n", - "\n", - " tuple(tuple(args), dictionary(kwargs))\n", - "\n", - " Where the first index of the tuple is itself a tuple of all positional arguments, and the second index is a dictionary of the keyword arguments.\n", - "\n", - " ---\n", - "\n", - "</details>\n", - "\n", - "\n", - "Now that we can access activations, we also want to do some post-processing on it. Let's find out which dimension of layer1's output has the highest value." - ], - "metadata": { - "id": "jk-U8zi33Gi-" - } - }, - { - "cell_type": "markdown", - "source": [ - "## Functions, Methods, and Operations" - ], - "metadata": { - "id": "p7Xo_PHyPr4p" - } - }, - { - "cell_type": "markdown", - "source": [ - "We could do this by calling `torch.argmax(...)` after the tracing context or we can just leverage the fact that `nnsight` handles functions and methods within the tracing context, by creating a Proxy request for it:" - ], - "metadata": { - "id": "ehSofWbx5DSx" - } - }, - { - "cell_type": "code", - "source": [ - "with model.trace(input):\n", - "\n", - " # Note we don't need to call .save() on the output,\n", - " # as we're only using its value within the tracing context.\n", - " l1_output = model.layer1.output\n", - "\n", - " l1_amax = torch.argmax(l1_output, dim=1).save()\n", - "\n", - "print(l1_amax[0])" - ], - "metadata": { - "id": "c5XCiSZn2p3k", - "colab": { - "base_uri": "https://localhost:8080/" + "74358c341daa44faa41a92b0eb073615": { + "model_module": "@jupyter-widgets/controls", + "model_module_version": "1.5.0", + "model_name": "HTMLModel", + "state": { + "_dom_classes": [], + "_model_module": "@jupyter-widgets/controls", + "_model_module_version": "1.5.0", + "_model_name": "HTMLModel", + "_view_count": null, + "_view_module": "@jupyter-widgets/controls", + "_view_module_version": "1.5.0", + "_view_name": "HTMLView", + "description": "", + "description_tooltip": null, + "layout": "IPY_MODEL_7a0f3d2deeb14bbba7f0df208133a7e2", + "placeholder": "", + "style": "IPY_MODEL_94cdcf1fb0914dfe99c3c615e68b7ec8", + "value": "config.json: 100%" + } + }, + "7a0f3d2deeb14bbba7f0df208133a7e2": { + "model_module": "@jupyter-widgets/base", + "model_module_version": "1.2.0", + "model_name": "LayoutModel", + "state": { + "_model_module": "@jupyter-widgets/base", + "_model_module_version": "1.2.0", + "_model_name": "LayoutModel", + "_view_count": null, + "_view_module": "@jupyter-widgets/base", + "_view_module_version": "1.2.0", + "_view_name": "LayoutView", + "align_content": null, + "align_items": null, + "align_self": null, + "border": null, + "bottom": null, + "display": null, + "flex": null, + "flex_flow": null, + "grid_area": null, + "grid_auto_columns": null, + "grid_auto_flow": null, + "grid_auto_rows": null, + "grid_column": null, + "grid_gap": null, + "grid_row": null, + "grid_template_areas": null, + "grid_template_columns": null, + "grid_template_rows": null, + "height": null, + "justify_content": null, + "justify_items": null, + "left": null, + "margin": null, + "max_height": null, + "max_width": null, + "min_height": null, + "min_width": null, + "object_fit": null, + "object_position": null, + "order": null, + "overflow": null, + "overflow_x": null, + "overflow_y": null, + "padding": null, + "right": null, + "top": null, + "visibility": null, + "width": null + } + }, + "7e6b41ec535648b190bfc655d9e430dd": { + "model_module": "@jupyter-widgets/controls", + "model_module_version": "1.5.0", + "model_name": "HTMLModel", + "state": { + "_dom_classes": [], + "_model_module": "@jupyter-widgets/controls", + "_model_module_version": "1.5.0", + "_model_name": "HTMLModel", + "_view_count": null, + "_view_module": "@jupyter-widgets/controls", + "_view_module_version": "1.5.0", + "_view_name": "HTMLView", + "description": "", + "description_tooltip": null, + "layout": "IPY_MODEL_beb4231b49024d8a9bb7ead19647d99d", + "placeholder": "", + "style": "IPY_MODEL_26d09aa072d64af0bf4df9a89309093a", + "value": " 1.36M/1.36M [00:00<00:00, 4.16MB/s]" + } }, - "outputId": "97d34dee-7bb5-4735-8c53-6aac6377af9e" - }, - "execution_count": null, - "outputs": [ - { - "output_type": "stream", - "name": "stdout", - "text": [ - "tensor(5)\n" - ] - } - ] - }, - { - "cell_type": "markdown", - "source": [ - "Nice! That worked seamlessly, but hold on, how come we didn't need to call `.value[0]` on the result? In previous sections, we were just being explicit to get an understanding of Proxies and their value. In practice, however, `nnsight` knows that when outside of the tracing context we only care about the actual value, and so printing, indexing, and applying functions all immediately return and reflect the data in `.value`. So for the rest of the tutorial we won't use it.\n", - "\n", - "The same principles work for methods and operations as well:" - ], - "metadata": { - "id": "BGPUJvWq_bOp" - } - }, - { - "cell_type": "code", - "source": [ - "with model.trace(input):\n", - "\n", - " value = (model.layer1.output.sum() + model.layer2.output.sum()).save()\n", - "\n", - "print(value)" - ], - "metadata": { - "id": "FIcOYeEjFuln", - "colab": { - "base_uri": "https://localhost:8080/" + "8776d04bfd3748b69e10708466e9c132": { + "model_module": "@jupyter-widgets/base", + "model_module_version": "1.2.0", + "model_name": "LayoutModel", + "state": { + "_model_module": "@jupyter-widgets/base", + "_model_module_version": "1.2.0", + "_model_name": "LayoutModel", + "_view_count": null, + "_view_module": "@jupyter-widgets/base", + "_view_module_version": "1.2.0", + "_view_name": "LayoutView", + "align_content": null, + "align_items": null, + "align_self": null, + "border": null, + "bottom": null, + "display": null, + "flex": null, + "flex_flow": null, + "grid_area": null, + "grid_auto_columns": null, + "grid_auto_flow": null, + "grid_auto_rows": null, + "grid_column": null, + "grid_gap": null, + "grid_row": null, + "grid_template_areas": null, + "grid_template_columns": null, + "grid_template_rows": null, + "height": null, + "justify_content": null, + "justify_items": null, + "left": null, + "margin": null, + "max_height": null, + "max_width": null, + "min_height": null, + "min_width": null, + "object_fit": null, + "object_position": null, + "order": null, + "overflow": null, + "overflow_x": null, + "overflow_y": null, + "padding": null, + "right": null, + "top": null, + "visibility": null, + "width": null + } }, - "outputId": "d836abd3-b713-44c5-aedc-95bf85abf11a" - }, - "execution_count": null, - "outputs": [ - { - "output_type": "stream", - "name": "stdout", - "text": [ - "tensor(2.3416)\n" - ] - } - ] - }, - { - "cell_type": "markdown", - "source": [ - "By default, torch functiona, methods and all operators work with `nnsight`. We also enable the use of the `einops` library.\n", - "\n", - "So to recap, the above code block is saying to `nnsight`, \"Run the model with the given `input`. When the output of layer1 is computed, take its sum. Then do the same for layer2. Now that both of those are computed, add them and make sure not to delete this value as I wish to use it outside of the tracing context.\"\n", - "\n", - "Getting and analyzing the activations from various points in a model can be really insightful, and a number of ML techniques do exactly that. However, often times we not only want to view the computation of a model, but influence it as well." - ], - "metadata": { - "id": "V0vNOJtJ2oFR" - } - }, - { - "cell_type": "markdown", - "source": [ - "## Setting" - ], - "metadata": { - "id": "m_5qH5gHPOT_" - } - }, - { - "cell_type": "markdown", - "source": [ - "To demonstrate the effect of editing the flow of information through the model, let's set the first dimension of the first layer's output to 0. `NNsight` makes this really easy using '=' operator:" - ], - "metadata": { - "id": "rgju-b_IOLlq" - } - }, - { - "cell_type": "code", - "source": [ - "with model.trace(input):\n", - "\n", - " # Save the output before the edit to compare.\n", - " # Notice we apply .clone() before saving as the setting operation is in-place.\n", - " l1_output_before = model.layer1.output.clone().save()\n", - "\n", - " # Access the 0th index of the hidden state dimension and set it to 0.\n", - " model.layer1.output[:, 0] = 0\n", - "\n", - " # Save the output after to see our edit.\n", - " l1_output_after = model.layer1.output.save()\n", - "\n", - "print(\"Before:\", l1_output_before)\n", - "print(\"After:\", l1_output_after)" - ], - "metadata": { - "id": "-6y2wzJqOz3a", - "colab": { - "base_uri": "https://localhost:8080/" + "8d7ef198c08b4a4ab49501e2204bdbb6": { + "model_module": "@jupyter-widgets/controls", + "model_module_version": "1.5.0", + "model_name": "DescriptionStyleModel", + "state": { + "_model_module": "@jupyter-widgets/controls", + "_model_module_version": "1.5.0", + "_model_name": "DescriptionStyleModel", + "_view_count": null, + "_view_module": "@jupyter-widgets/base", + "_view_module_version": "1.2.0", + "_view_name": "StyleView", + "description_width": "" + } + }, + "8de63414803f41c29e75242756cfd67f": { + "model_module": "@jupyter-widgets/base", + "model_module_version": "1.2.0", + "model_name": "LayoutModel", + "state": { + "_model_module": "@jupyter-widgets/base", + "_model_module_version": "1.2.0", + "_model_name": "LayoutModel", + "_view_count": null, + "_view_module": "@jupyter-widgets/base", + "_view_module_version": "1.2.0", + "_view_name": "LayoutView", + "align_content": null, + "align_items": null, + "align_self": null, + "border": null, + "bottom": null, + "display": null, + "flex": null, + "flex_flow": null, + "grid_area": null, + "grid_auto_columns": null, + "grid_auto_flow": null, + "grid_auto_rows": null, + "grid_column": null, + "grid_gap": null, + "grid_row": null, + "grid_template_areas": null, + "grid_template_columns": null, + "grid_template_rows": null, + "height": null, + "justify_content": null, + "justify_items": null, + "left": null, + "margin": null, + "max_height": null, + "max_width": null, + "min_height": null, + "min_width": null, + "object_fit": null, + "object_position": null, + "order": null, + "overflow": null, + "overflow_x": null, + "overflow_y": null, + "padding": null, + "right": null, + "top": null, + "visibility": null, + "width": null + } + }, + "9373fcd054014760b23e3acc928da0a5": { + "model_module": "@jupyter-widgets/controls", + "model_module_version": "1.5.0", + "model_name": "FloatProgressModel", + "state": { + "_dom_classes": [], + "_model_module": "@jupyter-widgets/controls", + "_model_module_version": "1.5.0", + "_model_name": "FloatProgressModel", + "_view_count": null, + "_view_module": "@jupyter-widgets/controls", + "_view_module_version": "1.5.0", + "_view_name": "ProgressView", + "bar_style": "success", + "description": "", + "description_tooltip": null, + "layout": "IPY_MODEL_0fddebfb42e847e08890d09f30161984", + "max": 1355256, + "min": 0, + "orientation": "horizontal", + "style": "IPY_MODEL_d07a0a49a2924092affc0e07bf0b5ef0", + "value": 1355256 + } }, - "outputId": "6274e5c5-c5fe-4e07-87c4-a119cfd09742" - }, - "execution_count": null, - "outputs": [ - { - "output_type": "stream", - "name": "stdout", - "text": [ - "Before: tensor([[ 0.0458, 0.5267, 0.7119, 0.4046, 0.2460, 0.7998, 0.4485, -0.2506,\n", - " 0.2968, -0.8834]])\n", - "After: tensor([[ 0.0000, 0.5267, 0.7119, 0.4046, 0.2460, 0.7998, 0.4485, -0.2506,\n", - " 0.2968, -0.8834]])\n" - ] - } - ] - }, - { - "cell_type": "markdown", - "source": [ - "Seems our change was reflected. Now the same for the last dimension:" - ], - "metadata": { - "id": "oZz9SMs3Y_iS" - } - }, - { - "cell_type": "code", - "source": [ - "with model.trace(input):\n", - "\n", - " # Save the output before the edit to compare.\n", - " # Notice we apply .clone() before saving as the setting operation is in-place.\n", - " l1_output_before = model.layer1.output.clone().save()\n", - "\n", - " # Access the last index of the hidden state dimension and set it to 0.\n", - " model.layer1.output[:, hidden_dims] = 0\n", - "\n", - " # Save the output after to see our edit.\n", - " l1_output_after = model.layer1.output.save()\n", - "\n", - "print(\"Before:\", l1_output_before)\n", - "print(\"After:\", l1_output_after)" - ], - "metadata": { - "colab": { - "base_uri": "https://localhost:8080/", - "height": 363 + "94cdcf1fb0914dfe99c3c615e68b7ec8": { + "model_module": "@jupyter-widgets/controls", + "model_module_version": "1.5.0", + "model_name": "DescriptionStyleModel", + "state": { + "_model_module": "@jupyter-widgets/controls", + "_model_module_version": "1.5.0", + "_model_name": "DescriptionStyleModel", + "_view_count": null, + "_view_module": "@jupyter-widgets/base", + "_view_module_version": "1.2.0", + "_view_name": "StyleView", + "description_width": "" + } }, - "id": "qwlqvHFcxld2", - "outputId": "28ae7175-5af0-4bd0-f9f0-21680a4fd748" - }, - "execution_count": null, - "outputs": [ - { - "output_type": "error", - "ename": "IndexError", - "evalue": "index 10 is out of bounds for dimension 1 with size 10", - "traceback": [ - "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m", - "\u001b[0;31mIndexError\u001b[0m Traceback (most recent call last)", - "\u001b[0;32m<ipython-input-42-a1e18ebd4137>\u001b[0m in \u001b[0;36m<cell line: 1>\u001b[0;34m()\u001b[0m\n\u001b[0;32m----> 1\u001b[0;31m \u001b[0;32mwith\u001b[0m \u001b[0mmodel\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mtrace\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0minput\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m 2\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 3\u001b[0m \u001b[0;31m# Save the output before the edit to compare.\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 4\u001b[0m \u001b[0;31m# Notice we apply .clone() before saving as the setting operation is in-place.\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 5\u001b[0m \u001b[0ml1_output_before\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mmodel\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mlayer1\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0moutput\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mclone\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0msave\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n", - "\u001b[0;32m/usr/local/lib/python3.10/dist-packages/nnsight/contexts/Runner.py\u001b[0m in \u001b[0;36m__exit__\u001b[0;34m(self, exc_type, exc_val, exc_tb)\u001b[0m\n\u001b[1;32m 40\u001b[0m \u001b[0;32mraise\u001b[0m \u001b[0mexc_val\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 41\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m---> 42\u001b[0;31m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_graph\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mtracing\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0;32mFalse\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m 43\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 44\u001b[0m \u001b[0;32mif\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mremote\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n", - "\u001b[0;32m<ipython-input-42-a1e18ebd4137>\u001b[0m in \u001b[0;36m<cell line: 1>\u001b[0;34m()\u001b[0m\n\u001b[1;32m 6\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 7\u001b[0m \u001b[0;31m# Access the last index of the hidden state dimension and set it to 0.\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m----> 8\u001b[0;31m \u001b[0mmodel\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mlayer1\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0moutput\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mhidden_dims\u001b[0m\u001b[0;34m]\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0;36m0\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m 9\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 10\u001b[0m \u001b[0;31m# Save the output after to see our edit.\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n", - "\u001b[0;32m/usr/local/lib/python3.10/dist-packages/nnsight/tracing/Proxy.py\u001b[0m in \u001b[0;36m__setitem__\u001b[0;34m(self, key, value)\u001b[0m\n\u001b[1;32m 90\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 91\u001b[0m \u001b[0;32mdef\u001b[0m \u001b[0m__setitem__\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mkey\u001b[0m\u001b[0;34m:\u001b[0m \u001b[0mUnion\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0mProxy\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mAny\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mvalue\u001b[0m\u001b[0;34m:\u001b[0m \u001b[0mUnion\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0mSelf\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mAny\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m)\u001b[0m \u001b[0;34m->\u001b[0m \u001b[0;32mNone\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m---> 92\u001b[0;31m self.node.graph.add(\n\u001b[0m\u001b[1;32m 93\u001b[0m \u001b[0mtarget\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0moperator\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0msetitem\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 94\u001b[0m \u001b[0margs\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mnode\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mkey\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mvalue\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n", - "\u001b[0;32m/usr/local/lib/python3.10/dist-packages/nnsight/tracing/Graph.py\u001b[0m in \u001b[0;36madd\u001b[0;34m(self, target, value, args, kwargs, name)\u001b[0m\n\u001b[1;32m 144\u001b[0m \u001b[0;32mtry\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 145\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 146\u001b[0;31m value = target(\n\u001b[0m\u001b[1;32m 147\u001b[0m \u001b[0;34m*\u001b[0m\u001b[0mNode\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mprepare_proxy_values\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0m_args\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 148\u001b[0m \u001b[0;34m**\u001b[0m\u001b[0mNode\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mprepare_proxy_values\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0m_kwargs\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n", - "\u001b[0;31mIndexError\u001b[0m: index 10 is out of bounds for dimension 1 with size 10" - ] - } - ] - }, - { - "cell_type": "markdown", - "source": [ - "Ah of course, we needed to index at `hidden_dims - 1` not `hidden_dims`. How did `nnsight` know there was this indexing error before leaving the tracing context?\n", - "\n", - "Earlier when discussing contexts in Python, we learned some logic happens upon entering, and some logic happens upon exiting. We know the model is actually run on exit, but what happens on enter? Our input IS actually run though the model, however under its own \"fake\" context. This means the input makes its way through all of the model operations, allowing `nnsight` to record the shapes and data types of module inputs and outputs! The operations are never executed using tensors with real values so it doesn't incur any memory costs. Then, when creating proxy requests like the setting one above, `nnsight` also attempts to execute the request on the \"fake\" values we recorded. Hence, it lets us know if our request is feasible before even running the model.\n", - "\n", - "<details>\n", - "<summary>On scanning</summary>\n", - "\n", - "---\n", - "\n", - "\"Scanning\" is what we call running \"fake\" inputs throught the model to collect information like shapes and types. \"Validating\" is what we call trying to execute your intervention proxies with \"fake\" inputs to see if they work. If you are doing anything in a loop where efficiency is important, you should turn off scanning and validating. You can turn off validating in `.trace(...)` like `.trace(..., validate=False)`. You can turn off scanning in `Tracer.invoke(...)` ([see the Batching section](#batching-id)) like `Tracer.invoke(..., scan=False)`\n", - "\n", - "---\n", - "\n", - "</details>\n", - "\n", - "Let's try again with the correct indexing, and view the shape of the output before leaving the tracing context:" - ], - "metadata": { - "id": "JCNR_Jkpxr8l" - } - }, - { - "cell_type": "code", - "source": [ - "with model.trace(input):\n", - "\n", - " # Save the output before the edit to compare.\n", - " # Notice we apply .clone() before saving as the setting operation is in-place.\n", - " l1_output_before = model.layer1.output.clone().save()\n", - "\n", - " print(f\"layer1 output shape: {model.layer1.output.shape}\")\n", - "\n", - " # Access the last index of the hidden state dimension and set it to 0.\n", - " model.layer1.output[:, hidden_dims - 1] = 0\n", - "\n", - " # Save the output after to see our edit.\n", - " l1_output_after = model.layer1.output.save()\n", - "\n", - "print(\"Before:\", l1_output_before)\n", - "print(\"After:\", l1_output_after)" - ], - "metadata": { - "colab": { - "base_uri": "https://localhost:8080/" + "95d303d83d3348a0adfdf6b5cb94c41c": { + "model_module": "@jupyter-widgets/controls", + "model_module_version": "1.5.0", + "model_name": "HTMLModel", + "state": { + "_dom_classes": [], + "_model_module": "@jupyter-widgets/controls", + "_model_module_version": "1.5.0", + "_model_name": "HTMLModel", + "_view_count": null, + "_view_module": "@jupyter-widgets/controls", + "_view_module_version": "1.5.0", + "_view_name": "HTMLView", + "description": "", + "description_tooltip": null, + "layout": "IPY_MODEL_eb0c489b913744a28304e5334bb66532", + "placeholder": "", + "style": "IPY_MODEL_c80603555c634e338f16c0021f752420", + "value": "tokenizer.json: 100%" + } + }, + "96a3b8c765374a5e98eaf713dd5e85d8": { + "model_module": "@jupyter-widgets/controls", + "model_module_version": "1.5.0", + "model_name": "HTMLModel", + "state": { + "_dom_classes": [], + "_model_module": "@jupyter-widgets/controls", + "_model_module_version": "1.5.0", + "_model_name": "HTMLModel", + "_view_count": null, + "_view_module": "@jupyter-widgets/controls", + "_view_module_version": "1.5.0", + "_view_name": "HTMLView", + "description": "", + "description_tooltip": null, + "layout": "IPY_MODEL_8776d04bfd3748b69e10708466e9c132", + "placeholder": "", + "style": "IPY_MODEL_56490c20aef04fe4a63ba7c990e71e77", + "value": " 1.04M/1.04M [00:00<00:00, 3.16MB/s]" + } + }, + "9fbbb0bf7d8e4c12939193791efc590f": { + "model_module": "@jupyter-widgets/base", + "model_module_version": "1.2.0", + "model_name": "LayoutModel", + "state": { + "_model_module": "@jupyter-widgets/base", + "_model_module_version": "1.2.0", + "_model_name": "LayoutModel", + "_view_count": null, + "_view_module": "@jupyter-widgets/base", + "_view_module_version": "1.2.0", + "_view_name": "LayoutView", + "align_content": null, + "align_items": null, + "align_self": null, + "border": null, + "bottom": null, + "display": null, + "flex": null, + "flex_flow": null, + "grid_area": null, + "grid_auto_columns": null, + "grid_auto_flow": null, + "grid_auto_rows": null, + "grid_column": null, + "grid_gap": null, + "grid_row": null, + "grid_template_areas": null, + "grid_template_columns": null, + "grid_template_rows": null, + "height": null, + "justify_content": null, + "justify_items": null, + "left": null, + "margin": null, + "max_height": null, + "max_width": null, + "min_height": null, + "min_width": null, + "object_fit": null, + "object_position": null, + "order": null, + "overflow": null, + "overflow_x": null, + "overflow_y": null, + "padding": null, + "right": null, + "top": null, + "visibility": null, + "width": null + } + }, + "b2748df9aff540df954f002310606a56": { + "model_module": "@jupyter-widgets/base", + "model_module_version": "1.2.0", + "model_name": "LayoutModel", + "state": { + "_model_module": "@jupyter-widgets/base", + "_model_module_version": "1.2.0", + "_model_name": "LayoutModel", + "_view_count": null, + "_view_module": "@jupyter-widgets/base", + "_view_module_version": "1.2.0", + "_view_name": "LayoutView", + "align_content": null, + "align_items": null, + "align_self": null, + "border": null, + "bottom": null, + "display": null, + "flex": null, + "flex_flow": null, + "grid_area": null, + "grid_auto_columns": null, + "grid_auto_flow": null, + "grid_auto_rows": null, + "grid_column": null, + "grid_gap": null, + "grid_row": null, + "grid_template_areas": null, + "grid_template_columns": null, + "grid_template_rows": null, + "height": null, + "justify_content": null, + "justify_items": null, + "left": null, + "margin": null, + "max_height": null, + "max_width": null, + "min_height": null, + "min_width": null, + "object_fit": null, + "object_position": null, + "order": null, + "overflow": null, + "overflow_x": null, + "overflow_y": null, + "padding": null, + "right": null, + "top": null, + "visibility": null, + "width": null + } }, - "id": "Wf8ugJKB2mru", - "outputId": "4eed6c19-cb8c-4deb-e0e1-25aa80959c64" - }, - "execution_count": null, - "outputs": [ - { - "output_type": "stream", - "name": "stdout", - "text": [ - "layer1 output shape: torch.Size([1, 10])\n", - "Before: tensor([[ 0.0458, 0.5267, 0.7119, 0.4046, 0.2460, 0.7998, 0.4485, -0.2506,\n", - " 0.2968, -0.8834]])\n", - "After: tensor([[ 0.0458, 0.5267, 0.7119, 0.4046, 0.2460, 0.7998, 0.4485, -0.2506,\n", - " 0.2968, 0.0000]])\n" - ] - } - ] - }, - { - "cell_type": "markdown", - "source": [ - "We can also just replace proxy inputs and outputs with tensors of the same shape and type. Let's use the shape information we have at our disposal to add noise to the output, and replace it with this new noised tensor:" - ], - "metadata": { - "id": "9DH7PeL13-WU" - } - }, - { - "cell_type": "code", - "source": [ - "with model.trace(input):\n", - "\n", - " # Save the output before the edit to compare.\n", - " # Notice we apply .clone() before saving as the setting operation is in-place.\n", - " l1_output_before = model.layer1.output.clone().save()\n", - "\n", - " # Create random noise with variance of .001\n", - " noise = (0.001**0.5)*torch.randn(l1_output_before.shape)\n", - "\n", - " # Add to original value and replace.\n", - " model.layer1.output = l1_output_before + noise\n", - "\n", - " # Save the output after to see our edit.\n", - " l1_output_after = model.layer1.output.save()\n", - "\n", - "print(\"Before:\", l1_output_before)\n", - "print(\"After:\", l1_output_after)" - ], - "metadata": { - "colab": { - "base_uri": "https://localhost:8080/" + "b3078a7a41c642839748d9864220cbde": { + "model_module": "@jupyter-widgets/controls", + "model_module_version": "1.5.0", + "model_name": "HBoxModel", + "state": { + "_dom_classes": [], + "_model_module": "@jupyter-widgets/controls", + "_model_module_version": "1.5.0", + "_model_name": "HBoxModel", + "_view_count": null, + "_view_module": "@jupyter-widgets/controls", + "_view_module_version": "1.5.0", + "_view_name": "HBoxView", + "box_style": "", + "children": [ + "IPY_MODEL_3aa1aef302474a0cba55152e311acd40", + "IPY_MODEL_dda1704b991a44dfb2b28191f307e081", + "IPY_MODEL_5c0847c658d3489ca529d1afc8c96ed5" + ], + "layout": "IPY_MODEL_9fbbb0bf7d8e4c12939193791efc590f" + } }, - "id": "AP5wRmz_4YE7", - "outputId": "4c30dc1d-232d-4900-bfdd-0c93a9b23ad7" - }, - "execution_count": null, - "outputs": [ - { - "output_type": "stream", - "name": "stdout", - "text": [ - "Before: tensor([[ 0.0458, 0.5267, 0.7119, 0.4046, 0.2460, 0.7998, 0.4485, -0.2506,\n", - " 0.2968, -0.8834]])\n", - "After: tensor([[ 0.0581, 0.5168, 0.6561, 0.4083, 0.2617, 0.7800, 0.4080, -0.2213,\n", - " 0.3394, -0.9187]])\n" - ] - } - ] - }, - { - "cell_type": "markdown", - "source": [ - "## Gradients\n", - "\n", - "`NNsight` can also let you apply backprop and access gradients with respect to a loss. Like `.input` and `.output` on modules, `nnsight` also exposes `.grad` on Proxies themselves (assuming they are proxies of tensors):" - ], - "metadata": { - "id": "s016CelFP8sx" - } - }, - { - "cell_type": "code", - "source": [ - "with model.trace(input):\n", - "\n", - " # We need to explicitly have the tensor require grad\n", - " # as the model we defined earlier turned off requiring grad.\n", - " model.layer1.output.requires_grad = True\n", - "\n", - " # We call .grad on a tensor Proxy to communicate we want to store its gradient.\n", - " # We need to call .save() of course as .grad is its own Proxy.\n", - " layer1_output_grad = model.layer1.output.grad.save()\n", - " layer2_output_grad = model.layer2.output.grad.save()\n", - "\n", - " # Need a loss to propagate through the later modules in order to have a grad.\n", - " loss = model.output.sum()\n", - " loss.backward()\n", - "\n", - "print(\"Layer 1 output gradient:\", layer1_output_grad)\n", - "print(\"Layer 2 output gradient:\", layer2_output_grad)\n" - ], - "metadata": { - "colab": { - "base_uri": "https://localhost:8080/" + "b49ce0d00e5b4df097b3a9fc4c0bd8e7": { + "model_module": "@jupyter-widgets/base", + "model_module_version": "1.2.0", + "model_name": "LayoutModel", + "state": { + "_model_module": "@jupyter-widgets/base", + "_model_module_version": "1.2.0", + "_model_name": "LayoutModel", + "_view_count": null, + "_view_module": "@jupyter-widgets/base", + "_view_module_version": "1.2.0", + "_view_name": "LayoutView", + "align_content": null, + "align_items": null, + "align_self": null, + "border": null, + "bottom": null, + "display": null, + "flex": null, + "flex_flow": null, + "grid_area": null, + "grid_auto_columns": null, + "grid_auto_flow": null, + "grid_auto_rows": null, + "grid_column": null, + "grid_gap": null, + "grid_row": null, + "grid_template_areas": null, + "grid_template_columns": null, + "grid_template_rows": null, + "height": null, + "justify_content": null, + "justify_items": null, + "left": null, + "margin": null, + "max_height": null, + "max_width": null, + "min_height": null, + "min_width": null, + "object_fit": null, + "object_position": null, + "order": null, + "overflow": null, + "overflow_x": null, + "overflow_y": null, + "padding": null, + "right": null, + "top": null, + "visibility": null, + "width": null + } }, - "id": "L-cYUNNyRDn1", - "outputId": "d03dafa1-ef1e-40f1-de28-4dc2f77fe4db" - }, - "execution_count": null, - "outputs": [ - { - "output_type": "stream", - "name": "stdout", - "text": [ - "Layer 1 output gradient: tensor([[ 0.4545, -0.0596, -0.2059, 0.4643, -0.4211, -0.2813, 0.2126, 0.5016,\n", - " -0.0126, -0.1564]])\n", - "Layer 2 output gradient: tensor([[1., 1.]])\n" - ] - } - ] - }, - { - "cell_type": "markdown", - "source": [ - "All of the features we learned previously, also apply to `.grad`. In other words, we can apply operations to and edit the gradients. Let's zero the grad of `layer1` and double the grad of `layer2`." - ], - "metadata": { - "id": "1o0JaaYvWlHG" - } - }, - { - "cell_type": "code", - "source": [ - "with model.trace(input):\n", - "\n", - " # We need to explicitly have the tensor require grad\n", - " # as the model we defined earlier turned off requiring grad.\n", - " model.layer1.output.requires_grad = True\n", - "\n", - " model.layer1.output.grad[:] = 0\n", - " model.layer2.output.grad = model.layer2.output.grad.clone() * 2\n", - "\n", - " layer1_output_grad = model.layer1.output.grad.save()\n", - " layer2_output_grad = model.layer2.output.grad.save()\n", - "\n", - " # Need a loss to propagate through the later modules in order to have a grad.\n", - " loss = model.output.sum()\n", - " loss.backward()\n", - "\n", - "print(\"Layer 1 output gradient:\", layer1_output_grad)\n", - "print(\"Layer 2 output gradient:\", layer2_output_grad)" - ], - "metadata": { - "colab": { - "base_uri": "https://localhost:8080/" + "b7be20bd738940658f710309729d8ccd": { + "model_module": "@jupyter-widgets/controls", + "model_module_version": "1.5.0", + "model_name": "ProgressStyleModel", + "state": { + "_model_module": "@jupyter-widgets/controls", + "_model_module_version": "1.5.0", + "_model_name": "ProgressStyleModel", + "_view_count": null, + "_view_module": "@jupyter-widgets/base", + "_view_module_version": "1.2.0", + "_view_name": "StyleView", + "bar_color": null, + "description_width": "" + } }, - "id": "9bFoaJpOWlRb", - "outputId": "1c026534-9ffe-40e7-d016-e4565ddde8ac" - }, - "execution_count": null, - "outputs": [ - { - "output_type": "stream", - "name": "stdout", - "text": [ - "Layer 1 output gradient: tensor([[0., 0., 0., 0., 0., 0., 0., 0., 0., 0.]])\n", - "Layer 2 output gradient: tensor([[2., 2.]])\n" - ] - } - ] - }, - { - "cell_type": "markdown", - "source": [ - "# 2 Bigger" - ], - "metadata": { - "id": "gvQ1nZgYQDG3" - } - }, - { - "cell_type": "markdown", - "source": [ - "Now that we have the basics of `nnsight` under our belt, we can scale our model up and combine the techniques we've learned into more interesting experiments.\n", - "\n", - "The `NNsight` class is very bare bones. It wraps a pre-defined model and does no pre-processing on the inputs we enter. It's designed to be extended with more complex and powerful types of models and we're excited to see what can be done to leverage its features." - ], - "metadata": { - "id": "miQgUY4NAmDQ" - } - }, - { - "cell_type": "markdown", - "source": [ - "## LanguageModel" - ], - "metadata": { - "id": "5TJDblHiQpp1" - } - }, - { - "cell_type": "markdown", - "source": [ - "`LanguageModel` is a subclass of `NNsight`. While we could define and create a model to pass in directly, `LanguageModel` includes special support for Huggingface language models, including automatically loading models from a Huggingface ID, and loading the model together with the appropriate tokenizer.\n", - "\n", - "Here is how you can use `LanguageModel` to load `GPT-2`:" - ], - "metadata": { - "id": "9l9mZOY5HFH2" - } - }, - { - "cell_type": "code", - "source": [ - "from nnsight import LanguageModel\n", - "\n", - "model = LanguageModel('openai-community/gpt2', device_map=\"auto\")\n", - "\n", - "print(model)" - ], - "metadata": { - "id": "1OD2z7d3HQJU", - "colab": { - "base_uri": "https://localhost:8080/", - "height": 769, - "referenced_widgets": [ - "0a676b8f2dc34fed9731f19a03e4bfa1", - "74358c341daa44faa41a92b0eb073615", - "c40e17a994ed44d79ef54421f4772858", - "431ffcb0dbb1419dbbb237a29842b905", - "c28c185a508e459b95edb29afedd9c79", - "7a0f3d2deeb14bbba7f0df208133a7e2", - "94cdcf1fb0914dfe99c3c615e68b7ec8", - "f3e24aa6619e4185a3e2f53f2e473e45", - "b7be20bd738940658f710309729d8ccd", - "562a287082eb4882af864062ae13c91b", - "11e99924bdf84739b3cfc7d676b797e5", - "60fe026e908741209637604edfeb0563", - "e7a73cfc3fa04425afc08368532ef0be", - "d839da1c31a34659988e3420ba520725", - "96a3b8c765374a5e98eaf713dd5e85d8", - "b49ce0d00e5b4df097b3a9fc4c0bd8e7", - "b2748df9aff540df954f002310606a56", - "f4caf2bc58704464b6ac7bf6da31b71d", - "3731a5747724403e86ef1abaf4fb389b", - "688bf676958a4050b73cc9597f5110de", - "8776d04bfd3748b69e10708466e9c132", - "56490c20aef04fe4a63ba7c990e71e77", - "b8ad433cc3dd4a8689cb06655158d545", - "e51620404d244a4e806eeaf3b96a474f", - "e4568f72971749ac92cc5577cb9c4f9c", - "1a1ada8b122d4502949c4c75861a82bf", - "18d4c2302f424bc18f9215f2d5b83393", - "f28f557210b7402b8b2d0fd83f038dc1", - "bc8190c700e949ebb6f874c8f8055161", - "ceebda1d07b545f5b605551c08e73be8", - "6aa96b80e3b746b48f5c13bcafe1134e", - "8de63414803f41c29e75242756cfd67f", - "f693dfed91df492c986afbd4ff925f89", - "cef2516aabc24f6d852079984afa8090", - "95d303d83d3348a0adfdf6b5cb94c41c", - "9373fcd054014760b23e3acc928da0a5", - "7e6b41ec535648b190bfc655d9e430dd", - "d0198fa98e5f412380f2ba3e9448aef3", - "eb0c489b913744a28304e5334bb66532", - "c80603555c634e338f16c0021f752420", - "0fddebfb42e847e08890d09f30161984", - "d07a0a49a2924092affc0e07bf0b5ef0", - "beb4231b49024d8a9bb7ead19647d99d", - "26d09aa072d64af0bf4df9a89309093a" - ] + "b8ad433cc3dd4a8689cb06655158d545": { + "model_module": "@jupyter-widgets/controls", + "model_module_version": "1.5.0", + "model_name": "HBoxModel", + "state": { + "_dom_classes": [], + "_model_module": "@jupyter-widgets/controls", + "_model_module_version": "1.5.0", + "_model_name": "HBoxModel", + "_view_count": null, + "_view_module": "@jupyter-widgets/controls", + "_view_module_version": "1.5.0", + "_view_name": "HBoxView", + "box_style": "", + "children": [ + "IPY_MODEL_e51620404d244a4e806eeaf3b96a474f", + "IPY_MODEL_e4568f72971749ac92cc5577cb9c4f9c", + "IPY_MODEL_1a1ada8b122d4502949c4c75861a82bf" + ], + "layout": "IPY_MODEL_18d4c2302f424bc18f9215f2d5b83393" + } }, - "outputId": "7a53864c-6965-4c25-9668-4b0d56db9f6b" - }, - "execution_count": null, - "outputs": [ - { - "output_type": "stream", - "name": "stderr", - "text": [ - "/usr/local/lib/python3.10/dist-packages/huggingface_hub/utils/_token.py:88: UserWarning: \n", - "The secret `HF_TOKEN` does not exist in your Colab secrets.\n", - "To authenticate with the Hugging Face Hub, create a token in your settings tab (https://huggingface.co/settings/tokens), set it as secret in your Google Colab and restart your session.\n", - "You will be able to reuse this secret in all of your notebooks.\n", - "Please note that authentication is recommended but still optional to access public models or datasets.\n", - " warnings.warn(\n" - ] + "bc8190c700e949ebb6f874c8f8055161": { + "model_module": "@jupyter-widgets/controls", + "model_module_version": "1.5.0", + "model_name": "DescriptionStyleModel", + "state": { + "_model_module": "@jupyter-widgets/controls", + "_model_module_version": "1.5.0", + "_model_name": "DescriptionStyleModel", + "_view_count": null, + "_view_module": "@jupyter-widgets/base", + "_view_module_version": "1.2.0", + "_view_name": "StyleView", + "description_width": "" + } + }, + "beb4231b49024d8a9bb7ead19647d99d": { + "model_module": "@jupyter-widgets/base", + "model_module_version": "1.2.0", + "model_name": "LayoutModel", + "state": { + "_model_module": "@jupyter-widgets/base", + "_model_module_version": "1.2.0", + "_model_name": "LayoutModel", + "_view_count": null, + "_view_module": "@jupyter-widgets/base", + "_view_module_version": "1.2.0", + "_view_name": "LayoutView", + "align_content": null, + "align_items": null, + "align_self": null, + "border": null, + "bottom": null, + "display": null, + "flex": null, + "flex_flow": null, + "grid_area": null, + "grid_auto_columns": null, + "grid_auto_flow": null, + "grid_auto_rows": null, + "grid_column": null, + "grid_gap": null, + "grid_row": null, + "grid_template_areas": null, + "grid_template_columns": null, + "grid_template_rows": null, + "height": null, + "justify_content": null, + "justify_items": null, + "left": null, + "margin": null, + "max_height": null, + "max_width": null, + "min_height": null, + "min_width": null, + "object_fit": null, + "object_position": null, + "order": null, + "overflow": null, + "overflow_x": null, + "overflow_y": null, + "padding": null, + "right": null, + "top": null, + "visibility": null, + "width": null + } + }, + "c28c185a508e459b95edb29afedd9c79": { + "model_module": "@jupyter-widgets/base", + "model_module_version": "1.2.0", + "model_name": "LayoutModel", + "state": { + "_model_module": "@jupyter-widgets/base", + "_model_module_version": "1.2.0", + "_model_name": "LayoutModel", + "_view_count": null, + "_view_module": "@jupyter-widgets/base", + "_view_module_version": "1.2.0", + "_view_name": "LayoutView", + "align_content": null, + "align_items": null, + "align_self": null, + "border": null, + "bottom": null, + "display": null, + "flex": null, + "flex_flow": null, + "grid_area": null, + "grid_auto_columns": null, + "grid_auto_flow": null, + "grid_auto_rows": null, + "grid_column": null, + "grid_gap": null, + "grid_row": null, + "grid_template_areas": null, + "grid_template_columns": null, + "grid_template_rows": null, + "height": null, + "justify_content": null, + "justify_items": null, + "left": null, + "margin": null, + "max_height": null, + "max_width": null, + "min_height": null, + "min_width": null, + "object_fit": null, + "object_position": null, + "order": null, + "overflow": null, + "overflow_x": null, + "overflow_y": null, + "padding": null, + "right": null, + "top": null, + "visibility": null, + "width": null + } }, - { - "output_type": "display_data", - "data": { - "text/plain": [ - "config.json: 0%| | 0.00/665 [00:00<?, ?B/s]" - ], - "application/vnd.jupyter.widget-view+json": { - "version_major": 2, - "version_minor": 0, - "model_id": "0a676b8f2dc34fed9731f19a03e4bfa1" - } - }, - "metadata": {} + "c40e17a994ed44d79ef54421f4772858": { + "model_module": "@jupyter-widgets/controls", + "model_module_version": "1.5.0", + "model_name": "FloatProgressModel", + "state": { + "_dom_classes": [], + "_model_module": "@jupyter-widgets/controls", + "_model_module_version": "1.5.0", + "_model_name": "FloatProgressModel", + "_view_count": null, + "_view_module": "@jupyter-widgets/controls", + "_view_module_version": "1.5.0", + "_view_name": "ProgressView", + "bar_style": "success", + "description": "", + "description_tooltip": null, + "layout": "IPY_MODEL_f3e24aa6619e4185a3e2f53f2e473e45", + "max": 665, + "min": 0, + "orientation": "horizontal", + "style": "IPY_MODEL_b7be20bd738940658f710309729d8ccd", + "value": 665 + } }, - { - "output_type": "display_data", - "data": { - "text/plain": [ - "vocab.json: 0%| | 0.00/1.04M [00:00<?, ?B/s]" - ], - "application/vnd.jupyter.widget-view+json": { - "version_major": 2, - "version_minor": 0, - "model_id": "60fe026e908741209637604edfeb0563" - } - }, - "metadata": {} + "c80603555c634e338f16c0021f752420": { + "model_module": "@jupyter-widgets/controls", + "model_module_version": "1.5.0", + "model_name": "DescriptionStyleModel", + "state": { + "_model_module": "@jupyter-widgets/controls", + "_model_module_version": "1.5.0", + "_model_name": "DescriptionStyleModel", + "_view_count": null, + "_view_module": "@jupyter-widgets/base", + "_view_module_version": "1.2.0", + "_view_name": "StyleView", + "description_width": "" + } }, - { - "output_type": "display_data", - "data": { - "text/plain": [ - "merges.txt: 0%| | 0.00/456k [00:00<?, ?B/s]" - ], - "application/vnd.jupyter.widget-view+json": { - "version_major": 2, - "version_minor": 0, - "model_id": "b8ad433cc3dd4a8689cb06655158d545" - } - }, - "metadata": {} + "ceebda1d07b545f5b605551c08e73be8": { + "model_module": "@jupyter-widgets/base", + "model_module_version": "1.2.0", + "model_name": "LayoutModel", + "state": { + "_model_module": "@jupyter-widgets/base", + "_model_module_version": "1.2.0", + "_model_name": "LayoutModel", + "_view_count": null, + "_view_module": "@jupyter-widgets/base", + "_view_module_version": "1.2.0", + "_view_name": "LayoutView", + "align_content": null, + "align_items": null, + "align_self": null, + "border": null, + "bottom": null, + "display": null, + "flex": null, + "flex_flow": null, + "grid_area": null, + "grid_auto_columns": null, + "grid_auto_flow": null, + "grid_auto_rows": null, + "grid_column": null, + "grid_gap": null, + "grid_row": null, + "grid_template_areas": null, + "grid_template_columns": null, + "grid_template_rows": null, + "height": null, + "justify_content": null, + "justify_items": null, + "left": null, + "margin": null, + "max_height": null, + "max_width": null, + "min_height": null, + "min_width": null, + "object_fit": null, + "object_position": null, + "order": null, + "overflow": null, + "overflow_x": null, + "overflow_y": null, + "padding": null, + "right": null, + "top": null, + "visibility": null, + "width": null + } }, - { - "output_type": "display_data", - "data": { - "text/plain": [ - "tokenizer.json: 0%| | 0.00/1.36M [00:00<?, ?B/s]" + "cef2516aabc24f6d852079984afa8090": { + "model_module": "@jupyter-widgets/controls", + "model_module_version": "1.5.0", + "model_name": "HBoxModel", + "state": { + "_dom_classes": [], + "_model_module": "@jupyter-widgets/controls", + "_model_module_version": "1.5.0", + "_model_name": "HBoxModel", + "_view_count": null, + "_view_module": "@jupyter-widgets/controls", + "_view_module_version": "1.5.0", + "_view_name": "HBoxView", + "box_style": "", + "children": [ + "IPY_MODEL_95d303d83d3348a0adfdf6b5cb94c41c", + "IPY_MODEL_9373fcd054014760b23e3acc928da0a5", + "IPY_MODEL_7e6b41ec535648b190bfc655d9e430dd" ], - "application/vnd.jupyter.widget-view+json": { - "version_major": 2, - "version_minor": 0, - "model_id": "cef2516aabc24f6d852079984afa8090" - } - }, - "metadata": {} + "layout": "IPY_MODEL_d0198fa98e5f412380f2ba3e9448aef3" + } }, - { - "output_type": "stream", - "name": "stdout", - "text": [ - "GPT2LMHeadModel(\n", - " (transformer): GPT2Model(\n", - " (wte): Embedding(50257, 768)\n", - " (wpe): Embedding(1024, 768)\n", - " (drop): Dropout(p=0.1, inplace=False)\n", - " (h): ModuleList(\n", - " (0-11): 12 x GPT2Block(\n", - " (ln_1): LayerNorm((768,), eps=1e-05, elementwise_affine=True)\n", - " (attn): GPT2Attention(\n", - " (c_attn): Conv1D()\n", - " (c_proj): Conv1D()\n", - " (attn_dropout): Dropout(p=0.1, inplace=False)\n", - " (resid_dropout): Dropout(p=0.1, inplace=False)\n", - " )\n", - " (ln_2): LayerNorm((768,), eps=1e-05, elementwise_affine=True)\n", - " (mlp): GPT2MLP(\n", - " (c_fc): Conv1D()\n", - " (c_proj): Conv1D()\n", - " (act): NewGELUActivation()\n", - " (dropout): Dropout(p=0.1, inplace=False)\n", - " )\n", - " )\n", - " )\n", - " (ln_f): LayerNorm((768,), eps=1e-05, elementwise_affine=True)\n", - " )\n", - " (lm_head): Linear(in_features=768, out_features=50257, bias=False)\n", - " (generator): WrapperModule()\n", - ")\n" - ] - } - ] - }, - { - "cell_type": "markdown", - "source": [ - "<details>\n", - "<summary>On Model Initialization</summary>\n", - "\n", - "---\n", - "\n", - "A few important things to note:\n", - "\n", - " Keyword arguments passed to the initialization of `LanguageModel` is forwarded to HuggingFace specific loading logic. In this case, `device_map` specifies which devices to use and its value `auto` indicates to evenly distribute it to all available GPUs (and cpu if no GPUs available). Other arguments can be found here: https://huggingface.co/docs/transformers/model_doc/auto#transformers.AutoModelForCausalLM\n", - "\n", - "\n", - " When we initialize `LanguageModel`, we aren't yet loading the parameters of the model into memory. We are actually loading a 'meta' version of the model which doesn't take up any memory, but still allows us to view and trace actions on it. After exiting the first tracing context, the model is then fully loaded into memory. To load into memory on initialization, you can pass `dispatch=True` into `LanguageModel` like `LanguageModel('openai-community/gpt2', device_map=\"auto\", dispatch=True)`.\n", - "\n", - "---\n", - "\n", - "</details>\n", - "\n", - "\n", - "Let's put together some of the features we applied to the small model, but now on `GPT-2`. Unlike `NNsight`, `LanguageModel` does define logic to pre-process inputs upon entering the tracing context. This makes interacting with the model simpler without having to directly access the tokenizer.\n", - "\n", - "In the following example, we ablate the value coming from the last layer's MLP module and decode the logits to see what token the model predicts without influence from that particular module:" - ], - "metadata": { - "id": "Hw2VaAr_ezkj" - } - }, - { - "cell_type": "code", - "source": [ - "with model.trace('The Eiffel Tower is in the city of'):\n", - "\n", - " # Access the last layer using h[-1] as it's a ModuleList\n", - " # Access the first index of .output as that's where the hidden states are.\n", - " model.transformer.h[-1].mlp.output[0][:] = 0\n", - "\n", - " # Logits come out of model.lm_head and we apply argmax to get the predicted token ids.\n", - " token_ids = model.lm_head.output.argmax(dim=-1).save()\n", - "\n", - "print(\"Token IDs:\", token_ids)\n", - "\n", - "# Apply the tokenizer to decode the ids into words after the tracing context.\n", - "print(\"Prediction:\", model.tokenizer.decode(token_ids[0][-1]))\n" - ], - "metadata": { - "colab": { - "base_uri": "https://localhost:8080/" + "d0198fa98e5f412380f2ba3e9448aef3": { + "model_module": "@jupyter-widgets/base", + "model_module_version": "1.2.0", + "model_name": "LayoutModel", + "state": { + "_model_module": "@jupyter-widgets/base", + "_model_module_version": "1.2.0", + "_model_name": "LayoutModel", + "_view_count": null, + "_view_module": "@jupyter-widgets/base", + "_view_module_version": "1.2.0", + "_view_name": "LayoutView", + "align_content": null, + "align_items": null, + "align_self": null, + "border": null, + "bottom": null, + "display": null, + "flex": null, + "flex_flow": null, + "grid_area": null, + "grid_auto_columns": null, + "grid_auto_flow": null, + "grid_auto_rows": null, + "grid_column": null, + "grid_gap": null, + "grid_row": null, + "grid_template_areas": null, + "grid_template_columns": null, + "grid_template_rows": null, + "height": null, + "justify_content": null, + "justify_items": null, + "left": null, + "margin": null, + "max_height": null, + "max_width": null, + "min_height": null, + "min_width": null, + "object_fit": null, + "object_position": null, + "order": null, + "overflow": null, + "overflow_x": null, + "overflow_y": null, + "padding": null, + "right": null, + "top": null, + "visibility": null, + "width": null + } }, - "id": "mLSapaMCgLNU", - "outputId": "192b8998-ecca-4f25-ecf2-5196c0d2d672" - }, - "execution_count": null, - "outputs": [ - { - "output_type": "stream", - "name": "stderr", - "text": [ - "You're using a GPT2TokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n" - ] + "d07a0a49a2924092affc0e07bf0b5ef0": { + "model_module": "@jupyter-widgets/controls", + "model_module_version": "1.5.0", + "model_name": "ProgressStyleModel", + "state": { + "_model_module": "@jupyter-widgets/controls", + "_model_module_version": "1.5.0", + "_model_name": "ProgressStyleModel", + "_view_count": null, + "_view_module": "@jupyter-widgets/base", + "_view_module_version": "1.2.0", + "_view_name": "StyleView", + "bar_color": null, + "description_width": "" + } }, - { - "output_type": "stream", - "name": "stdout", - "text": [ - "Token IDs: tensor([[ 262, 12, 417, 8765, 11, 257, 262, 3504, 338, 3576]])\n", - "Prediction: London\n" - ] - } - ] - }, - { - "cell_type": "markdown", - "source": [ - "You just ran a little intervention on a much more complex model with a lot more parameters! An important piece of information we're missing though is what the prediction would look like without our ablation.\n", - "\n", - "Of course we could just run two tracing contexts and compare the outputs. This, however, would require two forward passes through the model. `NNsight` can do better than that." - ], - "metadata": { - "id": "X2-HiCQhkv-B" - } - }, - { - "cell_type": "markdown", - "source": [ - "<a name=\"batching-id\"></a>\n", - "## Batching" - ], - "metadata": { - "id": "vLjHnRyRPXjp" - } - }, - { - "cell_type": "markdown", - "source": [ - "It's time to bring back the `Tracer` object we dropped before. See, when you call `.trace(...)` with some input, it's actually creating two different contexts behind the scenes. The second one is the invoker context. Being within this context just means that `.input` and `.output` should refer only to the input you've given invoke. Calling `.trace(...)` with some input just means there's only one input and therefore only one invoker context.\n", - "\n", - "We can call `.trace()` without input and call `Tracer.invoke(...)` to manually create the invoker context with our input. Now every subsequent time we call `.invoke(...)`, new interventions will only refer to the input in that particular invoke. When exiting the tracing context, the inputs from all of the invokers will be batched together, and they will be executed in one forward pass! So let's do the ablation experiment, and compute a 'control' output to compare to:\n", - "\n", - "<details>\n", - "<summary>On the invoker context</summary>\n", - "\n", - "---\n", - "\n", - "Note that when injecting data to only the relevant invoker interventions, `nnsight` tries, but can't guarantee, that it can narrow the data into the right batch idxs (in the case of an object as input or output). So there are cases where all invokes will get all of the data.\n", - "\n", - "Just like `.trace(...)` created a `Tracer` object, `.invoke(...)` creates an `Invoker` object. The `Invoker` object has post-processed inputs at `invoker.inputs`, which can be useful for seeing information about your input. If you are using `.trace(...)` with inputs, you can still access the invoker object at `tracer._invoker`.\n", - "\n", - "Keyword arguments given to `.invoke(..)` make its way to the input pre-processing. For example in `LanguageModel`, the keyword arguments are used to tokenize like `max_length` and `truncation`. If you need to pass in keyword arguments directly to one input `.trace(...)`, you can pass an `invoker_args` keyword argument that should be a dictionary of keyword arguments for the invoker. `.trace(..., invoker_args={...})`\n", - "\n", - "---\n", - "\n", - "</details>" - ], - "metadata": { - "id": "_Gc-Zr6NmEOn" - } - }, - { - "cell_type": "code", - "source": [ - "with model.trace() as tracer:\n", - "\n", - " with tracer.invoke('The Eiffel Tower is in the city of'):\n", - "\n", - " # Ablate the last MLP for only this batch.\n", - " model.transformer.h[-1].mlp.output[0][:] = 0\n", - "\n", - " # Get the output for only the intervened on batch.\n", - " token_ids_intervention = model.lm_head.output.argmax(dim=-1).save()\n", - "\n", - " with tracer.invoke('The Eiffel Tower is in the city of'):\n", - "\n", - " # Get the output for only the original batch.\n", - " token_ids_original = model.lm_head.output.argmax(dim=-1).save()\n", - "\n", - "print(\"Original token IDs:\", token_ids_original)\n", - "print(\"Intervention token IDs:\", token_ids_intervention)\n", - "\n", - "print(\"Original prediction:\", model.tokenizer.decode(token_ids_original[0][-1]))\n", - "print(\"Intervention prediction:\", model.tokenizer.decode(token_ids_intervention[0][-1]))" - ], - "metadata": { - "colab": { - "base_uri": "https://localhost:8080/" + "d839da1c31a34659988e3420ba520725": { + "model_module": "@jupyter-widgets/controls", + "model_module_version": "1.5.0", + "model_name": "FloatProgressModel", + "state": { + "_dom_classes": [], + "_model_module": "@jupyter-widgets/controls", + "_model_module_version": "1.5.0", + "_model_name": "FloatProgressModel", + "_view_count": null, + "_view_module": "@jupyter-widgets/controls", + "_view_module_version": "1.5.0", + "_view_name": "ProgressView", + "bar_style": "success", + "description": "", + "description_tooltip": null, + "layout": "IPY_MODEL_3731a5747724403e86ef1abaf4fb389b", + "max": 1042301, + "min": 0, + "orientation": "horizontal", + "style": "IPY_MODEL_688bf676958a4050b73cc9597f5110de", + "value": 1042301 + } + }, + "dda1704b991a44dfb2b28191f307e081": { + "model_module": "@jupyter-widgets/controls", + "model_module_version": "1.5.0", + "model_name": "FloatProgressModel", + "state": { + "_dom_classes": [], + "_model_module": "@jupyter-widgets/controls", + "_model_module_version": "1.5.0", + "_model_name": "FloatProgressModel", + "_view_count": null, + "_view_module": "@jupyter-widgets/controls", + "_view_module_version": "1.5.0", + "_view_name": "ProgressView", + "bar_style": "success", + "description": "", + "description_tooltip": null, + "layout": "IPY_MODEL_21f5a81dd29549e69b5cfa2661371e35", + "max": 1, + "min": 0, + "orientation": "horizontal", + "style": "IPY_MODEL_5f608f4313d64e458314ace8237708b1", + "value": 0 + } + }, + "ded9b4b26ce14ba0aae0b19e0274fcd0": { + "model_module": "@jupyter-widgets/base", + "model_module_version": "1.2.0", + "model_name": "LayoutModel", + "state": { + "_model_module": "@jupyter-widgets/base", + "_model_module_version": "1.2.0", + "_model_name": "LayoutModel", + "_view_count": null, + "_view_module": "@jupyter-widgets/base", + "_view_module_version": "1.2.0", + "_view_name": "LayoutView", + "align_content": null, + "align_items": null, + "align_self": null, + "border": null, + "bottom": null, + "display": null, + "flex": null, + "flex_flow": null, + "grid_area": null, + "grid_auto_columns": null, + "grid_auto_flow": null, + "grid_auto_rows": null, + "grid_column": null, + "grid_gap": null, + "grid_row": null, + "grid_template_areas": null, + "grid_template_columns": null, + "grid_template_rows": null, + "height": null, + "justify_content": null, + "justify_items": null, + "left": null, + "margin": null, + "max_height": null, + "max_width": null, + "min_height": null, + "min_width": null, + "object_fit": null, + "object_position": null, + "order": null, + "overflow": null, + "overflow_x": null, + "overflow_y": null, + "padding": null, + "right": null, + "top": null, + "visibility": null, + "width": null + } + }, + "e4568f72971749ac92cc5577cb9c4f9c": { + "model_module": "@jupyter-widgets/controls", + "model_module_version": "1.5.0", + "model_name": "FloatProgressModel", + "state": { + "_dom_classes": [], + "_model_module": "@jupyter-widgets/controls", + "_model_module_version": "1.5.0", + "_model_name": "FloatProgressModel", + "_view_count": null, + "_view_module": "@jupyter-widgets/controls", + "_view_module_version": "1.5.0", + "_view_name": "ProgressView", + "bar_style": "success", + "description": "", + "description_tooltip": null, + "layout": "IPY_MODEL_ceebda1d07b545f5b605551c08e73be8", + "max": 456318, + "min": 0, + "orientation": "horizontal", + "style": "IPY_MODEL_6aa96b80e3b746b48f5c13bcafe1134e", + "value": 456318 + } }, - "id": "Kdcq4oCNmEua", - "outputId": "e6d33c07-d22a-41af-b84a-91c9681c7524" - }, - "execution_count": null, - "outputs": [ - { - "output_type": "stream", - "name": "stdout", - "text": [ - "Original token IDs: tensor([[ 198, 12, 417, 8765, 318, 257, 262, 3504, 7372, 6342]])\n", - "Intervention token IDs: tensor([[ 262, 12, 417, 8765, 11, 257, 262, 3504, 338, 3576]])\n", - "Original prediction: Paris\n", - "Intervention prediction: London\n" - ] - } - ] - }, - { - "cell_type": "markdown", - "source": [ - "So it did end up affecting what the model predicted. That's pretty neat!\n", - "\n", - "Another cool thing with multiple invokes is that the Proxies can interact between them. Here we transfer the word token embeddings from a real prompt into another placeholder prompt. Therefore the latter prompt produces the output of the former prompt:" - ], - "metadata": { - "id": "8BsybgsK2Bbr" - } - }, - { - "cell_type": "code", - "source": [ - "with model.trace() as tracer:\n", - "\n", - " with tracer.invoke(\"The Eiffel Tower is in the city of\"):\n", - "\n", - " embeddings = model.transformer.wte.output\n", - "\n", - " with tracer.invoke(\"_ _ _ _ _ _ _ _ _ _\"):\n", - "\n", - " model.transformer.wte.output = embeddings\n", - "\n", - " token_ids_intervention = model.lm_head.output.argmax(dim=-1).save()\n", - "\n", - " with tracer.invoke(\"_ _ _ _ _ _ _ _ _ _\"):\n", - "\n", - " token_ids_original = model.lm_head.output.argmax(dim=-1).save()\n", - "\n", - "print(\"Original prediction:\", model.tokenizer.decode(token_ids_original[0][-1]))\n", - "print(\"Intervention prediction:\", model.tokenizer.decode(token_ids_intervention[0][-1]))" - ], - "metadata": { - "colab": { - "base_uri": "https://localhost:8080/" + "e51620404d244a4e806eeaf3b96a474f": { + "model_module": "@jupyter-widgets/controls", + "model_module_version": "1.5.0", + "model_name": "HTMLModel", + "state": { + "_dom_classes": [], + "_model_module": "@jupyter-widgets/controls", + "_model_module_version": "1.5.0", + "_model_name": "HTMLModel", + "_view_count": null, + "_view_module": "@jupyter-widgets/controls", + "_view_module_version": "1.5.0", + "_view_name": "HTMLView", + "description": "", + "description_tooltip": null, + "layout": "IPY_MODEL_f28f557210b7402b8b2d0fd83f038dc1", + "placeholder": "", + "style": "IPY_MODEL_bc8190c700e949ebb6f874c8f8055161", + "value": "merges.txt: 100%" + } }, - "id": "syKTI_KhpvCY", - "outputId": "b2238c5c-d2d8-47bf-b73a-730f1e03cf99" - }, - "execution_count": null, - "outputs": [ - { - "output_type": "stream", - "name": "stdout", - "text": [ - "Original prediction: _\n", - "Intervention prediction: Paris\n" - ] - } - ] - }, - { - "cell_type": "markdown", - "source": [ - "## .next()\n", - "\n", - "Some HuggingFace models define methods to generate multiple outputs at a time. `LanguageModel` wraps that functionality to provide the same tracing features by using `.generate(...)` instead of `.trace(...)`. This calls the underlying model's `.generate` method. It passes the output through a `model.generator` module that we've added onto the model, allowing you to get the generate output at `model.generator.output`.\n", - "\n", - "In a case like this, the underlying model is called more than once; the modules of said model produce more than one output. Which iteration should a given `module.output` refer to? That's where `Module.next()` comes in.\n", - "\n", - "Each module has a call idx associated with it and `.next()` simply increments that attribute. At the time of execution, data is injected into the intervention graph only at the iteration that matches the call idx." - ], - "metadata": { - "id": "vvWA-CWqQtah" - } - }, - { - "cell_type": "code", - "source": [ - "with model.generate(\"The Eiffel Tower is in the city of\", max_new_tokens=3):\n", - "\n", - " token_ids_1 = model.lm_head.output.argmax(dim=-1).save()\n", - "\n", - " token_ids_2 = model.lm_head.next().output.argmax(dim=-1).save()\n", - "\n", - " token_ids_3 = model.lm_head.next().output.argmax(dim=-1).save()\n", - "\n", - " output = model.generator.output.save()\n", - "\n", - "print(\"Prediction 1: \", model.tokenizer.decode(token_ids_1[0][-1]))\n", - "print(\"Prediction 2: \", model.tokenizer.decode(token_ids_2[0][-1]))\n", - "print(\"Prediction 3: \", model.tokenizer.decode(token_ids_3[0][-1]))\n", - "\n", - "print(\"All token ids: \", output)\n", - "\n", - "print(\"All prediction: \", model.tokenizer.batch_decode(output))" - ], - "metadata": { - "colab": { - "base_uri": "https://localhost:8080/" + "e7a73cfc3fa04425afc08368532ef0be": { + "model_module": "@jupyter-widgets/controls", + "model_module_version": "1.5.0", + "model_name": "HTMLModel", + "state": { + "_dom_classes": [], + "_model_module": "@jupyter-widgets/controls", + "_model_module_version": "1.5.0", + "_model_name": "HTMLModel", + "_view_count": null, + "_view_module": "@jupyter-widgets/controls", + "_view_module_version": "1.5.0", + "_view_name": "HTMLView", + "description": "", + "description_tooltip": null, + "layout": "IPY_MODEL_b2748df9aff540df954f002310606a56", + "placeholder": "", + "style": "IPY_MODEL_f4caf2bc58704464b6ac7bf6da31b71d", + "value": "vocab.json: 100%" + } + }, + "eb0c489b913744a28304e5334bb66532": { + "model_module": "@jupyter-widgets/base", + "model_module_version": "1.2.0", + "model_name": "LayoutModel", + "state": { + "_model_module": "@jupyter-widgets/base", + "_model_module_version": "1.2.0", + "_model_name": "LayoutModel", + "_view_count": null, + "_view_module": "@jupyter-widgets/base", + "_view_module_version": "1.2.0", + "_view_name": "LayoutView", + "align_content": null, + "align_items": null, + "align_self": null, + "border": null, + "bottom": null, + "display": null, + "flex": null, + "flex_flow": null, + "grid_area": null, + "grid_auto_columns": null, + "grid_auto_flow": null, + "grid_auto_rows": null, + "grid_column": null, + "grid_gap": null, + "grid_row": null, + "grid_template_areas": null, + "grid_template_columns": null, + "grid_template_rows": null, + "height": null, + "justify_content": null, + "justify_items": null, + "left": null, + "margin": null, + "max_height": null, + "max_width": null, + "min_height": null, + "min_width": null, + "object_fit": null, + "object_position": null, + "order": null, + "overflow": null, + "overflow_x": null, + "overflow_y": null, + "padding": null, + "right": null, + "top": null, + "visibility": null, + "width": null + } + }, + "f28f557210b7402b8b2d0fd83f038dc1": { + "model_module": "@jupyter-widgets/base", + "model_module_version": "1.2.0", + "model_name": "LayoutModel", + "state": { + "_model_module": "@jupyter-widgets/base", + "_model_module_version": "1.2.0", + "_model_name": "LayoutModel", + "_view_count": null, + "_view_module": "@jupyter-widgets/base", + "_view_module_version": "1.2.0", + "_view_name": "LayoutView", + "align_content": null, + "align_items": null, + "align_self": null, + "border": null, + "bottom": null, + "display": null, + "flex": null, + "flex_flow": null, + "grid_area": null, + "grid_auto_columns": null, + "grid_auto_flow": null, + "grid_auto_rows": null, + "grid_column": null, + "grid_gap": null, + "grid_row": null, + "grid_template_areas": null, + "grid_template_columns": null, + "grid_template_rows": null, + "height": null, + "justify_content": null, + "justify_items": null, + "left": null, + "margin": null, + "max_height": null, + "max_width": null, + "min_height": null, + "min_width": null, + "object_fit": null, + "object_position": null, + "order": null, + "overflow": null, + "overflow_x": null, + "overflow_y": null, + "padding": null, + "right": null, + "top": null, + "visibility": null, + "width": null + } }, - "id": "Yy5Z9NE1GkaN", - "outputId": "0b4c1bd3-a6f6-4ba1-9cb8-6019c222311f" - }, - "execution_count": null, - "outputs": [ - { - "output_type": "stream", - "name": "stderr", - "text": [ - "Setting `pad_token_id` to `eos_token_id`:50256 for open-end generation.\n", - "A decoder-only architecture is being used, but right-padding was detected! For correct generation results, please set `padding_side='left'` when initializing the tokenizer.\n", - "Setting `pad_token_id` to `eos_token_id`:50256 for open-end generation.\n" - ] + "f3e24aa6619e4185a3e2f53f2e473e45": { + "model_module": "@jupyter-widgets/base", + "model_module_version": "1.2.0", + "model_name": "LayoutModel", + "state": { + "_model_module": "@jupyter-widgets/base", + "_model_module_version": "1.2.0", + "_model_name": "LayoutModel", + "_view_count": null, + "_view_module": "@jupyter-widgets/base", + "_view_module_version": "1.2.0", + "_view_name": "LayoutView", + "align_content": null, + "align_items": null, + "align_self": null, + "border": null, + "bottom": null, + "display": null, + "flex": null, + "flex_flow": null, + "grid_area": null, + "grid_auto_columns": null, + "grid_auto_flow": null, + "grid_auto_rows": null, + "grid_column": null, + "grid_gap": null, + "grid_row": null, + "grid_template_areas": null, + "grid_template_columns": null, + "grid_template_rows": null, + "height": null, + "justify_content": null, + "justify_items": null, + "left": null, + "margin": null, + "max_height": null, + "max_width": null, + "min_height": null, + "min_width": null, + "object_fit": null, + "object_position": null, + "order": null, + "overflow": null, + "overflow_x": null, + "overflow_y": null, + "padding": null, + "right": null, + "top": null, + "visibility": null, + "width": null + } }, - { - "output_type": "stream", - "name": "stdout", - "text": [ - "Prediction 1: Paris\n", - "Prediction 2: ,\n", - "Prediction 3: and\n", - "All token ids: tensor([[ 464, 412, 733, 417, 8765, 318, 287, 262, 1748, 286, 6342, 11,\n", - " 290]])\n", - "All prediction: ['The Eiffel Tower is in the city of Paris, and']\n" - ] + "f4caf2bc58704464b6ac7bf6da31b71d": { + "model_module": "@jupyter-widgets/controls", + "model_module_version": "1.5.0", + "model_name": "DescriptionStyleModel", + "state": { + "_model_module": "@jupyter-widgets/controls", + "_model_module_version": "1.5.0", + "_model_name": "DescriptionStyleModel", + "_view_count": null, + "_view_module": "@jupyter-widgets/base", + "_view_module_version": "1.2.0", + "_view_name": "StyleView", + "description_width": "" + } + }, + "f693dfed91df492c986afbd4ff925f89": { + "model_module": "@jupyter-widgets/controls", + "model_module_version": "1.5.0", + "model_name": "DescriptionStyleModel", + "state": { + "_model_module": "@jupyter-widgets/controls", + "_model_module_version": "1.5.0", + "_model_name": "DescriptionStyleModel", + "_view_count": null, + "_view_module": "@jupyter-widgets/base", + "_view_module_version": "1.2.0", + "_view_name": "StyleView", + "description_width": "" + } } - ] - }, - { - "cell_type": "markdown", - "source": [ - "# 3 I thought you said huge models?" - ], - "metadata": { - "id": "rpkX-LwBQZHo" - } - }, - { - "cell_type": "markdown", - "source": [ - "`NNsight` is only one part of our project to democratize access to AI internals. The other half is `NDIF` (National Deep Inference Facility).\n", - "\n", - "The interaction between the two is fairly straightforward. The `intervention graph` we create via the tracing context can be encoded into a custom json format and sent via an http request to the `NDIF` servers. `NDIF` then decodes the `intervention graph` and `interleaves` it alongside the specified model.\n", - "\n", - "To see which models are currently being hosted, check out the following status page: https://nnsight.net/status/" - ], - "metadata": { - "id": "uCZR65VmILEb" - } - }, - { - "cell_type": "markdown", - "source": [ - "## Remote execution" - ], - "metadata": { - "id": "65Ks1LUvQaER" - } - }, - { - "cell_type": "markdown", - "source": [ - "In its current state, `NDIF` requires you to recieve an API key. Therefore, to run the rest of this colab, you would need one of your own. To get one, simply join the [NDIF discord](https://discord.gg/6uFJmCSwW7) and introduce yourself on the `#introductions` channel. Then DM either @JadenFK or @caden and we'll create one for you.\n", - "\n", - "Once you have one, to register your api key with `nnsight`, do the following:" - ], - "metadata": { - "id": "Aa-ZuVOFKS5k" - } - }, - { - "cell_type": "code", - "source": [ - "from nnsight import CONFIG\n", - "\n", - "CONFIG.set_default_api_key(\"<your api key here>\")" - ], - "metadata": { - "id": "ax1NWoS9MJeZ" - }, - "execution_count": null, - "outputs": [] - }, - { - "cell_type": "markdown", - "source": [ - "This only needs to be run once as it will save this api key as the default in a config file along with the `nnsight` installation.\n", - "\n", - "To amp things up a few levels, let's demonstrate using `nnsight`'s tracing context with one of the larger open source language models, `Llama-2-70b`!" - ], - "metadata": { - "id": "mvXLOh65MXmF" - } - }, - { - "cell_type": "code", - "source": [ - "import os\n", - "\n", - "# llama2 70b is a gated model and you need access via your huggingface token\n", - "os.environ['HF_TOKEN'] = \"<your huggingface token>\"\n", - "\n", - "# llama response object requires the version of transformers from github\n", - "!pip uninstall -y transformers\n", - "!pip install git+https://github.com/huggingface/transformers\n", - "\n", - "clear_output()" - ], - "metadata": { - "id": "Q1N04sJPZnJt" - }, - "execution_count": null, - "outputs": [] - }, - { - "cell_type": "code", - "source": [ - "# We'll never actually load the parameters so no need to specify a device_map.\n", - "model = LanguageModel(\"meta-llama/Llama-2-70b-hf\")\n", - "\n", - "# All we need to specify using NDIF vs executing locally is remote=True.\n", - "with model.trace('The Eiffel Tower is in the city of', remote=True) as runner:\n", - "\n", - " hidden_states = model.model.layers[-1].output.save()\n", - "\n", - " output = model.output.save()\n", - "\n", - "print(hidden_states)\n", - "\n", - "print(output['logits'])" - ], - "metadata": { - "id": "oqVzjNoyNGc7" - }, - "execution_count": null, - "outputs": [] - }, - { - "cell_type": "markdown", - "source": [ - "It really is as simple as `remote=True`. All of the techniques we went through in earlier sections work just the same when running locally and remotely.\n", - "\n", - "Note that both `nnsight`, but especially `NDIF`, is in active development and therefore there may be caveats, changes, and errors to work through." - ], - "metadata": { - "id": "IceVpnZcZ9F0" - } - }, - { - "cell_type": "markdown", - "source": [ - "# Getting Involved!" - ], - "metadata": { - "id": "P3zRm-7VRRov" - } - }, - { - "cell_type": "markdown", - "source": [ - "If you're interested in following updates to `nnsight`, contributing, giving feedback, or finding collaborators, please join the [NDIF discord](https://discord.gg/6uFJmCSwW7)!\n", - "\n", - "The [Mech Interp discord](https://discord.gg/km2RQBzaUn) is also a fantastic place to discuss all things mech interp with a really cool community.\n", - "\n", - "Our website [nnsight.net](https://nnsight.net/), has a bunch more tutorials detailing more complex interpretability techniques using `nnsight`. If you want to share any of the work you do using `nnsight`, let others know on either of the discords above and we might turn it into a tutorial on our website.\n", - "\n", - "💟\n", - "\n", - "\n", - "\n" - ], - "metadata": { - "id": "BnCc9xgjvxEP" } } - ] -} \ No newline at end of file + }, + "nbformat": 4, + "nbformat_minor": 0 +} diff --git a/README.md b/README.md index 195f14e4..fc8d9b8c 100644 --- a/README.md +++ b/README.md @@ -11,7 +11,7 @@ The `nnsight` package enables interpreting and manipulating the internals of de Install this package through pip by running: -`pip install nnsight==0.2.3.dev0` +`pip install nnsight` #### Examples diff --git a/docs/.DS_Store b/docs/.DS_Store index a7839fe9..78f1ef6b 100644 Binary files a/docs/.DS_Store and b/docs/.DS_Store differ diff --git a/docs/contributing.md b/docs/contributing.md new file mode 100644 index 00000000..9af41725 --- /dev/null +++ b/docs/contributing.md @@ -0,0 +1,27 @@ +# Installation + +Run `pip install -r requirements.txt` while in the `nnsight/docs` directory. + +# Adding Tutorials + +Tutorials are written as Jupyter Notebooks in `.ipynb` format, and automatically converted to html by the `nbsphinx` extension. + +The only requirement of `nbsphinx` is that you add a header to the notebook to act as the rendered title on Sphinx docs. To do this, create a markdown cell at the top of your notebook with a header `#`. + +Then, just add your notebook to the `nnsight/docs/source/notebooks/tutorials` directory. + +If you're adding a new notebook, navigate to `nnsight/docs/source/tutorials.rst`. At the bottom of the page, add the path to your notebook under the `toctree`. + +``` +.. toctree:: + :maxdepth: 1 + + notebooks/tutorials/walkthrough.ipynb + ... + <YOUR NOTEBOOK PATH> + +``` + +# Compiling Sphinx to HTML + +Run `make dirhtml` from the `nnsight/docs` directory. The build is located in `nnsight/docs/build/dirhtml`. You can run `python3 -m http.server <PORT>` from that directory, and see the site at `http://localhost:<PORT>`. diff --git a/docs/requirements.txt b/docs/requirements.txt index a49f80f4..cf864581 100644 --- a/docs/requirements.txt +++ b/docs/requirements.txt @@ -1,6 +1,8 @@ nnsight pydata_sphinx_theme sphinx -sphinx_charts sphinx-copybutton +sphinx-design +nbsphinx plotly +ipython \ No newline at end of file diff --git a/docs/source/.DS_Store b/docs/source/.DS_Store index d39d1092..836f5368 100644 Binary files a/docs/source/.DS_Store and b/docs/source/.DS_Store differ diff --git a/docs/source/_static/.DS_Store b/docs/source/_static/.DS_Store index 3548ed0e..9ea5ea8f 100644 Binary files a/docs/source/_static/.DS_Store and b/docs/source/_static/.DS_Store differ diff --git a/docs/source/_static/css/custom.css b/docs/source/_static/css/custom.css index a5ece848..db595237 100644 --- a/docs/source/_static/css/custom.css +++ b/docs/source/_static/css/custom.css @@ -1,3 +1,5 @@ +/* Website light and dark theme colors */ + html[data-theme="dark"] { --pst-color-secondary: #ce5d97; --pst-color-primary: #4385be; @@ -13,10 +15,29 @@ html[data-theme="dark"] { --pst-color-panel-background: #1C1B1A; } +html[data-theme="light"] { + --pst-color-secondary: #ce5d97; + --pst-color-primary: #4385be; + --pst-color-primary-highlight: #ce5d97; + --pst-color-background: #FFFCF0; + --pst-color-on-background: #F2F0E5; + --pst-color-surface: #E6E4D9; + --pst-color-on-surface: #100F0F; + --pst-color-text-base: #100F0F; + --pst-color-text-muted: #100F0F; + --pst-color-heading-color: #100F0F; + --pst-color-border: #F2F0E5; + --pst-color-panel-background: #F2F0E5; +} + +/* Editing the default width for pydata content */ + .bd-main .bd-content .bd-article-container { max-width: 80%; /* default is 60em */ } +/* Front page buttons and hero */ + .hero { height: 60vh !important; } @@ -39,34 +60,21 @@ html[data-theme="dark"] { color: var(--pst-color-on-surface) !important; } -.custom-card { - background-color : var(--pst-color-on-background) !important; - border-color : #878580 !important; - color: #878580 !important; -} +/* Hide the theme version note at the bottom of the page */ -.custom-card:hover, -.custom-card:hover h5, -.custom-card:hover p { - border-color: var(--pst-color-on-surface) !important; - color: var(--pst-color-on-surface) !important; +.theme-version { + visibility: hidden; } - - .text { color : var(--pst-color-on-surface) !important; } +/* Default surface style for cards */ + .surface { background-color : var(--pst-color-surface) !important; color : var(--pst-color-on-surface) !important; border-color: var(--pst-color-surface) !important; } - -pre { - border-color: var(--pst-color-surface) !important; -} - - diff --git a/docs/source/_static/css/front.css b/docs/source/_static/css/front.css index eb29888f..db44e323 100644 --- a/docs/source/_static/css/front.css +++ b/docs/source/_static/css/front.css @@ -1,26 +1,19 @@ /* Custom Styles for List Group Items */ .list-group-item { background-color: transparent; - /* Make the button background transparent */ border: none; - /* Remove border */ color: var(--pst-color-on-surface); - /* Set text color */ } .list-group-item:hover { background-color: var(--pst-color-on-background); color: var(--pst-color-on-surface); - /* Light grey hover effect, adjust as needed */ } .list-group-item.active { background-color: var(--pst-color-surface); - /* White background for active state */ color: var(--pst-color-on-surface); - /* Text color for active state */ border: none; - /* Ensure no border for active state */ } .tab-background { @@ -34,36 +27,23 @@ height: 100% !important; } -@media (max-width: 767px) { /* Targets screens smaller than the md breakpoint */ - .mid-section { - height: 50vh !important; - } - .static-on-small { - height: 35vh !important; - } -} - - +/* Middle section blurb */ .mid-section { + margin-top: 5vh; height: 20vh; -} - -.middle-stuff { margin-bottom: 15vh; } + +/* Custom Styles for Code Widget */ .code-surface { background-color : var(--pst-color-surface) !important; color : var(--pst-color-on-surface) !important; border-color: var(--pst-color-surface) !important; } - - -/* Custom Styles for Code Styles */ - .fill-height { height: 100% !important; } @@ -71,20 +51,3 @@ pre { margin: 0 !important; } - - -.hover-border { - transition: border 0.3s ease-in-out; -} - -.hover-border:hover { - border: 1px solid #fff; /* Adjust color to match the design */ -} - - -.center-text { - display: flex; - justify-content: center; - align-items: center; - text-align: center; -} diff --git a/docs/source/_static/css/hljs.css b/docs/source/_static/css/hljs-dark.css similarity index 94% rename from docs/source/_static/css/hljs.css rename to docs/source/_static/css/hljs-dark.css index a5b3fdf9..82abee3b 100644 --- a/docs/source/_static/css/hljs.css +++ b/docs/source/_static/css/hljs-dark.css @@ -1,3 +1,5 @@ +/* Custom Highlight.js styles for the front page snippet */ + .hljs { display: block !important; overflow-x: auto !important; diff --git a/docs/source/_static/css/hljs-light.css b/docs/source/_static/css/hljs-light.css new file mode 100644 index 00000000..c04c8e42 --- /dev/null +++ b/docs/source/_static/css/hljs-light.css @@ -0,0 +1,84 @@ +/* + +Orginal Style from ethanschoonover.com/solarized (c) Jeremy Hull <[email protected]> + +*/ + +.hljs { + display: block; + overflow-x: auto; + padding: 0.5em; + background: var(--pst-color-surface); + color: var(--pst-color-on-surface); + } + + .hljs-comment, + .hljs-quote { + color: #93a1a1; + } + + /* Solarized Green */ + .hljs-keyword, + .hljs-selector-tag, + .hljs-addition { + color: #859900; + } + + /* Solarized Cyan */ + .hljs-number, + .hljs-string, + .hljs-meta .hljs-meta-string, + .hljs-literal, + .hljs-doctag, + .hljs-regexp { + color: #2aa198; + } + + /* Solarized Blue */ + .hljs-title, + .hljs-section, + .hljs-name, + .hljs-selector-id, + .hljs-selector-class { + color: #268bd2; + } + + /* Solarized Yellow */ + .hljs-attribute, + .hljs-attr, + .hljs-variable, + .hljs-template-variable, + .hljs-class .hljs-title, + .hljs-type { + color: #b58900; + } + + /* Solarized Orange */ + .hljs-symbol, + .hljs-bullet, + .hljs-subst, + .hljs-meta, + .hljs-meta .hljs-keyword, + .hljs-selector-attr, + .hljs-selector-pseudo, + .hljs-link { + color: #cb4b16; + } + + /* Solarized Red */ + .hljs-built_in, + .hljs-deletion { + color: #dc322f; + } + + .hljs-formula { + background: #eee8d5; + } + + .hljs-emphasis { + font-style: italic; + } + + .hljs-strong { + font-weight: bold; + } \ No newline at end of file diff --git a/docs/source/_static/css/status.css b/docs/source/_static/css/status.css index 3979c12c..7f5f243f 100644 --- a/docs/source/_static/css/status.css +++ b/docs/source/_static/css/status.css @@ -1,7 +1,10 @@ -.status-body { - color: "white"; -} - -.accordion-header { - margin: 0 !important; -} \ No newline at end of file +.custom-accordion-header { + background-color: #007bff; /* Bootstrap's primary color for demonstration */ + color: white; /* Text color */ + } + + .custom-accordion-body { + background-color: #f8f9fa; /* A light gray background */ + color: #212529; /* Bootstrap default text color */ + } + \ No newline at end of file diff --git a/docs/source/_static/css/tutorials.css b/docs/source/_static/css/tutorials.css deleted file mode 100644 index 2349f603..00000000 --- a/docs/source/_static/css/tutorials.css +++ /dev/null @@ -1,16 +0,0 @@ -.tutorials-cards { - font-size: .9rem; -} - -.tutorial-card-section { - overflow: scroll; - height: 50vh; -} - -.tutorial-card-section::-webkit-scrollbar { - display: none; -} - -.page-info { - z-index: 100; -} \ No newline at end of file diff --git a/docs/source/_static/images/.DS_Store b/docs/source/_static/images/.DS_Store index c671bc88..1847e05f 100644 Binary files a/docs/source/_static/images/.DS_Store and b/docs/source/_static/images/.DS_Store differ diff --git a/docs/source/_static/images/execution.png b/docs/source/_static/images/execution.png new file mode 100644 index 00000000..f9094444 Binary files /dev/null and b/docs/source/_static/images/execution.png differ diff --git a/docs/source/_static/images/interleaved.png b/docs/source/_static/images/interleaved.png new file mode 100644 index 00000000..6d17f285 Binary files /dev/null and b/docs/source/_static/images/interleaved.png differ diff --git a/docs/source/_static/images/intrgraph.png b/docs/source/_static/images/intrgraph.png deleted file mode 100644 index 3bb50229..00000000 Binary files a/docs/source/_static/images/intrgraph.png and /dev/null differ diff --git a/docs/source/_static/images/remote_execution.png b/docs/source/_static/images/remote_execution.png deleted file mode 100644 index 7020d5b9..00000000 Binary files a/docs/source/_static/images/remote_execution.png and /dev/null differ diff --git a/docs/source/_static/images/tutorials/.DS_Store b/docs/source/_static/images/tutorials/.DS_Store index ba36697a..cc81a870 100644 Binary files a/docs/source/_static/images/tutorials/.DS_Store and b/docs/source/_static/images/tutorials/.DS_Store differ diff --git a/docs/source/_static/images/tutorials/activation_patching.png b/docs/source/_static/images/tutorials/activation_patching.png deleted file mode 100644 index 4f82b933..00000000 Binary files a/docs/source/_static/images/tutorials/activation_patching.png and /dev/null differ diff --git a/docs/source/_static/images/tutorials/activation_patching.webp b/docs/source/_static/images/tutorials/activation_patching.webp new file mode 100644 index 00000000..05604c7f Binary files /dev/null and b/docs/source/_static/images/tutorials/activation_patching.webp differ diff --git a/docs/source/_static/images/tutorials/attribution_patching.png b/docs/source/_static/images/tutorials/attribution_patching.png deleted file mode 100644 index 4f012886..00000000 Binary files a/docs/source/_static/images/tutorials/attribution_patching.png and /dev/null differ diff --git a/docs/source/_static/images/tutorials/attribution_patching.webp b/docs/source/_static/images/tutorials/attribution_patching.webp new file mode 100644 index 00000000..abd71d1d Binary files /dev/null and b/docs/source/_static/images/tutorials/attribution_patching.webp differ diff --git a/docs/source/_static/images/tutorials/dictionary_learning.png b/docs/source/_static/images/tutorials/dictionary_learning.png deleted file mode 100644 index 40440864..00000000 Binary files a/docs/source/_static/images/tutorials/dictionary_learning.png and /dev/null differ diff --git a/docs/source/_static/images/tutorials/dictionary_learning.webp b/docs/source/_static/images/tutorials/dictionary_learning.webp new file mode 100644 index 00000000..00a4cb45 Binary files /dev/null and b/docs/source/_static/images/tutorials/dictionary_learning.webp differ diff --git a/docs/source/_static/images/tutorials/function_vectors.png b/docs/source/_static/images/tutorials/function_vectors.png deleted file mode 100644 index bee6d3fa..00000000 Binary files a/docs/source/_static/images/tutorials/function_vectors.png and /dev/null differ diff --git a/docs/source/_static/images/tutorials/function_vectors.webp b/docs/source/_static/images/tutorials/function_vectors.webp new file mode 100644 index 00000000..041b1db0 Binary files /dev/null and b/docs/source/_static/images/tutorials/function_vectors.webp differ diff --git a/docs/source/_static/images/tutorials/future_lens.png b/docs/source/_static/images/tutorials/future_lens.png deleted file mode 100644 index 4d7e0cf3..00000000 Binary files a/docs/source/_static/images/tutorials/future_lens.png and /dev/null differ diff --git a/docs/source/_static/images/tutorials/future_lens.webp b/docs/source/_static/images/tutorials/future_lens.webp new file mode 100644 index 00000000..19e29c79 Binary files /dev/null and b/docs/source/_static/images/tutorials/future_lens.webp differ diff --git a/docs/source/_static/images/tutorials/logit_lens.png b/docs/source/_static/images/tutorials/logit_lens.png deleted file mode 100644 index da8af979..00000000 Binary files a/docs/source/_static/images/tutorials/logit_lens.png and /dev/null differ diff --git a/docs/source/_static/images/tutorials/logit_lens.webp b/docs/source/_static/images/tutorials/logit_lens.webp new file mode 100644 index 00000000..24d1beca Binary files /dev/null and b/docs/source/_static/images/tutorials/logit_lens.webp differ diff --git a/docs/source/_static/images/tutorials/walkthrough.png b/docs/source/_static/images/tutorials/walkthrough.png deleted file mode 100644 index c07ac8f4..00000000 Binary files a/docs/source/_static/images/tutorials/walkthrough.png and /dev/null differ diff --git a/docs/source/_static/images/tutorials/walkthrough.webp b/docs/source/_static/images/tutorials/walkthrough.webp new file mode 100644 index 00000000..15e5b375 Binary files /dev/null and b/docs/source/_static/images/tutorials/walkthrough.webp differ diff --git a/docs/source/_static/js/code.js b/docs/source/_static/js/code.js index 0d18c802..c4b8da0e 100644 --- a/docs/source/_static/js/code.js +++ b/docs/source/_static/js/code.js @@ -46,4 +46,37 @@ document.addEventListener('DOMContentLoaded', (event) => { document.querySelectorAll('code.language-python.multi').forEach(el => { el.textContent = multi; }); + document.querySelectorAll('div.output_area.stderr').forEach(el => { + el.style.visibility = "hidden"; + el.style.display = "none"; + }); + document.querySelectorAll('div.output_area.stderr').forEach(el => { + el.style.visibility = "hidden"; + el.style.display = "none"; + }); + document.querySelectorAll('div.output_area').forEach(el => { + el.style.marginBottom = "1em"; + }); + document.querySelectorAll('div.input_area').forEach(el => { + el.style.border = "0px"; + el.style.marginTop = "0.5em"; + el.style.marginBottom = "0.5em"; + }); + document.querySelectorAll('summary').forEach(el => { + el.style.marginTop = "1em"; + el.style.marginBottom = "1em"; + document.querySelectorAll('p').forEach(p => { + const span = document.createElement('span'); + + // Copy the innerHTML from the to the + span.innerHTML = p.innerHTML; + + // Optional: Copy styles from the to the + span.style.cssText = p.style.cssText; + + // Replace the with the in the DOM + p.parentNode.replaceChild(span, p); + }); + }); }); + diff --git a/docs/source/_static/js/custom.js b/docs/source/_static/js/custom.js index a72b393d..8d712853 100644 --- a/docs/source/_static/js/custom.js +++ b/docs/source/_static/js/custom.js @@ -1,85 +1,115 @@ require.config({ - paths: { - 'three': 'https://cdnjs.cloudflare.com/ajax/libs/three.js/r134/three.min.js?MEH=BLAH', - 'VANTA': 'https://cdn.jsdelivr.net/npm/[email protected]/dist/vanta.dots.min.js?MEH=BLAH' - }, - shim: { - 'VANTA': { - deps: ['three'] // Removed the exports line - } + paths: { + 'three': 'https://cdnjs.cloudflare.com/ajax/libs/three.js/r134/three.min.js?MEH=BLAH', + 'VANTA': 'https://cdn.jsdelivr.net/npm/[email protected]/dist/vanta.dots.min.js?MEH=BLAH' + }, + shim: { + 'VANTA': { + deps: ['three'] } + } }); -// require(['three', 'VANTA'], function(THREE) { -// // Assuming VANTA attaches itself to a global variable, directly use it here -// // Check the correct global variable name (VANTA, _vantaEffect, etc.) -// if (window.VANTA && window.VANTA.WAVES) { -// window.VANTA.WAVES({ -// el: "#hero" -// }); -// } else { -// console.error('VANTA.WAVES is not available.'); -// } -// }); - -require(['three'], function(THREE) { +var vantaEffect; + +function initVanta() { + require(['three'], function (THREE) { window.THREE = THREE; // Make THREE globally accessible - require(['VANTA'], function() { - if (VANTA.DOTS) { - VANTA.DOTS({ - el: "#canvas", - mouseControls: false, - touchControls: false, - gyroControls: false, - minHeight: 200.00, - minWidth: 200.00, - scale: 1.00, - scaleMobile: 1.00, - color: 0xb3a094, - color2: 0xffffff, - backgroundColor: 0x100F0F, - size: 0.50, - // spacing: 68.00, - spacing: 10.00, - showLines: false - }) - } + require(['VANTA'], function () { + if (vantaEffect) { + vantaEffect.destroy(); + } + + const darkMode = document.documentElement.dataset.theme === 'dark'; + const color = darkMode ? 0xCECDC3 : 0x100F0F; // Example dark mode color + const color2 = darkMode ? 0xCECDC3 : 0x100F0F; // Adjust color2 for dark mode as needed + const backgroundColor = darkMode ? 0x100F0F : 0xFFFCF0; // Adjust background color for dark mode + + vantaEffect = VANTA.DOTS({ + el: "#canvas", + mouseControls: false, + touchControls: false, + gyroControls: false, + minHeight: 200.00, + minWidth: 200.00, + scale: 1.00, + scaleMobile: 1.00, + color: color, + color2: color2, + backgroundColor: backgroundColor, + size: 0.50, + spacing: 10.00, + showLines: false + }); }); + }); +} + +// Initial call +initVanta(); + +// Setup a mutation observer to detect theme changes +var observer = new MutationObserver(function(mutations) { + mutations.forEach(function(mutation) { + if (mutation.attributeName === "data-theme") { + initVanta(); // Reinitialize Vanta with the new theme settings + } + }); }); +// Start observing the document element for attribute changes +observer.observe(document.documentElement, {attributes: true, attributeFilter: ['data-theme']}); + + document.addEventListener('DOMContentLoaded', (event) => { - var listItems = document.querySelectorAll('#list-tab .list-group-item'); - listItems.forEach(function(listItem) { - listItem.addEventListener('click', function(e) { - e.preventDefault(); - var tabPaneId = this.getAttribute('href'); - - // Remove active class from all list items - listItems.forEach(function(item) { - item.classList.remove('active'); - }); - - // Add active class to the clicked list item - this.classList.add('active'); - - // Hide all tab panes - var tabPanes = document.querySelectorAll('.tab-content .tab-pane'); - tabPanes.forEach(function(tabPane) { - tabPane.classList.remove('show'); - tabPane.classList.remove('active'); - }); - - // Show the clicked tab pane - var activeTabPane = document.querySelector(tabPaneId); - activeTabPane.classList.add('active'); - - // We need to manually add 'show' class after a slight delay to trigger fade effect - setTimeout(function() { - activeTabPane.classList.add('show'); - }, 150); // Adjust delay as needed + var listItems = document.querySelectorAll('#list-tab .list-group-item'); + listItems.forEach(function (listItem) { + listItem.addEventListener('click', function (e) { + e.preventDefault(); + var tabPaneId = this.getAttribute('href'); + + // Remove active class from all list items + listItems.forEach(function (item) { + item.classList.remove('active'); }); + + // Add active class to the clicked list item + this.classList.add('active'); + + // Hide all tab panes + var tabPanes = document.querySelectorAll('.tab-content .tab-pane'); + tabPanes.forEach(function (tabPane) { + tabPane.classList.remove('show'); + tabPane.classList.remove('active'); + }); + + // Show the clicked tab pane + var activeTabPane = document.querySelector(tabPaneId); + activeTabPane.classList.add('active'); + + // We need to manually add 'show' class after a slight delay to trigger fade effect + setTimeout(function () { + activeTabPane.classList.add('show'); + }, 150); // Adjust delay as needed }); }); - \ No newline at end of file +}); + +document.addEventListener("DOMContentLoaded", function() { + fetch("https://ndif.dev/ping") + .then(response => response.json()) + .then(data => { + const statusElement = document.querySelector(".ndif"); + const status = (data === "pong") ? "Available" : "Unavailable"; + if (statusElement) { + statusElement.setAttribute('data-bs-original-title', `Status: ${status}`); + } + }) + .catch(error => console.error('Error fetching NDIF status:', error)); + + const data = document.querySelector("p.copyright"); + data.innerHTML = "(c) 2024 <a href='https://ndif.us/'>NDIF</a>" + +}); diff --git a/docs/source/_static/pygments/example.html b/docs/source/_static/pygments/example.html deleted file mode 100644 index d1546fd9..00000000 --- a/docs/source/_static/pygments/example.html +++ /dev/null @@ -1,17 +0,0 @@ -<div class="highlight"><pre>import json -from nnsight import LanguageModel -import torch as t -from tqdm import tqdm - -aligned = [ - "truth_teller", - "genie", - "saint", -] - -misaligned = [ - "reward_maximizer", - "fitness_maximizer", - "money_maximizer", -] -</pre></div> diff --git a/docs/source/_static/pygments/example.py b/docs/source/_static/pygments/example.py deleted file mode 100644 index d6f3c5b5..00000000 --- a/docs/source/_static/pygments/example.py +++ /dev/null @@ -1,16 +0,0 @@ -import json -from nnsight import LanguageModel -import torch as t -from tqdm import tqdm - -aligned = [ - "truth_teller", - "genie", - "saint", -] - -misaligned = [ - "reward_maximizer", - "fitness_maximizer", - "money_maximizer", -] diff --git a/docs/source/_templates/layout.html b/docs/source/_templates/layout.html new file mode 100644 index 00000000..de1d1c49 --- /dev/null +++ b/docs/source/_templates/layout.html @@ -0,0 +1,7 @@ +{% extends "!layout.html" %} + +{% block extrahead %} +{{ super() }} +<link href="{{ pathto('_static/css/custom.css', 1) }}?v={{ version_identifier }}" rel="stylesheet" type="text/css" /> +<link href="{{ pathto('_static/css/front.css', 1) }}?v={{ version_identifier }}" rel="stylesheet" type="text/css" /> +{% endblock %} diff --git a/docs/source/_templates/ndif_status.html b/docs/source/_templates/ndif_status.html index 348c8d61..4ac071d8 100644 --- a/docs/source/_templates/ndif_status.html +++ b/docs/source/_templates/ndif_status.html @@ -1,5 +1,4 @@ <script> - fetch("{{ndif_url}}") .then((response) => { if (response.status == 200) { diff --git a/docs/source/about.rst b/docs/source/about.rst index c58f101f..08fef7e9 100644 --- a/docs/source/about.rst +++ b/docs/source/about.rst @@ -1,3 +1,18 @@ +.. raw:: html + + <style> + .accordion-header { + margin: 0 !important; + } + </style> + + <script> + document.addEventListener('DOMContentLoaded', (event) => { + document.querySelectorAll('figure.align-default').forEach(el => { + el.style.marginBottom = "0px"; + }); + }); + </script> About NNsight ============= @@ -16,10 +31,11 @@ An API for transparent science on black-box AI. it lets you run experiments on huge open models easily, with full transparent access. The nnsight library is also terrific for studying smaller local models. -.. figure:: _static/images/remote_execution.png +.. figure:: _static/images/interleaved.png .. card:: + An overview of the nnsight/NDIF pipeline. Researchers write simple Python code to run along with the neural network locally or remotely. Unlike commercial inference, the experiment code can read or write any of the internal states of the neural networks being studied. This code creates a computation graph that can be sent to the remote service and interleaved with the execution of the neural network. How do I use NNsight? @@ -61,7 +77,7 @@ When using nnsight, it is helpful to understand that the operations are not exec An example of one such intervention graph can be seen below: -.. figure:: _static/images/intrgraph.png +.. figure:: _static/images/execution.png .. card:: diff --git a/docs/source/charts/chart_schema.json b/docs/source/charts/chart_schema.json deleted file mode 100644 index 6733fb47..00000000 --- a/docs/source/charts/chart_schema.json +++ /dev/null @@ -1,1124 +0,0 @@ -{ - "data": [ - { - "coloraxis": "coloraxis", - "name": "0", - "z": [ - [ - 0.0, - 0.0, - 0.0, - 0.0, - 0.0, - 0.0, - 0.0, - 0.0, - 0.0, - 0.9999988079071045, - 0.0, - 0.0, - 0.0, - 0.0, - 0.0, - 0.0 - ], - [ - 0.0, - 0.0, - 0.0, - 0.0, - 0.0, - 0.0, - 0.0, - 0.0, - 0.0, - 0.9974687695503235, - 0.0017737559974193573, - 0.0, - 0.0, - 0.0, - -0.001089064753614366, - 0.0 - ], - [ - 0.0, - 0.0, - 0.0, - 0.0, - 0.0, - 0.0, - 0.0, - 0.0, - 0.0, - 0.9971288442611694, - 0.0026111311744898558, - 0.0, - 0.0, - 0.0, - 0.0, - 0.0 - ], - [ - 0.0, - 0.0, - 0.0, - 0.0, - 0.0, - 0.0, - 0.0, - 0.0, - 0.0, - 0.9950985908508301, - 0.004562620539218187, - 0.0, - 0.0, - 0.0, - 0.0, - -0.0021960220765322447 - ], - [ - 0.0, - 0.0, - 0.0, - 0.0, - 0.0, - 0.0, - 0.0, - 0.0, - 0.0, - 0.9868704080581665, - 0.020100580528378487, - 0.0, - 0.0011236571008339524, - 0.0, - 0.0, - -0.0019157042261213064 - ], - [ - 0.0, - 0.0, - 0.0, - 0.0, - 0.0, - 0.0, - 0.0, - 0.0, - 0.0, - 0.9520763754844666, - 0.07812518626451492, - 0.002582503017038107, - 0.0021328015718609095, - 0.0, - 0.0, - -0.001445724512450397 - ], - [ - 0.0, - 0.0, - 0.0, - 0.0, - 0.0, - 0.0, - 0.0, - 0.0, - 0.0, - 0.9528278708457947, - 0.0788850262761116, - 0.0019860819447785616, - 0.0, - 0.0, - 0.0, - -0.004969379864633083 - ], - [ - 0.0, - 0.0, - 0.0, - 0.0, - 0.0, - 0.0, - 0.0, - 0.0, - 0.0, - 0.9087952971458435, - 0.10825876146554947, - 0.003998406231403351, - 0.0, - 0.0, - 0.0, - 0.009814703837037086 - ], - [ - 0.0, - 0.0, - 0.0, - 0.0, - 0.0, - 0.0, - 0.0, - 0.0, - 0.0, - 0.6858435869216919, - 0.035843707621097565, - 0.0, - 0.0, - 0.0, - 0.0013097404735162854, - 0.4153654873371124 - ], - [ - 0.0, - 0.0, - 0.0, - 0.0, - 0.0, - 0.0, - 0.0, - 0.0, - 0.0, - 0.10977605730295181, - 0.024682287126779556, - 0.0, - 0.0, - 0.0, - 0.001402782159857452, - 0.8544625639915466 - ], - [ - 0.0, - 0.0, - 0.0, - 0.0, - 0.0, - 0.0, - 0.0, - 0.0, - 0.0, - 0.01987036131322384, - 0.015400783158838749, - 0.0, - 0.0, - 0.0, - 0.0, - 0.9079328775405884 - ], - [ - 0.0, - 0.0, - 0.0, - 0.0, - 0.0, - 0.0, - 0.0, - 0.0, - 0.0, - 0.019981296733021736, - 0.0062970127910375595, - 0.0, - 0.0, - 0.0, - 0.0, - 0.9069046378135681 - ] - ], - "type": "heatmap", - "xaxis": "x", - "yaxis": "y", - "hovertemplate": "Position: %{x}\u003cbr\u003eLayer: %{y}\u003cbr\u003ecolor: %{z}\u003cextra\u003e\u003c\u002fextra\u003e" - } - ], - "layout": { - "template": { - "data": { - "histogram2dcontour": [ - { - "type": "histogram2dcontour", - "colorbar": { - "outlinewidth": 0, - "ticks": "" - }, - "colorscale": [ - [ - 0.0, - "#0d0887" - ], - [ - 0.1111111111111111, - "#46039f" - ], - [ - 0.2222222222222222, - "#7201a8" - ], - [ - 0.3333333333333333, - "#9c179e" - ], - [ - 0.4444444444444444, - "#bd3786" - ], - [ - 0.5555555555555556, - "#d8576b" - ], - [ - 0.6666666666666666, - "#ed7953" - ], - [ - 0.7777777777777778, - "#fb9f3a" - ], - [ - 0.8888888888888888, - "#fdca26" - ], - [ - 1.0, - "#f0f921" - ] - ] - } - ], - "choropleth": [ - { - "type": "choropleth", - "colorbar": { - "outlinewidth": 0, - "ticks": "" - } - } - ], - "histogram2d": [ - { - "type": "histogram2d", - "colorbar": { - "outlinewidth": 0, - "ticks": "" - }, - "colorscale": [ - [ - 0.0, - "#0d0887" - ], - [ - 0.1111111111111111, - "#46039f" - ], - [ - 0.2222222222222222, - "#7201a8" - ], - [ - 0.3333333333333333, - "#9c179e" - ], - [ - 0.4444444444444444, - "#bd3786" - ], - [ - 0.5555555555555556, - "#d8576b" - ], - [ - 0.6666666666666666, - "#ed7953" - ], - [ - 0.7777777777777778, - "#fb9f3a" - ], - [ - 0.8888888888888888, - "#fdca26" - ], - [ - 1.0, - "#f0f921" - ] - ] - } - ], - "heatmap": [ - { - "type": "heatmap", - "colorbar": { - "outlinewidth": 0, - "ticks": "" - }, - "colorscale": [ - [ - 0.0, - "#0d0887" - ], - [ - 0.1111111111111111, - "#46039f" - ], - [ - 0.2222222222222222, - "#7201a8" - ], - [ - 0.3333333333333333, - "#9c179e" - ], - [ - 0.4444444444444444, - "#bd3786" - ], - [ - 0.5555555555555556, - "#d8576b" - ], - [ - 0.6666666666666666, - "#ed7953" - ], - [ - 0.7777777777777778, - "#fb9f3a" - ], - [ - 0.8888888888888888, - "#fdca26" - ], - [ - 1.0, - "#f0f921" - ] - ] - } - ], - "heatmapgl": [ - { - "type": "heatmapgl", - "colorbar": { - "outlinewidth": 0, - "ticks": "" - }, - "colorscale": [ - [ - 0.0, - "#0d0887" - ], - [ - 0.1111111111111111, - "#46039f" - ], - [ - 0.2222222222222222, - "#7201a8" - ], - [ - 0.3333333333333333, - "#9c179e" - ], - [ - 0.4444444444444444, - "#bd3786" - ], - [ - 0.5555555555555556, - "#d8576b" - ], - [ - 0.6666666666666666, - "#ed7953" - ], - [ - 0.7777777777777778, - "#fb9f3a" - ], - [ - 0.8888888888888888, - "#fdca26" - ], - [ - 1.0, - "#f0f921" - ] - ] - } - ], - "contourcarpet": [ - { - "type": "contourcarpet", - "colorbar": { - "outlinewidth": 0, - "ticks": "" - } - } - ], - "contour": [ - { - "type": "contour", - "colorbar": { - "outlinewidth": 0, - "ticks": "" - }, - "colorscale": [ - [ - 0.0, - "#0d0887" - ], - [ - 0.1111111111111111, - "#46039f" - ], - [ - 0.2222222222222222, - "#7201a8" - ], - [ - 0.3333333333333333, - "#9c179e" - ], - [ - 0.4444444444444444, - "#bd3786" - ], - [ - 0.5555555555555556, - "#d8576b" - ], - [ - 0.6666666666666666, - "#ed7953" - ], - [ - 0.7777777777777778, - "#fb9f3a" - ], - [ - 0.8888888888888888, - "#fdca26" - ], - [ - 1.0, - "#f0f921" - ] - ] - } - ], - "surface": [ - { - "type": "surface", - "colorbar": { - "outlinewidth": 0, - "ticks": "" - }, - "colorscale": [ - [ - 0.0, - "#0d0887" - ], - [ - 0.1111111111111111, - "#46039f" - ], - [ - 0.2222222222222222, - "#7201a8" - ], - [ - 0.3333333333333333, - "#9c179e" - ], - [ - 0.4444444444444444, - "#bd3786" - ], - [ - 0.5555555555555556, - "#d8576b" - ], - [ - 0.6666666666666666, - "#ed7953" - ], - [ - 0.7777777777777778, - "#fb9f3a" - ], - [ - 0.8888888888888888, - "#fdca26" - ], - [ - 1.0, - "#f0f921" - ] - ] - } - ], - "mesh3d": [ - { - "type": "mesh3d", - "colorbar": { - "outlinewidth": 0, - "ticks": "" - } - } - ], - "scatter": [ - { - "fillpattern": { - "fillmode": "overlay", - "size": 10, - "solidity": 0.2 - }, - "type": "scatter" - } - ], - "parcoords": [ - { - "type": "parcoords", - "line": { - "colorbar": { - "outlinewidth": 0, - "ticks": "" - } - } - } - ], - "scatterpolargl": [ - { - "type": "scatterpolargl", - "marker": { - "colorbar": { - "outlinewidth": 0, - "ticks": "" - } - } - } - ], - "bar": [ - { - "error_x": { - "color": "#2a3f5f" - }, - "error_y": { - "color": "#2a3f5f" - }, - "marker": { - "line": { - "color": "#E5ECF6", - "width": 0.5 - }, - "pattern": { - "fillmode": "overlay", - "size": 10, - "solidity": 0.2 - } - }, - "type": "bar" - } - ], - "scattergeo": [ - { - "type": "scattergeo", - "marker": { - "colorbar": { - "outlinewidth": 0, - "ticks": "" - } - } - } - ], - "scatterpolar": [ - { - "type": "scatterpolar", - "marker": { - "colorbar": { - "outlinewidth": 0, - "ticks": "" - } - } - } - ], - "histogram": [ - { - "marker": { - "pattern": { - "fillmode": "overlay", - "size": 10, - "solidity": 0.2 - } - }, - "type": "histogram" - } - ], - "scattergl": [ - { - "type": "scattergl", - "marker": { - "colorbar": { - "outlinewidth": 0, - "ticks": "" - } - } - } - ], - "scatter3d": [ - { - "type": "scatter3d", - "line": { - "colorbar": { - "outlinewidth": 0, - "ticks": "" - } - }, - "marker": { - "colorbar": { - "outlinewidth": 0, - "ticks": "" - } - } - } - ], - "scattermapbox": [ - { - "type": "scattermapbox", - "marker": { - "colorbar": { - "outlinewidth": 0, - "ticks": "" - } - } - } - ], - "scatterternary": [ - { - "type": "scatterternary", - "marker": { - "colorbar": { - "outlinewidth": 0, - "ticks": "" - } - } - } - ], - "scattercarpet": [ - { - "type": "scattercarpet", - "marker": { - "colorbar": { - "outlinewidth": 0, - "ticks": "" - } - } - } - ], - "carpet": [ - { - "aaxis": { - "endlinecolor": "#2a3f5f", - "gridcolor": "white", - "linecolor": "white", - "minorgridcolor": "white", - "startlinecolor": "#2a3f5f" - }, - "baxis": { - "endlinecolor": "#2a3f5f", - "gridcolor": "white", - "linecolor": "white", - "minorgridcolor": "white", - "startlinecolor": "#2a3f5f" - }, - "type": "carpet" - } - ], - "table": [ - { - "cells": { - "fill": { - "color": "#EBF0F8" - }, - "line": { - "color": "white" - } - }, - "header": { - "fill": { - "color": "#C8D4E3" - }, - "line": { - "color": "white" - } - }, - "type": "table" - } - ], - "barpolar": [ - { - "marker": { - "line": { - "color": "#E5ECF6", - "width": 0.5 - }, - "pattern": { - "fillmode": "overlay", - "size": 10, - "solidity": 0.2 - } - }, - "type": "barpolar" - } - ], - "pie": [ - { - "automargin": true, - "type": "pie" - } - ] - }, - "layout": { - "autotypenumbers": "strict", - "colorway": [ - "#636efa", - "#EF553B", - "#00cc96", - "#ab63fa", - "#FFA15A", - "#19d3f3", - "#FF6692", - "#B6E880", - "#FF97FF", - "#FECB52" - ], - "font": { - "color": "#2a3f5f" - }, - "hovermode": "closest", - "hoverlabel": { - "align": "left" - }, - "paper_bgcolor": "white", - "plot_bgcolor": "#E5ECF6", - "polar": { - "bgcolor": "#E5ECF6", - "angularaxis": { - "gridcolor": "white", - "linecolor": "white", - "ticks": "" - }, - "radialaxis": { - "gridcolor": "white", - "linecolor": "white", - "ticks": "" - } - }, - "ternary": { - "bgcolor": "#E5ECF6", - "aaxis": { - "gridcolor": "white", - "linecolor": "white", - "ticks": "" - }, - "baxis": { - "gridcolor": "white", - "linecolor": "white", - "ticks": "" - }, - "caxis": { - "gridcolor": "white", - "linecolor": "white", - "ticks": "" - } - }, - "coloraxis": { - "colorbar": { - "outlinewidth": 0, - "ticks": "" - } - }, - "colorscale": { - "sequential": [ - [ - 0.0, - "#0d0887" - ], - [ - 0.1111111111111111, - "#46039f" - ], - [ - 0.2222222222222222, - "#7201a8" - ], - [ - 0.3333333333333333, - "#9c179e" - ], - [ - 0.4444444444444444, - "#bd3786" - ], - [ - 0.5555555555555556, - "#d8576b" - ], - [ - 0.6666666666666666, - "#ed7953" - ], - [ - 0.7777777777777778, - "#fb9f3a" - ], - [ - 0.8888888888888888, - "#fdca26" - ], - [ - 1.0, - "#f0f921" - ] - ], - "sequentialminus": [ - [ - 0.0, - "#0d0887" - ], - [ - 0.1111111111111111, - "#46039f" - ], - [ - 0.2222222222222222, - "#7201a8" - ], - [ - 0.3333333333333333, - "#9c179e" - ], - [ - 0.4444444444444444, - "#bd3786" - ], - [ - 0.5555555555555556, - "#d8576b" - ], - [ - 0.6666666666666666, - "#ed7953" - ], - [ - 0.7777777777777778, - "#fb9f3a" - ], - [ - 0.8888888888888888, - "#fdca26" - ], - [ - 1.0, - "#f0f921" - ] - ], - "diverging": [ - [ - 0, - "#8e0152" - ], - [ - 0.1, - "#c51b7d" - ], - [ - 0.2, - "#de77ae" - ], - [ - 0.3, - "#f1b6da" - ], - [ - 0.4, - "#fde0ef" - ], - [ - 0.5, - "#f7f7f7" - ], - [ - 0.6, - "#e6f5d0" - ], - [ - 0.7, - "#b8e186" - ], - [ - 0.8, - "#7fbc41" - ], - [ - 0.9, - "#4d9221" - ], - [ - 1, - "#276419" - ] - ] - }, - "xaxis": { - "gridcolor": "white", - "linecolor": "white", - "ticks": "", - "title": { - "standoff": 15 - }, - "zerolinecolor": "white", - "automargin": true, - "zerolinewidth": 2 - }, - "yaxis": { - "gridcolor": "white", - "linecolor": "white", - "ticks": "", - "title": { - "standoff": 15 - }, - "zerolinecolor": "white", - "automargin": true, - "zerolinewidth": 2 - }, - "scene": { - "xaxis": { - "backgroundcolor": "#E5ECF6", - "gridcolor": "white", - "linecolor": "white", - "showbackground": true, - "ticks": "", - "zerolinecolor": "white", - "gridwidth": 2 - }, - "yaxis": { - "backgroundcolor": "#E5ECF6", - "gridcolor": "white", - "linecolor": "white", - "showbackground": true, - "ticks": "", - "zerolinecolor": "white", - "gridwidth": 2 - }, - "zaxis": { - "backgroundcolor": "#E5ECF6", - "gridcolor": "white", - "linecolor": "white", - "showbackground": true, - "ticks": "", - "zerolinecolor": "white", - "gridwidth": 2 - } - }, - "shapedefaults": { - "line": { - "color": "#2a3f5f" - } - }, - "annotationdefaults": { - "arrowcolor": "#2a3f5f", - "arrowhead": 0, - "arrowwidth": 1 - }, - "geo": { - "bgcolor": "white", - "landcolor": "#E5ECF6", - "subunitcolor": "white", - "showland": true, - "showlakes": true, - "lakecolor": "white" - }, - "title": { - "x": 0.05 - }, - "mapbox": { - "style": "light" - } - } - }, - "xaxis": { - "anchor": "y", - "domain": [ - 0.0, - 1.0 - ], - "scaleanchor": "y", - "constrain": "domain", - "title": { - "text": "Position" - } - }, - "yaxis": { - "anchor": "x", - "domain": [ - 0.0, - 1.0 - ], - "autorange": "reversed", - "constrain": "domain", - "title": { - "text": "Layer" - } - }, - "coloraxis": { - "colorscale": [ - [ - 0.0, - "rgb(103,0,31)" - ], - [ - 0.1, - "rgb(178,24,43)" - ], - [ - 0.2, - "rgb(214,96,77)" - ], - [ - 0.3, - "rgb(244,165,130)" - ], - [ - 0.4, - "rgb(253,219,199)" - ], - [ - 0.5, - "rgb(247,247,247)" - ], - [ - 0.6, - "rgb(209,229,240)" - ], - [ - 0.7, - "rgb(146,197,222)" - ], - [ - 0.8, - "rgb(67,147,195)" - ], - [ - 0.9, - "rgb(33,102,172)" - ], - [ - 1.0, - "rgb(5,48,97)" - ] - ], - "cmid": 0.0 - }, - "margin": { - "t": 60 - } - } -} \ No newline at end of file diff --git a/docs/source/charts/induction.json b/docs/source/charts/induction.json deleted file mode 100644 index 4946b16a..00000000 --- a/docs/source/charts/induction.json +++ /dev/null @@ -1,1 +0,0 @@ -{"data":[{"coloraxis":"coloraxis","name":"0","z":[[0.010070409625768661,7.986390846781433e-05,0.01258260291069746,2.1361253743634734e-07,0.00020078140369150788,5.013069312553853e-05,0.008831579238176346,0.000892787182237953,0.009880675002932549,0.010911175981163979,0.007516765501350164,0.015246091410517693],[0.0004512245941441506,0.0006454214453697205,0.0025818662252277136,0.01306276023387909,0.005299094598740339,0.013112545944750309,0.015744850039482117,0.013699539005756378,0.012578608468174934,0.015512274578213692,0.006809852551668882,0.001121349399909377],[0.004063299857079983,0.01543406117707491,0.0010489624692127109,0.0013296647230163217,0.011215250939130783,0.0018478287383913994,0.005366469733417034,0.008304793387651443,0.0030799838714301586,0.0010017683962360024,0.0008394559845328331,0.009666861966252327],[0.013943175785243511,0.005458588246256113,0.0022658391389995813,0.009771357290446758,0.019387023523449898,0.010070898570120335,0.001896885340102017,0.0006945269415155053,0.004318578634411097,0.009726484306156635,0.007587844040244818,0.005201588850468397],[0.012102176435291767,0.01121105533093214,0.012845116667449474,0.007275603711605072,0.01644388772547245,0.010665860958397388,0.007502487860620022,0.0023389223497360945,0.01597411371767521,0.011477481573820114,0.017345573753118515,3.745994181514334e-09],[0.16879354417324066,0.3332799971103668,0.01330574695020914,0.006573637947440147,0.009730679914355278,0.3288743495941162,0.008749036118388176,0.015621461905539036,0.020145757123827934,0.023961644619703293,0.01803380250930786,0.015584372915327549],[0.00760197639465332,0.01550981029868126,0.017369290813803673,0.014239192940294743,0.01672932133078575,0.011254709213972092,0.024911386892199516,0.010701530613005161,0.010776473209261894,0.32112863659858704,0.024556512013077736,0.010988256894052029],[0.009945792146027088,0.11099910736083984,0.2978491187095642,0.01551186852157116,0.016720473766326904,0.013705338351428509,0.02495485544204712,0.09114369004964828,0.01463283784687519,0.016146264970302582,0.3241707682609558,0.03573969751596451],[0.014393663965165615,0.1397504210472107,0.012329380959272385,0.023921499028801918,0.01609531044960022,0.011505285277962685,0.06444355845451355,0.012670181691646576,0.017619885504245758,0.02077418565750122,0.030146704986691475,0.014868587255477905],[0.09159916639328003,0.08219298720359802,0.046227119863033295,0.0109515730291605,0.04310443252325058,0.019227860495448112,0.16419260203838348,0.020021013915538788,0.023179948329925537,0.17912650108337402,0.01535245031118393,0.019395975396037102],[0.12493489682674408,0.18007728457450867,0.02339624986052513,0.062402497977018356,0.030824515968561172,0.013725594617426395,0.11222781985998154,0.1664256453514099,0.02788865752518177,0.013621106743812561,0.06758259236812592,0.10711898654699326],[0.014790158718824387,0.030200988054275513,0.02301931381225586,0.00981247890740633,0.019670825451612473,0.04539807513356209,0.029655294492840767,0.03308337181806564,0.008349732495844364,0.1181706115603447,0.1325961947441101,0.013514983467757702]],"type":"heatmap","xaxis":"x","yaxis":"y","hovertemplate":"Position: %{x}\u003cbr\u003eLayer: %{y}\u003cbr\u003ecolor: %{z}\u003cextra\u003e\u003c\u002fextra\u003e"}],"layout":{"template":{"data":{"histogram2dcontour":[{"type":"histogram2dcontour","colorbar":{"outlinewidth":0,"ticks":""},"colorscale":[[0.0,"#0d0887"],[0.1111111111111111,"#46039f"],[0.2222222222222222,"#7201a8"],[0.3333333333333333,"#9c179e"],[0.4444444444444444,"#bd3786"],[0.5555555555555556,"#d8576b"],[0.6666666666666666,"#ed7953"],[0.7777777777777778,"#fb9f3a"],[0.8888888888888888,"#fdca26"],[1.0,"#f0f921"]]}],"choropleth":[{"type":"choropleth","colorbar":{"outlinewidth":0,"ticks":""}}],"histogram2d":[{"type":"histogram2d","colorbar":{"outlinewidth":0,"ticks":""},"colorscale":[[0.0,"#0d0887"],[0.1111111111111111,"#46039f"],[0.2222222222222222,"#7201a8"],[0.3333333333333333,"#9c179e"],[0.4444444444444444,"#bd3786"],[0.5555555555555556,"#d8576b"],[0.6666666666666666,"#ed7953"],[0.7777777777777778,"#fb9f3a"],[0.8888888888888888,"#fdca26"],[1.0,"#f0f921"]]}],"heatmap":[{"type":"heatmap","colorbar":{"outlinewidth":0,"ticks":""},"colorscale":[[0.0,"#0d0887"],[0.1111111111111111,"#46039f"],[0.2222222222222222,"#7201a8"],[0.3333333333333333,"#9c179e"],[0.4444444444444444,"#bd3786"],[0.5555555555555556,"#d8576b"],[0.6666666666666666,"#ed7953"],[0.7777777777777778,"#fb9f3a"],[0.8888888888888888,"#fdca26"],[1.0,"#f0f921"]]}],"heatmapgl":[{"type":"heatmapgl","colorbar":{"outlinewidth":0,"ticks":""},"colorscale":[[0.0,"#0d0887"],[0.1111111111111111,"#46039f"],[0.2222222222222222,"#7201a8"],[0.3333333333333333,"#9c179e"],[0.4444444444444444,"#bd3786"],[0.5555555555555556,"#d8576b"],[0.6666666666666666,"#ed7953"],[0.7777777777777778,"#fb9f3a"],[0.8888888888888888,"#fdca26"],[1.0,"#f0f921"]]}],"contourcarpet":[{"type":"contourcarpet","colorbar":{"outlinewidth":0,"ticks":""}}],"contour":[{"type":"contour","colorbar":{"outlinewidth":0,"ticks":""},"colorscale":[[0.0,"#0d0887"],[0.1111111111111111,"#46039f"],[0.2222222222222222,"#7201a8"],[0.3333333333333333,"#9c179e"],[0.4444444444444444,"#bd3786"],[0.5555555555555556,"#d8576b"],[0.6666666666666666,"#ed7953"],[0.7777777777777778,"#fb9f3a"],[0.8888888888888888,"#fdca26"],[1.0,"#f0f921"]]}],"surface":[{"type":"surface","colorbar":{"outlinewidth":0,"ticks":""},"colorscale":[[0.0,"#0d0887"],[0.1111111111111111,"#46039f"],[0.2222222222222222,"#7201a8"],[0.3333333333333333,"#9c179e"],[0.4444444444444444,"#bd3786"],[0.5555555555555556,"#d8576b"],[0.6666666666666666,"#ed7953"],[0.7777777777777778,"#fb9f3a"],[0.8888888888888888,"#fdca26"],[1.0,"#f0f921"]]}],"mesh3d":[{"type":"mesh3d","colorbar":{"outlinewidth":0,"ticks":""}}],"scatter":[{"fillpattern":{"fillmode":"overlay","size":10,"solidity":0.2},"type":"scatter"}],"parcoords":[{"type":"parcoords","line":{"colorbar":{"outlinewidth":0,"ticks":""}}}],"scatterpolargl":[{"type":"scatterpolargl","marker":{"colorbar":{"outlinewidth":0,"ticks":""}}}],"bar":[{"error_x":{"color":"#2a3f5f"},"error_y":{"color":"#2a3f5f"},"marker":{"line":{"color":"#E5ECF6","width":0.5},"pattern":{"fillmode":"overlay","size":10,"solidity":0.2}},"type":"bar"}],"scattergeo":[{"type":"scattergeo","marker":{"colorbar":{"outlinewidth":0,"ticks":""}}}],"scatterpolar":[{"type":"scatterpolar","marker":{"colorbar":{"outlinewidth":0,"ticks":""}}}],"histogram":[{"marker":{"pattern":{"fillmode":"overlay","size":10,"solidity":0.2}},"type":"histogram"}],"scattergl":[{"type":"scattergl","marker":{"colorbar":{"outlinewidth":0,"ticks":""}}}],"scatter3d":[{"type":"scatter3d","line":{"colorbar":{"outlinewidth":0,"ticks":""}},"marker":{"colorbar":{"outlinewidth":0,"ticks":""}}}],"scattermapbox":[{"type":"scattermapbox","marker":{"colorbar":{"outlinewidth":0,"ticks":""}}}],"scatterternary":[{"type":"scatterternary","marker":{"colorbar":{"outlinewidth":0,"ticks":""}}}],"scattercarpet":[{"type":"scattercarpet","marker":{"colorbar":{"outlinewidth":0,"ticks":""}}}],"carpet":[{"aaxis":{"endlinecolor":"#2a3f5f","gridcolor":"white","linecolor":"white","minorgridcolor":"white","startlinecolor":"#2a3f5f"},"baxis":{"endlinecolor":"#2a3f5f","gridcolor":"white","linecolor":"white","minorgridcolor":"white","startlinecolor":"#2a3f5f"},"type":"carpet"}],"table":[{"cells":{"fill":{"color":"#EBF0F8"},"line":{"color":"white"}},"header":{"fill":{"color":"#C8D4E3"},"line":{"color":"white"}},"type":"table"}],"barpolar":[{"marker":{"line":{"color":"#E5ECF6","width":0.5},"pattern":{"fillmode":"overlay","size":10,"solidity":0.2}},"type":"barpolar"}],"pie":[{"automargin":true,"type":"pie"}]},"layout":{"autotypenumbers":"strict","colorway":["#636efa","#EF553B","#00cc96","#ab63fa","#FFA15A","#19d3f3","#FF6692","#B6E880","#FF97FF","#FECB52"],"font":{"color":"#2a3f5f"},"hovermode":"closest","hoverlabel":{"align":"left"},"paper_bgcolor":"white","plot_bgcolor":"#E5ECF6","polar":{"bgcolor":"#E5ECF6","angularaxis":{"gridcolor":"white","linecolor":"white","ticks":""},"radialaxis":{"gridcolor":"white","linecolor":"white","ticks":""}},"ternary":{"bgcolor":"#E5ECF6","aaxis":{"gridcolor":"white","linecolor":"white","ticks":""},"baxis":{"gridcolor":"white","linecolor":"white","ticks":""},"caxis":{"gridcolor":"white","linecolor":"white","ticks":""}},"coloraxis":{"colorbar":{"outlinewidth":0,"ticks":""}},"colorscale":{"sequential":[[0.0,"#0d0887"],[0.1111111111111111,"#46039f"],[0.2222222222222222,"#7201a8"],[0.3333333333333333,"#9c179e"],[0.4444444444444444,"#bd3786"],[0.5555555555555556,"#d8576b"],[0.6666666666666666,"#ed7953"],[0.7777777777777778,"#fb9f3a"],[0.8888888888888888,"#fdca26"],[1.0,"#f0f921"]],"sequentialminus":[[0.0,"#0d0887"],[0.1111111111111111,"#46039f"],[0.2222222222222222,"#7201a8"],[0.3333333333333333,"#9c179e"],[0.4444444444444444,"#bd3786"],[0.5555555555555556,"#d8576b"],[0.6666666666666666,"#ed7953"],[0.7777777777777778,"#fb9f3a"],[0.8888888888888888,"#fdca26"],[1.0,"#f0f921"]],"diverging":[[0,"#8e0152"],[0.1,"#c51b7d"],[0.2,"#de77ae"],[0.3,"#f1b6da"],[0.4,"#fde0ef"],[0.5,"#f7f7f7"],[0.6,"#e6f5d0"],[0.7,"#b8e186"],[0.8,"#7fbc41"],[0.9,"#4d9221"],[1,"#276419"]]},"xaxis":{"gridcolor":"white","linecolor":"white","ticks":"","title":{"standoff":15},"zerolinecolor":"white","automargin":true,"zerolinewidth":2},"yaxis":{"gridcolor":"white","linecolor":"white","ticks":"","title":{"standoff":15},"zerolinecolor":"white","automargin":true,"zerolinewidth":2},"scene":{"xaxis":{"backgroundcolor":"#E5ECF6","gridcolor":"white","linecolor":"white","showbackground":true,"ticks":"","zerolinecolor":"white","gridwidth":2},"yaxis":{"backgroundcolor":"#E5ECF6","gridcolor":"white","linecolor":"white","showbackground":true,"ticks":"","zerolinecolor":"white","gridwidth":2},"zaxis":{"backgroundcolor":"#E5ECF6","gridcolor":"white","linecolor":"white","showbackground":true,"ticks":"","zerolinecolor":"white","gridwidth":2}},"shapedefaults":{"line":{"color":"#2a3f5f"}},"annotationdefaults":{"arrowcolor":"#2a3f5f","arrowhead":0,"arrowwidth":1},"geo":{"bgcolor":"white","landcolor":"#E5ECF6","subunitcolor":"white","showland":true,"showlakes":true,"lakecolor":"white"},"title":{"x":0.05},"mapbox":{"style":"light"}}},"xaxis":{"anchor":"y","domain":[0.0,1.0],"scaleanchor":"y","constrain":"domain","title":{"text":"Position"}},"yaxis":{"anchor":"x","domain":[0.0,1.0],"autorange":"reversed","constrain":"domain","title":{"text":"Layer"}},"coloraxis":{"colorscale":[[0.0,"rgb(103,0,31)"],[0.1,"rgb(178,24,43)"],[0.2,"rgb(214,96,77)"],[0.3,"rgb(244,165,130)"],[0.4,"rgb(253,219,199)"],[0.5,"rgb(247,247,247)"],[0.6,"rgb(209,229,240)"],[0.7,"rgb(146,197,222)"],[0.8,"rgb(67,147,195)"],[0.9,"rgb(33,102,172)"],[1.0,"rgb(5,48,97)"]],"cmid":0.0},"margin":{"t":60}}} \ No newline at end of file diff --git a/docs/source/charts/line.json b/docs/source/charts/line.json deleted file mode 100644 index b3dae70c..00000000 --- a/docs/source/charts/line.json +++ /dev/null @@ -1,1083 +0,0 @@ -{ - "data": [ - { - "hovertemplate": "variable=0\u003cbr\u003eindex=%{x}\u003cbr\u003evalue=%{y}\u003cextra\u003e\u003c\u002fextra\u003e", - "legendgroup": "0", - "line": { - "color": "#636efa", - "dash": "solid" - }, - "marker": { - "symbol": "circle" - }, - "mode": "lines", - "name": "0", - "orientation": "v", - "showlegend": true, - "x": [ - 0, - 1, - 2, - 3, - 4, - 5, - 6, - 7, - 8, - 9, - 10, - 11, - 12, - 13, - 14, - 15, - 16, - 17, - 18, - 19, - 20, - 21, - 22, - 23, - 24, - 25, - 26, - 27, - 28, - 29, - 30, - 31, - 32, - 33, - 34, - 35, - 36, - 37, - 38, - 39, - 40, - 41, - 42, - 43, - 44, - 45, - 46, - 47, - 48, - 49, - 50, - 51, - 52, - 53, - 54, - 55, - 56, - 57, - 58, - 59, - 60, - 61, - 62, - 63, - 64, - 65, - 66, - 67, - 68, - 69, - 70, - 71, - 72, - 73, - 74, - 75, - 76, - 77, - 78, - 79, - 80, - 81, - 82, - 83, - 84, - 85, - 86, - 87, - 88, - 89, - 90, - 91, - 92, - 93, - 94, - 95, - 96, - 97, - 98, - 99, - 100, - 101, - 102, - 103, - 104 - ], - "xaxis": "x", - "y": [ - 6.721836090087891, - 15.863195419311523, - 14.229825019836426, - 14.620223999023438, - 12.628325462341309, - 14.076154708862305, - 13.74431037902832, - 14.042016983032227, - 12.240949630737305, - 12.101900100708008, - 12.948463439941406, - 13.237577438354492, - 11.980368614196777, - 12.275359153747559, - 12.520895004272461, - 12.39179801940918, - 11.573649406433105, - 11.729515075683594, - 12.298530578613281, - 11.90172004699707, - 12.044195175170898, - 12.171052932739258, - 12.23204231262207, - 10.854452133178711, - 10.37388801574707, - 12.753911972045898, - 12.342389106750488, - 11.449339866638184, - 12.227272033691406, - 12.660356521606445, - 12.355462074279785, - 11.243412017822266, - 11.748628616333008, - 12.428620338439941, - 11.892528533935547, - 12.418279647827148, - 12.380900382995605, - 12.087124824523926, - 11.42609977722168, - 12.237627983093262, - 12.026906967163086, - 10.750848770141602, - 11.18249225616455, - 10.964658737182617, - 12.21753978729248, - 11.411855697631836, - 12.105101585388184, - 11.151140213012695, - 11.415960311889648, - 10.983814239501953, - 10.562061309814453, - 12.14051628112793, - 10.84417724609375, - 10.993143081665039, - 7.57098913192749, - 5.991965293884277, - 2.293006420135498, - 1.1911180019378662, - 1.4175719022750854, - 0.5377484560012817, - 0.6352464556694031, - 0.3934612572193146, - 0.3168984055519104, - 1.2923715114593506, - 0.25427520275115967, - 0.12562693655490875, - 0.1655888557434082, - 0.05889231711626053, - 0.13178208470344543, - 0.11886248737573624, - 0.10017742961645126, - 0.28241199254989624, - 0.2901555001735687, - 0.07609564065933228, - 0.13399086892604828, - 0.20270712673664093, - 0.22564712166786194, - 0.1278029978275299, - 0.065616674721241, - 0.11973398923873901, - 0.03500190004706383, - 0.5889385938644409, - 0.05313155800104141, - 0.0918748527765274, - 0.1541168987751007, - 0.07680816948413849, - 0.18667154014110565, - 0.1084548830986023, - 0.3974376916885376, - 0.14485009014606476, - 0.13042660057544708, - 0.21075209975242615, - 0.02071751281619072, - 0.016112055629491806, - 0.04902868717908859, - 0.11065000295639038, - 0.11377827823162079, - 0.10953978449106216, - 0.10317499935626984, - 0.07640652358531952, - 0.4654570519924164, - 0.05069968104362488, - 0.09305594116449356, - 0.3460657298564911, - 0.14819812774658203 - ], - "yaxis": "y", - "type": "scatter" - } - ], - "layout": { - "template": { - "data": { - "histogram2dcontour": [ - { - "type": "histogram2dcontour", - "colorbar": { - "outlinewidth": 0, - "ticks": "" - }, - "colorscale": [ - [ - 0.0, - "#0d0887" - ], - [ - 0.1111111111111111, - "#46039f" - ], - [ - 0.2222222222222222, - "#7201a8" - ], - [ - 0.3333333333333333, - "#9c179e" - ], - [ - 0.4444444444444444, - "#bd3786" - ], - [ - 0.5555555555555556, - "#d8576b" - ], - [ - 0.6666666666666666, - "#ed7953" - ], - [ - 0.7777777777777778, - "#fb9f3a" - ], - [ - 0.8888888888888888, - "#fdca26" - ], - [ - 1.0, - "#f0f921" - ] - ] - } - ], - "choropleth": [ - { - "type": "choropleth", - "colorbar": { - "outlinewidth": 0, - "ticks": "" - } - } - ], - "histogram2d": [ - { - "type": "histogram2d", - "colorbar": { - "outlinewidth": 0, - "ticks": "" - }, - "colorscale": [ - [ - 0.0, - "#0d0887" - ], - [ - 0.1111111111111111, - "#46039f" - ], - [ - 0.2222222222222222, - "#7201a8" - ], - [ - 0.3333333333333333, - "#9c179e" - ], - [ - 0.4444444444444444, - "#bd3786" - ], - [ - 0.5555555555555556, - "#d8576b" - ], - [ - 0.6666666666666666, - "#ed7953" - ], - [ - 0.7777777777777778, - "#fb9f3a" - ], - [ - 0.8888888888888888, - "#fdca26" - ], - [ - 1.0, - "#f0f921" - ] - ] - } - ], - "heatmap": [ - { - "type": "heatmap", - "colorbar": { - "outlinewidth": 0, - "ticks": "" - }, - "colorscale": [ - [ - 0.0, - "#0d0887" - ], - [ - 0.1111111111111111, - "#46039f" - ], - [ - 0.2222222222222222, - "#7201a8" - ], - [ - 0.3333333333333333, - "#9c179e" - ], - [ - 0.4444444444444444, - "#bd3786" - ], - [ - 0.5555555555555556, - "#d8576b" - ], - [ - 0.6666666666666666, - "#ed7953" - ], - [ - 0.7777777777777778, - "#fb9f3a" - ], - [ - 0.8888888888888888, - "#fdca26" - ], - [ - 1.0, - "#f0f921" - ] - ] - } - ], - "heatmapgl": [ - { - "type": "heatmapgl", - "colorbar": { - "outlinewidth": 0, - "ticks": "" - }, - "colorscale": [ - [ - 0.0, - "#0d0887" - ], - [ - 0.1111111111111111, - "#46039f" - ], - [ - 0.2222222222222222, - "#7201a8" - ], - [ - 0.3333333333333333, - "#9c179e" - ], - [ - 0.4444444444444444, - "#bd3786" - ], - [ - 0.5555555555555556, - "#d8576b" - ], - [ - 0.6666666666666666, - "#ed7953" - ], - [ - 0.7777777777777778, - "#fb9f3a" - ], - [ - 0.8888888888888888, - "#fdca26" - ], - [ - 1.0, - "#f0f921" - ] - ] - } - ], - "contourcarpet": [ - { - "type": "contourcarpet", - "colorbar": { - "outlinewidth": 0, - "ticks": "" - } - } - ], - "contour": [ - { - "type": "contour", - "colorbar": { - "outlinewidth": 0, - "ticks": "" - }, - "colorscale": [ - [ - 0.0, - "#0d0887" - ], - [ - 0.1111111111111111, - "#46039f" - ], - [ - 0.2222222222222222, - "#7201a8" - ], - [ - 0.3333333333333333, - "#9c179e" - ], - [ - 0.4444444444444444, - "#bd3786" - ], - [ - 0.5555555555555556, - "#d8576b" - ], - [ - 0.6666666666666666, - "#ed7953" - ], - [ - 0.7777777777777778, - "#fb9f3a" - ], - [ - 0.8888888888888888, - "#fdca26" - ], - [ - 1.0, - "#f0f921" - ] - ] - } - ], - "surface": [ - { - "type": "surface", - "colorbar": { - "outlinewidth": 0, - "ticks": "" - }, - "colorscale": [ - [ - 0.0, - "#0d0887" - ], - [ - 0.1111111111111111, - "#46039f" - ], - [ - 0.2222222222222222, - "#7201a8" - ], - [ - 0.3333333333333333, - "#9c179e" - ], - [ - 0.4444444444444444, - "#bd3786" - ], - [ - 0.5555555555555556, - "#d8576b" - ], - [ - 0.6666666666666666, - "#ed7953" - ], - [ - 0.7777777777777778, - "#fb9f3a" - ], - [ - 0.8888888888888888, - "#fdca26" - ], - [ - 1.0, - "#f0f921" - ] - ] - } - ], - "mesh3d": [ - { - "type": "mesh3d", - "colorbar": { - "outlinewidth": 0, - "ticks": "" - } - } - ], - "scatter": [ - { - "fillpattern": { - "fillmode": "overlay", - "size": 10, - "solidity": 0.2 - }, - "type": "scatter" - } - ], - "parcoords": [ - { - "type": "parcoords", - "line": { - "colorbar": { - "outlinewidth": 0, - "ticks": "" - } - } - } - ], - "scatterpolargl": [ - { - "type": "scatterpolargl", - "marker": { - "colorbar": { - "outlinewidth": 0, - "ticks": "" - } - } - } - ], - "bar": [ - { - "error_x": { - "color": "#2a3f5f" - }, - "error_y": { - "color": "#2a3f5f" - }, - "marker": { - "line": { - "color": "#E5ECF6", - "width": 0.5 - }, - "pattern": { - "fillmode": "overlay", - "size": 10, - "solidity": 0.2 - } - }, - "type": "bar" - } - ], - "scattergeo": [ - { - "type": "scattergeo", - "marker": { - "colorbar": { - "outlinewidth": 0, - "ticks": "" - } - } - } - ], - "scatterpolar": [ - { - "type": "scatterpolar", - "marker": { - "colorbar": { - "outlinewidth": 0, - "ticks": "" - } - } - } - ], - "histogram": [ - { - "marker": { - "pattern": { - "fillmode": "overlay", - "size": 10, - "solidity": 0.2 - } - }, - "type": "histogram" - } - ], - "scattergl": [ - { - "type": "scattergl", - "marker": { - "colorbar": { - "outlinewidth": 0, - "ticks": "" - } - } - } - ], - "scatter3d": [ - { - "type": "scatter3d", - "line": { - "colorbar": { - "outlinewidth": 0, - "ticks": "" - } - }, - "marker": { - "colorbar": { - "outlinewidth": 0, - "ticks": "" - } - } - } - ], - "scattermapbox": [ - { - "type": "scattermapbox", - "marker": { - "colorbar": { - "outlinewidth": 0, - "ticks": "" - } - } - } - ], - "scatterternary": [ - { - "type": "scatterternary", - "marker": { - "colorbar": { - "outlinewidth": 0, - "ticks": "" - } - } - } - ], - "scattercarpet": [ - { - "type": "scattercarpet", - "marker": { - "colorbar": { - "outlinewidth": 0, - "ticks": "" - } - } - } - ], - "carpet": [ - { - "aaxis": { - "endlinecolor": "#2a3f5f", - "gridcolor": "white", - "linecolor": "white", - "minorgridcolor": "white", - "startlinecolor": "#2a3f5f" - }, - "baxis": { - "endlinecolor": "#2a3f5f", - "gridcolor": "white", - "linecolor": "white", - "minorgridcolor": "white", - "startlinecolor": "#2a3f5f" - }, - "type": "carpet" - } - ], - "table": [ - { - "cells": { - "fill": { - "color": "#EBF0F8" - }, - "line": { - "color": "white" - } - }, - "header": { - "fill": { - "color": "#C8D4E3" - }, - "line": { - "color": "white" - } - }, - "type": "table" - } - ], - "barpolar": [ - { - "marker": { - "line": { - "color": "#E5ECF6", - "width": 0.5 - }, - "pattern": { - "fillmode": "overlay", - "size": 10, - "solidity": 0.2 - } - }, - "type": "barpolar" - } - ], - "pie": [ - { - "automargin": true, - "type": "pie" - } - ] - }, - "layout": { - "autotypenumbers": "strict", - "colorway": [ - "#636efa", - "#EF553B", - "#00cc96", - "#ab63fa", - "#FFA15A", - "#19d3f3", - "#FF6692", - "#B6E880", - "#FF97FF", - "#FECB52" - ], - "font": { - "color": "#2a3f5f" - }, - "hovermode": "closest", - "hoverlabel": { - "align": "left" - }, - "paper_bgcolor": "white", - "plot_bgcolor": "#E5ECF6", - "polar": { - "bgcolor": "#E5ECF6", - "angularaxis": { - "gridcolor": "white", - "linecolor": "white", - "ticks": "" - }, - "radialaxis": { - "gridcolor": "white", - "linecolor": "white", - "ticks": "" - } - }, - "ternary": { - "bgcolor": "#E5ECF6", - "aaxis": { - "gridcolor": "white", - "linecolor": "white", - "ticks": "" - }, - "baxis": { - "gridcolor": "white", - "linecolor": "white", - "ticks": "" - }, - "caxis": { - "gridcolor": "white", - "linecolor": "white", - "ticks": "" - } - }, - "coloraxis": { - "colorbar": { - "outlinewidth": 0, - "ticks": "" - } - }, - "colorscale": { - "sequential": [ - [ - 0.0, - "#0d0887" - ], - [ - 0.1111111111111111, - "#46039f" - ], - [ - 0.2222222222222222, - "#7201a8" - ], - [ - 0.3333333333333333, - "#9c179e" - ], - [ - 0.4444444444444444, - "#bd3786" - ], - [ - 0.5555555555555556, - "#d8576b" - ], - [ - 0.6666666666666666, - "#ed7953" - ], - [ - 0.7777777777777778, - "#fb9f3a" - ], - [ - 0.8888888888888888, - "#fdca26" - ], - [ - 1.0, - "#f0f921" - ] - ], - "sequentialminus": [ - [ - 0.0, - "#0d0887" - ], - [ - 0.1111111111111111, - "#46039f" - ], - [ - 0.2222222222222222, - "#7201a8" - ], - [ - 0.3333333333333333, - "#9c179e" - ], - [ - 0.4444444444444444, - "#bd3786" - ], - [ - 0.5555555555555556, - "#d8576b" - ], - [ - 0.6666666666666666, - "#ed7953" - ], - [ - 0.7777777777777778, - "#fb9f3a" - ], - [ - 0.8888888888888888, - "#fdca26" - ], - [ - 1.0, - "#f0f921" - ] - ], - "diverging": [ - [ - 0, - "#8e0152" - ], - [ - 0.1, - "#c51b7d" - ], - [ - 0.2, - "#de77ae" - ], - [ - 0.3, - "#f1b6da" - ], - [ - 0.4, - "#fde0ef" - ], - [ - 0.5, - "#f7f7f7" - ], - [ - 0.6, - "#e6f5d0" - ], - [ - 0.7, - "#b8e186" - ], - [ - 0.8, - "#7fbc41" - ], - [ - 0.9, - "#4d9221" - ], - [ - 1, - "#276419" - ] - ] - }, - "xaxis": { - "gridcolor": "white", - "linecolor": "white", - "ticks": "", - "title": { - "standoff": 15 - }, - "zerolinecolor": "white", - "automargin": true, - "zerolinewidth": 2 - }, - "yaxis": { - "gridcolor": "white", - "linecolor": "white", - "ticks": "", - "title": { - "standoff": 15 - }, - "zerolinecolor": "white", - "automargin": true, - "zerolinewidth": 2 - }, - "scene": { - "xaxis": { - "backgroundcolor": "#E5ECF6", - "gridcolor": "white", - "linecolor": "white", - "showbackground": true, - "ticks": "", - "zerolinecolor": "white", - "gridwidth": 2 - }, - "yaxis": { - "backgroundcolor": "#E5ECF6", - "gridcolor": "white", - "linecolor": "white", - "showbackground": true, - "ticks": "", - "zerolinecolor": "white", - "gridwidth": 2 - }, - "zaxis": { - "backgroundcolor": "#E5ECF6", - "gridcolor": "white", - "linecolor": "white", - "showbackground": true, - "ticks": "", - "zerolinecolor": "white", - "gridwidth": 2 - } - }, - "shapedefaults": { - "line": { - "color": "#2a3f5f" - } - }, - "annotationdefaults": { - "arrowcolor": "#2a3f5f", - "arrowhead": 0, - "arrowwidth": 1 - }, - "geo": { - "bgcolor": "white", - "landcolor": "#E5ECF6", - "subunitcolor": "white", - "showland": true, - "showlakes": true, - "lakecolor": "white" - }, - "title": { - "x": 0.05 - }, - "mapbox": { - "style": "light" - } - } - }, - "xaxis": { - "anchor": "y", - "domain": [ - 0.0, - 1.0 - ], - "title": { - "text": "index" - } - }, - "yaxis": { - "anchor": "x", - "domain": [ - 0.0, - 1.0 - ], - "title": { - "text": "value" - } - }, - "legend": { - "title": { - "text": "variable" - }, - "tracegroupgap": 0 - }, - "margin": { - "t": 60 - } - } -} \ No newline at end of file diff --git a/docs/source/conf.py b/docs/source/conf.py index b09d7c01..12b179d4 100644 --- a/docs/source/conf.py +++ b/docs/source/conf.py @@ -1,8 +1,10 @@ +import time + # Configuration file for the Sphinx documentation builder. # Project Information project = "nnsight" -copyright = "2023, NDIF" +copyright = "2024 NDIF" author = "Jaden Fiotto-Kaufman" @@ -26,19 +28,21 @@ html_title = "nnsight" html_logo = "_static/images/nnsight_logo.svg" html_static_path = ["_static"] +html_show_sphinx = False html_favicon = "_static/images/icon.ico" html_show_sourcelink = False html_context = { "default_mode": "dark", - "ndif_url": "https://ndif.dev/ping" + "ndif_url": "https://ndif.dev/ping", + "version_identifier": str(int(time.time())), } html_theme_options = { "show_nav_level": 2, - "navbar_end": ["navbar-icon-links", "ndif_status"], + "navbar_end": ["theme-switcher","navbar-icon-links", "ndif_status"], "navbar_align": "left", "icon_links": [ { @@ -52,25 +56,17 @@ "icon": "fa-brands fa-discord", }, { - "name": "Status", + "name": "Status: Unknown", "url": "/status", "icon": "fa-solid fa-circle-check", "attributes": {"class": "ndif"}, }, ], "show_prev_next": False, - "default_mode": "dark", - "pygment_dark_style": "monokai" + "pygment_dark_style": "monokai", + "pygment_light_style": "solarized-light" } -pygments_style = "rrt" - -html_css_files = [ - "css/custom.css", - "css/front.css", - "css/hljs.css" -] - html_js_files = [ 'js/custom.js', 'js/code.js' diff --git a/docs/source/features.rst b/docs/source/features.rst index 947da101..65d81ef4 100644 --- a/docs/source/features.rst +++ b/docs/source/features.rst @@ -1,5 +1,5 @@ Features -========== +========= .. raw:: html @@ -22,8 +22,8 @@ Features .. grid-item-card:: :link: notebooks/features/getting.ipynb - :class-card: code-surface - :class-body: code-surface + :class-card: surface + :class-body: surface .. raw:: html @@ -31,7 +31,7 @@ Features <div class="d-flex align-items-center"> <div class="d-flex justify-content-center" style="min-width: 50px; margin-right: 20px; height: 100%;"> - </div>  + </div> <div> <h5 class="card-title">Getting</h5> Accessing values @@ -40,8 +40,8 @@ Features .. grid-item-card:: :link: notebooks/features/setting.ipynb - :class-card: code-surface - :class-body: code-surface + :class-card: surface + :class-body: surface .. raw:: html @@ -49,7 +49,7 @@ Features <div class="d-flex align-items-center"> <div class="d-flex justify-content-center" style="min-width: 50px; margin-right: 20px; height: 100%;"> - </div>  + </div> <div> <h5 class="card-title">Setting</h5> Intervening on values @@ -58,8 +58,8 @@ Features .. grid-item-card:: :link: notebooks/features/operations.ipynb - :class-card: code-surface - :class-body: code-surface + :class-card: surface + :class-body: surface .. raw:: html @@ -67,7 +67,7 @@ Features <div class="d-flex align-items-center"> <div class="d-flex justify-content-center" style="min-width: 50px; margin-right: 20px; height: 100%;"> - </div>  + </div> <div> <h5 class="card-title">Operations</h5> Editing values @@ -76,8 +76,8 @@ Features .. grid-item-card:: :link: notebooks/features/modules.ipynb - :class-card: code-surface - :class-body: code-surface + :class-card: surface + :class-body: surface .. raw:: html @@ -85,7 +85,7 @@ Features <div class="d-flex align-items-center"> <div class="d-flex justify-content-center" style="min-width: 50px; margin-right: 20px; height: 100%;"> - </div>  + </div> <div> <h5 class="card-title">Modules</h5> Applying modules @@ -94,8 +94,8 @@ Features .. grid-item-card:: :link: notebooks/features/cross_prompt.ipynb - :class-card: code-surface - :class-body: code-surface + :class-card: surface + :class-body: surface .. raw:: html @@ -112,8 +112,8 @@ Features .. grid-item-card:: :link: notebooks/features/multiple_token.ipynb - :class-card: code-surface - :class-body: code-surface + :class-card: surface + :class-body: surface .. raw:: html @@ -121,7 +121,7 @@ Features <div class="d-flex align-items-center"> <div class="d-flex justify-content-center" style="min-width: 50px; margin-right: 20px; height: 100%;"> - </div>  + </div> <div> <h5 class="card-title">Generation</h5> Multiple tokens @@ -130,8 +130,8 @@ Features .. grid-item-card:: :link: notebooks/features/gradients.ipynb - :class-card: code-surface - :class-body: code-surface + :class-card: surface + :class-body: surface .. raw:: html @@ -139,7 +139,7 @@ Features <div class="d-flex align-items-center"> <div class="d-flex justify-content-center" style="min-width: 50px; margin-right: 20px; height: 100%;"> - </div>  + </div> <div> <h5 class="card-title">Working With Gradients</h5> Intervene on gradients @@ -148,8 +148,8 @@ Features .. grid-item-card:: :link: notebooks/features/remote_execution.ipynb - :class-card: code-surface - :class-body: code-surface + :class-card: surface + :class-body: surface .. raw:: html @@ -157,7 +157,7 @@ Features <div class="d-flex align-items-center"> <div class="d-flex justify-content-center" style="min-width: 50px; margin-right: 20px; height: 100%;"> - </div>  + </div> <div> <h5 class="card-title">Remote Execution</h5> Transparent deep learning @@ -166,15 +166,15 @@ Features .. grid-item-card:: :link: notebooks/features/token_indexing.ipynb - :class-card: code-surface - :class-body: code-surface + :class-card: surface + :class-body: surface .. raw:: html <div class="d-flex align-items-center"> <div class="d-flex justify-content-center" style="min-width: 50px; margin-right: 20px; height: 100%;"> - </div>  + </div> <div> <h5 class="card-title">Multi-Token Indexing</h5> Easy token indexing diff --git a/docs/source/index.rst b/docs/source/index.rst index 8f4da88f..c57c20a2 100644 --- a/docs/source/index.rst +++ b/docs/source/index.rst @@ -33,31 +33,27 @@ nnsight </script> .. raw:: html - - <link rel="stylesheet" href="https://cdn.jsdelivr.net/npm/[email protected]/font/bootstrap-icons.min.css"> - <script src="dist/clipboard.min.js"></script> - - <script> - new ClipboardJS('.btn'); - </script> - - <script> - // Select the element with the class 'some-class' - var element = document.querySelector('.bd-container'); - - var newElement = document.createElement('div'); - newElement.id = 'canvas'; - - newElement.style.position = 'fixed'; - newElement.style.zIndex = '-1'; - newElement.style.height = '100vh'; - newElement.style.width = '100vw'; - newElement.style.top = '0'; - - element.appendChild(newElement); - - </script> + <script type="text/javascript"> + document.addEventListener('DOMContentLoaded', function() { + var observer = new MutationObserver(function(mutations) { + updateCSSLink(); + }); + observer.observe(document.documentElement, {attributes: true, attributeFilter: ['data-theme']}); + + function updateCSSLink() { + const dark = document.documentElement.dataset.theme == 'dark'; + var link = document.querySelector('link[rel="stylesheet"][href*="hljs"]'); + if(link) { + link.href = dark ? '../_static/css/hljs-dark.css?v=0.2' : '../_static/css/hljs-light.css?v=0.2'; + } + } + + // Initial check in case the theme is already set before the observer is added + updateCSSLink(); + }); + </script> + <link rel="stylesheet" href="../_static/css/hljs-dark.css?v=0.2"> .. raw:: html @@ -82,7 +78,7 @@ nnsight </div> </section> - <section class="d-flex align-items-center middle-stuff mid-section"> + <section class="d-flex align-items-center mid-section d-none d-md-block"> <div class="px-4 text-center"> <div class="col-lg-10 mx-auto"> Direct access to model internals, from one to one hundred billion parameters. Intervene on activations or gradients in transformers, train optimizations methods, perform out of order module applications, cross prompt interventions, and so much more. @@ -90,7 +86,7 @@ nnsight </div> </section> - <section class="d-flex align-items-center" style="height: 60vh;"> + <section class="d-flex align-items-center d-none d-md-block" style="height: 60vh;"> <div class="container"> <div class="row align-items-end pb-3 border-bottom mb-5"> <div class="col"> @@ -135,4 +131,28 @@ nnsight </div> </section> - <section style="height: 20vh;"/> + <section class="d-none d-md-block" style="height: 20vh;"/> + + +.. raw:: html + + <script> + + var element = document.querySelector('.bd-container'); + + if (element) { + // If .bd-container exists, proceed to create the new element + var newElement = document.createElement('div'); + newElement.id = 'canvas'; + + newElement.style.position = 'fixed'; + newElement.style.zIndex = '-1'; + newElement.style.height = '100vh'; + newElement.style.width = '100vw'; + newElement.style.top = '0'; + + element.appendChild(newElement); + } else { + console.log(".bd-container element does not exist."); + } + </script> \ No newline at end of file diff --git a/docs/source/notebooks/tutorials/walkthrough.ipynb b/docs/source/notebooks/tutorials/walkthrough.ipynb index b148dafc..4ad70d96 100644 --- a/docs/source/notebooks/tutorials/walkthrough.ipynb +++ b/docs/source/notebooks/tutorials/walkthrough.ipynb @@ -1,16 +1,18 @@ { "cells": [ + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "# Walkthrough" + ] + }, { "cell_type": "markdown", "metadata": { "id": "-Scfr942GwO4" }, "source": [ - "<img src=\"https://nnsight.net/_images/nnsight_logo.svg\" alt=\"drawing\" width=\"200\"/>\n", - "\n", - "# **NNsight**\n", - "## The API for a transparent science on black-box AI\n", - "\n", "An interactive version of this walkthrough can be found [here](https://colab.research.google.com/github/JadenFiotto-Kaufman/nnsight/blob/dev/NNsight_v0_2.ipynb)\n", "\n", "In this era of large-scale deep learning, the most interesting AI models are massive black boxes that are hard to run. Ordinary commercial inference service APIs let you interact with huge models, but they do not let you access model internals.\n", @@ -54,7 +56,7 @@ "outputs": [], "source": [ "# Install nnsight\n", - "!pip install nnsight==0.2.3.dev0\n", + "!pip install nnsight\n", "\n", "from IPython.display import clear_output\n", "\n", @@ -1530,7 +1532,8 @@ "name": "python3" }, "language_info": { - "name": "python" + "name": "python", + "version": "3.8.0" }, "widgets": { "application/vnd.jupyter.widget-state+json": { diff --git a/docs/source/start.rst b/docs/source/start.rst index 7bec657a..732ca5f1 100644 --- a/docs/source/start.rst +++ b/docs/source/start.rst @@ -12,7 +12,7 @@ To get started with NNsight, install it with ``pip``. .. code-block:: console - pip install nnsight==0.2.3.dev0 + pip install nnsight First Steps diff --git a/docs/source/status.rst b/docs/source/status.rst index 4f308d6a..00d0650e 100644 --- a/docs/source/status.rst +++ b/docs/source/status.rst @@ -3,7 +3,50 @@ .. raw:: html - <link rel="stylesheet" href="../_static/css/status.css"> + <style> + .accordion-header { + margin: 0 !important; + } + /* Custom accordion styles */ + .custom-accordion-header { + background-color: var(--pst-color-surface); /* Default state background color */ + color: var(--pst-color-text-base); /* Text color */ + border-bottom: var(--pst-color-border); /* Border color */ + } + + .custom-accordion-header: { + background-color: var(--pst-color-surface); /* Default state background color */ + color: var(--pst-color-text-base); /* Text color */ + border-bottom: var(--pst-color-border); /* Border color */ + } + + .accordion { + --bs-accordion-btn-icon: none; + --bs-accordion-btn-active-icon: none; + } + + .custom-accordion-header.collapsed { + background-color: var(--pst-color-on-background); /* Collapsed state background color */ + color: var(--pst-color-text-base); /* Text color */ + } + + .custom-accordion-header:not(.collapsed) { + background-color: var(--pst-color-surface); /* Active/Expanded state background color */ + color: var(--pst-color-text-base); /* Text color */ + } + + .custom-accordion-body { + background-color: var(--pst-color-on-background); /* Body background color */ + border-color: var(--pst-color-border); /* Border color */ + color: var(--pst-color-text-base); /* Text color */ + } + + .sd-card { + border-radius: 0 !important; + } + </style> + + <script> function autoFormatJsonString(jsonString) { @@ -24,13 +67,14 @@ fetch("https://ndif.dev/ping") .then((response) => { if (response.status == 200) { - Array.from(document.getElementsByClassName("status-container")).forEach((elm) => { - elm.style.backgroundColor = "#66800b"; - Array.from(elm.getElementsByClassName('sd-card-text')).forEach((text) => { - text.textContent = "All Systems Are Operational"; - text.style.color = "#CECDC3"; + document.querySelectorAll('div.sd-card-body.status-container').forEach(el => { + el.style.backgroundColor = "#66800b"; + el.querySelectorAll('p.sd-card-text').forEach(el => { + el.textContent = "All Systems Are Operational"; }); }); + + console.log('Ping success'); // Nested fetch to ndif.dev/stats @@ -82,7 +126,17 @@ prettyPrintedJson = prettyPrintedJson.replace(/"([^"]+)":/g, '"$1":'); let huggingFaceLink = `<a href="http://huggingface.co/${key}" target="_blank">HuggingFace Model Repository ↗</a>`; - infoString += `<div class="accordion-item"> <h2 class="accordion-header" id="${headingId}"> <button class="accordion-button collapsed" type="button" data-bs-toggle="collapse" data-bs-target="#${collapseId}" aria-expanded="false" aria-controls="${collapseId}"> (${summaryItem.number_of_copies}x) ${key} </button> </h2> <div id="${collapseId}" class="accordion-collapse collapse" aria-labelledby="${headingId}" data-bs-parent="#accordionExample"> <div class="accordion-body">${huggingFaceLink} <pre>${prettyPrintedJson}</pre></div> </div> </div>`; + infoString += `<div class="accordion-item"> + <h2 class="accordion-header" id="${headingId}"> + <button class="accordion-button custom-accordion-header collapsed" type="button" data-bs-toggle="collapse" data-bs-target="#${collapseId}" aria-expanded="false" aria-controls="${collapseId}"> + (${summaryItem.number_of_copies}x) ${key} + </button> + </h2> + <div id="${collapseId}" class="accordion-collapse collapse" aria-labelledby="${headingId}" data-bs-parent="#accordionExample"> + <div class="accordion-body custom-accordion-body">${huggingFaceLink}<pre>${prettyPrintedJson}</pre></div> + </div> + </div>`; + index++; }); @@ -92,8 +146,6 @@ elm.innerHTML = infoString; - - console.log('Stats success'); }).catch((jsonError) => { console.log('JSON parsing error:', jsonError); @@ -106,22 +158,20 @@ console.log('Stats error'); }); } else { - Array.from(document.getElementsByClassName("status-container")).forEach((elm) => { - elm.style.backgroundColor = "#F84F31"; - Array.from(elm.getElementsByClassName('sd-card-text')).forEach((text) => { - text.textContent = "NDIF Is Unavailable"; - text.style.color = "#CECDC3"; + document.querySelectorAll('div.sd-card-body.status-container').forEach(el => { + el.style.backgroundColor = "#F84F31"; + el.querySelectorAll('p.sd-card-text').forEach(el => { + el.textContent = "NDIF Is Unavailable"; }); }); console.log('Ping error'); } }) .catch((pingError) => { - Array.from(document.getElementsByClassName("status-container")).forEach((elm) => { - elm.style.backgroundColor = "#F84F31"; - Array.from(elm.getElementsByClassName('sd-card-text')).forEach((text) => { - text.textContent = "NDIF Is Unavailable"; - text.style.color = "#CECDC3"; + document.querySelectorAll('div.sd-card-body.status-container').forEach(el => { + el.style.backgroundColor = "#F84F31"; + el.querySelectorAll('p.sd-card-text').forEach(el => { + el.textContent = "NDIF Is Unavailable"; }); }); console.error('Ping fetch failed:', pingError); @@ -139,10 +189,16 @@ Status :hidden: .. card:: - :class-card: status-container + :class-body: status-container + :shadow: none All Systems Are Operational +.. card:: + :shadow: none + + The library can be used to run local models without a key. Running experiments on remote models requires a free server API key. To obtain a key, join the `NDIF Discord <https://discord.gg/6uFJmCSwW7>`_ and say hello on the #introductions channel. + .. raw:: html <div class="accordion accordion-flush" id="accordionHook"> diff --git a/docs/source/tutorials.rst b/docs/source/tutorials.rst index 1728168a..3ad61261 100644 --- a/docs/source/tutorials.rst +++ b/docs/source/tutorials.rst @@ -28,7 +28,7 @@ Tutorials .. raw:: html <div class="d-flex align-items-center"> - <img src="../_static/images/tutorials/walkthrough.png" class="img-fluid" style="max-width: 50px; margin-right: 20px;"> + <img src="../_static/images/tutorials/walkthrough.webp" class="img-fluid" style="max-width: 50px; margin-right: 20px;"> <div> <h5 class="code-surface card-title">Walkthrough</h5> Learn the basics @@ -43,7 +43,7 @@ Tutorials .. raw:: html <div class="d-flex align-items-center"> - <img src="../_static/images/tutorials/activation_patching.png" class="img-fluid" style="max-width: 50px; margin-right: 20px;"> + <img src="../_static/images/tutorials/activation_patching.webp" class="img-fluid" style="max-width: 50px; margin-right: 20px;"> <div> <h5 class="code-surface card-title">Activation Patching</h5> Causal intervention @@ -58,7 +58,7 @@ Tutorials .. raw:: html <div class="d-flex align-items-center"> - <img src="../_static/images/tutorials/attribution_patching.png" class="img-fluid" style="max-width: 50px; margin-right: 20px;"> + <img src="../_static/images/tutorials/attribution_patching.webp" class="img-fluid" style="max-width: 50px; margin-right: 20px;"> <div> <h5 class="code-surface card-title">Attribution Patching</h5> Approximate patching @@ -73,7 +73,7 @@ Tutorials .. raw:: html <div class="d-flex align-items-center"> - <img src="../_static/images/tutorials/logit_lens.png" class="img-fluid" style="max-width: 50px; margin-right: 20px;"> + <img src="../_static/images/tutorials/logit_lens.webp" class="img-fluid" style="max-width: 50px; margin-right: 20px;"> <div> <h5 class="code-surface card-title">Logit Lens</h5> Decode activations @@ -88,7 +88,7 @@ Tutorials .. raw:: html <div class="d-flex align-items-center"> - <img src="../_static/images/tutorials/future_lens.png" class="img-fluid" style="max-width: 50px; margin-right: 20px;"> + <img src="../_static/images/tutorials/future_lens.webp" class="img-fluid" style="max-width: 50px; margin-right: 20px;"> <div> <h5 class="code-surface card-title">Future Lens</h5> Probe future tokens @@ -103,7 +103,7 @@ Tutorials .. raw:: html <div class="d-flex align-items-center"> - <img src="../_static/images/tutorials/function_vectors.png" class="img-fluid" style="max-width: 50px; margin-right: 20px;"> + <img src="../_static/images/tutorials/function_vectors.webp" class="img-fluid" style="max-width: 50px; margin-right: 20px;"> <div> <h5 class="code-surface card-title">Function Vectors</h5> Lambdas @@ -118,7 +118,7 @@ Tutorials .. raw:: html <div class="d-flex align-items-center"> - <img src="../_static/images/tutorials/dictionary_learning.png" class="img-fluid" style="max-width: 50px; margin-right: 20px;"> + <img src="../_static/images/tutorials/dictionary_learning.webp" class="img-fluid" style="max-width: 50px; margin-right: 20px;"> <div> <h5 class="code-surface card-title">Dictionary Learning</h5> Sparse autoencoders diff --git a/src/.DS_Store b/src/.DS_Store new file mode 100644 index 00000000..1b454a83 Binary files /dev/null and b/src/.DS_Store differ diff --git a/src/nnsight/__init__.py b/src/nnsight/__init__.py index 5c9a94a2..218e7bb6 100644 --- a/src/nnsight/__init__.py +++ b/src/nnsight/__init__.py @@ -2,7 +2,7 @@ import os import yaml - +import torch from .patching import * from .pydantics.Config import ConfigModel @@ -41,20 +41,49 @@ def _bool(self): def fake_tensor_new_wrapper(fn): - + @wraps(fn) def inner(cls, fake_mode, elem, device, constant=None): - + if isinstance(elem, FakeTensor): return elem - + else: return fn(cls, fake_mode, elem, device, constant=constant) - + return inner -DEFAULT_PATCHER.add(Patch(FakeTensor, fake_tensor_new_wrapper(FakeTensor.__new__), "__new__")) + +DEFAULT_PATCHER.add( + Patch(FakeTensor, fake_tensor_new_wrapper(FakeTensor.__new__), "__new__") +) + + +def onehot_wrapper(fn): + @wraps(fn) + def onehot(input: torch.Tensor, num_classes=-1): + if input.device.type == "meta": + return torch.zeros((*input.shape, num_classes), device="meta") + + else: + return fn(input, num_classes=num_classes) + + return onehot + + +DEFAULT_PATCHER.add( + Patch(torch.nn.functional, onehot_wrapper(torch.nn.functional.one_hot), "one_hot") +) + +def noop_wrapper(fn): + @wraps(fn) + def noop(input: torch.Tensor, *args, **kwargs): + return input + + return noop + +DEFAULT_PATCHER.add(Patch(FakeTensor, noop_wrapper(FakeTensor.tolist), "tolist")) DEFAULT_PATCHER.__enter__() diff --git a/src/nnsight/config.yaml b/src/nnsight/config.yaml index ef2dc060..98cdc221 100644 --- a/src/nnsight/config.yaml +++ b/src/nnsight/config.yaml @@ -1,5 +1,5 @@ API: - APIKEY: vZ8HcTCC436SEX3qTfsq + APIKEY: null HOST: ndif.dev APP: LOGGING: false diff --git a/src/nnsight/contexts/Runner.py b/src/nnsight/contexts/Runner.py index 8dff9799..31366502 100644 --- a/src/nnsight/contexts/Runner.py +++ b/src/nnsight/contexts/Runner.py @@ -39,10 +39,10 @@ def __exit__(self, exc_type, exc_val, exc_tb) -> None: if isinstance(exc_val, BaseException): raise exc_val - self._graph.tracing = False - if self.remote: self.run_server() + + self._graph = None else: super().__exit__(exc_type, exc_val, exc_tb) @@ -110,7 +110,7 @@ def handle_response(self, event: str, data) -> bool: return True # Or if there was some error. elif response.status == pydantics.ResponseModel.JobStatus.ERROR: - return True + raise Exception(str(response)) return False diff --git a/src/nnsight/contexts/Tracer.py b/src/nnsight/contexts/Tracer.py index 7dfd626d..2110774b 100644 --- a/src/nnsight/contexts/Tracer.py +++ b/src/nnsight/contexts/Tracer.py @@ -64,8 +64,6 @@ def __exit__(self, exc_type, exc_val, exc_tb) -> None: if isinstance(exc_val, BaseException): raise exc_val - self._graph.tracing = False - output = self._model.interleave( self._model._execute, self._graph, @@ -73,6 +71,8 @@ def __exit__(self, exc_type, exc_val, exc_tb) -> None: **self._kwargs, ) + self._graph = None + def invoke(self, *inputs: Tuple[Any], **kwargs) -> Invoker: """Create an Invoker context dor a given input. diff --git a/src/nnsight/envoy.py b/src/nnsight/envoy.py index 352e11b8..55dc6322 100644 --- a/src/nnsight/envoy.py +++ b/src/nnsight/envoy.py @@ -1,5 +1,6 @@ from __future__ import annotations +import warnings from typing import Any, Dict, Iterator, List, Optional, Union import torch @@ -59,7 +60,45 @@ def __init__(self, module: torch.nn.Module, module_path: str = ""): self._sub_envoys.append(envoy) - setattr(self, name, envoy) + # If the module already has a sub-module named 'input' or 'output', + # mount the proxy access to 'nns_input' or 'nns_output instead. + if hasattr(Envoy, name): + + self._handle_overloaded_mount(envoy, name) + + else: + + setattr(self, name, envoy) + + def _handle_overloaded_mount(self, envoy: Envoy, mount_point: str): + + warnings.warn( + f"Module of type `{type(self._module)}` has pre-defined a `{mount_point}` attribute. nnsight access for `{mount_point}` will be mounted at `.nns_{mount_point}` instead of `.{mount_point}` for this module only." + ) + + # If we already shifted a mount point dont create another new class. + if "Preserved" in self.__class__.__name__: + + new_cls = self.__class__ + + else: + + new_cls = type( + f"{Envoy.__name__}.Preserved", + (Envoy,), + {}, + ) + + # Get the normal proxy mount point + mount = getattr(new_cls, mount_point) + + # Move it to nns_<mount point> + setattr(new_cls, f"nns_{mount_point}", mount) + # Set the sub-module/envoy to the normal mount point on the CLASS itself not the instance. + setattr(new_cls, mount_point, envoy) + + # Update the class on the instance + self.__class__ = new_cls def _set_tracer(self, tracer: Tracer, propagate=True): """Set tracer object on Envoy. @@ -145,14 +184,76 @@ def next(self, increment: int = 1, propagate: bool = False) -> Envoy: return self - def __str__(self) -> str: + def _repr_module_list(self): + + list_of_reprs = [repr(item) for item in self._sub_envoys] + if len(list_of_reprs) == 0: + return self._module._get_name() + "()" + + start_end_indices = [[0, 0]] + repeated_blocks = [list_of_reprs[0]] + for i, r in enumerate(list_of_reprs[1:], 1): + if r == repeated_blocks[-1]: + start_end_indices[-1][1] += 1 + continue + + start_end_indices.append([i, i]) + repeated_blocks.append(r) + + lines = [] + main_str = self._module._get_name() + "(" + for (start_id, end_id), b in zip(start_end_indices, repeated_blocks): + local_repr = f"({start_id}): {b}" # default repr + + if start_id != end_id: + n = end_id - start_id + 1 + local_repr = f"({start_id}-{end_id}): {n} x {b}" + + local_repr = torch.nn.modules.module._addindent(local_repr, 2) + lines.append(local_repr) + + main_str += "\n " + "\n ".join(lines) + "\n" + main_str += ")" + return main_str + + def __repr__(self) -> str: """Wrapper method for underlying module's string representation. Returns: str: String. """ - return str(self._module) + if isinstance(self._module, torch.nn.ModuleList): + + return self._repr_module_list() + + extra_lines = [] + extra_repr = self._module.extra_repr() + # empty string will be split into list [''] + if extra_repr: + extra_lines = extra_repr.split("\n") + child_lines = [] + for attribute_name, attribute in self.__dict__.items(): + + if isinstance(attribute, Envoy): + + mod_str = repr(attribute) + mod_str = torch.nn.modules.module._addindent(mod_str, 2) + child_lines.append("(" + attribute_name + "): " + mod_str) + + lines = extra_lines + child_lines + + main_str = self._module._get_name() + "(" + if lines: + # simple one-liner info, which most builtin Modules will use + if len(extra_lines) == 1 and not child_lines: + main_str += extra_lines[0] + else: + main_str += "\n " + "\n ".join(lines) + "\n" + + main_str += ")" + + return main_str def __iter__(self) -> Iterator[Envoy]: """Wrapper method for underlying ModuleList iterator. @@ -214,8 +315,9 @@ def __call__(self, *args: List[Any], **kwargs: Dict[str, Any]) -> InterventionPr torch._GLOBAL_DEVICE_CONTEXT = None return proxy - + torch.set_default_device + @property def output(self) -> InterventionProxy: """ diff --git a/src/nnsight/intervention.py b/src/nnsight/intervention.py index 0ec31e1b..c2b2af8f 100644 --- a/src/nnsight/intervention.py +++ b/src/nnsight/intervention.py @@ -64,7 +64,7 @@ def save(self) -> InterventionProxy: # Add a 'null' node with the save proxy as an argument to ensure the values are never deleted. # This is because 'null' nodes never actually get set and therefore there will always be a # dependency for the save proxy. - self.node.graph.add( + self.node.add( value=None, target="null", args=[self.node], @@ -83,8 +83,13 @@ def grad(self) -> InterventionProxy: """ if self._grad is None: - self.__dict__["_grad"] = self.node.graph.add( - value=self.node.proxy_value, target="grad", args=[self.node] + # We track how many times backward is called via an attribute on the Graph + if not hasattr(self.node.graph, 'n_backward_calls'): + + setattr(self.node.graph, 'n_backward_calls', 0) + + self.__dict__["_grad"] = self.node.add( + value=self.node.proxy_value, target="grad", args=[self.node,self.node.graph.n_backward_calls] ) return self._grad @@ -97,9 +102,45 @@ def grad(self, value: Union[InterventionProxy, Any]) -> None: Args: value (Union[InterventionProxy, Any]): Value to set output to. """ - self.node.graph.add(target="swap", args=[self.grad.node, value], value=True) + self.node.add(target="swap", args=[self.grad.node, value], value=True) - self.__dict__["_grad"] = None + def __call__(self, *args, **kwargs) -> Self: + + + # We don't want to call backward on fake tensors + if ( + self.node.target is util.fetch_attr + and isinstance(self.node.args[1], str) + and self.node.args[1] == "backward" + ): + # We track how many times backward is called via an attribute on the Graph + if not hasattr(self.node.graph, 'n_backward_calls'): + + setattr(self.node.graph, 'n_backward_calls', 0) + + # Clear all .grad proxies + for node in self.node.graph.nodes.values(): + + try: + + if node.proxy._grad is not None: + + node.proxy.__dict__['_grad'] = None + + except ReferenceError: + pass + + self.node.graph.n_backward_calls += 1 + + return self.node.add( + value=None, + target=Proxy.proxy_call, + args=[self.node] + list(args), + kwargs=kwargs, + ) + + + return super().__call__(*args, **kwargs) def __setattr__( self, key: Union[InterventionProxy, Any], value: Union[Self, Any] @@ -118,12 +159,26 @@ def shape(self) -> Collection[torch.Size]: Union[torch.Size,Collection[torch.Size]]: Proxy value shape or collection of shapes. """ - if not self.node.graph.tracing: + if self.node.is_graph_dereferenced(): return util.apply(self.value, lambda x: x.shape, torch.Tensor) return util.apply(self.node.proxy_value, lambda x: x.shape, torch.Tensor) + @property + def device(self) -> Collection[torch.device]: + """Property to retrieve the device of the traced proxy value or real value. + + Returns: + Union[torch.Size,Collection[torch.device]]: Proxy value shape or collection of shapes. + """ + + if self.node.is_graph_dereferenced(): + + return util.apply(self.value, lambda x: x.device, torch.Tensor) + + return util.apply(self.node.proxy_value, lambda x: x.device, torch.Tensor) + def concat( activations: Any, diff --git a/src/nnsight/tracing/Graph.py b/src/nnsight/tracing/Graph.py index c16e268c..3c63e00a 100644 --- a/src/nnsight/tracing/Graph.py +++ b/src/nnsight/tracing/Graph.py @@ -1,6 +1,7 @@ from __future__ import annotations import inspect +import weakref from typing import Any, Callable, Dict, List, Type, Union import torch @@ -40,8 +41,6 @@ def __init__( proxy_class: Type[Proxy] = Proxy, validate: bool = True, ) -> None: - - self.tracing = True self.proxy_class = proxy_class self.validate = validate @@ -56,7 +55,6 @@ def __init__( self.module_proxy = self.add( value=module, target="argument", args=["nnsight_root_module"] ) - def get_swap(self, value): if self.swap is not None: @@ -165,7 +163,7 @@ def add( node = Node( name=name, - graph=self, + graph=weakref.proxy(self), value=value, target=target, args=args, @@ -173,10 +171,6 @@ def add( meta={"line": proxy_frame.lineno, "file": proxy_frame.filename}, ) - if not self.tracing: - - return node.value - self.name_idx[target_name] += 1 self.nodes[name] = node diff --git a/src/nnsight/tracing/Node.py b/src/nnsight/tracing/Node.py index d43ac175..3a789e14 100644 --- a/src/nnsight/tracing/Node.py +++ b/src/nnsight/tracing/Node.py @@ -1,8 +1,7 @@ from __future__ import annotations import inspect -import weakref -from typing import TYPE_CHECKING, Any, Callable, Dict, List, Tuple, Union +from typing import TYPE_CHECKING, Any, Callable, Dict, List, Optional, Tuple, Union import torch @@ -28,7 +27,6 @@ class Node: listeners (List[Node]): Nodes that depend on this node. dependencies (List[Node]): Nodes that this node depends on. value (Any): Actual value to be populated during execution. - _proxy_device (torch.device): desc """ @staticmethod @@ -59,10 +57,12 @@ def slice_to_value(arg: slice): device = ( torch._GLOBAL_DEVICE_CONTEXT.device if torch._GLOBAL_DEVICE_CONTEXT is not None - else "cpu" + else None ) - values = util.apply(values, lambda x: x.to(device), torch.Tensor) + if device is not None: + + values = util.apply(values, lambda x: x.to(device), torch.Tensor) return values @@ -86,13 +86,15 @@ def __init__( meta = dict() self.name = name - self.graph: "Graph" = weakref.proxy(graph) + self.graph: "Graph" = graph self.proxy_value = value self.target = target self.args: List = util.apply(args, lambda x: x.node, Proxy) self.kwargs: Dict = util.apply(kwargs, lambda x: x.node, Proxy) self.meta = meta + self.proxy: Optional[Proxy] = None + self.value: Any = inspect._empty self.listeners: List[Node] = list() @@ -126,17 +128,62 @@ def __init__( self.remaining_listeners = 0 self.remaining_dependencies = 0 - self._proxy_device: torch.device = None - self.compile() - # (for when you want to apply things to proxies after model execution?) + # (for when you want to apply things to proxies after model execution) if self.fulfilled() and not isinstance(self.target, str): # So it doesn't get destroyed. self.remaining_listeners = 1 self.execute() + def is_graph_dereferenced(self) -> bool: + """Checks to see if the weakref to the Graph is deleted. If it is, we're no longer tracing. + + Returns: + bool: Is Graph dereferenced. + """ + + try: + + self.graph.add + + except: + return True + + return False + + def add( + self, + target: Union[Callable, str], + value: Any = inspect._empty, + args: List[Any] = None, + kwargs: Dict[str, Any] = None, + name: str = None, + ) -> Proxy: + """We use Node.add vs Graph.add in case the weakref to Graph is gone. + + Returns: + Proxy: Proxy + """ + + if self.is_graph_dereferenced(): + + return Proxy( + Node( + name=None, + graph=None, + value=None, + target=target, + args=args, + kwargs=kwargs, + ) + ).value + + return self.graph.add( + target=target, value=value, args=args, kwargs=kwargs, name=name + ) + def compile(self) -> None: """Resets this Nodes remaining_listeners and remaining_dependencies and sets its value to None.""" self.remaining_listeners = len(self.listeners) @@ -227,14 +274,30 @@ def execute(self) -> None: elif self.target == "grad": + tensor: torch.Tensor = args[0] + backward_idx: int = args[1] + def grad(value): - self.set_value(value) - value = self.graph.get_swap(value) + nonlocal backward_idx - return value + if backward_idx == 0: - tensor: torch.Tensor = args[0] + self.set_value(value) + + if not self.is_graph_dereferenced(): + + value = self.graph.get_swap(value) + + backward_idx = -1 + + return value + + else: + + backward_idx -= 1 + + return None tensor.register_hook(lambda value: grad(value)) diff --git a/src/nnsight/tracing/Proxy.py b/src/nnsight/tracing/Proxy.py index 3a981840..002a79e4 100644 --- a/src/nnsight/tracing/Proxy.py +++ b/src/nnsight/tracing/Proxy.py @@ -1,7 +1,7 @@ from __future__ import annotations -import inspect import operator +import weakref from typing import TYPE_CHECKING, Any, Callable, Union from typing_extensions import Self @@ -36,6 +36,7 @@ def __init__(self, node: "Node") -> None: self.__dict__["node"] = node self.node: "Node" + self.node.proxy = weakref.proxy(self) @property def value(self) -> Any: @@ -52,7 +53,7 @@ def value(self) -> Any: def __str__(self) -> str: - if not self.node.graph.tracing: + if self.node.is_graph_dereferenced(): return str(self.value) @@ -62,7 +63,7 @@ def __str__(self) -> str: def __repr__(self) -> str: - if not self.node.graph.tracing: + if self.node.is_graph_dereferenced(): return repr(self.value) @@ -76,38 +77,26 @@ def __call__(self, *args, **kwargs) -> Self: Proxy: New call proxy. """ - value = inspect._empty - - # We don't want to call backward on fake tensors - if ( - self.node.target is util.fetch_attr - and isinstance(self.node.args[1], str) - and self.node.args[1] == "backward" - ): - - value = None - - return self.node.graph.add( - value=value, + return self.node.add( target=Proxy.proxy_call, args=[self.node] + list(args), kwargs=kwargs, ) def __getitem__(self, key: Union[Proxy, Any]) -> Self: - return self.node.graph.add( + return self.node.add( target=operator.getitem, args=[self.node, key], ) def __setitem__(self, key: Union[Proxy, Any], value: Union[Self, Any]) -> None: - self.node.graph.add( + self.node.add( target=operator.setitem, args=[self.node, key, value], ) def __getattr__(self, key: Union[Proxy, Any]) -> Self: - return self.node.graph.add( + return self.node.add( target=util.fetch_attr, args=[self.node, key], ) @@ -120,109 +109,109 @@ def __setattr__(self, key: Union[Proxy, Any], value: Union[Self, Any]) -> None: return - return self.node.graph.add( + return self.node.add( target=setattr, args=[self.node, key, value], ) def __len__(self) -> Self: - return self.node.graph.add( + return self.node.add( target=len, args=[self.node], ) def __abs__(self) -> Self: - return self.node.graph.add( + return self.node.add( target=operator.abs, args=[self.node], ) def __invert__(self) -> Self: - return self.node.graph.add( + return self.node.add( target=operator.invert, args=[self.node], ) def __add__(self, other: Union[Proxy, Any]) -> Self: - return self.node.graph.add( + return self.node.add( target=operator.add, args=[self.node, other], ) def __radd__(self, other: Union[Proxy, Any]) -> Self: - return self.node.graph.add( + return self.node.add( target=operator.add, args=[other, self.node], ) def __sub__(self, other: Union[Proxy, Any]) -> Self: - return self.node.graph.add( + return self.node.add( target=operator.sub, args=[self.node, other], ) def __rsub__(self, other: Union[Proxy, Any]) -> Self: - return self.node.graph.add( + return self.node.add( target=operator.sub, args=[other, self.node], ) def __pow__(self, other: Union[Proxy, Any]) -> Self: - return self.node.graph.add( + return self.node.add( target=operator.pow, args=[self.node, other], ) def __rpow__(self, other: Union[Proxy, Any]) -> Self: - return self.node.graph.add( + return self.node.add( target=operator.pow, args=[other, self.node], ) def __mul__(self, other: Union[Proxy, Any]) -> Self: - return self.node.graph.add( + return self.node.add( target=operator.mul, args=[self.node, other], ) def __rmul__(self, other: Union[Proxy, Any]) -> Self: - return self.node.graph.add( + return self.node.add( target=operator.mul, args=[other, self.node], ) def __mod__(self, other: Union[Proxy, Any]) -> Self: - return self.node.graph.add( + return self.node.add( target=operator.mod, args=[self.node, other], ) def __rmod__(self, other: Union[Proxy, Any]) -> Self: - return self.node.graph.add( + return self.node.add( target=operator.mod, args=[other, self.node], ) def __matmul__(self, other: Union[Proxy, Any]) -> Self: - return self.node.graph.add( + return self.node.add( target=operator.matmul, args=[self.node, other], ) def __rmatmul__(self, other: Union[Proxy, Any]) -> Self: - return self.node.graph.add( + return self.node.add( target=operator.matmul, args=[other, self.node], ) def __truediv__(self, other: Union[Proxy, Any]) -> Self: - return self.node.graph.add( + return self.node.add( target=operator.truediv, args=[self.node, other], ) def __rtruediv__(self, other: Union[Proxy, Any]) -> Self: - return self.node.graph.add( + return self.node.add( target=operator.truediv, args=[other, self.node], ) @@ -252,7 +241,7 @@ def get_proxy(arg): util.apply(args, get_proxy, Proxy) - return proxy.node.graph.add( + return proxy.node.add( target=orig_method, args=args, kwargs=kwargs, @@ -288,7 +277,7 @@ def get_node(proxy: Proxy): util.apply(list(args) + list(kwargs.values()), get_node, Proxy) if node is not None: - return node.graph.add(target=fn, args=args, kwargs=kwargs) + return node.add(target=fn, args=args, kwargs=kwargs) else: return fn(*args, **kwargs)

Raise Exception on ERROR status of remote close #73

2024-02-23T20:01:02

0.0

[]

ndif-team/nnsight

ndif-team__nnsight-36

cab401e14545e1a24c2671132404ee3a7ac60456

diff --git a/.DS_Store b/.DS_Store new file mode 100644 index 00000000..87cd3f95 Binary files /dev/null and b/.DS_Store differ diff --git a/README.md b/README.md index aae77ba0..d4e8aa23 100644 --- a/README.md +++ b/README.md @@ -198,7 +198,7 @@ model = LanguageModel('gpt2', device_map='cuda') with model.generate(max_new_tokens=1) as generator: with generator.invoke('The Eiffel Tower is in the city of') as invoker: - hidden_states_pre = model.transformer.h[-1].mlp.output.save() + hidden_states_pre = model.transformer.h[-1].mlp.output.clone().save() noise = (0.001**0.5)*torch.randn(hidden_states_pre.shape) diff --git a/docs/.DS_Store b/docs/.DS_Store new file mode 100644 index 00000000..8bf5b4bc Binary files /dev/null and b/docs/.DS_Store differ diff --git a/docs/source/.DS_Store b/docs/source/.DS_Store new file mode 100644 index 00000000..bbf044e0 Binary files /dev/null and b/docs/source/.DS_Store differ diff --git a/docs/source/_static/.DS_Store b/docs/source/_static/.DS_Store new file mode 100644 index 00000000..f26591fc Binary files /dev/null and b/docs/source/_static/.DS_Store differ diff --git a/docs/source/_static/css/custom.css b/docs/source/_static/css/custom.css index 510f679a..d6257901 100644 --- a/docs/source/_static/css/custom.css +++ b/docs/source/_static/css/custom.css @@ -28,6 +28,10 @@ html[data-theme="light"] { } +.ndif { + color: "green"; +} + .bd-sidebar-primary { width: fit-content !important; padding-right: 40px !important @@ -48,6 +52,15 @@ img { pointer-events: none; } +.img-bottom-sizing { + max-height: 10vh; + width: auto +} + +.body-sizing{ + height: 10vh; +} + .title-bot { margin-bottom: -10px !important; line-height: normal !important; diff --git a/docs/source/_static/images/.DS_Store b/docs/source/_static/images/.DS_Store new file mode 100644 index 00000000..cc051dec Binary files /dev/null and b/docs/source/_static/images/.DS_Store differ diff --git a/docs/source/_static/images/attribute_lens.png b/docs/source/_static/images/attribute_lens.png new file mode 100644 index 00000000..ea63e99c Binary files /dev/null and b/docs/source/_static/images/attribute_lens.png differ diff --git a/docs/source/_static/images/ioi.png b/docs/source/_static/images/ioi.png new file mode 100644 index 00000000..6c323e55 Binary files /dev/null and b/docs/source/_static/images/ioi.png differ diff --git a/docs/source/_templates/ndif_status.html b/docs/source/_templates/ndif_status.html index 065ba225..b1e43700 100644 --- a/docs/source/_templates/ndif_status.html +++ b/docs/source/_templates/ndif_status.html @@ -1,27 +1,33 @@ -<style> - .ndif_status { - height: 18px; - width: 18px; - background-color: red; - border-radius: 50%; - display: inline-block; - margin-left: 5px; - vertical-align: middle; - } -</style> - -NDIF Status: - <script> + let statusIcon = document.querySelector('.ndif .fa-circle-check'); fetch("{{ndif_url}}") .then((response) => { if (response.status == 200) { - Array.from(document.getElementsByClassName("ndif_status")).forEach((elm) => elm.style.backgroundColor = "green"); + Array.from(document.getElementsByClassName("ndif")).forEach((ndifElement) => { + Array.from(ndifElement.getElementsByTagName('span')).forEach((spanElement) => { + spanElement.style.setProperty('color', '#23C552', 'important'); + }); + }); } else { - Array.from(document.getElementsByClassName("ndif_status")).forEach((elm) => elm.style.backgroundColor = "red"); + Array.from(document.getElementsByClassName("ndif")).forEach((ndifElement) => { + Array.from(ndifElement.getElementsByTagName('span')).forEach((spanElement) => { + spanElement.style.setProperty('color', '#F84F31', 'important'); + }); + }); + if (statusIcon) { + statusIcon.classList.replace('fa-circle-check', 'fa-circle-xmark'); + } + } + }) + .catch((response) => { + Array.from(document.getElementsByClassName("ndif")).forEach((ndifElement) => { + Array.from(ndifElement.getElementsByTagName('span')).forEach((spanElement) => { + spanElement.style.setProperty('color', '#F84F31', 'important'); + }); + }); + if (statusIcon) { + statusIcon.classList.replace('fa-circle-check', 'fa-circle-xmark'); } }) - .catch((response) => Array.from(document.getElementsByClassName("ndif_status")).forEach((elm) => elm.style.backgroundColor = "red")) - </script> \ No newline at end of file diff --git a/docs/source/about.rst b/docs/source/about.rst index 058f335a..bdbe3f06 100644 --- a/docs/source/about.rst +++ b/docs/source/about.rst @@ -38,7 +38,7 @@ Here is how it looks: model = LanguageModel('meta-llama/Llama-2-70b-hf') with model.forward(remote=True) as runner: with runner.invoke('The Eiffel Tower is in the city of ') as invoker: - hidden_state = model.layers[10].input.save() # save one hidden state + hidden_state = model.layers[10].input[0].save() # save one hidden state model.layers[11].mlp.output = 0 # change one MLP module output print('The model predicts', runner.output) print('The internal state was', hidden_state.value) diff --git a/docs/source/conf.py b/docs/source/conf.py index 5420d720..ba7bfc8f 100644 --- a/docs/source/conf.py +++ b/docs/source/conf.py @@ -31,9 +31,10 @@ html_favicon = "_static/images/icon.ico" html_show_sourcelink = False html_theme_options = { - "logo": {"text": "nnsight"}, + # "logo": {"text": "nnsight"}, "show_nav_level": 2, "navbar_end": ["navbar-icon-links", "ndif_status"], + # "navbar_end": ["navbar-icon-links"], "navbar_align": "left", "icon_links": [ { @@ -46,6 +47,12 @@ "url": "https://discord.gg/6uFJmCSwW7", "icon": "fa-brands fa-discord", }, + { + "name": "Status", + "url": "https://discord.gg/6uFJmCSwW7", + "icon": "fa-solid fa-circle-check", + "attributes": {"class": "ndif"}, + }, ], } diff --git a/docs/source/index.rst b/docs/source/index.rst index 6c6e2ce2..885a1d83 100644 --- a/docs/source/index.rst +++ b/docs/source/index.rst @@ -14,6 +14,10 @@ nnsight tutorials About <about> +.. toctree:: + :hidden: + status + .. grid:: 1 1 2 2 :class-container: hero :reverse: diff --git a/docs/source/notebooks/features/setting.ipynb b/docs/source/notebooks/features/setting.ipynb index d489854d..d2f1b571 100644 --- a/docs/source/notebooks/features/setting.ipynb +++ b/docs/source/notebooks/features/setting.ipynb @@ -30,13 +30,13 @@ "with model.generate(max_new_tokens=1) as generator:\n", " with generator.invoke('The Eiffel Tower is in the city of') as invoker:\n", "\n", - " hidden_states_pre = model.transformer.h[-1].output[0][:].save()\n", + " hidden_states_pre = model.transformer.h[-1].output[0].clone().save()\n", "\n", " noise = (0.001**0.5)*torch.randn(hidden_states_pre.shape)\n", "\n", " model.transformer.h[-1].output[0][:] = hidden_states_pre + noise\n", "\n", - " hidden_states_post = model.transformer.h[-1].output[0][:].save()" + " hidden_states_post = model.transformer.h[-1].output[0].save()" ] }, { diff --git a/docs/source/notebooks/features/token_indexing.ipynb b/docs/source/notebooks/features/token_indexing.ipynb index 79080e8c..948979b8 100644 --- a/docs/source/notebooks/features/token_indexing.ipynb +++ b/docs/source/notebooks/features/token_indexing.ipynb @@ -102,10 +102,10 @@ "\n", "with model.forward() as runner:\n", " with runner.invoke('The') as invoker:\n", - " incorrect_a = model.transformer.input[0][:,0].save()\n", + " incorrect_a = model.transformer.input[0][0][:,0].save()\n", " \n", " with runner.invoke('The Eiffel Tower is in the city of''The Eiffel Tower is in the city of') as invoker:\n", - " incorrect_b = model.transformer.input[0][:,0].save()\n", + " incorrect_b = model.transformer.input[0][0][:,0].save()\n", "\n", "print(f\"Shorter input: {incorrect_a.value}\")\n", "print(f\"Longer input: {incorrect_b.value}\")" @@ -157,10 +157,10 @@ "source": [ "with model.forward() as runner:\n", " with runner.invoke('The') as invoker:\n", - " correct_a = model.transformer.input[0].t[0].save()\n", + " correct_a = model.transformer.input[0][0].t[0].save()\n", " \n", " with runner.invoke('The Eiffel Tower is in the city of') as invoker:\n", - " correct_b = model.transformer.input[0].t[0].save()\n", + " correct_b = model.transformer.input[0][0].t[0].save()\n", "\n", "print(f\"Shorter input: {correct_a.value}\")\n", "print(f\"Longer input: {correct_b.value}\")" diff --git a/docs/source/notebooks/tutorials/attribution_patching.ipynb b/docs/source/notebooks/tutorials/attribution_patching.ipynb index 46998c72..cb2a98c1 100644 --- a/docs/source/notebooks/tutorials/attribution_patching.ipynb +++ b/docs/source/notebooks/tutorials/attribution_patching.ipynb @@ -154,7 +154,7 @@ " clean_logit_diff = get_logit_diff(clean_logits, answer_token_indices).item()\n", "\n", " clean_cache = [\n", - " model.transformer.h[i].attn.attn_dropout.input.save()\n", + " model.transformer.h[i].attn.attn_dropout.input[0].save()\n", " for i in range(len(model.transformer.h))\n", " ]\n", "\n", @@ -171,7 +171,7 @@ " ).item()\n", "\n", " corrupted_cache = [\n", - " model.transformer.h[i].attn.attn_dropout.input.save()\n", + " model.transformer.h[i].attn.attn_dropout.input[0].save()\n", " for i in range(len(model.transformer.h))\n", " ]\n", "\n", diff --git a/docs/source/notebooks/tutorials/dictionary_learning.ipynb b/docs/source/notebooks/tutorials/dictionary_learning.ipynb deleted file mode 100644 index fb9e75a5..00000000 --- a/docs/source/notebooks/tutorials/dictionary_learning.ipynb +++ /dev/null @@ -1,25 +0,0 @@ -{ - "cells": [ - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "# Dictionary Learning" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "Coming Soon!" - ] - } - ], - "metadata": { - "language_info": { - "name": "python" - } - }, - "nbformat": 4, - "nbformat_minor": 2 -} diff --git a/docs/source/notebooks/walkthrough.ipynb b/docs/source/notebooks/walkthrough.ipynb index bae480d8..94e85469 100644 --- a/docs/source/notebooks/walkthrough.ipynb +++ b/docs/source/notebooks/walkthrough.ipynb @@ -75,7 +75,7 @@ "\n", "Use `dispatch=True` to load the model onto the specified device locally on initialization, not on first use.\n", "\n", - "`LanguageModel` is one wrapper for `nnsight` functionality. Another one `DiffusionModel` for text-to-image disffusion models. We encourage you to create you own by inheriting from `AbstractModel`\n", + "`LanguageModel` is one wrapper for `nnsight` functionality. Another one `DiffusionModel` for text-to-image disffusion models. We encourage you to create your own by inheriting from `AbstractModel`\n", "\n", "</details>\n", "\n", @@ -1303,7 +1303,7 @@ "px.line(\n", " [layer_probs.value[mary_token].item() for layer_probs in all_probs],\n", " title=\"Probability of Mary after each layer, according to logit lens\",\n", - " labels={\"value\":\"Layer\", \"index\":\"Probability\"}\n", + " labels={\"value\":\"Probability\", \"index\":\"Layer\"}\n", ")" ] }, diff --git a/docs/source/status.rst b/docs/source/status.rst new file mode 100644 index 00000000..8165df30 --- /dev/null +++ b/docs/source/status.rst @@ -0,0 +1,77 @@ +:html_theme.sidebar_secondary.remove: +:sd_hide_title: + +status +====== + +.. toctree:: + :maxdepth: 1 + :hidden: + + + +.. card:: + + All Systems Are Operational + + + +.. div:: sd-fs-1 sd-font-weight-bold sd-text-center sd-text-primary sd-mb-5 + + Key Features + +.. grid:: 1 1 2 2 + :class-container: features + + .. grid-item:: + + .. div:: features-container + + .. image:: _static/images/one.png + :width: 250 + + .. div:: + + **Integration** + + Pass in a ID from any HuggingFace Transformer repo and access its weights with nnsight. + + .. grid-item:: + + .. div:: features-container + + .. image:: _static/images/two.png + :width: 250 + + .. div:: + + **Interpretability** + + Access the internal gradients and activations at any point or module in a model. + + .. grid-item:: + + .. div:: features-container + + .. image:: _static/images/two.png + :width: 250 + + .. div:: + + **Intuitive** + + Set up a context block and manipulate model internals with only a couple lines of code. + + .. grid-item:: + + .. div:: features-container + + .. image:: _static/images/one.png + :width: 250 + + .. div:: + + **Interoperable** + + Enable grad and train interventions like LORA or probes on any point in a model. + diff --git a/docs/source/tutorials.rst b/docs/source/tutorials.rst index bb5b8e17..bc01aca5 100644 --- a/docs/source/tutorials.rst +++ b/docs/source/tutorials.rst @@ -1,17 +1,54 @@ Tutorials ============ +.. grid:: 2 + :gutter: 2 + + .. grid-item-card:: Walkthrough + :link: notebooks/walkthrough.ipynb + + Main features of the nnsight library. + + .. grid-item-card:: IOI Patching + :link: notebooks/tutorials/ioi_patching.ipynb + + :bdg-primary:`arXiv:2211.00593` + + .. grid-item-card:: Attribution Patching + :link: notebooks/tutorials/attribution_patching.ipynb + + :bdg-primary:`arXiv:2308.09124` + + .. grid-item-card:: Logit Lens + :link: notebooks/tutorials/logit_lens.ipynb + + :bdg-primary:`nostalgebraist` + + .. grid-item-card:: Future Lens + :link: notebooks/tutorials/future_lens.ipynb + + :bdg-primary:`arXiv:2311.04897` + + .. grid-item-card:: Function Vectors + :link: notebooks/tutorials/function_vectors.ipynb + + :bdg-primary:`arXiv:2310.15213` + + .. grid-item-card:: Dictionary Learning + :link: notebooks/tutorials/sae.ipynb + + :bdg-primary:`arXiv:2309.08600` + + .. toctree:: :maxdepth: 1 + :hidden: notebooks/tutorials/ioi_patching.ipynb notebooks/tutorials/attribution_patching.ipynb notebooks/tutorials/logit_lens.ipynb notebooks/tutorials/future_lens.ipynb notebooks/tutorials/function_vectors.ipynb - notebooks/tutorials/dictionary_learning.ipynb + notebooks/tutorials/sae.ipynb -.. toctree:: - :hidden: - notebooks/tutorials/sae.ipynb diff --git a/src/nnsight/contexts/Runner.py b/src/nnsight/contexts/Runner.py index 15890f2f..cd53714e 100644 --- a/src/nnsight/contexts/Runner.py +++ b/src/nnsight/contexts/Runner.py @@ -1,5 +1,8 @@ from __future__ import annotations +import pickle + +import requests import socketio from .. import CONFIG, pydantics @@ -101,7 +104,7 @@ def blocking_request(self, request: pydantics.RequestModel): f"wss://{CONFIG.API.HOST}", socketio_path="/ws/socket.io", transports=["websocket"], - wait_timeout=10 + wait_timeout=10, ) # Called when receiving a response from the server. @@ -116,10 +119,15 @@ def blocking_response(data): # If the status of the response is completed, update the local nodes that the user specified to save. # Then disconnect and continue. if response.status == pydantics.ResponseModel.JobStatus.COMPLETED: - for name, value in response.saves.items(): + with requests.get( + url=f"https://{CONFIG.API.HOST}/retrieve/{response.id}", stream=True + ) as stream: + result_response = pydantics.ResponseModel(**pickle.load(stream.raw)) + + for name, value in result_response.result.saves.items(): self.graph.nodes[name].value = value - self.output = response.output + self.output = result_response.result.output sio.disconnect() # Or if there was some error. diff --git a/src/nnsight/intervention.py b/src/nnsight/intervention.py index e533fc51..8dcadb1c 100644 --- a/src/nnsight/intervention.py +++ b/src/nnsight/intervention.py @@ -151,6 +151,7 @@ def _concat(values): key: _concat([value[key] for value in values]) for key in values[0].keys() } + return values[0] # As interventions are scoped only to their relevant batch, if we want to swap in values for this batch # we need to concatenate the batches before and after the relevant batch with the new values. @@ -194,7 +195,7 @@ def intervene(activations: Any, module_path: str, graph: Graph, key: str): # Key to module activation argument nodes has format: <module path>.<output/input>.<generation index> module_path = f"{module_path}.{key}.{graph.generation_idx}" - + if module_path in graph.argument_node_names: argument_node_names = graph.argument_node_names[module_path] @@ -206,6 +207,7 @@ def intervene(activations: Any, module_path: str, graph: Graph, key: str): _, batch_size, batch_start = node.args # We set its result to the activations, indexed by only the relevant batch idxs. + value = util.apply( activations, lambda x: x.narrow(0, batch_start, batch_size), @@ -279,10 +281,10 @@ def __enter__(self) -> HookModel: if hook_type == "input": - def input_hook(module, input, module_path=module_path): - return self.input_hook(input, module_path) + def input_hook(module, input, kwargs, module_path=module_path): + return self.input_hook((input, kwargs), module_path) - self.handles.append(module.register_forward_pre_hook(input_hook)) + self.handles.append(module.register_forward_pre_hook(input_hook, with_kwargs=True)) elif hook_type == "output": diff --git a/src/nnsight/models/AbstractModel.py b/src/nnsight/models/AbstractModel.py index 1874e380..320eca99 100644 --- a/src/nnsight/models/AbstractModel.py +++ b/src/nnsight/models/AbstractModel.py @@ -92,7 +92,7 @@ def meta_deepcopy(self, memo): with Patcher() as patcher: patcher.add( Patch( - torch.nn.parameter.Parameter.__deepcopy__, meta_deepcopy + torch.nn.parameter.Parameter, meta_deepcopy, "__deepcopy__" ) ) diff --git a/src/nnsight/pydantics/Response.py b/src/nnsight/pydantics/Response.py index ea4a6ec3..effbb44c 100644 --- a/src/nnsight/pydantics/Response.py +++ b/src/nnsight/pydantics/Response.py @@ -9,6 +9,16 @@ from pydantic import BaseModel, field_validator +class ResultModel(BaseModel): + id: str + output: Any = None + saves: Any = None + + @field_validator("output", "saves") + def unpickle(cls, value: bytes): + return pickle.loads(value) + + class ResponseModel(BaseModel): class JobStatus(Enum): RECEIVED = "RECEIVED" @@ -22,11 +32,11 @@ class JobStatus(Enum): description: str received: datetime = None - saves: Union[bytes, Any] = None - output: Union[bytes, Any] = None session_id: str = None blocking: bool = False + result: Union[bytes, ResultModel] = None + def __str__(self) -> str: return f"{self.id} - {self.status.name}: {self.description}" @@ -37,8 +47,3 @@ def log(self, logger: logging.Logger) -> ResponseModel: logger.info(str(self)) return self - - @field_validator("output", "saves") - @classmethod - def unpickle(cls, value): - return pickle.loads(value)

Tutorial Page Reformat and Status Tweaks Changed status logo and reformatted tutorial page

2024-01-01T21:04:32

0.0

[]

Arize-ai/openinference

Arize-ai__openinference-1123

534908982d871ac7e481738109cb06d23f9294f9

diff --git a/js/.changeset/mean-walls-talk.md b/js/.changeset/mean-walls-talk.md new file mode 100644 index 000000000..579de1113 --- /dev/null +++ b/js/.changeset/mean-walls-talk.md @@ -0,0 +1,5 @@ +--- +"@arizeai/openinference-instrumentation-langchain": patch +--- + +fix: add support for capturing ChatBedrock token counts diff --git a/js/packages/openinference-instrumentation-langchain/src/utils.ts b/js/packages/openinference-instrumentation-langchain/src/utils.ts index d3f7abdb9..01423be37 100644 --- a/js/packages/openinference-instrumentation-langchain/src/utils.ts +++ b/js/packages/openinference-instrumentation-langchain/src/utils.ts @@ -478,6 +478,8 @@ function getTokenCount(maybeCount: unknown) { * Formats the token counts of a langchain run into OpenInference attributes. * @param outputs - The outputs of a langchain run * @returns The OpenInference attributes for the token counts + * + * @see https://github.com/langchain-ai/langchainjs/blob/main/langchain-core/src/language_models/chat_models.ts#L403 for how token counts get added to outputs */ function formatTokenCounts( outputs: Run["outputs"], @@ -485,6 +487,34 @@ function formatTokenCounts( if (!isObject(outputs)) { return null; } + const firstGeneration = getFirstOutputGeneration(outputs); + + /** + * Some community models have non standard output structures and show token counts in different places notable ChatBedrock + * @see https://github.com/langchain-ai/langchainjs/blob/a173e300ef9ada416220876a2739e024b3a7f268/libs/langchain-community/src/chat_models/bedrock/web.ts + */ + // Generations is an array of arrays containing messages + const maybeGenerationComponent = + firstGeneration != null ? firstGeneration[0] : null; + const maybeMessage = isObject(maybeGenerationComponent) + ? maybeGenerationComponent.message + : null; + const usageMetadata = isObject(maybeMessage) + ? maybeMessage.usage_metadata + : null; + if (isObject(usageMetadata)) { + return { + [SemanticConventions.LLM_TOKEN_COUNT_COMPLETION]: getTokenCount( + usageMetadata.output_tokens, + ), + [SemanticConventions.LLM_TOKEN_COUNT_PROMPT]: getTokenCount( + usageMetadata.input_tokens, + ), + [SemanticConventions.LLM_TOKEN_COUNT_TOTAL]: getTokenCount( + usageMetadata.total_tokens, + ), + }; + } const llmOutput = outputs.llmOutput; if (!isObject(llmOutput)) { return null; @@ -519,6 +549,33 @@ function formatTokenCounts( ), }; } + + /** + * In some cases community models have a different output structure do to the way they extend the base model + * Notably ChatBedrock may have tokens stored in this format instead of normalized + * @see https://github.com/langchain-ai/langchainjs/blob/a173e300ef9ada416220876a2739e024b3a7f268/libs/langchain-community/src/chat_models/bedrock/web.ts for ChatBedrock + * and + * @see https://github.com/langchain-ai/langchainjs/blob/main/langchain-core/src/language_models/chat_models.ts#L403 for nomalization + */ + if (isObject(llmOutput.usage)) { + const maybePromptTokens = getTokenCount(llmOutput.usage.input_tokens); + const maybeCompletionTokens = getTokenCount(llmOutput.usage.output_tokens); + let maybeTotalTokens = getTokenCount(llmOutput.usage.total_tokens); + if (maybeTotalTokens == null) { + maybeTotalTokens = + isNumber(maybePromptTokens) && isNumber(maybeCompletionTokens) + ? maybePromptTokens + maybeCompletionTokens + : undefined; + } + return { + [SemanticConventions.LLM_TOKEN_COUNT_COMPLETION]: getTokenCount( + maybeCompletionTokens, + ), + [SemanticConventions.LLM_TOKEN_COUNT_PROMPT]: + getTokenCount(maybePromptTokens), + [SemanticConventions.LLM_TOKEN_COUNT_TOTAL]: maybeTotalTokens, + }; + } return null; } @@ -589,7 +646,7 @@ function formatMetadata(run: Run) { return null; } return { - metadata: safelyJSONStringify(run.extra.metadata), + [SemanticConventions.METADATA]: safelyJSONStringify(run.extra.metadata), }; }

[bug][langchain][js] langchain js instrumentor missing token counts for claude 3.5 sonnet LangChain JS instrumentation reported to be missing token counts for Anthropic https://arize-ai.slack.com/archives/C04R3GXC8HK/p1731387916527799

2024-11-15T00:34:13

0.0

[]

Arize-ai/openinference

Arize-ai__openinference-1113

a0e6f30c84011c5c743625bb69b66ba055ac17bd

diff --git a/python/instrumentation/openinference-instrumentation-dspy/examples/requirements.txt b/python/instrumentation/openinference-instrumentation-dspy/examples/requirements.txt index 3c0ec027a..76e21f6e3 100644 --- a/python/instrumentation/openinference-instrumentation-dspy/examples/requirements.txt +++ b/python/instrumentation/openinference-instrumentation-dspy/examples/requirements.txt @@ -1,4 +1,4 @@ -dspy-ai >= 2.1.0 +dspy >= 2.5.0 opentelemetry-sdk opentelemetry-exporter-otlp openinference-instrumentation-dspy \ No newline at end of file diff --git a/python/instrumentation/openinference-instrumentation-dspy/pyproject.toml b/python/instrumentation/openinference-instrumentation-dspy/pyproject.toml index 5c0c64f4e..b407373df 100644 --- a/python/instrumentation/openinference-instrumentation-dspy/pyproject.toml +++ b/python/instrumentation/openinference-instrumentation-dspy/pyproject.toml @@ -35,10 +35,10 @@ dependencies = [ [project.optional-dependencies] instruments = [ - "dspy-ai >= 2.5.0", + "dspy >= 2.5.0", ] test = [ - "dspy-ai==2.5.0", + "dspy>=2.5.27", "google-generativeai", "opentelemetry-sdk", "pytest-recording", diff --git a/python/instrumentation/openinference-instrumentation-dspy/src/openinference/instrumentation/dspy/package.py b/python/instrumentation/openinference-instrumentation-dspy/src/openinference/instrumentation/dspy/package.py index 5825ce077..02cd1b833 100644 --- a/python/instrumentation/openinference-instrumentation-dspy/src/openinference/instrumentation/dspy/package.py +++ b/python/instrumentation/openinference-instrumentation-dspy/src/openinference/instrumentation/dspy/package.py @@ -1,2 +1,2 @@ -_instruments = ("dspy-ai >= 2.1.0",) +_instruments = ("dspy >= 2.5.0",) _supports_metrics = False

[bug] dspy instrumentor failing CI unit tests for the latest dspy version

2024-11-12T04:40:05

0.0

[]

Arize-ai/openinference

Arize-ai__openinference-1087

473701d3dd2f8bcda7afe12bb65a1145d0a4d893

diff --git a/python/instrumentation/openinference-instrumentation-anthropic/src/openinference/instrumentation/anthropic/_wrappers.py b/python/instrumentation/openinference-instrumentation-anthropic/src/openinference/instrumentation/anthropic/_wrappers.py index c5d0e8204..050200479 100644 --- a/python/instrumentation/openinference-instrumentation-anthropic/src/openinference/instrumentation/anthropic/_wrappers.py +++ b/python/instrumentation/openinference-instrumentation-anthropic/src/openinference/instrumentation/anthropic/_wrappers.py @@ -22,6 +22,7 @@ OpenInferenceMimeTypeValues, OpenInferenceSpanKindValues, SpanAttributes, + ToolAttributes, ToolCallAttributes, ) @@ -98,6 +99,7 @@ def __call__( _get_llm_prompts(llm_prompt), _get_inputs(arguments), _get_llm_invocation_parameters(llm_invocation_parameters), + _get_llm_tools(llm_invocation_parameters), ) ), ) as span: @@ -150,6 +152,7 @@ async def __call__( _get_llm_prompts(llm_prompt), _get_inputs(arguments), _get_llm_invocation_parameters(invocation_parameters), + _get_llm_tools(invocation_parameters), ) ), ) as span: @@ -201,6 +204,7 @@ def __call__( _get_llm_span_kind(), _get_llm_input_messages(llm_input_messages), _get_llm_invocation_parameters(invocation_parameters), + _get_llm_tools(invocation_parameters), _get_inputs(arguments), ) ), @@ -260,6 +264,7 @@ async def __call__( _get_llm_span_kind(), _get_llm_input_messages(llm_input_messages), _get_llm_invocation_parameters(invocation_parameters), + _get_llm_tools(invocation_parameters), _get_inputs(arguments), ) ), @@ -298,6 +303,12 @@ def _get_outputs(response: "BaseModel") -> Iterator[Tuple[str, Any]]: yield OUTPUT_MIME_TYPE, JSON +def _get_llm_tools(invocation_parameters: Mapping[str, Any]) -> Iterator[Tuple[str, Any]]: + if isinstance(tools := invocation_parameters.get("tools"), list): + for tool_index, tool_schema in enumerate(tools): + yield f"{LLM_TOOLS}.{tool_index}.{TOOL_JSON_SCHEMA}", safe_json_dumps(tool_schema) + + def _get_llm_span_kind() -> Iterator[Tuple[str, Any]]: yield OPENINFERENCE_SPAN_KIND, LLM @@ -464,6 +475,7 @@ def _validate_invocation_parameter(parameter: Any) -> bool: LLM_TOKEN_COUNT_COMPLETION = SpanAttributes.LLM_TOKEN_COUNT_COMPLETION LLM_TOKEN_COUNT_PROMPT = SpanAttributes.LLM_TOKEN_COUNT_PROMPT LLM_TOKEN_COUNT_TOTAL = SpanAttributes.LLM_TOKEN_COUNT_TOTAL +LLM_TOOLS = SpanAttributes.LLM_TOOLS MESSAGE_CONTENT = MessageAttributes.MESSAGE_CONTENT MESSAGE_CONTENTS = MessageAttributes.MESSAGE_CONTENTS MESSAGE_FUNCTION_CALL_ARGUMENTS_JSON = MessageAttributes.MESSAGE_FUNCTION_CALL_ARGUMENTS_JSON @@ -479,6 +491,7 @@ def _validate_invocation_parameter(parameter: Any) -> bool: TAG_TAGS = SpanAttributes.TAG_TAGS TOOL_CALL_FUNCTION_ARGUMENTS_JSON = ToolCallAttributes.TOOL_CALL_FUNCTION_ARGUMENTS_JSON TOOL_CALL_FUNCTION_NAME = ToolCallAttributes.TOOL_CALL_FUNCTION_NAME +TOOL_JSON_SCHEMA = ToolAttributes.TOOL_JSON_SCHEMA USER_ID = SpanAttributes.USER_ID LLM_PROVIDER = SpanAttributes.LLM_PROVIDER LLM_SYSTEM = SpanAttributes.LLM_SYSTEM

[feature request] capture tool_schema in anthropic requests see example: ```python from typing import Any, Dict from your_client_library import Client # Replace with actual import tool_schema = { "name": "record_travel_request_attributes", "description": "Records the attributes of a travel request", "input_schema": { "type": "object", "properties": { "location": { "type": "string", "description": 'The desired destination location. Use city, state, and country format when possible. If no destination is provided, return "not_stated".', }, "budget_level": { "type": "string", "enum": ["low", "medium", "high", "not_stated"], "description": 'The desired budget level. If no budget level is provided, return "not_stated".', }, "purpose": { "type": "string", "enum": ["business", "pleasure", "other", "not_stated"], "description": 'The purpose of the trip. If no purpose is provided, return "not_stated".', }, }, "required": ["location", "budget_level", "purpose"], }, } system_message = ( "You are an assistant that parses and records the attributes of a user's travel request." ) def extract_raw_travel_request_attributes_string( travel_request: str, client: Client, model: str = "claude-3-5-sonnet-20240620", ) -> str: response = client.messages.create( model=model, max_tokens=1024, messages=[ {"role": "user", "content": travel_request}, ], system=system_message, tools=[tool_schema], tool_choice={"type": "tool", "name": "record_travel_request_attributes"}, ) # Process the response to extract the relevant information # This is a placeholder - you'll need to implement the actual extraction logic extracted_attributes = process_response(response) return extracted_attributes def process_response(response: Any) -> str: # Implement the logic to extract the attributes from the response # This is a placeholder function pass ```

Related to playground

2024-10-30T06:15:55

0.0

[]

Arize-ai/openinference

Arize-ai__openinference-1048

ff43e9ddc0a5f683f80d09139247ad194d6c29af

diff --git a/js/.changeset/tame-poems-argue.md b/js/.changeset/tame-poems-argue.md new file mode 100644 index 000000000..47b95c3fb --- /dev/null +++ b/js/.changeset/tame-poems-argue.md @@ -0,0 +1,5 @@ +--- +"@arizeai/openinference-core": minor +--- + +add OITracer and trace config to allow for masking of sensitive information on spans diff --git a/js/packages/openinference-core/jest.config.js b/js/packages/openinference-core/jest.config.js index 2fd4f78cf..de51c20f7 100644 --- a/js/packages/openinference-core/jest.config.js +++ b/js/packages/openinference-core/jest.config.js @@ -3,4 +3,7 @@ module.exports = { preset: "ts-jest", testEnvironment: "node", prettierPath: null, + moduleNameMapper: { + "^@core/(.*)$": "<rootDir>/src/$1", + }, }; diff --git a/js/packages/openinference-core/package.json b/js/packages/openinference-core/package.json index b9caf8c8d..b4a1eac98 100644 --- a/js/packages/openinference-core/package.json +++ b/js/packages/openinference-core/package.json @@ -34,10 +34,8 @@ }, "dependencies": { "@arizeai/openinference-semantic-conventions": "workspace:*", - "@opentelemetry/core": "^1.25.1" - }, - "peerDependencies": { - "@opentelemetry/api": ">=1.0.0 <1.9.0" + "@opentelemetry/core": "^1.25.1", + "@opentelemetry/api": "^1.9.0" }, "devDependencies": { "@opentelemetry/context-async-hooks": "^1.25.1", diff --git a/js/packages/openinference-core/src/trace/index.ts b/js/packages/openinference-core/src/trace/index.ts index 671871650..3c8e70b9e 100644 --- a/js/packages/openinference-core/src/trace/index.ts +++ b/js/packages/openinference-core/src/trace/index.ts @@ -1,2 +1,3 @@ export * from "./contextAttributes"; export * from "./types"; +export * from "./trace-config"; diff --git a/js/packages/openinference-core/src/trace/trace-config/OISpan.ts b/js/packages/openinference-core/src/trace/trace-config/OISpan.ts new file mode 100644 index 000000000..b1acf6d8a --- /dev/null +++ b/js/packages/openinference-core/src/trace/trace-config/OISpan.ts @@ -0,0 +1,90 @@ +import { + Span, + SpanContext, + SpanStatus, + TimeInput, + Attributes, + Exception, + AttributeValue, + Link, +} from "@opentelemetry/api"; +import { TraceConfig } from "./types"; +import { mask } from "./maskingRules"; + +/** + * A wrapper around the OpenTelemetry {@link Span} interface that masks sensitive information based on the passed in {@link TraceConfig}. + */ +export class OISpan implements Span { + private readonly span: Span; + private readonly config: TraceConfig; + + constructor({ span, config }: { span: Span; config: TraceConfig }) { + this.span = span; + this.config = config; + } + + setAttribute(key: string, value: AttributeValue): this { + const maskedValue = mask({ config: this.config, key, value }); + if (maskedValue != null) { + this.span.setAttribute(key, maskedValue); + } + return this; + } + + setAttributes(attributes: Attributes): this { + const maskedAttributes = Object.entries(attributes).reduce( + (maskedAttributes, [key, value]) => { + maskedAttributes[key] = mask({ config: this.config, key, value }); + return maskedAttributes; + }, + {} as Attributes, + ); + this.span.setAttributes(maskedAttributes); + return this; + } + + addEvent( + name: string, + attributesOrStartTime?: Attributes | TimeInput, + startTime?: TimeInput, + ): this { + this.span.addEvent(name, attributesOrStartTime, startTime); + return this; + } + + setStatus(status: SpanStatus): this { + this.span.setStatus(status); + return this; + } + + updateName(name: string): this { + this.span.updateName(name); + return this; + } + + end(endTime?: TimeInput): void { + this.span.end(endTime); + } + + isRecording(): boolean { + return this.span.isRecording(); + } + + recordException(exception: Exception, time?: TimeInput): void { + this.span.recordException(exception, time); + } + + spanContext(): SpanContext { + return this.span.spanContext(); + } + + addLink(link: Link): this { + this.span.addLink(link); + return this; + } + + addLinks(links: Link[]): this { + this.span.addLinks(links); + return this; + } +} diff --git a/js/packages/openinference-core/src/trace/trace-config/OITracer.ts b/js/packages/openinference-core/src/trace/trace-config/OITracer.ts new file mode 100644 index 000000000..d026df606 --- /dev/null +++ b/js/packages/openinference-core/src/trace/trace-config/OITracer.ts @@ -0,0 +1,130 @@ +import { + context as apiContext, + Context, + Span, + SpanOptions, + Tracer, +} from "@opentelemetry/api"; +import { + OpenInferenceActiveSpanCallback, + TraceConfig, + TraceConfigOptions, +} from "./types"; +import { OISpan } from "./OISpan"; +import { generateTraceConfig } from "./traceConfig"; +import { getAttributesFromContext } from "../contextAttributes"; + +/** + * Formats the params for the startActiveSpan method + * The method has multiple overloads, so we need to format the arguments + * Taken from @see https://github.com/open-telemetry/opentelemetry-js/blob/main/packages/opentelemetry-sdk-trace-base/src/Tracer.ts#L220C3-L235C6 + * + */ +function formatStartActiveSpanParams<F extends OpenInferenceActiveSpanCallback>( + arg2?: SpanOptions | F, + arg3?: Context | F, + arg4?: F, +) { + let opts: SpanOptions | undefined; + let ctx: Context | undefined; + let fn: F; + + if (typeof arg2 === "function") { + fn = arg2; + } else if (typeof arg3 === "function") { + opts = arg2; + fn = arg3; + } else { + opts = arg2; + ctx = arg3; + fn = arg4 as F; + } + + opts = opts ?? {}; + ctx = ctx ?? apiContext.active(); + + return { opts, ctx, fn }; +} + +/** + * A wrapper around the OpenTelemetry {@link Tracer} interface that masks sensitive information based on the passed in {@link TraceConfig}. + */ +export class OITracer implements Tracer { + private readonly tracer: Tracer; + private readonly config: TraceConfig; + /** + * + * @param tracer The OpenTelemetry {@link Tracer} to wrap + * @param traceConfig The {@link TraceConfigOptions} to set to control the behavior of the tracer + */ + constructor({ + tracer, + traceConfig, + }: { + tracer: Tracer; + traceConfig?: TraceConfigOptions; + }) { + this.tracer = tracer; + this.config = generateTraceConfig(traceConfig); + } + startActiveSpan<F extends (span: OISpan) => unknown>( + name: string, + fn: F, + ): ReturnType<F>; + startActiveSpan<F extends (span: OISpan) => unknown>( + name: string, + options: SpanOptions, + fn: F, + ): ReturnType<F>; + startActiveSpan<F extends (span: OISpan) => unknown>( + name: string, + options: SpanOptions, + context: Context, + fn: F, + ): ReturnType<F>; + startActiveSpan<F extends (span: OISpan) => ReturnType<F>>( + name: string, + arg2?: F | SpanOptions, + arg3?: F | Context, + arg4?: F, + ): ReturnType<F> | undefined { + const formattedArgs = formatStartActiveSpanParams(arg2, arg3, arg4); + if (formattedArgs == null) { + return; + } + const { opts, ctx, fn } = formattedArgs; + const { attributes } = opts ?? {}; + const contextAttributes = getAttributesFromContext(ctx); + const mergedAttributes = { ...contextAttributes, ...attributes }; + return this.tracer.startActiveSpan( + name, + { ...opts, attributes: undefined }, + ctx, + (span: Span) => { + const openInferenceSpan = new OISpan({ + span, + config: this.config, + }); + openInferenceSpan.setAttributes(mergedAttributes); + return fn(openInferenceSpan); + }, + ); + } + + startSpan(name: string, options?: SpanOptions, context?: Context): OISpan { + const attributes = options?.attributes; + const ctx = context ?? apiContext.active(); + const contextAttributes = getAttributesFromContext(ctx); + const mergedAttributes = { ...contextAttributes, ...attributes }; + const span = new OISpan({ + span: this.tracer.startSpan( + name, + { ...options, attributes: undefined }, + ctx, + ), + config: this.config, + }); + span.setAttributes(mergedAttributes); + return span; + } +} diff --git a/js/packages/openinference-core/src/trace/trace-config/constants.ts b/js/packages/openinference-core/src/trace/trace-config/constants.ts new file mode 100644 index 000000000..a9b878e7c --- /dev/null +++ b/js/packages/openinference-core/src/trace/trace-config/constants.ts @@ -0,0 +1,107 @@ +import { TraceConfigKey, TraceConfig, TraceConfigFlag } from "./types"; + +/** Hides input value & messages */ +export const OPENINFERENCE_HIDE_INPUTS = "OPENINFERENCE_HIDE_INPUTS"; +/** Hides output value & messages */ +export const OPENINFERENCE_HIDE_OUTPUTS = "OPENINFERENCE_HIDE_OUTPUTS"; +/** Hides all input messages */ +export const OPENINFERENCE_HIDE_INPUT_MESSAGES = + "OPENINFERENCE_HIDE_INPUT_MESSAGES"; +/** Hides all output messages */ +export const OPENINFERENCE_HIDE_OUTPUT_MESSAGES = + "OPENINFERENCE_HIDE_OUTPUT_MESSAGES"; +/** Hides images from input messages */ +export const OPENINFERENCE_HIDE_INPUT_IMAGES = + "OPENINFERENCE_HIDE_INPUT_IMAGES"; +/** Hides text from input messages */ +export const OPENINFERENCE_HIDE_INPUT_TEXT = "OPENINFERENCE_HIDE_INPUT_TEXT"; +/** Hides text from output messages */ +export const OPENINFERENCE_HIDE_OUTPUT_TEXT = "OPENINFERENCE_HIDE_OUTPUT_TEXT"; +/** Hides embedding vectors */ +export const OPENINFERENCE_HIDE_EMBEDDING_VECTORS = + "OPENINFERENCE_HIDE_EMBEDDING_VECTORS"; +/** Limits characters of a base64 encoding of an image */ +export const OPENINFERENCE_BASE64_IMAGE_MAX_LENGTH = + "OPENINFERENCE_BASE64_IMAGE_MAX_LENGTH"; + +export const DEFAULT_HIDE_INPUTS = false; +export const DEFAULT_HIDE_OUTPUTS = false; + +export const DEFAULT_HIDE_INPUT_MESSAGES = false; +export const DEFAULT_HIDE_OUTPUT_MESSAGES = false; + +export const DEFAULT_HIDE_INPUT_IMAGES = false; +export const DEFAULT_HIDE_INPUT_TEXT = false; +export const DEFAULT_HIDE_OUTPUT_TEXT = false; + +export const DEFAULT_HIDE_EMBEDDING_VECTORS = false; +export const DEFAULT_BASE64_IMAGE_MAX_LENGTH = 32000; + +/** When a value is hidden, it will be replaced by this redacted value */ +export const REDACTED_VALUE = "__REDACTED__"; + +/** + * The default, environment, and type information for each value on the TraceConfig + * Used to generate a full TraceConfig object with the correct types and default values + */ +export const traceConfigMetadata: Readonly< + Record<TraceConfigKey, TraceConfigFlag> +> = { + hideInputs: { + default: DEFAULT_HIDE_INPUTS, + envKey: OPENINFERENCE_HIDE_INPUTS, + type: "boolean", + }, + hideOutputs: { + default: DEFAULT_HIDE_OUTPUTS, + envKey: OPENINFERENCE_HIDE_OUTPUTS, + type: "boolean", + }, + hideInputMessages: { + default: DEFAULT_HIDE_INPUT_MESSAGES, + envKey: OPENINFERENCE_HIDE_INPUT_MESSAGES, + type: "boolean", + }, + hideOutputMessages: { + default: DEFAULT_HIDE_OUTPUT_MESSAGES, + envKey: OPENINFERENCE_HIDE_OUTPUT_MESSAGES, + type: "boolean", + }, + hideInputImages: { + default: DEFAULT_HIDE_INPUT_IMAGES, + envKey: OPENINFERENCE_HIDE_INPUT_IMAGES, + type: "boolean", + }, + hideInputText: { + default: DEFAULT_HIDE_INPUT_TEXT, + envKey: OPENINFERENCE_HIDE_INPUT_TEXT, + type: "boolean", + }, + hideOutputText: { + default: DEFAULT_HIDE_OUTPUT_TEXT, + envKey: OPENINFERENCE_HIDE_OUTPUT_TEXT, + type: "boolean", + }, + hideEmbeddingVectors: { + default: DEFAULT_HIDE_EMBEDDING_VECTORS, + envKey: OPENINFERENCE_HIDE_EMBEDDING_VECTORS, + type: "boolean", + }, + base64ImageMaxLength: { + default: DEFAULT_BASE64_IMAGE_MAX_LENGTH, + envKey: OPENINFERENCE_BASE64_IMAGE_MAX_LENGTH, + type: "number", + }, +}; + +export const DefaultTraceConfig: TraceConfig = { + hideInputs: DEFAULT_HIDE_INPUTS, + hideOutputs: DEFAULT_HIDE_OUTPUTS, + hideInputMessages: DEFAULT_HIDE_INPUT_MESSAGES, + hideOutputMessages: DEFAULT_HIDE_OUTPUT_MESSAGES, + hideInputImages: DEFAULT_HIDE_INPUT_IMAGES, + hideInputText: DEFAULT_HIDE_INPUT_TEXT, + hideOutputText: DEFAULT_HIDE_OUTPUT_TEXT, + hideEmbeddingVectors: DEFAULT_HIDE_EMBEDDING_VECTORS, + base64ImageMaxLength: DEFAULT_BASE64_IMAGE_MAX_LENGTH, +}; diff --git a/js/packages/openinference-core/src/trace/trace-config/index.ts b/js/packages/openinference-core/src/trace/trace-config/index.ts new file mode 100644 index 000000000..f325bc598 --- /dev/null +++ b/js/packages/openinference-core/src/trace/trace-config/index.ts @@ -0,0 +1,24 @@ +export { TraceConfigOptions } from "./types"; +export { + OPENINFERENCE_BASE64_IMAGE_MAX_LENGTH, + OPENINFERENCE_HIDE_EMBEDDING_VECTORS, + DEFAULT_BASE64_IMAGE_MAX_LENGTH, + DEFAULT_HIDE_EMBEDDING_VECTORS, + DEFAULT_HIDE_INPUTS, + DEFAULT_HIDE_INPUT_IMAGES, + DEFAULT_HIDE_INPUT_MESSAGES, + DEFAULT_HIDE_INPUT_TEXT, + DEFAULT_HIDE_OUTPUTS, + DEFAULT_HIDE_OUTPUT_MESSAGES, + DEFAULT_HIDE_OUTPUT_TEXT, + OPENINFERENCE_HIDE_INPUTS, + OPENINFERENCE_HIDE_INPUT_IMAGES, + OPENINFERENCE_HIDE_INPUT_MESSAGES, + OPENINFERENCE_HIDE_INPUT_TEXT, + OPENINFERENCE_HIDE_OUTPUTS, + OPENINFERENCE_HIDE_OUTPUT_MESSAGES, + OPENINFERENCE_HIDE_OUTPUT_TEXT, + REDACTED_VALUE, +} from "./constants"; +export { OITracer } from "./OITracer"; +export { type OISpan } from "./OISpan"; diff --git a/js/packages/openinference-core/src/trace/trace-config/maskingRules.ts b/js/packages/openinference-core/src/trace/trace-config/maskingRules.ts new file mode 100644 index 000000000..c4cbed0f9 --- /dev/null +++ b/js/packages/openinference-core/src/trace/trace-config/maskingRules.ts @@ -0,0 +1,208 @@ +import { AttributeValue } from "@opentelemetry/api"; +import { REDACTED_VALUE } from "./constants"; +import { SemanticConventions } from "@arizeai/openinference-semantic-conventions"; +import { MaskingRule, MaskingRuleArgs } from "./types"; + +/** + * Masks (redacts) input text in LLM input messages. + * Will mask information stored under the key `llm.input_messages.[i].message.content`. + * @example + * ```typescript + * maskInputTextRule.condition({ + * config: {hideInputText: true}, + * key: "llm.input_messages.[i].message.content" + * }) // returns true so the rule applies and the value will be redacted + */ +const maskInputTextRule: MaskingRule = { + condition: ({ config, key }) => + config.hideInputText && + key.includes(SemanticConventions.LLM_INPUT_MESSAGES) && + key.includes(SemanticConventions.MESSAGE_CONTENT) && + !key.includes(SemanticConventions.MESSAGE_CONTENTS), + action: () => REDACTED_VALUE, +}; + +/** + * Masks (redacts) output text in LLM output messages. + * Will mask information stored under the key `llm.output_messages.[i].message.content`. + * @example + * ```typescript + * maskOutputTextRule.condition({ + * config: {hideOutputText: true}, + * key: "llm.output_messages.[i].message.content" + * }) // returns true so the rule applies and the value will be redacted + * ``` + */ +const maskOutputTextRule: MaskingRule = { + condition: ({ config, key }) => + config.hideOutputText && + key.includes(SemanticConventions.LLM_OUTPUT_MESSAGES) && + key.includes(SemanticConventions.MESSAGE_CONTENT) && + !key.includes(SemanticConventions.MESSAGE_CONTENTS), + action: () => REDACTED_VALUE, +}; + +/** + * Masks (redacts) input text content in LLM input messages. + * Will mask information stored under the key `llm.input_messages.[i].message.contents.[j].message_content.text`. + * @example + * ```typescript + * maskOutputTextRule.condition({ + * config: {hideInputText: true}, + * key: "llm.input_messages.[i].message.contents.[j].message_content.text" + * }) // returns true so the rule applies and the value will be redacted + */ +const maskInputTextContentRule: MaskingRule = { + condition: ({ config, key }) => + config.hideInputText && + key.includes(SemanticConventions.LLM_INPUT_MESSAGES) && + key.includes(SemanticConventions.MESSAGE_CONTENT_TEXT), + action: () => REDACTED_VALUE, +}; + +/** + * Masks (redacts) output text content in LLM output messages. + * @example + * ```typescript + * maskOutputTextRule.condition({ + * config: {hideOutputText: true}, + * key: "llm.output_messages.[i].message.contents.[j].message_content.text" + * }) // returns true so the rule applies and the value will be redacted + */ +const maskOutputTextContentRule: MaskingRule = { + condition: ({ config, key }) => + config.hideOutputText && + key.includes(SemanticConventions.LLM_OUTPUT_MESSAGES) && + key.includes(SemanticConventions.MESSAGE_CONTENT_TEXT), + action: () => REDACTED_VALUE, +}; + +/** + * Masks (removes) input images in LLM input messages. + * @example + * ```typescript + * maskOutputTextRule.condition({ + * config: {hideInputImages: true}, + * key: "llm.input_messages.[i].message.contents.[j].message_content.image" + * }) // returns true so the rule applies and the value will be removed + */ +const maskInputImagesRule: MaskingRule = { + condition: ({ config, key }) => + config.hideInputImages && + key.includes(SemanticConventions.LLM_INPUT_MESSAGES) && + key.includes(SemanticConventions.MESSAGE_CONTENT_IMAGE), + action: () => undefined, +}; + +function isBase64Url(url?: AttributeValue): boolean { + return ( + typeof url === "string" && + url.startsWith("data:image/") && + url.includes("base64") + ); +} + +/** + * Masks (redacts) base64 images that are too long. + * * @example + * ```typescript + * maskOutputTextRule.condition({ + * config: {base64ImageMaxLength: 10}, + * key: "llm.input_messages.[i].message.contents.[j].message_content.image.url", + * value: "data:image/base64,verylongbase64string" + * }) // returns true so the rule applies and the value will be redacted + */ +const maskLongBase64ImageRule: MaskingRule = { + condition: ({ config, key, value }) => + typeof value === "string" && + isBase64Url(value) && + value.length > config.base64ImageMaxLength && + key.includes(SemanticConventions.LLM_INPUT_MESSAGES) && + key.includes(SemanticConventions.MESSAGE_CONTENT_IMAGE) && + key.endsWith(SemanticConventions.IMAGE_URL), + action: () => REDACTED_VALUE, +}; + +/** + * Masks (removes) embedding vectors. + * * @example + * ```typescript + * maskOutputTextRule.condition({ + * config: {hideEmbeddingVectors: true}, + * key: "embedding.embeddings.[i].embedding.vector" + * }) // returns true so the rule applies and the value will be redacted + */ +const maskEmbeddingVectorsRule: MaskingRule = { + condition: ({ config, key }) => + config.hideEmbeddingVectors && + key.includes(SemanticConventions.EMBEDDING_EMBEDDINGS) && + key.includes(SemanticConventions.EMBEDDING_VECTOR), + action: () => undefined, +}; + +/** + * A list of {@link MaskingRule}s that are applied to span attributes to either redact or remove sensitive information. + * The order of these rules is important as it can ensure appropriate masking of information + * Rules should go from more specific to more general so that things like `llm.input_messages.[i].message.content` are masked with {@link REDACTED_VALUE} before the more generic masking of `llm.input_messages` might happen with `undefined` might happen. + */ +const maskingRules: MaskingRule[] = [ + { + condition: ({ config, key }) => + config.hideInputs && key === SemanticConventions.INPUT_VALUE, + action: () => REDACTED_VALUE, + }, + { + condition: ({ config, key }) => + config.hideInputs && key === SemanticConventions.INPUT_MIME_TYPE, + action: () => undefined, + }, + { + condition: ({ config, key }) => + config.hideOutputs && key === SemanticConventions.OUTPUT_VALUE, + action: () => REDACTED_VALUE, + }, + { + condition: ({ config, key }) => + config.hideOutputs && key === SemanticConventions.OUTPUT_MIME_TYPE, + action: () => undefined, + }, + { + condition: ({ config, key }) => + (config.hideInputs || config.hideInputMessages) && + key.includes(SemanticConventions.LLM_INPUT_MESSAGES), + action: () => undefined, + }, + { + condition: ({ config, key }) => + (config.hideOutputs || config.hideOutputMessages) && + key.includes(SemanticConventions.LLM_OUTPUT_MESSAGES), + action: () => undefined, + }, + maskInputTextRule, + maskOutputTextRule, + maskInputTextContentRule, + maskOutputTextContentRule, + maskInputImagesRule, + maskLongBase64ImageRule, + maskEmbeddingVectorsRule, +]; + +/** + * A function that masks (redacts or removes) sensitive information from span attributes based on the trace config. + * @param config The {@link TraceConfig} to use to determine if the value should be masked + * @param key The key of the attribute to mask + * @param value The value of the attribute to mask + * @returns The redacted value or undefined if the value should be masked, otherwise the original value + */ +export function mask({ + config, + key, + value, +}: MaskingRuleArgs): AttributeValue | undefined { + for (const rule of maskingRules) { + if (rule.condition({ config, key, value })) { + return rule.action(); + } + } + return value; +} diff --git a/js/packages/openinference-core/src/trace/trace-config/traceConfig.ts b/js/packages/openinference-core/src/trace/trace-config/traceConfig.ts new file mode 100644 index 000000000..2eef793c2 --- /dev/null +++ b/js/packages/openinference-core/src/trace/trace-config/traceConfig.ts @@ -0,0 +1,68 @@ +import { DefaultTraceConfig, traceConfigMetadata } from "./constants"; +import { TraceConfigKey, TraceConfig, TraceConfigOptions } from "./types"; +import { assertUnreachable, withSafety } from "../../utils"; + +const safelyParseInt = withSafety({ fn: parseInt }); + +type TraceConfigOptionMetadata = (typeof traceConfigMetadata)[TraceConfigKey]; + +/** + * Parses an option based on its type + * The order of precedence is: optionValue > envValue > defaultValue + * @param key - The key of the option. + * @param optionMetadata - The {@link TraceConfigOptionMetadata} for the option which includes its type, default value, and environment variable key. + * + */ +function parseOption({ + optionValue, + optionMetadata, +}: { + optionValue?: number | boolean; + optionMetadata: TraceConfigOptionMetadata; +}) { + if (optionValue !== undefined) { + return optionValue; + } + const envValue = process.env[optionMetadata.envKey]; + if (envValue !== undefined) { + switch (optionMetadata.type) { + case "number": { + const maybeEnvNumber = safelyParseInt(envValue); + return maybeEnvNumber != null && !isNaN(maybeEnvNumber) + ? maybeEnvNumber + : optionMetadata.default; + } + case "boolean": + return envValue.toLowerCase() === "true"; + default: + assertUnreachable(optionMetadata); + } + } + + return optionMetadata.default; +} + +/** + * Generates a full trace config object based on passed in options, environment variables, and default values. + * The order of precedence is: optionValue > envValue > defaultValue + * @param options - The user provided TraceConfigOptions. + * @returns A full TraceConfig object with all options set to their final values. + */ +export function generateTraceConfig(options?: TraceConfigOptions): TraceConfig { + if (options == null) { + return DefaultTraceConfig; + } + return Object.entries(traceConfigMetadata).reduce( + (config, [key, optionMetadata]) => { + const TraceConfigKey = key as TraceConfigKey; + return { + ...config, + [TraceConfigKey]: parseOption({ + optionValue: options[TraceConfigKey], + optionMetadata, + }), + }; + }, + {} as TraceConfig, + ); +} diff --git a/js/packages/openinference-core/src/trace/trace-config/types.ts b/js/packages/openinference-core/src/trace/trace-config/types.ts new file mode 100644 index 000000000..cb18e8439 --- /dev/null +++ b/js/packages/openinference-core/src/trace/trace-config/types.ts @@ -0,0 +1,120 @@ +import { AttributeValue } from "@opentelemetry/api"; +import { OISpan } from "./OISpan"; + +/** + * Tracing configuration options that can be set to hide or redact sensitive information from traces + */ +export type TraceConfigOptions = { + hideInputs?: boolean; + hideOutputs?: boolean; + hideInputMessages?: boolean; + hideOutputMessages?: boolean; + hideInputImages?: boolean; + hideInputText?: boolean; + hideOutputText?: boolean; + hideEmbeddingVectors?: boolean; + base64ImageMaxLength?: number; +}; + +/** + * A full tracing configuration object that includes all possible options with their final values + */ +export type TraceConfig = Readonly<Required<TraceConfigOptions>>; + +export type TraceConfigKey = keyof TraceConfig; + +type BooleanTraceConfigFlag = { + /** + * The default value for the flag + */ + default: boolean; + /** + * The environment variable key to use to override the default value + */ + envKey: string; + /** + * The type of the flag + */ + type: "boolean"; +}; + +type NumericTraceConfigFlag = { + /** + * The default value for the flag + */ + default: number; + /** + * The environment variable key to use to override the default value + */ + envKey: string; + /** + * The type of the flag + */ + type: "number"; +}; + +/** + * The default and environment information for a value on the TraceConfig + * Used to generate a full TraceConfig object + * + * @example + * ```typescript + * { + * hideInputs: { + * default: DEFAULT_HIDE_INPUTS, // false + * envKey: OPENINFERENCE_HIDE_INPUTS, // "OPENINFERENCE_HIDE_INPUTS" + * type: "boolean", + * } + * } + * ``` + */ +export type TraceConfigFlag = BooleanTraceConfigFlag | NumericTraceConfigFlag; + +/** + * The arguments for a masking rule, used to determine if a value should be masked (redacted or omitted) + */ +export type MaskingRuleArgs = { + /** + * The trace config to use to determine if the value should be masked + */ + config: TraceConfig; + /** + * The key of the attribute to mask + */ + key: string; + /** + * The value of the attribute to mask + */ + value?: AttributeValue; +}; + +/** + * A condition to determine if a value should be masked + * and an action that masks (removes or redacts) the value if the condition is met + * @example + * ```typescript + * const config = {hideInputText: true} + * const key = "llm.input_messages.0.message.content" + * if (maskInputTextRule.condition({ config, key })) { + * return maskInputTextRule.action() + * } + * ``` + */ +export type MaskingRule = { + /** + * + * @param args The {@linkcode MaskingRuleArgs} to determine if the value should be masked + * @returns true if the value should be masked, false otherwise + */ + condition: (args: MaskingRuleArgs) => boolean; + /** + * An action to be applied if the condition is met + * @returns A redacted value or undefined + */ + action: () => AttributeValue | undefined; +}; + +/** + * A callback that is called when a new active {@link OISpan} is created + */ +export type OpenInferenceActiveSpanCallback = (span: OISpan) => void; diff --git a/js/packages/openinference-core/src/utils/typeUtils.ts b/js/packages/openinference-core/src/utils/typeUtils.ts index 6d65f68dc..fae0fa660 100644 --- a/js/packages/openinference-core/src/utils/typeUtils.ts +++ b/js/packages/openinference-core/src/utils/typeUtils.ts @@ -48,3 +48,25 @@ export function isAttributes(value: unknown): value is Attributes { ) ); } + +/** + * A type check function to ensure that a switch or set of conditionals is exhaustive. + * Typscript will throw an error if the switch or conditionals are not exhaustive. + * @example + * ```typescript + * type MyType = "a" | "b"; + * function myFunction(value: MyType) { + * switch (value) { + * case "a": + * return "A"; + * case "b": + * return "B"; + * default: + * assertUnreachable(value); + * } + * } + * ``` + */ +export function assertUnreachable(value: never): never { + throw new Error(`Unreachable code reached with value: ${value}`); +} diff --git a/js/packages/openinference-core/tsconfig.esm.json b/js/packages/openinference-core/tsconfig.esm.json index 0b9947f92..737b185af 100644 --- a/js/packages/openinference-core/tsconfig.esm.json +++ b/js/packages/openinference-core/tsconfig.esm.json @@ -3,7 +3,8 @@ "compilerOptions": { "outDir": "dist/esm", "rootDir": "src", - "tsBuildInfoFile": "dist/esm/tsconfig.esm.tsbuildinfo" + "tsBuildInfoFile": "dist/esm/tsconfig.esm.tsbuildinfo", + "baseUrl": "./src" }, "include": ["src/**/*.ts"], "references": [] diff --git a/js/packages/openinference-core/tsconfig.esnext.json b/js/packages/openinference-core/tsconfig.esnext.json index 84a00950b..7f1cd3f1f 100644 --- a/js/packages/openinference-core/tsconfig.esnext.json +++ b/js/packages/openinference-core/tsconfig.esnext.json @@ -3,7 +3,8 @@ "compilerOptions": { "outDir": "dist/esnext", "rootDir": "src", - "tsBuildInfoFile": "dist/esnext/tsconfig.esnext.tsbuildinfo" + "tsBuildInfoFile": "dist/esnext/tsconfig.esnext.tsbuildinfo", + "baseUrl": "./src" }, "include": ["src/**/*.ts"], "references": [] diff --git a/js/packages/openinference-core/tsconfig.json b/js/packages/openinference-core/tsconfig.json index 64960dc0f..9099ee154 100644 --- a/js/packages/openinference-core/tsconfig.json +++ b/js/packages/openinference-core/tsconfig.json @@ -3,7 +3,8 @@ "compilerOptions": { "outDir": "dist", "rootDir": ".", - "types": ["node", "jest"] + "types": ["node", "jest"], + "baseUrl": "./src" }, "files": [], "include": ["src/**/*.ts", "test/**/*.ts"], diff --git a/js/packages/openinference-instrumentation-openai/package.json b/js/packages/openinference-instrumentation-openai/package.json index 9d5fe3d10..e1b16ec34 100644 --- a/js/packages/openinference-instrumentation-openai/package.json +++ b/js/packages/openinference-instrumentation-openai/package.json @@ -38,6 +38,7 @@ "@opentelemetry/sdk-trace-base": "^1.25.1", "@opentelemetry/sdk-trace-node": "^1.25.1", "@opentelemetry/semantic-conventions": "^1.25.1", + "jest": "^29.7.0", "openai": "^4.52.7" } } diff --git a/js/pnpm-lock.yaml b/js/pnpm-lock.yaml index 4c78f7a96..435a8855c 100644 --- a/js/pnpm-lock.yaml +++ b/js/pnpm-lock.yaml @@ -48,24 +48,24 @@ importers: specifier: workspace:* version: link:../openinference-semantic-conventions '@opentelemetry/api': - specifier: '>=1.0.0 <1.9.0' - version: 1.8.0 + specifier: ^1.9.0 + version: 1.9.0 '@opentelemetry/core': specifier: ^1.25.1 - version: 1.25.1(@opentelemetry/[email protected]) + version: 1.25.1(@opentelemetry/[email protected]) devDependencies: '@opentelemetry/context-async-hooks': specifier: ^1.25.1 - version: 1.25.1(@opentelemetry/[email protected]) + version: 1.25.1(@opentelemetry/[email protected]) '@opentelemetry/resources': specifier: ^1.19.0 - version: 1.25.1(@opentelemetry/[email protected]) + version: 1.25.1(@opentelemetry/[email protected]) '@opentelemetry/sdk-trace-base': specifier: ^1.19.0 - version: 1.25.1(@opentelemetry/[email protected]) + version: 1.25.1(@opentelemetry/[email protected]) '@opentelemetry/sdk-trace-node': specifier: ^1.19.0 - version: 1.25.1(@opentelemetry/[email protected]) + version: 1.25.1(@opentelemetry/[email protected]) '@opentelemetry/semantic-conventions': specifier: ^1.19.0 version: 1.26.0 @@ -191,6 +191,9 @@ importers: '@opentelemetry/semantic-conventions': specifier: ^1.25.1 version: 1.26.0 + jest: + specifier: ^29.7.0 + version: 29.7.0(@types/[email protected]) openai: specifier: ^4.52.7 version: 4.56.0([email protected]) @@ -3880,10 +3883,6 @@ snapshots: '@opentelemetry/[email protected]': {} - '@opentelemetry/[email protected](@opentelemetry/[email protected])': - dependencies: - '@opentelemetry/api': 1.8.0 - '@opentelemetry/[email protected](@opentelemetry/[email protected])': dependencies: '@opentelemetry/api': 1.9.0 @@ -3946,21 +3945,11 @@ snapshots: '@opentelemetry/sdk-metrics': 1.23.0(@opentelemetry/[email protected]) '@opentelemetry/sdk-trace-base': 1.23.0(@opentelemetry/[email protected]) - '@opentelemetry/[email protected](@opentelemetry/[email protected])': - dependencies: - '@opentelemetry/api': 1.8.0 - '@opentelemetry/core': 1.25.1(@opentelemetry/[email protected]) - '@opentelemetry/[email protected](@opentelemetry/[email protected])': dependencies: '@opentelemetry/api': 1.9.0 '@opentelemetry/core': 1.25.1(@opentelemetry/[email protected]) - '@opentelemetry/[email protected](@opentelemetry/[email protected])': - dependencies: - '@opentelemetry/api': 1.8.0 - '@opentelemetry/core': 1.25.1(@opentelemetry/[email protected]) - '@opentelemetry/[email protected](@opentelemetry/[email protected])': dependencies: '@opentelemetry/api': 1.9.0 @@ -4019,16 +4008,6 @@ snapshots: '@opentelemetry/resources': 1.25.1(@opentelemetry/[email protected]) '@opentelemetry/semantic-conventions': 1.25.1 - '@opentelemetry/[email protected](@opentelemetry/[email protected])': - dependencies: - '@opentelemetry/api': 1.8.0 - '@opentelemetry/context-async-hooks': 1.25.1(@opentelemetry/[email protected]) - '@opentelemetry/core': 1.25.1(@opentelemetry/[email protected]) - '@opentelemetry/propagator-b3': 1.25.1(@opentelemetry/[email protected]) - '@opentelemetry/propagator-jaeger': 1.25.1(@opentelemetry/[email protected]) - '@opentelemetry/sdk-trace-base': 1.25.1(@opentelemetry/[email protected]) - semver: 7.6.3 - '@opentelemetry/[email protected](@opentelemetry/[email protected])': dependencies: '@opentelemetry/api': 1.9.0

[instrumentation-js] TraceConfig for openinference-core Mimic the behavior of the trace config for python - e.g. in terms of scope and coherence with the environment variables

2024-09-19T00:01:49

0.0

[]

Arize-ai/openinference

Arize-ai__openinference-1045

50cd7d5f58ab0a7d424f8c9950a22eb912330663

diff --git a/python/instrumentation/openinference-instrumentation-langchain/pyproject.toml b/python/instrumentation/openinference-instrumentation-langchain/pyproject.toml index 3ebed9e39..77c603020 100644 --- a/python/instrumentation/openinference-instrumentation-langchain/pyproject.toml +++ b/python/instrumentation/openinference-instrumentation-langchain/pyproject.toml @@ -45,6 +45,10 @@ test = [ "opentelemetry-sdk", "respx", "pytest-recording", + "langsmith", + "portpicker", + "starlette", + "uvicorn", ] type-check = [ "langchain_core == 0.1.5", @@ -73,7 +77,6 @@ testpaths = [ [tool.mypy] strict = true -explicit_package_bases = true exclude = [ "examples", "dist", diff --git a/python/tox.ini b/python/tox.ini index 0aef11523..181ee95e3 100644 --- a/python/tox.ini +++ b/python/tox.ini @@ -10,7 +10,7 @@ envlist = py3{8,12}-ci-{vertexai,vertexai-latest} py3{8,12}-ci-{llama_index,llama_index-latest} py3{9,12}-ci-{dspy,dspy-latest} - py3{8,12}-ci-{langchain,langchain-latest} + py3{9,12}-ci-{langchain,langchain-latest} ; py3{9,12}-ci-{guardrails,guardrails-latest} ; py3{10,12}-ci-{crewai,crewai-latest} py3{8,12}-ci-{haystack,haystack-latest} @@ -79,8 +79,8 @@ commands = ruff: ruff format {posargs:.} ruff: ruff check --fix {posargs:.} mypy: mypy {posargs:.} - test: pytest -n auto --nf {posargs:.} - ci: ruff format --diff {posargs:.} - ci: ruff check --no-fix {posargs:.} - ci: mypy {posargs:.} + test: pytest -rfEs -l --nf {posargs:.} + ci: ruff format --diff . + ci: ruff check --no-fix . + ci: mypy . ci: pytest -n auto -x -ra {posargs:.}

[bug] langchain failing on latest 0.3.0 LangChain CI is failing on latest. This could be due to 0.3.0 moving to pydantic 2. Needs investigation.

2024-09-18T19:31:07

0.0

[]

Arize-ai/openinference

Arize-ai__openinference-1010

5bc4100c1a9683cff18e90538c857e5844159d31

diff --git a/python/openinference-instrumentation/src/openinference/instrumentation/config.py b/python/openinference-instrumentation/src/openinference/instrumentation/config.py index dd20abefc..1d441f6f8 100644 --- a/python/openinference-instrumentation/src/openinference/instrumentation/config.py +++ b/python/openinference-instrumentation/src/openinference/instrumentation/config.py @@ -1,6 +1,7 @@ import os from contextlib import contextmanager from dataclasses import dataclass, field, fields +from secrets import randbits from typing import ( Any, Callable, @@ -21,7 +22,16 @@ detach, set_value, ) -from opentelemetry.trace import Link, Span, SpanKind, Tracer, use_span +from opentelemetry.sdk.trace import IdGenerator +from opentelemetry.trace import ( + INVALID_SPAN_ID, + INVALID_TRACE_ID, + Link, + Span, + SpanKind, + Tracer, + use_span, +) from opentelemetry.util.types import Attributes, AttributeValue from openinference.semconv.trace import ( @@ -343,10 +353,32 @@ def end(self, end_time: Optional[int] = None) -> None: span.end(end_time) +class _IdGenerator(IdGenerator): + """ + An IdGenerator that uses a different source of randomness to + avoid being affected by seeds set by user application. + """ + + def generate_span_id(self) -> int: + while (span_id := randbits(64)) == INVALID_SPAN_ID: + continue + return span_id + + def generate_trace_id(self) -> int: + while (trace_id := randbits(128)) == INVALID_TRACE_ID: + continue + return trace_id + + class OITracer(wrapt.ObjectProxy): # type: ignore[misc] def __init__(self, wrapped: Tracer, config: TraceConfig) -> None: super().__init__(wrapped) self._self_config = config + self._self_id_generator = _IdGenerator() + + @property + def id_generator(self) -> IdGenerator: + return self._self_id_generator @contextmanager def start_as_current_span( @@ -391,7 +423,8 @@ def start_span( set_status_on_exception: bool = True, ) -> Span: tracer = cast(Tracer, self.__wrapped__) - span = tracer.start_span( + span = tracer.__class__.start_span( + self, name=name, context=context, kind=kind,

[bug] should use a different source of randomness, e.g. secrets, so span identifiers are not affected by seed from application [id_generator](https://github.com/open-telemetry/opentelemetry-python/blob/9697f805d1e5ba53cc55e2502ad1de0cc7c5900a/opentelemetry-sdk/src/opentelemetry/sdk/trace/id_generator.py) is affected by application seeds <img width="668" alt="Screenshot 2024-09-10 at 7 51 52 AM" src="https://github.com/user-attachments/assets/bc8c99ba-1d9c-411d-9e94-d0db2714b121">

2024-09-10T15:50:43

0.0

[]

cvxgrp/pymde

cvxgrp__pymde-55

2a271c70ec36fa86efd539ed432c87997b75078b

diff --git a/pymde/quadratic.py b/pymde/quadratic.py index 73a6eac..82e3a60 100644 --- a/pymde/quadratic.py +++ b/pymde/quadratic.py @@ -44,10 +44,27 @@ def pca(Y, embedding_dim): return np.sqrt(float(n)) * U[:, :embedding_dim] -def _laplacian(n, m, edges, weights): - A = util.adjacency_matrix(n, m, edges, weights) +def _laplacian(n, m, edges, weights, use_scipy=True): + A = util.adjacency_matrix(n, m, edges, weights, use_scipy=use_scipy) L = -A - L.setdiag((np.array(A.sum(axis=1)).squeeze())) + if use_scipy: + L.setdiag(np.asarray(A.sum(axis=1)).squeeze()) + else: + # diag is currently 0 + diag_vals = torch.sparse.sum(A, dim=1).to_dense() + diag_inds = torch.tensor( + [[i, i] for i in range(len(A))], device=L.device, dtype=L.dtype + ) + diag_inds = diag_inds.transpose(0, 1) + diag_matrix = torch.sparse_coo_tensor( + diag_inds, + diag_vals, + size=(n, n), + dtype=torch.float32, + device=L.device, + ) + L = L + diag_matrix + L = L.coalesce() return L @@ -62,8 +79,8 @@ def _spectral( device=None, ): n = L.shape[0] + k = m + 1 if not cg: - k = m + 1 num_lanczos_vectors = max(2 * k + 1, int(np.sqrt(L.shape[0]))) eigenvalues, eigenvectors = scipy.sparse.linalg.eigsh( L, @@ -74,33 +91,34 @@ def _spectral( v0=np.ones(L.shape[0]), maxiter=L.shape[0] * 5, ) - order = np.argsort(eigenvalues)[1:k] else: - k = m if warm_start: mde = problem.MDE( - n, m, edges, f=penalties.Quadratic(weights), device=device + n, + k, + edges, + distortion_function=penalties.Quadratic(weights), + device=device, ) - X_init = mde.fit(max_iter=40, use_line_search=False) + X_init = mde.embed(max_iter=40) else: X_init = util.proj_standardized( - torch.tensor(util.np_rng().randn(n, m), device=device), + torch.randn((n, k), device=device), demean=True, ) - eigenvalues, eigenvectors = scipy.sparse.linalg.lobpcg( + eigenvalues, eigenvectors = torch.lobpcg( A=L, - X=X_init.cpu().numpy(), - # Y: search in the orthogonal complement of the ones vector - Y=np.ones((L.shape[0], 1)), + X=X_init, tol=None, # largest: find the smallest eigenvalues largest=False, - maxiter=max_iter, + niter=max_iter, ) - order = np.argsort(eigenvalues)[0:k] + order = np.argsort(eigenvalues)[1:k] return eigenvectors[:, order] +# TODO(akshayka): Deprecate device argument, infer from edges/weights def spectral( n_items, embedding_dim, edges, weights, cg=False, max_iter=40, device="cpu" ): @@ -116,8 +134,6 @@ def spectral( \\mbox{subject to} & (1/n) X^T X = I, \quad d_{ij} = |x_i - x_j|_2. \\end{array} - The weights may be negative. - By default, the problem is solved using a Lanczos method. If cg=True, LOBPCG is used; LOBPCG is warm-started by running a projected quasi-newton method for a small number of iterations. Use cg=True when the number @@ -134,7 +150,7 @@ def spectral( edges: torch.Tensor(shape=(n_edges, 2)) A list of edges (i, j), 0 <= i < j < n_items weights: torch.Tensor(shape=(n_edges,)) - A list of weights associated with each edge + A list of nonnegative weights associated with each edge cg: bool If True, uses a preconditioned CG method to find the embedding, which requires that the Laplacian matrix plus the identity is @@ -151,9 +167,13 @@ def spectral( torch.Tensor(shape=(n_items, embedding_dim)) A spectral embedding, projected onto the standardization constraint """ - L = _laplacian(n_items, embedding_dim, edges, weights) + # use torch sparse and linalg for lobpcg + use_scipy = not cg + L = _laplacian(n_items, embedding_dim, edges, weights, use_scipy=use_scipy) emb = _spectral(L, embedding_dim, cg=cg, device=device, max_iter=max_iter) - emb -= emb.mean(axis=0) - return util.proj_standardized( - torch.tensor(emb, dtype=weights.dtype, device=device) - ) + if use_scipy: + emb -= emb.mean(axis=0) + emb = torch.tensor(emb, dtype=weights.dtype, device=device) + else: + emb -= emb.mean(dim=0) + return util.proj_standardized(emb) diff --git a/pymde/recipes.py b/pymde/recipes.py index 925e9be..5bd7397 100644 --- a/pymde/recipes.py +++ b/pymde/recipes.py @@ -292,7 +292,11 @@ def preserve_neighbors( If not None, neighborhoods are restricted to have a radius no greater than ``max_distance``. init: str - Initialization strategy; 'quadratic' or 'random'. + Initialization strategy; 'quadratic' or 'random'. If the quadratic + initialization takes too much time, try a random initialization. + When `device=='cuda'` and `init=='quadratic'` a torch-implemented + version will be used, which is fully on device, but uses a CG-based + technique (versus Lanczos). device: str (optional) Device for the embedding (eg, 'cpu', 'cuda'). verbose: bool @@ -357,9 +361,11 @@ def preserve_neighbors( if init == "quadratic": if verbose: - problem.LOGGER.info("Computing quadratic initialization.") + problem.LOGGER.info(f"Computing {init} initialization.") + # use cg + torch when using GPU + cg = device == "cuda" X_init = quadratic.spectral( - n, embedding_dim, edges, weights, device=device + n, embedding_dim, edges, weights, device=device, cg=cg ) if not isinstance( constraint, (constraints._Centered, constraints._Standardized) diff --git a/pymde/util.py b/pymde/util.py index 770c1e4..0359e76 100644 --- a/pymde/util.py +++ b/pymde/util.py @@ -168,16 +168,30 @@ def proj_standardized(X, demean=False, inplace=False): return proj -def adjacency_matrix(n, m, edges, weights): - if isinstance(weights, torch.Tensor): - weights = weights.detach().cpu().numpy() - if isinstance(edges, torch.Tensor): - edges = edges.detach().cpu().numpy() - A = scipy.sparse.coo_matrix( - (weights, (edges[:, 0], edges[:, 1])), shape=(n, n), dtype=np.float32 - ) - A = A + A.T - return A.tocoo() +def adjacency_matrix(n, m, edges, weights, use_scipy=True): + + if use_scipy: + if isinstance(weights, torch.Tensor): + weights = weights.detach().cpu().numpy() + if isinstance(edges, torch.Tensor): + edges = edges.detach().cpu().numpy() + A = scipy.sparse.coo_matrix( + (weights, (edges[:, 0], edges[:, 1])), + shape=(n, n), + dtype=np.float32, + ) + A = A + A.T + return A.tocoo() + else: + A = torch.sparse_coo_tensor( + edges.transpose(0, 1), + weights, + size=(n, n), + dtype=torch.float32, + device=edges.device, + ) + A = A + A.transpose(0, 1) + return A @tensor_arguments @@ -317,7 +331,7 @@ def align(source, target): def scale_delta(delta, d_nat): # scale delta so RMS(delta) == d_nat N = delta.nelement() - rms = torch.sqrt(1 / N * torch.sum(delta ** 2)) + rms = torch.sqrt(1 / N * torch.sum(delta**2)) return delta * d_nat / rms diff --git a/pyproject.toml b/pyproject.toml index 7f09190..0a52494 100644 --- a/pyproject.toml +++ b/pyproject.toml @@ -6,5 +6,8 @@ requires = [ "scipy >= 1.6", "setuptools", "wheel", - "cython" + "cython", ] + +[tool.black] +line-length = 80

Quadratic initialization on GPU The quadratic initialization takes the most time for a dataset with 150k observations and 15 nearest neighbors. Is it possible to use the GPU for this? PyTorch has support for both CUDA-backed COO matrices and [lobpcg](https://pytorch.org/docs/stable/generated/torch.lobpcg.html). See sparse supported linear algebra [operations](https://pytorch.org/docs/stable/sparse.html#supported-linear-algebra-operations).

> PyTorch has support for both CUDA-backed COO matrices and [lobpcg](https://pytorch.org/docs/stable/generated/torch.lobpcg.html). See sparse supported linear algebra [operations](https://pytorch.org/docs/stable/sparse.html#supported-linear-algebra-operations). Oh, that's awesome! I wasn't aware that sparse support had become this mature. I'm definitely open to having this in PyMDE. I could maybe try to implement it, but even if it was in there, it seems like it's not accessible through `preserve_neighbors` (`cg=True`) option for quadratic init

2022-02-25T19:42:47

0.0

[]

nickmaccarthy/python-datemath

nickmaccarthy__python-datemath-26

be87139fe3f3a2c8ca681186ab0d17668491ef21

diff --git a/.github/workflows/release.yaml b/.github/workflows/release.yaml new file mode 100644 index 0000000..e912349 --- /dev/null +++ b/.github/workflows/release.yaml @@ -0,0 +1,44 @@ +name: Release Packages +on: + release: + types: + - released +jobs: + release-python2: + runs-on: ubuntu-20.04 + needs: + - build-python2 + steps: + - uses: actions/checkout@v2 + - name: install python2 for ubuntu + run: sudo apt install python2 + - name: get pip script for python2 + run: curl https://bootstrap.pypa.io/pip/2.7/get-pip.py --output get-pip.py + - name: install pip for python2 + run: sudo python2 get-pip.py + - name: install requirements for python2 + run: pip2 install -r requirements-2.txt + - name: package and upload python2 + env: + TWINE_USERNAME: __token__ + TWINE_PASSWORD: ${{ secrets.PYPI_API_TOKEN }} + PYTHON_DATEMATH_VERSION: ${{ github.event.release.tag_name }} + run: | + python2.7 setup.py sdist bdist_wheel + twine upload dist/* + release-python3: + runs-on: ubuntu-20.04 + needs: + - build-python3 + steps: + - uses: actions/checkout@v2 + - name: install requirements for python3 + run: pip3 install -r requirements-3.txt + - name: package and upload python3 + env: + TWINE_USERNAME: __token__ + TWINE_PASSWORD: ${{ secrets.PYPI_API_TOKEN }} + PYTHON_DATEMATH_VERSION: ${{ github.event.release.tag_name }} + run: | + python3 setup.py sdist bdist_wheel + twine upload dist/* \ No newline at end of file diff --git a/.gitignore b/.gitignore index 79658d6..2849d3d 100644 --- a/.gitignore +++ b/.gitignore @@ -1,10 +1,6 @@ *.pyc -env -venv -arrow_test.py -dm_pretty.py -ranges.py -tztest.py +env* +venv* # general things to ignore build/ diff --git a/CHANGELOG.md b/CHANGELOG.md index 4baaa19..7b766d6 100644 --- a/CHANGELOG.md +++ b/CHANGELOG.md @@ -1,5 +1,8 @@ # Changelog +## 1.5.3 (2021-04-16) +* [FIX] [Issue #25](https://github.com/nickmaccarthy/python-datemath/issues/25) - Fixed an issue where if you provided an invalid timestamp, i.e. `datemath('2')` you would not get an DateMathException back. Also bumped dependencies. + ## 1.5.2 (2020-10-01) * [FIX] [Issue #21](https://github.com/nickmaccarthy/python-datemath/issues/21) - Fixed an issue where if timezone offset was in a datetime string (ISO8601), the timezone of the returned datemath object would be UTC and not the timezone as specified in the datetime string. diff --git a/VERSION.txt b/VERSION.txt index 4cda8f1..8af85be 100644 --- a/VERSION.txt +++ b/VERSION.txt @@ -1,1 +1,1 @@ -1.5.2 +1.5.3 diff --git a/datemath/__init__.py b/datemath/__init__.py index e0246bf..e0e771e 100644 --- a/datemath/__init__.py +++ b/datemath/__init__.py @@ -1,4 +1,4 @@ -from .helpers import parse +from .helpers import parse, DateMathException def dm(expr, **kwargs): ''' does our datemath and returns an arrow object ''' @@ -6,4 +6,4 @@ def dm(expr, **kwargs): def datemath(expr, **kwargs): ''' does our datemath and returns a datetime object ''' - return parse(expr, **kwargs).datetime + return parse(expr, **kwargs) diff --git a/datemath/helpers.py b/datemath/helpers.py index ef3d995..f953611 100644 --- a/datemath/helpers.py +++ b/datemath/helpers.py @@ -136,6 +136,7 @@ def parse(expression, now=None, tz='UTC', type=None, roundDown=True): math = expression time = now else: + if debug: print('parse() - Found and expression that will hit the catchall') math = '' time = parseTime(expression, tz) @@ -174,6 +175,9 @@ def parseTime(timestamp, timezone='UTC'): ts = ts.replace(tzinfo=timezone) return ts + else: + if debug: print('parseTime() - Doesnt look like we have a valid timestamp, raise an exception. timestamp={}'.format(timestamp)) + raise DateMathException('Valid length timestamp not provide, you gave me a timestamp of "{}", but I need something that has a len() >= 4'.format(timestamp)) def roundDate(now, unit, tz='UTC', roundDown=True): ''' diff --git a/requirements-2.txt b/requirements-2.txt index 60ec35e..6e5b16d 100644 --- a/requirements-2.txt +++ b/requirements-2.txt @@ -7,8 +7,6 @@ packaging==16.8 pkginfo==1.4.2 pyparsing==2.2.0 python-dateutil==2.6.0 -requests==2.13.0 -requests-toolbelt>=0.8.0 six==1.10.0 traceback2==1.4.0 twine>=1.8.1 diff --git a/requirements-3.txt b/requirements-3.txt index 5eba2ad..268d6ea 100644 --- a/requirements-3.txt +++ b/requirements-3.txt @@ -1,7 +1,7 @@ appdirs==1.4.3 args==0.1.0 arrow==0.15.2 -bleach==3.1.2 +bleach==3.3.0 certifi==2019.9.11 chardet==3.0.4 clint==0.5.1 @@ -9,8 +9,8 @@ docutils==0.15.2 idna==2.7 linecache2==1.0.0 packaging==16.8 -pkginfo==1.4.1 -Pygments==2.4.2 +pkginfo==1.4.2 +Pygments==2.7.4 pyparsing==2.2.0 python-dateutil==2.6.0 readme-renderer==24.0

Invalid datemath expression can lead to NoneType Hi, when datemath gets an invalid argument, it can happen that `None `is returned instead of an exception. This is can lead to a weird NoneType if `datemath.datemath` is used instead of `datemath.parse`. ```python import datemath type(datemath.parse("2")) # --> None datemath.datemath("2") # AttributeError: 'NoneType' object has no attribute 'datetime' ``` I use version 1.5.2 (latest). It would be great if both cases could raise a parse exception. Thanks, Kai

2021-04-16T14:16:57

0.0

[]

nickmaccarthy/python-datemath

nickmaccarthy__python-datemath-22

b71536403908faf74f513fd458ff577f6c41bffc

diff --git a/.gitignore b/.gitignore index 546d997..79658d6 100644 --- a/.gitignore +++ b/.gitignore @@ -1,5 +1,6 @@ *.pyc env +venv arrow_test.py dm_pretty.py ranges.py diff --git a/CHANGELOG.md b/CHANGELOG.md index 46d3cfe..4baaa19 100644 --- a/CHANGELOG.md +++ b/CHANGELOG.md @@ -1,5 +1,8 @@ # Changelog +## 1.5.2 (2020-10-01) +* [FIX] [Issue #21](https://github.com/nickmaccarthy/python-datemath/issues/21) - Fixed an issue where if timezone offset was in a datetime string (ISO8601), the timezone of the returned datemath object would be UTC and not the timezone as specified in the datetime string. + ## 1.5.1 (2020-03-25) * [FIX] [Issue #15](https://github.com/nickmaccarthy/python-datemath/issues/15) - Fixed issue with parser finding invalid timeunits and throwing correct errors diff --git a/README.md b/README.md index 55539ce..7d0f9cd 100644 --- a/README.md +++ b/README.md @@ -1,32 +1,37 @@ [![Build Status](https://travis-ci.org/nickmaccarthy/python-datemath.svg?branch=master)](https://travis-ci.org/nickmaccarthy/python-datemath.svg?branch=master) +# Python Datemath + +## What? -# What? A date math (aka datemath) parser compatiable with the elasticsearch 'date math' format -# Why? -Working with date objects in python has always been interesting. Having a background in php, I have been looking for quite some time ( no pun intended ) for a way to do date time interpolation similar to php's ```strtotime()``` function. While the arrow module comes close, I needed something that could turn date math type strings into datetime objects for use in tattle.io and other projects I use in elasticsearch. I have found even more uses for it, including AWS cloudwatch and various other projects and hopefully you will too. +## Why? + +Working with date objects in python has always been interesting. Having a background in php, I have been looking for quite some time ( no pun intended ) for a way to do date time interpolation similar to php's ```strtotime()``` function. While the arrow module comes close, I needed something that could turn date math type strings into datetime objects for use in [tattle.io](http://tattle.io) and other projects I use in elasticsearch. I have found even more uses for it, including AWS cloudwatch and various other projects and hopefully you will too. + +## What is date math? -# What is date math ? Date Math is the short hand arithmetic to find relative time to fixed moments in date and time. Similar to the SOLR date math format, Elasticsearch has its own built in format for short hand date math and this module aims to support that same coverage in python. Documentation from elasticsearch: -http://www.elasticsearch.org/guide/en/elasticsearch/reference/current/mapping-date-format.html#date-math +[http://www.elasticsearch.org/guide/en/elasticsearch/reference/current/mapping-date-format.html#date-math](http://www.elasticsearch.org/guide/en/elasticsearch/reference/current/mapping-date-format.html#date-math) > The date type supports using date math expression when using it in a query/filter (mainly makes sense in range query/filter). - +> > The expression starts with an "anchor" date, which can be either now or a date string (in the applicable format) ending with `||`. - +> > It can then follow by a math expression, supporting `+`, `-` and `/` (rounding). - +> > The units supported are `y` (year), `M` (month), `w` (week), `d` (day), `h` (hour), `m` (minute), and `s` (second). - +> > Here are some samples: `now+1h`, `now+1h+1m`, `now+1h/d`, `2012-01-01||+1M/d`. - +> > Note, when doing range type searches, and the upper value is inclusive, the rounding will properly be rounded to the ceiling instead of flooring it. -# Unit Maps -``` +## Unit Maps + +```yaml y or Y = 'year' M = 'month' m = 'minute' @@ -36,13 +41,17 @@ h or H = 'hour' s or S = 'second' ``` -# Install +## Install + ```python pip install python-datemath ``` -# Examples -Assuming our datetime is currently: '2016-01-01T00:00:00-00:00' -``` + +## Examples + +Assuming our datetime is currently: `2016-01-01T00:00:00-00:00` + +```yaml Expression: Result: now-1h 2015-12-31T23:00:00+00:00 now-1y 2015-01-01T00:00:00+00:00 @@ -62,10 +71,11 @@ now/d 2016-01-01T23:59:59+00:00 now/Y 2016-12-31T23:59:59+00:00 ``` -# Usage +## Usage + By default datemath return an arrow date object representing your timestamp. -``` +```python >>> from datemath import dm >>> >>> dm('now+1h') @@ -92,7 +102,8 @@ By default datemath return an arrow date object representing your timestamp. If you would rather have a string, you can use arrow's ```.format()``` method. > For for info on string formatting, check out arrows tokens section: http://crsmithdev.com/arrow/#tokens -``` + +```python >>> from datemath import dm >>> >>> src_timestamp = dm('2016-01-01') @@ -105,11 +116,11 @@ If you would rather have a string, you can use arrow's ```.format()``` method. >>> >>> new_timestamp.format('YYYY.MM.DD') u'2015.12.18' ->>> ``` Rather have a python datetime object instead? Just pass along the 'datetime' type -``` + +```python from datemath import dm >>> dm('now', type='datetime') datetime.datetime(2016, 1, 22, 22, 58, 28, 338060, tzinfo=tzutc()) @@ -119,7 +130,8 @@ datetime.datetime(2016, 1, 24, 22, 57, 45, 394470, tzinfo=tzutc()) ``` Or you can just import the `datemath` module, this will always give us a native `datetime` object -``` + +```python >>> from datemath import datemath >>> >>> datemath('2016-01-01T16:20:00||/d', roundDown=False) @@ -129,20 +141,28 @@ datetime.datetime(2016, 1, 1, 23, 59, 59, 999999, tzinfo=tzutc()) >>> # roundDown=True is default and implied >>> datemath('2016-01-01T16:20:00||/d') datetime.datetime(2016, 1, 1, 0, 0, tzinfo=tzutc()) ->>> ``` If you want a Epoch timestamp back instead, we can do that. + ```python >>> dm('now+2d-1m', type='timestamp') 1453676321 ``` -# What timezone are my objects in? -By default all objects returned by datemath are in UTC. If you want them them in a different timezone, just pass along the ```tz``` argument. -Timezone list can be found here: https://gist.github.com/pamelafox/986163 -``` -from datemath import dm +## What timezone are my objects in? + +By default all object returned by datemath are in UTC. + +If you want them them back in a different timezone, just pass along the ```tz``` argument. Timezone list can be found here: [https://gist.github.com/pamelafox/986163](https://gist.github.com/pamelafox/986163) + +If you provide a timezone offset in your timestring, datemath will return your time object as that timezone offset in the string. + +Note - currently timestrings with a timezone offset and the usage of the ```tz``` argument will result in the time object being returned with the timezone of what was in the timezone offset in the original string + +```python +>>> from datemath import dm +>>> >>> dm('now') <Arrow [2016-01-26T01:00:53.601088+00:00]> >>> @@ -154,9 +174,26 @@ from datemath import dm >>> >>> dm('2017-10-20 09:15:20', tz='US/Pacific') <Arrow [2017-10-20T09:15:20.000000-08:00]> +>>> +>>> # Timestring with TZ offset in the string (ISO8601 format only) +>>> dm('2016-01-01T00:00:00-05:00') +<Arrow [2016-01-01T00:00:00-05:00]> +>>> +>>> # Timestring with TZ offset with datemath added (again, TS must be in ISO8601) +>>> dm('2016-01-01T00:00:00-05:00||+2d+3h+5m') +<Arrow [2016-01-03T03:05:00-05:00]> +>>> +>>> # Note, timestrings with TZ offsets will be returned as the timezone of the offset in the string even if the "tz" option is used. +>>> dm('2016-01-01T00:00:00-05:00', tz='US/Central') +<Arrow [2016-01-01T00:00:00-05:00]> ``` +## Debugging + +If you would like more verbose output to debug the process of what datemath is doing, simply set `export DATEMATH_DEBUG=true` in your shell then run some datemath tests. To stop debugging, run `unset DATEMATH_DEBUG`. + ## Changes + See CHANGELOG.md # Happy date math'ing! diff --git a/VERSION.txt b/VERSION.txt index 26ca594..4cda8f1 100644 --- a/VERSION.txt +++ b/VERSION.txt @@ -1,1 +1,1 @@ -1.5.1 +1.5.2 diff --git a/datemath/helpers.py b/datemath/helpers.py index df9aba7..ef3d995 100644 --- a/datemath/helpers.py +++ b/datemath/helpers.py @@ -42,9 +42,11 @@ import re import os from dateutil import tz +import dateutil import sys +from pprint import pprint -debug = True if os.environ.get('DEBUG') == 'true' else False +debug = True if os.environ.get('DATEMATH_DEBUG') else False class DateMathException(Exception): pass @@ -75,7 +77,7 @@ def unitMap(c): def as_datetime(expression, now, tz='UTC'): ''' - returs our datemath expression as a python datetime object + returns our datemath expression as a python datetime object note: this has been deprecated and the 'type' argument in parse is the current way ''' return parse(expression, now, tz) @@ -89,26 +91,28 @@ def parse(expression, now=None, tz='UTC', type=None, roundDown=True): :param type - if we are dealing with a arrow or datetime object :param roundDown - wether or not we should round up or round down on this. default is roundDown=True, which means if it was 12:00:00, `/d` would be '00:00:00', and with roundDown=False, `/d` would be '29:59:59' ''' + if debug: print("parse() - starting for expression: {}".format(expression)) if now is None: + if debug: print("parse() - Now is None, setting now to utcnow()") now = arrow.utcnow() - if debug: print("Orig Expression: {0}".format(expression)) + if debug: print("parse() - Orig Expression: {0}".format(expression)) math = '' time = '' if 'UTC' not in tz: - if debug: print("will now convert tz to {0}".format(tz)) + if debug: print("parse() - will now convert tz to {0}".format(tz)) now = now.to(tz) if expression == 'now': - if debug: print("Now, no dm: {0}".format(now)) + if debug: print("parse() - Now, no dm: {0}".format(now)) if type: return getattr(now, type) else: return now elif re.match('\d{10,}', str(expression)): - if debug: print('found an epoch timestamp') + if debug: print('parse() - found an epoch timestamp') if len(str(expression)) == 13: raise DateMathException('Unable to parse epoch timestamps in millis, please convert to the nearest second to continue - i.e. 1451610061 / 1000') ts = arrow.get(int(expression)) @@ -118,7 +122,7 @@ def parse(expression, now=None, tz='UTC', type=None, roundDown=True): ''' parse our standard "now+1d" kind of queries ''' math = expression[3:] time = now - if debug: print('now expression: {0}'.format(now)) + if debug: print('parse() - now expression: {0}'.format(now)) else: ''' parse out datemath with date, ex "2015-10-20||+1d" ''' if '||' in expression: @@ -137,6 +141,7 @@ def parse(expression, now=None, tz='UTC', type=None, roundDown=True): if not math or math == '': rettime = time + rettime = evaluate(math, time, tz, roundDown) if type: @@ -144,17 +149,32 @@ def parse(expression, now=None, tz='UTC', type=None, roundDown=True): else: return rettime - def parseTime(timestamp, timezone='UTC'): ''' - parses a date/time stamp and returns and arrow object + parses a datetime string and returns and arrow object ''' if timestamp and len(timestamp) >= 4: ts = arrow.get(timestamp) - ts = ts.replace(tzinfo=timezone) - return ts + if debug: print("parseTime() - ts = {} :: vars :: {}".format(ts, vars(ts))) + if debug: print("parseTime() - ts timezone = {}".format(ts.tzinfo)) + if debug: print("parseTime() - tzinfo type = {}".format(type(ts.tzinfo))) + if debug: print("parseTime() - timezone that came in = {}".format(timezone)) + + if ts.tzinfo: + import dateutil + if isinstance(ts.tzinfo, dateutil.tz.tz.tzoffset): + # this means our TZ probably came in via our datetime string + # then lets set our tz to whatever tzoffset is + ts = ts.replace(tzinfo=ts.tzinfo) + elif isinstance(ts.tzinfo, dateutil.tz.tz.tzutc): + # otherwise if we are utc, then lets just set it to be as such + ts = ts.replace(tzinfo=timezone) + else: + # otherwise lets just ensure its set to whatever timezone came in + ts = ts.replace(tzinfo=timezone) - + return ts + def roundDate(now, unit, tz='UTC', roundDown=True): ''' rounds our date object @@ -163,7 +183,7 @@ def roundDate(now, unit, tz='UTC', roundDown=True): now = now.floor(unit) else: now = now.ceil(unit) - if debug: print("roundDate Now: {0}".format(now)) + if debug: print("roundDate() Now: {0}".format(now)) return now def calculate(now, offsetval, unit): @@ -175,7 +195,7 @@ def calculate(now, offsetval, unit): offsetval = int(offsetval) try: now = now.shift(**{unit: offsetval}) - if debug: print("Calculate called: now: {}, offsetval: {}, offsetval-type: {}, unit: {}".format(now, offsetval, type(offsetval), unit)) + if debug: print("calculate() called: now: {}, offsetval: {}, offsetval-type: {}, unit: {}".format(now, offsetval, type(offsetval), unit)) return now except Exception as e: raise DateMathException('Unable to calculate date: now: {0}, offsetvalue: {1}, unit: {2} - reason: {3}'.format(now,offsetval,unit,e)) @@ -184,8 +204,8 @@ def evaluate(expression, now, timeZone='UTC', roundDown=True): ''' evaluates our datemath style expression ''' - if debug: print('Expression: {0}'.format(expression)) - if debug: print('Now: {0}'.format(now)) + if debug: print('evaluate() - Expression: {0}'.format(expression)) + if debug: print('evaluate() - Now: {0}'.format(now)) val = 0 i = 0 while i < len(expression): @@ -221,7 +241,7 @@ def evaluate(expression, now, timeZone='UTC', roundDown=True): raise DateMathException(''''{}' is not a valid timeunit for expression: '{}' '''.format(char, expression)) i += 1 - if debug: print("Fin: {0}".format(now)) + if debug: print("evaluate() - Finished: {0}".format(now)) if debug: print('\n\n') return now

Timezone in string doesn't work/isn't blocked Hi, I tried some examples with this library and it accepts timezones in the string but seems to ignore them: ```python datemath("2016-01-01T16:20:00.5+12:00") datetime.datetime(2016, 1, 1, 16, 20, 0, 500000, tzinfo=tzutc()) ``` If you then try to do math on it it fails: ```python datemath("2016-01-01T16:20:00.5+12:00||+1d") datemath.helpers.DateMathException: Unable to calculate date: now: 2016-01-01T16:20:00.500000+00:00, offsetvalue: 1.0, unit: days - reason: 'Arrow' object has no attribute 'shift' ``` Can you please check if timezones in the string are supported and if not, block them? datemath version: 1.5.1 python version: 3.7.1 PS: I'm aware of the tz parameter but was curious if it's supported inside the string

2020-10-01T20:17:31

0.0

[]

pybamm-team/PyBaMM

pybamm-team__PyBaMM-4641

c471f0db7d841033d9edbccd54d19b99d8c70a15

diff --git a/src/pybamm/experiment/step/base_step.py b/src/pybamm/experiment/step/base_step.py index a7dfa9c9ba..8dd829935e 100644 --- a/src/pybamm/experiment/step/base_step.py +++ b/src/pybamm/experiment/step/base_step.py @@ -1,6 +1,4 @@ -# -# Private classes and functions for experiment steps -# +from __future__ import annotations import pybamm import numpy as np from datetime import datetime @@ -492,6 +490,10 @@ def set_up(self, new_model, new_parameter_values): } +def get_unit_from(a_string: str) -> str: + return a_string.lstrip("0123456789.-eE ") + + def _convert_time_to_seconds(time_and_units): """Convert a time in seconds, minutes or hours to a time in seconds""" if time_and_units is None: @@ -501,11 +503,10 @@ def _convert_time_to_seconds(time_and_units): if isinstance(time_and_units, numbers.Number): if time_and_units <= 0: raise ValueError("time must be positive") - else: - return time_and_units + return time_and_units # Split number and units - units = time_and_units.lstrip("0123456789.- ") + units = get_unit_from(time_and_units) time = time_and_units[: -len(units)] if units in ["second", "seconds", "s", "sec"]: time_in_seconds = float(time) @@ -528,7 +529,7 @@ def _convert_temperature_to_kelvin(temperature_and_units): return temperature_and_units # Split number and units - units = temperature_and_units.lstrip("0123456789.- ") + units = get_unit_from(temperature_and_units) temperature = temperature_and_units[: -len(units)] if units in ["K"]: temperature_in_kelvin = float(temperature) @@ -547,7 +548,7 @@ def _convert_electric(value_string): value = 1 / float(value_string[2:]) else: # All other cases e.g. 4 A, 2.5 V, 1.5 Ohm - unit = value_string.lstrip("0123456789.- ") + unit = get_unit_from(value_string) value = float(value_string[: -len(unit)]) # Catch milli- prefix if unit.startswith("m"):

[Bug]: pybam.Experiment parse units error ### PyBaMM Version 24.9.0 ### Python Version python 3.10 ### Describe the bug in https://github.com/pybamm-team/PyBaMM/blob/develop/src/pybamm/experiment/step/base_step.py function def _convert_electric(value_string), if the value_string float value contains 'e' or 'E', the units incorrect parsed: Python 3.12.3 (main, Nov 6 2024, 18:32:19) [GCC 13.2.0] on linux Type "help", "copyright", "credits" or "license" for more information. >>> value_string = "-1.1729285959021311e-05 W" >>> unit = value_string.lstrip("0123456789.- ") >>> unit 'e-05 W' >>> unit = value_string.lstrip("0123456789eE.- ") >>> unit 'W' >>> for python3, according to https://docs.python.org/3/library/functions.html#float, should accept 'e' or 'E' as well ### Steps to Reproduce 1. value_string = "-1.1729285959021311e-05 W" 2. unit = value_string.lstrip("0123456789.- ") get incorrect unit 'e-05 W', it should be 'W' ### Relevant log output ```shell ```

Thanks @chuckliu1979 I will take a look at this

2024-12-03T02:00:53

0.0

[]

pybamm-team/PyBaMM

pybamm-team__PyBaMM-4622

0a46a3b7b833c86e3d50f3a9358022551dec4b1f

diff --git a/CHANGELOG.md b/CHANGELOG.md index ff376c2edc..5748b9f227 100644 --- a/CHANGELOG.md +++ b/CHANGELOG.md @@ -3,6 +3,10 @@ ## Features - Added two more submodels (options) for the SEI: Lars von Kolzenberg (2020) model and Tunneling Limit model ([#4394](https://github.com/pybamm-team/PyBaMM/pull/4394)) +## Bug fixes + +- Reverted modifications to quickplot from [#4529](https://github.com/pybamm-team/PyBaMM/pull/4529) which caused issues with the plots displaying correct variable names. ([#4622](https://github.com/pybamm-team/PyBaMM/pull/4622)) + ## Breaking changes - Double-layer SEI models have been removed (with the corresponding parameters). All models assume now a single SEI layer. ([#4470](https://github.com/pybamm-team/PyBaMM/pull/4470)) diff --git a/src/pybamm/plotting/quick_plot.py b/src/pybamm/plotting/quick_plot.py index e3ad40855e..babfd2e761 100644 --- a/src/pybamm/plotting/quick_plot.py +++ b/src/pybamm/plotting/quick_plot.py @@ -1,7 +1,6 @@ # # Class for quick plotting of variables from models # -from __future__ import annotations import os import numpy as np import pybamm @@ -480,13 +479,13 @@ def reset_axis(self): ): # pragma: no cover raise ValueError(f"Axis limits cannot be NaN for variables '{key}'") - def plot(self, t: float | list[float], dynamic: bool = False): + def plot(self, t, dynamic=False): """Produces a quick plot with the internal states at time t. Parameters ---------- - t : float or list of float - Dimensional time (in 'time_units') at which to plot. Can be a single time or a list of times. + t : float + Dimensional time (in 'time_units') at which to plot. dynamic : bool, optional Determine whether to allocate space for a slider at the bottom of the plot when generating a dynamic plot. If True, creates a dynamic plot with a slider. @@ -494,10 +493,10 @@ def plot(self, t: float | list[float], dynamic: bool = False): plt = import_optional_dependency("matplotlib.pyplot") gridspec = import_optional_dependency("matplotlib.gridspec") + cm = import_optional_dependency("matplotlib", "cm") + colors = import_optional_dependency("matplotlib", "colors") - if not isinstance(t, list): - t = [t] - + t_in_seconds = t * self.time_scaling_factor self.fig = plt.figure(figsize=self.figsize) self.gridspec = gridspec.GridSpec(self.n_rows, self.n_cols) @@ -509,11 +508,6 @@ def plot(self, t: float | list[float], dynamic: bool = False): # initialize empty handles, to be created only if the appropriate plots are made solution_handles = [] - # Generate distinct colors for each time point - time_colors = plt.cm.coolwarm( - np.linspace(0, 1, len(t)) - ) # Use a colormap for distinct colors - for k, (key, variable_lists) in enumerate(self.variables.items()): ax = self.fig.add_subplot(self.gridspec[k]) self.axes.add(key, ax) @@ -524,17 +518,19 @@ def plot(self, t: float | list[float], dynamic: bool = False): ax.xaxis.set_major_locator(plt.MaxNLocator(3)) self.plots[key] = defaultdict(dict) variable_handles = [] - + # Set labels for the first subplot only (avoid repetition) if variable_lists[0][0].dimensions == 0: - # 0D plot: plot as a function of time, indicating multiple times with lines + # 0D plot: plot as a function of time, indicating time t with a line ax.set_xlabel(f"Time [{self.time_unit}]") for i, variable_list in enumerate(variable_lists): for j, variable in enumerate(variable_list): - linestyle = ( - self.linestyles[i] - if len(variable_list) == 1 - else self.linestyles[j] - ) + if len(variable_list) == 1: + # single variable -> use linestyle to differentiate model + linestyle = self.linestyles[i] + else: + # multiple variables -> use linestyle to differentiate + # variables (color differentiates models) + linestyle = self.linestyles[j] full_t = self.ts_seconds[i] (self.plots[key][i][j],) = ax.plot( full_t / self.time_scaling_factor, @@ -546,104 +542,128 @@ def plot(self, t: float | list[float], dynamic: bool = False): solution_handles.append(self.plots[key][i][0]) y_min, y_max = ax.get_ylim() ax.set_ylim(y_min, y_max) - - # Add vertical lines for each time in the list, using different colors for each time - for idx, t_single in enumerate(t): - t_in_seconds = t_single * self.time_scaling_factor - (self.time_lines[key],) = ax.plot( - [ - t_in_seconds / self.time_scaling_factor, - t_in_seconds / self.time_scaling_factor, - ], - [y_min, y_max], - "--", # Dashed lines - lw=1.5, - color=time_colors[idx], # Different color for each time - label=f"t = {t_single:.2f} {self.time_unit}", - ) - ax.legend() - + (self.time_lines[key],) = ax.plot( + [ + t_in_seconds / self.time_scaling_factor, + t_in_seconds / self.time_scaling_factor, + ], + [y_min, y_max], + "k--", + lw=1.5, + ) elif variable_lists[0][0].dimensions == 1: - # 1D plot: plot as a function of x at different times + # 1D plot: plot as a function of x at time t + # Read dictionary of spatial variables spatial_vars = self.spatial_variable_dict[key] spatial_var_name = next(iter(spatial_vars.keys())) - ax.set_xlabel(f"{spatial_var_name} [{self.spatial_unit}]") - - for idx, t_single in enumerate(t): - t_in_seconds = t_single * self.time_scaling_factor - - for i, variable_list in enumerate(variable_lists): - for j, variable in enumerate(variable_list): - linestyle = ( - self.linestyles[i] - if len(variable_list) == 1 - else self.linestyles[j] - ) - (self.plots[key][i][j],) = ax.plot( - self.first_spatial_variable[key], - variable(t_in_seconds, **spatial_vars), - color=time_colors[idx], # Different color for each time - linestyle=linestyle, - label=f"t = {t_single:.2f} {self.time_unit}", # Add time label - zorder=10, - ) - variable_handles.append(self.plots[key][0][j]) - solution_handles.append(self.plots[key][i][0]) - - # Add a legend to indicate which plot corresponds to which time - ax.legend() - + ax.set_xlabel( + f"{spatial_var_name} [{self.spatial_unit}]", + ) + for i, variable_list in enumerate(variable_lists): + for j, variable in enumerate(variable_list): + if len(variable_list) == 1: + # single variable -> use linestyle to differentiate model + linestyle = self.linestyles[i] + else: + # multiple variables -> use linestyle to differentiate + # variables (color differentiates models) + linestyle = self.linestyles[j] + (self.plots[key][i][j],) = ax.plot( + self.first_spatial_variable[key], + variable(t_in_seconds, **spatial_vars), + color=self.colors[i], + linestyle=linestyle, + zorder=10, + ) + variable_handles.append(self.plots[key][0][j]) + solution_handles.append(self.plots[key][i][0]) + # add lines for boundaries between subdomains + for boundary in variable_lists[0][0].internal_boundaries: + boundary_scaled = boundary * self.spatial_factor + ax.axvline(boundary_scaled, color="0.5", lw=1, zorder=0) elif variable_lists[0][0].dimensions == 2: - # 2D plot: superimpose plots at different times + # Read dictionary of spatial variables spatial_vars = self.spatial_variable_dict[key] + # there can only be one entry in the variable list variable = variable_lists[0][0] - - for t_single in t: - t_in_seconds = t_single * self.time_scaling_factor + # different order based on whether the domains are x-r, x-z or y-z, etc + if self.x_first_and_y_second[key] is False: + x_name = list(spatial_vars.keys())[1][0] + y_name = next(iter(spatial_vars.keys()))[0] + x = self.second_spatial_variable[key] + y = self.first_spatial_variable[key] + var = variable(t_in_seconds, **spatial_vars) + else: + x_name = next(iter(spatial_vars.keys()))[0] + y_name = list(spatial_vars.keys())[1][0] x = self.first_spatial_variable[key] y = self.second_spatial_variable[key] var = variable(t_in_seconds, **spatial_vars).T - - ax.set_xlabel( - f"{next(iter(spatial_vars.keys()))[0]} [{self.spatial_unit}]" - ) - ax.set_ylabel( - f"{list(spatial_vars.keys())[1][0]} [{self.spatial_unit}]" + ax.set_xlabel(f"{x_name} [{self.spatial_unit}]") + ax.set_ylabel(f"{y_name} [{self.spatial_unit}]") + vmin, vmax = self.variable_limits[key] + # store the plot and the var data (for testing) as cant access + # z data from QuadMesh or QuadContourSet object + if self.is_y_z[key] is True: + self.plots[key][0][0] = ax.pcolormesh( + x, + y, + var, + vmin=vmin, + vmax=vmax, + shading=self.shading, ) - vmin, vmax = self.variable_limits[key] - - # Use contourf and colorbars to represent the values - contour_plot = ax.contourf( - x, y, var, levels=100, vmin=vmin, vmax=vmax, cmap="coolwarm" + else: + self.plots[key][0][0] = ax.contourf( + x, y, var, levels=100, vmin=vmin, vmax=vmax ) - self.plots[key][0][0] = contour_plot - self.colorbars[key] = self.fig.colorbar(contour_plot, ax=ax) - - self.plots[key][0][1] = var - - ax.set_title(f"t = {t_single:.2f} {self.time_unit}") + self.plots[key][0][1] = var + if vmin is None and vmax is None: + vmin = ax_min(var) + vmax = ax_max(var) + self.colorbars[key] = self.fig.colorbar( + cm.ScalarMappable(colors.Normalize(vmin=vmin, vmax=vmax)), + ax=ax, + ) + # Set either y label or legend entries + if len(key) == 1: + title = split_long_string(key[0]) + ax.set_title(title, fontsize="medium") + else: + ax.legend( + variable_handles, + [split_long_string(s, 6) for s in key], + bbox_to_anchor=(0.5, 1), + loc="lower center", + ) - # Set global legend if there are multiple models + # Set global legend if len(self.labels) > 1: fig_legend = self.fig.legend( solution_handles, self.labels, loc="lower right" ) + # Get the position of the top of the legend in relative figure units + # There may be a better way ... + try: + legend_top_inches = fig_legend.get_window_extent( + renderer=self.fig.canvas.get_renderer() + ).get_points()[1, 1] + fig_height_inches = (self.fig.get_size_inches() * self.fig.dpi)[1] + legend_top = legend_top_inches / fig_height_inches + except AttributeError: # pragma: no cover + # When testing the examples we set the matplotlib backend to "Template" + # which means that the above code doesn't work. Since this is just for + # that particular test we can just skip it + legend_top = 0 else: - fig_legend = None + legend_top = 0 - # Fix layout for sliders if dynamic + # Fix layout if dynamic: slider_top = 0.05 else: slider_top = 0 - bottom = max( - fig_legend.get_window_extent( - renderer=self.fig.canvas.get_renderer() - ).get_points()[1, 1] - if fig_legend - else 0, - slider_top, - ) + bottom = max(legend_top, slider_top) self.gridspec.tight_layout(self.fig, rect=[0, bottom, 1, 1]) def dynamic_plot(self, show_plot=True, step=None):

[Bug]: Variables names missing from plot ### PyBaMM Version 24.11.1 ### Python Version 3.11 ### Describe the bug Dynamic plot has no variable labels and shows the legend entries as `t=0.00h` for all plots ![Image](https://github.com/user-attachments/assets/4ab94412-5ffd-4298-8c7a-fa93a5c76e35) ### Steps to Reproduce ```python3 import pybamm model = pybamm.lithium_ion.SPM() sim = pybamm.Simulation(model) sim.solve([0, 3600]) sim.plot() ``` ### Relevant log output ```shell ```

From https://github.com/pybamm-team/PyBaMM/pull/4529 @rtimms Sorry missed in the review cc: @medha-14, this might be worth a look! Ah, I must have missed this , will get to fixing this in a few days. Another issue that I believe is related: when plotting 2D variables the colourbar range is wrong. Edit: digging a bit, I suspect the issue for the colourmaps is also to do with the plotting multiple times in the same plot. For 2D this should not be possible so the fix there only requires reverting that code to the previous version and throwing an error if multiple times are passed for a 2D variable. ## 24.9 ![Image](https://github.com/user-attachments/assets/31be8b94-3128-4fdd-8886-a936f8dd036b) ## 24.11 ![Image](https://github.com/user-attachments/assets/d807d9f8-2ba9-4751-8396-13e027d1c62f) Digging a bit deeper, there also some inconsistencies when plotting multiple solutions in one plot. Given that there are multiple issues to fix (basically the issue is a lot harder than I anticipated in #4271 😅) and plotting is quite a major feature, I suggest reverting #4529 and making a minor release. We can then address this for the next major release. ![Image](https://github.com/user-attachments/assets/9aa0319c-f71b-4489-add9-3ad444a46f5d)

2024-11-27T15:45:19

0.0

[]

pybamm-team/PyBaMM

pybamm-team__PyBaMM-4591

e17b549fb20f036cf6da474329f7c58ca11365fe

diff --git a/.github/workflows/benchmark_on_push.yml b/.github/workflows/benchmark_on_push.yml index 2883eb5f26..3aa9fce9c0 100644 --- a/.github/workflows/benchmark_on_push.yml +++ b/.github/workflows/benchmark_on_push.yml @@ -8,6 +8,9 @@ concurrency: group: ${{ github.workflow }}-${{ github.ref }} cancel-in-progress: true +env: + PYBAMM_DISABLE_TELEMETRY: "true" + jobs: benchmarks: runs-on: ubuntu-latest diff --git a/.github/workflows/periodic_benchmarks.yml b/.github/workflows/periodic_benchmarks.yml index 14f97c55d8..30603a0ea9 100644 --- a/.github/workflows/periodic_benchmarks.yml +++ b/.github/workflows/periodic_benchmarks.yml @@ -15,6 +15,9 @@ on: # workflow manually workflow_dispatch: +env: + PYBAMM_DISABLE_TELEMETRY: "true" + jobs: benchmarks: runs-on: ubuntu-latest diff --git a/.github/workflows/publish_pypi.yml b/.github/workflows/publish_pypi.yml index 89c9b057c1..944c900e9a 100644 --- a/.github/workflows/publish_pypi.yml +++ b/.github/workflows/publish_pypi.yml @@ -18,6 +18,7 @@ on: # Set options available for all jobs that use cibuildwheel env: + PYBAMM_DISABLE_TELEMETRY: "true" # Increase pip debugging output, equivalent to `pip -vv` CIBW_BUILD_VERBOSITY: 2 # Disable build isolation to allow pre-installing build-time dependencies. @@ -75,6 +76,7 @@ jobs: run: pipx run cibuildwheel --output-dir wheelhouse env: CIBW_ENVIRONMENT: > + PYBAMM_DISABLE_TELEMETRY="true" PYBAMM_USE_VCPKG=ON VCPKG_ROOT_DIR=C:\vcpkg VCPKG_DEFAULT_TRIPLET=x64-windows-static-md @@ -116,6 +118,8 @@ jobs: - name: Build wheels on Linux run: pipx run cibuildwheel --output-dir wheelhouse env: + CIBW_ENVIRONMENT: > + PYBAMM_DISABLE_TELEMETRY="true" CIBW_ARCHS_LINUX: x86_64 CIBW_BEFORE_ALL_LINUX: > yum -y install openblas-devel lapack-devel && @@ -242,6 +246,8 @@ jobs: python scripts/install_KLU_Sundials.py python -m cibuildwheel --output-dir wheelhouse env: + CIBW_ENVIRONMENT: > + PYBAMM_DISABLE_TELEMETRY="true" # 10.13 for Intel (macos-13), 11.0 for Apple Silicon (macos-14 and macos-latest) MACOSX_DEPLOYMENT_TARGET: ${{ matrix.os == 'macos-14' && '11.0' || '10.13' }} CIBW_ARCHS_MACOS: auto diff --git a/.github/workflows/run_benchmarks_over_history.yml b/.github/workflows/run_benchmarks_over_history.yml index d66704a635..28960fb4da 100644 --- a/.github/workflows/run_benchmarks_over_history.yml +++ b/.github/workflows/run_benchmarks_over_history.yml @@ -18,6 +18,10 @@ on: ncommits: description: "Number of commits to benchmark between commit_start and commit_end" default: "100" + +env: + PYBAMM_DISABLE_TELEMETRY: "true" + jobs: benchmarks: runs-on: ubuntu-latest diff --git a/.github/workflows/work_precision_sets.yml b/.github/workflows/work_precision_sets.yml index c167bf5747..5810956786 100644 --- a/.github/workflows/work_precision_sets.yml +++ b/.github/workflows/work_precision_sets.yml @@ -5,6 +5,9 @@ on: types: [published] workflow_dispatch: +env: + PYBAMM_DISABLE_TELEMETRY: "true" + jobs: benchmarks_on_release: if: github.repository_owner == 'pybamm-team' diff --git a/noxfile.py b/noxfile.py index d65812b8ed..5ab32f463f 100644 --- a/noxfile.py +++ b/noxfile.py @@ -49,7 +49,6 @@ def set_iree_state(): "IREE_INDEX_URL": os.getenv( "IREE_INDEX_URL", "https://iree.dev/pip-release-links.html" ), - "PYBAMM_DISABLE_TELEMETRY": "true", } VENV_DIR = Path("./venv").resolve() diff --git a/src/pybamm/config.py b/src/pybamm/config.py index ba7171e5d2..0bd5c96eb6 100644 --- a/src/pybamm/config.py +++ b/src/pybamm/config.py @@ -8,6 +8,18 @@ import time +def check_env_opt_out(): + return os.getenv("PYBAMM_DISABLE_TELEMETRY", "false").lower() != "false" + + +def check_opt_out(): + opt_out = check_env_opt_out() + config = pybamm.config.read() + if config: + opt_out = opt_out or not config["enable_telemetry"] + return opt_out + + def is_running_tests(): # pragma: no cover """ Detect if the code is being run as part of a test suite or building docs with Sphinx. @@ -21,20 +33,18 @@ def is_running_tests(): # pragma: no cover ): return True - # Check for GitHub Actions environment variable - if "GITHUB_ACTIONS" in os.environ: - return True - # Check for other common CI environment variables - ci_env_vars = ["CI", "TRAVIS", "CIRCLECI", "JENKINS_URL", "GITLAB_CI"] + ci_env_vars = [ + "GITHUB_ACTIONS", + "CI", + "TRAVIS", + "CIRCLECI", + "JENKINS_URL", + "GITLAB_CI", + ] if any(var in os.environ for var in ci_env_vars): return True - # Check for common test runner names in command-line arguments - test_runners = ["pytest", "unittest", "nose", "trial", "nox", "tox"] - if any(runner in arg.lower() for arg in sys.argv for runner in test_runners): - return True - # Check if building docs with Sphinx if any(mod == "sphinx" or mod.startswith("sphinx.") for mod in sys.modules): print( @@ -110,7 +120,7 @@ def get_input(): # pragma: no cover def generate(): - if is_running_tests(): + if is_running_tests() or check_opt_out(): return # Check if the config file already exists diff --git a/src/pybamm/telemetry.py b/src/pybamm/telemetry.py index b3319b9237..2dad19f814 100644 --- a/src/pybamm/telemetry.py +++ b/src/pybamm/telemetry.py @@ -1,38 +1,45 @@ from posthog import Posthog -import os import pybamm import sys -_posthog = Posthog( - # this is the public, write only API key, so it's ok to include it here - project_api_key="phc_bLZKBW03XjgiRhbWnPsnKPr0iw0z03fA6ZZYjxgW7ej", - host="https://us.i.posthog.com", -) -_posthog.log.setLevel("CRITICAL") +class MockTelemetry: + def __init__(self): + self.disabled = True + @staticmethod + def capture(**kwargs): # pragma: no cover + pass -def disable(): - _posthog.disabled = True + +if pybamm.config.check_opt_out(): + _posthog = MockTelemetry() +else: # pragma: no cover + _posthog = Posthog( + # this is the public, write only API key, so it's ok to include it here + project_api_key="phc_bLZKBW03XjgiRhbWnPsnKPr0iw0z03fA6ZZYjxgW7ej", + host="https://us.i.posthog.com", + ) + _posthog.log.setLevel("CRITICAL") -_opt_out = os.getenv("PYBAMM_DISABLE_TELEMETRY", "false").lower() -if _opt_out != "false": # pragma: no cover - disable() +def disable(): + _posthog.disabled = True def capture(event): # pragma: no cover - # don't capture events in automated testing if pybamm.config.is_running_tests() or _posthog.disabled: return - properties = { - "python_version": sys.version, - "pybamm_version": pybamm.__version__, - } + if pybamm.config.check_opt_out(): + disable() + return config = pybamm.config.read() if config: - if config["enable_telemetry"]: - user_id = config["uuid"] - _posthog.capture(user_id, event, properties=properties) + properties = { + "python_version": sys.version, + "pybamm_version": pybamm.__version__, + } + user_id = config["uuid"] + _posthog.capture(distinct_id=user_id, event=event, properties=properties)

Fortnightly build for wheels failed The build is failing with the following logs - https://github.com/pybamm-team/PyBaMM/actions/runs/11854441290

2024-11-15T22:02:44

0.0

[]

pybamm-team/PyBaMM

pybamm-team__PyBaMM-4562

bf8f6238cfbd6fc6dadec8a527307d14533dc9c1

diff --git a/CHANGELOG.md b/CHANGELOG.md index bf376c1a13..eee7de05d7 100644 --- a/CHANGELOG.md +++ b/CHANGELOG.md @@ -2,6 +2,7 @@ ## Features +- Added `CoupledVariable` which provides a placeholder variable whose equation can be elsewhere in the model. ([#4556](https://github.com/pybamm-team/PyBaMM/pull/4556)) - Adds support to `pybamm.Experiment` for the `output_variables` option in the `IDAKLUSolver`. ([#4534](https://github.com/pybamm-team/PyBaMM/pull/4534)) - Adds an option "voltage as a state" that can be "false" (default) or "true". If "true" adds an explicit algebraic equation for the voltage. ([#4507](https://github.com/pybamm-team/PyBaMM/pull/4507)) - Improved `QuickPlot` accuracy for simulations with Hermite interpolation. ([#4483](https://github.com/pybamm-team/PyBaMM/pull/4483)) @@ -21,6 +22,8 @@ - Removed the `start_step_offset` setting and disabled minimum `dt` warnings for drive cycles with the (`IDAKLUSolver`). ([#4416](https://github.com/pybamm-team/PyBaMM/pull/4416)) ## Bug Fixes +- Added error for binary operators on two concatenations with different numbers of children. Previously, the extra children were dropped. Also fixed bug where Q_rxn was dropped from the total heating term in half-cell models. ([#4562](https://github.com/pybamm-team/PyBaMM/pull/4562)) +- Fixed bug where Q_rxn was set to 0 for the negative electrode in half-cell models. ([#4557](https://github.com/pybamm-team/PyBaMM/pull/4557)) - Fixed bug in post-processing solutions with infeasible experiments using the (`IDAKLUSolver`). ([#4541](https://github.com/pybamm-team/PyBaMM/pull/4541)) - Disabled IREE on MacOS due to compatibility issues and added the CasADI path to the environment to resolve issues on MacOS and Linux. Windows diff --git a/src/pybamm/expression_tree/binary_operators.py b/src/pybamm/expression_tree/binary_operators.py index 42f2f74e24..be3df653ad 100644 --- a/src/pybamm/expression_tree/binary_operators.py +++ b/src/pybamm/expression_tree/binary_operators.py @@ -856,12 +856,17 @@ def _simplified_binary_broadcast_concatenation( elif isinstance(right, pybamm.Concatenation) and not isinstance( right, pybamm.ConcatenationVariable ): - return left.create_copy( - [ - operator(left_child, right_child) - for left_child, right_child in zip(left.orphans, right.orphans) - ] - ) + if len(left.orphans) == len(right.orphans): + return left.create_copy( + [ + operator(left_child, right_child) + for left_child, right_child in zip(left.orphans, right.orphans) + ] + ) + else: + raise AssertionError( + "Concatenations must have the same number of children" + ) if isinstance(right, pybamm.Concatenation) and not isinstance( right, pybamm.ConcatenationVariable ): diff --git a/src/pybamm/models/full_battery_models/base_battery_model.py b/src/pybamm/models/full_battery_models/base_battery_model.py index a445f47d19..153c447ed3 100644 --- a/src/pybamm/models/full_battery_models/base_battery_model.py +++ b/src/pybamm/models/full_battery_models/base_battery_model.py @@ -989,7 +989,6 @@ def options(self, extra_options): raise pybamm.OptionError( f"must use surface formulation to solve {self!s} with hydrolysis" ) - self._options = options def set_standard_output_variables(self): diff --git a/src/pybamm/models/submodels/thermal/base_thermal.py b/src/pybamm/models/submodels/thermal/base_thermal.py index 384cb1723a..d5a39435fd 100644 --- a/src/pybamm/models/submodels/thermal/base_thermal.py +++ b/src/pybamm/models/submodels/thermal/base_thermal.py @@ -88,6 +88,12 @@ def _get_standard_coupled_variables(self, variables): # TODO: change full stefan-maxwell conductivity so that i_e is always # a Concatenation i_e = variables["Electrolyte current density [A.m-2]"] + # Special case for half cell -- i_e has to be a concatenation for this to work due to a mismatch with Q_ohm, so we make a new i_e which is a concatenation. + if (not isinstance(i_e, pybamm.Concatenation)) and ( + self.options.electrode_types["negative"] == "planar" + ): + i_e = self._get_i_e_for_half_cell_thermal(variables) + phi_e = variables["Electrolyte potential [V]"] if isinstance(i_e, pybamm.Concatenation): # compute by domain if possible @@ -411,3 +417,23 @@ def _yz_average(self, var): return pybamm.z_average(var) elif self.options["dimensionality"] == 2: return pybamm.yz_average(var) + + def _get_i_e_for_half_cell_thermal(self, variables): + c_e_s = variables["Separator electrolyte concentration [mol.m-3]"] + c_e_p = variables["Positive electrolyte concentration [mol.m-3]"] + grad_phi_e_s = variables["Gradient of separator electrolyte potential [V.m-1]"] + grad_phi_e_p = variables["Gradient of positive electrolyte potential [V.m-1]"] + grad_c_e_s = pybamm.grad(c_e_s) + grad_c_e_p = pybamm.grad(c_e_p) + T_s = variables["Separator temperature [K]"] + T_p = variables["Positive electrode temperature [K]"] + tor_s = variables["Separator electrolyte transport efficiency"] + tor_p = variables["Positive electrolyte transport efficiency"] + i_e_s = (self.param.kappa_e(c_e_s, T_s) * tor_s) * ( + self.param.chiRT_over_Fc(c_e_s, T_s) * grad_c_e_s - grad_phi_e_s + ) + i_e_p = (self.param.kappa_e(c_e_p, T_p) * tor_p) * ( + self.param.chiRT_over_Fc(c_e_p, T_p) * grad_c_e_p - grad_phi_e_p + ) + i_e = pybamm.concatenation(i_e_s, i_e_p) + return i_e

[Bug]: Binary operators on concatenations assume that each concatenation has the same number of orphans ### PyBaMM Version develop ### Python Version 3.12 ### Describe the bug If a binary operator, such as `+`, is invoked with 2 concatenations (i.e. `isinstance(left, pybamm.Concatenation) and isinstance(right, pybamm.Concatenation) == True`), and the concatenations have a different number of orphans, the "extra orphans" are dropped. This can be easily seen [here](https://github.com/aabills/PyBaMM/blob/8ce4a20090fee6bae2251d54fd28b4b0d49abdfb/src/pybamm/expression_tree/binary_operators.py#L859). This is causing reaction heating to not be present in DFN half-cell models (see comments in #4557) ### Steps to Reproduce no MWE yet, but should be self evident in the code. ### Relevant log output _No response_

I think there are two different options for "solving" this: 1) Raise an error when this situation occurs 2) Pass the identity of the "unmatched orphans". I implemented (1) to check how impactful it will be, and the only test that fails is the specific case I found this bug in: Half-cell DFN models. Therefore, I am of the opinion that we should implement (1) and fix the underlying issue that causes it in the half-cell case. On board with option 1.

2024-11-04T20:02:11

0.0

[]

pybamm-team/PyBaMM

pybamm-team__PyBaMM-4456

1390ea306e4200e04e6d63a9eac42333a75961e8

diff --git a/CHANGELOG.md b/CHANGELOG.md index 7070ebaf52..f70aae0272 100644 --- a/CHANGELOG.md +++ b/CHANGELOG.md @@ -10,6 +10,7 @@ ## Optimizations +- Performance refactor of JAX BDF Solver with default Jax method set to `"BDF"`. ([#4456](https://github.com/pybamm-team/PyBaMM/pull/4456)) - Improved performance of initialization and reinitialization of ODEs in the (`IDAKLUSolver`). ([#4453](https://github.com/pybamm-team/PyBaMM/pull/4453)) - Removed the `start_step_offset` setting and disabled minimum `dt` warnings for drive cycles with the (`IDAKLUSolver`). ([#4416](https://github.com/pybamm-team/PyBaMM/pull/4416)) diff --git a/examples/scripts/multiprocess_jax_solver.py b/examples/scripts/multiprocess_jax_solver.py new file mode 100644 index 0000000000..8192256ed1 --- /dev/null +++ b/examples/scripts/multiprocess_jax_solver.py @@ -0,0 +1,57 @@ +import pybamm +import time +import numpy as np + + +# This script provides an example for massively vectorised +# model solves using the JAX BDF solver. First, +# we set up the model and process parameters +model = pybamm.lithium_ion.SPM() +model.convert_to_format = "jax" +model.events = [] # remove events (not supported in jax) +geometry = model.default_geometry +param = pybamm.ParameterValues("Chen2020") +param.update({"Current function [A]": "[input]"}) +param.process_geometry(geometry) +param.process_model(model) + +# Discretise and setup solver +mesh = pybamm.Mesh(geometry, model.default_submesh_types, model.default_var_pts) +disc = pybamm.Discretisation(mesh, model.default_spatial_methods) +disc.process_model(model) +t_eval = np.linspace(0, 3600, 100) +solver = pybamm.JaxSolver(atol=1e-6, rtol=1e-6, method="BDF") + +# Set number of vectorised solves +values = np.linspace(0.01, 1.0, 1000) +inputs = [{"Current function [A]": value} for value in values] + +# Run solve for all inputs, with a just-in-time compilation +# occurring on the first solve. All sequential solves will +# use the compiled code, with a large performance improvement. +start_time = time.time() +sol = solver.solve(model, t_eval, inputs=inputs) +print(f"Time taken: {time.time() - start_time}") # 1.3s + +# Rerun the vectorised solve, showing performance improvement +start_time = time.time() +compiled_sol = solver.solve(model, t_eval, inputs=inputs) +print(f"Compiled time taken: {time.time() - start_time}") # 0.42s + +# Plot one of the solves +plot = pybamm.QuickPlot( + sol[5], + [ + "Negative particle concentration [mol.m-3]", + "Electrolyte concentration [mol.m-3]", + "Positive particle concentration [mol.m-3]", + "Current [A]", + "Negative electrode potential [V]", + "Electrolyte potential [V]", + "Positive electrode potential [V]", + "Voltage [V]", + ], + time_unit="seconds", + spatial_unit="um", +) +plot.dynamic_plot() diff --git a/noxfile.py b/noxfile.py index 6567ed167c..14bafcca47 100644 --- a/noxfile.py +++ b/noxfile.py @@ -222,7 +222,7 @@ def run_scripts(session): # https://bitbucket.org/pybtex-devs/pybtex/issues/169/replace-pkg_resources-with # is fixed session.install("setuptools", silent=False) - session.install("-e", ".[all,dev]", silent=False) + session.install("-e", ".[all,dev,jax]", silent=False) session.run("python", "-m", "pytest", "-m", "scripts") diff --git a/src/pybamm/solvers/base_solver.py b/src/pybamm/solvers/base_solver.py index 6334169cf0..a7fc79fe3a 100644 --- a/src/pybamm/solvers/base_solver.py +++ b/src/pybamm/solvers/base_solver.py @@ -90,6 +90,10 @@ def root_method(self): def supports_parallel_solve(self): return False + @property + def requires_explicit_sensitivities(self): + return True + @root_method.setter def root_method(self, method): if method == "casadi": @@ -141,7 +145,7 @@ def set_up(self, model, inputs=None, t_eval=None, ics_only=False): # see if we need to form the explicit sensitivity equations calculate_sensitivities_explicit = ( - model.calculate_sensitivities and not isinstance(self, pybamm.IDAKLUSolver) + model.calculate_sensitivities and self.requires_explicit_sensitivities ) self._set_up_model_sensitivities_inplace( @@ -494,11 +498,7 @@ def _set_up_model_sensitivities_inplace( # if we have a mass matrix, we need to extend it def extend_mass_matrix(M): M_extend = [M.entries] * (num_parameters + 1) - M_extend_pybamm = pybamm.Matrix(block_diag(M_extend, format="csr")) - return M_extend_pybamm - - model.mass_matrix = extend_mass_matrix(model.mass_matrix) - model.mass_matrix = extend_mass_matrix(model.mass_matrix) + return pybamm.Matrix(block_diag(M_extend, format="csr")) model.mass_matrix = extend_mass_matrix(model.mass_matrix) diff --git a/src/pybamm/solvers/idaklu_solver.py b/src/pybamm/solvers/idaklu_solver.py index 32048d89c0..2b2d852697 100644 --- a/src/pybamm/solvers/idaklu_solver.py +++ b/src/pybamm/solvers/idaklu_solver.py @@ -736,6 +736,10 @@ def _demote_64_to_32(self, x: pybamm.EvaluatorJax): def supports_parallel_solve(self): return True + @property + def requires_explicit_sensitivities(self): + return False + def _integrate(self, model, t_eval, inputs_list=None, t_interp=None): """ Solve a DAE model defined by residuals with initial conditions y0. diff --git a/src/pybamm/solvers/jax_bdf_solver.py b/src/pybamm/solvers/jax_bdf_solver.py index 6f0c62b9a8..3db82ca0da 100644 --- a/src/pybamm/solvers/jax_bdf_solver.py +++ b/src/pybamm/solvers/jax_bdf_solver.py @@ -111,68 +111,63 @@ def fun_bind_inputs(y, t): jac_bind_inputs = jax.jacfwd(fun_bind_inputs, argnums=0) - t0 = t_eval[0] - h0 = t_eval[1] - t0 - + t0, h0 = t_eval[0], t_eval[1] - t_eval[0] stepper = _bdf_init( fun_bind_inputs, jac_bind_inputs, mass, t0, y0, h0, rtol, atol ) - i = 0 y_out = jnp.empty((len(t_eval), len(y0)), dtype=y0.dtype) - init_state = [stepper, t_eval, i, y_out] - def cond_fun(state): - _, t_eval, i, _ = state + _, _, i, _ = state return i < len(t_eval) def body_fun(state): stepper, t_eval, i, y_out = state stepper = _bdf_step(stepper, fun_bind_inputs, jac_bind_inputs) - index = jnp.searchsorted(t_eval, stepper.t) - index = index.astype( - "int" + t_eval.dtype.name[-2:] - ) # Coerce index to correct type + index = jnp.searchsorted(t_eval, stepper.t).astype(jnp.int32) - def for_body(j, y_out): - t = t_eval[j] - y_out = y_out.at[jnp.index_exp[j, :]].set(_bdf_interpolate(stepper, t)) - return y_out + def interpolate_and_update(j, y_out): + y = _bdf_interpolate(stepper, t_eval[j]) + return y_out.at[j].set(y) - y_out = jax.lax.fori_loop(i, index, for_body, y_out) - return [stepper, t_eval, index, y_out] + y_out = jax.lax.fori_loop(i, index, interpolate_and_update, y_out) + return stepper, t_eval, index, y_out + + init_state = (stepper, t_eval, 0, y_out) + _, _, _, y_out = jax.lax.while_loop(cond_fun, body_fun, init_state) - stepper, t_eval, i, y_out = jax.lax.while_loop(cond_fun, body_fun, init_state) return y_out - BDFInternalStates = [ - "t", - "atol", - "rtol", - "M", - "newton_tol", - "order", - "h", - "n_equal_steps", - "D", - "y0", - "scale_y0", - "kappa", - "gamma", - "alpha", - "c", - "error_const", - "J", - "LU", - "U", - "psi", - "n_function_evals", - "n_jacobian_evals", - "n_lu_decompositions", - "n_steps", - "consistent_y0_failed", - ] - BDFState = collections.namedtuple("BDFState", BDFInternalStates) + BDFState = collections.namedtuple( + "BDFState", + [ + "t", + "atol", + "rtol", + "M", + "newton_tol", + "order", + "h", + "n_equal_steps", + "D", + "y0", + "scale_y0", + "kappa", + "gamma", + "alpha", + "c", + "error_const", + "J", + "LU", + "U", + "psi", + "n_function_evals", + "n_jacobian_evals", + "n_lu_decompositions", + "n_steps", + "consistent_y0_failed", + ], + ) jax.tree_util.register_pytree_node( BDFState, lambda xs: (tuple(xs), None), lambda _, xs: BDFState(*xs) @@ -211,62 +206,70 @@ def _bdf_init(fun, jac, mass, t0, y0, h0, rtol, atol): absolute tolerance for the solver """ - state = {} - state["t"] = t0 - state["atol"] = atol - state["rtol"] = rtol - state["M"] = mass EPS = jnp.finfo(y0.dtype).eps - state["newton_tol"] = jnp.maximum(10 * EPS / rtol, jnp.minimum(0.03, rtol**0.5)) + # Scaling and tolerance initialisation scale_y0 = atol + rtol * jnp.abs(y0) + newton_tol = jnp.maximum(10 * EPS / rtol, jnp.minimum(0.03, rtol**0.5)) + y0, not_converged = _select_initial_conditions( - fun, mass, t0, y0, state["newton_tol"], scale_y0 + fun, mass, t0, y0, newton_tol, scale_y0 ) - state["consistent_y0_failed"] = not_converged + # Compute initial function and step size f0 = fun(y0, t0) - order = 1 - state["order"] = order - state["h"] = _select_initial_step(atol, rtol, fun, t0, y0, f0, h0) - state["n_equal_steps"] = 0 + h = _select_initial_step(atol, rtol, fun, t0, y0, f0, h0) + + # Initialise the difference matrix, D D = jnp.empty((MAX_ORDER + 1, len(y0)), dtype=y0.dtype) D = D.at[jnp.index_exp[0, :]].set(y0) - D = D.at[jnp.index_exp[1, :]].set(f0 * state["h"]) - state["D"] = D - state["y0"] = y0 - state["scale_y0"] = scale_y0 + D = D.at[jnp.index_exp[1, :]].set(f0 * h) # kappa values for difference orders, taken from Table 1 of [1] kappa = jnp.array([0, -0.1850, -1 / 9, -0.0823, -0.0415, 0]) gamma = jnp.hstack((0, jnp.cumsum(1 / jnp.arange(1, MAX_ORDER + 1)))) alpha = 1.0 / ((1 - kappa) * gamma) - c = state["h"] * alpha[order] + c = h * alpha[1] error_const = kappa * gamma + 1 / jnp.arange(1, MAX_ORDER + 2) - state["kappa"] = kappa - state["gamma"] = gamma - state["alpha"] = alpha - state["c"] = c - state["error_const"] = error_const - + # Jacobian and LU decomp J = jac(y0, t0) - state["J"] = J - - state["LU"] = jax.scipy.linalg.lu_factor(state["M"] - c * J) - - state["U"] = _compute_R(order, 1) - state["psi"] = None - - state["n_function_evals"] = 2 - state["n_jacobian_evals"] = 1 - state["n_lu_decompositions"] = 1 - state["n_steps"] = 0 + LU = jax.scipy.linalg.lu_factor(mass - c * J) + U = _compute_R(1, 1) # Order 1 + + # Create initial BDFState + state = BDFState( + t=t0, + atol=atol, + rtol=rtol, + M=mass, + newton_tol=newton_tol, + consistent_y0_failed=not_converged, + order=1, + h=h, + n_equal_steps=0, + D=D, + y0=y0, + scale_y0=scale_y0, + kappa=kappa, + gamma=gamma, + alpha=alpha, + c=c, + error_const=error_const, + J=J, + LU=LU, + U=U, + psi=None, + n_function_evals=2, + n_jacobian_evals=1, + n_lu_decompositions=1, + n_steps=0, + ) - tuple_state = BDFState(*[state[k] for k in BDFInternalStates]) - y0, scale_y0 = _predict(tuple_state, D) - psi = _update_psi(tuple_state, D) - return tuple_state._replace(y0=y0, scale_y0=scale_y0, psi=psi) + # Predict initial y0, scale_yo, update state + y0, scale_y0 = _predict(state, D) + psi = _update_psi(state, D) + return state._replace(y0=y0, scale_y0=scale_y0, psi=psi) def _compute_R(order, factor): """ @@ -374,10 +377,8 @@ def _predict(state, D): """ predict forward to new step (eq 2 in [1]) """ - n = len(state.y0) - order = state.order - orders = jnp.repeat(jnp.arange(MAX_ORDER + 1).reshape(-1, 1), n, axis=1) - subD = jnp.where(orders <= order, D, 0) + orders = jnp.arange(MAX_ORDER + 1)[:, None] + subD = jnp.where(orders <= state.order, D, 0) y0 = jnp.sum(subD, axis=0) scale_y0 = state.atol + state.rtol * jnp.abs(state.y0) return y0, scale_y0 @@ -397,7 +398,7 @@ def _update_psi(state, D): def _update_difference_for_next_step(state, d): """ - update of difference equations can be done efficiently + Update of difference equations can be done efficiently by reusing d and D. From first equation on page 4 of [1]: @@ -409,34 +410,21 @@ def _update_difference_for_next_step(state, d): Combining these gives the following algorithm """ order = state.order - D = state.D - D = D.at[jnp.index_exp[order + 2]].set(d - D[order + 1]) - D = D.at[jnp.index_exp[order + 1]].set(d) - i = order - while_state = [i, D] - - def while_cond(while_state): - i, _ = while_state - return i >= 0 + D = state.D.at[order + 2].set(d - state.D[order + 1]) + D = D.at[order + 1].set(d) - def while_body(while_state): - i, D = while_state - D = D.at[jnp.index_exp[i]].add(D[i + 1]) - i -= 1 - return [i, D] + def update_D(i, D): + return D.at[order - i].add(D[order - i + 1]) - i, D = jax.lax.while_loop(while_cond, while_body, while_state) - - return D + return jax.lax.fori_loop(0, order + 1, update_D, D) def _update_step_size_and_lu(state, factor): + """ + Update step size and recompute LU decomposition. + """ state = _update_step_size(state, factor) - - # redo lu (c has changed) LU = jax.scipy.linalg.lu_factor(state.M - state.c * state.J) - n_lu_decompositions = state.n_lu_decompositions + 1 - - return state._replace(LU=LU, n_lu_decompositions=n_lu_decompositions) + return state._replace(LU=LU, n_lu_decompositions=state.n_lu_decompositions + 1) def _update_step_size(state, factor): """ @@ -449,7 +437,6 @@ def _update_step_size(state, factor): """ order = state.order h = state.h * factor - n_equal_steps = 0 c = h * state.alpha[order] # update D using equations in section 3.2 of [1] @@ -461,19 +448,14 @@ def _update_step_size(state, factor): RU = jnp.where( jnp.logical_and(I <= order, J <= order), RU, jnp.identity(MAX_ORDER + 1) ) - D = state.D - D = jnp.dot(RU.T, D) - # D = jax.ops.index_update(D, jax.ops.index[:order + 1], - # jnp.dot(RU.T, D[:order + 1])) + D = jnp.dot(RU.T, state.D) - # update psi (D has changed) + # update psi, y0 (D has changed) psi = _update_psi(state, D) - - # update y0 (D has changed) y0, scale_y0 = _predict(state, D) return state._replace( - n_equal_steps=n_equal_steps, + n_equal_steps=0, h=h, c=c, D=D, @@ -484,27 +466,23 @@ def _update_step_size(state, factor): def _update_jacobian(state, jac): """ - we update the jacobian using J(t_{n+1}, y^0_{n+1}) + Update the jacobian using J(t_{n+1}, y^0_{n+1}) following the scipy bdf implementation rather than J(t_n, y_n) as per [1] """ J = jac(state.y0, state.t + state.h) - n_jacobian_evals = state.n_jacobian_evals + 1 LU = jax.scipy.linalg.lu_factor(state.M - state.c * J) - n_lu_decompositions = state.n_lu_decompositions + 1 return state._replace( J=J, - n_jacobian_evals=n_jacobian_evals, + n_jacobian_evals=state.n_jacobian_evals + 1, LU=LU, - n_lu_decompositions=n_lu_decompositions, + n_lu_decompositions=state.n_lu_decompositions + 1, ) def _newton_iteration(state, fun): - tol = state.newton_tol - c = state.c - psi = state.psi + """ + Perform Newton iteration to solve the system. + """ y0 = state.y0 - LU = state.LU - M = state.M scale_y0 = state.scale_y0 t = state.t + state.h d = jnp.zeros(y0.shape, dtype=y0.dtype) @@ -522,17 +500,20 @@ def while_cond(while_state): def while_body(while_state): k, converged, dy_norm_old, d, y, n_function_evals = while_state + f_eval = fun(y, t) n_function_evals += 1 - b = c * f_eval - M @ (psi + d) - dy = jax.scipy.linalg.lu_solve(LU, b) + b = state.c * f_eval - state.M @ (state.psi + d) + dy = jax.scipy.linalg.lu_solve(state.LU, b) dy_norm = jnp.sqrt(jnp.mean((dy / scale_y0) ** 2)) rate = dy_norm / dy_norm_old # if iteration is not going to converge in NEWTON_MAXITER # (assuming the current rate), then abort pred = rate >= 1 - pred += rate ** (NEWTON_MAXITER - k) / (1 - rate) * dy_norm > tol + pred += ( + rate ** (NEWTON_MAXITER - k) / (1 - rate) * dy_norm > state.newton_tol + ) pred *= dy_norm_old >= 0 k += pred * (NEWTON_MAXITER - k - 1) @@ -541,7 +522,7 @@ def while_body(while_state): # if converged then break out of iteration early pred = dy_norm_old >= 0.0 - pred *= rate / (1 - rate) * dy_norm < tol + pred *= rate / (1 - rate) * dy_norm < state.newton_tol converged = (dy_norm == 0.0) + pred dy_norm_old = dy_norm @@ -564,7 +545,6 @@ def _prepare_next_step(state, d): def _prepare_next_step_order_change(state, d, y, n_iter): order = state.order - D = _update_difference_for_next_step(state, d) # Note: we are recalculating these from the while loop above, could re-use? @@ -586,7 +566,6 @@ def _prepare_next_step_order_change(state, d, y, n_iter): rms_norm(state.error_const[order + 1] * D[order + 2] / scale_y), jnp.inf, ) - error_norms = jnp.array([error_m_norm, error_norm, error_p_norm]) factors = error_norms ** (-1 / (jnp.arange(3) + order)) @@ -595,111 +574,89 @@ def _prepare_next_step_order_change(state, d, y, n_iter): max_index = jnp.argmax(factors) order += max_index - 1 + # New step size factor factor = jnp.minimum(MAX_FACTOR, safety * factors[max_index]) - - new_state = _update_step_size_and_lu(state._replace(D=D, order=order), factor) - return new_state + new_state = state._replace(D=D, order=order) + return _update_step_size_and_lu(new_state, factor) def _bdf_step(state, fun, jac): - # print('bdf_step', state.t, state.h) - # we will try and use the old jacobian unless convergence of newton iteration - # fails - updated_jacobian = False - # initialise step size and try to make the step, - # iterate, reducing step size until error is in bounds - step_accepted = False - y = jnp.empty_like(state.y0) - d = jnp.empty_like(state.y0) - n_iter = -1 - - # loop until step is accepted - while_state = [state, step_accepted, updated_jacobian, y, d, n_iter] + """ + Perform a BDF step. - def while_cond(while_state): - _, step_accepted, _, _, _, _ = while_state - return step_accepted == False # noqa: E712 + We will try and use the old jacobian unless + convergence of newton iteration fails. + """ - def while_body(while_state): - state, step_accepted, updated_jacobian, y, d, n_iter = while_state + def step_iteration(while_state): + state, updated_jacobian = while_state - # solve BDF equation using y0 as starting point + # Solve BDF equation using Newton iteration converged, n_iter, y, d, state = _newton_iteration(state, fun) - not_converged = converged == False # noqa: E712 - - # newton iteration did not converge, but jacobian has already been - # evaluated so reduce step size by 0.3 (as per [1]) and try again - state = tree_map( - partial(jnp.where, not_converged * updated_jacobian), - _update_step_size_and_lu(state, 0.3), - state, - ) - # if not_converged * updated_jacobian: - # print('not converged, update step size by 0.3') - # if not_converged * (updated_jacobian == False): - # print('not converged, update jacobian') - - # if not converged and jacobian not updated, then update the jacobian and - # try again - (state, updated_jacobian) = tree_map( - partial( - jnp.where, - not_converged * (updated_jacobian == False), # noqa: E712 + # Update Jacobian or reduce step size if not converged + # Evaluated so reduce step size by 0.3 (as per [1]) and try again + state, updated_jacobian = jax.lax.cond( + ~converged, + lambda s, uj: jax.lax.cond( + uj, + lambda s: (_update_step_size_and_lu(s, 0.3), True), + lambda s: (_update_jacobian(s, jac), True), + s, ), - (_update_jacobian(state, jac), True), - (state, False + updated_jacobian), + lambda s, uj: (s, uj), + state, + updated_jacobian, ) safety = 0.9 * (2 * NEWTON_MAXITER + 1) / (2 * NEWTON_MAXITER + n_iter) scale_y = state.atol + state.rtol * jnp.abs(y) + # Calculate error and updated step size factor # combine eq 3, 4 and 6 from [1] to obtain error # Note that error = C_k * h^{k+1} y^{k+1} # and d = D^{k+1} y_{n+1} \approx h^{k+1} y^{k+1} error = state.error_const[state.order] * d - error_norm = rms_norm(error / scale_y) - # calculate optimal step size factor as per eq 2.46 of [2] - factor = jnp.maximum( - MIN_FACTOR, safety * error_norm ** (-1 / (state.order + 1)) + # Calculate optimal step size factor as per eq 2.46 of [2] + factor = jnp.clip( + safety * error_norm ** (-1 / (state.order + 1)), MIN_FACTOR, None ) - # if converged * (error_norm > 1): - # print( - # "converged, but error is too large", - # error_norm, - # factor, - # d, - # scale_y, - # ) - - (state, step_accepted) = tree_map( - partial(jnp.where, converged * (error_norm > 1)), - (_update_step_size_and_lu(state, factor), False), - (state, converged), + # Update step size if error is too large + state = jax.lax.cond( + converged & (error_norm > 1), + lambda s: _update_step_size_and_lu(s, factor), + lambda s: s, + state, ) - return [state, step_accepted, updated_jacobian, y, d, n_iter] - - state, step_accepted, updated_jacobian, y, d, n_iter = jax.lax.while_loop( - while_cond, while_body, while_state + step_accepted = converged & (error_norm <= 1) + return (state, updated_jacobian), (step_accepted, y, d, n_iter) + + # Iterate until step is accepted + (state, _), (_, y, d, n_iter) = jax.lax.while_loop( + lambda carry_and_aux: ~carry_and_aux[1][0], + lambda carry_and_aux: step_iteration(carry_and_aux[0]), + ( + (state, False), + (False, jnp.empty_like(state.y0), jnp.empty_like(state.y0), -1), + ), ) - # take the accepted step + # Update state for the accepted step n_steps = state.n_steps + 1 t = state.t + state.h - - # a change in order is only done after running at order k for k + 1 steps - # (see page 83 of [2]) n_equal_steps = state.n_equal_steps + 1 - state = state._replace(n_equal_steps=n_equal_steps, t=t, n_steps=n_steps) - state = tree_map( - partial(jnp.where, n_equal_steps < state.order + 1), - _prepare_next_step(state, d), - _prepare_next_step_order_change(state, d, y, n_iter), + # Prepare for the next step, potentially changing order + # (see page 83 of [2]) + state = jax.lax.cond( + n_equal_steps < state.order + 1, + lambda s: _prepare_next_step(s, d), + lambda s: _prepare_next_step_order_change(s, d, y, n_iter), + state, ) return state @@ -710,8 +667,6 @@ def _bdf_interpolate(state, t_eval): definition of the interpolating polynomial can be found on page 7 of [1] """ - order = state.order - t = state.t h = state.h D = state.D j = 0 @@ -721,11 +676,11 @@ def _bdf_interpolate(state, t_eval): def while_cond(while_state): j, _, _ = while_state - return j < order + return j < state.order def while_body(while_state): j, time_factor, order_summation = while_state - time_factor *= (t_eval - (t - h * j)) / (h * (1 + j)) + time_factor *= (t_eval - (state.t - h * j)) / (h * (1 + j)) order_summation += D[j + 1] * time_factor j += 1 return [j, time_factor, order_summation] @@ -972,13 +927,13 @@ def jax_bdf_integrate(func, y0, t_eval, *args, rtol=1e-6, atol=1e-6, mass=None): """ Backward Difference formula (BDF) implicit multistep integrator. The basic algorithm is derived in :footcite:t:`byrne1975polyalgorithm`. This particular implementation - follows that implemented in the Matlab routine ode15s described in - :footcite:t:`shampine1997matlab` and the SciPy implementation - :footcite:t:`Virtanen2020` which features the NDF formulas for improved stability, - with associated differences in the error constants, and calculates the jacobian at - J(t_{n+1}, y^0_{n+1}). This implementation was based on that implemented in the - SciPy library :footcite:t:`Virtanen2020`, which also mainly follows - :footcite:t:`shampine1997matlab` but uses the more standard jacobian update. + follows the Matlab routine ode15s described in :footcite:t:`shampine1997matlab` + and the SciPy implementation :footcite:t:`Virtanen2020` which features + the NDF formulas for improved stability, with associated differences in the + error constants, and calculates the jacobian at J(t_{n+1}, y^0_{n+1}). This + implementation was based on that implemented in the SciPy library + :footcite:t:`Virtanen2020`, which also mainly follows :footcite:t:`shampine1997matlab` + but uses the more standard jacobian update. Parameters ---------- diff --git a/src/pybamm/solvers/jax_solver.py b/src/pybamm/solvers/jax_solver.py index bfcdef1882..a1f1733ed6 100644 --- a/src/pybamm/solvers/jax_solver.py +++ b/src/pybamm/solvers/jax_solver.py @@ -31,10 +31,10 @@ class JaxSolver(pybamm.BaseSolver): Parameters ---------- method: str, optional (see `jax.experimental.ode.odeint` for details) - * 'RK45' (default) uses jax.experimental.ode.odeint - * 'BDF' uses custom jax_bdf_integrate (see `jax_bdf_integrate.py` for details) + * 'BDF' (default) uses custom jax_bdf_integrate (see `jax_bdf_integrate.py` for details) + * 'RK45' uses jax.experimental.ode.odeint root_method: str, optional - Method to use to calculate consistent initial conditions. By default this uses + Method to use to calculate consistent initial conditions. By default, this uses the newton chord method internal to the jax bdf solver, otherwise choose from the set of default options defined in docs for pybamm.BaseSolver rtol : float, optional @@ -52,7 +52,7 @@ class JaxSolver(pybamm.BaseSolver): def __init__( self, - method="RK45", + method="BDF", root_method=None, rtol=1e-6, atol=1e-6, @@ -189,6 +189,10 @@ def solve_model_bdf(inputs): def supports_parallel_solve(self): return True + @property + def requires_explicit_sensitivities(self): + return False + def _integrate(self, model, t_eval, inputs=None, t_interp=None): """ Solve a model defined by dydt with initial conditions y0.

[Bug]: JAX BDF solver sensitivities bug ### PyBaMM Version develop ### Python Version 3.12 ### Describe the bug While sensitivities are not currently supported with the JAX solvers, passing the `calculate_sensitivities` argument currently results in an internal error due to extending the mass matrix, but not the corresponding initial conditions. ### Steps to Reproduce Pass `calculate_sensitivities` to Jax BDF `solver.solve()`. I.e, ```python ... solver = pybamm.JaxSolver(atol=1e-6, rtol=1e-6, method="BDF") sol = solver.solve(model, t_eval, calculate_sensitivities=True) ``` ### Relevant log output ```shell Traceback (most recent call last): File "/Users/Documents/Git/forks/PyBaMM/examples/scripts/compare_lithium_ion.py", line 31, in <module> solution = solver.solve(model, t_eval, inputs=inputs, calculate_sensitivities=True) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "/Users/Documents/Git/forks/PyBaMM/src/pybamm/solvers/base_solver.py", line 905, in solve new_solutions = self._integrate( ^^^^^^^^^^^^^^^^ File "/Users/Documents/Git/forks/PyBaMM/src/pybamm/solvers/jax_solver.py", line 231, in _integrate y = asyncio.run(solve_model_for_inputs()) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "/Users/.pyenv/versions/3.12.2/lib/python3.12/asyncio/runners.py", line 194, in run return runner.run(main) ^^^^^^^^^^^^^^^^ File "/Users/.pyenv/versions/3.12.2/lib/python3.12/asyncio/runners.py", line 118, in run return self._loop.run_until_complete(task) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "/Users/.pyenv/versions/3.12.2/lib/python3.12/asyncio/base_events.py", line 685, in run_until_complete return future.result() ^^^^^^^^^^^^^^^ File "/Users/Documents/Git/forks/PyBaMM/src/pybamm/solvers/jax_solver.py", line 229, in solve_model_for_inputs return await asyncio.gather(*coro) ^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "/Users/Documents/Git/forks/PyBaMM/src/pybamm/solvers/jax_solver.py", line 224, in solve_model_async return self._cached_solves[model](inputs_v) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "/Users/Documents/Git/forks/PyBaMM/src/pybamm/solvers/jax_solver.py", line 171, in solve_model_bdf y = pybamm.jax_bdf_integrate( ^^^^^^^^^^^^^^^^^^^^^^^^^ File "/Users/Documents/Git/forks/PyBaMM/src/pybamm/solvers/jax_bdf_solver.py", line 1030, in jax_bdf_integrate return _bdf_odeint_wrapper(converted, mass, rtol, atol, y0, t_eval, *consts, *args) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "/Users/Documents/Git/forks/PyBaMM/src/pybamm/solvers/jax_bdf_solver.py", line 63, in caller return callee(*args) ^^^^^^^^^^^^^ File "/Users/Documents/Git/forks/PyBaMM/src/pybamm/solvers/jax_bdf_solver.py", line 57, in callee return fun(*args) ^^^^^^^^^^ File "/Users/Documents/Git/forks/PyBaMM/src/pybamm/solvers/jax_bdf_solver.py", line 812, in _bdf_odeint_wrapper out = _bdf_odeint(func, mass, rtol, atol, y0, ts, *args) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "/Users/Documents/Git/forks/PyBaMM/src/pybamm/solvers/jax_bdf_solver.py", line 117, in _bdf_odeint stepper = _bdf_init( ^^^^^^^^^^ File "/Users/Documents/Git/forks/PyBaMM/src/pybamm/solvers/jax_bdf_solver.py", line 256, in _bdf_init state["LU"] = jax.scipy.linalg.lu_factor(state["M"] - c * J) ~~~~~~~~~~~^~~~~~~ File "/Users/.pyenv/versions/pybamm-v24.5/lib/python3.12/site-packages/jax/_src/numpy/array_methods.py", line 737, in op return getattr(self.aval, f"_{name}")(self, *args) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "/Users/.pyenv/versions/pybamm-v24.5/lib/python3.12/site-packages/jax/_src/numpy/array_methods.py", line 265, in deferring_binary_op return binary_op(*args) ^^^^^^^^^^^^^^^^ File "/Users/.pyenv/versions/pybamm-v24.5/lib/python3.12/site-packages/jax/_src/numpy/ufuncs.py", line 86, in <lambda> fn = lambda x1, x2, /: lax_fn(*promote_args(numpy_fn.__name__, x1, x2)) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ TypeError: sub got incompatible shapes for broadcasting: (84, 84), (42, 42). ``` [Bug]: JAX BDF solver sensitivities bug ### PyBaMM Version develop ### Python Version 3.12 ### Describe the bug While sensitivities are not currently supported with the JAX solvers, passing the `calculate_sensitivities` argument currently results in an internal error due to extending the mass matrix, but not the corresponding initial conditions. ### Steps to Reproduce Pass `calculate_sensitivities` to Jax BDF `solver.solve()`. I.e, ```python ... solver = pybamm.JaxSolver(atol=1e-6, rtol=1e-6, method="BDF") sol = solver.solve(model, t_eval, calculate_sensitivities=True) ``` ### Relevant log output ```shell Traceback (most recent call last): File "/Users/Documents/Git/forks/PyBaMM/examples/scripts/compare_lithium_ion.py", line 31, in <module> solution = solver.solve(model, t_eval, inputs=inputs, calculate_sensitivities=True) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "/Users/Documents/Git/forks/PyBaMM/src/pybamm/solvers/base_solver.py", line 905, in solve new_solutions = self._integrate( ^^^^^^^^^^^^^^^^ File "/Users/Documents/Git/forks/PyBaMM/src/pybamm/solvers/jax_solver.py", line 231, in _integrate y = asyncio.run(solve_model_for_inputs()) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "/Users/.pyenv/versions/3.12.2/lib/python3.12/asyncio/runners.py", line 194, in run return runner.run(main) ^^^^^^^^^^^^^^^^ File "/Users/.pyenv/versions/3.12.2/lib/python3.12/asyncio/runners.py", line 118, in run return self._loop.run_until_complete(task) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "/Users/.pyenv/versions/3.12.2/lib/python3.12/asyncio/base_events.py", line 685, in run_until_complete return future.result() ^^^^^^^^^^^^^^^ File "/Users/Documents/Git/forks/PyBaMM/src/pybamm/solvers/jax_solver.py", line 229, in solve_model_for_inputs return await asyncio.gather(*coro) ^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "/Users/Documents/Git/forks/PyBaMM/src/pybamm/solvers/jax_solver.py", line 224, in solve_model_async return self._cached_solves[model](inputs_v) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "/Users/Documents/Git/forks/PyBaMM/src/pybamm/solvers/jax_solver.py", line 171, in solve_model_bdf y = pybamm.jax_bdf_integrate( ^^^^^^^^^^^^^^^^^^^^^^^^^ File "/Users/Documents/Git/forks/PyBaMM/src/pybamm/solvers/jax_bdf_solver.py", line 1030, in jax_bdf_integrate return _bdf_odeint_wrapper(converted, mass, rtol, atol, y0, t_eval, *consts, *args) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "/Users/Documents/Git/forks/PyBaMM/src/pybamm/solvers/jax_bdf_solver.py", line 63, in caller return callee(*args) ^^^^^^^^^^^^^ File "/Users/Documents/Git/forks/PyBaMM/src/pybamm/solvers/jax_bdf_solver.py", line 57, in callee return fun(*args) ^^^^^^^^^^ File "/Users/Documents/Git/forks/PyBaMM/src/pybamm/solvers/jax_bdf_solver.py", line 812, in _bdf_odeint_wrapper out = _bdf_odeint(func, mass, rtol, atol, y0, ts, *args) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "/Users/Documents/Git/forks/PyBaMM/src/pybamm/solvers/jax_bdf_solver.py", line 117, in _bdf_odeint stepper = _bdf_init( ^^^^^^^^^^ File "/Users/Documents/Git/forks/PyBaMM/src/pybamm/solvers/jax_bdf_solver.py", line 256, in _bdf_init state["LU"] = jax.scipy.linalg.lu_factor(state["M"] - c * J) ~~~~~~~~~~~^~~~~~~ File "/Users/.pyenv/versions/pybamm-v24.5/lib/python3.12/site-packages/jax/_src/numpy/array_methods.py", line 737, in op return getattr(self.aval, f"_{name}")(self, *args) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "/Users/.pyenv/versions/pybamm-v24.5/lib/python3.12/site-packages/jax/_src/numpy/array_methods.py", line 265, in deferring_binary_op return binary_op(*args) ^^^^^^^^^^^^^^^^ File "/Users/.pyenv/versions/pybamm-v24.5/lib/python3.12/site-packages/jax/_src/numpy/ufuncs.py", line 86, in <lambda> fn = lambda x1, x2, /: lax_fn(*promote_args(numpy_fn.__name__, x1, x2)) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ TypeError: sub got incompatible shapes for broadcasting: (84, 84), (42, 42). ```

2024-09-23T12:19:12

0.0

[]

pybamm-team/PyBaMM

pybamm-team__PyBaMM-4300

5408ce3063a9d0f0e2a21c9deafc003703e531f6

diff --git a/CHANGELOG.md b/CHANGELOG.md index 044c85609b..1de6947420 100644 --- a/CHANGELOG.md +++ b/CHANGELOG.md @@ -4,6 +4,10 @@ - Added additional user-configurable options to the (`IDAKLUSolver`) and adjusted the default values to improve performance. ([#4282](https://github.com/pybamm-team/PyBaMM/pull/4282)) +## Bug Fixes + +- Fixed bug where IDAKLU solver failed when `output variables` were specified and an event triggered. ([#4300](https://github.com/pybamm-team/PyBaMM/pull/4300)) + # [v24.5](https://github.com/pybamm-team/PyBaMM/tree/v24.5) - 2024-07-26 ## Features diff --git a/pybamm/solvers/c_solvers/idaklu.cpp b/pybamm/solvers/c_solvers/idaklu.cpp index 3afed5faa8..3427c01853 100644 --- a/pybamm/solvers/c_solvers/idaklu.cpp +++ b/pybamm/solvers/c_solvers/idaklu.cpp @@ -187,5 +187,6 @@ PYBIND11_MODULE(idaklu, m) .def_readwrite("t", &Solution::t) .def_readwrite("y", &Solution::y) .def_readwrite("yS", &Solution::yS) + .def_readwrite("y_term", &Solution::y_term) .def_readwrite("flag", &Solution::flag); } diff --git a/pybamm/solvers/c_solvers/idaklu/IDAKLUSolverOpenMP.inl b/pybamm/solvers/c_solvers/idaklu/IDAKLUSolverOpenMP.inl index 340baa2d30..96ebb40d10 100644 --- a/pybamm/solvers/c_solvers/idaklu/IDAKLUSolverOpenMP.inl +++ b/pybamm/solvers/c_solvers/idaklu/IDAKLUSolverOpenMP.inl @@ -401,6 +401,7 @@ Solution IDAKLUSolverOpenMP<ExprSet>::solve( // set return vectors int length_of_return_vector = 0; + int length_of_final_sv_slice = 0; size_t max_res_size = 0; // maximum result size (for common result buffer) size_t max_res_dvar_dy = 0, max_res_dvar_dp = 0; if (functions->var_fcns.size() > 0) { @@ -414,6 +415,7 @@ Solution IDAKLUSolverOpenMP<ExprSet>::solve( for (auto& dvar_fcn : functions->dvar_dp_fcns) { max_res_dvar_dp = std::max(max_res_dvar_dp, size_t(dvar_fcn->out_shape(0))); } + length_of_final_sv_slice = number_of_states; } } else { // Return full y state-vector @@ -425,6 +427,7 @@ Solution IDAKLUSolverOpenMP<ExprSet>::solve( realtype *yS_return = new realtype[number_of_parameters * number_of_timesteps * length_of_return_vector]; + realtype *yterm_return = new realtype[length_of_final_sv_slice]; res.resize(max_res_size); res_dvar_dy.resize(max_res_dvar_dy); @@ -451,6 +454,13 @@ Solution IDAKLUSolverOpenMP<ExprSet>::solve( delete[] vect; } ); + py::capsule free_yterm_when_done( + yterm_return, + [](void *f) { + realtype *vect = reinterpret_cast<realtype *>(f); + delete[] vect; + } + ); // Initial state (t_i=0) int t_i = 0; @@ -518,6 +528,10 @@ Solution IDAKLUSolverOpenMP<ExprSet>::solve( t_i += 1; if (retval == IDA_SUCCESS || retval == IDA_ROOT_RETURN) { + if (functions->var_fcns.size() > 0) { + // store final state slice if outout variables are specified + yterm_return = yval; + } break; } } @@ -532,7 +546,7 @@ Solution IDAKLUSolverOpenMP<ExprSet>::solve( &y_return[0], free_y_when_done ); - // Note: Ordering of vector is differnet if computing variables vs returning + // Note: Ordering of vector is different if computing variables vs returning // the complete state vector np_array yS_ret; if (functions->var_fcns.size() > 0) { @@ -556,8 +570,13 @@ Solution IDAKLUSolverOpenMP<ExprSet>::solve( free_yS_when_done ); } + np_array y_term = np_array( + length_of_final_sv_slice, + &yterm_return[0], + free_yterm_when_done + ); - Solution sol(retval, t_ret, y_ret, yS_ret); + Solution sol(retval, t_ret, y_ret, yS_ret, y_term); if (solver_opts.print_stats) { long nsteps, nrevals, nlinsetups, netfails; diff --git a/pybamm/solvers/c_solvers/idaklu/Solution.hpp b/pybamm/solvers/c_solvers/idaklu/Solution.hpp index 92e22d02b6..ad0ea06762 100644 --- a/pybamm/solvers/c_solvers/idaklu/Solution.hpp +++ b/pybamm/solvers/c_solvers/idaklu/Solution.hpp @@ -12,8 +12,8 @@ class Solution /** * @brief Constructor */ - Solution(int retval, np_array t_np, np_array y_np, np_array yS_np) - : flag(retval), t(t_np), y(y_np), yS(yS_np) + Solution(int retval, np_array t_np, np_array y_np, np_array yS_np, np_array y_term_np) + : flag(retval), t(t_np), y(y_np), yS(yS_np), y_term(y_term_np) { } @@ -21,6 +21,7 @@ class Solution np_array t; np_array y; np_array yS; + np_array y_term; }; #endif // PYBAMM_IDAKLU_COMMON_HPP diff --git a/pybamm/solvers/c_solvers/idaklu/python.cpp b/pybamm/solvers/c_solvers/idaklu/python.cpp index 03090c9850..015f504086 100644 --- a/pybamm/solvers/c_solvers/idaklu/python.cpp +++ b/pybamm/solvers/c_solvers/idaklu/python.cpp @@ -478,8 +478,9 @@ Solution solve_python(np_array t_np, np_array y0_np, np_array yp0_np, std::vector<ptrdiff_t> {number_of_parameters, number_of_timesteps, number_of_states}, &yS_return[0] ); + np_array yterm_ret = np_array(0); - Solution sol(retval, t_ret, y_ret, yS_ret); + Solution sol(retval, t_ret, y_ret, yS_ret, yterm_ret); return sol; } diff --git a/pybamm/solvers/idaklu_solver.py b/pybamm/solvers/idaklu_solver.py index 8741a595f8..fa92563e3b 100644 --- a/pybamm/solvers/idaklu_solver.py +++ b/pybamm/solvers/idaklu_solver.py @@ -816,6 +816,7 @@ def _make_iree_function(self, fcn, *args, sparse_index=False): def fcn(*args): return fcn_inner(*args)[coo.row, coo.col] + elif coo.nnz != iree_fcn.numel: iree_fcn.nnz = iree_fcn.numel iree_fcn.col = list(range(iree_fcn.numel)) @@ -969,8 +970,10 @@ def _integrate(self, model, t_eval, inputs_dict=None): if self.output_variables: # Substitute empty vectors for state vector 'y' y_out = np.zeros((number_of_timesteps * number_of_states, 0)) + y_event = sol.y_term else: y_out = sol.y.reshape((number_of_timesteps, number_of_states)) + y_event = y_out[-1] # return sensitivity solution, we need to flatten yS to # (#timesteps * #states (where t is changing the quickest),) @@ -1000,7 +1003,7 @@ def _integrate(self, model, t_eval, inputs_dict=None): model, inputs_dict, np.array([t[-1]]), - np.transpose(y_out[-1])[:, np.newaxis], + np.transpose(y_event)[:, np.newaxis], termination, sensitivities=yS_out, ) diff --git a/pybamm/solvers/solution.py b/pybamm/solvers/solution.py index 39281ef4b0..c3c8451634 100644 --- a/pybamm/solvers/solution.py +++ b/pybamm/solvers/solution.py @@ -100,10 +100,6 @@ def __init__( self.sensitivities = sensitivities - self._t_event = t_event - self._y_event = y_event - self._termination = termination - # Check no ys are too large if check_solution: self.check_ys_are_not_too_large()

[Bug]: output variables option in idaklu solver fails when event triggered ### PyBaMM Version 24.1 ### Python Version 3.10 ### Describe the bug When the output_variables option of the IDAKLU solver is used in a simulation with an event, the solve fails with the error message below when trying to get the event reason ### Steps to Reproduce ```python def test(solver): model = pybamm.lithium_ion.DFN() parameter_values = pybamm.ParameterValues("Chen2020") sim = pybamm.Simulation( model, parameter_values=parameter_values, solver=solver, ) sol = sim.solve(np.linspace(0, 3600, 1000)) sol = sol["Terminal voltage [V]"].entries test(pybamm.IDAKLUSolver(output_variables=["Terminal voltage [V]"])) ``` ### Relevant log output ```shell Traceback (most recent call last): File "/home/mrobins/git/tmp/test.py", line 17, in <module> test(pybamm.IDAKLUSolver(output_variables=["Terminal voltage [V]"])) File "/home/mrobins/git/tmp/test.py", line 14, in test sol = sim.solve(np.linspace(0, 3600, 1000)) File "/home/mrobins/git/tmp/env/lib/python3.10/site-packages/pybamm/simulation.py", line 584, in solve self._solution = solver.solve(self.built_model, t_eval, **kwargs) File "/home/mrobins/git/tmp/env/lib/python3.10/site-packages/pybamm/solvers/base_solver.py", line 959, in solve solutions[i], termination = self.get_termination_reason( File "/home/mrobins/git/tmp/env/lib/python3.10/site-packages/pybamm/solvers/base_solver.py", line 1303, in get_termination_reason final_event_values[event.name] = event.expression.evaluate( File "/home/mrobins/git/tmp/env/lib/python3.10/site-packages/pybamm/expression_tree/binary_operators.py", line 132, in evaluate right = self.right.evaluate(t, y, y_dot, inputs) File "/home/mrobins/git/tmp/env/lib/python3.10/site-packages/pybamm/expression_tree/binary_operators.py", line 132, in evaluate right = self.right.evaluate(t, y, y_dot, inputs) File "/home/mrobins/git/tmp/env/lib/python3.10/site-packages/pybamm/expression_tree/binary_operators.py", line 132, in evaluate right = self.right.evaluate(t, y, y_dot, inputs) File "/home/mrobins/git/tmp/env/lib/python3.10/site-packages/pybamm/expression_tree/symbol.py", line 778, in evaluate return self._base_evaluate(t, y, y_dot, inputs) File "/home/mrobins/git/tmp/env/lib/python3.10/site-packages/pybamm/expression_tree/state_vector.py", line 284, in _base_evaluate raise ValueError( ValueError: y is too short, so value with slice is smaller than expected (env) ➜ tmp pip list |grep pybamm ```

2024-07-29T11:29:43

0.0

[]

pybamm-team/PyBaMM

pybamm-team__PyBaMM-4239

5e04abdd13fdebb1d8d01f7ba53c4bd02f8a7dfd

diff --git a/docs/source/examples/notebooks/getting_started/tutorial-5-run-experiments.ipynb b/docs/source/examples/notebooks/getting_started/tutorial-5-run-experiments.ipynb index eb82f59719..85be34e421 100644 --- a/docs/source/examples/notebooks/getting_started/tutorial-5-run-experiments.ipynb +++ b/docs/source/examples/notebooks/getting_started/tutorial-5-run-experiments.ipynb @@ -25,18 +25,8 @@ "name": "stdout", "output_type": "stream", "text": [ - "\n", - "\u001b[1m[\u001b[0m\u001b[34;49mnotice\u001b[0m\u001b[1;39;49m]\u001b[0m\u001b[39;49m A new release of pip is available: \u001b[0m\u001b[31;49m23.3.1\u001b[0m\u001b[39;49m -> \u001b[0m\u001b[32;49m24.0\u001b[0m\n", - "\u001b[1m[\u001b[0m\u001b[34;49mnotice\u001b[0m\u001b[1;39;49m]\u001b[0m\u001b[39;49m To update, run: \u001b[0m\u001b[32;49mpip install --upgrade pip\u001b[0m\n", "Note: you may need to restart the kernel to use updated packages.\n" ] - }, - { - "name": "stderr", - "output_type": "stream", - "text": [ - "An NVIDIA GPU may be present on this machine, but a CUDA-enabled jaxlib is not installed. Falling back to cpu.\n" - ] } ], "source": [ @@ -163,18 +153,19 @@ "name": "stderr", "output_type": "stream", "text": [ - "At t = 522.66 and h = 1.1556e-13, the corrector convergence failed repeatedly or with |h| = hmin.\n" + "At t = 339.952 and h = 1.4337e-18, the corrector convergence failed repeatedly or with |h| = hmin.\n", + "At t = 522.687 and h = 4.04917e-14, the corrector convergence failed repeatedly or with |h| = hmin.\n" ] }, { "data": { "application/vnd.jupyter.widget-view+json": { - "model_id": "5ab1f22de6af4878b6ca43d27ffc01c5", + "model_id": "93feca98298f4111909ae487e2a1e273", "version_major": 2, "version_minor": 0 }, "text/plain": [ - "interactive(children=(FloatSlider(value=0.0, description='t', max=40.132949019384355, step=0.40132949019384356…" + "interactive(children=(FloatSlider(value=0.0, description='t', max=40.13268704803602, step=0.4013268704803602),…" ] }, "metadata": {}, @@ -183,7 +174,7 @@ { "data": { "text/plain": [ - "<pybamm.plotting.quick_plot.QuickPlot at 0x7f1a11f76690>" + "<pybamm.plotting.quick_plot.QuickPlot at 0x16520e690>" ] }, "execution_count": 5, @@ -211,12 +202,12 @@ { "data": { "application/vnd.jupyter.widget-view+json": { - "model_id": "7cdac234d74241a2814918053454f6a6", + "model_id": "4d6e43032f4e4aa6be5843c4916b4b50", "version_major": 2, "version_minor": 0 }, "text/plain": [ - "interactive(children=(FloatSlider(value=0.0, description='t', max=13.076977041121545, step=0.13076977041121546…" + "interactive(children=(FloatSlider(value=0.0, description='t', max=13.076887099589111, step=0.1307688709958911)…" ] }, "metadata": {}, @@ -225,7 +216,7 @@ { "data": { "text/plain": [ - "<pybamm.plotting.quick_plot.QuickPlot at 0x7f1a105ecb50>" + "<pybamm.plotting.quick_plot.QuickPlot at 0x1696cf290>" ] }, "execution_count": 6, @@ -255,7 +246,7 @@ { "data": { "text/plain": [ - "_Step(C-rate, 1.0, duration=1 hour, period=1 minute, temperature=25oC, tags=['tag1'], description=Discharge at 1C for 1 hour)" + "Step(1.0, duration=1 hour, period=1 minute, temperature=25oC, tags=['tag1'], description=Discharge at 1C for 1 hour, direction=Discharge)" ] }, "execution_count": 7, @@ -293,7 +284,7 @@ { "data": { "text/plain": [ - "_Step(current, 1, duration=1 hour, termination=2.5 V)" + "Step(1, duration=1 hour, termination=2.5 V, direction=Discharge)" ] }, "execution_count": 8, @@ -321,7 +312,7 @@ { "data": { "text/plain": [ - "_Step(current, 1.0, duration=1 hour, termination=2.5V, description=Discharge at 1A for 1 hour or until 2.5V)" + "Step(1.0, duration=1 hour, termination=2.5V, description=Discharge at 1A for 1 hour or until 2.5V, direction=Discharge)" ] }, "execution_count": 9, @@ -348,10 +339,78 @@ "execution_count": 10, "metadata": {}, "outputs": [ + { + "name": "stderr", + "output_type": "stream", + "text": [ + "2024-07-10 14:41:02.625 - [WARNING] callbacks.on_experiment_infeasible_time(240): \n", + "\n", + "\tExperiment is infeasible: default duration (1.0 seconds) was reached during 'Step([[ 0.00000000e+00 0.00000000e+00]\n", + " [ 1.69491525e-02 5.31467428e-02]\n", + " [ 3.38983051e-02 1.05691312e-01]\n", + " [ 5.08474576e-02 1.57038356e-01]\n", + " [ 6.77966102e-02 2.06606093e-01]\n", + " [ 8.47457627e-02 2.53832900e-01]\n", + " [ 1.01694915e-01 2.98183679e-01]\n", + " [ 1.18644068e-01 3.39155918e-01]\n", + " [ 1.35593220e-01 3.76285385e-01]\n", + " [ 1.52542373e-01 4.09151388e-01]\n", + " [ 1.69491525e-01 4.37381542e-01]\n", + " [ 1.86440678e-01 4.60655989e-01]\n", + " [ 2.03389831e-01 4.78711019e-01]\n", + " [ 2.20338983e-01 4.91342062e-01]\n", + " [ 2.37288136e-01 4.98406004e-01]\n", + " [ 2.54237288e-01 4.99822806e-01]\n", + " [ 2.71186441e-01 4.95576416e-01]\n", + " [ 2.88135593e-01 4.85714947e-01]\n", + " [ 3.05084746e-01 4.70350133e-01]\n", + " [ 3.22033898e-01 4.49656065e-01]\n", + " [ 3.38983051e-01 4.23867214e-01]\n", + " [ 3.55932203e-01 3.93275778e-01]\n", + " [ 3.72881356e-01 3.58228370e-01]\n", + " [ 3.89830508e-01 3.19122092e-01]\n", + " [ 4.06779661e-01 2.76400033e-01]\n", + " [ 4.23728814e-01 2.30546251e-01]\n", + " [ 4.40677966e-01 1.82080288e-01]\n", + " [ 4.57627119e-01 1.31551282e-01]\n", + " [ 4.74576271e-01 7.95317480e-02]\n", + " [ 4.91525424e-01 2.66110874e-02]\n", + " [ 5.08474576e-01 -2.66110874e-02]\n", + " [ 5.25423729e-01 -7.95317480e-02]\n", + " [ 5.42372881e-01 -1.31551282e-01]\n", + " [ 5.59322034e-01 -1.82080288e-01]\n", + " [ 5.76271186e-01 -2.30546251e-01]\n", + " [ 5.93220339e-01 -2.76400033e-01]\n", + " [ 6.10169492e-01 -3.19122092e-01]\n", + " [ 6.27118644e-01 -3.58228370e-01]\n", + " [ 6.44067797e-01 -3.93275778e-01]\n", + " [ 6.61016949e-01 -4.23867214e-01]\n", + " [ 6.77966102e-01 -4.49656065e-01]\n", + " [ 6.94915254e-01 -4.70350133e-01]\n", + " [ 7.11864407e-01 -4.85714947e-01]\n", + " [ 7.28813559e-01 -4.95576416e-01]\n", + " [ 7.45762712e-01 -4.99822806e-01]\n", + " [ 7.62711864e-01 -4.98406004e-01]\n", + " [ 7.79661017e-01 -4.91342062e-01]\n", + " [ 7.96610169e-01 -4.78711019e-01]\n", + " [ 8.13559322e-01 -4.60655989e-01]\n", + " [ 8.30508475e-01 -4.37381542e-01]\n", + " [ 8.47457627e-01 -4.09151388e-01]\n", + " [ 8.64406780e-01 -3.76285385e-01]\n", + " [ 8.81355932e-01 -3.39155918e-01]\n", + " [ 8.98305085e-01 -2.98183679e-01]\n", + " [ 9.15254237e-01 -2.53832900e-01]\n", + " [ 9.32203390e-01 -2.06606093e-01]\n", + " [ 9.49152542e-01 -1.57038356e-01]\n", + " [ 9.66101695e-01 -1.05691312e-01]\n", + " [ 9.83050847e-01 -5.31467428e-02]\n", + " [ 1.00000000e+00 -1.22464680e-16]], duration=1.0, period=0.016949152542372836, direction=Rest)'. The returned solution only contains up to step 1 of cycle 1. Please specify a duration in the step instructions.\n" + ] + }, { "data": { "application/vnd.jupyter.widget-view+json": { - "model_id": "730d5e19b17e447ebde5679de68c46ef", + "model_id": "6364b4579fc447e2a607f2f8414172ba", "version_major": 2, "version_minor": 0 }, @@ -365,7 +424,7 @@ { "data": { "text/plain": [ - "<pybamm.plotting.quick_plot.QuickPlot at 0x7f1a1a4c2ed0>" + "<pybamm.plotting.quick_plot.QuickPlot at 0x16a1d8d90>" ] }, "execution_count": 10, @@ -419,13 +478,14 @@ "output_type": "stream", "text": [ "[1] Joel A. E. Andersson, Joris Gillis, Greg Horn, James B. Rawlings, and Moritz Diehl. CasADi – A software framework for nonlinear optimization and optimal control. Mathematical Programming Computation, 11(1):1–36, 2019. doi:10.1007/s12532-018-0139-4.\n", - "[2] Marc Doyle, Thomas F. Fuller, and John Newman. Modeling of galvanostatic charge and discharge of the lithium/polymer/insertion cell. Journal of the Electrochemical society, 140(6):1526–1533, 1993. doi:10.1149/1.2221597.\n", - "[3] Charles R. Harris, K. Jarrod Millman, Stéfan J. van der Walt, Ralf Gommers, Pauli Virtanen, David Cournapeau, Eric Wieser, Julian Taylor, Sebastian Berg, Nathaniel J. Smith, and others. Array programming with NumPy. Nature, 585(7825):357–362, 2020. doi:10.1038/s41586-020-2649-2.\n", - "[4] Scott G. Marquis, Valentin Sulzer, Robert Timms, Colin P. Please, and S. Jon Chapman. An asymptotic derivation of a single particle model with electrolyte. Journal of The Electrochemical Society, 166(15):A3693–A3706, 2019. doi:10.1149/2.0341915jes.\n", - "[5] Peyman Mohtat, Suhak Lee, Jason B Siegel, and Anna G Stefanopoulou. Towards better estimability of electrode-specific state of health: decoding the cell expansion. Journal of Power Sources, 427:101–111, 2019.\n", - "[6] Valentin Sulzer, Scott G. Marquis, Robert Timms, Martin Robinson, and S. Jon Chapman. Python Battery Mathematical Modelling (PyBaMM). Journal of Open Research Software, 9(1):14, 2021. doi:10.5334/jors.309.\n", - "[7] Pauli Virtanen, Ralf Gommers, Travis E. Oliphant, Matt Haberland, Tyler Reddy, David Cournapeau, Evgeni Burovski, Pearu Peterson, Warren Weckesser, Jonathan Bright, and others. SciPy 1.0: fundamental algorithms for scientific computing in Python. Nature Methods, 17(3):261–272, 2020. doi:10.1038/s41592-019-0686-2.\n", - "[8] Andrew Weng, Jason B Siegel, and Anna Stefanopoulou. Differential voltage analysis for battery manufacturing process control. arXiv preprint arXiv:2303.07088, 2023.\n", + "[2] Von DAG Bruggeman. Berechnung verschiedener physikalischer konstanten von heterogenen substanzen. i. dielektrizitätskonstanten und leitfähigkeiten der mischkörper aus isotropen substanzen. Annalen der physik, 416(7):636–664, 1935.\n", + "[3] Marc Doyle, Thomas F. Fuller, and John Newman. Modeling of galvanostatic charge and discharge of the lithium/polymer/insertion cell. Journal of the Electrochemical society, 140(6):1526–1533, 1993. doi:10.1149/1.2221597.\n", + "[4] Charles R. Harris, K. Jarrod Millman, Stéfan J. van der Walt, Ralf Gommers, Pauli Virtanen, David Cournapeau, Eric Wieser, Julian Taylor, Sebastian Berg, Nathaniel J. Smith, and others. Array programming with NumPy. Nature, 585(7825):357–362, 2020. doi:10.1038/s41586-020-2649-2.\n", + "[5] Scott G. Marquis, Valentin Sulzer, Robert Timms, Colin P. Please, and S. Jon Chapman. An asymptotic derivation of a single particle model with electrolyte. Journal of The Electrochemical Society, 166(15):A3693–A3706, 2019. doi:10.1149/2.0341915jes.\n", + "[6] Peyman Mohtat, Suhak Lee, Jason B Siegel, and Anna G Stefanopoulou. Towards better estimability of electrode-specific state of health: decoding the cell expansion. Journal of Power Sources, 427:101–111, 2019.\n", + "[7] Valentin Sulzer, Scott G. Marquis, Robert Timms, Martin Robinson, and S. Jon Chapman. Python Battery Mathematical Modelling (PyBaMM). Journal of Open Research Software, 9(1):14, 2021. doi:10.5334/jors.309.\n", + "[8] Pauli Virtanen, Ralf Gommers, Travis E. Oliphant, Matt Haberland, Tyler Reddy, David Cournapeau, Evgeni Burovski, Pearu Peterson, Warren Weckesser, Jonathan Bright, and others. SciPy 1.0: fundamental algorithms for scientific computing in Python. Nature Methods, 17(3):261–272, 2020. doi:10.1038/s41592-019-0686-2.\n", + "[9] Andrew Weng, Jason B Siegel, and Anna Stefanopoulou. Differential voltage analysis for battery manufacturing process control. arXiv preprint arXiv:2303.07088, 2023.\n", "\n" ] } diff --git a/pybamm/callbacks.py b/pybamm/callbacks.py index 57f2426d0f..4e8c67c8be 100644 --- a/pybamm/callbacks.py +++ b/pybamm/callbacks.py @@ -36,37 +36,37 @@ def on_experiment_start(self, logs): """ Called at the start of an experiment simulation. """ - pass + pass # pragma: no cover def on_cycle_start(self, logs): """ Called at the start of each cycle in an experiment simulation. """ - pass + pass # pragma: no cover def on_step_start(self, logs): """ Called at the start of each step in an experiment simulation. """ - pass + pass # pragma: no cover def on_step_end(self, logs): """ Called at the end of each step in an experiment simulation. """ - pass + pass # pragma: no cover def on_cycle_end(self, logs): """ Called at the end of each cycle in an experiment simulation. """ - pass + pass # pragma: no cover def on_experiment_end(self, logs): """ Called at the end of an experiment simulation. """ - pass + pass # pragma: no cover def on_experiment_error(self, logs): """ @@ -75,13 +75,19 @@ def on_experiment_error(self, logs): For example, this could be used to send an error alert with a bug report when running batch simulations in the cloud. """ - pass + pass # pragma: no cover - def on_experiment_infeasible(self, logs): + def on_experiment_infeasible_time(self, logs): """ - Called when an experiment simulation is infeasible. + Called when an experiment simulation is infeasible due to reaching maximum time. """ - pass + pass # pragma: no cover + + def on_experiment_infeasible_event(self, logs): + """ + Called when an experiment simulation is infeasible due to an event. + """ + pass # pragma: no cover ######################################################################################## @@ -226,7 +232,19 @@ def on_experiment_error(self, logs): error = logs["error"] pybamm.logger.error(f"Simulation error: {error}") - def on_experiment_infeasible(self, logs): + def on_experiment_infeasible_time(self, logs): + duration = logs["step duration"] + cycle_num = logs["cycle number"][0] + step_num = logs["step number"][0] + operating_conditions = logs["step operating conditions"] + self.logger.warning( + f"\n\n\tExperiment is infeasible: default duration ({duration} seconds) " + f"was reached during '{operating_conditions}'. The returned solution only " + f"contains up to step {step_num} of cycle {cycle_num}. " + "Please specify a duration in the step instructions." + ) + + def on_experiment_infeasible_event(self, logs): termination = logs["termination"] cycle_num = logs["cycle number"][0] step_num = logs["step number"][0] diff --git a/pybamm/experiment/step/base_step.py b/pybamm/experiment/step/base_step.py index 16224a6afa..6b77bed2cf 100644 --- a/pybamm/experiment/step/base_step.py +++ b/pybamm/experiment/step/base_step.py @@ -71,6 +71,8 @@ def __init__( description=None, direction=None, ): + self.input_duration = duration + self.input_value = value # Check if drive cycle is_drive_cycle = isinstance(value, np.ndarray) is_python_function = callable(value) @@ -100,8 +102,11 @@ def __init__( f"Input function must return a real number output at t = {t0}" ) + # Record whether the step uses the default duration + # This will be used by the experiment to check whether the step is feasible + self.uses_default_duration = duration is None # Set duration - if duration is None: + if self.uses_default_duration: duration = self.default_duration(value) self.duration = _convert_time_to_seconds(duration) @@ -195,8 +200,8 @@ def copy(self): A copy of the step. """ return self.__class__( - self.value, - duration=self.duration, + self.input_value, + duration=self.input_duration, termination=self.termination, period=self.period, temperature=self.temperature, @@ -259,7 +264,7 @@ def default_duration(self, value): t = value[:, 0] return t[-1] else: - return 24 * 3600 # 24 hours in seconds + return 24 * 3600 # one day in seconds def process_model(self, model, parameter_values): new_model = model.new_copy() @@ -411,10 +416,16 @@ def set_up(self, new_model, new_parameter_values): def _convert_time_to_seconds(time_and_units): """Convert a time in seconds, minutes or hours to a time in seconds""" - # If the time is a number, assume it is in seconds - if isinstance(time_and_units, numbers.Number) or time_and_units is None: + if time_and_units is None: return time_and_units + # If the time is a number, assume it is in seconds + if isinstance(time_and_units, numbers.Number): + if time_and_units <= 0: + raise ValueError("time must be positive") + else: + return time_and_units + # Split number and units units = time_and_units.lstrip("0123456789.- ") time = time_and_units[: -len(units)] diff --git a/pybamm/experiment/step/steps.py b/pybamm/experiment/step/steps.py index e3322104bf..e66178dc81 100644 --- a/pybamm/experiment/step/steps.py +++ b/pybamm/experiment/step/steps.py @@ -156,6 +156,11 @@ def __init__(self, value, **kwargs): def current_value(self, variables): return self.value * pybamm.Parameter("Nominal cell capacity [A.h]") + def default_duration(self, value): + # "value" is C-rate, so duration is "1 / value" hours in seconds + # with a 2x safety factor + return 1 / abs(value) * 3600 * 2 + def c_rate(value, **kwargs): """ diff --git a/pybamm/simulation.py b/pybamm/simulation.py index 8ec85d67d4..a55310870e 100644 --- a/pybamm/simulation.py +++ b/pybamm/simulation.py @@ -679,6 +679,7 @@ def solve( logs["step number"] = (step_num, cycle_length) logs["step operating conditions"] = step_str + logs["step duration"] = step.duration callbacks.on_step_start(logs) inputs = { @@ -767,23 +768,33 @@ def solve( callbacks.on_step_end(logs) logs["termination"] = step_solution.termination - # Only allow events specified by experiment - if not ( + + # Check for some cases that would make the experiment end early + if step_termination == "final time" and step.uses_default_duration: + # reached the default duration of a step (typically we should + # reach an event before the default duration) + callbacks.on_experiment_infeasible_time(logs) + feasible = False + break + + elif not ( isinstance(step_solution, pybamm.EmptySolution) or step_termination == "final time" or "[experiment]" in step_termination ): - callbacks.on_experiment_infeasible(logs) + # Step has reached an event that is not specified in the + # experiment + callbacks.on_experiment_infeasible_event(logs) feasible = False break - if time_stop is not None: - max_time = cycle_solution.t[-1] - if max_time >= time_stop: - break + elif time_stop is not None and logs["experiment time"] >= time_stop: + # reached the time limit of the experiment + break - # Increment index for next iteration - idx += 1 + else: + # Increment index for next iteration, then continue + idx += 1 if save_this_cycle or feasible is False: self._solution = self._solution + cycle_solution

[Bug]: [24.5rc0] Simulations with `pybamm.Experiment` terminate after 24 h ### PyBaMM Version 24.5rc0 ### Python Version 3.9.13 ### Describe the bug Simulations using `pybamm.Experiment` with a voltage cut-off only appear to terminate with a time cut-off after 24 h. I'm afraid I haven't had time to get into logs to figure out why this is happening! ### Steps to Reproduce The following simulation should proceed for approximately 100 h = 360000 s, until the specified voltage cut-off. However, it terminates after 24 h. ```python import pybamm parameter_values = pybamm.ParameterValues("Chen2020") model = pybamm.lithium_ion.DFN() experiment = pybamm.Experiment( [ "Discharge at 0.01C until 2.5 V", ] ) sim = pybamm.Simulation( model, parameter_values=parameter_values, experiment=experiment, ) sol = sim.solve() sim.plot() print(f'Last time step at {sim.solution["Time [s]"].entries[-1] / 3600} h') ``` ### Relevant log output _No response_

Hi, @ejfdickinson – looks like a duplicate of https://github.com/pybamm-team/PyBaMM/issues/4155? @agriyakhetarpal Yes, it is. v24.1 does not do this, so it should be in your release notes for v24.5rc0. Any recommendations on how to work around? > Any recommendations on how to work around? @ejfdickinson you can include a duration in the experiment step, e.g., ```python experiment = pybamm.Experiment( [ "Discharge at 0.01C for 100 hours or until 2.5 V", ] ) ``` and it won't cut off at 24 hours. <img width="1484" alt="image" src="https://github.com/pybamm-team/PyBaMM/assets/34451391/cb30fa0d-f333-464d-951a-4dc644437e0a"> Thanks @MarcBerliner ! > v24.1 does not do this, so it should be in your release notes for v24.5rc0. I think we should mark this as a breaking change, @valentinsulzer? IMO we should fix this before 24.5 proper. I don't know what logic changed or if there was a longer default before. Update: the logic used to be ```python3 if op_conds.type == "current": # Current control: max simulation time: 3h / C-rate Crate = op_conds.value / capacity dt = 3 / abs(Crate) * 3600 # seconds else: # max simulation time: 1 day dt = 24 * 3600 # seconds ``` Alternatively, should we require a duration to avoid unexpected behaviour? Why did we move away from that logic? As in, if we bring it back will it break anything? It got moved to the `Step` class instead of the `Simulation` class. It's easy to put the same logic back in for C-rate, but not current, as the default duration needs to return a float, not a symbol (see the commit above). It's fixable but needs a bit of work, and I'm leaning towards requiring a duration as it's less ambiguous. It would be a breaking change but easy for users to fix. @rtimms For our use cases, a maximum duration would be fine, so long as it's declared on simulation exit like a `"Time"` stop rather than occurring silently. As noted in #4155 , 24 h is well within typical simulation times for normal experiments, so if the purpose is just to avoid hanging the solver due to an absurd duration, adding a few orders of magnitude to the default might be more sensible!

2024-07-03T17:32:40

0.0

[]

pybamm-team/PyBaMM

pybamm-team__PyBaMM-4221

b50dfa595dd93a2943aa2a5a8aa14ac9468aea8e

diff --git a/pybamm/models/full_battery_models/base_battery_model.py b/pybamm/models/full_battery_models/base_battery_model.py index fb8d001f93..8a2e443338 100644 --- a/pybamm/models/full_battery_models/base_battery_model.py +++ b/pybamm/models/full_battery_models/base_battery_model.py @@ -633,6 +633,20 @@ def __init__(self, extra_options): "be 'none': 'particle mechanics', 'loss of active material'" ) + if "true" in options["SEI on cracks"]: + sei_on_cr = options["SEI on cracks"] + p_mechanics = options["particle mechanics"] + if isinstance(p_mechanics, str) and isinstance(sei_on_cr, tuple): + p_mechanics = (p_mechanics, p_mechanics) + if any( + sei == "true" and mech != "swelling and cracking" + for mech, sei in zip(p_mechanics, sei_on_cr) + ): + raise pybamm.OptionError( + "If 'SEI on cracks' is 'true' then 'particle mechanics' must be " + "'swelling and cracking'." + ) + # Check options are valid for option, value in options.items(): if isinstance(value, str) or option in [

Raise OptionError if "SEI on cracks" is "true" but cracking isn't enabled. If you do e.g. ```python3 model = pybamm.lithium_ion.SPMe( { "SEI": "solvent-diffusion limited", "SEI porosity change": "true", "lithium plating": "partially reversible", "lithium plating porosity change": "true", # alias for "SEI porosity change" "particle mechanics": ("swelling only", "swelling only"), "SEI on cracks": "true", "loss of active material": "stress-driven", } ) ``` you get a `ZeroDivisionError`. It doesn't make sense to have SEI n cracks if there are no cracks, so we should catch this and raise an `OptionError`

As an alternative the cracking could be turned on if the "SEI on cracks" is enabled. Going by the name of the parameter, I think I would expect that it also enables cracking

2024-06-26T16:48:05

0.0

[]

pybamm-team/PyBaMM

pybamm-team__PyBaMM-4208

9b5766921af1f6d192e8b1974343eb7ccbd72efe

diff --git a/pybamm/parameters/parameter_values.py b/pybamm/parameters/parameter_values.py index c8011c735c..57ebf65058 100644 --- a/pybamm/parameters/parameter_values.py +++ b/pybamm/parameters/parameter_values.py @@ -65,7 +65,10 @@ def __init__(self, values, chemistry=None): values.pop("chemistry", None) self.update(values, check_already_exists=False) else: - raise ValueError("Invalid Parameter Value") + valid_sets = "\n".join(pybamm.parameter_sets.keys()) + raise ValueError( + f"'{values}' is not a valid parameter set. Parameter set must be one of:\n{valid_sets}" + ) # Initialise empty _processed_symbols dict (for caching) self._processed_symbols = {}

Better error on unrecognised parameter set Doing `pybamm.ParameterValues("bad")` gives `ValueError: Invalid Parameter Value`. It would be better to say something like `ValueError: 'bad' is not a recognised parameter set` and possibly list the valid options (including custom parameter sets that have been registered as entry points).

2024-06-24T15:06:53

0.0

[]

pybamm-team/PyBaMM

pybamm-team__PyBaMM-4092

33d9ea7f28c26fcbe16461857adbada1349fe0ff

diff --git a/.github/workflows/publish_pypi.yml b/.github/workflows/publish_pypi.yml index fb4a203e5d..01bf7ad95f 100644 --- a/.github/workflows/publish_pypi.yml +++ b/.github/workflows/publish_pypi.yml @@ -99,7 +99,7 @@ jobs: strategy: fail-fast: false matrix: - os: [ubuntu-latest, macos-12] + os: [ubuntu-latest, macos-13] steps: - uses: actions/checkout@v4 name: Check out PyBaMM repository @@ -112,15 +112,6 @@ jobs: - name: Clone pybind11 repo (no history) run: git clone --depth 1 --branch v2.11.1 https://github.com/pybind/pybind11.git - # sometimes gfortran cannot be found, so reinstall gcc just to be sure - - name: Install SuiteSparse and SUNDIALS on macOS - if: matrix.os == 'macos-12' - run: | - brew install graphviz libomp - brew reinstall gcc - python -m pip install cmake wget - python scripts/install_KLU_Sundials.py - - name: Build wheels on Linux run: pipx run cibuildwheel --output-dir wheelhouse if: matrix.os == 'ubuntu-latest' @@ -134,11 +125,50 @@ jobs: CIBW_TEST_COMMAND: python -c "import pybamm; pybamm.IDAKLUSolver()" - name: Build wheels on macOS amd64 - if: matrix.os == 'macos-12' + if: matrix.os == 'macos-13' run: pipx run cibuildwheel --output-dir wheelhouse env: - CIBW_BEFORE_BUILD_MACOS: > - python -m pip install --upgrade cmake casadi setuptools wheel + MACOSX_DEPLOYMENT_TARGET: 11.0 + # Sourced from + # License: BSD-3-Clause + # https://github.com/scipy/scipy/blob/f2d4775e7762fad984f8f0acd8227c725ff21630/tools/wheels/cibw_before_build_macos.sh#L23-L49 + CIBW_BEFORE_ALL_MACOS: | + set -e -x + + # download gfortran with proper macos minimum version (11.0) + curl -L https://github.com/isuruf/gcc/releases/download/gcc-11.3.0-2/gfortran-darwin-x86_64-native.tar.gz -o gfortran.tar.gz + + GFORTRAN_SHA256=$(shasum --algorithm 256 gfortran.tar.gz) + KNOWN_SHA256="981367dd0ad4335613e91bbee453d60b6669f5d7e976d18c7bdb7f1966f26ae4 gfortran.tar.gz" + if [ "$GFORTRAN_SHA256" != "$KNOWN_SHA256" ]; then + echo "SHA256 mismatch for gfortran.tar.gz" + echo "expected: $KNOWN_SHA256" + echo "got: $GFORTRAN_SHA256" + exit 1 + fi + + mkdir -p gfortran_installed/ + tar -xv -C gfortran_installed/ -f gfortran.tar.gz + + export FC=$(pwd)/gfortran_installed/gfortran-darwin-x86_64-native/bin/gfortran + export PATH=$(pwd)/gfortran_installed/gfortran-darwin-x86_64-native/bin:$PATH + + # link libgfortran.5.dylib, libgfortran.dylib, libquadmath.0.dylib, libquadmath.dylib, libgcc_s.1.dylib, libgcc_s.1.1.dylib + # and place them in $HOME/.local/lib, and then change rpath + # to $HOME/.local/lib + + mkdir -p $HOME/.local/lib + lib_dir=$(pwd)/gfortran_installed/gfortran-darwin-x86_64-native/lib + for lib in libgfortran.5.dylib libgfortran.dylib libquadmath.0.dylib libquadmath.dylib libgcc_s.1.dylib libgcc_s.1.1.dylib; do + cp $lib_dir/$lib $HOME/.local/lib/ + install_name_tool -id $HOME/.local/lib/$lib $HOME/.local/lib/$lib + done + + export SDKROOT=${SDKROOT:-$(xcrun --show-sdk-path)} + + python -m pip install cmake wget + python scripts/install_KLU_Sundials.py + CIBW_BEFORE_BUILD_MACOS: python -m pip install --upgrade pip cmake casadi setuptools wheel delocate CIBW_REPAIR_WHEEL_COMMAND_MACOS: delocate-listdeps {wheel} && delocate-wheel -v -w {dest_dir} {wheel} CIBW_TEST_COMMAND: python -c "import pybamm; pybamm.IDAKLUSolver()" @@ -162,18 +192,57 @@ jobs: - name: Clone pybind11 repo (no history) run: git clone --depth 1 --branch v2.11.1 https://github.com/pybind/pybind11.git - - name: Install SuiteSparse and SUNDIALS on macOS - run: | - brew install graphviz libomp - brew reinstall gcc - python -m pip install cmake pipx - python scripts/install_KLU_Sundials.py - - name: Build wheels on macOS arm64 - run: python -m pipx run cibuildwheel --output-dir wheelhouse + run: | + python -m pip install cibuildwheel + python -m cibuildwheel --output-dir wheelhouse env: - CIBW_BEFORE_BUILD: python -m pip install cmake casadi setuptools wheel delocate - CIBW_REPAIR_WHEEL_COMMAND: delocate-listdeps {wheel} && delocate-wheel -v -w {dest_dir} {wheel} + MACOSX_DEPLOYMENT_TARGET: 11.0 + # Sourced from + # License: BSD-3-Clause + # https://github.com/scipy/scipy/blob/f2d4775e7762fad984f8f0acd8227c725ff21630/tools/wheels/cibw_before_build_macos.sh#L23-L49 + CIBW_BEFORE_ALL_MACOS: | + set -e -x + + # download gfortran with proper macos minimum version (11.0) + curl -L https://github.com/isuruf/gcc/releases/download/gcc-11.3.0-2/gfortran-darwin-arm64-native.tar.gz -o gfortran.tar.gz + + GFORTRAN_SHA256=$(shasum --algorithm 256 gfortran.tar.gz) + KNOWN_SHA256="84364eee32ba843d883fb8124867e2bf61a0cd73b6416d9897ceff7b85a24604 gfortran.tar.gz" + if [ "$GFORTRAN_SHA256" != "$KNOWN_SHA256" ]; then + echo "SHA256 mismatch for gfortran.tar.gz" + echo "expected: $KNOWN_SHA256" + echo "got: $GFORTRAN_SHA256" + exit 1 + fi + + mkdir -p gfortran_installed/ + tar -xv -C gfortran_installed/ -f gfortran.tar.gz + + export FC=$(pwd)/gfortran_installed/gfortran-darwin-arm64-native/bin/gfortran + export PATH=$(pwd)/gfortran_installed/gfortran-darwin-arm64-native/bin:$PATH + + # link libgfortran.5.dylib, libgfortran.dylib, libquadmath.0.dylib, libquadmath.dylib, libgcc_s.1.1.dylib + # and place them in $HOME/.local/lib, and then change rpath + # note: libgcc_s.1.dylib not present for arm64, skip for now + # to $HOME/.local/lib + + mkdir -p $HOME/.local/lib + lib_dir=$(pwd)/gfortran_installed/gfortran-darwin-arm64-native/lib + for lib in libgfortran.5.dylib libgfortran.dylib libquadmath.0.dylib libquadmath.dylib libgcc_s.1.1.dylib; do + cp $lib_dir/$lib $HOME/.local/lib/ + install_name_tool -id $HOME/.local/lib/$lib $HOME/.local/lib/$lib + done + + export SDKROOT=${SDKROOT:-$(xcrun --show-sdk-path)} + + python -m pip install cmake wget + python scripts/install_KLU_Sundials.py + CIBW_BEFORE_BUILD_MACOS: python -m pip install --upgrade pip cmake casadi setuptools wheel delocate + # Use higher macOS target for now: https://github.com/casadi/casadi/issues/3698 + CIBW_REPAIR_WHEEL_COMMAND_MACOS: | + delocate-listdeps {wheel} && delocate-wheel -v -w {dest_dir} {wheel} --require-target-macos-version 11.1 + for file in {dest_dir}/*.whl; do mv "$file" "${file//macosx_11_1/macosx_11_0}"; done CIBW_TEST_COMMAND: python -c "import pybamm; pybamm.IDAKLUSolver()" - name: Upload wheels for macOS arm64 diff --git a/.gitignore b/.gitignore index 03750e18b2..6f1f201617 100644 --- a/.gitignore +++ b/.gitignore @@ -137,3 +137,7 @@ results/ # tests test_callback.log + +# openmp downloads +install_KLU_Sundials/openmp-* +install_KLU_Sundials/usr/ diff --git a/.lycheeignore b/.lycheeignore index 338303b4ba..1543b69f03 100644 --- a/.lycheeignore +++ b/.lycheeignore @@ -1,6 +1,7 @@ # a list of links/files to be ignored by lychee link checker (see workflow file) https://github.com/DrTimothyAldenDavis/SuiteSparse/archive/v%7BSUITESPARSE_VERSION%7D.tar.gz https://github.com/LLNL/sundials/releases/download/v%7BSUNDIALS_VERSION%7D/sundials-%7BSUNDIALS_VERSION%7D.tar.gz +https://mac.r-project.org/openmp/openmp-%7BOPENMP_VERSION%7D-darwin20-Release.tar.gz # Errors in docs/source/user_guide/getting_started.md file:///home/runner/work/PyBaMM/PyBaMM/docs/source/user_guide/api_docs diff --git a/docs/source/user_guide/installation/install-from-source.rst b/docs/source/user_guide/installation/install-from-source.rst index ea664b4a5b..a2a7637c9e 100644 --- a/docs/source/user_guide/installation/install-from-source.rst +++ b/docs/source/user_guide/installation/install-from-source.rst @@ -43,11 +43,11 @@ You can install the above with Where ``X`` is the version sub-number. -.. tab:: MacOS +.. tab:: macOS .. code:: bash - brew install python openblas gcc gfortran graphviz libomp cmake pandoc + brew install python openblas gcc gfortran graphviz cmake pandoc .. note:: @@ -82,8 +82,9 @@ If you are running windows, you can simply skip this section and jump to :ref:`p # in the PyBaMM/ directory nox -s pybamm-requires -This will download, compile and install the SuiteSparse and SUNDIALS libraries. -Both libraries are installed in ``~/.local``. +This will download, compile and install the SuiteSparse and SUNDIALS (with OpenMP) libraries +and the ``pybind11`` headers. +SuiteSparse and SUNDIALS are installed in ``~/.local`` by default. For users requiring more control over the installation process, the ``pybamm-requires`` session supports additional command-line arguments: @@ -110,7 +111,7 @@ If you'd rather do things yourself, 1. Make sure you have CMake installed 2. Compile and install SuiteSparse (PyBaMM only requires the ``KLU`` component). -3. Compile and install SUNDIALS. +3. Compile and install SUNDIALS with `OpenMP support <https://mac.r-project.org/openmp/>`_. 4. Clone the pybind11 repository in the ``PyBaMM/`` directory (make sure the directory is named ``pybind11``). @@ -315,12 +316,12 @@ source files to your current directory. ``ValueError: Integrator name ida does not exist``, or ``ValueError: Integrator name cvode does not exist``. -**Solution:** This could mean that you have not installed -``scikits.odes`` correctly, check the instructions given above and make +**Solution:** This could mean that you have not linked to +SUNDIALS correctly, check the instructions given above and make sure each command was successful. One possibility is that you have not set your ``LD_LIBRARY_PATH`` to -point to the sundials library, type ``echo $LD_LIBRARY_PATH`` and make +point to the SUNDIALS library, type ``echo $LD_LIBRARY_PATH`` and make sure one of the directories printed out corresponds to where the SUNDIALS libraries are located. diff --git a/scripts/install_KLU_Sundials.py b/scripts/install_KLU_Sundials.py index 6224e40776..c0e046c48d 100755 --- a/scripts/install_KLU_Sundials.py +++ b/scripts/install_KLU_Sundials.py @@ -14,18 +14,35 @@ SUITESPARSE_VERSION = "6.0.3" SUNDIALS_VERSION = "6.5.0" + SUITESPARSE_URL = f"https://github.com/DrTimothyAldenDavis/SuiteSparse/archive/v{SUITESPARSE_VERSION}.tar.gz" SUNDIALS_URL = f"https://github.com/LLNL/sundials/releases/download/v{SUNDIALS_VERSION}/sundials-{SUNDIALS_VERSION}.tar.gz" + SUITESPARSE_CHECKSUM = ( "7111b505c1207f6f4bd0be9740d0b2897e1146b845d73787df07901b4f5c1fb7" ) SUNDIALS_CHECKSUM = "4e0b998dff292a2617e179609b539b511eb80836f5faacf800e688a886288502" + +# universal binaries for macOS 11.0 and later; sourced from https://mac.r-project.org/openmp/ +OPENMP_VERSION = "16.0.4" +OPENMP_URL = ( + f"https://mac.r-project.org/openmp/openmp-{OPENMP_VERSION}-darwin20-Release.tar.gz" +) +OPENMP_CHECKSUM = "a763f0bdc9115c4f4933accc81f514f3087d56d6528778f38419c2a0d2231972" + + DEFAULT_INSTALL_DIR = os.path.join(os.getenv("HOME"), ".local") def safe_remove_dir(path): - if os.path.exists(path): - shutil.rmtree(path) + """Remove a directory or file if it exists.""" + try: + if os.path.isfile(path): + os.remove(path) + elif os.path.isdir(path): + shutil.rmtree(path) + except Exception as e: + print(f"Error while removing {path}: {e}") def install_suitesparse(download_dir): @@ -51,11 +68,11 @@ def install_suitesparse(download_dir): env = os.environ.copy() for libdir in ["SuiteSparse_config", "AMD", "COLAMD", "BTF", "KLU"]: build_dir = os.path.join(suitesparse_src, libdir) - # We want to ensure that libsuitesparseconfig.dylib is not repeated in - # multiple paths at the time of wheel repair. Therefore, it should not be - # built with an RPATH since it is copied to the install prefix. + # Set an RPATH in order for libsuitesparseconfig.dylib to find libomp.dylib if libdir == "SuiteSparse_config": - env["CMAKE_OPTIONS"] = f"-DCMAKE_INSTALL_PREFIX={install_dir}" + env["CMAKE_OPTIONS"] = ( + f"-DCMAKE_INSTALL_PREFIX={install_dir} -DCMAKE_INSTALL_RPATH={install_dir}/lib" + ) else: # For AMD, COLAMD, BTF and KLU; do not set a BUILD RPATH but use an # INSTALL RPATH in order to ensure that the dynamic libraries are found @@ -93,20 +110,9 @@ def install_sundials(download_dir, install_dir): # try to find OpenMP on mac if platform.system() == "Darwin": # flags to find OpenMP on mac - if platform.processor() == "arm": - OpenMP_C_FLAGS = ( - "-Xpreprocessor -fopenmp -I/opt/homebrew/opt/libomp/include" - ) - OpenMP_C_LIB_NAMES = "omp" - OpenMP_omp_LIBRARY = "/opt/homebrew/opt/libomp/lib/libomp.dylib" - elif platform.processor() == "i386": - OpenMP_C_FLAGS = "-Xpreprocessor -fopenmp -I/usr/local/opt/libomp/include" - OpenMP_C_LIB_NAMES = "omp" - OpenMP_omp_LIBRARY = "/usr/local/opt/libomp/lib/libomp.dylib" - else: - raise NotImplementedError( - f"Unsupported processor architecture: {platform.processor()}. Only 'arm' and 'i386' architectures are supported." - ) + OpenMP_C_FLAGS = f"-Xpreprocessor -fopenmp -lomp -L{os.path.join(KLU_LIBRARY_DIR)} -I{os.path.join(KLU_INCLUDE_DIR)}" + OpenMP_C_LIB_NAMES = "omp" + OpenMP_omp_LIBRARY = os.path.join(KLU_LIBRARY_DIR, "libomp.dylib") cmake_args += [ "-DOpenMP_C_FLAGS=" + OpenMP_C_FLAGS, @@ -130,6 +136,47 @@ def install_sundials(download_dir, install_dir): subprocess.run(make_cmd, cwd=build_dir, check=True) +# Relevant for macOS only because recent Xcode Clang versions do not include OpenMP headers. +# Other compilers (e.g. GCC) include the OpenMP specification by default. +def set_up_openmp(download_dir, install_dir): + print("-" * 10, "Extracting OpenMP archive", "-" * 40) + + openmp_dir = f"openmp-{OPENMP_VERSION}" + openmp_src = os.path.join(download_dir, openmp_dir) + + # extract OpenMP archive + with tarfile.open( + os.path.join(download_dir, f"{openmp_dir}-darwin20-Release.tar.gz") + ) as tar: + tar.extractall(openmp_src) + + # create directories + os.makedirs(os.path.join(install_dir, "lib"), exist_ok=True) + os.makedirs(os.path.join(install_dir, "include"), exist_ok=True) + + # copy files + shutil.copy( + os.path.join(openmp_src, "usr", "local", "lib", "libomp.dylib"), + os.path.join(install_dir, "lib"), + ) + for file in os.listdir(os.path.join(openmp_src, "usr", "local", "include")): + shutil.copy( + os.path.join(openmp_src, "usr", "local", "include", file), + os.path.join(install_dir, "include"), + ) + + # fix rpath; for some reason the downloaded dylib has an absolute path + # to /usr/local/lib/, so use self-referential rpath + subprocess.check_call( + [ + "install_name_tool", + "-id", + "@rpath/libomp.dylib", + f"{os.path.join(install_dir, 'lib', 'libomp.dylib')}", + ] + ) + + def check_libraries_installed(install_dir): # Define the directories to check for SUNDIALS and SuiteSparse libraries lib_dirs = [install_dir] @@ -177,7 +224,7 @@ def check_libraries_installed(install_dir): suitesparse_files = [file + ".dylib" for file in suitesparse_files] else: raise NotImplementedError( - f"Unsupported operating system: {platform.system()}. This script currently supports only Linux and macOS." + f"Unsupported operating system: {platform.system()}. This script supports only Linux and macOS." ) suitesparse_lib_found = True @@ -199,6 +246,16 @@ def check_libraries_installed(install_dir): return sundials_lib_found, suitesparse_lib_found +def check_openmp_installed_on_macos(install_dir): + openmp_lib_found = isfile(join(install_dir, "lib", "libomp.dylib")) + openmp_headers_found = isfile(join(install_dir, "include", "omp.h")) + if not openmp_lib_found or not openmp_headers_found: + print("libomp.dylib or omp.h not found. Proceeding with OpenMP installation.") + else: + print(f"libomp.dylib and omp.h found in {install_dir}.") + return openmp_lib_found + + def calculate_sha256(file_path): sha256_hash = hashlib.sha256() with open(file_path, "rb") as f: @@ -274,7 +331,7 @@ def parallel_download(urls, download_dir): # Get installation location parser = argparse.ArgumentParser( - description="Download, compile and install Sundials and SuiteSparse." + description="Download, compile and install SUNDIALS and SuiteSparse." ) parser.add_argument( "--force", @@ -296,10 +353,20 @@ def parallel_download(urls, download_dir): safe_remove_dir(os.path.join(download_dir, "build_sundials")) safe_remove_dir(os.path.join(download_dir, f"SuiteSparse-{SUITESPARSE_VERSION}")) safe_remove_dir(os.path.join(download_dir, f"sundials-{SUNDIALS_VERSION}")) - sundials_found, suitesparse_found = False, False + if platform.system() == "Darwin": + safe_remove_dir(os.path.join(install_dir, "lib", "libomp.dylib")) + safe_remove_dir(os.path.join(install_dir, "include", "omp.h")) + sundials_found, suitesparse_found, openmp_found = False, False, False + else: + sundials_found, suitesparse_found = False, False else: # Check whether the libraries are installed - sundials_found, suitesparse_found = check_libraries_installed(install_dir) + if platform.system() == "Darwin": + sundials_found, suitesparse_found = check_libraries_installed(install_dir) + openmp_found = check_openmp_installed_on_macos(install_dir) + else: # Linux + sundials_found, suitesparse_found = check_libraries_installed(install_dir) + if __name__ == "__main__": # Determine which libraries to download based on whether they were found @@ -312,12 +379,22 @@ def parallel_download(urls, download_dir): ], download_dir, ) + + if platform.system() == "Darwin" and not openmp_found: + download_extract_library(OPENMP_URL, OPENMP_CHECKSUM, download_dir) + set_up_openmp(download_dir, install_dir) + install_suitesparse(download_dir) install_sundials(download_dir, install_dir) + else: if not sundials_found: # Only SUNDIALS is missing, download and install it parallel_download([(SUNDIALS_URL, SUNDIALS_CHECKSUM)], download_dir) + if platform.system() == "Darwin" and not openmp_found: + download_extract_library(OPENMP_URL, OPENMP_CHECKSUM, download_dir) + set_up_openmp(download_dir, install_dir) + # openmp needed for SUNDIALS on macOS install_sundials(download_dir, install_dir) if not suitesparse_found: # Only SuiteSparse is missing, download and install it

Fortnightly build for wheels failed / PyBaMM import kills process on macOS M-series The build is failing with the following logs - https://github.com/pybamm-team/PyBaMM/actions/runs/9093914939 Note: after #4092, this is failing for the self-hosted runners as well after the wheels were fixed: https://github.com/pybamm-team/PyBaMM/actions/runs/9184757819/job/25257629262 most likely due to a SIGKILL or some sort of memory leak.

Something is up either on the `cibuildwheel` side or when we build SuiteSparse and SUNDIALS. Looks like `MACOSX_DEPLOYMENT_TARGET` wasn't set for the latter. I shall take a look at this.

2024-05-15T13:18:18

0.0

[]

pybamm-team/PyBaMM

pybamm-team__PyBaMM-4083

b972782ab9a8e5b308c29c04a7212fddcb71a273

diff --git a/pybamm/models/submodels/thermal/base_thermal.py b/pybamm/models/submodels/thermal/base_thermal.py index 757181655b..65a3f72a59 100644 --- a/pybamm/models/submodels/thermal/base_thermal.py +++ b/pybamm/models/submodels/thermal/base_thermal.py @@ -138,7 +138,7 @@ def _get_standard_coupled_variables(self, variables): num_phases = int(self.options.negative["particle phases"]) phase_names = [""] if num_phases > 1: - phase_names = ["primary", "secondary"] + phase_names = ["primary ", "secondary "] if self.options.electrode_types["negative"] == "planar": Q_rxn_n = pybamm.FullBroadcast(

[Bug]: thermal option does not work in 2 phase dfn model ### PyBaMM Version 24.1 ### Python Version 3.9 ### Describe the bug thermal option does not work in 2 phase dfn model ### Steps to Reproduce model = pybamm.lithium_ion.DFN( { "particle phases": ("2", "1"), "open-circuit potential": (("single", "current sigmoid"), "single"), "thermal": "lumped", "cell geometry": "pouch", } ) ### Relevant log output ```shell ModelError: Missing variable for submodel 'thermal': 'Negative electrode primaryvolumetric interfacial current density [A.m-3]'. Check the selected submodels provide all of the required variables. ```

Adding a space after {phase} in base_thermal.py _get_standard_coupled_variables() seems to solve the problem Thanks, the way to fix this is to replace [this line](https://github.com/pybamm-team/PyBaMM/blob/1dae0bddd25c2b998ef6b635313c7bd0e6d8a33c/pybamm/models/submodels/thermal/base_thermal.py#L141) with ` phase_names = ["primary ", "secondary "]` so that you don't end up with an extra space in the single phase model. Could you open a PR with a fix? I will do a quick patch for this so we can get this into the release

2024-05-13T15:44:20

0.0

[]

pybamm-team/PyBaMM

pybamm-team__PyBaMM-4020

5d22165e7d97ae48ad2e567d8a7742cfbde1eb08

diff --git a/CHANGELOG.md b/CHANGELOG.md index ea458b4b60..a8d7d155de 100644 --- a/CHANGELOG.md +++ b/CHANGELOG.md @@ -2,6 +2,7 @@ ## Features +- Added `plot_thermal_components` to plot the contributions to the total heat generation in a battery ([#4021](https://github.com/pybamm-team/PyBaMM/pull/4021)) - Added functions for normal probability density function (`pybamm.normal_pdf`) and cumulative distribution function (`pybamm.normal_cdf`) ([#3999](https://github.com/pybamm-team/PyBaMM/pull/3999)) - Updates multiprocess `Pool` in `BaseSolver.solve()` to be constructed with context `fork`. Adds small example for multiprocess inputs. ([#3974](https://github.com/pybamm-team/PyBaMM/pull/3974)) - Added custom experiment steps ([#3835](https://github.com/pybamm-team/PyBaMM/pull/3835)) @@ -14,6 +15,7 @@ ## Bug Fixes +- Set the `remove_independent_variables_from_rhs` to `False` by default, and moved the option from `Discretisation.process_model` to `Discretisation.__init__`. This fixes a bug related to the discharge capacity, but may make the simulation slower in some cases. To set the option to `True`, use `Simulation(..., discretisation_kwargs={"remove_independent_variables_from_rhs": True})`. ([#4020](https://github.com/pybamm-team/PyBaMM/pull/4020)) - Fixed a bug where independent variables were removed from models even if they appeared in events ([#4019](https://github.com/pybamm-team/PyBaMM/pull/4019)) - Fix bug with upwind and downwind schemes producing the wrong discretised system ([#3979](https://github.com/pybamm-team/PyBaMM/pull/3979)) - Allow evaluation of an `Interpolant` object with a number ([#3932](https://github.com/pybamm-team/PyBaMM/pull/3932)) @@ -33,6 +35,7 @@ ## Breaking changes +- Removed `check_model` argument from `Simulation.solve`. To change the `check_model` option, use `Simulation(..., discretisation_kwargs={"check_model": False})`. ([#4020](https://github.com/pybamm-team/PyBaMM/pull/4020)) - Removed multiple Docker images. Here on, a single Docker image tagged `pybamm/pybamm:latest` will be provided with both solvers (`IDAKLU` and `JAX`) pre-installed. ([#3992](https://github.com/pybamm-team/PyBaMM/pull/3992)) - Removed support for Python 3.8 ([#3961](https://github.com/pybamm-team/PyBaMM/pull/3961)) - Renamed "ocp_soc_0_dimensional" to "ocp_soc_0" and "ocp_soc_100_dimensional" to "ocp_soc_100" ([#3942](https://github.com/pybamm-team/PyBaMM/pull/3942)) diff --git a/pybamm/batch_study.py b/pybamm/batch_study.py index 1503c3048c..e854c94e00 100644 --- a/pybamm/batch_study.py +++ b/pybamm/batch_study.py @@ -102,7 +102,6 @@ def solve( self, t_eval=None, solver=None, - check_model=True, save_at_cycles=None, calc_esoh=True, starting_solution=None, @@ -158,7 +157,6 @@ def solve( sol = sim.solve( t_eval, solver, - check_model, save_at_cycles, calc_esoh, starting_solution, diff --git a/pybamm/discretisations/discretisation.py b/pybamm/discretisations/discretisation.py index 92e9e4e24f..d61d8a4d60 100644 --- a/pybamm/discretisations/discretisation.py +++ b/pybamm/discretisations/discretisation.py @@ -26,14 +26,32 @@ class Discretisation: Parameters ---------- mesh : pybamm.Mesh - contains all submeshes to be used on each domain + contains all submeshes to be used on each domain spatial_methods : dict - a dictionary of the spatial methods to be used on each - domain. The keys correspond to the model domains and the - values to the spatial method. + a dictionary of the spatial methods to be used on each + domain. The keys correspond to the model domains and the + values to the spatial method. + check_model : bool, optional + If True, model checks are performed after discretisation. For large + systems these checks can be slow, so can be skipped by setting this + option to False. When developing, testing or debugging it is recommended + to leave this option as True as it may help to identify any errors. + Default is True. + remove_independent_variables_from_rhs : bool, optional + If True, model checks to see whether any variables from the RHS are used + in any other equation. If a variable meets all of the following criteria + (not used anywhere in the model, len(rhs)>1), then the variable + is moved to be explicitly integrated when called by the solution object. + Default is False. """ - def __init__(self, mesh=None, spatial_methods=None): + def __init__( + self, + mesh=None, + spatial_methods=None, + check_model=True, + remove_independent_variables_from_rhs=False, + ): self._mesh = mesh if mesh is None: self._spatial_methods = {} @@ -60,6 +78,10 @@ def __init__(self, mesh=None, spatial_methods=None): self._bcs = {} self.y_slices = {} self._discretised_symbols = {} + self._check_model_flag = check_model + self._remove_independent_variables_from_rhs_flag = ( + remove_independent_variables_from_rhs + ) @property def mesh(self): @@ -90,13 +112,7 @@ def bcs(self, value): # reset discretised_symbols self._discretised_symbols = {} - def process_model( - self, - model, - inplace=True, - check_model=True, - remove_independent_variables_from_rhs=True, - ): + def process_model(self, model, inplace=True): """ Discretise a model. Currently inplace, could be changed to return a new model. @@ -108,18 +124,6 @@ def process_model( inplace : bool, optional If True, discretise the model in place. Otherwise, return a new discretised model. Default is True. - check_model : bool, optional - If True, model checks are performed after discretisation. For large - systems these checks can be slow, so can be skipped by setting this - option to False. When developing, testing or debugging it is recommended - to leave this option as True as it may help to identify any errors. - Default is True. - remove_independent_variables_from_rhs : bool, optional - If True, model checks to see whether any variables from the RHS are used - in any other equation. If a variable meets all of the following criteria - (not used anywhere in the model, len(rhs)>1), then the variable - is moved to be explicitly integrated when called by the solution object. - Default is True. Returns ------- @@ -158,7 +162,7 @@ def process_model( # set variables (we require the full variable not just id) # Search Equations for Independence - if remove_independent_variables_from_rhs: + if self._remove_independent_variables_from_rhs_flag: model = self.remove_independent_variables_from_rhs(model) variables = list(model.rhs.keys()) + list(model.algebraic.keys()) # Find those RHS's that are constant @@ -240,7 +244,7 @@ def process_model( model_disc._geometry = getattr(self.mesh, "_geometry", None) # Check that resulting model makes sense - if check_model: + if self._check_model_flag: pybamm.logger.verbose(f"Performing model checks for {model.name}") self.check_model(model_disc) @@ -1123,6 +1127,7 @@ def remove_independent_variables_from_rhs(self, model): # in variables twice under different names for key in model.variables: if model.variables[key] == var: + print("here") model.variables[key] = model.variables[var.name] del model.rhs[var] del model.initial_conditions[var] diff --git a/pybamm/simulation.py b/pybamm/simulation.py index 060d1cdd8c..a81f7b62b2 100644 --- a/pybamm/simulation.py +++ b/pybamm/simulation.py @@ -63,6 +63,9 @@ class Simulation: A list of variables to plot automatically C_rate: float (optional) The C-rate at which you would like to run a constant current (dis)charge. + discretisation_kwargs: dict (optional) + Any keyword arguments to pass to the Discretisation class. + See :class:`pybamm.Discretisation` for details. """ def __init__( @@ -77,6 +80,7 @@ def __init__( solver=None, output_variables=None, C_rate=None, + discretisation_kwargs=None, ): self._parameter_values = parameter_values or model.default_parameter_values self._unprocessed_parameter_values = self._parameter_values @@ -126,6 +130,7 @@ def __init__( self._spatial_methods = spatial_methods or self._model.default_spatial_methods self._solver = solver or self._model.default_solver self._output_variables = output_variables + self._discretisation_kwargs = discretisation_kwargs or {} # Initialize empty built states self._model_with_set_params = None @@ -268,7 +273,7 @@ def set_initial_soc(self, initial_soc, inputs=None): # Save solved initial SOC in case we need to re-build the model self._built_initial_soc = initial_soc - def build(self, check_model=True, initial_soc=None, inputs=None): + def build(self, initial_soc=None, inputs=None): """ A method to build the model into a system of matrices and vectors suitable for performing numerical computations. If the model has already been built or @@ -278,13 +283,12 @@ def build(self, check_model=True, initial_soc=None, inputs=None): Parameters ---------- - check_model : bool, optional - If True, model checks are performed after discretisation (see - :meth:`pybamm.Discretisation.process_model`). Default is True. initial_soc : float, optional Initial State of Charge (SOC) for the simulation. Must be between 0 and 1. If given, overwrites the initial concentrations provided in the parameter set. + inputs : dict, optional + A dictionary of input parameters to pass to the model when solving. """ if initial_soc is not None: self.set_initial_soc(initial_soc, inputs=inputs) @@ -297,14 +301,16 @@ def build(self, check_model=True, initial_soc=None, inputs=None): else: self.set_parameters() self._mesh = pybamm.Mesh(self._geometry, self._submesh_types, self._var_pts) - self._disc = pybamm.Discretisation(self._mesh, self._spatial_methods) + self._disc = pybamm.Discretisation( + self._mesh, self._spatial_methods, **self._discretisation_kwargs + ) self._built_model = self._disc.process_model( - self._model_with_set_params, inplace=False, check_model=check_model + self._model_with_set_params, inplace=False ) # rebuilt model so clear solver setup self._solver._model_set_up = {} - def build_for_experiment(self, check_model=True, initial_soc=None, inputs=None): + def build_for_experiment(self, initial_soc=None, inputs=None): """ Similar to :meth:`Simulation.build`, but for the case of simulating an experiment, where there may be several models and solvers to build. @@ -322,7 +328,9 @@ def build_for_experiment(self, check_model=True, initial_soc=None, inputs=None): self._parameter_values.process_geometry(self._geometry) # Only needs to set up mesh and discretisation once self._mesh = pybamm.Mesh(self._geometry, self._submesh_types, self._var_pts) - self._disc = pybamm.Discretisation(self._mesh, self._spatial_methods) + self._disc = pybamm.Discretisation( + self._mesh, self._spatial_methods, **self._discretisation_kwargs + ) # Process all the different models self.steps_to_built_models = {} self.steps_to_built_solvers = {} @@ -333,7 +341,7 @@ def build_for_experiment(self, check_model=True, initial_soc=None, inputs=None): # It's ok to modify the model with set parameters in place as it's # not returned anywhere built_model = self._disc.process_model( - model_with_set_params, inplace=True, check_model=check_model + model_with_set_params, inplace=True ) solver = self._solver.copy() self.steps_to_built_solvers[step] = solver @@ -343,7 +351,6 @@ def solve( self, t_eval=None, solver=None, - check_model=True, save_at_cycles=None, calc_esoh=True, starting_solution=None, @@ -377,9 +384,6 @@ def solve( provided in the data. solver : :class:`pybamm.BaseSolver`, optional The solver to use to solve the model. If None, Simulation.solver is used - check_model : bool, optional - If True, model checks are performed after discretisation (see - :meth:`pybamm.Discretisation.process_model`). Default is True. save_at_cycles : int or list of ints, optional Which cycles to save the full sub-solutions for. If None, all cycles are saved. If int, every multiple of save_at_cycles is saved. If list, every @@ -416,7 +420,7 @@ def solve( inputs = inputs or {} if self.operating_mode in ["without experiment", "drive cycle"]: - self.build(check_model=check_model, initial_soc=initial_soc, inputs=inputs) + self.build(initial_soc=initial_soc, inputs=inputs) if save_at_cycles is not None: raise ValueError( "'save_at_cycles' option can only be used if simulating an " @@ -493,9 +497,7 @@ def solve( elif self.operating_mode == "with experiment": callbacks.on_experiment_start(logs) - self.build_for_experiment( - check_model=check_model, initial_soc=initial_soc, inputs=inputs - ) + self.build_for_experiment(initial_soc=initial_soc, inputs=inputs) if t_eval is not None: pybamm.logger.warning( "Ignoring t_eval as solution times are specified by the experiment"

"Throughput capacity [A.h]" is different for sol.cycles[0] and sol.cycles[0].steps[0] ### Discussed in https://github.com/pybamm-team/PyBaMM/discussions/2523 <div type='discussions-op-text'> Originally posted by **RuiheLi** December 5, 2022 Dear PyBaMM, I am using PyBaMM 22.9 and Python 3.9.6. I find when using that set_initial_conditions_from, "Throughput capacity [A.h]" is different for sol.cycles[0] and sol.cycles[0].steps[0] for a later solution. Figure below: ![image](https://user-images.githubusercontent.com/84007676/205644495-558eabbe-1f36-4dd5-9935-44f131704896.png) Script below: ``` import pybamm import matplotlib as mpl fs=17; font = {'family' : 'DejaVu Sans','size' : fs} mpl.rc('font', **font) V_max = 4.0; V_min = 2.7; model_0 = pybamm.lithium_ion.DFN(options={ "calculate discharge energy":"true", },) exp_AGE_text = [(f"Discharge at 1 C until {V_min} V", f"Charge at 1 C until {V_max} V", ), ] Experiment = pybamm.Experiment( exp_AGE_text * 1) Para_0 = pybamm.ParameterValues("OKane2022" ) Sim_0= pybamm.Simulation( model_0, experiment = Experiment, parameter_values=Para_0, solver = pybamm.CasadiSolver(),) Sol_0 = Sim_0.solve(calc_esoh=False,) model_1 = model_0.set_initial_conditions_from(Sol_0, inplace=False) Sim_1= pybamm.Simulation( model_1, experiment = Experiment, parameter_values=Para_0, solver = pybamm.CasadiSolver(),) Sol_1 = Sim_1.solve(calc_esoh=False,) model_2 = model_0.set_initial_conditions_from(Sol_1, inplace=False) Sim_2= pybamm.Simulation( model_2, experiment = Experiment, parameter_values=Para_0, solver = pybamm.CasadiSolver(),) Sol_2 = Sim_2.solve(calc_esoh=False,) label = ["Sol_0","Sol_1","Sol_2",] output_variables3 = [ "Throughput energy [W.h]", "Throughput capacity [A.h]", "Terminal voltage [V]", ] quick_plot = pybamm.QuickPlot( [Sol_0,Sol_1,Sol_2,], output_variables3,label,variable_limits='fixed', time_unit='hours',n_rows=1) #figsize = (18,12), quick_plot.dynamic_plot() label = ["Sol_2 whole","Sol_2 cycles","Sol_2 steps",] quick_plot = pybamm.QuickPlot( [Sol_2,Sol_2.cycles[0],Sol_2.cycles[0].steps[0]], output_variables3,label,variable_limits='fixed', time_unit='hours',n_rows=1) #figsize = (18,12), quick_plot.dynamic_plot() ``` Do you have any ideas on this? Big thanks! </div>

Basically the problem is that `initial_value` for `ExplicitTimeIntegral` objects, which is used here https://github.com/pybamm-team/PyBaMM/blob/b072d22f1344f062fc84ca44e0c540873669b6f7/pybamm/solvers/solution.py#L472 is not updated properly when doing `solution.last_state`, which is used here https://github.com/pybamm-team/PyBaMM/blob/b072d22f1344f062fc84ca44e0c540873669b6f7/pybamm/models/base_model.py#L596 To solve this properly will require reformatting how we do `ExplicitTimeIntegral`, which in turn will depend on other improvements to the backend that we are working on. In the meantime, what you could do is set this to `remove_independent_variables_from_rhs=False` in your branch https://github.com/pybamm-team/PyBaMM/blob/b072d22f1344f062fc84ca44e0c540873669b6f7/pybamm/discretisations/discretisation.py#L102 (but note that it might take longer to solve the model as a result) Possibly related to this, I encountered a similar issue in a much simpler problem (free fall problem) ```python3 # Create model model = pybamm.BaseModel() u = pybamm.Variable("u") v = pybamm.Variable("v") model.rhs = {u: v, v: -1} model.initial_conditions = {u: 2, v: 0} model.variables = {"u": u, "v": v} # Discretise using default discretisation disc = pybamm.Discretisation() disc.process_model(model); # Solve t_eval = np.linspace(0, 5, 30) ode_solver = pybamm.ScipySolver() solution = ode_solver.solve(model, t_eval) # Plot solution.plot(["u", "v"]) ``` It seems that after the first step, the variable `u` is set to zero and then evolves according to the equation. ![image](https://github.com/pybamm-team/PyBaMM/assets/28443643/3401215a-570b-42ff-9a70-9e0fd4b10c0a)

2024-04-17T14:42:40

0.0

[]

pybamm-team/PyBaMM

pybamm-team__PyBaMM-4001

8cd1cec79134e60aed9dfa04b2f14883b9b5a8b5

diff --git a/pybamm/solvers/base_solver.py b/pybamm/solvers/base_solver.py index 28dd4bdd9d..0e60aca189 100644 --- a/pybamm/solvers/base_solver.py +++ b/pybamm/solvers/base_solver.py @@ -13,10 +13,6 @@ import pybamm from pybamm.expression_tree.binary_operators import _Heaviside -# Set context for parallel processing depending on the platform -if platform.system() == "Darwin" or platform.system() == "Linux": - mp.set_start_method("fork") - class BaseSolver: """Solve a discretised model. @@ -70,6 +66,7 @@ def __init__( self.algebraic_solver = False self._on_extrapolation = "warn" self.computed_var_fcns = {} + self._mp_context = self.get_platform_context(platform.system()) @property def root_method(self): @@ -917,7 +914,7 @@ def solve( model_inputs_list, ) else: - with mp.Pool(processes=nproc) as p: + with mp.get_context(self._mp_context).Pool(processes=nproc) as p: new_solutions = p.starmap( self._integrate, zip( @@ -1395,6 +1392,12 @@ def check_extrapolation(self, solution, events): "outside these bounds." ) + def get_platform_context(self, system_type: str): + # Set context for parallel processing depending on the platform + if system_type.lower() in ["linux", "darwin"]: + return "fork" + return "spawn" + @staticmethod def _set_up_model_inputs(model, inputs): """Set up input parameters"""

[Bug]: PyBaMM fails to import when user has already imported a library that uses `mp.set_start_method` ### PyBaMM Version develop ### Python Version 3.11 ### Describe the bug PyBaMM fails to import when importing a library that uses multiprocessing and has already set the start method to something other than fork, as a result of #3975. This should be apparent when reading the documentation for `mp.set_start_method` Pending a long-term solution, I think that #3975 should at least have `mp.set_start_method` wrapped in a try-except -- most people, most of the time, do not really need the multiprocessing functionality, and I don't think pybamm should force those users to modify their multiprocessing configuration. It can throw a warning saying multiprocessing is not available or whatever. Other solutions are welcome, of course. ### Steps to Reproduce module `thing`: ``` import multiprocessing as mp mp.set_start_method("spawn") def hello(): print("hello, world") ``` script: ``` import thing import pybamm ``` ### Relevant log output ```shell ❯ python main.py Traceback (most recent call last): File "/Users/abills/.../main.py", line 2, in <module> import pybamm File "/Users/abills/.../python3.11/site-packages/pybamm/__init__.py", line 204, in <module> from .solvers.base_solver import BaseSolver File "/Users/abills/.../python3.11/site-packages/pybamm/solvers/base_solver.py", line 18, in <module> mp.set_start_method("fork") File "/Users/abills/.../python3.11/multiprocessing/context.py", line 247, in set_start_method raise RuntimeError('context has already been set') RuntimeError: context has already been set ```

@BradyPlanden Tagging you since the bug most likely comes from #3975

2024-04-12T08:03:36

0.0

[]

pybamm-team/PyBaMM

pybamm-team__PyBaMM-3988

51c1f76efabbd9a1444f472bab42c70dfba3b4f9

diff --git a/docs/source/api/util.rst b/docs/source/api/util.rst index 03e7388e3e..f187cfbabb 100644 --- a/docs/source/api/util.rst +++ b/docs/source/api/util.rst @@ -3,8 +3,6 @@ Utility functions .. autofunction:: pybamm.get_git_commit_info -.. autofunction:: pybamm.rmse - .. autofunction:: pybamm.root_dir .. autoclass:: pybamm.Timer diff --git a/pybamm/__init__.py b/pybamm/__init__.py index 68195ca279..6d6e1b855c 100644 --- a/pybamm/__init__.py +++ b/pybamm/__init__.py @@ -1,38 +1,12 @@ import sys -import os - -# -# Version info -# from pybamm.version import __version__ -# -# Constants -# -# Float format: a float can be converted to a 17 digit decimal and back without -# loss of information -FLOAT_FORMAT = "{: .17e}" -# Absolute path to the PyBaMM repo -script_path = os.path.abspath(__file__) - -from .util import root_dir - -ABSOLUTE_PATH = root_dir() -PARAMETER_PATH = [ - root_dir(), - os.getcwd(), - os.path.join(root_dir(), "pybamm", "input", "parameters"), -] - - -# # Utility classes and methods -# +from .util import root_dir from .util import Timer, TimerTime, FuzzyDict from .util import ( root_dir, - rmse, load, is_constant_and_can_evaluate, ) @@ -48,9 +22,7 @@ from .settings import settings from .citations import Citations, citations, print_citations -# # Classes for the Expression Tree -# from .expression_tree.symbol import * from .expression_tree.binary_operators import * from .expression_tree.concatenations import * @@ -88,9 +60,7 @@ from .expression_tree.operations.convert_to_casadi import CasadiConverter from .expression_tree.operations.unpack_symbols import SymbolUnpacker -# # Model classes -# from .models.base_model import BaseModel from .models.event import Event from .models.event import EventType @@ -104,9 +74,7 @@ from .models.full_battery_models import lithium_ion from .models.full_battery_models import equivalent_circuit -# # Submodel classes -# from .models.submodels.base_submodel import BaseSubModel from .models.submodels import ( @@ -132,18 +100,14 @@ from .models.submodels.interface import interface_utilisation from .models.submodels.interface import open_circuit_potential -# # Geometry -# from .geometry.geometry import Geometry from .geometry.battery_geometry import battery_geometry from .expression_tree.independent_variable import KNOWN_COORD_SYS from .geometry import standard_spatial_vars -# # Parameter classes and methods -# from .parameters.parameter_values import ParameterValues from .parameters import constants from .parameters.geometric_parameters import geometric_parameters, GeometricParameters @@ -158,9 +122,7 @@ from .parameters.size_distribution_parameters import * from .parameters.parameter_sets import parameter_sets -# # Mesh and Discretisation classes -# from .discretisations.discretisation import Discretisation from .discretisations.discretisation import has_bc_of_form from .meshes.meshes import Mesh, SubMesh, MeshGenerator @@ -181,23 +143,17 @@ UserSupplied2DSubMesh, ) -# # Serialisation -# from .models.base_model import load_model -# # Spatial Methods -# from .spatial_methods.spatial_method import SpatialMethod from .spatial_methods.zero_dimensional_method import ZeroDimensionalSpatialMethod from .spatial_methods.finite_volume import FiniteVolume from .spatial_methods.spectral_volume import SpectralVolume from .spatial_methods.scikit_finite_element import ScikitFiniteElement -# # Solver classes -# from .solvers.solution import Solution, EmptySolution, make_cycle_solution from .solvers.processed_variable import ProcessedVariable from .solvers.processed_variable_computed import ProcessedVariableComputed @@ -214,17 +170,12 @@ from .solvers.idaklu_jax import IDAKLUJax from .solvers.idaklu_solver import IDAKLUSolver, have_idaklu -# # Experiments -# from .experiment.experiment import Experiment from . import experiment from .experiment import step - -# # Plotting -# from .plotting.quick_plot import QuickPlot, close_plots, QuickPlotAxes from .plotting.plot import plot from .plotting.plot2D import plot2D @@ -232,22 +183,14 @@ from .plotting.plot_summary_variables import plot_summary_variables from .plotting.dynamic_plot import dynamic_plot -# # Simulation -# from .simulation import Simulation, load_sim, is_notebook -# # Batch Study -# from .batch_study import BatchStudy -# # Callbacks -# from . import callbacks -# # Remove any imported modules, so we don't expose them as part of pybamm -# del sys diff --git a/pybamm/util.py b/pybamm/util.py index c3d7b4e263..f37248dabd 100644 --- a/pybamm/util.py +++ b/pybamm/util.py @@ -18,7 +18,6 @@ import difflib from warnings import warn -import numpy as np import pybamm # Versions of jax and jaxlib compatible with PyBaMM. Note: these are also defined in @@ -244,16 +243,6 @@ def __eq__(self, other): return self.value == other.value -def rmse(x, y): - """ - Calculate the root-mean-square-error between two vectors x and y, ignoring NaNs - """ - # Check lengths - if len(x) != len(y): - raise ValueError("Vectors must have the same length") - return np.sqrt(np.nanmean((x - y) ** 2)) - - def load(filename): """Load a saved object""" with open(filename, "rb") as f:

pybamm.rmse. document return type and args types ### PyBaMM Version 23.9 ### Python Version 3.11.7 ### Describe the bug Minor issue for the PyBaMM training workshop ### Steps to Reproduce Read the documentation of pybamm.rmse ### Relevant log output _No response_

Let's get rid of pybamm.rmse, it doesn't fit within the scope of the package

2024-04-10T20:46:33

0.0

[]

pybamm-team/PyBaMM

pybamm-team__PyBaMM-3968

09ee982ba84055ec2c56f73b4cc63742c2ca0721

diff --git a/CONTRIBUTING.md b/CONTRIBUTING.md index fc8e848bb5..aeb62b9b00 100644 --- a/CONTRIBUTING.md +++ b/CONTRIBUTING.md @@ -126,26 +126,18 @@ This allows people to (1) use PyBaMM without importing optional dependencies by **Writing Tests for Optional Dependencies** -Whenever a new optional dependency is added for optional functionality, it is recommended to write a corresponding unit test in `test_util.py`. This ensures that an error is raised upon the absence of said dependency. Here's an example: +Below, we list the currently available test functions to provide an overview. If you find it useful to add new test cases please do so within `tests/unit/test_util.py`. -```python -from tests import TestCase -import pybamm - - -class TestUtil(TestCase): - def test_optional_dependency(self): - # Test that an error is raised when pybtex is not available - with self.assertRaisesRegex( - ModuleNotFoundError, "Optional dependency pybtex is not available" - ): - sys.modules["pybtex"] = None - pybamm.function_using_pybtex(x, y, z) - - # Test that the function works when pybtex is available - sys.modules["pybtex"] = pybamm.util.import_optional_dependency("pybtex") - pybamm.function_using_pybtex(x, y, z) -``` +Currently, there are three functions to test what concerns optional dependencies: +- `test_import_optional_dependency` +- `test_pybamm_import` +- `test_optional_dependencies` + +The `test_import_optional_dependency` function extracts the optional dependencies installed in the setup environment, makes them unimportable (by setting them to `None` among the `sys.modules`), and tests that the `pybamm.util.import_optional_dependency` function throws a `ModuleNotFoundError` exception when their import is attempted. + +The `test_pybamm_import` function extracts the optional dependencies installed in the setup environment and makes them unimportable (by setting them to `None` among the `sys.modules`), unloads `pybamm` and its sub-modules, and finally tests that `pybamm` can be imported successfully. In fact, it is essential that the `pybamm` package is importable with only the mandatory dependencies. + +The `test_optional_dependencies` function extracts `pybamm` mandatory distribution packages and verifies that they are not present in the optional distribution packages list in `pyproject.toml`. This test is crucial for ensuring the consistency of the released package information and potential updates to dependencies during development. ## Testing diff --git a/pybamm/citations.py b/pybamm/citations.py index ff1851bfaa..ae18f9adc7 100644 --- a/pybamm/citations.py +++ b/pybamm/citations.py @@ -67,29 +67,41 @@ def read_citations(self): """Reads the citations in `pybamm.CITATIONS.bib`. Other works can be cited by passing a BibTeX citation to :meth:`register`. """ - parse_file = import_optional_dependency("pybtex.database", "parse_file") - citations_file = os.path.join(pybamm.root_dir(), "pybamm", "CITATIONS.bib") - bib_data = parse_file(citations_file, bib_format="bibtex") - for key, entry in bib_data.entries.items(): - self._add_citation(key, entry) + try: + parse_file = import_optional_dependency("pybtex.database", "parse_file") + citations_file = os.path.join(pybamm.root_dir(), "pybamm", "CITATIONS.bib") + bib_data = parse_file(citations_file, bib_format="bibtex") + for key, entry in bib_data.entries.items(): + self._add_citation(key, entry) + except ModuleNotFoundError: # pragma: no cover + pybamm.logger.warning( + "Citations could not be read because the 'pybtex' library is not installed. " + "Install 'pybamm[cite]' to enable citation reading." + ) def _add_citation(self, key, entry): """Adds `entry` to `self._all_citations` under `key`, warning the user if a previous entry is overwritten """ - Entry = import_optional_dependency("pybtex.database", "Entry") - # Check input types are correct - if not isinstance(key, str) or not isinstance(entry, Entry): - raise TypeError() - - # Warn if overwriting a previous citation - new_citation = entry.to_string("bibtex") - if key in self._all_citations and new_citation != self._all_citations[key]: - warnings.warn(f"Replacing citation for {key}", stacklevel=2) - - # Add to database - self._all_citations[key] = new_citation + try: + Entry = import_optional_dependency("pybtex.database", "Entry") + # Check input types are correct + if not isinstance(key, str) or not isinstance(entry, Entry): + raise TypeError() + + # Warn if overwriting a previous citation + new_citation = entry.to_string("bibtex") + if key in self._all_citations and new_citation != self._all_citations[key]: + warnings.warn(f"Replacing citation for {key}", stacklevel=2) + + # Add to database + self._all_citations[key] = new_citation + except ModuleNotFoundError: # pragma: no cover + pybamm.logger.warning( + f"Could not add citation for '{key}' because the 'pybtex' library is not installed. " + "Install 'pybamm[cite]' to enable adding citations." + ) def _add_citation_tag(self, key, entry): """Adds a tag for a citation key in the dict, which represents the name of the @@ -143,24 +155,32 @@ def _parse_citation(self, key): key: str A BibTeX formatted citation """ - PybtexError = import_optional_dependency("pybtex.scanner", "PybtexError") - parse_string = import_optional_dependency("pybtex.database", "parse_string") try: - # Parse string as a bibtex citation, and check that a citation was found - bib_data = parse_string(key, bib_format="bibtex") - if not bib_data.entries: - raise PybtexError("no entries found") - - # Add and register all citations - for key, entry in bib_data.entries.items(): - # Add to _all_citations dictionary - self._add_citation(key, entry) - # Add to _papers_to_cite set - self._papers_to_cite.add(key) - return - except PybtexError as error: - # Unable to parse / unknown key - raise KeyError(f"Not a bibtex citation or known citation: {key}") from error + PybtexError = import_optional_dependency("pybtex.scanner", "PybtexError") + parse_string = import_optional_dependency("pybtex.database", "parse_string") + try: + # Parse string as a bibtex citation, and check that a citation was found + bib_data = parse_string(key, bib_format="bibtex") + if not bib_data.entries: + raise PybtexError("no entries found") + + # Add and register all citations + for key, entry in bib_data.entries.items(): + # Add to _all_citations dictionary + self._add_citation(key, entry) + # Add to _papers_to_cite set + self._papers_to_cite.add(key) + return + except PybtexError as error: + # Unable to parse / unknown key + raise KeyError( + f"Not a bibtex citation or known citation: {key}" + ) from error + except ModuleNotFoundError: # pragma: no cover + pybamm.logger.warning( + f"Could not parse citation for '{key}' because the 'pybtex' library is not installed. " + "Install 'pybamm[cite]' to enable citation parsing." + ) def _tag_citations(self): """Prints the citation tags for the citations that have been registered @@ -211,38 +231,44 @@ def print(self, filename=None, output_format="text", verbose=False): """ # Parse citations that were not known keys at registration, but do not # fail if they cannot be parsed - pybtex = import_optional_dependency("pybtex") try: - for key in self._unknown_citations: - self._parse_citation(key) - except KeyError: # pragma: no cover - warnings.warn( - message=f'\nCitation with key "{key}" is invalid. Please try again\n', - category=UserWarning, - stacklevel=2, - ) - # delete the invalid citation from the set - self._unknown_citations.remove(key) - - if output_format == "text": - citations = pybtex.format_from_strings( - self._cited, style="plain", output_backend="plaintext" + pybtex = import_optional_dependency("pybtex") + try: + for key in self._unknown_citations: + self._parse_citation(key) + except KeyError: # pragma: no cover + warnings.warn( + message=f'\nCitation with key "{key}" is invalid. Please try again\n', + category=UserWarning, + stacklevel=2, + ) + # delete the invalid citation from the set + self._unknown_citations.remove(key) + + if output_format == "text": + citations = pybtex.format_from_strings( + self._cited, style="plain", output_backend="plaintext" + ) + elif output_format == "bibtex": + citations = "\n".join(self._cited) + else: + raise pybamm.OptionError( + f"Output format {output_format} not recognised." + "It should be 'text' or 'bibtex'." + ) + + if filename is None: + print(citations) + if verbose: + self._tag_citations() # pragma: no cover + else: + with open(filename, "w") as f: + f.write(citations) + except ModuleNotFoundError: # pragma: no cover + pybamm.logger.warning( + "Could not print citations because the 'pybtex' library is not installed. " + "Please, install 'pybamm[cite]' to print citations." ) - elif output_format == "bibtex": - citations = "\n".join(self._cited) - else: - raise pybamm.OptionError( - f"Output format {output_format} not recognised." - "It should be 'text' or 'bibtex'." - ) - - if filename is None: - print(citations) - if verbose: - self._tag_citations() # pragma: no cover - else: - with open(filename, "w") as f: - f.write(citations) def print_citations(filename=None, output_format="text", verbose=False):

PyBaMM import error - Cannot run anything (import broken due to `pybtex`) ### Discussed in https://github.com/pybamm-team/PyBaMM/discussions/3928 <div type='discussions-op-text'> Originally posted by **abhishek-appana** March 25, 2024 Hello, I am running Python 3.11.8 and installed PyBaMM 24.1 using pip on a Windows system. Earlier I was using Python 3.10 and everything was fine. But I am getting the below error after upgrading Python. When I run `import pybamm`, it gives me the error: ModuleNotFoundError: Optional dependency pybtex.database is not available. See https://docs.pybamm.org/en/latest/source/user_guide/installation/index.html#optional-dependencies for more details. I tried installing pybamm using `pip install pybamm[all]`. I was still getting this error. How do I resolve this?</div> <hr> I think that this is a valid bug and it is still failing, and `import pybamm` should work without installing `pybamm[all]`. It can be noticed in the logs here: https://github.com/pybamm-team/PyBaMM/actions/runs/8461852686/job/23182296194, so I've converted this discussion to an issue instead (cc: @Saransh-cpp @arjxn-py). Perhaps this isn't being properly checked through #3892, @lorenzofavaro?

Perhaps #3886 broke this where the try-catch blocks everywhere were removed? I haven't looked at this in enough detail, but the breakage is coming from this instantiation of the `Citations` class at the end of the file. ```python citations = Citations() ``` where `__init__` tries to call the function `_reset()`, and that in-turn is trying to register the PyBaMM and NumPy citation, which will fail? Hmm, I somehow can't reproduce this locally. I'm not sure where is this originating from. Me neither I haven't looked at this again, but I think having it fail in a fresh CI environment means there is indeed a bug, even if we can't reproduce it on our own machines. I added the "high priority" label for now. Maybe a fix can be to register the PyBaMM and NumPy papers only inside the `print_citations()` method and only if they don't exist in the `_papers_to_cite` variable, so that they get cited when someone actually runs `pybamm.print_citations()`. I'm not sure if this is a complete fix, though. Let me look into this. ~I have looked into this and looks like this is failing as we merged #3932 & haven't removed the jobs to test `pybamm_install_odes`~ Oh wait, I see that they've been removed already I had a chance to look at the issue and understand the problem. Actually at the moment pybamm has a hidden mandatory dependency, namely `pybtex`. It's imported during `pybamm/citations.py:41`, that in turn trigger `pybamm/citations.py:70`: ```python parse_file = import_optional_dependency("pybtex.database", "parse_file") ``` During #3892 the error was still masked in `pybamm/citations` in the `except`: ```python try: self.read_citations() self._reset() except Exception as e: # pragma: no cover self._citation_err_msg = e ``` Beyond that, the error was not reported after for a subtle issue. The issue is that `test_pybamm_import` is done inside `tests/unit/test_util.py` and if we look at the module `tests/unit/_init__.py` we find that it imports `StandardModelTest` which in turn imports `pybamm`. When we **don't have** the optional libraries installed locally (e.g., from local with `pybamm.[dev]`), this import avoid the tests execution during the `nox` session (it stops before), even though it reports them as `PASSED`. In fact, if you try to run the specific tests individually (i.e. `python -m unittest tests.unit.test_util.TestUtil.test_pybamm_import`), you would get an exception before the test run. Instead, when we **have** the optional libraries installed (e.g. from CI with `pybamm.[all]`), we don't notice the error because digging deep into the `importlib` documentation, `importlib.reload(...)` does not reload all sub-modules. So when it initially in `StandardModelTest` imports `pybamm`, it also imports `pybtex.database` which will no longer be imported even with the reload (not triggering the error we expect). To solve this, we should edit the `test_pybamm_import` function. Possible solution steps: 1. Remove the hidden mandatory dependency from `pybtex` - I know that @prady0t is investigating to solve this. 2. Fix error detection - Edit `tests/unit/test_util.py`, in particular edit `test_pybamm_import`: - Do not use `importlib.reload(...)` to reload `pybamm`, instead remove `pybamm` from the `sys.modules` and all its sub-modules (so also `pybamm.citations` etc..). - In the `sys.modules` set to `None` all the optional modules (e.g. `pybtex`) and all their sub-modules (so also `pybtex.database`, etc..). I could work on this in a shared PR with @prady0t. Thanks for posting this here, @lorenzofavaro, I think this is a reasonable solution based on our conversation earlier. You mentioned that you're working on some aspects of this too, so I'll assign you to this alongside @prady0t.

2024-04-07T13:55:34

0.0

[]

pybamm-team/PyBaMM

pybamm-team__PyBaMM-3963

426e441fb6d26d390f6add9d32631ccf219cd940

diff --git a/CHANGELOG.md b/CHANGELOG.md index d13c17ad25..351b644ac8 100644 --- a/CHANGELOG.md +++ b/CHANGELOG.md @@ -28,6 +28,7 @@ ## Breaking changes +- Removed support for Python 3.8 ([#3961](https://github.com/pybamm-team/PyBaMM/pull/3961)) - Renamed "ocp_soc_0_dimensional" to "ocp_soc_0" and "ocp_soc_100_dimensional" to "ocp_soc_100" ([#3942](https://github.com/pybamm-team/PyBaMM/pull/3942)) - The ODES solver was removed due to compatibility issues. Users should use IDAKLU, Casadi, or JAX instead. ([#3932](https://github.com/pybamm-team/PyBaMM/pull/3932)) - Integrated the `[pandas]` extra into the core PyBaMM package, deprecating the `pybamm[pandas]` optional dependency. Pandas is now a required dependency and will be installed upon installing PyBaMM ([#3892](https://github.com/pybamm-team/PyBaMM/pull/3892))

Flaky JAX test cases While running unit-tests, some tests appaers to be flaky. See https://github.com/pybamm-team/PyBaMM/pull/3857#issuecomment-2015300918 #3857 Is a prerequisite to this. Once it gets merged, we can discuss more such tests here and come up with a solution for them.

For additional context – we have seen this happen just with the JAX evaluator tests, where we do not receive consistent outputs. Right now, a stopgap solution is to run these tests again by adding `pytest-flaky` as a dependency and using a `@pytest.mark.flaky` decorator on them – it will then run those tests thrice until they (hopefully) pass Upon a closer look, these failures seem to be coming from comparing the time taken to get to a second solution after the first solution, which speaks more about how the test has been written rather than the computation, I feel. It can be inherently flaky, especially with how we have been noticing that the these tests in parallel mode are quite slow, cf. #3948.

2024-04-03T22:19:39

0.0

[]

pybamm-team/PyBaMM

pybamm-team__PyBaMM-3961

6bb959c933808a91f01feb7047e2943da41fb761

diff --git a/.github/workflows/benchmark_on_push.yml b/.github/workflows/benchmark_on_push.yml index 41eeb7d637..b0da71461e 100644 --- a/.github/workflows/benchmark_on_push.yml +++ b/.github/workflows/benchmark_on_push.yml @@ -13,10 +13,10 @@ jobs: runs-on: ubuntu-latest steps: - uses: actions/checkout@v4 - - name: Set up Python 3.8 + - name: Set up Python 3.12 uses: actions/setup-python@v5 with: - python-version: 3.8 + python-version: 3.12 - name: Install Linux system dependencies run: | diff --git a/.github/workflows/periodic_benchmarks.yml b/.github/workflows/periodic_benchmarks.yml index 276724d274..dd5cc39173 100644 --- a/.github/workflows/periodic_benchmarks.yml +++ b/.github/workflows/periodic_benchmarks.yml @@ -21,10 +21,10 @@ jobs: steps: - uses: actions/checkout@v4 - - name: Set up Python 3.8 + - name: Set up Python 3.12 uses: actions/setup-python@v5 with: - python-version: 3.8 + python-version: 3.12 - name: Install Linux system dependencies run: | @@ -58,10 +58,10 @@ jobs: needs: benchmarks runs-on: ubuntu-latest steps: - - name: Set up Python 3.8 + - name: Set up Python 3.12 uses: actions/setup-python@v5 with: - python-version: 3.8 + python-version: 3.12 - name: Install asv run: pip install asv diff --git a/.github/workflows/publish_pypi.yml b/.github/workflows/publish_pypi.yml index bbff62201d..d1c960ec43 100644 --- a/.github/workflows/publish_pypi.yml +++ b/.github/workflows/publish_pypi.yml @@ -35,7 +35,7 @@ jobs: - uses: actions/checkout@v4 - uses: actions/setup-python@v5 with: - python-version: 3.8 + python-version: 3.11 - name: Get number of cores on Windows id: get_num_cores @@ -107,7 +107,7 @@ jobs: - uses: actions/setup-python@v5 name: Set up Python with: - python-version: 3.8 + python-version: 3.11 - name: Clone pybind11 repo (no history) run: git clone --depth 1 --branch v2.11.1 https://github.com/pybind/pybind11.git diff --git a/.github/workflows/run_benchmarks_over_history.yml b/.github/workflows/run_benchmarks_over_history.yml index 4f7302a4a5..c8d6892e66 100644 --- a/.github/workflows/run_benchmarks_over_history.yml +++ b/.github/workflows/run_benchmarks_over_history.yml @@ -23,10 +23,10 @@ jobs: runs-on: ubuntu-latest steps: - uses: actions/checkout@v4 - - name: Set up Python 3.8 + - name: Set up Python 3.12 uses: actions/setup-python@v5 with: - python-version: 3.8 + python-version: 3.12 - name: Install nox and asv run: pip install -U pip nox asv - name: Fetch develop branch @@ -53,10 +53,10 @@ jobs: needs: benchmarks runs-on: ubuntu-latest steps: - - name: Set up Python 3.8 + - name: Set up Python 3.12 uses: actions/setup-python@v5 with: - python-version: 3.8 + python-version: 3.12 - name: Install asv run: pip install asv - name: Checkout pybamm-bench repo diff --git a/.github/workflows/update_version.yml b/.github/workflows/update_version.yml index 85f85cebe8..eb62469778 100644 --- a/.github/workflows/update_version.yml +++ b/.github/workflows/update_version.yml @@ -44,7 +44,7 @@ jobs: - name: Set up Python uses: actions/setup-python@v5 with: - python-version: 3.8 + python-version: 3.12 - name: Install dependencies run: | diff --git a/docs/source/user_guide/installation/gnu-linux-mac.rst b/docs/source/user_guide/installation/gnu-linux-mac.rst index 46f3d1281f..0be4b98e4c 100644 --- a/docs/source/user_guide/installation/gnu-linux-mac.rst +++ b/docs/source/user_guide/installation/gnu-linux-mac.rst @@ -95,10 +95,6 @@ Optional - JaxSolver Users can install ``jax`` and ``jaxlib`` to use the Jax solver. -.. note:: - - The Jax solver is only supported for Python versions 3.9 through 3.12. - .. code:: bash pip install "pybamm[jax]" diff --git a/docs/source/user_guide/installation/install-from-source.rst b/docs/source/user_guide/installation/install-from-source.rst index edcbdf79aa..827f6c2750 100644 --- a/docs/source/user_guide/installation/install-from-source.rst +++ b/docs/source/user_guide/installation/install-from-source.rst @@ -25,7 +25,7 @@ or download the source archive on the repository's homepage. To install PyBaMM, you will need: -- Python 3 (PyBaMM supports versions 3.8, 3.9, 3.10, 3.11, and 3.12) +- Python 3 (PyBaMM supports versions 3.9, 3.10, 3.11, and 3.12) - The Python headers file for your current Python version. - A BLAS library (for instance `openblas <https://www.openblas.net/>`_). - A C compiler (ex: ``gcc``). diff --git a/docs/source/user_guide/installation/windows.rst b/docs/source/user_guide/installation/windows.rst index d99d1f2eb2..02d9f8dd29 100644 --- a/docs/source/user_guide/installation/windows.rst +++ b/docs/source/user_guide/installation/windows.rst @@ -6,7 +6,7 @@ Windows Prerequisites ------------- -To use PyBaMM, you must have Python 3.8, 3.9, 3.10, 3.11, or 3.12 installed. +To use PyBaMM, you must have Python 3.9, 3.10, 3.11, or 3.12 installed. To install Python 3 download the installation files from `Python’s website <https://www.python.org/downloads/windows/>`__. Make sure to @@ -71,10 +71,6 @@ Optional - JaxSolver Users can install ``jax`` and ``jaxlib`` to use the Jax solver. -.. note:: - - The Jax solver is only supported for Python versions 3.9 through 3.12. - .. code:: bash pip install "pybamm[jax]" diff --git a/noxfile.py b/noxfile.py index 17d466e690..c93f3d808d 100644 --- a/noxfile.py +++ b/noxfile.py @@ -61,10 +61,7 @@ def run_coverage(session): """Run the coverage tests and generate an XML report.""" set_environment_variables(PYBAMM_ENV, session=session) session.install("coverage", silent=False) - if sys.version_info < (3, 9): - session.install("-e", ".[all,dev]", silent=False) - else: - session.install("-e", ".[all,dev,jax]", silent=False) + session.install("-e", ".[all,dev,jax]", silent=False) session.run("pytest", "--cov=pybamm", "--cov-report=xml", "tests/unit") @@ -72,10 +69,7 @@ def run_coverage(session): def run_integration(session): """Run the integration tests.""" set_environment_variables(PYBAMM_ENV, session=session) - if sys.version_info < (3, 9): - session.install("-e", ".[all,dev]", silent=False) - else: - session.install("-e", ".[all,dev,jax]", silent=False) + session.install("-e", ".[all,dev,jax]", silent=False) session.run("python", "run-tests.py", "--integration") @@ -90,10 +84,7 @@ def run_doctests(session): def run_unit(session): """Run the unit tests.""" set_environment_variables(PYBAMM_ENV, session=session) - if sys.version_info < (3, 9): - session.install("-e", ".[all,dev]", silent=False) - else: - session.install("-e", ".[all,dev,jax]", silent=False) + session.install("-e", ".[all,dev,jax]", silent=False) session.run("python", "run-tests.py", "--unit") @@ -129,36 +120,22 @@ def set_dev(session): # https://bitbucket.org/pybtex-devs/pybtex/issues/169/replace-pkg_resources-with # is fixed session.run(python, "-m", "pip", "install", "setuptools", external=True) - if sys.version_info < (3, 9): - session.run( - python, - "-m", - "pip", - "install", - "-e", - ".[all,dev]", - external=True, - ) - else: - session.run( - python, - "-m", - "pip", - "install", - "-e", - ".[all,dev,jax]", - external=True, - ) + session.run( + python, + "-m", + "pip", + "install", + "-e", + ".[all,dev,jax]", + external=True, + ) @nox.session(name="tests") def run_tests(session): """Run the unit tests and integration tests sequentially.""" set_environment_variables(PYBAMM_ENV, session=session) - if sys.version_info < (3, 9): - session.install("-e", ".[all,dev]", silent=False) - else: - session.install("-e", ".[all,dev,jax]", silent=False) + session.install("-e", ".[all,dev,jax]", silent=False) session.run("python", "run-tests.py", "--all") diff --git a/pybamm/expression_tree/concatenations.py b/pybamm/expression_tree/concatenations.py index a25ac8e0b3..11e695425f 100644 --- a/pybamm/expression_tree/concatenations.py +++ b/pybamm/expression_tree/concatenations.py @@ -8,7 +8,7 @@ import numpy as np import sympy from scipy.sparse import issparse, vstack -from typing import Sequence +from collections.abc import Sequence import pybamm diff --git a/pybamm/expression_tree/functions.py b/pybamm/expression_tree/functions.py index 7f43f0b7c2..6eb301a24c 100644 --- a/pybamm/expression_tree/functions.py +++ b/pybamm/expression_tree/functions.py @@ -6,7 +6,8 @@ import numpy as np from scipy import special import sympy -from typing import Sequence, Callable +from typing import Callable +from collections.abc import Sequence from typing_extensions import TypeVar import pybamm diff --git a/pybamm/expression_tree/interpolant.py b/pybamm/expression_tree/interpolant.py index 16714f9db3..10881d3084 100644 --- a/pybamm/expression_tree/interpolant.py +++ b/pybamm/expression_tree/interpolant.py @@ -4,7 +4,7 @@ from __future__ import annotations import numpy as np from scipy import interpolate -from typing import Sequence +from collections.abc import Sequence import numbers import pybamm diff --git a/pybamm/expression_tree/operations/unpack_symbols.py b/pybamm/expression_tree/operations/unpack_symbols.py index 1933eada76..97e1d08d98 100644 --- a/pybamm/expression_tree/operations/unpack_symbols.py +++ b/pybamm/expression_tree/operations/unpack_symbols.py @@ -2,7 +2,8 @@ # Helper function to unpack a symbol # from __future__ import annotations -from typing import TYPE_CHECKING, Sequence +from typing import TYPE_CHECKING +from collections.abc import Sequence if TYPE_CHECKING: # pragma: no cover import pybamm diff --git a/pybamm/expression_tree/symbol.py b/pybamm/expression_tree/symbol.py index 466e5b7be6..d50b8a6c89 100644 --- a/pybamm/expression_tree/symbol.py +++ b/pybamm/expression_tree/symbol.py @@ -8,7 +8,8 @@ import sympy from scipy.sparse import csr_matrix, issparse from functools import cached_property -from typing import TYPE_CHECKING, Sequence, cast +from typing import TYPE_CHECKING, cast +from collections.abc import Sequence import pybamm from pybamm.util import import_optional_dependency diff --git a/pybamm/solvers/idaklu_jax.py b/pybamm/solvers/idaklu_jax.py index 9a5315e120..ba1c80c1c4 100644 --- a/pybamm/solvers/idaklu_jax.py +++ b/pybamm/solvers/idaklu_jax.py @@ -5,7 +5,6 @@ import numbers from typing import Union -from typing import List from functools import lru_cache @@ -275,7 +274,7 @@ def jax_value( self, t: np.ndarray = None, inputs: Union[dict, None] = None, - output_variables: Union[List[str], None] = None, + output_variables: Union[list[str], None] = None, ): """Helper function to compute the gradient of a jaxified expression @@ -308,7 +307,7 @@ def jax_grad( self, t: np.ndarray = None, inputs: Union[dict, None] = None, - output_variables: Union[List[str], None] = None, + output_variables: Union[list[str], None] = None, ): """Helper function to compute the gradient of a jaxified expression diff --git a/pybamm/type_definitions.py b/pybamm/type_definitions.py index c3d0a56faa..376189b41a 100644 --- a/pybamm/type_definitions.py +++ b/pybamm/type_definitions.py @@ -1,9 +1,6 @@ -# -# Common type definitions for PyBaMM -# from __future__ import annotations -from typing import Union, List, Dict +from typing import Union from typing_extensions import TypeAlias import numpy as np import pybamm @@ -15,6 +12,6 @@ ChildValue: TypeAlias = Union[float, np.ndarray] ChildSymbol: TypeAlias = Union[float, np.ndarray, pybamm.Symbol] -DomainType: TypeAlias = Union[List[str], str, None] -AuxiliaryDomainType: TypeAlias = Union[Dict[str, str], None] -DomainsType: TypeAlias = Union[Dict[str, Union[List[str], str]], None] +DomainType: TypeAlias = Union[list[str], str, None] +AuxiliaryDomainType: TypeAlias = Union[dict[str, str], None] +DomainsType: TypeAlias = Union[dict[str, Union[list[str], str]], None] diff --git a/pyproject.toml b/pyproject.toml index 07b2909983..b8b1469ece 100644 --- a/pyproject.toml +++ b/pyproject.toml @@ -18,7 +18,7 @@ license = { file = "LICENSE.txt" } description = "Python Battery Mathematical Modelling" authors = [{name = "The PyBaMM Team", email = "[email protected]"}] maintainers = [{name = "The PyBaMM Team", email = "[email protected]"}] -requires-python = ">=3.8, <3.13" +requires-python = ">=3.9, <3.13" readme = {file = "README.md", content-type = "text/markdown"} classifiers = [ "Development Status :: 5 - Production/Stable", @@ -28,7 +28,6 @@ classifiers = [ "Programming Language :: Python", "Programming Language :: Python :: 3", "Programming Language :: Python :: 3 :: Only", - "Programming Language :: Python :: 3.8", "Programming Language :: Python :: 3.9", "Programming Language :: Python :: 3.10", "Programming Language :: Python :: 3.11", @@ -54,7 +53,6 @@ Releases = "https://github.com/pybamm-team/PyBaMM/releases" Changelog = "https://github.com/pybamm-team/PyBaMM/blob/develop/CHANGELOG.md" [project.optional-dependencies] -# For the generation of documentation docs = [ "sphinx>=6", "sphinx_rtd_theme>=0.5", @@ -77,7 +75,6 @@ docs = [ examples = [ "jupyter", ] -# Plotting functionality plot = [ "imageio>=2.3.0", # Note: matplotlib is loaded for debug plots, but to ensure PyBaMM runs @@ -85,7 +82,6 @@ plot = [ # outside of plot() methods. "matplotlib>=3.6.0", ] -# For the Citations class cite = [ "pybtex>=0.24.0", ] @@ -97,7 +93,6 @@ bpx = [ tqdm = [ "tqdm", ] -# Dependencies intended for use by developers dev = [ # For working with pre-commit hooks "pre-commit", diff --git a/scripts/install_KLU_Sundials.py b/scripts/install_KLU_Sundials.py index c30a3cf3e5..6224e40776 100755 --- a/scripts/install_KLU_Sundials.py +++ b/scripts/install_KLU_Sundials.py @@ -1,14 +1,3 @@ -# /// pyproject -# [run] -# requires-python = "">=3.8, <3.13"" -# dependencies = [ -# "cmake", -# ] -# -# [additional-info] -# repository = "https://github.com/pybamm-team/PyBaMM" -# documentation = "https://docs.pybamm.org" -# /// import os import subprocess import tarfile

Drop support for Python `3.8`, support Python `3.12`? NumPy>=1.26 and SciPy>=1.11 support Python 3.9+ now and already have wheels for Python 3.12. The blocker should be CasADi, which might have a new release soon because the [pre-release](https://github.com/casadi/casadi/releases/tag/nightly-py312) is available. In any case, we should be able to drop Python `3.8` support at this time and `3.12` can wait

I don't think we should drop support for Python 3.8 unless PyBaMM breaks on 3.8 or 3.8 reaches `end-of-life`. We can actually start testing PyBaMM on 3.12 in the CI. IIRC, `pybind11` has already migrated to Python 3.12, and we can fetch CasADi's nightly wheels in the 3.12 job. Yes, it would be good to try supporting Python 3.12, which comes tomorrow. `xarray` doesn't run CI against 3.12 yet, but they list their Python dependency as >=3.9. I think we should be fine with using the nightly wheels for CasADi, should we let the optional dependencies catch up later? > I don't think we should drop support for Python 3.8 unless PyBaMM breaks on 3.8 or 3.8 reaches end-of-life. That makes sense, I just checked and Python 3.8 EOL is in October 2024, so we have just about a year to remove it We should only build and ship pybamm wheels once every dependency works, but testing pybamm (just with the required dependencies) in the CI should be good. (release candidates of 3.12 have been available on GH Actions for quite some time now) Good idea, we could also do these checks in the nightly releases when ready (#3251) since it would help catch errors and deprecations early Copying from slack - > https://scientific-python.org/specs/ - like PEPs but for the Scientific Python ecosystem. I think we should follow them to stay up-to-date with other libraries. > > The first SPEC suggests removing Python 3.9 support by the end of this year, but I’m not sure if we should do that :grimacing: There are a lot of interesting things available in the Scientific Python ecosystem, perhaps we should also try using [`sp-repo-review`](https://learn.scientific-python.org/development/guides/repo-review/)? I guess this would be better off as a separate issue, please let me know if I should open one Some of the failing checks in repo review are intentional from our end (for example - we removed codespell, prettier, and isort as pre-commit hooks because they were doing more harm than good), and some of them are potential GSoC projects (for example - pybamm wide typing hints), but yes an issue to track repo review's suggestions would be nice. We could either implement the suggestions, discard them, or mark them as implemented. PS: we weren't using repo review before because it was Scikit-HEP specific, but now they've moved it to Scientific Python, which is nice! Makes sense. I will run the tool locally and see what we are missing, I cannot get it to to work inside the browser at all for some reason. There has been some discussion about a PR for adding type hints to PyBaMM models after they have been serialised in #3397 (or they were probably for the expression tree: not sure) in the developer meeting yesterday > Makes sense. I will run the tool locally and see what we are missing, I cannot get it to to work inside the browser at all for some reason. There has been some discussion about a PR for adding type hints to PyBaMM models after they have been serialised in #3397 (or they were probably for the expression tree: not sure) in the developer meeting yesterday I've been going through and adding type hints for the expression trees yes (they're more of a by-product from trying various methods for serialisation) - I'll make a PR asap for those. `casadi` now supports Python 3.12 (https://github.com/casadi/casadi/releases/tag/3.6.4) which checks all four of our core dependencies. I installed all dependencies with Python 3.12 locally without issues, except for `[jax]` (which will now require bumping the versions) and `[odes]` (I see that `numpy` removed `distutils` with Python 3.12+ which broke the wheel building process – I can file an issue upstream requesting for a fix). @Saransh-cpp: should we try for Python 3.12? We can still keep Python 3.8 as discussed above. I was supposed to work on #3443, which I can then close in the same PR to avoid merge conflicts. In any case, it will be merged only after the stable release, so we have a lot of time to look into any failures that occur. > should we try for Python 3.12? We can still keep Python 3.8 as discussed above. Yes! We can add a note in the docs about optional dependencies not yet supported for Python 3.12. > and [odes] Ah, I remember the maintainers of scikits.odes were waiting for NumPy's migration to meson to stop using `numpy.distutils`, but the repository is inactive. See my old PR https://github.com/bmcage/odes/pull/137 for some discussion on this. Creating an issue there should be good.

2024-04-03T17:37:07

0.0

[]

pybamm-team/PyBaMM

pybamm-team__PyBaMM-3866

7a8268710a4c6ca72f650c08f6f323a23f73b105

diff --git a/CHANGELOG.md b/CHANGELOG.md index 069093015a..41856b3fd7 100644 --- a/CHANGELOG.md +++ b/CHANGELOG.md @@ -18,6 +18,7 @@ ## Breaking changes +- Renamed "have_optional_dependency" to "import_optional_dependency" ([#3866](https://github.com/pybamm-team/PyBaMM/pull/3866)) - Integrated the `[latexify]` extra into the core PyBaMM package, deprecating the `pybamm[latexify]` set of optional dependencies. SymPy is now a required dependency and will be installed upon installing PyBaMM ([#3848](https://github.com/pybamm-team/PyBaMM/pull/3848)) - Renamed "testing" argument for plots to "show_plot" and flipped its meaning (show_plot=True is now the default and shows the plot) ([#3842](https://github.com/pybamm-team/PyBaMM/pull/3842)) - Dropped support for BPX version 0.3.0 and below ([#3414](https://github.com/pybamm-team/PyBaMM/pull/3414)) diff --git a/CONTRIBUTING.md b/CONTRIBUTING.md index b9800dcd61..fc8e848bb5 100644 --- a/CONTRIBUTING.md +++ b/CONTRIBUTING.md @@ -104,13 +104,13 @@ Only 'core pybamm' is installed by default. The others have to be specified expl PyBaMM utilizes optional dependencies to allow users to choose which additional libraries they want to use. Managing these optional dependencies and their imports is essential to provide flexibility to PyBaMM users. -PyBaMM provides a utility function `have_optional_dependency`, to check for the availability of optional dependencies within methods. This function can be used to conditionally import optional dependencies only if they are available. Here's how to use it: +PyBaMM provides a utility function `import_optional_dependency`, to check for the availability of optional dependencies within methods. This function can be used to conditionally import optional dependencies only if they are available. Here's how to use it: Optional dependencies should never be imported at the module level, but always inside methods. For example: ``` def use_pybtex(x,y,z): - pybtex = have_optional_dependency("pybtex") + pybtex = import_optional_dependency("pybtex") ... ``` @@ -118,7 +118,7 @@ While importing a specific module instead of an entire package/library: ```python def use_parse_file(x, y, z): - parse_file = have_optional_dependency("pybtex.database", "parse_file") + parse_file = import_optional_dependency("pybtex.database", "parse_file") ... ``` @@ -143,7 +143,7 @@ class TestUtil(TestCase): pybamm.function_using_pybtex(x, y, z) # Test that the function works when pybtex is available - sys.modules["pybtex"] = pybamm.util.have_optional_dependency("pybtex") + sys.modules["pybtex"] = pybamm.util.import_optional_dependency("pybtex") pybamm.function_using_pybtex(x, y, z) ``` diff --git a/pybamm/__init__.py b/pybamm/__init__.py index ab2e72ed28..c2654ea9cf 100644 --- a/pybamm/__init__.py +++ b/pybamm/__init__.py @@ -47,7 +47,7 @@ get_parameters_filepath, have_jax, install_jax, - have_optional_dependency, + import_optional_dependency, is_jax_compatible, get_git_commit_info, ) diff --git a/pybamm/citations.py b/pybamm/citations.py index da371bbd84..74f477c117 100644 --- a/pybamm/citations.py +++ b/pybamm/citations.py @@ -7,7 +7,7 @@ import os import warnings from sys import _getframe -from pybamm.util import have_optional_dependency +from pybamm.util import import_optional_dependency class Citations: @@ -74,7 +74,7 @@ def read_citations(self): """Reads the citations in `pybamm.CITATIONS.bib`. Other works can be cited by passing a BibTeX citation to :meth:`register`. """ - parse_file = have_optional_dependency("pybtex.database", "parse_file") + parse_file = import_optional_dependency("pybtex.database", "parse_file") citations_file = os.path.join(pybamm.root_dir(), "pybamm", "CITATIONS.bib") bib_data = parse_file(citations_file, bib_format="bibtex") for key, entry in bib_data.entries.items(): @@ -85,7 +85,7 @@ def _add_citation(self, key, entry): previous entry is overwritten """ - Entry = have_optional_dependency("pybtex.database", "Entry") + Entry = import_optional_dependency("pybtex.database", "Entry") # Check input types are correct if not isinstance(key, str) or not isinstance(entry, Entry): raise TypeError() @@ -151,8 +151,8 @@ def _parse_citation(self, key): key: str A BibTeX formatted citation """ - PybtexError = have_optional_dependency("pybtex.scanner", "PybtexError") - parse_string = have_optional_dependency("pybtex.database", "parse_string") + PybtexError = import_optional_dependency("pybtex.scanner", "PybtexError") + parse_string = import_optional_dependency("pybtex.database", "parse_string") try: # Parse string as a bibtex citation, and check that a citation was found bib_data = parse_string(key, bib_format="bibtex") @@ -219,7 +219,7 @@ def print(self, filename=None, output_format="text", verbose=False): """ # Parse citations that were not known keys at registration, but do not # fail if they cannot be parsed - pybtex = have_optional_dependency("pybtex") + pybtex = import_optional_dependency("pybtex") try: for key in self._unknown_citations: self._parse_citation(key) diff --git a/pybamm/expression_tree/functions.py b/pybamm/expression_tree/functions.py index 7f0986441f..7f43f0b7c2 100644 --- a/pybamm/expression_tree/functions.py +++ b/pybamm/expression_tree/functions.py @@ -10,7 +10,7 @@ from typing_extensions import TypeVar import pybamm -from pybamm.util import have_optional_dependency +from pybamm.util import import_optional_dependency class Function(pybamm.Symbol): @@ -98,7 +98,7 @@ def _function_diff(self, children: Sequence[pybamm.Symbol], idx: float): Derivative with respect to child number 'idx'. See :meth:`pybamm.Symbol._diff()`. """ - autograd = have_optional_dependency("autograd") + autograd = import_optional_dependency("autograd") # Store differentiated function, needed in case we want to convert to CasADi if self.derivative == "autograd": return Function( diff --git a/pybamm/expression_tree/symbol.py b/pybamm/expression_tree/symbol.py index 5c4bd5f097..e68e436749 100644 --- a/pybamm/expression_tree/symbol.py +++ b/pybamm/expression_tree/symbol.py @@ -11,7 +11,7 @@ from typing import TYPE_CHECKING, Sequence, cast import pybamm -from pybamm.util import have_optional_dependency +from pybamm.util import import_optional_dependency from pybamm.expression_tree.printing.print_name import prettify_print_name if TYPE_CHECKING: # pragma: no cover @@ -479,7 +479,7 @@ def render(self): # pragma: no cover """ Print out a visual representation of the tree (this node and its children) """ - anytree = have_optional_dependency("anytree") + anytree = import_optional_dependency("anytree") for pre, _, node in anytree.RenderTree(self): if isinstance(node, pybamm.Scalar) and node.name != str(node.value): print(f"{pre}{node.name} = {node.value}") @@ -498,7 +498,7 @@ def visualise(self, filename: str): filename to output, must end in ".png" """ - DotExporter = have_optional_dependency("anytree.exporter", "DotExporter") + DotExporter = import_optional_dependency("anytree.exporter", "DotExporter") # check that filename ends in .png. if filename[-4:] != ".png": raise ValueError("filename should end in .png") @@ -518,7 +518,7 @@ def relabel_tree(self, symbol: Symbol, counter: int): Finds all children of a symbol and assigns them a new id so that they can be visualised properly using the graphviz output """ - anytree = have_optional_dependency("anytree") + anytree = import_optional_dependency("anytree") name = symbol.name if name == "div": name = "∇⋅" @@ -561,7 +561,7 @@ def pre_order(self): a b """ - anytree = have_optional_dependency("anytree") + anytree = import_optional_dependency("anytree") return anytree.PreOrderIter(self) def __str__(self): diff --git a/pybamm/expression_tree/unary_operators.py b/pybamm/expression_tree/unary_operators.py index cb5bc905fd..9669c2596a 100644 --- a/pybamm/expression_tree/unary_operators.py +++ b/pybamm/expression_tree/unary_operators.py @@ -7,7 +7,7 @@ from scipy.sparse import csr_matrix, issparse import sympy import pybamm -from pybamm.util import have_optional_dependency +from pybamm.util import import_optional_dependency from pybamm.type_definitions import DomainsType @@ -450,7 +450,9 @@ def _unary_new_copy(self, child): def _sympy_operator(self, child): """Override :meth:`pybamm.UnaryOperator._sympy_operator`""" - sympy_Gradient = have_optional_dependency("sympy.vector.operators", "Gradient") + sympy_Gradient = import_optional_dependency( + "sympy.vector.operators", "Gradient" + ) return sympy_Gradient(child) @@ -484,7 +486,7 @@ def _unary_new_copy(self, child): def _sympy_operator(self, child): """Override :meth:`pybamm.UnaryOperator._sympy_operator`""" - sympy_Divergence = have_optional_dependency( + sympy_Divergence = import_optional_dependency( "sympy.vector.operators", "Divergence" ) return sympy_Divergence(child) diff --git a/pybamm/meshes/scikit_fem_submeshes.py b/pybamm/meshes/scikit_fem_submeshes.py index 82a7bd72f1..e52f58f069 100644 --- a/pybamm/meshes/scikit_fem_submeshes.py +++ b/pybamm/meshes/scikit_fem_submeshes.py @@ -5,7 +5,7 @@ from .meshes import SubMesh import numpy as np -from pybamm.util import have_optional_dependency +from pybamm.util import import_optional_dependency class ScikitSubMesh2D(SubMesh): @@ -27,7 +27,7 @@ class ScikitSubMesh2D(SubMesh): """ def __init__(self, edges, coord_sys, tabs): - skfem = have_optional_dependency("skfem") + skfem = import_optional_dependency("skfem") self.edges = edges self.nodes = dict.fromkeys(["y", "z"]) for var in self.nodes.keys(): diff --git a/pybamm/models/base_model.py b/pybamm/models/base_model.py index 194b8938af..007897370f 100644 --- a/pybamm/models/base_model.py +++ b/pybamm/models/base_model.py @@ -13,7 +13,6 @@ import pybamm from pybamm.expression_tree.operations.serialise import Serialise -import sympy class BaseModel: @@ -1185,8 +1184,7 @@ def latexify(self, filename=None, newline=True, output_variables=None): This will return first five model equations >>> model.latexify(newline=False)[1:5] """ - if sympy: - from pybamm.expression_tree.operations.latexify import Latexify + from pybamm.expression_tree.operations.latexify import Latexify return Latexify(self, filename, newline).latexify( output_variables=output_variables diff --git a/pybamm/plotting/plot.py b/pybamm/plotting/plot.py index cf5c972a87..4037ab8fbf 100644 --- a/pybamm/plotting/plot.py +++ b/pybamm/plotting/plot.py @@ -3,7 +3,7 @@ # import pybamm from .quick_plot import ax_min, ax_max -from pybamm.util import have_optional_dependency +from pybamm.util import import_optional_dependency def plot(x, y, ax=None, show_plot=True, **kwargs): @@ -27,7 +27,7 @@ def plot(x, y, ax=None, show_plot=True, **kwargs): Keyword arguments, passed to plt.plot """ - plt = have_optional_dependency("matplotlib.pyplot") + plt = import_optional_dependency("matplotlib.pyplot") if not isinstance(x, pybamm.Array): raise TypeError("x must be 'pybamm.Array'") diff --git a/pybamm/plotting/plot2D.py b/pybamm/plotting/plot2D.py index a37cd1e2ed..7d1f3c6bae 100644 --- a/pybamm/plotting/plot2D.py +++ b/pybamm/plotting/plot2D.py @@ -3,7 +3,7 @@ # import pybamm from .quick_plot import ax_min, ax_max -from pybamm.util import have_optional_dependency +from pybamm.util import import_optional_dependency def plot2D(x, y, z, ax=None, show_plot=True, **kwargs): @@ -27,7 +27,7 @@ def plot2D(x, y, z, ax=None, show_plot=True, **kwargs): only display the plot after plt.show() has been called. """ - plt = have_optional_dependency("matplotlib.pyplot") + plt = import_optional_dependency("matplotlib.pyplot") if not isinstance(x, pybamm.Array): raise TypeError("x must be 'pybamm.Array'") diff --git a/pybamm/plotting/plot_summary_variables.py b/pybamm/plotting/plot_summary_variables.py index 33642c4d5a..bd4db0ee6c 100644 --- a/pybamm/plotting/plot_summary_variables.py +++ b/pybamm/plotting/plot_summary_variables.py @@ -3,7 +3,7 @@ # import numpy as np import pybamm -from pybamm.util import have_optional_dependency +from pybamm.util import import_optional_dependency def plot_summary_variables( @@ -27,7 +27,7 @@ def plot_summary_variables( Keyword arguments, passed to plt.subplots. """ - plt = have_optional_dependency("matplotlib.pyplot") + plt = import_optional_dependency("matplotlib.pyplot") if isinstance(solutions, pybamm.Solution): solutions = [solutions] diff --git a/pybamm/plotting/plot_voltage_components.py b/pybamm/plotting/plot_voltage_components.py index 3b155b71de..0d1bb7b573 100644 --- a/pybamm/plotting/plot_voltage_components.py +++ b/pybamm/plotting/plot_voltage_components.py @@ -3,7 +3,7 @@ # import numpy as np -from pybamm.util import have_optional_dependency +from pybamm.util import import_optional_dependency from pybamm.simulation import Simulation from pybamm.solvers.solution import Solution @@ -42,7 +42,7 @@ def plot_voltage_components( solution = input_data.solution elif isinstance(input_data, Solution): solution = input_data - plt = have_optional_dependency("matplotlib.pyplot") + plt = import_optional_dependency("matplotlib.pyplot") # Set a default value for alpha, the opacity kwargs_fill = {"alpha": 0.6, **kwargs_fill} diff --git a/pybamm/plotting/quick_plot.py b/pybamm/plotting/quick_plot.py index c6d3054abb..dbb8f37555 100644 --- a/pybamm/plotting/quick_plot.py +++ b/pybamm/plotting/quick_plot.py @@ -5,7 +5,7 @@ import numpy as np import pybamm from collections import defaultdict -from pybamm.util import have_optional_dependency +from pybamm.util import import_optional_dependency class LoopList(list): @@ -46,7 +46,7 @@ def split_long_string(title, max_words=None): def close_plots(): """Close all open figures""" - plt = have_optional_dependency("matplotlib.pyplot") + plt = import_optional_dependency("matplotlib.pyplot") plt.close("all") @@ -473,10 +473,10 @@ def plot(self, t, dynamic=False): Dimensional time (in 'time_units') at which to plot. """ - plt = have_optional_dependency("matplotlib.pyplot") - gridspec = have_optional_dependency("matplotlib.gridspec") - cm = have_optional_dependency("matplotlib", "cm") - colors = have_optional_dependency("matplotlib", "colors") + plt = import_optional_dependency("matplotlib.pyplot") + gridspec = import_optional_dependency("matplotlib.gridspec") + cm = import_optional_dependency("matplotlib", "cm") + colors = import_optional_dependency("matplotlib", "colors") t_in_seconds = t * self.time_scaling_factor self.fig = plt.figure(figsize=self.figsize) @@ -674,8 +674,8 @@ def dynamic_plot(self, show_plot=True, step=None): continuous_update=False, ) else: - plt = have_optional_dependency("matplotlib.pyplot") - Slider = have_optional_dependency("matplotlib.widgets", "Slider") + plt = import_optional_dependency("matplotlib.pyplot") + Slider = import_optional_dependency("matplotlib.widgets", "Slider") # create an initial plot at time self.min_t self.plot(self.min_t, dynamic=True) @@ -779,8 +779,8 @@ def create_gif(self, number_of_images=80, duration=0.1, output_filename="plot.gi Name of the generated GIF file. """ - imageio = have_optional_dependency("imageio.v2") - plt = have_optional_dependency("matplotlib.pyplot") + imageio = import_optional_dependency("imageio.v2") + plt = import_optional_dependency("matplotlib.pyplot") # time stamps at which the images/plots will be created time_array = np.linspace(self.min_t, self.max_t, num=number_of_images) diff --git a/pybamm/simulation.py b/pybamm/simulation.py index c1d62a47e9..aa58022b6d 100644 --- a/pybamm/simulation.py +++ b/pybamm/simulation.py @@ -11,7 +11,7 @@ import sys from functools import lru_cache from datetime import timedelta -from pybamm.util import have_optional_dependency +from pybamm.util import import_optional_dependency from pybamm.expression_tree.operations.serialise import Serialise @@ -699,7 +699,7 @@ def solve( # check if a user has tqdm installed if showprogress: - tqdm = have_optional_dependency("tqdm") + tqdm = import_optional_dependency("tqdm") cycle_lengths = tqdm.tqdm( self.experiment.cycle_lengths, desc="Cycling", diff --git a/pybamm/spatial_methods/scikit_finite_element.py b/pybamm/spatial_methods/scikit_finite_element.py index 07a3c0e1be..e65e29f7f8 100644 --- a/pybamm/spatial_methods/scikit_finite_element.py +++ b/pybamm/spatial_methods/scikit_finite_element.py @@ -7,7 +7,7 @@ from scipy.sparse.linalg import inv import numpy as np -from pybamm.util import have_optional_dependency +from pybamm.util import import_optional_dependency class ScikitFiniteElement(pybamm.SpatialMethod): @@ -88,7 +88,7 @@ def gradient(self, symbol, discretised_symbol, boundary_conditions): to the y-component of the gradient and the second column corresponds to the z component of the gradient. """ - skfem = have_optional_dependency("skfem") + skfem = import_optional_dependency("skfem") domain = symbol.domain[0] mesh = self.mesh[domain] @@ -144,7 +144,7 @@ def gradient_matrix(self, symbol, boundary_conditions): :class:`pybamm.Matrix` The (sparse) finite element gradient matrix for the domain """ - skfem = have_optional_dependency("skfem") + skfem = import_optional_dependency("skfem") # get primary domain mesh domain = symbol.domain[0] mesh = self.mesh[domain] @@ -190,7 +190,7 @@ def laplacian(self, symbol, discretised_symbol, boundary_conditions): Contains the result of acting the discretised gradient on the child discretised_symbol """ - skfem = have_optional_dependency("skfem") + skfem = import_optional_dependency("skfem") domain = symbol.domain[0] mesh = self.mesh[domain] @@ -258,7 +258,7 @@ def stiffness_matrix(self, symbol, boundary_conditions): :class:`pybamm.Matrix` The (sparse) finite element stiffness matrix for the domain """ - skfem = have_optional_dependency("skfem") + skfem = import_optional_dependency("skfem") # get primary domain mesh domain = symbol.domain[0] mesh = self.mesh[domain] @@ -321,7 +321,7 @@ def definite_integral_matrix(self, child, vector_type="row"): :class:`pybamm.Matrix` The finite element integral vector for the domain """ - skfem = have_optional_dependency("skfem") + skfem = import_optional_dependency("skfem") # get primary domain mesh domain = child.domain[0] mesh = self.mesh[domain] @@ -383,7 +383,7 @@ def boundary_integral_vector(self, domain, region): :class:`pybamm.Matrix` The finite element integral vector for the domain """ - skfem = have_optional_dependency("skfem") + skfem = import_optional_dependency("skfem") # get primary domain mesh mesh = self.mesh[domain[0]] @@ -501,7 +501,7 @@ def assemble_mass_form(self, symbol, boundary_conditions, region="interior"): :class:`pybamm.Matrix` The (sparse) mass matrix for the spatial method. """ - skfem = have_optional_dependency("skfem") + skfem = import_optional_dependency("skfem") # get primary domain mesh domain = symbol.domain[0] mesh = self.mesh[domain] diff --git a/pybamm/util.py b/pybamm/util.py index 1149327cf7..0722b8174b 100644 --- a/pybamm/util.py +++ b/pybamm/util.py @@ -353,24 +353,19 @@ def install_jax(arguments=None): # pragma: no cover # https://docs.pybamm.org/en/latest/source/user_guide/contributing.html#managing-optional-dependencies-and-their-imports -def have_optional_dependency(module_name, attribute=None): +def import_optional_dependency(module_name, attribute=None): err_msg = f"Optional dependency {module_name} is not available. See https://docs.pybamm.org/en/latest/source/user_guide/installation/index.html#optional-dependencies for more details." try: - # Attempt to import the specified module module = importlib.import_module(module_name) - if attribute: - # If an attribute is specified, check if it's available if hasattr(module, attribute): imported_attribute = getattr(module, attribute) - return imported_attribute # Return the imported attribute + # Return the imported attribute + return imported_attribute else: - # Raise an ModuleNotFoundError if the attribute is not available raise ModuleNotFoundError(err_msg) # pragma: no cover else: # Return the entire module if no attribute is specified return module - except ModuleNotFoundError: - # Raise an ModuleNotFoundError if the module or attribute is not available raise ModuleNotFoundError(err_msg)

Rename `have_optional_dependency` to `import_optional_dependency` We use `have_optional_dependency()` to import optional dependencies across codebase which is a wrapper over `importlib.import_module()` to throw a helpful error message.

2024-03-05T14:44:19

0.0

[]

pybamm-team/PyBaMM

pybamm-team__PyBaMM-3848

ed6df7e8b444f7aa1fd18dea48254592ef2eb3f4

diff --git a/CHANGELOG.md b/CHANGELOG.md index 6cade77684..069093015a 100644 --- a/CHANGELOG.md +++ b/CHANGELOG.md @@ -18,6 +18,7 @@ ## Breaking changes +- Integrated the `[latexify]` extra into the core PyBaMM package, deprecating the `pybamm[latexify]` set of optional dependencies. SymPy is now a required dependency and will be installed upon installing PyBaMM ([#3848](https://github.com/pybamm-team/PyBaMM/pull/3848)) - Renamed "testing" argument for plots to "show_plot" and flipped its meaning (show_plot=True is now the default and shows the plot) ([#3842](https://github.com/pybamm-team/PyBaMM/pull/3842)) - Dropped support for BPX version 0.3.0 and below ([#3414](https://github.com/pybamm-team/PyBaMM/pull/3414)) diff --git a/docs/source/user_guide/installation/index.rst b/docs/source/user_guide/installation/index.rst index bf83e38835..1a773c65a5 100644 --- a/docs/source/user_guide/installation/index.rst +++ b/docs/source/user_guide/installation/index.rst @@ -59,15 +59,17 @@ Required dependencies PyBaMM requires the following dependencies. -================================================================ ========================== -Package Minimum supported version -================================================================ ========================== -`NumPy <https://numpy.org>`__ 1.23.5 -`SciPy <https://docs.scipy.org/doc/scipy/>`__ 1.9.3 -`CasADi <https://web.casadi.org/docs/>`__ 3.6.3 -`Xarray <https://docs.xarray.dev/en/stable/>`__ 2022.6.0 -`Anytree <https://anytree.readthedocs.io/en/stable/>`__ 2.8.0 -================================================================ ========================== +=================================================================== ========================== +Package Minimum supported version +=================================================================== ========================== +`NumPy <https://numpy.org>`__ 1.23.5 +`SciPy <https://docs.scipy.org/doc/scipy/>`__ 1.9.3 +`CasADi <https://web.casadi.org/docs/>`__ 3.6.3 +`Xarray <https://docs.xarray.dev/en/stable/>`__ 2022.6.0 +`Anytree <https://anytree.readthedocs.io/en/stable/>`__ 2.8.0 +`SymPy <https://docs.sympy.org/latest/index.html>`__ 1.9.3 +`typing-extensions <https://pypi.org/project/typing-extensions/>`__ 4.10.0 +=================================================================== ========================== .. _install.optional_dependencies: @@ -174,19 +176,6 @@ Dependency Minimum Version p `pybtex <https://docs.pybtex.org/>`__ 0.24.0 cite BibTeX-compatible bibliography processor. =========================================================== ================== ================== ========================================= -.. _install.latexify_dependencies: - -Latexify dependencies -^^^^^^^^^^^^^^^^^^^^^ - -Installable with ``pip install "pybamm[latexify]"`` - -=========================================================== ================== ================== ========================= -Dependency Minimum Version pip extra Notes -=========================================================== ================== ================== ========================= -`sympy <https://docs.sympy.org/latest/index.html>`__ 1.9.3 latexify For symbolic mathematics. -=========================================================== ================== ================== ========================= - .. _install.bpx_dependencies: bpx dependencies diff --git a/pybamm/expression_tree/array.py b/pybamm/expression_tree/array.py index 0bb6168a7c..dbaa5cd144 100644 --- a/pybamm/expression_tree/array.py +++ b/pybamm/expression_tree/array.py @@ -4,14 +4,10 @@ from __future__ import annotations import numpy as np from scipy.sparse import csr_matrix, issparse -from typing import TYPE_CHECKING import pybamm -from pybamm.util import have_optional_dependency from pybamm.type_definitions import DomainType, AuxiliaryDomainType, DomainsType - -if TYPE_CHECKING: # pragma: no cover - import sympy +import sympy class Array(pybamm.Symbol): @@ -157,7 +153,6 @@ def is_constant(self): def to_equation(self) -> sympy.Array: """Returns the value returned by the node when evaluated.""" - sympy = have_optional_dependency("sympy") entries_list = self.entries.tolist() return sympy.Array(entries_list) diff --git a/pybamm/expression_tree/binary_operators.py b/pybamm/expression_tree/binary_operators.py index d10844798b..a8086ebdd1 100644 --- a/pybamm/expression_tree/binary_operators.py +++ b/pybamm/expression_tree/binary_operators.py @@ -5,11 +5,11 @@ import numbers import numpy as np +import sympy from scipy.sparse import csr_matrix, issparse import functools import pybamm -from pybamm.util import have_optional_dependency from typing import Callable, cast @@ -180,7 +180,6 @@ def _sympy_operator(self, left, right): def to_equation(self): """Convert the node and its subtree into a SymPy equation.""" - sympy = have_optional_dependency("sympy") if self.print_name is not None: return sympy.Symbol(self.print_name) else: @@ -388,7 +387,6 @@ def _binary_evaluate(self, left, right): def _sympy_operator(self, left, right): """Override :meth:`pybamm.BinaryOperator._sympy_operator`""" - sympy = have_optional_dependency("sympy") left = sympy.Matrix(left) right = sympy.Matrix(right) return left * right @@ -737,7 +735,6 @@ def _binary_new_copy( def _sympy_operator(self, left, right): """Override :meth:`pybamm.BinaryOperator._sympy_operator`""" - sympy = have_optional_dependency("sympy") return sympy.Min(left, right) @@ -782,7 +779,6 @@ def _binary_new_copy( def _sympy_operator(self, left, right): """Override :meth:`pybamm.BinaryOperator._sympy_operator`""" - sympy = have_optional_dependency("sympy") return sympy.Max(left, right) diff --git a/pybamm/expression_tree/concatenations.py b/pybamm/expression_tree/concatenations.py index 06718c311d..52d22b062b 100644 --- a/pybamm/expression_tree/concatenations.py +++ b/pybamm/expression_tree/concatenations.py @@ -6,11 +6,11 @@ from collections import defaultdict import numpy as np +import sympy from scipy.sparse import issparse, vstack from typing import Sequence import pybamm -from pybamm.util import have_optional_dependency class Concatenation(pybamm.Symbol): @@ -159,7 +159,6 @@ def is_constant(self): def _sympy_operator(self, *children): """Apply appropriate SymPy operators.""" - sympy = have_optional_dependency("sympy") self.concat_latex = tuple(map(sympy.latex, children)) if self.print_name is not None: diff --git a/pybamm/expression_tree/functions.py b/pybamm/expression_tree/functions.py index 72c9d4074a..7f0986441f 100644 --- a/pybamm/expression_tree/functions.py +++ b/pybamm/expression_tree/functions.py @@ -5,6 +5,7 @@ import numpy as np from scipy import special +import sympy from typing import Sequence, Callable from typing_extensions import TypeVar @@ -210,7 +211,6 @@ def _sympy_operator(self, child): def to_equation(self): """Convert the node and its subtree into a SymPy equation.""" - sympy = have_optional_dependency("sympy") if self.print_name is not None: return sympy.Symbol(self.print_name) else: @@ -275,7 +275,6 @@ def _function_new_copy(self, children): def _sympy_operator(self, child): """Apply appropriate SymPy operators.""" - sympy = have_optional_dependency("sympy") class_name = self.__class__.__name__.lower() sympy_function = getattr(sympy, class_name) return sympy_function(child) @@ -332,7 +331,6 @@ def _function_diff(self, children, idx): def _sympy_operator(self, child): """Override :meth:`pybamm.Function._sympy_operator`""" - sympy = have_optional_dependency("sympy") return sympy.asinh(child) @@ -360,7 +358,6 @@ def _function_diff(self, children, idx): def _sympy_operator(self, child): """Override :meth:`pybamm.Function._sympy_operator`""" - sympy = have_optional_dependency("sympy") return sympy.atan(child) diff --git a/pybamm/expression_tree/independent_variable.py b/pybamm/expression_tree/independent_variable.py index 0dca6dba46..01e9dd661a 100644 --- a/pybamm/expression_tree/independent_variable.py +++ b/pybamm/expression_tree/independent_variable.py @@ -6,7 +6,6 @@ import numpy as np import pybamm -from pybamm.util import have_optional_dependency from pybamm.type_definitions import DomainType, AuxiliaryDomainType, DomainsType KNOWN_COORD_SYS = ["cartesian", "cylindrical polar", "spherical polar"] @@ -58,7 +57,6 @@ def _jac(self, variable) -> pybamm.Scalar: def to_equation(self) -> sympy.Symbol: """Convert the node and its subtree into a SymPy equation.""" - sympy = have_optional_dependency("sympy") if self.print_name is not None: return sympy.Symbol(self.print_name) else: @@ -102,7 +100,6 @@ def _evaluate_for_shape(self): def to_equation(self): """Convert the node and its subtree into a SymPy equation.""" - sympy = have_optional_dependency("sympy") return sympy.Symbol("t") diff --git a/pybamm/expression_tree/operations/latexify.py b/pybamm/expression_tree/operations/latexify.py index c16ab4b83d..f1c3734deb 100644 --- a/pybamm/expression_tree/operations/latexify.py +++ b/pybamm/expression_tree/operations/latexify.py @@ -9,7 +9,7 @@ import pybamm from pybamm.expression_tree.printing.sympy_overrides import custom_print_func -from pybamm.util import have_optional_dependency +import sympy def get_rng_min_max_name(rng, min_or_max): @@ -89,7 +89,6 @@ def _get_bcs_displays(self, var): Returns a list of boundary condition equations with ranges in front of the equations. """ - sympy = have_optional_dependency("sympy") bcs_eqn_list = [] bcs = self.model.boundary_conditions.get(var, None) @@ -120,7 +119,6 @@ def _get_bcs_displays(self, var): def _get_param_var(self, node): """Returns a list of parameters and a list of variables.""" - sympy = have_optional_dependency("sympy") param_list = [] var_list = [] dfs_nodes = [node] @@ -163,7 +161,6 @@ def _get_param_var(self, node): return param_list, var_list def latexify(self, output_variables=None): - sympy = have_optional_dependency("sympy") # Voltage is the default output variable if it exists if output_variables is None: if "Voltage [V]" in self.model.variables: diff --git a/pybamm/expression_tree/parameter.py b/pybamm/expression_tree/parameter.py index e646ff234d..4294d7ebc1 100644 --- a/pybamm/expression_tree/parameter.py +++ b/pybamm/expression_tree/parameter.py @@ -5,13 +5,11 @@ import sys import numpy as np -from typing import TYPE_CHECKING, Literal +from typing import Literal -if TYPE_CHECKING: # pragma: no cover - import sympy +import sympy import pybamm -from pybamm.util import have_optional_dependency class Parameter(pybamm.Symbol): @@ -48,7 +46,6 @@ def is_constant(self) -> Literal[False]: def to_equation(self) -> sympy.Symbol: """Convert the node and its subtree into a SymPy equation.""" - sympy = have_optional_dependency("sympy") if self.print_name is not None: return sympy.Symbol(self.print_name) else: @@ -244,7 +241,6 @@ def _evaluate_for_shape(self): def to_equation(self) -> sympy.Symbol: """Convert the node and its subtree into a SymPy equation.""" - sympy = have_optional_dependency("sympy") if self.print_name is not None: return sympy.Symbol(self.print_name) else: diff --git a/pybamm/expression_tree/scalar.py b/pybamm/expression_tree/scalar.py index 26bbabfcf0..a860ae45ec 100644 --- a/pybamm/expression_tree/scalar.py +++ b/pybamm/expression_tree/scalar.py @@ -3,10 +3,10 @@ # from __future__ import annotations import numpy as np +import sympy from typing import Literal import pybamm -from pybamm.util import have_optional_dependency from pybamm.type_definitions import Numeric @@ -87,7 +87,6 @@ def is_constant(self) -> Literal[True]: def to_equation(self): """Returns the value returned by the node when evaluated.""" - sympy = have_optional_dependency("sympy") if self.print_name is not None: return sympy.Symbol(self.print_name) else: diff --git a/pybamm/expression_tree/symbol.py b/pybamm/expression_tree/symbol.py index 70f4e82db6..5c4bd5f097 100644 --- a/pybamm/expression_tree/symbol.py +++ b/pybamm/expression_tree/symbol.py @@ -5,6 +5,7 @@ import numbers import numpy as np +import sympy from scipy.sparse import csr_matrix, issparse from functools import lru_cache, cached_property from typing import TYPE_CHECKING, Sequence, cast @@ -1056,7 +1057,6 @@ def print_name(self, name): self._print_name = prettify_print_name(name) def to_equation(self): - sympy = have_optional_dependency("sympy") return sympy.Symbol(str(self.name)) def to_json(self): diff --git a/pybamm/expression_tree/unary_operators.py b/pybamm/expression_tree/unary_operators.py index 319499a9fc..cb5bc905fd 100644 --- a/pybamm/expression_tree/unary_operators.py +++ b/pybamm/expression_tree/unary_operators.py @@ -5,6 +5,7 @@ import numpy as np from scipy.sparse import csr_matrix, issparse +import sympy import pybamm from pybamm.util import have_optional_dependency from pybamm.type_definitions import DomainsType @@ -108,7 +109,6 @@ def _sympy_operator(self, child): def to_equation(self): """Convert the node and its subtree into a SymPy equation.""" - sympy = have_optional_dependency("sympy") if self.print_name is not None: return sympy.Symbol(self.print_name) else: @@ -672,7 +672,6 @@ def _evaluates_on_edges(self, dimension: str) -> bool: def _sympy_operator(self, child): """Override :meth:`pybamm.UnaryOperator._sympy_operator`""" - sympy = have_optional_dependency("sympy") return sympy.Integral(child, sympy.Symbol("xn")) @@ -996,7 +995,6 @@ def _unary_new_copy(self, child): def _sympy_operator(self, child): """Override :meth:`pybamm.UnaryOperator._sympy_operator`""" - sympy = have_optional_dependency("sympy") if ( self.child.domain[0] in ["negative particle", "positive particle"] and self.side == "right" diff --git a/pybamm/expression_tree/variable.py b/pybamm/expression_tree/variable.py index eb0d90cdb6..238dab05fc 100644 --- a/pybamm/expression_tree/variable.py +++ b/pybamm/expression_tree/variable.py @@ -5,7 +5,7 @@ import numpy as np import numbers import pybamm -from pybamm.util import have_optional_dependency +import sympy from pybamm.type_definitions import ( DomainType, AuxiliaryDomainType, @@ -135,7 +135,6 @@ def _evaluate_for_shape(self): def to_equation(self): """Convert the node and its subtree into a SymPy equation.""" - sympy = have_optional_dependency("sympy") if self.print_name is not None: return sympy.Symbol(self.print_name) else: diff --git a/pybamm/models/base_model.py b/pybamm/models/base_model.py index b6b5a9b2da..194b8938af 100644 --- a/pybamm/models/base_model.py +++ b/pybamm/models/base_model.py @@ -13,7 +13,7 @@ import pybamm from pybamm.expression_tree.operations.serialise import Serialise -from pybamm.util import have_optional_dependency +import sympy class BaseModel: @@ -1185,7 +1185,6 @@ def latexify(self, filename=None, newline=True, output_variables=None): This will return first five model equations >>> model.latexify(newline=False)[1:5] """ - sympy = have_optional_dependency("sympy") if sympy: from pybamm.expression_tree.operations.latexify import Latexify diff --git a/pyproject.toml b/pyproject.toml index 13ea6df786..65e8f8fb48 100644 --- a/pyproject.toml +++ b/pyproject.toml @@ -41,6 +41,8 @@ dependencies = [ "casadi>=3.6.3", "xarray>=2022.6.0", "anytree>=2.8.0", + "sympy>=1.12", + "typing-extensions>=4.10.0", ] [project.urls] @@ -86,10 +88,6 @@ plot = [ cite = [ "pybtex>=0.24.0", ] -# To generate LaTeX strings -latexify = [ - "sympy>=1.12", -] # Battery Parameter eXchange format bpx = [ "bpx>=0.4.0",

[Bug]: non-required dependency `typing_extensions` is imported by default ### PyBaMM Version 24.1 ### Python Version 3.8-3.12 ### Describe the bug There might be some instance in #3578 where non required dependency has been imported by default, which is causing failure, [related build logs](https://github.com/arjxn-py/PyBaMM/actions/runs/8032837736/job/21942676799). ### Steps to Reproduce * try installing latest built wheels on develop * do `import pybamm` ### Relevant log output _No response_

Note: this happens with `sympy` as well Also, it's weird that the tests for `have_optional_dependency` in `test_util.py` didn't catch `typing-extensions` and `sympy` – is this because we are not mocking all of the optional dependencies inside that test case? IMO, we should mock all of the optional dependencies to ensure that this problem doesn't get repeated in the future.

2024-02-28T17:50:20

0.0

[]

pybamm-team/PyBaMM

pybamm-team__PyBaMM-3829

84eb098e3140d774e39ab3d3c842c22462faf4aa

diff --git a/CHANGELOG.md b/CHANGELOG.md index bdacb00ae3..067e6ee025 100644 --- a/CHANGELOG.md +++ b/CHANGELOG.md @@ -9,6 +9,7 @@ ## Bug Fixes +- Fixed a bug where 1+1D and 2+1D models would not work with voltage or power controlled experiments([#3829](https://github.com/pybamm-team/PyBaMM/pull/3829)) - Updated `_steps_util.py` to throw a specific exception when drive cycle starts at t>0 ([#3756](https://github.com/pybamm-team/PyBaMM/pull/3756)) - Updated `plot_voltage_components.py` to support both `Simulation` and `Solution` objects. Added new methods in both `Simulation` and `Solution` classes for allow the syntax `simulation.plot_voltage_components` and `solution.plot_voltage_components`. Updated `test_plot_voltage_components.py` to reflect these changes ([#3723](https://github.com/pybamm-team/PyBaMM/pull/3723)). - The SEI thickness decreased at some intervals when the 'electron-migration limited' model was used. It has been corrected ([#3622](https://github.com/pybamm-team/PyBaMM/pull/3622)) diff --git a/pybamm/models/submodels/current_collector/potential_pair.py b/pybamm/models/submodels/current_collector/potential_pair.py index f2fd3aee83..68a9066da3 100644 --- a/pybamm/models/submodels/current_collector/potential_pair.py +++ b/pybamm/models/submodels/current_collector/potential_pair.py @@ -58,13 +58,12 @@ def set_algebraic(self, variables): } def set_initial_conditions(self, variables): - applied_current = self.param.current_with_time phi_s_cn = variables["Negative current collector potential [V]"] i_boundary_cc = variables["Current collector current density [A.m-2]"] self.initial_conditions = { phi_s_cn: pybamm.Scalar(0), - i_boundary_cc: applied_current, + i_boundary_cc: pybamm.Scalar(0), }

[Bug]: pouch models fail with voltage or power control ### PyBaMM Version 24.1 ### Python Version 3.11 ### Describe the bug Pouch cell simulations with voltage or power control fail with the error ``` TypeError: Initial conditions cannot contain 'Variable' objects, but 'Variable(-0x4b072f39e284004e, Current variable [A], children=[], domains={})' found in initial conditions for 'Current collector current density [A.m-2]' ``` ### Steps to Reproduce ```python3 import pybamm model = pybamm.lithium_ion.DFN( { "cell geometry": "pouch", "current collector": "potential pair", "dimensionality": 1, "thermal": "x-lumped" }, ) experiment = pybamm.Experiment(["Hold at 4.2 V until C/20 (5 seconds period)"]) sim = pybamm.Simulation(model, experiment = experiment) sol = sim.solve() ``` ### Relevant log output _No response_

2024-02-19T22:57:01

0.0

[]

pybamm-team/PyBaMM

pybamm-team__PyBaMM-3811

84bfd6159baace67bad87d231c945f241ac20d50

diff --git a/noxfile.py b/noxfile.py index 380fae6af6..b74cfbab64 100644 --- a/noxfile.py +++ b/noxfile.py @@ -16,6 +16,7 @@ PYBAMM_ENV = { "SUNDIALS_INST": f"{homedir}/.local", "LD_LIBRARY_PATH": f"{homedir}/.local/lib", + "PIP_NO_BINARY": "scikits.odes", } VENV_DIR = Path("./venv").resolve() @@ -68,6 +69,15 @@ def run_coverage(session): if sys.version_info > (3, 12): session.install("-e", ".[all,jax]", silent=False) else: + session.run_always( + sys.executable, + "-m", + "pip", + "cache", + "remove", + "scikits.odes", + external=True, + ) session.install("-e", ".[all,jax,odes]", silent=False) else: if sys.version_info < (3, 9): @@ -91,6 +101,15 @@ def run_integration(session): if sys.version_info > (3, 12): session.install("-e", ".[all,jax]", silent=False) else: + session.run_always( + sys.executable, + "-m", + "pip", + "cache", + "remove", + "scikits.odes", + external=True, + ) session.install("-e", ".[all,jax,odes]", silent=False) else: if sys.version_info < (3, 9): @@ -123,6 +142,15 @@ def run_unit(session): if sys.version_info > (3, 12): session.install("-e", ".[all,jax]", silent=False) else: + session.run_always( + sys.executable, + "-m", + "pip", + "cache", + "remove", + "scikits.odes", + external=True, + ) session.install("-e", ".[all,jax,odes]", silent=False) else: if sys.version_info < (3, 9): @@ -180,6 +208,15 @@ def set_dev(session): external=True, ) else: + session.run_always( + sys.executable, + "-m", + "pip", + "cache", + "remove", + "scikits.odes", + external=True, + ) session.run( python, "-m", @@ -224,6 +261,15 @@ def run_tests(session): if sys.version_info > (3, 12): session.install("-e", ".[all,jax]", silent=False) else: + session.run_always( + sys.executable, + "-m", + "pip", + "cache", + "remove", + "scikits.odes", + external=True, + ) session.install("-e", ".[all,jax,odes]", silent=False) else: if sys.version_info < (3, 9):

Fortran compiler (`FC` variable) missing in macOS M-series runners, breaks SUNDIALS compilation See https://github.com/agriyakhetarpal/PyBaMM/actions/runs/7794200267/job/21255141901

I will look into the issue of the failing Python 3.11 unit tests here since that is a small fix

2024-02-07T18:48:07

0.0

[]

pybamm-team/PyBaMM

pybamm-team__PyBaMM-3795

705baff8689106b546ab06746b91a4b8e672123f

diff --git a/pybamm/batch_study.py b/pybamm/batch_study.py index 4f1c37e71c..1503c3048c 100644 --- a/pybamm/batch_study.py +++ b/pybamm/batch_study.py @@ -196,7 +196,8 @@ def create_gif(self, number_of_images=80, duration=0.1, output_filename="plot.gi Name of the generated GIF file. """ - + if not hasattr(self, "sims"): + raise ValueError("The simulations have not been solved yet.") if self.quick_plot is None: self.quick_plot = pybamm.QuickPlot(self.sims) diff --git a/pybamm/simulation.py b/pybamm/simulation.py index 73c8188be7..a4c0d4bb32 100644 --- a/pybamm/simulation.py +++ b/pybamm/simulation.py @@ -1072,6 +1072,8 @@ def create_gif(self, number_of_images=80, duration=0.1, output_filename="plot.gi Name of the generated GIF file. """ + if self.solution is None: + raise ValueError("The simulation has not been solved yet.") if self.quick_plot is None: self.quick_plot = pybamm.QuickPlot(self._solution)

Check if the solution exists before creating a GIF (`Simulation.create_gif` and `BatchStudy.create_gif`) I liked how #3723 checks if the simulations have been solved before proceeding ahead to the plotting functionality. It would be nice to have similar checks in - https://github.com/pybamm-team/PyBaMM/blob/c3def31a04df163769851733657d9b49b3700bca/pybamm/batch_study.py#L184-L187 and https://github.com/pybamm-team/PyBaMM/blob/c3def31a04df163769851733657d9b49b3700bca/pybamm/simulation.py#L1047-L1050

Would like to take this on. Hi @prady0t are you still working on this issue? yes @Saransh-cpp BatchStudy class doesn't have an attribute `solution` unlike Simulation. How to check for empty solutions there? You can check if the `self.sims` list is empty and throw an error if it is - https://github.com/pybamm-team/PyBaMM/blob/c3def31a04df163769851733657d9b49b3700bca/pybamm/batch_study.py#L116 The simulations are being initialised and solved in the `solve()` method, so this would be a good indication of if the user has called the `solve()` method before calling the `create_gif()` method. why is https://github.com/pybamm-team/PyBaMM/blob/063a3830afd5547f7a78e283ae81058465cbdd3e/pybamm/batch_study.py#L116 declared outside `__init__` should'nt it be inside it?

2024-02-01T09:30:59

0.0

[]

pybamm-team/PyBaMM

pybamm-team__PyBaMM-3708

64681dd7ee9ca6e9c6c18a9e704e5d984a90466c

diff --git a/CHANGELOG.md b/CHANGELOG.md index 17adc3a31f..89b104298c 100644 --- a/CHANGELOG.md +++ b/CHANGELOG.md @@ -1,5 +1,8 @@ # [Unreleased](https://github.com/pybamm-team/PyBaMM/) +## Bug Fixes + +- Fixed a bug where if the first step(s) in a cycle are skipped then the cycle solution started from the model's initial conditions instead of from the last state of the previous cycle ([#3708](https://github.com/pybamm-team/PyBaMM/pull/3708)) # [v24.1rc0](https://github.com/pybamm-team/PyBaMM/tree/v24.1rc0) - 2024-01-31 ## Features diff --git a/pybamm/simulation.py b/pybamm/simulation.py index c95ab3039c..8a6150cc4e 100644 --- a/pybamm/simulation.py +++ b/pybamm/simulation.py @@ -839,7 +839,20 @@ def solve( steps.append(step_solution) - cycle_solution = cycle_solution + step_solution + # If there haven't been any successful steps yet in this cycle, then + # carry the solution over from the previous cycle (but + # `step_solution` should still be an EmptySolution so that in the + # list of returned step solutions we can see which steps were + # skipped) + if ( + cycle_solution is None + and isinstance(step_solution, pybamm.EmptySolution) + and not isinstance(current_solution, pybamm.EmptySolution) + ): + cycle_solution = current_solution.last_state + else: + cycle_solution = cycle_solution + step_solution + current_solution = cycle_solution callbacks.on_step_end(logs)

[Bug]: Some variables not copied correctly if experiment steps are skipped ### PyBaMM Version 23,9 ### Python Version 3.11 ### Describe the bug Sometimes experiement steps are skipped (e.g. skip a charge if you're already at that voltage). When this happens the solver returns an `EmptySolution`. The next step of the experiment then starts from the wrong point. ### Steps to Reproduce ```python3 import pybamm import matplotlib.pyplot as plt model = pybamm.lithium_ion.SPMe( { "SEI": "solvent-diffusion limited", } ) exp = pybamm.Experiment( [(f"Rest for 24 hours (1 hour period)",)] + [ ( "Charge at C/3 until 4.1 V", "Hold at 4.1V until C/20", "Discharge at C/3 until 2.5 V", ) ] ) sim1 = pybamm.Simulation(model, experiment=exp) sim1.solve(initial_soc=1) sim2 = pybamm.Simulation(model, experiment=exp) sim2.solve(initial_soc=0.9) sols = [sim1.solution, sim2.solution] labels = ["SOC=1", "SOC=0.9"] fig, ax = plt.subplots(2, 2, figsize=(10, 4)) for sol, label in zip(sols, labels): ax[0,0].plot(sol["Time [h]"].data, sol["Terminal voltage [V]"].data, label=label) ax[0,1].plot(sol["Time [h]"].data, sol["X-averaged negative particle surface concentration [mol.m-3]"].data, label=label) ax[1,0].plot(sol["Time [h]"].data, sol["Total capacity lost to side reactions [A.h]"].data, label=label) ax[1,1].plot(sol["Time [h]"].data, sol["X-averaged negative total SEI thickness [m]"].data, label=label) ax[0, 0].legend() sim1.solution.cycles[1].steps sim2.solution.cycles[1].steps ``` ![image](https://github.com/pybamm-team/PyBaMM/assets/43040151/d3d0d3cf-9a58-4ee0-a69a-8df6e657ea00) ### Relevant log output _No response_

2024-01-10T16:13:59

0.0

[]

pybamm-team/PyBaMM

pybamm-team__PyBaMM-3627

2e354155e447fc3b3947c75375012942b271dd1d

diff --git a/CHANGELOG.md b/CHANGELOG.md index fdaf360d6f..b1e093f452 100644 --- a/CHANGELOG.md +++ b/CHANGELOG.md @@ -3,9 +3,10 @@ ## Features - Added support for macOS arm64 (M-series) platforms. ([#3789](https://github.com/pybamm-team/PyBaMM/pull/3789)) +- Added the ability to specify a custom solver tolerance in `get_initial_stoichiometries` and related functions ([#3714](https://github.com/pybamm-team/PyBaMM/pull/3714)) +- Modified `step` function to take an array of time `t_eval` as an argument and deprecated use of `npts`. ([#3627](https://github.com/pybamm-team/PyBaMM/pull/3627)) - Renamed "electrode diffusivity" to "particle diffusivity" as a non-breaking change with a deprecation warning ([#3624](https://github.com/pybamm-team/PyBaMM/pull/3624)) - Add support for BPX version 0.4.0 which allows for blended electrodes and user-defined parameters in BPX([#3414](https://github.com/pybamm-team/PyBaMM/pull/3414)) -- Added the ability to specify a custom solver tolerance in `get_initial_stoichiometries` and related functions ([#3714](https://github.com/pybamm-team/PyBaMM/pull/3714)) ## Bug Fixes diff --git a/examples/scripts/SPMe_step.py b/examples/scripts/SPMe_step.py index 90d6f3d017..f277c0e790 100644 --- a/examples/scripts/SPMe_step.py +++ b/examples/scripts/SPMe_step.py @@ -32,12 +32,14 @@ # step model dt = 500 +# t_eval is an array of time in the interval 0 to dt, dt being size of the step. +t_eval = np.array([0, 50, 100, 200, 500]) time = 0 end_time = solution.t[-1] step_solver = pybamm.CasadiSolver() step_solution = None while time < end_time: - step_solution = step_solver.step(step_solution, model, dt=dt, npts=10) + step_solution = step_solver.step(step_solution, model, dt=dt, t_eval=t_eval) time += dt # plot diff --git a/pybamm/simulation.py b/pybamm/simulation.py index a2b260ab43..3dfdac94b5 100644 --- a/pybamm/simulation.py +++ b/pybamm/simulation.py @@ -789,7 +789,7 @@ def solve( current_solution, model, dt, - npts=npts, + t_eval=np.linspace(0, dt, npts), save=False, **kwargs, ) @@ -970,7 +970,7 @@ def run_padding_rest(self, kwargs, rest_time, step_solution): step_solution, model, rest_time, - npts=npts, + t_eval=np.linspace(0, rest_time, npts), save=False, **kwargs, ) @@ -978,7 +978,13 @@ def run_padding_rest(self, kwargs, rest_time, step_solution): return step_solution_with_rest def step( - self, dt, solver=None, npts=2, save=True, starting_solution=None, **kwargs + self, + dt, + solver=None, + t_eval=None, + save=True, + starting_solution=None, + **kwargs, ): """ A method to step the model forward one timestep. This method will @@ -990,9 +996,10 @@ def step( The timestep over which to step the solution solver : :class:`pybamm.BaseSolver` The solver to use to solve the model. - npts : int, optional - The number of points at which the solution will be returned during - the step dt. Default is 2 (returns the solution at t0 and t0 + dt). + t_eval : list or numpy.ndarray, optional + An array of times at which to return the solution during the step + (Note: t_eval is the time measured from the start of the step, so should start at 0 and end at dt). + By default, the solution is returned at t0 and t0 + dt. save : bool Turn on to store the solution of all previous timesteps starting_solution : :class:`pybamm.Solution` @@ -1012,7 +1019,12 @@ def step( starting_solution = self._solution self._solution = solver.step( - starting_solution, self.built_model, dt, npts=npts, save=save, **kwargs + starting_solution, + self.built_model, + dt, + t_eval=t_eval, + save=save, + **kwargs, ) return self.solution diff --git a/pybamm/solvers/base_solver.py b/pybamm/solvers/base_solver.py index a2b4c305c2..be26306bbd 100644 --- a/pybamm/solvers/base_solver.py +++ b/pybamm/solvers/base_solver.py @@ -1090,7 +1090,8 @@ def step( old_solution, model, dt, - npts=2, + t_eval=None, + npts=None, inputs=None, save=True, ): @@ -1108,9 +1109,11 @@ def step( initial_conditions dt : numeric type The timestep (in seconds) over which to step the solution - npts : int, optional - The number of points at which the solution will be returned during - the step dt. default is 2 (returns the solution at t0 and t0 + dt). + t_eval : list or numpy.ndarray, optional + An array of times at which to return the solution during the step + (Note: t_eval is the time measured from the start of the step, so should start at 0 and end at dt). + By default, the solution is returned at t0 and t0 + dt. + npts : deprecated inputs : dict, optional Any input parameters to pass to the model when solving save : bool @@ -1149,10 +1152,28 @@ def step( f"Step time must be at least {pybamm.TimerTime(step_start_offset)}" ) + # Raise deprecation warning for npts and convert it to t_eval + if npts is not None: + warnings.warn( + "The 'npts' parameter is deprecated, use 't_eval' instead.", + DeprecationWarning, + stacklevel=2, + ) + t_eval = np.linspace(0, dt, npts) + + if t_eval is not None: + # Checking if t_eval lies within range + if t_eval[0] != 0 or t_eval[-1] != dt: + raise pybamm.SolverError( + "Elements inside array t_eval must lie in the closed interval 0 to dt" + ) + + else: + t_eval = np.array([0, dt]) + t_start = old_solution.t[-1] + t_eval = t_start + t_eval t_end = t_start + dt - # Calculate t_eval - t_eval = np.linspace(t_start, t_end, npts) if t_start == 0: t_start_shifted = t_start

Pass array of times to `step` When you call `solver.step` you pass in `dt`, the time time increment you want to step forward in. You can also pass `npts` to get the solution back at a linearly spaced array of times between the initial time, `t0`, and `t0+dt`. It would be good to instead be able to pass an array of times, `t_eval`, like you pass to solve, so you can get step solutions returned at non-linearly spaced times. The motivation for this is that right now you can't pass a `starting_solution` to `Simualtion.solve` if you aren't using an experiment, so if you want to chain solutions together you have to call `Simulation.step` instead. An alternative is that we allow starting solutions without experiments in `Simulation.solve`. The whole logic for `step` and `solve` is pretty clunky both in `Simulation` and `BaseSolver`, but that's a separate issue...

Hey I would like to work on this. Can you please give examples of how to use step() function to get solutions which I can use for debugging? I'm quite new here and I cant find it in project doc. Sorry for the slow reply. Take a look [here](https://github.com/pybamm-team/PyBaMM/blob/develop/examples/scripts/SPMe_step.py). Thanks 😄 Maybe if we change step function to something like: ```py def step( self, old_solution, model, dt=0, non_linear_time=None, npts=2, inputs=None, save=True, ) ``` Here `non_linear_time` is an array of time (example: [0,10,15,17,20]). We can assign `dt = non_linear_time[i] - non_linear_time[i-1]` inside a loop and make changes to the function accordingly. Is this a correct solution? > Here `non_linear_time` is an array of time (example: [0,10,15,17,20]). Hi @prady0t, theoretically it looks promising, though it would be nicer to open a PR to keep things moving & also would be easier for us to suggest changes there 🙂 Sure will do!

2023-12-17T07:27:09

0.0

[]

pybamm-team/PyBaMM

pybamm-team__PyBaMM-3557

a4632463c7a9bed4f644bc7c1549726c9d976769

diff --git a/.coveragerc b/.coveragerc deleted file mode 100644 index a174f5aced..0000000000 --- a/.coveragerc +++ /dev/null @@ -1,3 +0,0 @@ -[run] -source = pybamm -concurrency = multiprocessing diff --git a/.git-blame-ignore-revs b/.git-blame-ignore-revs index 228a76373e..9e59bd7f07 100644 --- a/.git-blame-ignore-revs +++ b/.git-blame-ignore-revs @@ -4,3 +4,5 @@ a63e49ece0f9336d1f5c2562f7459e555c6e6693 # activated standard pre-commits - https://github.com/pybamm-team/PyBaMM/pull/3192 5273214b585c5a4286609aed40e0b092d0e05f42 +# migrate config to pyproject.toml - https://github.com/pybamm-team/PyBaMM/pull/3557 +12c5d77203bd93542785d237bac00bad5ed5469a diff --git a/benchmarks/work_precision_sets/time_vs_abstols.py b/benchmarks/work_precision_sets/time_vs_abstols.py index 6447884083..9a96f07514 100644 --- a/benchmarks/work_precision_sets/time_vs_abstols.py +++ b/benchmarks/work_precision_sets/time_vs_abstols.py @@ -96,7 +96,7 @@ plt.savefig(f"benchmarks/benchmark_images/time_vs_abstols_{pybamm.__version__}.png") -content = f"# PyBaMM {pybamm.__version__}\n## Solve Time vs Abstols\n<img src='./benchmark_images/time_vs_abstols_{pybamm.__version__}.png'>\n" # noqa +content = f"# PyBaMM {pybamm.__version__}\n## Solve Time vs Abstols\n<img src='./benchmark_images/time_vs_abstols_{pybamm.__version__}.png'>\n" with open("./benchmarks/release_work_precision_sets.md", "r") as original: data = original.read() diff --git a/benchmarks/work_precision_sets/time_vs_dt_max.py b/benchmarks/work_precision_sets/time_vs_dt_max.py index 3926a4bcd6..3e428b702c 100644 --- a/benchmarks/work_precision_sets/time_vs_dt_max.py +++ b/benchmarks/work_precision_sets/time_vs_dt_max.py @@ -98,7 +98,7 @@ plt.savefig(f"benchmarks/benchmark_images/time_vs_dt_max_{pybamm.__version__}.png") -content = f"## Solve Time vs dt_max\n<img src='./benchmark_images/time_vs_dt_max_{pybamm.__version__}.png'>\n" # noqa +content = f"## Solve Time vs dt_max\n<img src='./benchmark_images/time_vs_dt_max_{pybamm.__version__}.png'>\n" with open("./benchmarks/release_work_precision_sets.md", "r") as original: data = original.read() diff --git a/benchmarks/work_precision_sets/time_vs_mesh_size.py b/benchmarks/work_precision_sets/time_vs_mesh_size.py index 7b4d4145d4..f0f13f706b 100644 --- a/benchmarks/work_precision_sets/time_vs_mesh_size.py +++ b/benchmarks/work_precision_sets/time_vs_mesh_size.py @@ -78,7 +78,7 @@ plt.savefig(f"benchmarks/benchmark_images/time_vs_mesh_size_{pybamm.__version__}.png") -content = f"## Solve Time vs Mesh size\n<img src='./benchmark_images/time_vs_mesh_size_{pybamm.__version__}.png'>\n" # noqa +content = f"## Solve Time vs Mesh size\n<img src='./benchmark_images/time_vs_mesh_size_{pybamm.__version__}.png'>\n" with open("./benchmarks/release_work_precision_sets.md", "r") as original: data = original.read() diff --git a/benchmarks/work_precision_sets/time_vs_no_of_states.py b/benchmarks/work_precision_sets/time_vs_no_of_states.py index 0a88ac8b52..eb27aba322 100644 --- a/benchmarks/work_precision_sets/time_vs_no_of_states.py +++ b/benchmarks/work_precision_sets/time_vs_no_of_states.py @@ -82,7 +82,7 @@ ) -content = f"## Solve Time vs Number of states\n<img src='./benchmark_images/time_vs_no_of_states_{pybamm.__version__}.png'>\n" # noqa +content = f"## Solve Time vs Number of states\n<img src='./benchmark_images/time_vs_no_of_states_{pybamm.__version__}.png'>\n" with open("./benchmarks/release_work_precision_sets.md", "r") as original: data = original.read() diff --git a/benchmarks/work_precision_sets/time_vs_reltols.py b/benchmarks/work_precision_sets/time_vs_reltols.py index 12e41b526f..93964910a8 100644 --- a/benchmarks/work_precision_sets/time_vs_reltols.py +++ b/benchmarks/work_precision_sets/time_vs_reltols.py @@ -102,7 +102,7 @@ plt.savefig(f"benchmarks/benchmark_images/time_vs_reltols_{pybamm.__version__}.png") -content = f"## Solve Time vs Reltols\n<img src='./benchmark_images/time_vs_reltols_{pybamm.__version__}.png'>\n" # noqa +content = f"## Solve Time vs Reltols\n<img src='./benchmark_images/time_vs_reltols_{pybamm.__version__}.png'>\n" with open("./benchmarks/release_work_precision_sets.md", "r") as original: data = original.read() diff --git a/docs/source/examples/notebooks/models/jelly-roll-model.ipynb b/docs/source/examples/notebooks/models/jelly-roll-model.ipynb index fe6173f1ce..557366099a 100644 --- a/docs/source/examples/notebooks/models/jelly-roll-model.ipynb +++ b/docs/source/examples/notebooks/models/jelly-roll-model.ipynb @@ -84,7 +84,7 @@ "delta = pybamm.Parameter(\"Current collector thickness\")\n", "delta_p = delta # assume same thickness\n", "delta_n = delta # assume same thickness\n", - "l = 1/2 - delta_p - delta_n # active material thickness # noqa: E741\n", + "l = 1/2 - delta_p - delta_n # active material thickness\n", "sigma_p = pybamm.Parameter(\"Positive current collector conductivity\")\n", "sigma_n = pybamm.Parameter(\"Negative current collector conductivity\")\n", "sigma_a = pybamm.Parameter(\"Active material conductivity\")" diff --git a/docs/source/examples/notebooks/models/latexify.ipynb b/docs/source/examples/notebooks/models/latexify.ipynb index 63e7c0d519..c2a45ff2c8 100644 --- a/docs/source/examples/notebooks/models/latexify.ipynb +++ b/docs/source/examples/notebooks/models/latexify.ipynb @@ -1252,7 +1252,7 @@ "source": [ "spme_latex = model_spme.latexify(newline=False)\n", "for line in spme_latex:\n", - " display(line) # noqa: F821" + " display(line)" ] }, { diff --git a/docs/source/user_guide/installation/index.rst b/docs/source/user_guide/installation/index.rst index 983f66842e..e771611a37 100644 --- a/docs/source/user_guide/installation/index.rst +++ b/docs/source/user_guide/installation/index.rst @@ -154,6 +154,7 @@ Dependency `pre-commit <https://pre-commit.com/index.html>`__ \- dev For managing and maintaining multi-language pre-commit hooks. `ruff <https://beta.ruff.rs/docs/>`__ \- dev For code formatting. `nox <https://nox.thea.codes/en/stable/>`__ \- dev For running testing sessions in multiple environments. +`coverage <https://coverage.readthedocs.io/en/>`__ \- dev For calculating coverage of tests. `pytest <https://docs.pytest.org/en/stable/>`__ 6.0.0 dev For running Jupyter notebooks tests. `pytest-xdist <https://pytest-xdist.readthedocs.io/en/latest/>`__ \- dev For running tests in parallel across distributed workers. `nbmake <https://github.com/treebeardtech/nbmake/>`__ \- dev A ``pytest`` plugin for executing Jupyter notebooks. diff --git a/examples/scripts/print_model_parameter_combinations.py b/examples/scripts/print_model_parameter_combinations.py index 8d24f919c3..f7f00714fb 100644 --- a/examples/scripts/print_model_parameter_combinations.py +++ b/examples/scripts/print_model_parameter_combinations.py @@ -19,13 +19,13 @@ try: model = pybamm.lithium_ion.SPM(options.copy()) except pybamm.OptionError as e: - print(f"Cannot create model with {options}. (OptionError: {str(e)})") + print(f"Cannot create model with {options}. (OptionError: {e!s})") except pybamm.ModelError as e: # todo: properly resolve the cases that raise these errors - print(f"Cannot create model with {options}. (ModelError: {str(e)})") + print(f"Cannot create model with {options}. (ModelError: {e!s})") except AttributeError as e: # todo: properly resolve the cases that raise these errors - print(f"Cannot create model with {options}. (AttributeError: {str(e)})") + print(f"Cannot create model with {options}. (AttributeError: {e!s})") else: output = f"{options} with {parameter_set} parameters: " try: diff --git a/noxfile.py b/noxfile.py index 7a57ad5820..5db08ee92f 100644 --- a/noxfile.py +++ b/noxfile.py @@ -38,7 +38,7 @@ def set_environment_variables(env_dict, session): @nox.session(name="pybamm-requires") def run_pybamm_requires(session): - """Download, compile, and install the build-time requirements for Linux and macOS: the SuiteSparse and SUNDIALS libraries.""" # noqa: E501 + """Download, compile, and install the build-time requirements for Linux and macOS: the SuiteSparse and SUNDIALS libraries.""" set_environment_variables(PYBAMM_ENV, session=session) if sys.platform != "win32": session.install("wget", "cmake", silent=False) @@ -64,7 +64,7 @@ def run_coverage(session): session.install("-e", ".[all,odes,jax]", silent=False) else: session.install("-e", ".[all]", silent=False) - session.run("coverage", "run", "--rcfile=.coveragerc", "run-tests.py", "--nosub") + session.run("coverage", "run", "run-tests.py", "--nosub") session.run("coverage", "combine") session.run("coverage", "xml") diff --git a/pybamm/citations.py b/pybamm/citations.py index b72262989b..e73351a4c6 100644 --- a/pybamm/citations.py +++ b/pybamm/citations.py @@ -267,7 +267,7 @@ def print_citations(filename=None, output_format="text", verbose=False): if verbose: # pragma: no cover if filename is not None: # pragma: no cover raise Exception( - "Verbose output is available only for the terminal and not for printing to files", # noqa: E501 + "Verbose output is available only for the terminal and not for printing to files", ) else: citations.print(filename, output_format, verbose=True) diff --git a/pybamm/discretisations/discretisation.py b/pybamm/discretisations/discretisation.py index a120adecc0..bb6e678f4c 100644 --- a/pybamm/discretisations/discretisation.py +++ b/pybamm/discretisations/discretisation.py @@ -461,7 +461,7 @@ def process_boundary_conditions(self, model): if ( self.mesh[subdomain].coord_sys in ["spherical polar", "cylindrical polar"] - and list(self.mesh.geometry[subdomain].values())[0]["min"] == 0 + and next(iter(self.mesh.geometry[subdomain].values()))["min"] == 0 ): if bcs["left"][0].value != 0 or bcs["left"][1] != "Neumann": raise pybamm.ModelError( @@ -753,7 +753,7 @@ def _process_symbol(self, symbol): spatial_method = self.spatial_methods[symbol.domain[0]] # If boundary conditions are provided, need to check for BCs on tabs if self.bcs: - key_id = list(self.bcs.keys())[0] + key_id = next(iter(self.bcs.keys())) if any("tab" in side for side in list(self.bcs[key_id].keys())): self.bcs[key_id] = self.check_tab_conditions( symbol, self.bcs[key_id] diff --git a/pybamm/expression_tree/array.py b/pybamm/expression_tree/array.py index 2736886d95..adbdc88dc2 100644 --- a/pybamm/expression_tree/array.py +++ b/pybamm/expression_tree/array.py @@ -97,7 +97,7 @@ def entries_string(self, value): def set_id(self): """See :meth:`pybamm.Symbol.set_id()`.""" self._id = hash( - (self.__class__, self.name) + self.entries_string + tuple(self.domain) + (self.__class__, self.name, *self.entries_string, *tuple(self.domain)) ) def _jac(self, variable): diff --git a/pybamm/expression_tree/interpolant.py b/pybamm/expression_tree/interpolant.py index cd0df4d077..28188ce68f 100644 --- a/pybamm/expression_tree/interpolant.py +++ b/pybamm/expression_tree/interpolant.py @@ -222,9 +222,7 @@ def entries_string(self, value): def set_id(self): """See :meth:`pybamm.Symbol.set_id()`.""" self._id = hash( - (self.__class__, self.name, self.entries_string) - + tuple([child.id for child in self.children]) - + tuple(self.domain) + (self.__class__, self.name, self.entries_string, *tuple([child.id for child in self.children]), *tuple(self.domain)) ) def _function_new_copy(self, children): diff --git a/pybamm/expression_tree/operations/evaluate_python.py b/pybamm/expression_tree/operations/evaluate_python.py index 1f44a69784..d0cd4c776d 100644 --- a/pybamm/expression_tree/operations/evaluate_python.py +++ b/pybamm/expression_tree/operations/evaluate_python.py @@ -42,7 +42,7 @@ class JaxCooMatrix: def __init__(self, row, col, data, shape): if not pybamm.have_jax(): # pragma: no cover raise ModuleNotFoundError( - "Jax or jaxlib is not installed, please see https://docs.pybamm.org/en/latest/source/user_guide/installation/GNU-linux.html#optional-jaxsolver" # noqa: E501 + "Jax or jaxlib is not installed, please see https://docs.pybamm.org/en/latest/source/user_guide/installation/GNU-linux.html#optional-jaxsolver" ) self.row = jax.numpy.array(row) @@ -537,7 +537,7 @@ class EvaluatorJax: def __init__(self, symbol): if not pybamm.have_jax(): # pragma: no cover raise ModuleNotFoundError( - "Jax or jaxlib is not installed, please see https://docs.pybamm.org/en/latest/source/user_guide/installation/GNU-linux.html#optional-jaxsolver" # noqa: E501 + "Jax or jaxlib is not installed, please see https://docs.pybamm.org/en/latest/source/user_guide/installation/GNU-linux.html#optional-jaxsolver" ) constants, python_str = pybamm.to_python(symbol, debug=False, output_jax=True) diff --git a/pybamm/expression_tree/operations/latexify.py b/pybamm/expression_tree/operations/latexify.py index 9f2949069e..572f01a560 100644 --- a/pybamm/expression_tree/operations/latexify.py +++ b/pybamm/expression_tree/operations/latexify.py @@ -93,9 +93,9 @@ def _get_bcs_displays(self, var): if bcs: # Take range minimum from the first domain - var_name = list(self.model.default_geometry[var.domain[0]].keys())[0] - rng_left = list(self.model.default_geometry[var.domain[0]].values())[0] - rng_right = list(self.model.default_geometry[var.domain[-1]].values())[0] + var_name = next(iter(self.model.default_geometry[var.domain[0]].keys())) + rng_left = next(iter(self.model.default_geometry[var.domain[0]].values())) + rng_right = next(iter(self.model.default_geometry[var.domain[-1]].values())) # Trim name (r_n --> r) var_name = re.findall(r"(.)_*.*", str(var_name))[0] diff --git a/pybamm/expression_tree/parameter.py b/pybamm/expression_tree/parameter.py index eebe77ad2f..00a28017b8 100644 --- a/pybamm/expression_tree/parameter.py +++ b/pybamm/expression_tree/parameter.py @@ -152,9 +152,7 @@ def input_names(self, inp=None): def set_id(self): """See :meth:`pybamm.Symbol.set_id`""" self._id = hash( - (self.__class__, self.name, self.diff_variable) - + tuple([child.id for child in self.children]) - + tuple(self.domain) + (self.__class__, self.name, self.diff_variable, *tuple([child.id for child in self.children]), *tuple(self.domain)) ) def diff(self, variable): diff --git a/pybamm/expression_tree/state_vector.py b/pybamm/expression_tree/state_vector.py index 6ef8bee904..437ba752ed 100644 --- a/pybamm/expression_tree/state_vector.py +++ b/pybamm/expression_tree/state_vector.py @@ -107,8 +107,7 @@ def set_evaluation_array(self, y_slices, evaluation_array): def set_id(self): """See :meth:`pybamm.Symbol.set_id()`""" self._id = hash( - (self.__class__, self.name, tuple(self.evaluation_array)) - + tuple(self.domain) + (self.__class__, self.name, tuple(self.evaluation_array), *tuple(self.domain)) ) def _jac_diff_vector(self, variable): diff --git a/pybamm/expression_tree/symbol.py b/pybamm/expression_tree/symbol.py index 8f1608e7ba..5fe765af33 100644 --- a/pybamm/expression_tree/symbol.py +++ b/pybamm/expression_tree/symbol.py @@ -403,9 +403,7 @@ def set_id(self): need to hash once. """ self._id = hash( - (self.__class__, self.name) - + tuple([child.id for child in self.children]) - + tuple([(k, tuple(v)) for k, v in self.domains.items() if v != []]) + (self.__class__, self.name, *tuple([child.id for child in self.children]), *tuple([(k, tuple(v)) for k, v in self.domains.items() if v != []])) ) @property diff --git a/pybamm/expression_tree/unary_operators.py b/pybamm/expression_tree/unary_operators.py index 81c3dc28c2..95306ebad5 100644 --- a/pybamm/expression_tree/unary_operators.py +++ b/pybamm/expression_tree/unary_operators.py @@ -287,14 +287,7 @@ def _unary_jac(self, child_jac): def set_id(self): """See :meth:`pybamm.Symbol.set_id()`""" self._id = hash( - ( - self.__class__, - self.name, - self.slice.start, - self.slice.stop, - self.children[0].id, - ) - + tuple(self.domain) + (self.__class__, self.name, self.slice.start, self.slice.stop, self.children[0].id, *tuple(self.domain)) ) def _unary_evaluate(self, child): @@ -554,15 +547,7 @@ def integration_variable(self): def set_id(self): """See :meth:`pybamm.Symbol.set_id()`""" self._id = hash( - (self.__class__, self.name) - + tuple( - [ - integration_variable.id - for integration_variable in self.integration_variable - ] - ) - + (self.children[0].id,) - + tuple(self.domain) + (self.__class__, self.name, *tuple([integration_variable.id for integration_variable in self.integration_variable]), self.children[0].id, *tuple(self.domain)) ) def _unary_new_copy(self, child): @@ -702,9 +687,7 @@ def __init__(self, child, vector_type="row"): def set_id(self): """See :meth:`pybamm.Symbol.set_id()`""" self._id = hash( - (self.__class__, self.name, self.vector_type) - + (self.children[0].id,) - + tuple(self.domain) + (self.__class__, self.name, self.vector_type, self.children[0].id, *tuple(self.domain)) ) def _unary_new_copy(self, child): @@ -757,7 +740,7 @@ def __init__(self, child, region="entire"): def set_id(self): """See :meth:`pybamm.Symbol.set_id()`""" self._id = hash( - (self.__class__, self.name) + (self.children[0].id,) + tuple(self.domain) + (self.__class__, self.name, self.children[0].id, *tuple(self.domain)) ) def _unary_new_copy(self, child): @@ -798,8 +781,7 @@ def __init__(self, child, side, domain): def set_id(self): """See :meth:`pybamm.Symbol.set_id()`""" self._id = hash( - (self.__class__, self.name, self.side, self.children[0].id) - + tuple([(k, tuple(v)) for k, v in self.domains.items()]) + (self.__class__, self.name, self.side, self.children[0].id, *tuple([(k, tuple(v)) for k, v in self.domains.items()])) ) def _evaluates_on_edges(self, dimension): @@ -857,8 +839,7 @@ def __init__(self, name, child, side): def set_id(self): """See :meth:`pybamm.Symbol.set_id()`""" self._id = hash( - (self.__class__, self.name, self.side, self.children[0].id) - + tuple([(k, tuple(v)) for k, v in self.domains.items()]) + (self.__class__, self.name, self.side, self.children[0].id, *tuple([(k, tuple(v)) for k, v in self.domains.items()])) ) def _unary_new_copy(self, child): diff --git a/pybamm/expression_tree/variable.py b/pybamm/expression_tree/variable.py index 0d1e1fd424..22b176b6fc 100644 --- a/pybamm/expression_tree/variable.py +++ b/pybamm/expression_tree/variable.py @@ -103,8 +103,7 @@ def bounds(self, values): def set_id(self): self._id = hash( - (self.__class__, self.name, self.scale, self.reference) - + tuple([(k, tuple(v)) for k, v in self.domains.items() if v != []]) + (self.__class__, self.name, self.scale, self.reference, *tuple([(k, tuple(v)) for k, v in self.domains.items() if v != []])) ) def create_copy(self): diff --git a/pybamm/install_odes.py b/pybamm/install_odes.py index 4bf310a0f2..0fbbcdc637 100644 --- a/pybamm/install_odes.py +++ b/pybamm/install_odes.py @@ -66,7 +66,7 @@ def install_sundials(download_dir, install_dir): print("-" * 10, "Running CMake prepare", "-" * 40) subprocess.run( - ["cmake", "../sundials-{}".format(sundials_version)] + cmake_args, + ["cmake", "../sundials-{}".format(sundials_version), *cmake_args], cwd=build_directory, check=True, ) diff --git a/pybamm/meshes/zero_dimensional_submesh.py b/pybamm/meshes/zero_dimensional_submesh.py index 5b2f38e29f..82e8cb6524 100644 --- a/pybamm/meshes/zero_dimensional_submesh.py +++ b/pybamm/meshes/zero_dimensional_submesh.py @@ -31,7 +31,7 @@ def __init__(self, position, npts=None): raise pybamm.GeometryError("position should only contain a single variable") # extract the position - position = list(position.values())[0] + position = next(iter(position.values())) spatial_position = position["position"] self.nodes = np.array([spatial_position]) self.edges = np.array([spatial_position]) diff --git a/pybamm/models/base_model.py b/pybamm/models/base_model.py index 08890757b7..a88f6f4255 100644 --- a/pybamm/models/base_model.py +++ b/pybamm/models/base_model.py @@ -1205,7 +1205,7 @@ def check_and_convert_equations(self, equations): for var, eqn in equations.items(): if eqn.has_symbol_of_classes(pybamm.Variable): unpacker = pybamm.SymbolUnpacker(pybamm.Variable) - variable_in_equation = list(unpacker.unpack_symbol(eqn))[0] + variable_in_equation = next(iter(unpacker.unpack_symbol(eqn))) raise TypeError( "Initial conditions cannot contain 'Variable' objects, " "but '{!r}' found in initial conditions for '{}'".format( diff --git a/pybamm/models/submodels/electrolyte_conductivity/surface_potential_form/full_surface_form_conductivity.py b/pybamm/models/submodels/electrolyte_conductivity/surface_potential_form/full_surface_form_conductivity.py index 7afeeac47e..83bcfb8027 100644 --- a/pybamm/models/submodels/electrolyte_conductivity/surface_potential_form/full_surface_form_conductivity.py +++ b/pybamm/models/submodels/electrolyte_conductivity/surface_potential_form/full_surface_form_conductivity.py @@ -224,7 +224,7 @@ class FullAlgebraic(BaseModel): The parameters to use for this submodel options : dict, optional A dictionary of options to be passed to the model. - """ # noqa: E501 + """ def __init__(self, param, domain, options=None): super().__init__(param, domain, options) @@ -258,7 +258,7 @@ class FullDifferential(BaseModel): The parameters to use for this submodel options : dict, optional A dictionary of options to be passed to the model. - """ # noqa: E501 + """ def __init__(self, param, domain, options=None): super().__init__(param, domain, options) diff --git a/pybamm/parameters_cli.py b/pybamm/parameters_cli.py index 90950b23ee..e3d4a273b8 100644 --- a/pybamm/parameters_cli.py +++ b/pybamm/parameters_cli.py @@ -2,7 +2,7 @@ def raise_error(): raise NotImplementedError( "parameters cli has been deprecated. " "Parameters should now be defined via python files (see " - "https://github.com/pybamm-team/PyBaMM/tree/develop/pybamm/input/parameters/lithium_ion/Ai2020.py" # noqa: E501 + "https://github.com/pybamm-team/PyBaMM/tree/develop/pybamm/input/parameters/lithium_ion/Ai2020.py" " for example)" ) diff --git a/pybamm/plotting/quick_plot.py b/pybamm/plotting/quick_plot.py index ff657ee375..0e56c17c75 100644 --- a/pybamm/plotting/quick_plot.py +++ b/pybamm/plotting/quick_plot.py @@ -534,7 +534,7 @@ def plot(self, t, dynamic=False): # 1D plot: plot as a function of x at time t # Read dictionary of spatial variables spatial_vars = self.spatial_variable_dict[key] - spatial_var_name = list(spatial_vars.keys())[0] + spatial_var_name = next(iter(spatial_vars.keys())) ax.set_xlabel( "{} [{}]".format(spatial_var_name, self.spatial_unit), ) @@ -568,12 +568,12 @@ def plot(self, t, dynamic=False): # different order based on whether the domains are x-r, x-z or y-z, etc if self.x_first_and_y_second[key] is False: x_name = list(spatial_vars.keys())[1][0] - y_name = list(spatial_vars.keys())[0][0] + y_name = next(iter(spatial_vars.keys()))[0] x = self.second_spatial_variable[key] y = self.first_spatial_variable[key] var = variable(t_in_seconds, **spatial_vars, warn=False) else: - x_name = list(spatial_vars.keys())[0][0] + x_name = next(iter(spatial_vars.keys()))[0] y_name = list(spatial_vars.keys())[1][0] x = self.first_spatial_variable[key] y = self.second_spatial_variable[key] diff --git a/pybamm/simulation.py b/pybamm/simulation.py index f9aebb1c54..bdd97f6894 100644 --- a/pybamm/simulation.py +++ b/pybamm/simulation.py @@ -375,7 +375,7 @@ def set_initial_soc(self, initial_soc): param = self._model.param if options["open-circuit potential"] == "MSMR": self._parameter_values = ( - self._unprocessed_parameter_values.set_initial_ocps( # noqa: E501 + self._unprocessed_parameter_values.set_initial_ocps( initial_soc, param=param, inplace=False, options=options ) ) diff --git a/pybamm/solvers/jax_bdf_solver.py b/pybamm/solvers/jax_bdf_solver.py index 2f334ed8ec..8f5b8ed817 100644 --- a/pybamm/solvers/jax_bdf_solver.py +++ b/pybamm/solvers/jax_bdf_solver.py @@ -676,7 +676,7 @@ def while_body(while_state): # ) (state, step_accepted) = tree_map( - partial(jnp.where, converged * (error_norm > 1)), # noqa: E712 + partial(jnp.where, converged * (error_norm > 1)), (_update_step_size_and_lu(state, factor), False), (state, converged), ) @@ -883,9 +883,7 @@ def arg_dicts_to_values(args): """ return sum((tuple(b.values()) for b in args if isinstance(b, dict)), ()) - aug_mass = (mass, mass, onp.array(1.0)) + arg_dicts_to_values( - tree_map(arg_to_identity, args) - ) + aug_mass = (mass, mass, onp.array(1.0), *arg_dicts_to_values(tree_map(arg_to_identity, args))) def scan_fun(carry, i): y_bar, t0_bar, args_bar = carry @@ -961,7 +959,7 @@ def ravel_first_arg(f, unravel): @lu.transformation def ravel_first_arg_(unravel, y_flat, *args): y = unravel(y_flat) - ans = yield (y,) + args, {} + ans = yield (y, *args), {} ans_flat, _ = ravel_pytree(ans) yield ans_flat @@ -1007,7 +1005,7 @@ def jax_bdf_integrate(func, y0, t_eval, *args, rtol=1e-6, atol=1e-6, mass=None): """ if not pybamm.have_jax(): raise ModuleNotFoundError( - "Jax or jaxlib is not installed, please see https://docs.pybamm.org/en/latest/source/user_guide/installation/GNU-linux.html#optional-jaxsolver" # noqa: E501 + "Jax or jaxlib is not installed, please see https://docs.pybamm.org/en/latest/source/user_guide/installation/GNU-linux.html#optional-jaxsolver" ) def _check_arg(arg): diff --git a/pybamm/solvers/jax_solver.py b/pybamm/solvers/jax_solver.py index 4c9759008a..313fddc208 100644 --- a/pybamm/solvers/jax_solver.py +++ b/pybamm/solvers/jax_solver.py @@ -61,7 +61,7 @@ def __init__( ): if not pybamm.have_jax(): raise ModuleNotFoundError( - "Jax or jaxlib is not installed, please see https://docs.pybamm.org/en/latest/source/user_guide/installation/GNU-linux.html#optional-jaxsolver" # noqa: E501 + "Jax or jaxlib is not installed, please see https://docs.pybamm.org/en/latest/source/user_guide/installation/GNU-linux.html#optional-jaxsolver" ) # note: bdf solver itself calculates consistent initial conditions so can set diff --git a/pybamm/spatial_methods/finite_volume.py b/pybamm/spatial_methods/finite_volume.py index ecbae69796..5c32e5a2c0 100644 --- a/pybamm/spatial_methods/finite_volume.py +++ b/pybamm/spatial_methods/finite_volume.py @@ -620,10 +620,10 @@ def add_ghost_nodes(self, symbol, discretised_symbol, bcs): # Dirichlet boundary conditions n_bcs = 0 if lbc_type == "Dirichlet": - domain = [domain[0] + "_left ghost cell"] + domain + domain = [domain[0] + "_left ghost cell", *domain] n_bcs += 1 if rbc_type == "Dirichlet": - domain = domain + [domain[-1] + "_right ghost cell"] + domain = [*domain, domain[-1] + "_right ghost cell"] n_bcs += 1 # Calculate values for ghost nodes for any Dirichlet boundary conditions diff --git a/pyproject.toml b/pyproject.toml index 4569c7c6c3..eae0575117 100644 --- a/pyproject.toml +++ b/pyproject.toml @@ -102,6 +102,8 @@ dev = [ "ruff", # For running testing sessions "nox", + # For coverage + "coverage[toml]", # For testing Jupyter notebooks "pytest>=6", "pytest-xdist", @@ -172,3 +174,76 @@ pybamm = [ [tool.setuptools.packages.find] include = ["pybamm", "pybamm.*"] + +[tool.ruff] +extend-include = ["*.ipynb"] +extend-exclude = ["__init__.py"] + +[tool.ruff.lint] +extend-select = [ + # "B", # flake8-bugbear + # "I", # isort + # "ARG", # flake8-unused-arguments + # "C4", # flake8-comprehensions + # "ICN", # flake8-import-conventions + # "ISC", # flake8-implicit-str-concat + # "PGH", # pygrep-hooks + # "PIE", # flake8-pie + # "PL", # pylint + # "PT", # flake8-pytest-style + # "PTH", # flake8-use-pathlib + # "RET", # flake8-return + "RUF", # Ruff-specific + # "SIM", # flake8-simplify + # "T20", # flake8-print + # "UP", # pyupgrade + "YTT", # flake8-2020 +] +ignore = [ + "E741", # Ambiguous variable name + "RUF012", # Mutable class attributes should be annotated with `typing.ClassVar` + "SIM108", # Use ternary operator + "ARG001", # Unused function argument: + "ARG002", # Unused method arguments + "PLR2004", # Magic value used in comparison + "PLR0915", # Too many statements + "PLR0913", # Too many arguments + "PLR0912", # Too many branches + "RET504", # Unnecessary assignment + "RET505", # Unnecessary `else` + "RET506", # Unnecessary `elif` + "B018", # Found useless expression + "RUF002", # Docstring contains ambiguous +] + +[tool.ruff.lint.per-file-ignores] +"tests/*" = ["T20"] +"docs/*" = ["T20"] +"examples/*" = ["T20"] +"**.ipynb" = ["E402", "E703"] + +# NOTE: currently used only for notebook tests with the nbmake plugin. +[tool.pytest.ini_options] +# Use pytest-xdist to run tests in parallel by default, exit with +# error if not installed +required_plugins = [ + "pytest-xdist", +] +addopts = [ + "-nauto", + "-v", +] +testpaths = [ + "docs/source/examples/", +] +console_output_style = "progress" + +# Logging configuration +log_cli = "true" +log_cli_level = "INFO" +log_format = "%(asctime)s - %(name)s - %(levelname)s - %(message)s" +log_date_format = "%Y-%m-%d %H:%M:%S" + +[tool.coverage.run] +source = ["pybamm"] +concurrency = ["multiprocessing"] diff --git a/ruff.toml b/ruff.toml deleted file mode 100644 index 7304d64570..0000000000 --- a/ruff.toml +++ /dev/null @@ -1,8 +0,0 @@ -extend-include = ["*.ipynb"] -extend-exclude = ["__init__.py"] - -[lint] -ignore = ["E741"] - -[lint.per-file-ignores] -"**.ipynb" = ["E402", "E703"] diff --git a/scripts/fix_casadi_rpath_mac.py b/scripts/fix_casadi_rpath_mac.py index 23c8a32d59..3f7f71e834 100644 --- a/scripts/fix_casadi_rpath_mac.py +++ b/scripts/fix_casadi_rpath_mac.py @@ -30,15 +30,15 @@ os.path.join(casadi_dir, libcasadi_37_name), ] -subprocess.run(["otool"] + ["-L", os.path.join(casadi_dir, libcasadi_name)]) +subprocess.run(["otool", "-L", os.path.join(casadi_dir, libcasadi_name)]) -print(" ".join(["install_name_tool"] + install_name_tool_args_for_libcasadi_name)) -subprocess.run(["install_name_tool"] + install_name_tool_args_for_libcasadi_name) +print(" ".join(["install_name_tool", *install_name_tool_args_for_libcasadi_name])) +subprocess.run(["install_name_tool", *install_name_tool_args_for_libcasadi_name]) -print(" ".join(["install_name_tool"] + install_name_tool_args_for_libcasadi_37_name)) -subprocess.run(["install_name_tool"] + install_name_tool_args_for_libcasadi_37_name) +print(" ".join(["install_name_tool", *install_name_tool_args_for_libcasadi_37_name])) +subprocess.run(["install_name_tool", *install_name_tool_args_for_libcasadi_37_name]) -subprocess.run(["otool"] + ["-L", os.path.join(casadi_dir, libcasadi_name)]) +subprocess.run(["otool", "-L", os.path.join(casadi_dir, libcasadi_name)]) install_name_tool_args = [ "-change", @@ -46,12 +46,12 @@ os.path.join(casadi_dir, libcppabi_name), os.path.join(casadi_dir, libcpp_name), ] -subprocess.run(["otool"] + ["-L", os.path.join(casadi_dir, libcpp_name)]) +subprocess.run(["otool", "-L", os.path.join(casadi_dir, libcpp_name)]) -print(" ".join(["install_name_tool"] + install_name_tool_args)) -subprocess.run(["install_name_tool"] + install_name_tool_args) +print(" ".join(["install_name_tool", *install_name_tool_args])) +subprocess.run(["install_name_tool", *install_name_tool_args]) -subprocess.run(["otool"] + ["-L", os.path.join(casadi_dir, libcpp_name)]) +subprocess.run(["otool", "-L", os.path.join(casadi_dir, libcpp_name)]) # Copy libcasadi.3.7.dylib and libc++.1.0.dylib to LD_LIBRARY_PATH # This is needed for the casadi python bindings to work while repairing the wheel diff --git a/scripts/install_KLU_Sundials.py b/scripts/install_KLU_Sundials.py index 96e60aeb0e..8f41f5969a 100755 --- a/scripts/install_KLU_Sundials.py +++ b/scripts/install_KLU_Sundials.py @@ -156,7 +156,7 @@ def download_extract_library(url, download_dir): sundials_src = "../sundials-{}".format(sundials_version) print("-" * 10, "Running CMake prepare", "-" * 40) -subprocess.run(["cmake", sundials_src] + cmake_args, cwd=build_dir, check=True) +subprocess.run(["cmake", sundials_src, *cmake_args], cwd=build_dir, check=True) print("-" * 10, "Building the sundials", "-" * 40) make_cmd = ["make", "install"] diff --git a/scripts/update_version.py b/scripts/update_version.py index 8a2d832e59..ccdec661e2 100644 --- a/scripts/update_version.py +++ b/scripts/update_version.py @@ -78,7 +78,7 @@ def update_version(): file.write(replace_version) # Get latest commit id from pybamm-team/sundials-vcpkg-registry - cmd = "git ls-remote https://github.com/pybamm-team/sundials-vcpkg-registry | grep refs/heads/main | cut -f 1 | tr -d '\n'" # noqa: E501 + cmd = "git ls-remote https://github.com/pybamm-team/sundials-vcpkg-registry | grep refs/heads/main | cut -f 1 | tr -d '\n'" latest_commit_id = os.popen(cmd).read() # vcpkg-configuration.json @@ -93,7 +93,7 @@ def update_version(): file.write(replace_commit_id) changelog_line1 = "# [Unreleased](https://github.com/pybamm-team/PyBaMM/)\n" - changelog_line2 = f"# [v{release_version}](https://github.com/pybamm-team/PyBaMM/tree/v{release_version}) - {last_day_of_month}\n\n" # noqa: E501 + changelog_line2 = f"# [v{release_version}](https://github.com/pybamm-team/PyBaMM/tree/v{release_version}) - {last_day_of_month}\n\n" # CHANGELOG.md with open(os.path.join(pybamm.root_dir(), "CHANGELOG.md"), "r+") as file: diff --git a/setup.py b/setup.py index 9cfc4df4ff..ef82e65e70 100644 --- a/setup.py +++ b/setup.py @@ -42,10 +42,7 @@ def set_vcpkg_environment_variables(): # ---------- CMakeBuild class (custom build_ext for IDAKLU target) --------------------- class CMakeBuild(build_ext): - user_options = build_ext.user_options + [ - ("suitesparse-root=", None, "suitesparse source location"), - ("sundials-root=", None, "sundials source location"), - ] + user_options = [*build_ext.user_options, ("suitesparse-root=", None, "suitesparse source location"), ("sundials-root=", None, "sundials source location")] def initialize_options(self): build_ext.initialize_options(self) @@ -136,7 +133,7 @@ def run(self): cmake_list_dir = os.path.abspath(os.path.dirname(__file__)) print("-" * 10, "Running CMake for IDAKLU solver", "-" * 40) subprocess.run( - ["cmake", cmake_list_dir] + cmake_args, cwd=build_dir, env=build_env + ["cmake", cmake_list_dir, *cmake_args], cwd=build_dir, env=build_env , check=True) if os.path.isfile(os.path.join(build_dir, "CMakeError.log")): @@ -144,7 +141,7 @@ def run(self): "cmake configuration steps encountered errors, and the IDAKLU module" " could not be built. Make sure dependencies are correctly " "installed. See " - "https://docs.pybamm.org/en/latest/source/user_guide/installation/install-from-source.html" # noqa: E501 + "https://docs.pybamm.org/en/latest/source/user_guide/installation/install-from-source.html" ) raise RuntimeError(msg) else: @@ -201,10 +198,7 @@ def move_output(self, ext): class CustomInstall(install): """A custom install command to add 2 build options""" - user_options = install.user_options + [ - ("suitesparse-root=", None, "suitesparse source location"), - ("sundials-root=", None, "sundials source location"), - ] + user_options = [*install.user_options, ("suitesparse-root=", None, "suitesparse source location"), ("sundials-root=", None, "sundials source location")] def initialize_options(self): install.initialize_options(self) @@ -228,10 +222,7 @@ def run(self): class bdist_wheel(orig.bdist_wheel): """A custom install command to add 2 build options""" - user_options = orig.bdist_wheel.user_options + [ - ("suitesparse-root=", None, "suitesparse source location"), - ("sundials-root=", None, "sundials source location"), - ] + user_options = [*orig.bdist_wheel.user_options, ("suitesparse-root=", None, "suitesparse source location"), ("sundials-root=", None, "sundials source location")] def initialize_options(self): orig.bdist_wheel.initialize_options(self)

Move all the extra configurations to `pyproject.toml` Once #3301 is merged, we can move the configurations of pytest, ruff, and others to `pyproject.toml`

I was going to open this issue, you beat me to it! 😄 Some more tools whose configuration can be moved: `cibuildwheel`, `coverage.py`, and subsequently `towncrier` – after #3469 is merged)

2023-11-25T15:20:47

0.0

[]

eclipse-volttron/volttron-openadr-ven

eclipse-volttron__volttron-openadr-ven-17

da1dd24743d0b35c59fa7f8d90cf909654168979

diff --git a/.gitignore b/.gitignore index 94cae34..6063ad4 100644 --- a/.gitignore +++ b/.gitignore @@ -132,3 +132,4 @@ dmypy.json *secret config_local.json +.DS_Store diff --git a/.gitmodules b/.gitmodules deleted file mode 100644 index 8dd3851..0000000 --- a/.gitmodules +++ /dev/null @@ -1,3 +0,0 @@ -[submodule "openleadr"] - path = openleadr - url = [email protected]:bonicim/openleadr-python.git diff --git a/README.md b/README.md index 933493b..0a6db97 100644 --- a/README.md +++ b/README.md @@ -14,15 +14,14 @@ correct configuration with optional parameters added. { "ven_name": "PNNLVEN", "vtn_url": "https://eiss2demo.ipkeys.com/oadr2/OpenADR2/Simple/2.0b", + "openadr_client_type": "IPKeysClient" # the list of valid client types are the class names of the OpenADRClient subclasses in ~volttron_openadr_ven/volttron_openadr_client.py # below are optional parameters - # if you want/need to sign outgoing messages using a public-private key pair, you need to add a directory named "secret" in the top level - # of this repo; then provide the relative path to the cert and key - # below is an example - # test certificates for IP Keys VTN's can be obtained at https://testcerts.kyrio.com/#/ - "cert": "secret/TEST_RSA_VEN_210923215148_certs/TEST_RSA_VEN_210923215148_cert.pem", - "key": "secret/TEST_RSA_VEN_210923215148_certs/TEST_RSA_VEN_210923215148_privkey.pem", + # if you want/need to sign outgoing messages using a public-private key pair, provide the relative path to the cert_path and key_path + # in this example, the keypair is stored in the directory named '~/.ssh/secret' + "cert_path": "~/.ssh/secret/TEST_RSA_VEN_210923215148_cert.pem", + "key_path": "~/.ssh/secret/TEST_RSA_VEN_210923215148_privkey.pem", # other optional parameters "debug": true, @@ -30,32 +29,58 @@ correct configuration with optional parameters added. } ``` -## Volttron integration +Save this configuration in a JSON file in your preferred location; we recommend saving it in the `volttron_openadr_ven` directory. +An example of such a configuration is saved in the root of this repository; the file is named `config_example1.json` -To ensure that this agent is created within the Volttron framework, ensure that you set the following environmental variable -to the absolute path of the configuration JSON file for this agent. For example, on Linux, use the following example to set -the environmental variable AGENT_CONFIG to a path to your JSON config file: -`export AGENT_CONFIG /home/usr/path-to-my-config-file` +# Quickstart -# Development +## Installing the agent on Monolith Volttron -## Prerequisites +1. Start the Volttron platform on your machine + +```shell +volttron -vv -l volttron.log & +``` + +2. Tail the logs so you can observe OpenADRClient logs + +```shell +tail -f volttron.log +``` + +3. Install the agent on Volttron + +```shell +vctl -v install <path to root directory of volttron-openadr-ven> \ +--tag openadr \ +--agent-config <path to agent config> +``` + +4. Verify status of agent +```shell +vctl status +``` + +5. Start the agent +```shell +vctl start --tag openadr +``` + +## Starting the agent on Modular Volttron + + +### Prerequisites * Python >=3.7 * Pip >=20.1 * Poetry >=1.16 - -## Environment setup +### 1. Environment setup Note: This repo uses [Poetry](https://python-poetry.org/), a dependency management and packaging tool for Python. If you don't have Poetry installed on your machine, follow [these steps](https://python-poetry.org/docs/#installation) to install it on your machine. To check if Poetry is installed, run `poetry --version`. If you receive the error 'command not found: poetry', add the following line to your '~/.bashrc' script: ```export PATH=$PATH:$HOME/.poetry/bin```. -0. Install the required submodules - -```git submodule update --init --recursive``` - 1. Create virtual environment By default, poetry creates a virtual environment in {cache-dir}/virtualenvs @@ -71,11 +96,6 @@ Then to create the virtual environment itself, run the following command: poetry shell ``` -NOTE: Poetry needs the command 'python' in order to create a new shell. On Linux, if you only have python3 installed, run the following to link 'python' to 'python3': -```shell -sudo apt install python-is-python3 -``` - 2. Install requirements ```shell @@ -83,80 +103,141 @@ poetry install ``` -## Code Quality and Standardization +The environment is now setup to run the OpenADRVen agent. Next, we need to set up a local Volttron platform and configure the +remote VTN server that this OpenADRVen agent will connect to. -## Git Hooks and Pre-commit +### 2. Local Volttron platform setup -This repo provides pre-commit hooks to enforce code quality and standardization before pushing any commit to the repo. To use this tool locally, -run the following command. +We need to start a local Volttron platform on the same machine that is hosting this OpenADRVen agent. Follow the [installation steps](https://volttron.readthedocs.io/en/develop/introduction/platform-install.html) +for installing a single instance of the Volttron platform. After you have installed and started the Volttron platform, we need +to get the platform's server key and create a public and secret keys. We will use these keys to configure the credentials of the +OpenADRVen agent so that the local Volttron platform can successfully authenticate the OpenADRVen agent. + +**Steps** + +1. With the local Volttron platform running, open a new shell and run `vctl auth serverkey`. Copy and paste output in the agent's config file. See example below: ```shell -pre-commit install +$ vctl auth serverkey +FDIH6OBg0m3L2T6Jv_zUuQlSxYdgvTD3QOEye-vM-iI + +# Paste the output in config file for agent + +{ + { + "ven_name": "PNNLVEN", + ... + "server_key": "FDIH6OBg0m3L2T6Jv_zUuQlSxYdgvTD3QOEye-vM-iI" + } +} ``` -To check that pre-commit scripts are installed and properly run, run pre-commit on all the files: +2. On the same shell, run `vctl auth keypair`. Copy the values of `public` and `secret` and paste them in the agent's config file. See example below: +```shell +$ vctl auth keypair +{ + "public": "dyMf8g58ptiQTqI7EHbNr7PPFsAr2y-OIy1J-he_DCU", + "secret": "iJa4mlZBomb3bDP8oRAN4IK8uXEqpl1MDg_te3aJzMo" +} + +# Paste the output in config file for agent + +{ + { + "ven_name": "PNNLVEN", + ... + "server_key": "FDIH6OBg0m3L2T6Jv_zUuQlSxYdgvTD3QOEye-vM-iI", + "agent_public": "csJBQqQDZ-pP_8E9FIgM9hvkAak6HriLkIQhP46ZFl4", + "agent_secret": "4rv_l3oEzgmRxbPkGma2-tMhfPu47yyhi48ygF5hVQY" + } +} ``` -pre-commit run --all + +3. Add the public key, which was generated in the previous step, to the Volttron platform. Use `vctl auth add --credentials <public key>'. +An example of the command being run is shown below: + +```shell +$ vctl auth add --credentials 'csJBQqQDZ-pP_8E9FIgM9hvkAak6HriLkIQhP46ZFl4' ``` -With pre-commit installed and configured, every time you run `git commit`, git will check your commit against a set of code linters. For details, see .pre-commit-config.yaml +### 3. VTN Server setup -This project installs the following tools to aid in code standardization and best practices: +Depending on the type of VTN that you are using, you need to configure your VTN to send events so that the OpenADRVen agent +can receive such events from your VTN. -## Black -This repo uses [Black](https://pypi.org/project/black/), an opinionated Python code formatter. To use Black to check and automagically reformat your code, -run the following command on your chosen Python file: +#### IPKeys VTN configuration -```shell -black <path to file> -``` +The community is currently testing this OpenADRVen agent against a IPKeys VTN. To configure the agent with the right +certificates, follow the instructions below: + +Get VEN certificates at https://testcerts.kyrio.com/#/. Certificates can be stored anywhere on your machine, however, +it is recommended to place your keypair files in your ~/.ssh/ folder and make those files read-only. + +* IPKeys VTN web browser console: https://eiss2demo.ipkeys.com/ui/home/#/login + +To create an event, click "Events" tab. Then click the "+" icon to create an event. Use the template "PNNLTestEvent" or +"PNNLnontestevent" to create an event. + +![Alt text](screenshots/test_event_screenshot_ipkeys.png?raw=true "Screenshot of Events page of IPKeys GUI") + +### 4. Starting the OpenADRVen agent + +Now that we have both the environment for OpenADRVen agent setup and the Volttron platform and VTN's configured properly, +we can finally run our OpenADRVen agent. Note: we are NOT installing this agent on the local Volttron platform. We will +simply run this agent on the host machine. Navigate to the root level of this repo. Then run the ~/volttron_openadr_ven/agent.py +module with the path to the agent configuration file that you created as noted in the "Agent Configuration" section at the beginning of this README. -## MyPy -This repo uses [MyPy](https://mypy.readthedocs.io/en/stable/), a static type checker for Python. To use MyPy on a Python file, run the following command: ```shell -mypy <path to file> +poetry run python -m volttron_openadr_ven.agent <path to the config file> ``` -# CI/CD -TODO: +# Development, Code Quality, and Standardization -# Building Wheel and Publishing to PyPi +## OpenLeadr Notes (IMPORTANT) -TODO: +The OpenADRVen agent uses a third-party library, OpenLeader[https://openleadr.org/docs/index.html], as the actual client. +However, the OpenADRVen agent had to modify some parts of that client in order to connect to the IPKeys VTN for testing. Specifically, +OpenADRVen agent extends the OpenLeadr client class and overrides some class methods. -Build wheel -```python setup.py -q sdist bdist_wheel``` +Those parts that were modified are located in ~/volttron_openadr_ven/volttron_openleadr.py. Future releases of OpenLeadr could potentially and adversely +affect OpenADRVen agent if those releases directly or indirectly affect the modified code blocks. Thus, maintenance of this agent should closely monitor future changes to OpenLeadr. -Check wheel -```check-wheel-contents dist/*.whl``` -```tar tzf <>``` -Test uploading -```twine upload --repository-url https://test.pypi.org/legacy/ dist/*``` +## Git Hooks and Pre-commit -Upload to PyPi -```python -m twin upload dist/*``` +This repo provides pre-commit hooks to enforce code quality and standardization before pushing any commit to the repo. To use this tool locally, +run the following command. +```shell +pre-commit install +``` -# Testing +To check that pre-commit scripts are installed and properly run, run pre-commit on all the files: -## Ven Certificates +``` +pre-commit run --all +``` -https://testcerts.kyrio.com/#/ +With pre-commit installed and configured, every time you run `git commit`, git will check your commit against a set of code linters. For details, see .pre-commit-config.yaml -## Test VTN's +This project installs the following tools to aid in code standardization and best practices: -* IPKeys: https://eiss2demo.ipkeys.com/ui/home/#/login +## Black -# Other +This repo uses [Black](https://pypi.org/project/black/), an opinionated Python code formatter. To use Black to check and automagically reformat your code, +run the following command on your chosen Python file: -## Installing dependencies behind a proxy +```shell +black <path to file> +``` -If running this behind a PNNL VPN, run the following so Poetry can install all the dependencies: -See: https://leifengblog.net/blog/how-to-use-pip-behind-a-proxy/ +## MyPy -```export https_proxy=http://proxy01.pnl.gov:3128``` +This repo uses [MyPy](https://mypy.readthedocs.io/en/stable/), a static type checker for Python. To use MyPy on a Python file, run the following command: +```shell +mypy <path to file> +``` diff --git a/config_example1.json b/config_example1.json new file mode 100644 index 0000000..bfed53d --- /dev/null +++ b/config_example1.json @@ -0,0 +1,13 @@ +{ + "ven_name": "PNNLVEN", + "vtn_url": "https://eiss2demo.ipkeys.com/oadr2/OpenADR2/Simple/2.0b", + "cert_path": "~/.ssh/secret/path_to_cert.pem", + "key_path": "~/.ssh/secret/path_to_privkey.pem", + "debug": true, + "disable_signature": true, + "vip_address": "tcp://127.0.0.1:22916", + "server_key": "FDIH6OBg0m3L2T6Jv_zUuQlSxYdgvTD3QOEye-vM-iI", + "agent_public": "csJBQqQDZ-pP_8E9FIgM9hvkAak6HriLkIQhP46ZFl4", + "agent_secret": "4rv_l3oEzgmRxbPkGma2-tMhfPu47yyhi48ygF5hVQY", + "openadr_client_type": "IPKeysClient" +} diff --git a/openleadr b/openleadr deleted file mode 160000 index e306798..0000000 --- a/openleadr +++ /dev/null @@ -1,1 +0,0 @@ -Subproject commit e306798f9f4af36f0349c10fec4607aaa18d34c8 diff --git a/poetry.lock b/poetry.lock index b2f2e1c..0568674 100644 --- a/poetry.lock +++ b/poetry.lock @@ -1,25 +1,37 @@ [[package]] name = "aiohttp" -version = "3.7.4.post0" +version = "3.8.1" description = "Async http client/server framework (asyncio)" category = "main" optional = false python-versions = ">=3.6" [package.dependencies] -async-timeout = ">=3.0,<4.0" +aiosignal = ">=1.1.2" +async-timeout = ">=4.0.0a3,<5.0" attrs = ">=17.3.0" -chardet = ">=2.0,<5.0" +charset-normalizer = ">=2.0,<3.0" +frozenlist = ">=1.1.1" multidict = ">=4.5,<7.0" -typing-extensions = ">=3.6.5" yarl = ">=1.0,<2.0" [package.extras] -speedups = ["aiodns", "brotlipy", "cchardet"] +speedups = ["aiodns", "brotli", "cchardet"] + +[[package]] +name = "aiosignal" +version = "1.2.0" +description = "aiosignal: a list of registered asynchronous callbacks" +category = "main" +optional = false +python-versions = ">=3.6" + +[package.dependencies] +frozenlist = ">=1.1.0" [[package]] name = "apscheduler" -version = "3.8.0" +version = "3.8.1" description = "In-process task scheduler with Cron-like capabilities" category = "main" optional = false @@ -28,7 +40,7 @@ python-versions = ">=2.7, !=3.0.*, !=3.1.*, !=3.2.*, !=3.3.*, !=3.4.*, <4" [package.dependencies] pytz = "*" six = ">=1.4.0" -tzlocal = ">=2.0,<3.0" +tzlocal = ">=2.0,<3.0.0 || >=4.0.0" [package.extras] asyncio = ["trollius"] @@ -53,11 +65,14 @@ python-versions = "*" [[package]] name = "async-timeout" -version = "3.0.1" +version = "4.0.1" description = "Timeout context manager for asyncio programs" category = "main" optional = false -python-versions = ">=3.5.3" +python-versions = ">=3.6" + +[package.dependencies] +typing-extensions = ">=3.6.5" [[package]] name = "atomicwrites" @@ -83,7 +98,7 @@ tests_no_zope = ["coverage[toml] (>=5.0.2)", "hypothesis", "pympler", "pytest (> [[package]] name = "backports.entry-points-selectable" -version = "1.1.0" +version = "1.1.1" description = "Compatibility shim providing selectable entry points for older implementations" category = "main" optional = false @@ -91,11 +106,11 @@ python-versions = ">=2.7" [package.extras] docs = ["sphinx", "jaraco.packaging (>=8.2)", "rst.linker (>=1.9)"] -testing = ["pytest (>=4.6)", "pytest-flake8", "pytest-cov", "pytest-black (>=0.3.7)", "pytest-mypy", "pytest-checkdocs (>=2.4)", "pytest-enabler (>=1.0.1)"] +testing = ["pytest", "pytest-flake8", "pytest-cov", "pytest-black (>=0.3.7)", "pytest-mypy", "pytest-checkdocs (>=2.4)", "pytest-enabler (>=1.0.1)"] [[package]] name = "black" -version = "21.9b0" +version = "21.11b1" description = "The uncompromising code formatter." category = "main" optional = false @@ -106,7 +121,7 @@ click = ">=7.1.2" mypy-extensions = ">=0.4.3" pathspec = ">=0.9.0,<1" platformdirs = ">=2" -regex = ">=2020.1.8" +regex = ">=2021.4.4" tomli = ">=0.2.6,<2.0.0" typing-extensions = [ {version = ">=3.10.0.0", markers = "python_version < \"3.10\""}, @@ -115,9 +130,9 @@ typing-extensions = [ [package.extras] colorama = ["colorama (>=0.4.3)"] -d = ["aiohttp (>=3.6.0)", "aiohttp-cors (>=0.4.0)"] +d = ["aiohttp (>=3.7.4)"] jupyter = ["ipython (>=7.8.0)", "tokenize-rt (>=3.2.0)"] -python2 = ["typed-ast (>=1.4.2)"] +python2 = ["typed-ast (>=1.4.3)"] uvloop = ["uvloop (>=0.15.2)"] [[package]] @@ -147,17 +162,9 @@ category = "main" optional = false python-versions = ">=3.6.1" -[[package]] -name = "chardet" -version = "4.0.0" -description = "Universal encoding detector for Python 2 and 3" -category = "main" -optional = false -python-versions = ">=2.7, !=3.0.*, !=3.1.*, !=3.2.*, !=3.3.*, !=3.4.*" - [[package]] name = "charset-normalizer" -version = "2.0.7" +version = "2.0.9" description = "The Real First Universal Charset Detector. Open, modern and actively maintained alternative to Chardet." category = "main" optional = false @@ -238,7 +245,7 @@ future = ">=0.16,<0.19" [[package]] name = "filelock" -version = "3.3.1" +version = "3.4.0" description = "A platform independent file lock." category = "main" optional = false @@ -248,6 +255,14 @@ python-versions = ">=3.6" docs = ["furo (>=2021.8.17b43)", "sphinx (>=4.1)", "sphinx-autodoc-typehints (>=1.12)"] testing = ["covdefaults (>=1.2.0)", "coverage (>=4)", "pytest (>=4)", "pytest-cov", "pytest-timeout (>=1.4.2)"] +[[package]] +name = "frozenlist" +version = "1.2.0" +description = "A list-like structure which implements collections.abc.MutableSequence" +category = "main" +optional = false +python-versions = ">=3.6" + [[package]] name = "future" version = "0.18.2" @@ -299,22 +314,22 @@ requests = "*" [[package]] name = "identify" -version = "2.3.0" +version = "2.4.0" description = "File identification library for Python" category = "main" optional = false python-versions = ">=3.6.1" [package.extras] -license = ["editdistance-s"] +license = ["ukkonen"] [[package]] name = "idna" -version = "2.10" +version = "3.3" description = "Internationalized Domain Names in Applications (IDNA)" category = "main" optional = false -python-versions = ">=2.7, !=3.0.*, !=3.1.*, !=3.2.*, !=3.3.*" +python-versions = ">=3.5" [[package]] name = "iniconfig" @@ -326,7 +341,7 @@ python-versions = "*" [[package]] name = "jinja2" -version = "3.0.2" +version = "3.0.3" description = "A very fast and expressive template engine." category = "main" optional = false @@ -340,7 +355,7 @@ i18n = ["Babel (>=2.7)"] [[package]] name = "lxml" -version = "4.6.3" +version = "4.6.4" description = "Powerful and Pythonic XML processing library combining libxml2/libxslt with the ElementTree API." category = "main" optional = false @@ -403,11 +418,10 @@ python-versions = "*" [[package]] name = "openleadr" version = "0.5.24" -description = "" +description = "Python3 library for building OpenADR Clients (VENs) and Servers (VTNs)" category = "main" optional = false python-versions = ">=3.7.0" -develop = true [package.dependencies] aiohttp = "*" @@ -416,20 +430,16 @@ jinja2 = "*" signxml-openadr = "2.9.1" xmltodict = "*" -[package.source] -type = "directory" -url = "openleadr" - [[package]] name = "packaging" -version = "21.0" +version = "21.3" description = "Core utilities for Python packages" category = "dev" optional = false python-versions = ">=3.6" [package.dependencies] -pyparsing = ">=2.0.2" +pyparsing = ">=2.0.2,<3.0.5 || >3.0.5" [[package]] name = "pathspec" @@ -439,6 +449,14 @@ category = "main" optional = false python-versions = "!=3.0.*,!=3.1.*,!=3.2.*,!=3.3.*,!=3.4.*,>=2.7" +[[package]] +name = "pbr" +version = "5.8.0" +description = "Python Build Reasonableness" +category = "main" +optional = false +python-versions = ">=2.6" + [[package]] name = "platformdirs" version = "2.4.0" @@ -465,7 +483,7 @@ testing = ["pytest", "pytest-benchmark"] [[package]] name = "pre-commit" -version = "2.15.0" +version = "2.16.0" description = "A framework for managing and maintaining multi-language pre-commit hooks." category = "main" optional = false @@ -492,15 +510,15 @@ test = ["ipaddress", "mock", "unittest2", "enum34", "pywin32", "wmi"] [[package]] name = "py" -version = "1.10.0" +version = "1.11.0" description = "library with cross-python path, ini-parsing, io, code, log facilities" category = "main" optional = false -python-versions = ">=2.7, !=3.0.*, !=3.1.*, !=3.2.*, !=3.3.*" +python-versions = ">=2.7, !=3.0.*, !=3.1.*, !=3.2.*, !=3.3.*, !=3.4.*" [[package]] name = "pycparser" -version = "2.20" +version = "2.21" description = "C parser in Python" category = "main" optional = false @@ -524,11 +542,14 @@ test = ["flaky", "pretend", "pytest (>=3.0.1)"] [[package]] name = "pyparsing" -version = "2.4.7" +version = "3.0.6" description = "Python parsing module" category = "dev" optional = false -python-versions = ">=2.6, !=3.0.*, !=3.1.*, !=3.2.*" +python-versions = ">=3.6" + +[package.extras] +diagrams = ["jinja2", "railroad-diagrams"] [[package]] name = "pytest" @@ -572,11 +593,11 @@ python-versions = "*" [[package]] name = "pyyaml" -version = "5.4.1" +version = "6.0" description = "YAML parser and emitter for Python" category = "main" optional = false -python-versions = ">=2.7, !=3.0.*, !=3.1.*, !=3.2.*, !=3.3.*, !=3.4.*, !=3.5.*" +python-versions = ">=3.6" [[package]] name = "pyzmq" @@ -592,7 +613,7 @@ py = {version = "*", markers = "implementation_name == \"pypy\""} [[package]] name = "regex" -version = "2021.10.21" +version = "2021.11.10" description = "Alternative regular expression module, to replace re." category = "main" optional = false @@ -649,7 +670,7 @@ python-versions = ">=2.6, !=3.0.*, !=3.1.*, !=3.2.*" [[package]] name = "tomli" -version = "1.2.1" +version = "1.2.2" description = "A lil' TOML parser" category = "main" optional = false @@ -665,11 +686,11 @@ python-versions = "*" [[package]] name = "typing-extensions" -version = "3.10.0.2" -description = "Backported and Experimental Type Hints for Python 3.5+" +version = "4.0.1" +description = "Backported and Experimental Type Hints for Python 3.6+" category = "main" optional = false -python-versions = "*" +python-versions = ">=3.6" [[package]] name = "tzlocal" @@ -697,7 +718,7 @@ socks = ["PySocks (>=1.5.6,!=1.5.7,<2.0)"] [[package]] name = "virtualenv" -version = "20.8.1" +version = "20.10.0" description = "Virtual Python Environment builder" category = "main" optional = false @@ -706,49 +727,83 @@ python-versions = "!=3.0.*,!=3.1.*,!=3.2.*,!=3.3.*,!=3.4.*,>=2.7" [package.dependencies] "backports.entry-points-selectable" = ">=1.0.4" distlib = ">=0.3.1,<1" -filelock = ">=3.0.0,<4" +filelock = ">=3.2,<4" platformdirs = ">=2,<3" six = ">=1.9.0,<2" [package.extras] -docs = ["proselint (>=0.10.2)", "sphinx (>=3)", "sphinx-argparse (>=0.2.5)", "sphinx-rtd-theme (>=0.4.3)", "towncrier (>=19.9.0rc1)"] +docs = ["proselint (>=0.10.2)", "sphinx (>=3)", "sphinx-argparse (>=0.2.5)", "sphinx-rtd-theme (>=0.4.3)", "towncrier (>=21.3)"] testing = ["coverage (>=4)", "coverage-enable-subprocess (>=1)", "flaky (>=3)", "pytest (>=4)", "pytest-env (>=0.6.2)", "pytest-freezegun (>=0.4.1)", "pytest-mock (>=2)", "pytest-randomly (>=1)", "pytest-timeout (>=1)", "packaging (>=20.0)"] [[package]] name = "volttron-client" -version = "0.3.4" -description = "Client for connecting to a volttron server" +version = "0.3.8" +description = "\"VOLTTRON client library to connect to volttron servers\"" category = "main" optional = false -python-versions = ">=3.6.2,<4" +python-versions = ">=3.7, <4" [package.dependencies] -dateutils = ">=0.6,<0.7" -grequests = ">=0.6.0,<0.7.0" -idna = ">=2.5,<3" -requests = ">=2.26.0,<3.0.0" -toml = ">=0.10.2,<0.11.0" -volttron-utils = ">=0.3,<0.4" +asn1crypto = "1.4.0" +certifi = "2021.10.8" +cffi = "1.15.0" +charset-normalizer = {version = "2.0.9", markers = "python_version >= \"3\""} +cryptography = "2.3" +dateutils = "0.6.12" +gevent = "20.6.1" +greenlet = {version = "0.4.16", markers = "platform_python_implementation == \"CPython\""} +grequests = "0.6.0" +idna = {version = "3.3", markers = "python_version >= \"3\""} +pbr = "5.8.0" +psutil = "5.8.0" +pycparser = "2.21" +pyopenssl = "19.0.0" +python-dateutil = "2.8.2" +pytz = "2021.3" +pyyaml = "6.0" +pyzmq = "22.3.0" +requests = "2.26.0" +six = "1.16.0" +tzlocal = "2.1" +urllib3 = "1.26.7" +volttron-utils = "0.3.4" +watchdog = "2.1.6" +watchdog-gevent = "0.1.1" +zmq = "0.0.0" +"zope.event" = "4.5.0" +"zope.interface" = "5.4.0" [[package]] name = "volttron-utils" -version = "0.3.2" +version = "0.3.4" description = "Utility classes and methods for VOLTTRON" category = "main" optional = false -python-versions = ">=3.6.2,<4.0" +python-versions = ">=3.7, <4" [package.dependencies] +asn1crypto = "1.4.0" +cffi = "1.15.0" cryptography = "2.3" -dateutils = ">=0.6.7,<0.7.0" +dateutils = "0.6.12" gevent = "20.6.1" -greenlet = "0.4.16" -psutil = ">=5.7.0,<6.0.0" -pyOpenSSL = "19.0.0" -PyYAML = ">=5.4.1,<6.0.0" +greenlet = {version = "0.4.16", markers = "platform_python_implementation == \"CPython\""} +idna = "3.3" +pbr = "5.8.0" +psutil = "5.8.0" +pycparser = "2.21" +pyopenssl = "19.0.0" +python-dateutil = "2.8.2" +pytz = "2021.3" +pyyaml = "6.0" +pyzmq = "22.3.0" +six = "1.16.0" tzlocal = "2.1" -watchdog-gevent = ">=0.1.1,<0.2.0" -zmq = ">=0.0.0,<0.0.1" +watchdog = "2.1.6" +watchdog-gevent = "0.1.1" +zmq = "0.0.0" +"zope.event" = "4.5.0" +"zope.interface" = "5.4.0" [[package]] name = "watchdog" @@ -783,7 +838,7 @@ python-versions = ">=2.7, !=3.0.*, !=3.1.*, !=3.2.*, !=3.3.*" [[package]] name = "yarl" -version = "1.7.0" +version = "1.7.2" description = "Yet another URL library" category = "main" optional = false @@ -832,59 +887,98 @@ testing = ["coverage (>=5.0.3)", "zope.event", "zope.testing"] [metadata] lock-version = "1.1" python-versions = "^3.8" -content-hash = "05aefd0e959b8da3ef1df6dbd3f3ec194bfe8b7cbc8d49f393a5919cd78c075e" +content-hash = "a9c58b9265431cbbc61fc25e65851b944f320d1f21f4c74be7f043d40e4c7c6b" [metadata.files] aiohttp = [ - {file = "aiohttp-3.7.4.post0-cp36-cp36m-macosx_10_14_x86_64.whl", hash = "sha256:3cf75f7cdc2397ed4442594b935a11ed5569961333d49b7539ea741be2cc79d5"}, - {file = "aiohttp-3.7.4.post0-cp36-cp36m-manylinux1_i686.whl", hash = "sha256:4b302b45040890cea949ad092479e01ba25911a15e648429c7c5aae9650c67a8"}, - {file = "aiohttp-3.7.4.post0-cp36-cp36m-manylinux2014_aarch64.whl", hash = "sha256:fe60131d21b31fd1a14bd43e6bb88256f69dfc3188b3a89d736d6c71ed43ec95"}, - {file = "aiohttp-3.7.4.post0-cp36-cp36m-manylinux2014_i686.whl", hash = "sha256:393f389841e8f2dfc86f774ad22f00923fdee66d238af89b70ea314c4aefd290"}, - {file = "aiohttp-3.7.4.post0-cp36-cp36m-manylinux2014_ppc64le.whl", hash = "sha256:c6e9dcb4cb338d91a73f178d866d051efe7c62a7166653a91e7d9fb18274058f"}, - {file = "aiohttp-3.7.4.post0-cp36-cp36m-manylinux2014_s390x.whl", hash = "sha256:5df68496d19f849921f05f14f31bd6ef53ad4b00245da3195048c69934521809"}, - {file = "aiohttp-3.7.4.post0-cp36-cp36m-manylinux2014_x86_64.whl", hash = "sha256:0563c1b3826945eecd62186f3f5c7d31abb7391fedc893b7e2b26303b5a9f3fe"}, - {file = "aiohttp-3.7.4.post0-cp36-cp36m-win32.whl", hash = "sha256:3d78619672183be860b96ed96f533046ec97ca067fd46ac1f6a09cd9b7484287"}, - {file = "aiohttp-3.7.4.post0-cp36-cp36m-win_amd64.whl", hash = "sha256:f705e12750171c0ab4ef2a3c76b9a4024a62c4103e3a55dd6f99265b9bc6fcfc"}, - {file = "aiohttp-3.7.4.post0-cp37-cp37m-macosx_10_14_x86_64.whl", hash = "sha256:230a8f7e24298dea47659251abc0fd8b3c4e38a664c59d4b89cca7f6c09c9e87"}, - {file = "aiohttp-3.7.4.post0-cp37-cp37m-manylinux1_i686.whl", hash = "sha256:2e19413bf84934d651344783c9f5e22dee452e251cfd220ebadbed2d9931dbf0"}, - {file = "aiohttp-3.7.4.post0-cp37-cp37m-manylinux2014_aarch64.whl", hash = "sha256:e4b2b334e68b18ac9817d828ba44d8fcb391f6acb398bcc5062b14b2cbeac970"}, - {file = "aiohttp-3.7.4.post0-cp37-cp37m-manylinux2014_i686.whl", hash = "sha256:d012ad7911653a906425d8473a1465caa9f8dea7fcf07b6d870397b774ea7c0f"}, - {file = "aiohttp-3.7.4.post0-cp37-cp37m-manylinux2014_ppc64le.whl", hash = "sha256:40eced07f07a9e60e825554a31f923e8d3997cfc7fb31dbc1328c70826e04cde"}, - {file = "aiohttp-3.7.4.post0-cp37-cp37m-manylinux2014_s390x.whl", hash = "sha256:209b4a8ee987eccc91e2bd3ac36adee0e53a5970b8ac52c273f7f8fd4872c94c"}, - {file = "aiohttp-3.7.4.post0-cp37-cp37m-manylinux2014_x86_64.whl", hash = "sha256:14762875b22d0055f05d12abc7f7d61d5fd4fe4642ce1a249abdf8c700bf1fd8"}, - {file = "aiohttp-3.7.4.post0-cp37-cp37m-win32.whl", hash = "sha256:7615dab56bb07bff74bc865307aeb89a8bfd9941d2ef9d817b9436da3a0ea54f"}, - {file = "aiohttp-3.7.4.post0-cp37-cp37m-win_amd64.whl", hash = "sha256:d9e13b33afd39ddeb377eff2c1c4f00544e191e1d1dee5b6c51ddee8ea6f0cf5"}, - {file = "aiohttp-3.7.4.post0-cp38-cp38-macosx_10_14_x86_64.whl", hash = "sha256:547da6cacac20666422d4882cfcd51298d45f7ccb60a04ec27424d2f36ba3eaf"}, - {file = "aiohttp-3.7.4.post0-cp38-cp38-manylinux1_i686.whl", hash = "sha256:af9aa9ef5ba1fd5b8c948bb11f44891968ab30356d65fd0cc6707d989cd521df"}, - {file = "aiohttp-3.7.4.post0-cp38-cp38-manylinux2014_aarch64.whl", hash = "sha256:64322071e046020e8797117b3658b9c2f80e3267daec409b350b6a7a05041213"}, - {file = "aiohttp-3.7.4.post0-cp38-cp38-manylinux2014_i686.whl", hash = "sha256:bb437315738aa441251214dad17428cafda9cdc9729499f1d6001748e1d432f4"}, - {file = "aiohttp-3.7.4.post0-cp38-cp38-manylinux2014_ppc64le.whl", hash = "sha256:e54962802d4b8b18b6207d4a927032826af39395a3bd9196a5af43fc4e60b009"}, - {file = "aiohttp-3.7.4.post0-cp38-cp38-manylinux2014_s390x.whl", hash = "sha256:a00bb73540af068ca7390e636c01cbc4f644961896fa9363154ff43fd37af2f5"}, - {file = "aiohttp-3.7.4.post0-cp38-cp38-manylinux2014_x86_64.whl", hash = "sha256:79ebfc238612123a713a457d92afb4096e2148be17df6c50fb9bf7a81c2f8013"}, - {file = "aiohttp-3.7.4.post0-cp38-cp38-win32.whl", hash = "sha256:515dfef7f869a0feb2afee66b957cc7bbe9ad0cdee45aec7fdc623f4ecd4fb16"}, - {file = "aiohttp-3.7.4.post0-cp38-cp38-win_amd64.whl", hash = "sha256:114b281e4d68302a324dd33abb04778e8557d88947875cbf4e842c2c01a030c5"}, - {file = "aiohttp-3.7.4.post0-cp39-cp39-macosx_10_14_x86_64.whl", hash = "sha256:7b18b97cf8ee5452fa5f4e3af95d01d84d86d32c5e2bfa260cf041749d66360b"}, - {file = "aiohttp-3.7.4.post0-cp39-cp39-manylinux1_i686.whl", hash = "sha256:15492a6368d985b76a2a5fdd2166cddfea5d24e69eefed4630cbaae5c81d89bd"}, - {file = "aiohttp-3.7.4.post0-cp39-cp39-manylinux2014_aarch64.whl", hash = "sha256:bdb230b4943891321e06fc7def63c7aace16095be7d9cf3b1e01be2f10fba439"}, - {file = "aiohttp-3.7.4.post0-cp39-cp39-manylinux2014_i686.whl", hash = "sha256:cffe3ab27871bc3ea47df5d8f7013945712c46a3cc5a95b6bee15887f1675c22"}, - {file = "aiohttp-3.7.4.post0-cp39-cp39-manylinux2014_ppc64le.whl", hash = "sha256:f881853d2643a29e643609da57b96d5f9c9b93f62429dcc1cbb413c7d07f0e1a"}, - {file = "aiohttp-3.7.4.post0-cp39-cp39-manylinux2014_s390x.whl", hash = "sha256:a5ca29ee66f8343ed336816c553e82d6cade48a3ad702b9ffa6125d187e2dedb"}, - {file = "aiohttp-3.7.4.post0-cp39-cp39-manylinux2014_x86_64.whl", hash = "sha256:17c073de315745a1510393a96e680d20af8e67e324f70b42accbd4cb3315c9fb"}, - {file = "aiohttp-3.7.4.post0-cp39-cp39-win32.whl", hash = "sha256:932bb1ea39a54e9ea27fc9232163059a0b8855256f4052e776357ad9add6f1c9"}, - {file = "aiohttp-3.7.4.post0-cp39-cp39-win_amd64.whl", hash = "sha256:02f46fc0e3c5ac58b80d4d56eb0a7c7d97fcef69ace9326289fb9f1955e65cfe"}, - {file = "aiohttp-3.7.4.post0.tar.gz", hash = "sha256:493d3299ebe5f5a7c66b9819eacdcfbbaaf1a8e84911ddffcdc48888497afecf"}, + {file = "aiohttp-3.8.1-cp310-cp310-macosx_10_9_universal2.whl", hash = "sha256:1ed0b6477896559f17b9eaeb6d38e07f7f9ffe40b9f0f9627ae8b9926ae260a8"}, + {file = "aiohttp-3.8.1-cp310-cp310-macosx_10_9_x86_64.whl", hash = "sha256:7dadf3c307b31e0e61689cbf9e06be7a867c563d5a63ce9dca578f956609abf8"}, + {file = "aiohttp-3.8.1-cp310-cp310-macosx_11_0_arm64.whl", hash = "sha256:a79004bb58748f31ae1cbe9fa891054baaa46fb106c2dc7af9f8e3304dc30316"}, + {file = "aiohttp-3.8.1-cp310-cp310-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:12de6add4038df8f72fac606dff775791a60f113a725c960f2bab01d8b8e6b15"}, + {file = "aiohttp-3.8.1-cp310-cp310-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:6f0d5f33feb5f69ddd57a4a4bd3d56c719a141080b445cbf18f238973c5c9923"}, + {file = "aiohttp-3.8.1-cp310-cp310-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:eaba923151d9deea315be1f3e2b31cc39a6d1d2f682f942905951f4e40200922"}, + {file = "aiohttp-3.8.1-cp310-cp310-manylinux_2_5_i686.manylinux1_i686.manylinux_2_12_i686.manylinux2010_i686.whl", hash = "sha256:099ebd2c37ac74cce10a3527d2b49af80243e2a4fa39e7bce41617fbc35fa3c1"}, + {file = "aiohttp-3.8.1-cp310-cp310-manylinux_2_5_x86_64.manylinux1_x86_64.manylinux_2_12_x86_64.manylinux2010_x86_64.whl", hash = "sha256:2e5d962cf7e1d426aa0e528a7e198658cdc8aa4fe87f781d039ad75dcd52c516"}, + {file = "aiohttp-3.8.1-cp310-cp310-musllinux_1_1_aarch64.whl", hash = "sha256:fa0ffcace9b3aa34d205d8130f7873fcfefcb6a4dd3dd705b0dab69af6712642"}, + {file = "aiohttp-3.8.1-cp310-cp310-musllinux_1_1_i686.whl", hash = "sha256:61bfc23df345d8c9716d03717c2ed5e27374e0fe6f659ea64edcd27b4b044cf7"}, + {file = "aiohttp-3.8.1-cp310-cp310-musllinux_1_1_ppc64le.whl", hash = "sha256:31560d268ff62143e92423ef183680b9829b1b482c011713ae941997921eebc8"}, + {file = "aiohttp-3.8.1-cp310-cp310-musllinux_1_1_s390x.whl", hash = "sha256:01d7bdb774a9acc838e6b8f1d114f45303841b89b95984cbb7d80ea41172a9e3"}, + {file = "aiohttp-3.8.1-cp310-cp310-musllinux_1_1_x86_64.whl", hash = "sha256:97ef77eb6b044134c0b3a96e16abcb05ecce892965a2124c566af0fd60f717e2"}, + {file = "aiohttp-3.8.1-cp310-cp310-win32.whl", hash = "sha256:c2aef4703f1f2ddc6df17519885dbfa3514929149d3ff900b73f45998f2532fa"}, + {file = "aiohttp-3.8.1-cp310-cp310-win_amd64.whl", hash = "sha256:713ac174a629d39b7c6a3aa757b337599798da4c1157114a314e4e391cd28e32"}, + {file = "aiohttp-3.8.1-cp36-cp36m-macosx_10_9_x86_64.whl", hash = "sha256:473d93d4450880fe278696549f2e7aed8cd23708c3c1997981464475f32137db"}, + {file = "aiohttp-3.8.1-cp36-cp36m-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:99b5eeae8e019e7aad8af8bb314fb908dd2e028b3cdaad87ec05095394cce632"}, + {file = "aiohttp-3.8.1-cp36-cp36m-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:3af642b43ce56c24d063325dd2cf20ee012d2b9ba4c3c008755a301aaea720ad"}, + {file = "aiohttp-3.8.1-cp36-cp36m-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:c3630c3ef435c0a7c549ba170a0633a56e92629aeed0e707fec832dee313fb7a"}, + {file = "aiohttp-3.8.1-cp36-cp36m-manylinux_2_5_i686.manylinux1_i686.manylinux_2_12_i686.manylinux2010_i686.whl", hash = "sha256:4a4a4e30bf1edcad13fb0804300557aedd07a92cabc74382fdd0ba6ca2661091"}, + {file = "aiohttp-3.8.1-cp36-cp36m-manylinux_2_5_x86_64.manylinux1_x86_64.manylinux_2_12_x86_64.manylinux2010_x86_64.whl", hash = "sha256:6f8b01295e26c68b3a1b90efb7a89029110d3a4139270b24fda961893216c440"}, + {file = "aiohttp-3.8.1-cp36-cp36m-musllinux_1_1_aarch64.whl", hash = "sha256:a25fa703a527158aaf10dafd956f7d42ac6d30ec80e9a70846253dd13e2f067b"}, + {file = "aiohttp-3.8.1-cp36-cp36m-musllinux_1_1_i686.whl", hash = "sha256:5bfde62d1d2641a1f5173b8c8c2d96ceb4854f54a44c23102e2ccc7e02f003ec"}, + {file = "aiohttp-3.8.1-cp36-cp36m-musllinux_1_1_ppc64le.whl", hash = "sha256:51467000f3647d519272392f484126aa716f747859794ac9924a7aafa86cd411"}, + {file = "aiohttp-3.8.1-cp36-cp36m-musllinux_1_1_s390x.whl", hash = "sha256:03a6d5349c9ee8f79ab3ff3694d6ce1cfc3ced1c9d36200cb8f08ba06bd3b782"}, + {file = "aiohttp-3.8.1-cp36-cp36m-musllinux_1_1_x86_64.whl", hash = "sha256:102e487eeb82afac440581e5d7f8f44560b36cf0bdd11abc51a46c1cd88914d4"}, + {file = "aiohttp-3.8.1-cp36-cp36m-win32.whl", hash = "sha256:4aed991a28ea3ce320dc8ce655875e1e00a11bdd29fe9444dd4f88c30d558602"}, + {file = "aiohttp-3.8.1-cp36-cp36m-win_amd64.whl", hash = "sha256:b0e20cddbd676ab8a64c774fefa0ad787cc506afd844de95da56060348021e96"}, + {file = "aiohttp-3.8.1-cp37-cp37m-macosx_10_9_x86_64.whl", hash = "sha256:37951ad2f4a6df6506750a23f7cbabad24c73c65f23f72e95897bb2cecbae676"}, + {file = "aiohttp-3.8.1-cp37-cp37m-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:5c23b1ad869653bc818e972b7a3a79852d0e494e9ab7e1a701a3decc49c20d51"}, + {file = "aiohttp-3.8.1-cp37-cp37m-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:15b09b06dae900777833fe7fc4b4aa426556ce95847a3e8d7548e2d19e34edb8"}, + {file = "aiohttp-3.8.1-cp37-cp37m-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:477c3ea0ba410b2b56b7efb072c36fa91b1e6fc331761798fa3f28bb224830dd"}, + {file = "aiohttp-3.8.1-cp37-cp37m-manylinux_2_5_i686.manylinux1_i686.manylinux_2_12_i686.manylinux2010_i686.whl", hash = "sha256:2f2f69dca064926e79997f45b2f34e202b320fd3782f17a91941f7eb85502ee2"}, + {file = "aiohttp-3.8.1-cp37-cp37m-manylinux_2_5_x86_64.manylinux1_x86_64.manylinux_2_12_x86_64.manylinux2010_x86_64.whl", hash = "sha256:ef9612483cb35171d51d9173647eed5d0069eaa2ee812793a75373447d487aa4"}, + {file = "aiohttp-3.8.1-cp37-cp37m-musllinux_1_1_aarch64.whl", hash = "sha256:6d69f36d445c45cda7b3b26afef2fc34ef5ac0cdc75584a87ef307ee3c8c6d00"}, + {file = "aiohttp-3.8.1-cp37-cp37m-musllinux_1_1_i686.whl", hash = "sha256:55c3d1072704d27401c92339144d199d9de7b52627f724a949fc7d5fc56d8b93"}, + {file = "aiohttp-3.8.1-cp37-cp37m-musllinux_1_1_ppc64le.whl", hash = "sha256:b9d00268fcb9f66fbcc7cd9fe423741d90c75ee029a1d15c09b22d23253c0a44"}, + {file = "aiohttp-3.8.1-cp37-cp37m-musllinux_1_1_s390x.whl", hash = "sha256:07b05cd3305e8a73112103c834e91cd27ce5b4bd07850c4b4dbd1877d3f45be7"}, + {file = "aiohttp-3.8.1-cp37-cp37m-musllinux_1_1_x86_64.whl", hash = "sha256:c34dc4958b232ef6188c4318cb7b2c2d80521c9a56c52449f8f93ab7bc2a8a1c"}, + {file = "aiohttp-3.8.1-cp37-cp37m-win32.whl", hash = "sha256:d2f9b69293c33aaa53d923032fe227feac867f81682f002ce33ffae978f0a9a9"}, + {file = "aiohttp-3.8.1-cp37-cp37m-win_amd64.whl", hash = "sha256:6ae828d3a003f03ae31915c31fa684b9890ea44c9c989056fea96e3d12a9fa17"}, + {file = "aiohttp-3.8.1-cp38-cp38-macosx_10_9_universal2.whl", hash = "sha256:0c7ebbbde809ff4e970824b2b6cb7e4222be6b95a296e46c03cf050878fc1785"}, + {file = "aiohttp-3.8.1-cp38-cp38-macosx_10_9_x86_64.whl", hash = "sha256:8b7ef7cbd4fec9a1e811a5de813311ed4f7ac7d93e0fda233c9b3e1428f7dd7b"}, + {file = "aiohttp-3.8.1-cp38-cp38-macosx_11_0_arm64.whl", hash = "sha256:c3d6a4d0619e09dcd61021debf7059955c2004fa29f48788a3dfaf9c9901a7cd"}, + {file = "aiohttp-3.8.1-cp38-cp38-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:718626a174e7e467f0558954f94af117b7d4695d48eb980146016afa4b580b2e"}, + {file = "aiohttp-3.8.1-cp38-cp38-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:589c72667a5febd36f1315aa6e5f56dd4aa4862df295cb51c769d16142ddd7cd"}, + {file = "aiohttp-3.8.1-cp38-cp38-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:2ed076098b171573161eb146afcb9129b5ff63308960aeca4b676d9d3c35e700"}, + {file = "aiohttp-3.8.1-cp38-cp38-manylinux_2_5_i686.manylinux1_i686.manylinux_2_12_i686.manylinux2010_i686.whl", hash = "sha256:086f92daf51a032d062ec5f58af5ca6a44d082c35299c96376a41cbb33034675"}, + {file = "aiohttp-3.8.1-cp38-cp38-manylinux_2_5_x86_64.manylinux1_x86_64.manylinux_2_12_x86_64.manylinux2010_x86_64.whl", hash = "sha256:11691cf4dc5b94236ccc609b70fec991234e7ef8d4c02dd0c9668d1e486f5abf"}, + {file = "aiohttp-3.8.1-cp38-cp38-musllinux_1_1_aarch64.whl", hash = "sha256:31d1e1c0dbf19ebccbfd62eff461518dcb1e307b195e93bba60c965a4dcf1ba0"}, + {file = "aiohttp-3.8.1-cp38-cp38-musllinux_1_1_i686.whl", hash = "sha256:11a67c0d562e07067c4e86bffc1553f2cf5b664d6111c894671b2b8712f3aba5"}, + {file = "aiohttp-3.8.1-cp38-cp38-musllinux_1_1_ppc64le.whl", hash = "sha256:bb01ba6b0d3f6c68b89fce7305080145d4877ad3acaed424bae4d4ee75faa950"}, + {file = "aiohttp-3.8.1-cp38-cp38-musllinux_1_1_s390x.whl", hash = "sha256:44db35a9e15d6fe5c40d74952e803b1d96e964f683b5a78c3cc64eb177878155"}, + {file = "aiohttp-3.8.1-cp38-cp38-musllinux_1_1_x86_64.whl", hash = "sha256:844a9b460871ee0a0b0b68a64890dae9c415e513db0f4a7e3cab41a0f2fedf33"}, + {file = "aiohttp-3.8.1-cp38-cp38-win32.whl", hash = "sha256:7d08744e9bae2ca9c382581f7dce1273fe3c9bae94ff572c3626e8da5b193c6a"}, + {file = "aiohttp-3.8.1-cp38-cp38-win_amd64.whl", hash = "sha256:04d48b8ce6ab3cf2097b1855e1505181bdd05586ca275f2505514a6e274e8e75"}, + {file = "aiohttp-3.8.1-cp39-cp39-macosx_10_9_universal2.whl", hash = "sha256:f5315a2eb0239185af1bddb1abf472d877fede3cc8d143c6cddad37678293237"}, + {file = "aiohttp-3.8.1-cp39-cp39-macosx_10_9_x86_64.whl", hash = "sha256:a996d01ca39b8dfe77440f3cd600825d05841088fd6bc0144cc6c2ec14cc5f74"}, + {file = "aiohttp-3.8.1-cp39-cp39-macosx_11_0_arm64.whl", hash = "sha256:13487abd2f761d4be7c8ff9080de2671e53fff69711d46de703c310c4c9317ca"}, + {file = "aiohttp-3.8.1-cp39-cp39-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:ea302f34477fda3f85560a06d9ebdc7fa41e82420e892fc50b577e35fc6a50b2"}, + {file = "aiohttp-3.8.1-cp39-cp39-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:a2f635ce61a89c5732537a7896b6319a8fcfa23ba09bec36e1b1ac0ab31270d2"}, + {file = "aiohttp-3.8.1-cp39-cp39-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:e999f2d0e12eea01caeecb17b653f3713d758f6dcc770417cf29ef08d3931421"}, + {file = "aiohttp-3.8.1-cp39-cp39-manylinux_2_5_i686.manylinux1_i686.manylinux_2_12_i686.manylinux2010_i686.whl", hash = "sha256:0770e2806a30e744b4e21c9d73b7bee18a1cfa3c47991ee2e5a65b887c49d5cf"}, + {file = "aiohttp-3.8.1-cp39-cp39-manylinux_2_5_x86_64.manylinux1_x86_64.manylinux_2_12_x86_64.manylinux2010_x86_64.whl", hash = "sha256:d15367ce87c8e9e09b0f989bfd72dc641bcd04ba091c68cd305312d00962addd"}, + {file = "aiohttp-3.8.1-cp39-cp39-musllinux_1_1_aarch64.whl", hash = "sha256:6c7cefb4b0640703eb1069835c02486669312bf2f12b48a748e0a7756d0de33d"}, + {file = "aiohttp-3.8.1-cp39-cp39-musllinux_1_1_i686.whl", hash = "sha256:71927042ed6365a09a98a6377501af5c9f0a4d38083652bcd2281a06a5976724"}, + {file = "aiohttp-3.8.1-cp39-cp39-musllinux_1_1_ppc64le.whl", hash = "sha256:28d490af82bc6b7ce53ff31337a18a10498303fe66f701ab65ef27e143c3b0ef"}, + {file = "aiohttp-3.8.1-cp39-cp39-musllinux_1_1_s390x.whl", hash = "sha256:b6613280ccedf24354406caf785db748bebbddcf31408b20c0b48cb86af76866"}, + {file = "aiohttp-3.8.1-cp39-cp39-musllinux_1_1_x86_64.whl", hash = "sha256:81e3d8c34c623ca4e36c46524a3530e99c0bc95ed068fd6e9b55cb721d408fb2"}, + {file = "aiohttp-3.8.1-cp39-cp39-win32.whl", hash = "sha256:7187a76598bdb895af0adbd2fb7474d7f6025d170bc0a1130242da817ce9e7d1"}, + {file = "aiohttp-3.8.1-cp39-cp39-win_amd64.whl", hash = "sha256:1c182cb873bc91b411e184dab7a2b664d4fea2743df0e4d57402f7f3fa644bac"}, + {file = "aiohttp-3.8.1.tar.gz", hash = "sha256:fc5471e1a54de15ef71c1bc6ebe80d4dc681ea600e68bfd1cbce40427f0b7578"}, +] +aiosignal = [ + {file = "aiosignal-1.2.0-py3-none-any.whl", hash = "sha256:26e62109036cd181df6e6ad646f91f0dcfd05fe16d0cb924138ff2ab75d64e3a"}, + {file = "aiosignal-1.2.0.tar.gz", hash = "sha256:78ed67db6c7b7ced4f98e495e572106d5c432a93e1ddd1bf475e1dc05f5b7df2"}, ] apscheduler = [ - {file = "APScheduler-3.8.0-py2.py3-none-any.whl", hash = "sha256:82d6d21b2f0343510d07bb35618333a794653439a265635555b12935647b460a"}, - {file = "APScheduler-3.8.0.tar.gz", hash = "sha256:793b2d37c52ece53e34626619e6142e99b20b59a12155f39e1e6932e324f079d"}, + {file = "APScheduler-3.8.1-py2.py3-none-any.whl", hash = "sha256:c22cb14b411a31435eb2c530dfbbec948ac63015b517087c7978adb61b574865"}, + {file = "APScheduler-3.8.1.tar.gz", hash = "sha256:5cf344ebcfbdaa48ae178c029c055cec7bc7a4a47c21e315e4d1f08bd35f2355"}, ] asn1crypto = [ {file = "asn1crypto-1.4.0-py2.py3-none-any.whl", hash = "sha256:4bcdf33c861c7d40bdcd74d8e4dd7661aac320fcdf40b9a3f95b4ee12fde2fa8"}, {file = "asn1crypto-1.4.0.tar.gz", hash = "sha256:f4f6e119474e58e04a2b1af817eb585b4fd72bdd89b998624712b5c99be7641c"}, ] async-timeout = [ - {file = "async-timeout-3.0.1.tar.gz", hash = "sha256:0c3c816a028d47f659d6ff5c745cb2acf1f966da1fe5c19c77a70282b25f4c5f"}, - {file = "async_timeout-3.0.1-py3-none-any.whl", hash = "sha256:4291ca197d287d274d0b6cb5d6f8f8f82d434ed288f962539ff18cc9012f9ea3"}, + {file = "async-timeout-4.0.1.tar.gz", hash = "sha256:b930cb161a39042f9222f6efb7301399c87eeab394727ec5437924a36d6eef51"}, + {file = "async_timeout-4.0.1-py3-none-any.whl", hash = "sha256:a22c0b311af23337eb05fcf05a8b51c3ea53729d46fb5460af62bee033cec690"}, ] atomicwrites = [ {file = "atomicwrites-1.4.0-py2.py3-none-any.whl", hash = "sha256:6d1784dea7c0c8d4a5172b6c620f40b6e4cbfdf96d783691f2e1302a7b88e197"}, @@ -895,12 +989,12 @@ attrs = [ {file = "attrs-21.2.0.tar.gz", hash = "sha256:ef6aaac3ca6cd92904cdd0d83f629a15f18053ec84e6432106f7a4d04ae4f5fb"}, ] "backports.entry-points-selectable" = [ - {file = "backports.entry_points_selectable-1.1.0-py2.py3-none-any.whl", hash = "sha256:a6d9a871cde5e15b4c4a53e3d43ba890cc6861ec1332c9c2428c92f977192acc"}, - {file = "backports.entry_points_selectable-1.1.0.tar.gz", hash = "sha256:988468260ec1c196dab6ae1149260e2f5472c9110334e5d51adcb77867361f6a"}, + {file = "backports.entry_points_selectable-1.1.1-py2.py3-none-any.whl", hash = "sha256:7fceed9532a7aa2bd888654a7314f864a3c16a4e710b34a58cfc0f08114c663b"}, + {file = "backports.entry_points_selectable-1.1.1.tar.gz", hash = "sha256:914b21a479fde881635f7af5adc7f6e38d6b274be32269070c53b698c60d5386"}, ] black = [ - {file = "black-21.9b0-py3-none-any.whl", hash = "sha256:380f1b5da05e5a1429225676655dddb96f5ae8c75bdf91e53d798871b902a115"}, - {file = "black-21.9b0.tar.gz", hash = "sha256:7de4cfc7eb6b710de325712d40125689101d21d25283eed7e9998722cf10eb91"}, + {file = "black-21.11b1-py3-none-any.whl", hash = "sha256:802c6c30b637b28645b7fde282ed2569c0cd777dbe493a41b6a03c1d903f99ac"}, + {file = "black-21.11b1.tar.gz", hash = "sha256:a042adbb18b3262faad5aff4e834ff186bb893f95ba3a8013f09de1e5569def2"}, ] certifi = [ {file = "certifi-2021.10.8-py2.py3-none-any.whl", hash = "sha256:d62a0163eb4c2344ac042ab2bdf75399a71a2d8c7d47eac2e2ee91b9d6339569"}, @@ -962,13 +1056,9 @@ cfgv = [ {file = "cfgv-3.3.1-py2.py3-none-any.whl", hash = "sha256:c6a0883f3917a037485059700b9e75da2464e6c27051014ad85ba6aaa5884426"}, {file = "cfgv-3.3.1.tar.gz", hash = "sha256:f5a830efb9ce7a445376bb66ec94c638a9787422f96264c98edc6bdeed8ab736"}, ] -chardet = [ - {file = "chardet-4.0.0-py2.py3-none-any.whl", hash = "sha256:f864054d66fd9118f2e67044ac8981a54775ec5b67aed0441892edb553d21da5"}, - {file = "chardet-4.0.0.tar.gz", hash = "sha256:0d6f53a15db4120f2b08c94f11e7d93d2c911ee118b6b30a04ec3ee8310179fa"}, -] charset-normalizer = [ - {file = "charset-normalizer-2.0.7.tar.gz", hash = "sha256:e019de665e2bcf9c2b64e2e5aa025fa991da8720daa3c1138cadd2fd1856aed0"}, - {file = "charset_normalizer-2.0.7-py3-none-any.whl", hash = "sha256:f7af805c321bfa1ce6714c51f254e0d5bb5e5834039bc17db7ebe3a4cec9492b"}, + {file = "charset-normalizer-2.0.9.tar.gz", hash = "sha256:b0b883e8e874edfdece9c28f314e3dd5badf067342e42fb162203335ae61aa2c"}, + {file = "charset_normalizer-2.0.9-py3-none-any.whl", hash = "sha256:1eecaa09422db5be9e29d7fc65664e6c33bd06f9ced7838578ba40d58bdf3721"}, ] click = [ {file = "click-8.0.3-py3-none-any.whl", hash = "sha256:353f466495adaeb40b6b5f592f9f91cb22372351c84caeb068132442a4518ef3"}, @@ -1012,8 +1102,82 @@ eight = [ {file = "eight-1.0.1.tar.gz", hash = "sha256:6465970d3c12a1d759151690629b10b50ffb124c9b62e912999ee9233378a562"}, ] filelock = [ - {file = "filelock-3.3.1-py3-none-any.whl", hash = "sha256:2b5eb3589e7fdda14599e7eb1a50e09b4cc14f34ed98b8ba56d33bfaafcbef2f"}, - {file = "filelock-3.3.1.tar.gz", hash = "sha256:34a9f35f95c441e7b38209775d6e0337f9a3759f3565f6c5798f19618527c76f"}, + {file = "filelock-3.4.0-py3-none-any.whl", hash = "sha256:2e139a228bcf56dd8b2274a65174d005c4a6b68540ee0bdbb92c76f43f29f7e8"}, + {file = "filelock-3.4.0.tar.gz", hash = "sha256:93d512b32a23baf4cac44ffd72ccf70732aeff7b8050fcaf6d3ec406d954baf4"}, +] +frozenlist = [ + {file = "frozenlist-1.2.0-cp310-cp310-macosx_10_9_universal2.whl", hash = "sha256:977a1438d0e0d96573fd679d291a1542097ea9f4918a8b6494b06610dfeefbf9"}, + {file = "frozenlist-1.2.0-cp310-cp310-macosx_10_9_x86_64.whl", hash = "sha256:a8d86547a5e98d9edd47c432f7a14b0c5592624b496ae9880fb6332f34af1edc"}, + {file = "frozenlist-1.2.0-cp310-cp310-macosx_11_0_arm64.whl", hash = "sha256:181754275d5d32487431a0a29add4f897968b7157204bc1eaaf0a0ce80c5ba7d"}, + {file = "frozenlist-1.2.0-cp310-cp310-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:5df31bb2b974f379d230a25943d9bf0d3bc666b4b0807394b131a28fca2b0e5f"}, + {file = "frozenlist-1.2.0-cp310-cp310-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:4766632cd8a68e4f10f156a12c9acd7b1609941525569dd3636d859d79279ed3"}, + {file = "frozenlist-1.2.0-cp310-cp310-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:16eef427c51cb1203a7c0ab59d1b8abccaba9a4f58c4bfca6ed278fc896dc193"}, + {file = "frozenlist-1.2.0-cp310-cp310-manylinux_2_5_i686.manylinux1_i686.manylinux_2_12_i686.manylinux2010_i686.whl", hash = "sha256:01d79515ed5aa3d699b05f6bdcf1fe9087d61d6b53882aa599a10853f0479c6c"}, + {file = "frozenlist-1.2.0-cp310-cp310-manylinux_2_5_x86_64.manylinux1_x86_64.manylinux_2_12_x86_64.manylinux2010_x86_64.whl", hash = "sha256:28e164722ea0df0cf6d48c4d5bdf3d19e87aaa6dfb39b0ba91153f224b912020"}, + {file = "frozenlist-1.2.0-cp310-cp310-musllinux_1_1_aarch64.whl", hash = "sha256:e63ad0beef6ece06475d29f47d1f2f29727805376e09850ebf64f90777962792"}, + {file = "frozenlist-1.2.0-cp310-cp310-musllinux_1_1_i686.whl", hash = "sha256:41de4db9b9501679cf7cddc16d07ac0f10ef7eb58c525a1c8cbff43022bddca4"}, + {file = "frozenlist-1.2.0-cp310-cp310-musllinux_1_1_ppc64le.whl", hash = "sha256:c6a9d84ee6427b65a81fc24e6ef589cb794009f5ca4150151251c062773e7ed2"}, + {file = "frozenlist-1.2.0-cp310-cp310-musllinux_1_1_s390x.whl", hash = "sha256:f5f3b2942c3b8b9bfe76b408bbaba3d3bb305ee3693e8b1d631fe0a0d4f93673"}, + {file = "frozenlist-1.2.0-cp310-cp310-musllinux_1_1_x86_64.whl", hash = "sha256:c98d3c04701773ad60d9545cd96df94d955329efc7743fdb96422c4b669c633b"}, + {file = "frozenlist-1.2.0-cp310-cp310-win32.whl", hash = "sha256:72cfbeab7a920ea9e74b19aa0afe3b4ad9c89471e3badc985d08756efa9b813b"}, + {file = "frozenlist-1.2.0-cp310-cp310-win_amd64.whl", hash = "sha256:11ff401951b5ac8c0701a804f503d72c048173208490c54ebb8d7bb7c07a6d00"}, + {file = "frozenlist-1.2.0-cp36-cp36m-macosx_10_9_x86_64.whl", hash = "sha256:b46f997d5ed6d222a863b02cdc9c299101ee27974d9bbb2fd1b3c8441311c408"}, + {file = "frozenlist-1.2.0-cp36-cp36m-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:351686ca020d1bcd238596b1fa5c8efcbc21bffda9d0efe237aaa60348421e2a"}, + {file = "frozenlist-1.2.0-cp36-cp36m-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:bfbaa08cf1452acad9cb1c1d7b89394a41e712f88df522cea1a0f296b57782a0"}, + {file = "frozenlist-1.2.0-cp36-cp36m-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:b2ae2f5e9fa10805fb1c9adbfefaaecedd9e31849434be462c3960a0139ed729"}, + {file = "frozenlist-1.2.0-cp36-cp36m-manylinux_2_5_i686.manylinux1_i686.manylinux_2_12_i686.manylinux2010_i686.whl", hash = "sha256:6790b8d96bbb74b7a6f4594b6f131bd23056c25f2aa5d816bd177d95245a30e3"}, + {file = "frozenlist-1.2.0-cp36-cp36m-manylinux_2_5_x86_64.manylinux1_x86_64.manylinux_2_12_x86_64.manylinux2010_x86_64.whl", hash = "sha256:41f62468af1bd4e4b42b5508a3fe8cc46a693f0cdd0ca2f443f51f207893d837"}, + {file = "frozenlist-1.2.0-cp36-cp36m-musllinux_1_1_aarch64.whl", hash = "sha256:ec6cf345771cdb00791d271af9a0a6fbfc2b6dd44cb753f1eeaa256e21622adb"}, + {file = "frozenlist-1.2.0-cp36-cp36m-musllinux_1_1_i686.whl", hash = "sha256:14a5cef795ae3e28fb504b73e797c1800e9249f950e1c964bb6bdc8d77871161"}, + {file = "frozenlist-1.2.0-cp36-cp36m-musllinux_1_1_ppc64le.whl", hash = "sha256:8b54cdd2fda15467b9b0bfa78cee2ddf6dbb4585ef23a16e14926f4b076dfae4"}, + {file = "frozenlist-1.2.0-cp36-cp36m-musllinux_1_1_s390x.whl", hash = "sha256:f025f1d6825725b09c0038775acab9ae94264453a696cc797ce20c0769a7b367"}, + {file = "frozenlist-1.2.0-cp36-cp36m-musllinux_1_1_x86_64.whl", hash = "sha256:84e97f59211b5b9083a2e7a45abf91cfb441369e8bb6d1f5287382c1c526def3"}, + {file = "frozenlist-1.2.0-cp36-cp36m-win32.whl", hash = "sha256:c5328ed53fdb0a73c8a50105306a3bc013e5ca36cca714ec4f7bd31d38d8a97f"}, + {file = "frozenlist-1.2.0-cp36-cp36m-win_amd64.whl", hash = "sha256:9ade70aea559ca98f4b1b1e5650c45678052e76a8ab2f76d90f2ac64180215a2"}, + {file = "frozenlist-1.2.0-cp37-cp37m-macosx_10_9_x86_64.whl", hash = "sha256:a0d3ffa8772464441b52489b985d46001e2853a3b082c655ec5fad9fb6a3d618"}, + {file = "frozenlist-1.2.0-cp37-cp37m-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:3457f8cf86deb6ce1ba67e120f1b0128fcba1332a180722756597253c465fc1d"}, + {file = "frozenlist-1.2.0-cp37-cp37m-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:5a72eecf37eface331636951249d878750db84034927c997d47f7f78a573b72b"}, + {file = "frozenlist-1.2.0-cp37-cp37m-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:acc4614e8d1feb9f46dd829a8e771b8f5c4b1051365d02efb27a3229048ade8a"}, + {file = "frozenlist-1.2.0-cp37-cp37m-manylinux_2_5_i686.manylinux1_i686.manylinux_2_12_i686.manylinux2010_i686.whl", hash = "sha256:87521e32e18a2223311afc2492ef2d99946337da0779ddcda77b82ee7319df59"}, + {file = "frozenlist-1.2.0-cp37-cp37m-manylinux_2_5_x86_64.manylinux1_x86_64.manylinux_2_12_x86_64.manylinux2010_x86_64.whl", hash = "sha256:8b4c7665a17c3a5430edb663e4ad4e1ad457614d1b2f2b7f87052e2ef4fa45ca"}, + {file = "frozenlist-1.2.0-cp37-cp37m-musllinux_1_1_aarch64.whl", hash = "sha256:ed58803563a8c87cf4c0771366cf0ad1aa265b6b0ae54cbbb53013480c7ad74d"}, + {file = "frozenlist-1.2.0-cp37-cp37m-musllinux_1_1_i686.whl", hash = "sha256:aa44c4740b4e23fcfa259e9dd52315d2b1770064cde9507457e4c4a65a04c397"}, + {file = "frozenlist-1.2.0-cp37-cp37m-musllinux_1_1_ppc64le.whl", hash = "sha256:2de5b931701257d50771a032bba4e448ff958076380b049fd36ed8738fdb375b"}, + {file = "frozenlist-1.2.0-cp37-cp37m-musllinux_1_1_s390x.whl", hash = "sha256:6e105013fa84623c057a4381dc8ea0361f4d682c11f3816cc80f49a1f3bc17c6"}, + {file = "frozenlist-1.2.0-cp37-cp37m-musllinux_1_1_x86_64.whl", hash = "sha256:705c184b77565955a99dc360f359e8249580c6b7eaa4dc0227caa861ef46b27a"}, + {file = "frozenlist-1.2.0-cp37-cp37m-win32.whl", hash = "sha256:a37594ad6356e50073fe4f60aa4187b97d15329f2138124d252a5a19c8553ea4"}, + {file = "frozenlist-1.2.0-cp37-cp37m-win_amd64.whl", hash = "sha256:25b358aaa7dba5891b05968dd539f5856d69f522b6de0bf34e61f133e077c1a4"}, + {file = "frozenlist-1.2.0-cp38-cp38-macosx_10_9_universal2.whl", hash = "sha256:af2a51c8a381d76eabb76f228f565ed4c3701441ecec101dd18be70ebd483cfd"}, + {file = "frozenlist-1.2.0-cp38-cp38-macosx_10_9_x86_64.whl", hash = "sha256:82d22f6e6f2916e837c91c860140ef9947e31194c82aaeda843d6551cec92f19"}, + {file = "frozenlist-1.2.0-cp38-cp38-macosx_11_0_arm64.whl", hash = "sha256:1cfe6fef507f8bac40f009c85c7eddfed88c1c0d38c75e72fe10476cef94e10f"}, + {file = "frozenlist-1.2.0-cp38-cp38-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:26f602e380a5132880fa245c92030abb0fc6ff34e0c5500600366cedc6adb06a"}, + {file = "frozenlist-1.2.0-cp38-cp38-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:4ad065b2ebd09f32511ff2be35c5dfafee6192978b5a1e9d279a5c6e121e3b03"}, + {file = "frozenlist-1.2.0-cp38-cp38-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:bc93f5f62df3bdc1f677066327fc81f92b83644852a31c6aa9b32c2dde86ea7d"}, + {file = "frozenlist-1.2.0-cp38-cp38-manylinux_2_5_i686.manylinux1_i686.manylinux_2_12_i686.manylinux2010_i686.whl", hash = "sha256:89fdfc84c6bf0bff2ff3170bb34ecba8a6911b260d318d377171429c4be18c73"}, + {file = "frozenlist-1.2.0-cp38-cp38-manylinux_2_5_x86_64.manylinux1_x86_64.manylinux_2_12_x86_64.manylinux2010_x86_64.whl", hash = "sha256:47b2848e464883d0bbdcd9493c67443e5e695a84694efff0476f9059b4cb6257"}, + {file = "frozenlist-1.2.0-cp38-cp38-musllinux_1_1_aarch64.whl", hash = "sha256:4f52d0732e56906f8ddea4bd856192984650282424049c956857fed43697ea43"}, + {file = "frozenlist-1.2.0-cp38-cp38-musllinux_1_1_i686.whl", hash = "sha256:16ef7dd5b7d17495404a2e7a49bac1bc13d6d20c16d11f4133c757dd94c4144c"}, + {file = "frozenlist-1.2.0-cp38-cp38-musllinux_1_1_ppc64le.whl", hash = "sha256:1cf63243bc5f5c19762943b0aa9e0d3fb3723d0c514d820a18a9b9a5ef864315"}, + {file = "frozenlist-1.2.0-cp38-cp38-musllinux_1_1_s390x.whl", hash = "sha256:54a1e09ab7a69f843cd28fefd2bcaf23edb9e3a8d7680032c8968b8ac934587d"}, + {file = "frozenlist-1.2.0-cp38-cp38-musllinux_1_1_x86_64.whl", hash = "sha256:954b154a4533ef28bd3e83ffdf4eadf39deeda9e38fb8feaf066d6069885e034"}, + {file = "frozenlist-1.2.0-cp38-cp38-win32.whl", hash = "sha256:cb3957c39668d10e2b486acc85f94153520a23263b6401e8f59422ef65b9520d"}, + {file = "frozenlist-1.2.0-cp38-cp38-win_amd64.whl", hash = "sha256:0a7c7cce70e41bc13d7d50f0e5dd175f14a4f1837a8549b0936ed0cbe6170bf9"}, + {file = "frozenlist-1.2.0-cp39-cp39-macosx_10_9_universal2.whl", hash = "sha256:4c457220468d734e3077580a3642b7f682f5fd9507f17ddf1029452450912cdc"}, + {file = "frozenlist-1.2.0-cp39-cp39-macosx_10_9_x86_64.whl", hash = "sha256:e74f8b4d8677ebb4015ac01fcaf05f34e8a1f22775db1f304f497f2f88fdc697"}, + {file = "frozenlist-1.2.0-cp39-cp39-macosx_11_0_arm64.whl", hash = "sha256:fbd4844ff111449f3bbe20ba24fbb906b5b1c2384d0f3287c9f7da2354ce6d23"}, + {file = "frozenlist-1.2.0-cp39-cp39-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:f0081a623c886197ff8de9e635528fd7e6a387dccef432149e25c13946cb0cd0"}, + {file = "frozenlist-1.2.0-cp39-cp39-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:9b6e21e5770df2dea06cb7b6323fbc008b13c4a4e3b52cb54685276479ee7676"}, + {file = "frozenlist-1.2.0-cp39-cp39-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:406aeb340613b4b559db78d86864485f68919b7141dec82aba24d1477fd2976f"}, + {file = "frozenlist-1.2.0-cp39-cp39-manylinux_2_5_i686.manylinux1_i686.manylinux_2_12_i686.manylinux2010_i686.whl", hash = "sha256:878ebe074839d649a1cdb03a61077d05760624f36d196884a5cafb12290e187b"}, + {file = "frozenlist-1.2.0-cp39-cp39-manylinux_2_5_x86_64.manylinux1_x86_64.manylinux_2_12_x86_64.manylinux2010_x86_64.whl", hash = "sha256:1fef737fd1388f9b93bba8808c5f63058113c10f4e3c0763ced68431773f72f9"}, + {file = "frozenlist-1.2.0-cp39-cp39-musllinux_1_1_aarch64.whl", hash = "sha256:4a495c3d513573b0b3f935bfa887a85d9ae09f0627cf47cad17d0cc9b9ba5c38"}, + {file = "frozenlist-1.2.0-cp39-cp39-musllinux_1_1_i686.whl", hash = "sha256:e7d0dd3e727c70c2680f5f09a0775525229809f1a35d8552b92ff10b2b14f2c2"}, + {file = "frozenlist-1.2.0-cp39-cp39-musllinux_1_1_ppc64le.whl", hash = "sha256:66a518731a21a55b7d3e087b430f1956a36793acc15912e2878431c7aec54210"}, + {file = "frozenlist-1.2.0-cp39-cp39-musllinux_1_1_s390x.whl", hash = "sha256:94728f97ddf603d23c8c3dd5cae2644fa12d33116e69f49b1644a71bb77b89ae"}, + {file = "frozenlist-1.2.0-cp39-cp39-musllinux_1_1_x86_64.whl", hash = "sha256:c1e8e9033d34c2c9e186e58279879d78c94dd365068a3607af33f2bc99357a53"}, + {file = "frozenlist-1.2.0-cp39-cp39-win32.whl", hash = "sha256:83334e84a290a158c0c4cc4d22e8c7cfe0bba5b76d37f1c2509dabd22acafe15"}, + {file = "frozenlist-1.2.0-cp39-cp39-win_amd64.whl", hash = "sha256:735f386ec522e384f511614c01d2ef9cf799f051353876b4c6fb93ef67a6d1ee"}, + {file = "frozenlist-1.2.0.tar.gz", hash = "sha256:68201be60ac56aff972dc18085800b6ee07973c49103a8aba669dee3d71079de"}, ] future = [ {file = "future-0.18.2.tar.gz", hash = "sha256:b1bead90b70cf6ec3f0710ae53a525360fa360d306a86583adc6bf83a4db537d"}, @@ -1067,70 +1231,82 @@ grequests = [ {file = "grequests-0.6.0.tar.gz", hash = "sha256:7dec890c6668e6755a1ea968565535867956639301268394d24df67b478df666"}, ] identify = [ - {file = "identify-2.3.0-py2.py3-none-any.whl", hash = "sha256:d1e82c83d063571bb88087676f81261a4eae913c492dafde184067c584bc7c05"}, - {file = "identify-2.3.0.tar.gz", hash = "sha256:fd08c97f23ceee72784081f1ce5125c8f53a02d3f2716dde79a6ab8f1039fea5"}, + {file = "identify-2.4.0-py2.py3-none-any.whl", hash = "sha256:eba31ca80258de6bb51453084bff4a923187cd2193b9c13710f2516ab30732cc"}, + {file = "identify-2.4.0.tar.gz", hash = "sha256:a33ae873287e81651c7800ca309dc1f84679b763c9c8b30680e16fbfa82f0107"}, ] idna = [ - {file = "idna-2.10-py2.py3-none-any.whl", hash = "sha256:b97d804b1e9b523befed77c48dacec60e6dcb0b5391d57af6a65a312a90648c0"}, - {file = "idna-2.10.tar.gz", hash = "sha256:b307872f855b18632ce0c21c5e45be78c0ea7ae4c15c828c20788b26921eb3f6"}, + {file = "idna-3.3-py3-none-any.whl", hash = "sha256:84d9dd047ffa80596e0f246e2eab0b391788b0503584e8945f2368256d2735ff"}, + {file = "idna-3.3.tar.gz", hash = "sha256:9d643ff0a55b762d5cdb124b8eaa99c66322e2157b69160bc32796e824360e6d"}, ] iniconfig = [ {file = "iniconfig-1.1.1-py2.py3-none-any.whl", hash = "sha256:011e24c64b7f47f6ebd835bb12a743f2fbe9a26d4cecaa7f53bc4f35ee9da8b3"}, {file = "iniconfig-1.1.1.tar.gz", hash = "sha256:bc3af051d7d14b2ee5ef9969666def0cd1a000e121eaea580d4a313df4b37f32"}, ] jinja2 = [ - {file = "Jinja2-3.0.2-py3-none-any.whl", hash = "sha256:8569982d3f0889eed11dd620c706d39b60c36d6d25843961f33f77fb6bc6b20c"}, - {file = "Jinja2-3.0.2.tar.gz", hash = "sha256:827a0e32839ab1600d4eb1c4c33ec5a8edfbc5cb42dafa13b81f182f97784b45"}, + {file = "Jinja2-3.0.3-py3-none-any.whl", hash = "sha256:077ce6014f7b40d03b47d1f1ca4b0fc8328a692bd284016f806ed0eaca390ad8"}, + {file = "Jinja2-3.0.3.tar.gz", hash = "sha256:611bb273cd68f3b993fabdc4064fc858c5b47a973cb5aa7999ec1ba405c87cd7"}, ] lxml = [ - {file = "lxml-4.6.3-cp27-cp27m-macosx_10_9_x86_64.whl", hash = "sha256:df7c53783a46febb0e70f6b05df2ba104610f2fb0d27023409734a3ecbb78fb2"}, - {file = "lxml-4.6.3-cp27-cp27m-manylinux1_i686.whl", hash = "sha256:1b7584d421d254ab86d4f0b13ec662a9014397678a7c4265a02a6d7c2b18a75f"}, - {file = "lxml-4.6.3-cp27-cp27m-manylinux1_x86_64.whl", hash = "sha256:079f3ae844f38982d156efce585bc540c16a926d4436712cf4baee0cce487a3d"}, - {file = "lxml-4.6.3-cp27-cp27m-win32.whl", hash = "sha256:bc4313cbeb0e7a416a488d72f9680fffffc645f8a838bd2193809881c67dd106"}, - {file = "lxml-4.6.3-cp27-cp27m-win_amd64.whl", hash = "sha256:8157dadbb09a34a6bd95a50690595e1fa0af1a99445e2744110e3dca7831c4ee"}, - {file = "lxml-4.6.3-cp27-cp27mu-manylinux1_i686.whl", hash = "sha256:7728e05c35412ba36d3e9795ae8995e3c86958179c9770e65558ec3fdfd3724f"}, - {file = "lxml-4.6.3-cp27-cp27mu-manylinux1_x86_64.whl", hash = "sha256:4bff24dfeea62f2e56f5bab929b4428ae6caba2d1eea0c2d6eb618e30a71e6d4"}, - {file = "lxml-4.6.3-cp310-cp310-manylinux_2_12_i686.manylinux2010_i686.manylinux_2_24_i686.whl", hash = "sha256:64812391546a18896adaa86c77c59a4998f33c24788cadc35789e55b727a37f4"}, - {file = "lxml-4.6.3-cp310-cp310-manylinux_2_17_x86_64.manylinux2014_x86_64.manylinux_2_24_x86_64.whl", hash = "sha256:c1a40c06fd5ba37ad39caa0b3144eb3772e813b5fb5b084198a985431c2f1e8d"}, - {file = "lxml-4.6.3-cp35-cp35m-manylinux1_i686.whl", hash = "sha256:74f7d8d439b18fa4c385f3f5dfd11144bb87c1da034a466c5b5577d23a1d9b51"}, - {file = "lxml-4.6.3-cp35-cp35m-manylinux1_x86_64.whl", hash = "sha256:f90ba11136bfdd25cae3951af8da2e95121c9b9b93727b1b896e3fa105b2f586"}, - {file = "lxml-4.6.3-cp35-cp35m-manylinux2010_i686.whl", hash = "sha256:4c61b3a0db43a1607d6264166b230438f85bfed02e8cff20c22e564d0faff354"}, - {file = "lxml-4.6.3-cp35-cp35m-manylinux2014_x86_64.whl", hash = "sha256:5c8c163396cc0df3fd151b927e74f6e4acd67160d6c33304e805b84293351d16"}, - {file = "lxml-4.6.3-cp35-cp35m-win32.whl", hash = "sha256:f2380a6376dfa090227b663f9678150ef27543483055cc327555fb592c5967e2"}, - {file = "lxml-4.6.3-cp35-cp35m-win_amd64.whl", hash = "sha256:c4f05c5a7c49d2fb70223d0d5bcfbe474cf928310ac9fa6a7c6dddc831d0b1d4"}, - {file = "lxml-4.6.3-cp36-cp36m-macosx_10_9_x86_64.whl", hash = "sha256:d2e35d7bf1c1ac8c538f88d26b396e73dd81440d59c1ef8522e1ea77b345ede4"}, - {file = "lxml-4.6.3-cp36-cp36m-manylinux1_i686.whl", hash = "sha256:289e9ca1a9287f08daaf796d96e06cb2bc2958891d7911ac7cae1c5f9e1e0ee3"}, - {file = "lxml-4.6.3-cp36-cp36m-manylinux1_x86_64.whl", hash = "sha256:bccbfc27563652de7dc9bdc595cb25e90b59c5f8e23e806ed0fd623755b6565d"}, - {file = "lxml-4.6.3-cp36-cp36m-manylinux2010_i686.whl", hash = "sha256:d916d31fd85b2f78c76400d625076d9124de3e4bda8b016d25a050cc7d603f24"}, - {file = "lxml-4.6.3-cp36-cp36m-manylinux2014_aarch64.whl", hash = "sha256:820628b7b3135403540202e60551e741f9b6d3304371712521be939470b454ec"}, - {file = "lxml-4.6.3-cp36-cp36m-manylinux2014_x86_64.whl", hash = "sha256:c47ff7e0a36d4efac9fd692cfa33fbd0636674c102e9e8d9b26e1b93a94e7617"}, - {file = "lxml-4.6.3-cp36-cp36m-win32.whl", hash = "sha256:5a0a14e264069c03e46f926be0d8919f4105c1623d620e7ec0e612a2e9bf1c04"}, - {file = "lxml-4.6.3-cp36-cp36m-win_amd64.whl", hash = "sha256:92e821e43ad382332eade6812e298dc9701c75fe289f2a2d39c7960b43d1e92a"}, - {file = "lxml-4.6.3-cp37-cp37m-macosx_10_9_x86_64.whl", hash = "sha256:efd7a09678fd8b53117f6bae4fa3825e0a22b03ef0a932e070c0bdbb3a35e654"}, - {file = "lxml-4.6.3-cp37-cp37m-manylinux1_i686.whl", hash = "sha256:efac139c3f0bf4f0939f9375af4b02c5ad83a622de52d6dfa8e438e8e01d0eb0"}, - {file = "lxml-4.6.3-cp37-cp37m-manylinux1_x86_64.whl", hash = "sha256:0fbcf5565ac01dff87cbfc0ff323515c823081c5777a9fc7703ff58388c258c3"}, - {file = "lxml-4.6.3-cp37-cp37m-manylinux2010_i686.whl", hash = "sha256:36108c73739985979bf302006527cf8a20515ce444ba916281d1c43938b8bb96"}, - {file = "lxml-4.6.3-cp37-cp37m-manylinux2014_aarch64.whl", hash = "sha256:122fba10466c7bd4178b07dba427aa516286b846b2cbd6f6169141917283aae2"}, - {file = "lxml-4.6.3-cp37-cp37m-manylinux2014_x86_64.whl", hash = "sha256:cdaf11d2bd275bf391b5308f86731e5194a21af45fbaaaf1d9e8147b9160ea92"}, - {file = "lxml-4.6.3-cp37-cp37m-win32.whl", hash = "sha256:3439c71103ef0e904ea0a1901611863e51f50b5cd5e8654a151740fde5e1cade"}, - {file = "lxml-4.6.3-cp37-cp37m-win_amd64.whl", hash = "sha256:4289728b5e2000a4ad4ab8da6e1db2e093c63c08bdc0414799ee776a3f78da4b"}, - {file = "lxml-4.6.3-cp38-cp38-macosx_10_9_x86_64.whl", hash = "sha256:b007cbb845b28db4fb8b6a5cdcbf65bacb16a8bd328b53cbc0698688a68e1caa"}, - {file = "lxml-4.6.3-cp38-cp38-manylinux1_i686.whl", hash = "sha256:76fa7b1362d19f8fbd3e75fe2fb7c79359b0af8747e6f7141c338f0bee2f871a"}, - {file = "lxml-4.6.3-cp38-cp38-manylinux1_x86_64.whl", hash = "sha256:26e761ab5b07adf5f555ee82fb4bfc35bf93750499c6c7614bd64d12aaa67927"}, - {file = "lxml-4.6.3-cp38-cp38-manylinux2010_i686.whl", hash = "sha256:e1cbd3f19a61e27e011e02f9600837b921ac661f0c40560eefb366e4e4fb275e"}, - {file = "lxml-4.6.3-cp38-cp38-manylinux2014_aarch64.whl", hash = "sha256:66e575c62792c3f9ca47cb8b6fab9e35bab91360c783d1606f758761810c9791"}, - {file = "lxml-4.6.3-cp38-cp38-manylinux2014_x86_64.whl", hash = "sha256:1b38116b6e628118dea5b2186ee6820ab138dbb1e24a13e478490c7db2f326ae"}, - {file = "lxml-4.6.3-cp38-cp38-win32.whl", hash = "sha256:89b8b22a5ff72d89d48d0e62abb14340d9e99fd637d046c27b8b257a01ffbe28"}, - {file = "lxml-4.6.3-cp38-cp38-win_amd64.whl", hash = "sha256:2a9d50e69aac3ebee695424f7dbd7b8c6d6eb7de2a2eb6b0f6c7db6aa41e02b7"}, - {file = "lxml-4.6.3-cp39-cp39-macosx_10_9_x86_64.whl", hash = "sha256:ce256aaa50f6cc9a649c51be3cd4ff142d67295bfc4f490c9134d0f9f6d58ef0"}, - {file = "lxml-4.6.3-cp39-cp39-manylinux1_i686.whl", hash = "sha256:7610b8c31688f0b1be0ef882889817939490a36d0ee880ea562a4e1399c447a1"}, - {file = "lxml-4.6.3-cp39-cp39-manylinux1_x86_64.whl", hash = "sha256:f8380c03e45cf09f8557bdaa41e1fa7c81f3ae22828e1db470ab2a6c96d8bc23"}, - {file = "lxml-4.6.3-cp39-cp39-manylinux2010_i686.whl", hash = "sha256:3082c518be8e97324390614dacd041bb1358c882d77108ca1957ba47738d9d59"}, - {file = "lxml-4.6.3-cp39-cp39-manylinux2014_aarch64.whl", hash = "sha256:884ab9b29feaca361f7f88d811b1eea9bfca36cf3da27768d28ad45c3ee6f969"}, - {file = "lxml-4.6.3-cp39-cp39-manylinux2014_x86_64.whl", hash = "sha256:6f12e1427285008fd32a6025e38e977d44d6382cf28e7201ed10d6c1698d2a9a"}, - {file = "lxml-4.6.3-cp39-cp39-win32.whl", hash = "sha256:33bb934a044cf32157c12bfcfbb6649807da20aa92c062ef51903415c704704f"}, - {file = "lxml-4.6.3-cp39-cp39-win_amd64.whl", hash = "sha256:542d454665a3e277f76954418124d67516c5f88e51a900365ed54a9806122b83"}, - {file = "lxml-4.6.3.tar.gz", hash = "sha256:39b78571b3b30645ac77b95f7c69d1bffc4cf8c3b157c435a34da72e78c82468"}, + {file = "lxml-4.6.4-cp27-cp27m-macosx_10_14_x86_64.whl", hash = "sha256:bbf2dc330bd44bfc0254ab37677ec60f7c7ecea55ad8ba1b8b2ea7bf20c265f5"}, + {file = "lxml-4.6.4-cp27-cp27m-manylinux_2_5_i686.manylinux1_i686.whl", hash = "sha256:b667c51682fe9b9788c69465956baa8b6999531876ccedcafc895c74ad716cd8"}, + {file = "lxml-4.6.4-cp27-cp27m-manylinux_2_5_x86_64.manylinux1_x86_64.whl", hash = "sha256:72e730d33fe2e302fd07285f14624fca5e5e2fb2bb4fb2c3941e318c41c443d1"}, + {file = "lxml-4.6.4-cp27-cp27m-win32.whl", hash = "sha256:433df8c7dde0f9e41cbf4f36b0829d50a378116ef5e962ba3881f2f5f025c7be"}, + {file = "lxml-4.6.4-cp27-cp27m-win_amd64.whl", hash = "sha256:35752ee40f7bbf6adc9ff4e1f4b84794a3593736dcce80db32e3c2aa85e294ac"}, + {file = "lxml-4.6.4-cp27-cp27mu-manylinux_2_5_i686.manylinux1_i686.whl", hash = "sha256:5ff5bb2a198ea67403bb6818705e9a4f90e0313f2215428ec51001ce56d939fb"}, + {file = "lxml-4.6.4-cp27-cp27mu-manylinux_2_5_x86_64.manylinux1_x86_64.whl", hash = "sha256:9b87727561c1150c0cc91c5d9d389448b37a7d15f0ba939ed3d1acb2f11bf6c5"}, + {file = "lxml-4.6.4-cp310-cp310-macosx_10_14_x86_64.whl", hash = "sha256:45fdb2899c755138722797161547a40b3e2a06feda620cc41195ee7e97806d81"}, + {file = "lxml-4.6.4-cp310-cp310-manylinux_2_12_i686.manylinux2010_i686.manylinux_2_24_i686.whl", hash = "sha256:38b9de0de3aa689fe9fb9877ae1be1e83b8cf9621f7e62049d0436b9ecf4ad64"}, + {file = "lxml-4.6.4-cp310-cp310-manylinux_2_17_aarch64.manylinux2014_aarch64.manylinux_2_24_aarch64.whl", hash = "sha256:662523cd2a0246740225c7e32531f2e766544122e58bee70e700a024cfc0cf81"}, + {file = "lxml-4.6.4-cp310-cp310-manylinux_2_17_x86_64.manylinux2014_x86_64.manylinux_2_24_x86_64.whl", hash = "sha256:4aa349c5567651f34d4eaae7de6ed5b523f6d70a288f9c6fbac22d13a0784e04"}, + {file = "lxml-4.6.4-cp310-cp310-musllinux_1_1_x86_64.whl", hash = "sha256:08eb9200d88b376a8ed5e50f1dc1d1a45b49305169674002a3b5929943390591"}, + {file = "lxml-4.6.4-cp310-cp310-win32.whl", hash = "sha256:bdc224f216ead849e902151112efef6e96c41ee1322e15d4e5f7c8a826929aee"}, + {file = "lxml-4.6.4-cp310-cp310-win_amd64.whl", hash = "sha256:ab6db93a2b6b66cbf62b4e4a7135f476e708e8c5c990d186584142c77d7f975a"}, + {file = "lxml-4.6.4-cp35-cp35m-manylinux_2_5_i686.manylinux1_i686.whl", hash = "sha256:50790313df028aa05cf22be9a8da033b86c42fa32523e4fd944827b482b17bf0"}, + {file = "lxml-4.6.4-cp35-cp35m-manylinux_2_5_x86_64.manylinux1_x86_64.whl", hash = "sha256:6764998345552b1dfc9326a932d2bad6367c6b37a176bb73ada6b9486bf602f7"}, + {file = "lxml-4.6.4-cp35-cp35m-win32.whl", hash = "sha256:543b239b191bb3b6d9bef5f09f1fb2be5b7eb09ab4d386aa655e4d53fbe9ff47"}, + {file = "lxml-4.6.4-cp35-cp35m-win_amd64.whl", hash = "sha256:a75c1ad05eedb1a3ff2a34a52a4f0836cfaa892e12796ba39a7732c82701eff4"}, + {file = "lxml-4.6.4-cp36-cp36m-macosx_10_14_x86_64.whl", hash = "sha256:47e955112ce64241fdb357acf0216081f9f3255b3ac9c502ca4b3323ec1ca558"}, + {file = "lxml-4.6.4-cp36-cp36m-manylinux_2_12_i686.manylinux2010_i686.manylinux_2_24_i686.whl", hash = "sha256:20d7c8d90d449c6a353b15ee0459abae8395dbe59ad01e406ccbf30cd81c6f98"}, + {file = "lxml-4.6.4-cp36-cp36m-manylinux_2_17_x86_64.manylinux2014_x86_64.manylinux_2_24_x86_64.whl", hash = "sha256:240db6f3228d26e3c6f4fad914b9ddaaf8707254e8b3efd564dc680c8ec3c264"}, + {file = "lxml-4.6.4-cp36-cp36m-manylinux_2_5_i686.manylinux1_i686.whl", hash = "sha256:351482da8dd028834028537f08724b1de22d40dcf3bb723b469446564f409074"}, + {file = "lxml-4.6.4-cp36-cp36m-manylinux_2_5_x86_64.manylinux1_x86_64.whl", hash = "sha256:e678a643177c0e5ec947b645fa7bc84260dfb9b6bf8fb1fdd83008dfc2ca5928"}, + {file = "lxml-4.6.4-cp36-cp36m-musllinux_1_1_x86_64.whl", hash = "sha256:15d0381feb56f08f78c5cc4fc385ddfe0bde1456e37f54a9322833371aec4060"}, + {file = "lxml-4.6.4-cp36-cp36m-win32.whl", hash = "sha256:4ba74afe5ee5cb5e28d83b513a6e8f0875fda1dc1a9aea42cc0065f029160d2a"}, + {file = "lxml-4.6.4-cp36-cp36m-win_amd64.whl", hash = "sha256:9c91a73971a922c13070fd8fa5a114c858251791ba2122a941e6aa781c713e44"}, + {file = "lxml-4.6.4-cp37-cp37m-macosx_10_14_x86_64.whl", hash = "sha256:6020c70ff695106bf80651953a23e37718ef1fee9abd060dcad8e32ab2dc13f3"}, + {file = "lxml-4.6.4-cp37-cp37m-manylinux_2_12_i686.manylinux2010_i686.manylinux_2_24_i686.whl", hash = "sha256:f5dd358536b8a964bf6bd48de038754c1609e72e5f17f5d21efe2dda17594dbf"}, + {file = "lxml-4.6.4-cp37-cp37m-manylinux_2_17_aarch64.manylinux2014_aarch64.manylinux_2_24_aarch64.whl", hash = "sha256:7ae7089d81fc502df4b217ad77f03c54039fe90dac0acbe70448d7e53bfbc57e"}, + {file = "lxml-4.6.4-cp37-cp37m-manylinux_2_17_x86_64.manylinux2014_x86_64.manylinux_2_24_x86_64.whl", hash = "sha256:80d10d53d3184837445ff8562021bdd37f57c4cadacbf9d8726cc16220a00d54"}, + {file = "lxml-4.6.4-cp37-cp37m-manylinux_2_5_i686.manylinux1_i686.whl", hash = "sha256:e95da348d57eb448d226a44b868ff2ca5786fbcbe417ac99ff62d0a7d724b9c7"}, + {file = "lxml-4.6.4-cp37-cp37m-manylinux_2_5_x86_64.manylinux1_x86_64.whl", hash = "sha256:ffd65cfa33fed01735c82aca640fde4cc63f0414775cba11e06f84fae2085a6e"}, + {file = "lxml-4.6.4-cp37-cp37m-musllinux_1_1_x86_64.whl", hash = "sha256:877666418598f6cb289546c77ff87590cfd212f903b522b0afa0b9fb73b3ccfb"}, + {file = "lxml-4.6.4-cp37-cp37m-win32.whl", hash = "sha256:e91d24623e747eeb2d8121f4a94c6a7ad27dc48e747e2dc95bfe88632bd028a2"}, + {file = "lxml-4.6.4-cp37-cp37m-win_amd64.whl", hash = "sha256:4ec9a80dd5704ecfde54319b6964368daf02848c8954d3bacb9b64d1c7659159"}, + {file = "lxml-4.6.4-cp38-cp38-macosx_10_14_x86_64.whl", hash = "sha256:2901625f4a878a055d275beedc20ba9cb359cefc4386a967222fee29eb236038"}, + {file = "lxml-4.6.4-cp38-cp38-manylinux_2_12_i686.manylinux2010_i686.manylinux_2_24_i686.whl", hash = "sha256:b567178a74a2261345890eac66fbf394692a6e002709d329f28a673ca6042473"}, + {file = "lxml-4.6.4-cp38-cp38-manylinux_2_17_aarch64.manylinux2014_aarch64.manylinux_2_24_aarch64.whl", hash = "sha256:4717123f7c11c81e0da69989e5a64079c3f402b0efeb4c6241db6c369d657bd8"}, + {file = "lxml-4.6.4-cp38-cp38-manylinux_2_17_x86_64.manylinux2014_x86_64.manylinux_2_24_x86_64.whl", hash = "sha256:cf201bf5594d1aab139fe53e3fca457e4f8204a5bbd65d48ab3b82a16f517868"}, + {file = "lxml-4.6.4-cp38-cp38-manylinux_2_5_i686.manylinux1_i686.whl", hash = "sha256:a77a3470ba37e11872c75ca95baf9b3312133a3d5a5dc720803b23098c653976"}, + {file = "lxml-4.6.4-cp38-cp38-manylinux_2_5_x86_64.manylinux1_x86_64.whl", hash = "sha256:619c6d2b552bba00491e96c0518aad94002651c108a0f7364ff2d7798812c00e"}, + {file = "lxml-4.6.4-cp38-cp38-musllinux_1_1_x86_64.whl", hash = "sha256:601f0ab75538b280aaf1e720eb9d68d4fa104ac274e1e9e6971df488f4dcdb0f"}, + {file = "lxml-4.6.4-cp38-cp38-win32.whl", hash = "sha256:75d3c5bbc0ddbad03bb68b9be638599f67e4b98ed3dcd0fec9f6f39e41ee96cb"}, + {file = "lxml-4.6.4-cp38-cp38-win_amd64.whl", hash = "sha256:4341d135f5660db10184963d9c3418c3e28d7f868aaf8b11a323ebf85813f7f4"}, + {file = "lxml-4.6.4-cp39-cp39-macosx_10_14_x86_64.whl", hash = "sha256:9db24803fa71e3305fe4a7812782b708da21a0b774b130dd1860cf40a6d7a3ee"}, + {file = "lxml-4.6.4-cp39-cp39-manylinux_2_12_i686.manylinux2010_i686.manylinux_2_24_i686.whl", hash = "sha256:afd60230ad9d8bcba005945ec3a343722f09e0b7f8ae804246e5d2cfc6bd71a6"}, + {file = "lxml-4.6.4-cp39-cp39-manylinux_2_17_aarch64.manylinux2014_aarch64.manylinux_2_24_aarch64.whl", hash = "sha256:0c15e1cd55055956e77b0732270f1c6005850696bc3ef3e03d01e78af84eaa42"}, + {file = "lxml-4.6.4-cp39-cp39-manylinux_2_17_x86_64.manylinux2014_x86_64.manylinux_2_24_x86_64.whl", hash = "sha256:6d422b3c729737d8a39279a25fa156c983a56458f8b2f97661ee6fb22b80b1d6"}, + {file = "lxml-4.6.4-cp39-cp39-manylinux_2_5_i686.manylinux1_i686.whl", hash = "sha256:2eb90f6ec3c236ef2f1bb38aee7c0d23e77d423d395af6326e7cca637519a4cb"}, + {file = "lxml-4.6.4-cp39-cp39-manylinux_2_5_x86_64.manylinux1_x86_64.whl", hash = "sha256:51a0e5d243687596f46e24e464121d4b232ad772e2d1785b2a2c0eb413c285d4"}, + {file = "lxml-4.6.4-cp39-cp39-musllinux_1_1_x86_64.whl", hash = "sha256:d43bd68714049c84e297c005456a15ecdec818f7b5aa5868c8b0a865cfb78a44"}, + {file = "lxml-4.6.4-cp39-cp39-win32.whl", hash = "sha256:ee9e4b07b0eba4b6a521509e9e1877476729c1243246b6959de697ebea739643"}, + {file = "lxml-4.6.4-cp39-cp39-win_amd64.whl", hash = "sha256:48eaac2991b3036175b42ee8d3c23f4cca13f2be8426bf29401a690ab58c88f4"}, + {file = "lxml-4.6.4-pp37-pypy37_pp73-macosx_10_14_x86_64.whl", hash = "sha256:2b06a91cf7b8acea7793006e4ae50646cef0fe35ce5acd4f5cb1c77eb228e4a1"}, + {file = "lxml-4.6.4-pp37-pypy37_pp73-manylinux_2_12_i686.manylinux2010_i686.manylinux_2_24_i686.whl", hash = "sha256:523f195948a1ba4f9f5b7294d83c6cd876547dc741820750a7e5e893a24bbe38"}, + {file = "lxml-4.6.4-pp37-pypy37_pp73-manylinux_2_17_x86_64.manylinux2014_x86_64.manylinux_2_24_x86_64.whl", hash = "sha256:b0ca0ada9d3bc18bd6f611bd001a28abdd49ab9698bd6d717f7f5394c8e94628"}, + {file = "lxml-4.6.4-pp38-pypy38_pp73-macosx_10_14_x86_64.whl", hash = "sha256:197b7cb7a753cf553a45115739afd8458464a28913da00f5c525063f94cd3f48"}, + {file = "lxml-4.6.4-pp38-pypy38_pp73-manylinux_2_12_i686.manylinux2010_i686.manylinux_2_24_i686.whl", hash = "sha256:6298f5b42a26581206ef63fffa97c754245d329414108707c525512a5197f2ba"}, + {file = "lxml-4.6.4-pp38-pypy38_pp73-manylinux_2_17_x86_64.manylinux2014_x86_64.manylinux_2_24_x86_64.whl", hash = "sha256:0b12c95542f04d10cba46b3ff28ea52ea56995b78cf918f0b11b05e75812bb79"}, + {file = "lxml-4.6.4.tar.gz", hash = "sha256:daf9bd1fee31f1c7a5928b3e1059e09a8d683ea58fb3ffc773b6c88cb8d1399c"}, ] markupsafe = [ {file = "MarkupSafe-2.0.1-cp310-cp310-macosx_10_9_universal2.whl", hash = "sha256:d8446c54dc28c01e5a2dbac5a25f071f6653e6e40f3a8818e8b45d790fe6ef53"}, @@ -1294,15 +1470,22 @@ nodeenv = [ {file = "nodeenv-1.6.0-py2.py3-none-any.whl", hash = "sha256:621e6b7076565ddcacd2db0294c0381e01fd28945ab36bcf00f41c5daf63bef7"}, {file = "nodeenv-1.6.0.tar.gz", hash = "sha256:3ef13ff90291ba2a4a7a4ff9a979b63ffdd00a464dbe04acf0ea6471517a4c2b"}, ] -openleadr = [] +openleadr = [ + {file = "openleadr-0.5.24-py3-none-any.whl", hash = "sha256:8723ff25eb02ab8b31b0cdca7d2429f9bb21e337b9eb809978237c644bbf5390"}, + {file = "openleadr-0.5.24.tar.gz", hash = "sha256:c06078040ca4246c85ca6e8378c386737df8daf7df86dbebe82a74d666185de8"}, +] packaging = [ - {file = "packaging-21.0-py3-none-any.whl", hash = "sha256:c86254f9220d55e31cc94d69bade760f0847da8000def4dfe1c6b872fd14ff14"}, - {file = "packaging-21.0.tar.gz", hash = "sha256:7dc96269f53a4ccec5c0670940a4281106dd0bb343f47b7471f779df49c2fbe7"}, + {file = "packaging-21.3-py3-none-any.whl", hash = "sha256:ef103e05f519cdc783ae24ea4e2e0f508a9c99b2d4969652eed6a2e1ea5bd522"}, + {file = "packaging-21.3.tar.gz", hash = "sha256:dd47c42927d89ab911e606518907cc2d3a1f38bbd026385970643f9c5b8ecfeb"}, ] pathspec = [ {file = "pathspec-0.9.0-py2.py3-none-any.whl", hash = "sha256:7d15c4ddb0b5c802d161efc417ec1a2558ea2653c2e8ad9c19098201dc1c993a"}, {file = "pathspec-0.9.0.tar.gz", hash = "sha256:e564499435a2673d586f6b2130bb5b95f04a3ba06f81b8f895b651a3c76aabb1"}, ] +pbr = [ + {file = "pbr-5.8.0-py2.py3-none-any.whl", hash = "sha256:176e8560eaf61e127817ef93d8a844803abb27a4d4637f0ff3bb783129be2e0a"}, + {file = "pbr-5.8.0.tar.gz", hash = "sha256:672d8ebee84921862110f23fcec2acea191ef58543d34dfe9ef3d9f13c31cddf"}, +] platformdirs = [ {file = "platformdirs-2.4.0-py3-none-any.whl", hash = "sha256:8868bbe3c3c80d42f20156f22e7131d2fb321f5bc86a2a345375c6481a67021d"}, {file = "platformdirs-2.4.0.tar.gz", hash = "sha256:367a5e80b3d04d2428ffa76d33f124cf11e8fff2acdaa9b43d545f5c7d661ef2"}, @@ -1312,8 +1495,8 @@ pluggy = [ {file = "pluggy-1.0.0.tar.gz", hash = "sha256:4224373bacce55f955a878bf9cfa763c1e360858e330072059e10bad68531159"}, ] pre-commit = [ - {file = "pre_commit-2.15.0-py2.py3-none-any.whl", hash = "sha256:a4ed01000afcb484d9eb8d504272e642c4c4099bbad3a6b27e519bd6a3e928a6"}, - {file = "pre_commit-2.15.0.tar.gz", hash = "sha256:3c25add78dbdfb6a28a651780d5c311ac40dd17f160eb3954a0c59da40a505a7"}, + {file = "pre_commit-2.16.0-py2.py3-none-any.whl", hash = "sha256:758d1dc9b62c2ed8881585c254976d66eae0889919ab9b859064fc2fe3c7743e"}, + {file = "pre_commit-2.16.0.tar.gz", hash = "sha256:fe9897cac830aa7164dbd02a4e7b90cae49630451ce88464bca73db486ba9f65"}, ] psutil = [ {file = "psutil-5.8.0-cp27-cp27m-macosx_10_9_x86_64.whl", hash = "sha256:0066a82f7b1b37d334e68697faba68e5ad5e858279fd6351c8ca6024e8d6ba64"}, @@ -1346,20 +1529,20 @@ psutil = [ {file = "psutil-5.8.0.tar.gz", hash = "sha256:0c9ccb99ab76025f2f0bbecf341d4656e9c1351db8cc8a03ccd62e318ab4b5c6"}, ] py = [ - {file = "py-1.10.0-py2.py3-none-any.whl", hash = "sha256:3b80836aa6d1feeaa108e046da6423ab8f6ceda6468545ae8d02d9d58d18818a"}, - {file = "py-1.10.0.tar.gz", hash = "sha256:21b81bda15b66ef5e1a777a21c4dcd9c20ad3efd0b3f817e7a809035269e1bd3"}, + {file = "py-1.11.0-py2.py3-none-any.whl", hash = "sha256:607c53218732647dff4acdfcd50cb62615cedf612e72d1724fb1a0cc6405b378"}, + {file = "py-1.11.0.tar.gz", hash = "sha256:51c75c4126074b472f746a24399ad32f6053d1b34b68d2fa41e558e6f4a98719"}, ] pycparser = [ - {file = "pycparser-2.20-py2.py3-none-any.whl", hash = "sha256:7582ad22678f0fcd81102833f60ef8d0e57288b6b5fb00323d101be910e35705"}, - {file = "pycparser-2.20.tar.gz", hash = "sha256:2d475327684562c3a96cc71adf7dc8c4f0565175cf86b6d7a404ff4c771f15f0"}, + {file = "pycparser-2.21-py2.py3-none-any.whl", hash = "sha256:8ee45429555515e1f6b185e78100aea234072576aa43ab53aefcae078162fca9"}, + {file = "pycparser-2.21.tar.gz", hash = "sha256:e644fdec12f7872f86c58ff790da456218b10f863970249516d60a5eaca77206"}, ] pyopenssl = [ {file = "pyOpenSSL-19.0.0-py2.py3-none-any.whl", hash = "sha256:c727930ad54b10fc157015014b666f2d8b41f70c0d03e83ab67624fd3dd5d1e6"}, {file = "pyOpenSSL-19.0.0.tar.gz", hash = "sha256:aeca66338f6de19d1aa46ed634c3b9ae519a64b458f8468aec688e7e3c20f200"}, ] pyparsing = [ - {file = "pyparsing-2.4.7-py2.py3-none-any.whl", hash = "sha256:ef9d7589ef3c200abe66653d3f1ab1033c3c419ae9b9bdb1240a85b024efc88b"}, - {file = "pyparsing-2.4.7.tar.gz", hash = "sha256:c203ec8783bf771a155b207279b9bccb8dea02d8f0c9e5f8ead507bc3246ecc1"}, + {file = "pyparsing-3.0.6-py3-none-any.whl", hash = "sha256:04ff808a5b90911829c55c4e26f75fa5ca8a2f5f36aa3a51f68e27033341d3e4"}, + {file = "pyparsing-3.0.6.tar.gz", hash = "sha256:d9bdec0013ef1eb5a84ab39a3b3868911598afa494f5faa038647101504e2b81"}, ] pytest = [ {file = "pytest-6.2.5-py3-none-any.whl", hash = "sha256:7310f8d27bc79ced999e760ca304d69f6ba6c6649c0b60fb0e04a4a77cacc134"}, @@ -1374,35 +1557,39 @@ pytz = [ {file = "pytz-2021.3.tar.gz", hash = "sha256:acad2d8b20a1af07d4e4c9d2e9285c5ed9104354062f275f3fcd88dcef4f1326"}, ] pyyaml = [ - {file = "PyYAML-5.4.1-cp27-cp27m-macosx_10_9_x86_64.whl", hash = "sha256:3b2b1824fe7112845700f815ff6a489360226a5609b96ec2190a45e62a9fc922"}, - {file = "PyYAML-5.4.1-cp27-cp27m-win32.whl", hash = "sha256:129def1b7c1bf22faffd67b8f3724645203b79d8f4cc81f674654d9902cb4393"}, - {file = "PyYAML-5.4.1-cp27-cp27m-win_amd64.whl", hash = "sha256:4465124ef1b18d9ace298060f4eccc64b0850899ac4ac53294547536533800c8"}, - {file = "PyYAML-5.4.1-cp27-cp27mu-manylinux1_x86_64.whl", hash = "sha256:bb4191dfc9306777bc594117aee052446b3fa88737cd13b7188d0e7aa8162185"}, - {file = "PyYAML-5.4.1-cp36-cp36m-macosx_10_9_x86_64.whl", hash = "sha256:6c78645d400265a062508ae399b60b8c167bf003db364ecb26dcab2bda048253"}, - {file = "PyYAML-5.4.1-cp36-cp36m-manylinux1_x86_64.whl", hash = "sha256:4e0583d24c881e14342eaf4ec5fbc97f934b999a6828693a99157fde912540cc"}, - {file = "PyYAML-5.4.1-cp36-cp36m-manylinux2014_aarch64.whl", hash = "sha256:72a01f726a9c7851ca9bfad6fd09ca4e090a023c00945ea05ba1638c09dc3347"}, - {file = "PyYAML-5.4.1-cp36-cp36m-manylinux2014_s390x.whl", hash = "sha256:895f61ef02e8fed38159bb70f7e100e00f471eae2bc838cd0f4ebb21e28f8541"}, - {file = "PyYAML-5.4.1-cp36-cp36m-win32.whl", hash = "sha256:3bd0e463264cf257d1ffd2e40223b197271046d09dadf73a0fe82b9c1fc385a5"}, - {file = "PyYAML-5.4.1-cp36-cp36m-win_amd64.whl", hash = "sha256:e4fac90784481d221a8e4b1162afa7c47ed953be40d31ab4629ae917510051df"}, - {file = "PyYAML-5.4.1-cp37-cp37m-macosx_10_9_x86_64.whl", hash = "sha256:5accb17103e43963b80e6f837831f38d314a0495500067cb25afab2e8d7a4018"}, - {file = "PyYAML-5.4.1-cp37-cp37m-manylinux1_x86_64.whl", hash = "sha256:e1d4970ea66be07ae37a3c2e48b5ec63f7ba6804bdddfdbd3cfd954d25a82e63"}, - {file = "PyYAML-5.4.1-cp37-cp37m-manylinux2014_aarch64.whl", hash = "sha256:cb333c16912324fd5f769fff6bc5de372e9e7a202247b48870bc251ed40239aa"}, - {file = "PyYAML-5.4.1-cp37-cp37m-manylinux2014_s390x.whl", hash = "sha256:fe69978f3f768926cfa37b867e3843918e012cf83f680806599ddce33c2c68b0"}, - {file = "PyYAML-5.4.1-cp37-cp37m-win32.whl", hash = "sha256:dd5de0646207f053eb0d6c74ae45ba98c3395a571a2891858e87df7c9b9bd51b"}, - {file = "PyYAML-5.4.1-cp37-cp37m-win_amd64.whl", hash = "sha256:08682f6b72c722394747bddaf0aa62277e02557c0fd1c42cb853016a38f8dedf"}, - {file = "PyYAML-5.4.1-cp38-cp38-macosx_10_9_x86_64.whl", hash = "sha256:d2d9808ea7b4af864f35ea216be506ecec180628aced0704e34aca0b040ffe46"}, - {file = "PyYAML-5.4.1-cp38-cp38-manylinux1_x86_64.whl", hash = "sha256:8c1be557ee92a20f184922c7b6424e8ab6691788e6d86137c5d93c1a6ec1b8fb"}, - {file = "PyYAML-5.4.1-cp38-cp38-manylinux2014_aarch64.whl", hash = "sha256:fd7f6999a8070df521b6384004ef42833b9bd62cfee11a09bda1079b4b704247"}, - {file = "PyYAML-5.4.1-cp38-cp38-manylinux2014_s390x.whl", hash = "sha256:bfb51918d4ff3d77c1c856a9699f8492c612cde32fd3bcd344af9be34999bfdc"}, - {file = "PyYAML-5.4.1-cp38-cp38-win32.whl", hash = "sha256:fa5ae20527d8e831e8230cbffd9f8fe952815b2b7dae6ffec25318803a7528fc"}, - {file = "PyYAML-5.4.1-cp38-cp38-win_amd64.whl", hash = "sha256:0f5f5786c0e09baddcd8b4b45f20a7b5d61a7e7e99846e3c799b05c7c53fa696"}, - {file = "PyYAML-5.4.1-cp39-cp39-macosx_10_9_x86_64.whl", hash = "sha256:294db365efa064d00b8d1ef65d8ea2c3426ac366c0c4368d930bf1c5fb497f77"}, - {file = "PyYAML-5.4.1-cp39-cp39-manylinux1_x86_64.whl", hash = "sha256:74c1485f7707cf707a7aef42ef6322b8f97921bd89be2ab6317fd782c2d53183"}, - {file = "PyYAML-5.4.1-cp39-cp39-manylinux2014_aarch64.whl", hash = "sha256:d483ad4e639292c90170eb6f7783ad19490e7a8defb3e46f97dfe4bacae89122"}, - {file = "PyYAML-5.4.1-cp39-cp39-manylinux2014_s390x.whl", hash = "sha256:fdc842473cd33f45ff6bce46aea678a54e3d21f1b61a7750ce3c498eedfe25d6"}, - {file = "PyYAML-5.4.1-cp39-cp39-win32.whl", hash = "sha256:49d4cdd9065b9b6e206d0595fee27a96b5dd22618e7520c33204a4a3239d5b10"}, - {file = "PyYAML-5.4.1-cp39-cp39-win_amd64.whl", hash = "sha256:c20cfa2d49991c8b4147af39859b167664f2ad4561704ee74c1de03318e898db"}, - {file = "PyYAML-5.4.1.tar.gz", hash = "sha256:607774cbba28732bfa802b54baa7484215f530991055bb562efbed5b2f20a45e"}, + {file = "PyYAML-6.0-cp310-cp310-macosx_10_9_x86_64.whl", hash = "sha256:d4db7c7aef085872ef65a8fd7d6d09a14ae91f691dec3e87ee5ee0539d516f53"}, + {file = "PyYAML-6.0-cp310-cp310-macosx_11_0_arm64.whl", hash = "sha256:9df7ed3b3d2e0ecfe09e14741b857df43adb5a3ddadc919a2d94fbdf78fea53c"}, + {file = "PyYAML-6.0-cp310-cp310-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:77f396e6ef4c73fdc33a9157446466f1cff553d979bd00ecb64385760c6babdc"}, + {file = "PyYAML-6.0-cp310-cp310-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:a80a78046a72361de73f8f395f1f1e49f956c6be882eed58505a15f3e430962b"}, + {file = "PyYAML-6.0-cp310-cp310-manylinux_2_5_x86_64.manylinux1_x86_64.manylinux_2_12_x86_64.manylinux2010_x86_64.whl", hash = "sha256:f84fbc98b019fef2ee9a1cb3ce93e3187a6df0b2538a651bfb890254ba9f90b5"}, + {file = "PyYAML-6.0-cp310-cp310-win32.whl", hash = "sha256:2cd5df3de48857ed0544b34e2d40e9fac445930039f3cfe4bcc592a1f836d513"}, + {file = "PyYAML-6.0-cp310-cp310-win_amd64.whl", hash = "sha256:daf496c58a8c52083df09b80c860005194014c3698698d1a57cbcfa182142a3a"}, + {file = "PyYAML-6.0-cp36-cp36m-macosx_10_9_x86_64.whl", hash = "sha256:897b80890765f037df3403d22bab41627ca8811ae55e9a722fd0392850ec4d86"}, + {file = "PyYAML-6.0-cp36-cp36m-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:50602afada6d6cbfad699b0c7bb50d5ccffa7e46a3d738092afddc1f9758427f"}, + {file = "PyYAML-6.0-cp36-cp36m-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:48c346915c114f5fdb3ead70312bd042a953a8ce5c7106d5bfb1a5254e47da92"}, + {file = "PyYAML-6.0-cp36-cp36m-manylinux_2_5_x86_64.manylinux1_x86_64.manylinux_2_12_x86_64.manylinux2010_x86_64.whl", hash = "sha256:98c4d36e99714e55cfbaaee6dd5badbc9a1ec339ebfc3b1f52e293aee6bb71a4"}, + {file = "PyYAML-6.0-cp36-cp36m-win32.whl", hash = "sha256:0283c35a6a9fbf047493e3a0ce8d79ef5030852c51e9d911a27badfde0605293"}, + {file = "PyYAML-6.0-cp36-cp36m-win_amd64.whl", hash = "sha256:07751360502caac1c067a8132d150cf3d61339af5691fe9e87803040dbc5db57"}, + {file = "PyYAML-6.0-cp37-cp37m-macosx_10_9_x86_64.whl", hash = "sha256:819b3830a1543db06c4d4b865e70ded25be52a2e0631ccd2f6a47a2822f2fd7c"}, + {file = "PyYAML-6.0-cp37-cp37m-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:473f9edb243cb1935ab5a084eb238d842fb8f404ed2193a915d1784b5a6b5fc0"}, + {file = "PyYAML-6.0-cp37-cp37m-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:0ce82d761c532fe4ec3f87fc45688bdd3a4c1dc5e0b4a19814b9009a29baefd4"}, + {file = "PyYAML-6.0-cp37-cp37m-manylinux_2_5_x86_64.manylinux1_x86_64.manylinux_2_12_x86_64.manylinux2010_x86_64.whl", hash = "sha256:231710d57adfd809ef5d34183b8ed1eeae3f76459c18fb4a0b373ad56bedcdd9"}, + {file = "PyYAML-6.0-cp37-cp37m-win32.whl", hash = "sha256:c5687b8d43cf58545ade1fe3e055f70eac7a5a1a0bf42824308d868289a95737"}, + {file = "PyYAML-6.0-cp37-cp37m-win_amd64.whl", hash = "sha256:d15a181d1ecd0d4270dc32edb46f7cb7733c7c508857278d3d378d14d606db2d"}, + {file = "PyYAML-6.0-cp38-cp38-macosx_10_9_x86_64.whl", hash = "sha256:0b4624f379dab24d3725ffde76559cff63d9ec94e1736b556dacdfebe5ab6d4b"}, + {file = "PyYAML-6.0-cp38-cp38-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:213c60cd50106436cc818accf5baa1aba61c0189ff610f64f4a3e8c6726218ba"}, + {file = "PyYAML-6.0-cp38-cp38-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:9fa600030013c4de8165339db93d182b9431076eb98eb40ee068700c9c813e34"}, + {file = "PyYAML-6.0-cp38-cp38-manylinux_2_5_x86_64.manylinux1_x86_64.manylinux_2_12_x86_64.manylinux2010_x86_64.whl", hash = "sha256:277a0ef2981ca40581a47093e9e2d13b3f1fbbeffae064c1d21bfceba2030287"}, + {file = "PyYAML-6.0-cp38-cp38-win32.whl", hash = "sha256:d4eccecf9adf6fbcc6861a38015c2a64f38b9d94838ac1810a9023a0609e1b78"}, + {file = "PyYAML-6.0-cp38-cp38-win_amd64.whl", hash = "sha256:1e4747bc279b4f613a09eb64bba2ba602d8a6664c6ce6396a4d0cd413a50ce07"}, + {file = "PyYAML-6.0-cp39-cp39-macosx_10_9_x86_64.whl", hash = "sha256:055d937d65826939cb044fc8c9b08889e8c743fdc6a32b33e2390f66013e449b"}, + {file = "PyYAML-6.0-cp39-cp39-macosx_11_0_arm64.whl", hash = "sha256:e61ceaab6f49fb8bdfaa0f92c4b57bcfbea54c09277b1b4f7ac376bfb7a7c174"}, + {file = "PyYAML-6.0-cp39-cp39-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:d67d839ede4ed1b28a4e8909735fc992a923cdb84e618544973d7dfc71540803"}, + {file = "PyYAML-6.0-cp39-cp39-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:cba8c411ef271aa037d7357a2bc8f9ee8b58b9965831d9e51baf703280dc73d3"}, + {file = "PyYAML-6.0-cp39-cp39-manylinux_2_5_x86_64.manylinux1_x86_64.manylinux_2_12_x86_64.manylinux2010_x86_64.whl", hash = "sha256:40527857252b61eacd1d9af500c3337ba8deb8fc298940291486c465c8b46ec0"}, + {file = "PyYAML-6.0-cp39-cp39-win32.whl", hash = "sha256:b5b9eccad747aabaaffbc6064800670f0c297e52c12754eb1d976c57e4f74dcb"}, + {file = "PyYAML-6.0-cp39-cp39-win_amd64.whl", hash = "sha256:b3d267842bf12586ba6c734f89d1f5b871df0273157918b0ccefa29deb05c21c"}, + {file = "PyYAML-6.0.tar.gz", hash = "sha256:68fb519c14306fec9720a2a5b45bc9f0c8d1b9c72adf45c37baedfcd949c35a2"}, ] pyzmq = [ {file = "pyzmq-22.3.0-cp310-cp310-macosx_10_15_universal2.whl", hash = "sha256:6b217b8f9dfb6628f74b94bdaf9f7408708cb02167d644edca33f38746ca12dd"}, @@ -1444,42 +1631,80 @@ pyzmq = [ {file = "pyzmq-22.3.0.tar.gz", hash = "sha256:8eddc033e716f8c91c6a2112f0a8ebc5e00532b4a6ae1eb0ccc48e027f9c671c"}, ] regex = [ - {file = "regex-2021.10.21-cp310-cp310-macosx_10_9_x86_64.whl", hash = "sha256:edff4e31d159672a7b9d70164b21289e4b53b239ce1dc945bf9643d266537573"}, - {file = "regex-2021.10.21-cp310-cp310-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:d6432daf42f2c487b357e1aa0bdc43193f050ff53a3188bfab20b88202b53027"}, - {file = "regex-2021.10.21-cp310-cp310-manylinux_2_5_i686.manylinux1_i686.manylinux_2_12_i686.manylinux2010_i686.whl", hash = "sha256:468de52dd3f20187ab5ca4fd265c1bea61a5346baef01ad0333a5e89fa9fad29"}, - {file = "regex-2021.10.21-cp310-cp310-manylinux_2_5_i686.manylinux1_i686.manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:c5a2ac760f2fc13a1c58131ec217779911890899ce1a0a63c9409bd23fecde6f"}, - {file = "regex-2021.10.21-cp310-cp310-manylinux_2_5_x86_64.manylinux1_x86_64.manylinux_2_12_x86_64.manylinux2010_x86_64.whl", hash = "sha256:ad1fedca001fefc3030d1e9022b038af429e58dc06a7e9c55e40bd1f834582ec"}, - {file = "regex-2021.10.21-cp310-cp310-win32.whl", hash = "sha256:9c613d797a3790f6b12e78a61e1cd29df7fc88135218467cf8b0891353292b9c"}, - {file = "regex-2021.10.21-cp310-cp310-win_amd64.whl", hash = "sha256:678d9a4ce79e1eaa4ebe88bc9769df52919eb30c597576a0deba1f3cf2360e65"}, - {file = "regex-2021.10.21-cp36-cp36m-macosx_10_9_x86_64.whl", hash = "sha256:2044174af237bb9c56ecc07294cf38623ee379e8dca14b01e970f8b015c71917"}, - {file = "regex-2021.10.21-cp36-cp36m-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:98743a2d827a135bf3390452be18d95839b947a099734d53c17e09a64fc09480"}, - {file = "regex-2021.10.21-cp36-cp36m-manylinux_2_5_i686.manylinux1_i686.manylinux_2_12_i686.manylinux2010_i686.whl", hash = "sha256:f1b23304855303bd97b5954edab63b8ddd56c91c41c6d4eba408228c0bae95f3"}, - {file = "regex-2021.10.21-cp36-cp36m-manylinux_2_5_i686.manylinux1_i686.manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:19c4fd59747236423016ccd89b9a6485d958bf1aa7a8a902a6ba28029107a87f"}, - {file = "regex-2021.10.21-cp36-cp36m-manylinux_2_5_x86_64.manylinux1_x86_64.manylinux_2_12_x86_64.manylinux2010_x86_64.whl", hash = "sha256:130a002fa386c976615a2f6d6dff0fcc25da24858994a36b14d2e3129dce7de2"}, - {file = "regex-2021.10.21-cp36-cp36m-win32.whl", hash = "sha256:8bd83d9b8ee125350cd666b55294f4bc9993c4f0d9b1be9344a318d0762e94cc"}, - {file = "regex-2021.10.21-cp36-cp36m-win_amd64.whl", hash = "sha256:98fe0e1b07a314f0a86dc58af4e717c379d48a403eddd8d966ab9b8bf91ce164"}, - {file = "regex-2021.10.21-cp37-cp37m-macosx_10_9_x86_64.whl", hash = "sha256:ded4748c7be6f31fb207387ee83a3a0f625e700defe32f268cb1d350ed6e4a66"}, - {file = "regex-2021.10.21-cp37-cp37m-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:f3da121de36a9ead0f32b44ea720ee8c87edbb59dca6bb980d18377d84ad58a3"}, - {file = "regex-2021.10.21-cp37-cp37m-manylinux_2_5_i686.manylinux1_i686.manylinux_2_12_i686.manylinux2010_i686.whl", hash = "sha256:b9dfba513eae785e3d868803f5a7e21a032cb2b038fa4a1ea7ec691037426ad3"}, - {file = "regex-2021.10.21-cp37-cp37m-manylinux_2_5_i686.manylinux1_i686.manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:2ff91696888755e96230138355cbe8ce2965d930d967d6cff7c636082d038c78"}, - {file = "regex-2021.10.21-cp37-cp37m-manylinux_2_5_x86_64.manylinux1_x86_64.manylinux_2_12_x86_64.manylinux2010_x86_64.whl", hash = "sha256:0f82de529d7595011a40573cc0f27422e59cafa94943b64a4d17d966d75f2c01"}, - {file = "regex-2021.10.21-cp37-cp37m-win32.whl", hash = "sha256:164e51ace4d00f07c519f85ec2209e8faaeab18bc77be6b35685c18d4ac1c22a"}, - {file = "regex-2021.10.21-cp37-cp37m-win_amd64.whl", hash = "sha256:e39eafa854e469d7225066c806c76b9a0acba5ff5ce36c82c0224b75e24888f2"}, - {file = "regex-2021.10.21-cp38-cp38-macosx_10_9_x86_64.whl", hash = "sha256:740a28580520b099b804776db1e919360fcbf30a734a14c5985d5e39a39e7237"}, - {file = "regex-2021.10.21-cp38-cp38-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:de7dbf72ae80f06e79444ff9614fb5e3a7956645d513b0e12d1bbe6f3ccebd11"}, - {file = "regex-2021.10.21-cp38-cp38-manylinux_2_5_i686.manylinux1_i686.manylinux_2_12_i686.manylinux2010_i686.whl", hash = "sha256:dc1a9bedf389bf3d3627a4d2b21cbdc5fe5e0f029d1f465972f4437833dcc946"}, - {file = "regex-2021.10.21-cp38-cp38-manylinux_2_5_i686.manylinux1_i686.manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:9cd14f22425beecf727f6dbdf5c893e46ecbc5ff16197c16a6f38a9066f2d4d5"}, - {file = "regex-2021.10.21-cp38-cp38-manylinux_2_5_x86_64.manylinux1_x86_64.manylinux_2_12_x86_64.manylinux2010_x86_64.whl", hash = "sha256:5b75a3db3aab0bfa51b6af3f820760779d360eb79f59e32c88c7fba648990b4f"}, - {file = "regex-2021.10.21-cp38-cp38-win32.whl", hash = "sha256:f68c71aabb10b1352a06515e25a425a703ba85660ae04cf074da5eb91c0af5e5"}, - {file = "regex-2021.10.21-cp38-cp38-win_amd64.whl", hash = "sha256:c0f49f1f03be3e4a5faaadc35db7afa2b83a871943b889f9f7bba56e0e2e8bd5"}, - {file = "regex-2021.10.21-cp39-cp39-macosx_10_9_x86_64.whl", hash = "sha256:201890fdc8a65396cfb6aa4493201353b2a6378e27d2de65234446f8329233cb"}, - {file = "regex-2021.10.21-cp39-cp39-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:fd1bfc6b7347de9f0ae1fb6f9080426bed6a9ca55b5766fa4fdf7b3a29ccae9c"}, - {file = "regex-2021.10.21-cp39-cp39-manylinux_2_5_i686.manylinux1_i686.manylinux_2_12_i686.manylinux2010_i686.whl", hash = "sha256:72a0b98d41c4508ed23a96eef41090f78630b44ba746e28cd621ecbe961e0a16"}, - {file = "regex-2021.10.21-cp39-cp39-manylinux_2_5_i686.manylinux1_i686.manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:3b5a0660a63b0703380758a7141b96cc1c1a13dee2b8e9c280a2522962fd12af"}, - {file = "regex-2021.10.21-cp39-cp39-manylinux_2_5_x86_64.manylinux1_x86_64.manylinux_2_12_x86_64.manylinux2010_x86_64.whl", hash = "sha256:f82d3adde46ac9188db3aa7e6e1690865ebb6448d245df5a3ea22284f70d9e46"}, - {file = "regex-2021.10.21-cp39-cp39-win32.whl", hash = "sha256:bc4637390235f1e3e2fcdd3e904ca0b42aa655ae28a78072248b2992b4ad4c08"}, - {file = "regex-2021.10.21-cp39-cp39-win_amd64.whl", hash = "sha256:74d03c256cf0aed81997e87be8e24297b5792c9718f3a735f5055ddfad392f06"}, - {file = "regex-2021.10.21.tar.gz", hash = "sha256:4832736b3f24617e63dc919ce8c4215680ba94250a5d9e710fcc0c5f457b5028"}, + {file = "regex-2021.11.10-cp310-cp310-macosx_10_9_x86_64.whl", hash = "sha256:9345b6f7ee578bad8e475129ed40123d265464c4cfead6c261fd60fc9de00bcf"}, + {file = "regex-2021.11.10-cp310-cp310-macosx_11_0_arm64.whl", hash = "sha256:416c5f1a188c91e3eb41e9c8787288e707f7d2ebe66e0a6563af280d9b68478f"}, + {file = "regex-2021.11.10-cp310-cp310-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:e0538c43565ee6e703d3a7c3bdfe4037a5209250e8502c98f20fea6f5fdf2965"}, + {file = "regex-2021.11.10-cp310-cp310-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:7ee1227cf08b6716c85504aebc49ac827eb88fcc6e51564f010f11a406c0a667"}, + {file = "regex-2021.11.10-cp310-cp310-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:6650f16365f1924d6014d2ea770bde8555b4a39dc9576abb95e3cd1ff0263b36"}, + {file = "regex-2021.11.10-cp310-cp310-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:30ab804ea73972049b7a2a5c62d97687d69b5a60a67adca07eb73a0ddbc9e29f"}, + {file = "regex-2021.11.10-cp310-cp310-manylinux_2_5_i686.manylinux1_i686.manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:68a067c11463de2a37157930d8b153005085e42bcb7ad9ca562d77ba7d1404e0"}, + {file = "regex-2021.11.10-cp310-cp310-manylinux_2_5_x86_64.manylinux1_x86_64.manylinux_2_12_x86_64.manylinux2010_x86_64.whl", hash = "sha256:162abfd74e88001d20cb73ceaffbfe601469923e875caf9118333b1a4aaafdc4"}, + {file = "regex-2021.11.10-cp310-cp310-musllinux_1_1_aarch64.whl", hash = "sha256:b9ed0b1e5e0759d6b7f8e2f143894b2a7f3edd313f38cf44e1e15d360e11749b"}, + {file = "regex-2021.11.10-cp310-cp310-musllinux_1_1_i686.whl", hash = "sha256:473e67837f786404570eae33c3b64a4b9635ae9f00145250851a1292f484c063"}, + {file = "regex-2021.11.10-cp310-cp310-musllinux_1_1_ppc64le.whl", hash = "sha256:2fee3ed82a011184807d2127f1733b4f6b2ff6ec7151d83ef3477f3b96a13d03"}, + {file = "regex-2021.11.10-cp310-cp310-musllinux_1_1_s390x.whl", hash = "sha256:d5fd67df77bab0d3f4ea1d7afca9ef15c2ee35dfb348c7b57ffb9782a6e4db6e"}, + {file = "regex-2021.11.10-cp310-cp310-musllinux_1_1_x86_64.whl", hash = "sha256:5d408a642a5484b9b4d11dea15a489ea0928c7e410c7525cd892f4d04f2f617b"}, + {file = "regex-2021.11.10-cp310-cp310-win32.whl", hash = "sha256:98ba568e8ae26beb726aeea2273053c717641933836568c2a0278a84987b2a1a"}, + {file = "regex-2021.11.10-cp310-cp310-win_amd64.whl", hash = "sha256:780b48456a0f0ba4d390e8b5f7c661fdd218934388cde1a974010a965e200e12"}, + {file = "regex-2021.11.10-cp36-cp36m-macosx_10_9_x86_64.whl", hash = "sha256:dba70f30fd81f8ce6d32ddeef37d91c8948e5d5a4c63242d16a2b2df8143aafc"}, + {file = "regex-2021.11.10-cp36-cp36m-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:e1f54b9b4b6c53369f40028d2dd07a8c374583417ee6ec0ea304e710a20f80a0"}, + {file = "regex-2021.11.10-cp36-cp36m-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:fbb9dc00e39f3e6c0ef48edee202f9520dafb233e8b51b06b8428cfcb92abd30"}, + {file = "regex-2021.11.10-cp36-cp36m-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:666abff54e474d28ff42756d94544cdfd42e2ee97065857413b72e8a2d6a6345"}, + {file = "regex-2021.11.10-cp36-cp36m-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:5537f71b6d646f7f5f340562ec4c77b6e1c915f8baae822ea0b7e46c1f09b733"}, + {file = "regex-2021.11.10-cp36-cp36m-manylinux_2_5_i686.manylinux1_i686.manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:ed2e07c6a26ed4bea91b897ee2b0835c21716d9a469a96c3e878dc5f8c55bb23"}, + {file = "regex-2021.11.10-cp36-cp36m-manylinux_2_5_x86_64.manylinux1_x86_64.manylinux_2_12_x86_64.manylinux2010_x86_64.whl", hash = "sha256:ca5f18a75e1256ce07494e245cdb146f5a9267d3c702ebf9b65c7f8bd843431e"}, + {file = "regex-2021.11.10-cp36-cp36m-musllinux_1_1_aarch64.whl", hash = "sha256:74cbeac0451f27d4f50e6e8a8f3a52ca074b5e2da9f7b505c4201a57a8ed6286"}, + {file = "regex-2021.11.10-cp36-cp36m-musllinux_1_1_i686.whl", hash = "sha256:3598893bde43091ee5ca0a6ad20f08a0435e93a69255eeb5f81b85e81e329264"}, + {file = "regex-2021.11.10-cp36-cp36m-musllinux_1_1_ppc64le.whl", hash = "sha256:50a7ddf3d131dc5633dccdb51417e2d1910d25cbcf842115a3a5893509140a3a"}, + {file = "regex-2021.11.10-cp36-cp36m-musllinux_1_1_s390x.whl", hash = "sha256:61600a7ca4bcf78a96a68a27c2ae9389763b5b94b63943d5158f2a377e09d29a"}, + {file = "regex-2021.11.10-cp36-cp36m-musllinux_1_1_x86_64.whl", hash = "sha256:563d5f9354e15e048465061509403f68424fef37d5add3064038c2511c8f5e00"}, + {file = "regex-2021.11.10-cp36-cp36m-win32.whl", hash = "sha256:93a5051fcf5fad72de73b96f07d30bc29665697fb8ecdfbc474f3452c78adcf4"}, + {file = "regex-2021.11.10-cp36-cp36m-win_amd64.whl", hash = "sha256:b483c9d00a565633c87abd0aaf27eb5016de23fed952e054ecc19ce32f6a9e7e"}, + {file = "regex-2021.11.10-cp37-cp37m-macosx_10_9_x86_64.whl", hash = "sha256:fff55f3ce50a3ff63ec8e2a8d3dd924f1941b250b0aac3d3d42b687eeff07a8e"}, + {file = "regex-2021.11.10-cp37-cp37m-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:e32d2a2b02ccbef10145df9135751abea1f9f076e67a4e261b05f24b94219e36"}, + {file = "regex-2021.11.10-cp37-cp37m-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:53db2c6be8a2710b359bfd3d3aa17ba38f8aa72a82309a12ae99d3c0c3dcd74d"}, + {file = "regex-2021.11.10-cp37-cp37m-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:2207ae4f64ad3af399e2d30dde66f0b36ae5c3129b52885f1bffc2f05ec505c8"}, + {file = "regex-2021.11.10-cp37-cp37m-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:d5ca078bb666c4a9d1287a379fe617a6dccd18c3e8a7e6c7e1eb8974330c626a"}, + {file = "regex-2021.11.10-cp37-cp37m-manylinux_2_5_i686.manylinux1_i686.manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:dd33eb9bdcfbabab3459c9ee651d94c842bc8a05fabc95edf4ee0c15a072495e"}, + {file = "regex-2021.11.10-cp37-cp37m-manylinux_2_5_x86_64.manylinux1_x86_64.manylinux_2_12_x86_64.manylinux2010_x86_64.whl", hash = "sha256:05b7d6d7e64efe309972adab77fc2af8907bb93217ec60aa9fe12a0dad35874f"}, + {file = "regex-2021.11.10-cp37-cp37m-musllinux_1_1_aarch64.whl", hash = "sha256:42b50fa6666b0d50c30a990527127334d6b96dd969011e843e726a64011485da"}, + {file = "regex-2021.11.10-cp37-cp37m-musllinux_1_1_i686.whl", hash = "sha256:6e1d2cc79e8dae442b3fa4a26c5794428b98f81389af90623ffcc650ce9f6732"}, + {file = "regex-2021.11.10-cp37-cp37m-musllinux_1_1_ppc64le.whl", hash = "sha256:0416f7399e918c4b0e074a0f66e5191077ee2ca32a0f99d4c187a62beb47aa05"}, + {file = "regex-2021.11.10-cp37-cp37m-musllinux_1_1_s390x.whl", hash = "sha256:ce298e3d0c65bd03fa65ffcc6db0e2b578e8f626d468db64fdf8457731052942"}, + {file = "regex-2021.11.10-cp37-cp37m-musllinux_1_1_x86_64.whl", hash = "sha256:dc07f021ee80510f3cd3af2cad5b6a3b3a10b057521d9e6aaeb621730d320c5a"}, + {file = "regex-2021.11.10-cp37-cp37m-win32.whl", hash = "sha256:e71255ba42567d34a13c03968736c5d39bb4a97ce98188fafb27ce981115beec"}, + {file = "regex-2021.11.10-cp37-cp37m-win_amd64.whl", hash = "sha256:07856afef5ffcc052e7eccf3213317fbb94e4a5cd8177a2caa69c980657b3cb4"}, + {file = "regex-2021.11.10-cp38-cp38-macosx_10_9_x86_64.whl", hash = "sha256:ba05430e819e58544e840a68b03b28b6d328aff2e41579037e8bab7653b37d83"}, + {file = "regex-2021.11.10-cp38-cp38-macosx_11_0_arm64.whl", hash = "sha256:7f301b11b9d214f83ddaf689181051e7f48905568b0c7017c04c06dfd065e244"}, + {file = "regex-2021.11.10-cp38-cp38-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:4aaa4e0705ef2b73dd8e36eeb4c868f80f8393f5f4d855e94025ce7ad8525f50"}, + {file = "regex-2021.11.10-cp38-cp38-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:788aef3549f1924d5c38263104dae7395bf020a42776d5ec5ea2b0d3d85d6646"}, + {file = "regex-2021.11.10-cp38-cp38-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:f8af619e3be812a2059b212064ea7a640aff0568d972cd1b9e920837469eb3cb"}, + {file = "regex-2021.11.10-cp38-cp38-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:85bfa6a5413be0ee6c5c4a663668a2cad2cbecdee367630d097d7823041bdeec"}, + {file = "regex-2021.11.10-cp38-cp38-manylinux_2_5_i686.manylinux1_i686.manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:f23222527b307970e383433daec128d769ff778d9b29343fb3496472dc20dabe"}, + {file = "regex-2021.11.10-cp38-cp38-manylinux_2_5_x86_64.manylinux1_x86_64.manylinux_2_12_x86_64.manylinux2010_x86_64.whl", hash = "sha256:da1a90c1ddb7531b1d5ff1e171b4ee61f6345119be7351104b67ff413843fe94"}, + {file = "regex-2021.11.10-cp38-cp38-musllinux_1_1_aarch64.whl", hash = "sha256:f5be7805e53dafe94d295399cfbe5227f39995a997f4fd8539bf3cbdc8f47ca8"}, + {file = "regex-2021.11.10-cp38-cp38-musllinux_1_1_i686.whl", hash = "sha256:a955b747d620a50408b7fdf948e04359d6e762ff8a85f5775d907ceced715129"}, + {file = "regex-2021.11.10-cp38-cp38-musllinux_1_1_ppc64le.whl", hash = "sha256:139a23d1f5d30db2cc6c7fd9c6d6497872a672db22c4ae1910be22d4f4b2068a"}, + {file = "regex-2021.11.10-cp38-cp38-musllinux_1_1_s390x.whl", hash = "sha256:ca49e1ab99593438b204e00f3970e7a5f70d045267051dfa6b5f4304fcfa1dbf"}, + {file = "regex-2021.11.10-cp38-cp38-musllinux_1_1_x86_64.whl", hash = "sha256:96fc32c16ea6d60d3ca7f63397bff5c75c5a562f7db6dec7d412f7c4d2e78ec0"}, + {file = "regex-2021.11.10-cp38-cp38-win32.whl", hash = "sha256:0617383e2fe465732af4509e61648b77cbe3aee68b6ac8c0b6fe934db90be5cc"}, + {file = "regex-2021.11.10-cp38-cp38-win_amd64.whl", hash = "sha256:a3feefd5e95871872673b08636f96b61ebef62971eab044f5124fb4dea39919d"}, + {file = "regex-2021.11.10-cp39-cp39-macosx_10_9_x86_64.whl", hash = "sha256:f7f325be2804246a75a4f45c72d4ce80d2443ab815063cdf70ee8fb2ca59ee1b"}, + {file = "regex-2021.11.10-cp39-cp39-macosx_11_0_arm64.whl", hash = "sha256:537ca6a3586931b16a85ac38c08cc48f10fc870a5b25e51794c74df843e9966d"}, + {file = "regex-2021.11.10-cp39-cp39-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:eef2afb0fd1747f33f1ee3e209bce1ed582d1896b240ccc5e2697e3275f037c7"}, + {file = "regex-2021.11.10-cp39-cp39-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:432bd15d40ed835a51617521d60d0125867f7b88acf653e4ed994a1f8e4995dc"}, + {file = "regex-2021.11.10-cp39-cp39-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:b43c2b8a330a490daaef5a47ab114935002b13b3f9dc5da56d5322ff218eeadb"}, + {file = "regex-2021.11.10-cp39-cp39-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:962b9a917dd7ceacbe5cd424556914cb0d636001e393b43dc886ba31d2a1e449"}, + {file = "regex-2021.11.10-cp39-cp39-manylinux_2_5_i686.manylinux1_i686.manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:fa8c626d6441e2d04b6ee703ef2d1e17608ad44c7cb75258c09dd42bacdfc64b"}, + {file = "regex-2021.11.10-cp39-cp39-manylinux_2_5_x86_64.manylinux1_x86_64.manylinux_2_12_x86_64.manylinux2010_x86_64.whl", hash = "sha256:3c5fb32cc6077abad3bbf0323067636d93307c9fa93e072771cf9a64d1c0f3ef"}, + {file = "regex-2021.11.10-cp39-cp39-musllinux_1_1_aarch64.whl", hash = "sha256:cd410a1cbb2d297c67d8521759ab2ee3f1d66206d2e4328502a487589a2cb21b"}, + {file = "regex-2021.11.10-cp39-cp39-musllinux_1_1_i686.whl", hash = "sha256:e6096b0688e6e14af6a1b10eaad86b4ff17935c49aa774eac7c95a57a4e8c296"}, + {file = "regex-2021.11.10-cp39-cp39-musllinux_1_1_ppc64le.whl", hash = "sha256:529801a0d58809b60b3531ee804d3e3be4b412c94b5d267daa3de7fadef00f49"}, + {file = "regex-2021.11.10-cp39-cp39-musllinux_1_1_s390x.whl", hash = "sha256:0f594b96fe2e0821d026365f72ac7b4f0b487487fb3d4aaf10dd9d97d88a9737"}, + {file = "regex-2021.11.10-cp39-cp39-musllinux_1_1_x86_64.whl", hash = "sha256:2409b5c9cef7054dde93a9803156b411b677affc84fca69e908b1cb2c540025d"}, + {file = "regex-2021.11.10-cp39-cp39-win32.whl", hash = "sha256:3b5df18db1fccd66de15aa59c41e4f853b5df7550723d26aa6cb7f40e5d9da5a"}, + {file = "regex-2021.11.10-cp39-cp39-win_amd64.whl", hash = "sha256:83ee89483672b11f8952b158640d0c0ff02dc43d9cb1b70c1564b49abe92ce29"}, + {file = "regex-2021.11.10.tar.gz", hash = "sha256:f341ee2df0999bfdf7a95e448075effe0db212a59387de1a70690e4acb03d4c6"}, ] requests = [ {file = "requests-2.26.0-py2.py3-none-any.whl", hash = "sha256:6c1246513ecd5ecd4528a0906f910e8f0f9c6b8ec72030dc9fd154dc1a6efd24"}, @@ -1498,8 +1723,8 @@ toml = [ {file = "toml-0.10.2.tar.gz", hash = "sha256:b3bda1d108d5dd99f4a20d24d9c348e91c4db7ab1b749200bded2f839ccbe68f"}, ] tomli = [ - {file = "tomli-1.2.1-py3-none-any.whl", hash = "sha256:8dd0e9524d6f386271a36b41dbf6c57d8e32fd96fd22b6584679dc569d20899f"}, - {file = "tomli-1.2.1.tar.gz", hash = "sha256:a5b75cb6f3968abb47af1b40c1819dc519ea82bcc065776a866e8d74c5ca9442"}, + {file = "tomli-1.2.2-py3-none-any.whl", hash = "sha256:f04066f68f5554911363063a30b108d2b5a5b1a010aa8b6132af78489fe3aade"}, + {file = "tomli-1.2.2.tar.gz", hash = "sha256:c6ce0015eb38820eaf32b5db832dbc26deb3dd427bd5f6556cf0acac2c214fee"}, ] typed-ast = [ {file = "typed_ast-1.4.3-cp35-cp35m-manylinux1_i686.whl", hash = "sha256:2068531575a125b87a41802130fa7e29f26c09a2833fea68d9a40cf33902eba6"}, @@ -1534,9 +1759,8 @@ typed-ast = [ {file = "typed_ast-1.4.3.tar.gz", hash = "sha256:fb1bbeac803adea29cedd70781399c99138358c26d05fcbd23c13016b7f5ec65"}, ] typing-extensions = [ - {file = "typing_extensions-3.10.0.2-py2-none-any.whl", hash = "sha256:d8226d10bc02a29bcc81df19a26e56a9647f8b0a6d4a83924139f4a8b01f17b7"}, - {file = "typing_extensions-3.10.0.2-py3-none-any.whl", hash = "sha256:f1d25edafde516b146ecd0613dabcc61409817af4766fbbcfb8d1ad4ec441a34"}, - {file = "typing_extensions-3.10.0.2.tar.gz", hash = "sha256:49f75d16ff11f1cd258e1b988ccff82a3ca5570217d7ad8c5f48205dd99a677e"}, + {file = "typing_extensions-4.0.1-py3-none-any.whl", hash = "sha256:7f001e5ac290a0c0401508864c7ec868be4e701886d5b573a9528ed3973d9d3b"}, + {file = "typing_extensions-4.0.1.tar.gz", hash = "sha256:4ca091dea149f945ec56afb48dae714f21e8692ef22a395223bcd328961b6a0e"}, ] tzlocal = [ {file = "tzlocal-2.1-py2.py3-none-any.whl", hash = "sha256:e2cb6c6b5b604af38597403e9852872d7f534962ae2954c7f35efcb1ccacf4a4"}, @@ -1547,16 +1771,16 @@ urllib3 = [ {file = "urllib3-1.26.7.tar.gz", hash = "sha256:4987c65554f7a2dbf30c18fd48778ef124af6fab771a377103da0585e2336ece"}, ] virtualenv = [ - {file = "virtualenv-20.8.1-py2.py3-none-any.whl", hash = "sha256:10062e34c204b5e4ec5f62e6ef2473f8ba76513a9a617e873f1f8fb4a519d300"}, - {file = "virtualenv-20.8.1.tar.gz", hash = "sha256:bcc17f0b3a29670dd777d6f0755a4c04f28815395bca279cdcb213b97199a6b8"}, + {file = "virtualenv-20.10.0-py2.py3-none-any.whl", hash = "sha256:4b02e52a624336eece99c96e3ab7111f469c24ba226a53ec474e8e787b365814"}, + {file = "virtualenv-20.10.0.tar.gz", hash = "sha256:576d05b46eace16a9c348085f7d0dc8ef28713a2cabaa1cf0aea41e8f12c9218"}, ] volttron-client = [ - {file = "volttron-client-0.3.4.tar.gz", hash = "sha256:a426efd8c5d10c5f574b78501082c3ac76714903c29d4fe9c8c9d335abc03d47"}, - {file = "volttron_client-0.3.4-py3-none-any.whl", hash = "sha256:60ba299b75431e5d249ff0cf871e1978ab75e65814a28b6252a5f7612f1ec658"}, + {file = "volttron-client-0.3.8.tar.gz", hash = "sha256:e8ec4953f27f28de851a2ebdd55c89eb5df3163ad594c1c8eec3b4ae035632e9"}, + {file = "volttron_client-0.3.8-py3-none-any.whl", hash = "sha256:75c2fef11b5032b664e47105d1d9f73e67ddab28d8c7febf45cba5109e695a1b"}, ] volttron-utils = [ - {file = "volttron-utils-0.3.2.tar.gz", hash = "sha256:abb35d335ba60ecc06e9e1bbcbb5dd0b3dbc9445e58a366dfaa6071e4ce5ce80"}, - {file = "volttron_utils-0.3.2-py3-none-any.whl", hash = "sha256:240e6f84ef164d7740b0c035b01efbd00c194c1b5b12bb3b73fe28386fff6f5f"}, + {file = "volttron-utils-0.3.4.tar.gz", hash = "sha256:0d499f5ab228d669e5c99b39987af3e0fbba9ad46a4cbb9bc6c5d8e320c28ca2"}, + {file = "volttron_utils-0.3.4-py3-none-any.whl", hash = "sha256:e565dc2535a9cd1dd24006ad61c35627e06d701e923f3ac75f8ed2d8959a715c"}, ] watchdog = [ {file = "watchdog-2.1.6-cp310-cp310-macosx_10_9_universal2.whl", hash = "sha256:9693f35162dc6208d10b10ddf0458cc09ad70c30ba689d9206e02cd836ce28a3"}, @@ -1592,78 +1816,78 @@ xmltodict = [ {file = "xmltodict-0.12.0.tar.gz", hash = "sha256:50d8c638ed7ecb88d90561beedbf720c9b4e851a9fa6c47ebd64e99d166d8a21"}, ] yarl = [ - {file = "yarl-1.7.0-cp310-cp310-macosx_10_9_universal2.whl", hash = "sha256:e35d8230e4b08d86ea65c32450533b906a8267a87b873f2954adeaecede85169"}, - {file = "yarl-1.7.0-cp310-cp310-macosx_10_9_x86_64.whl", hash = "sha256:eb4b3f277880c314e47720b4b6bb2c85114ab3c04c5442c9bc7006b3787904d8"}, - {file = "yarl-1.7.0-cp310-cp310-macosx_11_0_arm64.whl", hash = "sha256:c7015dcedb91d90a138eebdc7e432aec8966e0147ab2a55f2df27b1904fa7291"}, - {file = "yarl-1.7.0-cp310-cp310-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:bb3e478175e15e00d659fb0354a6a8db71a7811a2a5052aed98048bc972e5d2b"}, - {file = "yarl-1.7.0-cp310-cp310-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:8b8c409aa3a7966647e7c1c524846b362a6bcbbe120bf8a176431f940d2b9a2e"}, - {file = "yarl-1.7.0-cp310-cp310-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:b22ea41c7e98170474a01e3eded1377d46b2dfaef45888a0005c683eaaa49285"}, - {file = "yarl-1.7.0-cp310-cp310-manylinux_2_5_i686.manylinux1_i686.manylinux_2_12_i686.manylinux2010_i686.whl", hash = "sha256:a7dfc46add4cfe5578013dbc4127893edc69fe19132d2836ff2f6e49edc5ecd6"}, - {file = "yarl-1.7.0-cp310-cp310-manylinux_2_5_x86_64.manylinux1_x86_64.manylinux_2_12_x86_64.manylinux2010_x86_64.whl", hash = "sha256:82ff6f85f67500a4f74885d81659cd270eb24dfe692fe44e622b8a2fd57e7279"}, - {file = "yarl-1.7.0-cp310-cp310-musllinux_1_1_aarch64.whl", hash = "sha256:f3cd2158b2ed0fb25c6811adfdcc47224efe075f2d68a750071dacc03a7a66e4"}, - {file = "yarl-1.7.0-cp310-cp310-musllinux_1_1_i686.whl", hash = "sha256:59c0f13f9592820c51280d1cf811294d753e4a18baf90f0139d1dc93d4b6fc5f"}, - {file = "yarl-1.7.0-cp310-cp310-musllinux_1_1_ppc64le.whl", hash = "sha256:7f7655ad83d1a8afa48435a449bf2f3009293da1604f5dd95b5ddcf5f673bd69"}, - {file = "yarl-1.7.0-cp310-cp310-musllinux_1_1_s390x.whl", hash = "sha256:aa9f0d9b62d15182341b3e9816582f46182cab91c1a57b2d308b9a3c4e2c4f78"}, - {file = "yarl-1.7.0-cp310-cp310-musllinux_1_1_x86_64.whl", hash = "sha256:fdd1b90c225a653b1bd1c0cae8edf1957892b9a09c8bf7ee6321eeb8208eac0f"}, - {file = "yarl-1.7.0-cp310-cp310-win32.whl", hash = "sha256:7c8d0bb76eabc5299db203e952ec55f8f4c53f08e0df4285aac8c92bd9e12675"}, - {file = "yarl-1.7.0-cp310-cp310-win_amd64.whl", hash = "sha256:622a36fa779efb4ff9eff5fe52730ff17521431379851a31e040958fc251670c"}, - {file = "yarl-1.7.0-cp36-cp36m-macosx_10_9_x86_64.whl", hash = "sha256:3d461b7a8e139b9e4b41f62eb417ffa0b98d1c46d4caf14c845e6a3b349c0bb1"}, - {file = "yarl-1.7.0-cp36-cp36m-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:81cfacdd1e40bc931b5519499342efa388d24d262c30a3d31187bfa04f4a7001"}, - {file = "yarl-1.7.0-cp36-cp36m-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:821b978f2152be7695d4331ef0621d207aedf9bbd591ba23a63412a3efc29a01"}, - {file = "yarl-1.7.0-cp36-cp36m-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:b64bd24c8c9a487f4a12260dc26732bf41028816dbf0c458f17864fbebdb3131"}, - {file = "yarl-1.7.0-cp36-cp36m-manylinux_2_5_i686.manylinux1_i686.manylinux_2_12_i686.manylinux2010_i686.whl", hash = "sha256:98c9ddb92b60a83c21be42c776d3d9d5ec632a762a094c41bda37b7dfbd2cd83"}, - {file = "yarl-1.7.0-cp36-cp36m-manylinux_2_5_x86_64.manylinux1_x86_64.manylinux_2_12_x86_64.manylinux2010_x86_64.whl", hash = "sha256:a532d75ca74431c053a88a802e161fb3d651b8bf5821a3440bc3616e38754583"}, - {file = "yarl-1.7.0-cp36-cp36m-musllinux_1_1_aarch64.whl", hash = "sha256:053e09817eafb892e94e172d05406c1b3a22a93bc68f6eff5198363a3d764459"}, - {file = "yarl-1.7.0-cp36-cp36m-musllinux_1_1_i686.whl", hash = "sha256:98c51f02d542945d306c8e934aa2c1e66ba5e9c1c86b5bf37f3a51c8a747067e"}, - {file = "yarl-1.7.0-cp36-cp36m-musllinux_1_1_ppc64le.whl", hash = "sha256:15ec41a5a5fdb7bace6d7b16701f9440007a82734f69127c0fbf6d87e10f4a1e"}, - {file = "yarl-1.7.0-cp36-cp36m-musllinux_1_1_s390x.whl", hash = "sha256:a7f08819dba1e1255d6991ed37448a1bf4b1352c004bcd899b9da0c47958513d"}, - {file = "yarl-1.7.0-cp36-cp36m-musllinux_1_1_x86_64.whl", hash = "sha256:8e3ffab21db0542ffd1887f3b9575ddd58961f2cf61429cb6458afc00c4581e0"}, - {file = "yarl-1.7.0-cp36-cp36m-win32.whl", hash = "sha256:50127634f519b2956005891507e3aa4ac345f66a7ea7bbc2d7dcba7401f41898"}, - {file = "yarl-1.7.0-cp36-cp36m-win_amd64.whl", hash = "sha256:36ec44f15193f6d5288d42ebb8e751b967ebdfb72d6830983838d45ab18edb4f"}, - {file = "yarl-1.7.0-cp37-cp37m-macosx_10_9_x86_64.whl", hash = "sha256:ec1b5a25a25c880c976d0bb3d107def085bb08dbb3db7f4442e0a2b980359d24"}, - {file = "yarl-1.7.0-cp37-cp37m-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:b36f5a63c891f813c6f04ef19675b382efc190fd5ce7e10ab19386d2548bca06"}, - {file = "yarl-1.7.0-cp37-cp37m-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:38173b8c3a29945e7ecade9a3f6ff39581eee8201338ee6a2c8882db5df3e806"}, - {file = "yarl-1.7.0-cp37-cp37m-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:8ba402f32184f0b405fb281b93bd0d8ab7e3257735b57b62a6ed2e94cdf4fe50"}, - {file = "yarl-1.7.0-cp37-cp37m-manylinux_2_5_i686.manylinux1_i686.manylinux_2_12_i686.manylinux2010_i686.whl", hash = "sha256:be52bc5208d767cdd8308a9e93059b3b36d1e048fecbea0e0346d0d24a76adc0"}, - {file = "yarl-1.7.0-cp37-cp37m-manylinux_2_5_x86_64.manylinux1_x86_64.manylinux_2_12_x86_64.manylinux2010_x86_64.whl", hash = "sha256:08c2044a956f4ef30405f2f433ce77f1f57c2c773bf81ae43201917831044d5a"}, - {file = "yarl-1.7.0-cp37-cp37m-musllinux_1_1_aarch64.whl", hash = "sha256:484d61c047c45670ef5967653a1d0783e232c54bf9dd786a7737036828fa8d54"}, - {file = "yarl-1.7.0-cp37-cp37m-musllinux_1_1_i686.whl", hash = "sha256:b7de92a4af85cfcaf4081f8aa6165b1d63ee5de150af3ee85f954145f93105a7"}, - {file = "yarl-1.7.0-cp37-cp37m-musllinux_1_1_ppc64le.whl", hash = "sha256:376e41775aab79c5575534924a386c8e0f1a5d91db69fc6133fd27a489bcaf10"}, - {file = "yarl-1.7.0-cp37-cp37m-musllinux_1_1_s390x.whl", hash = "sha256:8a8b10d0e7bac154f959b709fcea593cda527b234119311eb950096653816a86"}, - {file = "yarl-1.7.0-cp37-cp37m-musllinux_1_1_x86_64.whl", hash = "sha256:f46cd4c43e6175030e2a56def8f1d83b64e6706eeb2bb9ab0ef4756f65eab23f"}, - {file = "yarl-1.7.0-cp37-cp37m-win32.whl", hash = "sha256:b28cfb46140efe1a6092b8c5c4994a1fe70dc83c38fbcea4992401e0c6fb9cce"}, - {file = "yarl-1.7.0-cp37-cp37m-win_amd64.whl", hash = "sha256:9624154ec9c02a776802da1086eed7f5034bd1971977f5146233869c2ac80297"}, - {file = "yarl-1.7.0-cp38-cp38-macosx_10_9_universal2.whl", hash = "sha256:69945d13e1bbf81784a9bc48824feb9cd66491e6a503d4e83f6cd7c7cc861361"}, - {file = "yarl-1.7.0-cp38-cp38-macosx_10_9_x86_64.whl", hash = "sha256:46a742ed9e363bd01be64160ce7520e92e11989bd4cb224403cfd31c101cc83d"}, - {file = "yarl-1.7.0-cp38-cp38-macosx_11_0_arm64.whl", hash = "sha256:cb4ff1ac7cb4500f43581b3f4cbd627d702143aa6be1fdc1fa3ebffaf4dc1be5"}, - {file = "yarl-1.7.0-cp38-cp38-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:3ad51e17cd65ea3debb0e10f0120cf8dd987c741fe423ed2285087368090b33d"}, - {file = "yarl-1.7.0-cp38-cp38-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:7e37786ea89a5d3ffbbf318ea9790926f8dfda83858544f128553c347ad143c6"}, - {file = "yarl-1.7.0-cp38-cp38-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:c63c1e208f800daad71715786bfeb1cecdc595d87e2e9b1cd234fd6e597fd71d"}, - {file = "yarl-1.7.0-cp38-cp38-manylinux_2_5_i686.manylinux1_i686.manylinux_2_12_i686.manylinux2010_i686.whl", hash = "sha256:91cbe24300c11835ef186436363352b3257db7af165e0a767f4f17aa25761388"}, - {file = "yarl-1.7.0-cp38-cp38-manylinux_2_5_x86_64.manylinux1_x86_64.manylinux_2_12_x86_64.manylinux2010_x86_64.whl", hash = "sha256:e510dbec7c59d32eaa61ffa48173d5e3d7170a67f4a03e8f5e2e9e3971aca622"}, - {file = "yarl-1.7.0-cp38-cp38-musllinux_1_1_aarch64.whl", hash = "sha256:3def6e681cc02397e5d8141ee97b41d02932b2bcf0fb34532ad62855eab7c60e"}, - {file = "yarl-1.7.0-cp38-cp38-musllinux_1_1_i686.whl", hash = "sha256:263c81b94e6431942b27f6f671fa62f430a0a5c14bb255f2ab69eeb9b2b66ff7"}, - {file = "yarl-1.7.0-cp38-cp38-musllinux_1_1_ppc64le.whl", hash = "sha256:e78c91faefe88d601ddd16e3882918dbde20577a2438e2320f8239c8b7507b8f"}, - {file = "yarl-1.7.0-cp38-cp38-musllinux_1_1_s390x.whl", hash = "sha256:22b2430c49713bfb2f0a0dd4a8d7aab218b28476ba86fd1c78ad8899462cbcf2"}, - {file = "yarl-1.7.0-cp38-cp38-musllinux_1_1_x86_64.whl", hash = "sha256:2e7ad9db939082f5d0b9269cfd92c025cb8f2fbbb1f1b9dc5a393c639db5bd92"}, - {file = "yarl-1.7.0-cp38-cp38-win32.whl", hash = "sha256:3a31e4a8dcb1beaf167b7e7af61b88cb961b220db8d3ba1c839723630e57eef7"}, - {file = "yarl-1.7.0-cp38-cp38-win_amd64.whl", hash = "sha256:d579957439933d752358c6a300c93110f84aae67b63dd0c19dde6ecbf4056f6b"}, - {file = "yarl-1.7.0-cp39-cp39-macosx_10_9_universal2.whl", hash = "sha256:87721b549505a546eb003252185103b5ec8147de6d3ad3714d148a5a67b6fe53"}, - {file = "yarl-1.7.0-cp39-cp39-macosx_10_9_x86_64.whl", hash = "sha256:a1fa866fa24d9f4108f9e58ea8a2135655419885cdb443e36b39a346e1181532"}, - {file = "yarl-1.7.0-cp39-cp39-macosx_11_0_arm64.whl", hash = "sha256:1d3b8449dfedfe94eaff2b77954258b09b24949f6818dfa444b05dbb05ae1b7e"}, - {file = "yarl-1.7.0-cp39-cp39-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:db2372e350794ce8b9f810feb094c606b7e0e4aa6807141ac4fadfe5ddd75bb0"}, - {file = "yarl-1.7.0-cp39-cp39-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:a06d9d0b9a97fa99b84fee71d9dd11e69e21ac8a27229089f07b5e5e50e8d63c"}, - {file = "yarl-1.7.0-cp39-cp39-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:a3455c2456d6307bcfa80bc1157b8603f7d93573291f5bdc7144489ca0df4628"}, - {file = "yarl-1.7.0-cp39-cp39-manylinux_2_5_i686.manylinux1_i686.manylinux_2_12_i686.manylinux2010_i686.whl", hash = "sha256:d30d67e3486aea61bb2cbf7cf81385364c2e4f7ce7469a76ed72af76a5cdfe6b"}, - {file = "yarl-1.7.0-cp39-cp39-manylinux_2_5_x86_64.manylinux1_x86_64.manylinux_2_12_x86_64.manylinux2010_x86_64.whl", hash = "sha256:c18a4b286e8d780c3a40c31d7b79836aa93b720f71d5743f20c08b7e049ca073"}, - {file = "yarl-1.7.0-cp39-cp39-musllinux_1_1_aarch64.whl", hash = "sha256:d54c925396e7891666cabc0199366ca55b27d003393465acef63fd29b8b7aa92"}, - {file = "yarl-1.7.0-cp39-cp39-musllinux_1_1_i686.whl", hash = "sha256:64773840952de17851a1c7346ad7f71688c77e74248d1f0bc230e96680f84028"}, - {file = "yarl-1.7.0-cp39-cp39-musllinux_1_1_ppc64le.whl", hash = "sha256:acbf1756d9dc7cd0ae943d883be72e84e04396f6c2ff93a6ddeca929d562039f"}, - {file = "yarl-1.7.0-cp39-cp39-musllinux_1_1_s390x.whl", hash = "sha256:2e48f27936aa838939c798f466c851ba4ae79e347e8dfce43b009c64b930df12"}, - {file = "yarl-1.7.0-cp39-cp39-musllinux_1_1_x86_64.whl", hash = "sha256:1beef4734ca1ad40a9d8c6b20a76ab46e3a2ed09f38561f01e4aa2ea82cafcef"}, - {file = "yarl-1.7.0-cp39-cp39-win32.whl", hash = "sha256:8ee78c9a5f3c642219d4607680a4693b59239c27a3aa608b64ef79ddc9698039"}, - {file = "yarl-1.7.0-cp39-cp39-win_amd64.whl", hash = "sha256:d750503682605088a14d29a4701548c15c510da4f13c8b17409c4097d5b04c52"}, - {file = "yarl-1.7.0.tar.gz", hash = "sha256:8e7ebaf62e19c2feb097ffb7c94deb0f0c9fab52590784c8cd679d30ab009162"}, + {file = "yarl-1.7.2-cp310-cp310-macosx_10_9_universal2.whl", hash = "sha256:f2a8508f7350512434e41065684076f640ecce176d262a7d54f0da41d99c5a95"}, + {file = "yarl-1.7.2-cp310-cp310-macosx_10_9_x86_64.whl", hash = "sha256:da6df107b9ccfe52d3a48165e48d72db0eca3e3029b5b8cb4fe6ee3cb870ba8b"}, + {file = "yarl-1.7.2-cp310-cp310-macosx_11_0_arm64.whl", hash = "sha256:a1d0894f238763717bdcfea74558c94e3bc34aeacd3351d769460c1a586a8b05"}, + {file = "yarl-1.7.2-cp310-cp310-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:dfe4b95b7e00c6635a72e2d00b478e8a28bfb122dc76349a06e20792eb53a523"}, + {file = "yarl-1.7.2-cp310-cp310-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:c145ab54702334c42237a6c6c4cc08703b6aa9b94e2f227ceb3d477d20c36c63"}, + {file = "yarl-1.7.2-cp310-cp310-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:1ca56f002eaf7998b5fcf73b2421790da9d2586331805f38acd9997743114e98"}, + {file = "yarl-1.7.2-cp310-cp310-manylinux_2_5_i686.manylinux1_i686.manylinux_2_12_i686.manylinux2010_i686.whl", hash = "sha256:1d3d5ad8ea96bd6d643d80c7b8d5977b4e2fb1bab6c9da7322616fd26203d125"}, + {file = "yarl-1.7.2-cp310-cp310-manylinux_2_5_x86_64.manylinux1_x86_64.manylinux_2_12_x86_64.manylinux2010_x86_64.whl", hash = "sha256:167ab7f64e409e9bdd99333fe8c67b5574a1f0495dcfd905bc7454e766729b9e"}, + {file = "yarl-1.7.2-cp310-cp310-musllinux_1_1_aarch64.whl", hash = "sha256:95a1873b6c0dd1c437fb3bb4a4aaa699a48c218ac7ca1e74b0bee0ab16c7d60d"}, + {file = "yarl-1.7.2-cp310-cp310-musllinux_1_1_i686.whl", hash = "sha256:6152224d0a1eb254f97df3997d79dadd8bb2c1a02ef283dbb34b97d4f8492d23"}, + {file = "yarl-1.7.2-cp310-cp310-musllinux_1_1_ppc64le.whl", hash = "sha256:5bb7d54b8f61ba6eee541fba4b83d22b8a046b4ef4d8eb7f15a7e35db2e1e245"}, + {file = "yarl-1.7.2-cp310-cp310-musllinux_1_1_s390x.whl", hash = "sha256:9c1f083e7e71b2dd01f7cd7434a5f88c15213194df38bc29b388ccdf1492b739"}, + {file = "yarl-1.7.2-cp310-cp310-musllinux_1_1_x86_64.whl", hash = "sha256:f44477ae29025d8ea87ec308539f95963ffdc31a82f42ca9deecf2d505242e72"}, + {file = "yarl-1.7.2-cp310-cp310-win32.whl", hash = "sha256:cff3ba513db55cc6a35076f32c4cdc27032bd075c9faef31fec749e64b45d26c"}, + {file = "yarl-1.7.2-cp310-cp310-win_amd64.whl", hash = "sha256:c9c6d927e098c2d360695f2e9d38870b2e92e0919be07dbe339aefa32a090265"}, + {file = "yarl-1.7.2-cp36-cp36m-macosx_10_9_x86_64.whl", hash = "sha256:9b4c77d92d56a4c5027572752aa35082e40c561eec776048330d2907aead891d"}, + {file = "yarl-1.7.2-cp36-cp36m-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:c01a89a44bb672c38f42b49cdb0ad667b116d731b3f4c896f72302ff77d71656"}, + {file = "yarl-1.7.2-cp36-cp36m-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:c19324a1c5399b602f3b6e7db9478e5b1adf5cf58901996fc973fe4fccd73eed"}, + {file = "yarl-1.7.2-cp36-cp36m-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:3abddf0b8e41445426d29f955b24aeecc83fa1072be1be4e0d194134a7d9baee"}, + {file = "yarl-1.7.2-cp36-cp36m-manylinux_2_5_i686.manylinux1_i686.manylinux_2_12_i686.manylinux2010_i686.whl", hash = "sha256:6a1a9fe17621af43e9b9fcea8bd088ba682c8192d744b386ee3c47b56eaabb2c"}, + {file = "yarl-1.7.2-cp36-cp36m-manylinux_2_5_x86_64.manylinux1_x86_64.manylinux_2_12_x86_64.manylinux2010_x86_64.whl", hash = "sha256:8b0915ee85150963a9504c10de4e4729ae700af11df0dc5550e6587ed7891e92"}, + {file = "yarl-1.7.2-cp36-cp36m-musllinux_1_1_aarch64.whl", hash = "sha256:29e0656d5497733dcddc21797da5a2ab990c0cb9719f1f969e58a4abac66234d"}, + {file = "yarl-1.7.2-cp36-cp36m-musllinux_1_1_i686.whl", hash = "sha256:bf19725fec28452474d9887a128e98dd67eee7b7d52e932e6949c532d820dc3b"}, + {file = "yarl-1.7.2-cp36-cp36m-musllinux_1_1_ppc64le.whl", hash = "sha256:d6f3d62e16c10e88d2168ba2d065aa374e3c538998ed04996cd373ff2036d64c"}, + {file = "yarl-1.7.2-cp36-cp36m-musllinux_1_1_s390x.whl", hash = "sha256:ac10bbac36cd89eac19f4e51c032ba6b412b3892b685076f4acd2de18ca990aa"}, + {file = "yarl-1.7.2-cp36-cp36m-musllinux_1_1_x86_64.whl", hash = "sha256:aa32aaa97d8b2ed4e54dc65d241a0da1c627454950f7d7b1f95b13985afd6c5d"}, + {file = "yarl-1.7.2-cp36-cp36m-win32.whl", hash = "sha256:87f6e082bce21464857ba58b569370e7b547d239ca22248be68ea5d6b51464a1"}, + {file = "yarl-1.7.2-cp36-cp36m-win_amd64.whl", hash = "sha256:ac35ccde589ab6a1870a484ed136d49a26bcd06b6a1c6397b1967ca13ceb3913"}, + {file = "yarl-1.7.2-cp37-cp37m-macosx_10_9_x86_64.whl", hash = "sha256:a467a431a0817a292121c13cbe637348b546e6ef47ca14a790aa2fa8cc93df63"}, + {file = "yarl-1.7.2-cp37-cp37m-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:6ab0c3274d0a846840bf6c27d2c60ba771a12e4d7586bf550eefc2df0b56b3b4"}, + {file = "yarl-1.7.2-cp37-cp37m-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:d260d4dc495c05d6600264a197d9d6f7fc9347f21d2594926202fd08cf89a8ba"}, + {file = "yarl-1.7.2-cp37-cp37m-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:fc4dd8b01a8112809e6b636b00f487846956402834a7fd59d46d4f4267181c41"}, + {file = "yarl-1.7.2-cp37-cp37m-manylinux_2_5_i686.manylinux1_i686.manylinux_2_12_i686.manylinux2010_i686.whl", hash = "sha256:c1164a2eac148d85bbdd23e07dfcc930f2e633220f3eb3c3e2a25f6148c2819e"}, + {file = "yarl-1.7.2-cp37-cp37m-manylinux_2_5_x86_64.manylinux1_x86_64.manylinux_2_12_x86_64.manylinux2010_x86_64.whl", hash = "sha256:67e94028817defe5e705079b10a8438b8cb56e7115fa01640e9c0bb3edf67332"}, + {file = "yarl-1.7.2-cp37-cp37m-musllinux_1_1_aarch64.whl", hash = "sha256:89ccbf58e6a0ab89d487c92a490cb5660d06c3a47ca08872859672f9c511fc52"}, + {file = "yarl-1.7.2-cp37-cp37m-musllinux_1_1_i686.whl", hash = "sha256:8cce6f9fa3df25f55521fbb5c7e4a736683148bcc0c75b21863789e5185f9185"}, + {file = "yarl-1.7.2-cp37-cp37m-musllinux_1_1_ppc64le.whl", hash = "sha256:211fcd65c58bf250fb994b53bc45a442ddc9f441f6fec53e65de8cba48ded986"}, + {file = "yarl-1.7.2-cp37-cp37m-musllinux_1_1_s390x.whl", hash = "sha256:c10ea1e80a697cf7d80d1ed414b5cb8f1eec07d618f54637067ae3c0334133c4"}, + {file = "yarl-1.7.2-cp37-cp37m-musllinux_1_1_x86_64.whl", hash = "sha256:52690eb521d690ab041c3919666bea13ab9fbff80d615ec16fa81a297131276b"}, + {file = "yarl-1.7.2-cp37-cp37m-win32.whl", hash = "sha256:695ba021a9e04418507fa930d5f0704edbce47076bdcfeeaba1c83683e5649d1"}, + {file = "yarl-1.7.2-cp37-cp37m-win_amd64.whl", hash = "sha256:c17965ff3706beedafd458c452bf15bac693ecd146a60a06a214614dc097a271"}, + {file = "yarl-1.7.2-cp38-cp38-macosx_10_9_universal2.whl", hash = "sha256:fce78593346c014d0d986b7ebc80d782b7f5e19843ca798ed62f8e3ba8728576"}, + {file = "yarl-1.7.2-cp38-cp38-macosx_10_9_x86_64.whl", hash = "sha256:c2a1ac41a6aa980db03d098a5531f13985edcb451bcd9d00670b03129922cd0d"}, + {file = "yarl-1.7.2-cp38-cp38-macosx_11_0_arm64.whl", hash = "sha256:39d5493c5ecd75c8093fa7700a2fb5c94fe28c839c8e40144b7ab7ccba6938c8"}, + {file = "yarl-1.7.2-cp38-cp38-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:1eb6480ef366d75b54c68164094a6a560c247370a68c02dddb11f20c4c6d3c9d"}, + {file = "yarl-1.7.2-cp38-cp38-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:5ba63585a89c9885f18331a55d25fe81dc2d82b71311ff8bd378fc8004202ff6"}, + {file = "yarl-1.7.2-cp38-cp38-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:e39378894ee6ae9f555ae2de332d513a5763276a9265f8e7cbaeb1b1ee74623a"}, + {file = "yarl-1.7.2-cp38-cp38-manylinux_2_5_i686.manylinux1_i686.manylinux_2_12_i686.manylinux2010_i686.whl", hash = "sha256:c0910c6b6c31359d2f6184828888c983d54d09d581a4a23547a35f1d0b9484b1"}, + {file = "yarl-1.7.2-cp38-cp38-manylinux_2_5_x86_64.manylinux1_x86_64.manylinux_2_12_x86_64.manylinux2010_x86_64.whl", hash = "sha256:6feca8b6bfb9eef6ee057628e71e1734caf520a907b6ec0d62839e8293e945c0"}, + {file = "yarl-1.7.2-cp38-cp38-musllinux_1_1_aarch64.whl", hash = "sha256:8300401dc88cad23f5b4e4c1226f44a5aa696436a4026e456fe0e5d2f7f486e6"}, + {file = "yarl-1.7.2-cp38-cp38-musllinux_1_1_i686.whl", hash = "sha256:788713c2896f426a4e166b11f4ec538b5736294ebf7d5f654ae445fd44270832"}, + {file = "yarl-1.7.2-cp38-cp38-musllinux_1_1_ppc64le.whl", hash = "sha256:fd547ec596d90c8676e369dd8a581a21227fe9b4ad37d0dc7feb4ccf544c2d59"}, + {file = "yarl-1.7.2-cp38-cp38-musllinux_1_1_s390x.whl", hash = "sha256:737e401cd0c493f7e3dd4db72aca11cfe069531c9761b8ea474926936b3c57c8"}, + {file = "yarl-1.7.2-cp38-cp38-musllinux_1_1_x86_64.whl", hash = "sha256:baf81561f2972fb895e7844882898bda1eef4b07b5b385bcd308d2098f1a767b"}, + {file = "yarl-1.7.2-cp38-cp38-win32.whl", hash = "sha256:ede3b46cdb719c794427dcce9d8beb4abe8b9aa1e97526cc20de9bd6583ad1ef"}, + {file = "yarl-1.7.2-cp38-cp38-win_amd64.whl", hash = "sha256:cc8b7a7254c0fc3187d43d6cb54b5032d2365efd1df0cd1749c0c4df5f0ad45f"}, + {file = "yarl-1.7.2-cp39-cp39-macosx_10_9_universal2.whl", hash = "sha256:580c1f15500e137a8c37053e4cbf6058944d4c114701fa59944607505c2fe3a0"}, + {file = "yarl-1.7.2-cp39-cp39-macosx_10_9_x86_64.whl", hash = "sha256:3ec1d9a0d7780416e657f1e405ba35ec1ba453a4f1511eb8b9fbab81cb8b3ce1"}, + {file = "yarl-1.7.2-cp39-cp39-macosx_11_0_arm64.whl", hash = "sha256:3bf8cfe8856708ede6a73907bf0501f2dc4e104085e070a41f5d88e7faf237f3"}, + {file = "yarl-1.7.2-cp39-cp39-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:1be4bbb3d27a4e9aa5f3df2ab61e3701ce8fcbd3e9846dbce7c033a7e8136746"}, + {file = "yarl-1.7.2-cp39-cp39-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:534b047277a9a19d858cde163aba93f3e1677d5acd92f7d10ace419d478540de"}, + {file = "yarl-1.7.2-cp39-cp39-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:c6ddcd80d79c96eb19c354d9dca95291589c5954099836b7c8d29278a7ec0bda"}, + {file = "yarl-1.7.2-cp39-cp39-manylinux_2_5_i686.manylinux1_i686.manylinux_2_12_i686.manylinux2010_i686.whl", hash = "sha256:9bfcd43c65fbb339dc7086b5315750efa42a34eefad0256ba114cd8ad3896f4b"}, + {file = "yarl-1.7.2-cp39-cp39-manylinux_2_5_x86_64.manylinux1_x86_64.manylinux_2_12_x86_64.manylinux2010_x86_64.whl", hash = "sha256:f64394bd7ceef1237cc604b5a89bf748c95982a84bcd3c4bbeb40f685c810794"}, + {file = "yarl-1.7.2-cp39-cp39-musllinux_1_1_aarch64.whl", hash = "sha256:044daf3012e43d4b3538562da94a88fb12a6490652dbc29fb19adfa02cf72eac"}, + {file = "yarl-1.7.2-cp39-cp39-musllinux_1_1_i686.whl", hash = "sha256:368bcf400247318382cc150aaa632582d0780b28ee6053cd80268c7e72796dec"}, + {file = "yarl-1.7.2-cp39-cp39-musllinux_1_1_ppc64le.whl", hash = "sha256:bab827163113177aee910adb1f48ff7af31ee0289f434f7e22d10baf624a6dfe"}, + {file = "yarl-1.7.2-cp39-cp39-musllinux_1_1_s390x.whl", hash = "sha256:0cba38120db72123db7c58322fa69e3c0efa933040ffb586c3a87c063ec7cae8"}, + {file = "yarl-1.7.2-cp39-cp39-musllinux_1_1_x86_64.whl", hash = "sha256:59218fef177296451b23214c91ea3aba7858b4ae3306dde120224cfe0f7a6ee8"}, + {file = "yarl-1.7.2-cp39-cp39-win32.whl", hash = "sha256:1edc172dcca3f11b38a9d5c7505c83c1913c0addc99cd28e993efeaafdfaa18d"}, + {file = "yarl-1.7.2-cp39-cp39-win_amd64.whl", hash = "sha256:797c2c412b04403d2da075fb93c123df35239cd7b4cc4e0cd9e5839b73f52c58"}, + {file = "yarl-1.7.2.tar.gz", hash = "sha256:45399b46d60c253327a460e99856752009fcee5f5d3c80b2f7c0cae1c38d56dd"}, ] zmq = [ {file = "zmq-0.0.0.tar.gz", hash = "sha256:6b1a1de53338646e8c8405803cffb659e8eb7bb02fff4c9be62a7acfac8370c9"}, diff --git a/pyproject.toml b/pyproject.toml index 3f39fdf..64002f8 100644 --- a/pyproject.toml +++ b/pyproject.toml @@ -10,9 +10,8 @@ black = "^21.5b2" mypy = "^0.812" pre-commit = "^2.13.0" volttron-client = "^0.3.4" +openleadr = "^0.5.24" -# local dependencies -openleadr= {path="./openleadr", develop=true} [tool.poetry.dev-dependencies] pytest = "^6.2.4" diff --git a/requirements.txt b/requirements.txt new file mode 100644 index 0000000..27803e4 --- /dev/null +++ b/requirements.txt @@ -0,0 +1,1 @@ +openleadr==0.5.24 diff --git a/scripts/__init__.py b/scripts/__init__.py new file mode 100644 index 0000000..e69de29 diff --git a/scripts/run_ipkeys_openadr_client.py b/scripts/run_ipkeys_openadr_client.py new file mode 100644 index 0000000..a365235 --- /dev/null +++ b/scripts/run_ipkeys_openadr_client.py @@ -0,0 +1,51 @@ +import logging +import asyncio + +from pprint import pformat +from openleadr.enums import OPT +from volttron_openadr_ven.volttron_openadr_client import IPKeysClient +from volttron_openadr_ven.constants import ( + VEN_NAME, + VTN_URL, + DEBUG, + DISABLE_SIGNATURE, + CERT, + KEY, +) + +logging.basicConfig(level=logging.INFO) +_log = logging.getLogger(__name__) +__version__ = "1.0" + + +def handle_event(event): + _log.info(f"Received event:\n {pformat(event)}") + return OPT.OPT_IN + + +config = { + # Change the paths to the location of the cert and key to the paths of the cert and key in your local machine + CERT: "secret/TEST_RSA_VEN_210923215148_certs/TEST_RSA_VEN_210923215148_cert.pem", + KEY: "secret/TEST_RSA_VEN_210923215148_certs/TEST_RSA_VEN_210923215148_privkey.pem", + # Do not change these values + VEN_NAME: "PNNLVEN", + VTN_URL: "https://eiss2demo.ipkeys.com/oadr2/OpenADR2/Simple/2.0b", + DEBUG: True, + DISABLE_SIGNATURE: True, +} + + +ven = IPKeysClient( + ven_name=config[VEN_NAME], + vtn_url=config[VTN_URL], + debug=True, + disable_signature=True, + cert=config[CERT], + key=config[KEY], +) +ven.add_handler("on_event", handle_event) + + +loop = asyncio.get_event_loop() +loop.create_task(ven.run()) +loop.run_forever() diff --git a/setup.cfg b/setup.cfg new file mode 100644 index 0000000..4930738 --- /dev/null +++ b/setup.cfg @@ -0,0 +1,43 @@ +[metadata] +name = volttron-openadr-ven +description = OpenAdr VEN Implementation +description-file = README.md +description-content-type = text/markdown +author = The VOLTTRON Team +author-email = [email protected] +license = Apache 2 +keywords = volttron, iot platform, building automation, ADR, openadr, Automated Demand Response +project_urls = + Bug Reports = https://github.com/VOLTTRON/volttron-openadr-ven/issues + Source = https://github.com/VOLTTRON/volttron-openadr-ven/ + +# Classifiers help users find your project by categorizing it. +# +# For a list of valid classifiers, see https://pypi.org/classifiers/ +classifiers= + # How mature is this project? Common values are + # 3 - Alpha + # 4 - Beta + # 5 - Production/Stable + Development Status :: 3 - Alpha + # Indicate who your project is intended for + Intended Audience :: Developers + Intended Audience :: Science/Research + # Pick your license as you wish + License :: OSI Approved :: Apache Software License + # Specify the Python versions you support here. In particular, ensure + # that you indicate whether you support Python 2, Python 3 or both. + # These classifiers are *not* checked by 'pip install'. See instead + # 'python_requires' below. + Programming Language :: Python :: 3.7 + Programming Language :: Python :: 3.8 + Programming Language :: Python :: 3.9 + +[files] +packages = + volttron_openadr_ven + +;[entry_points] +;console_scripts = +; volttron = volttron.server.__main__:main + diff --git a/setup.py b/setup.py new file mode 100644 index 0000000..8e2caba --- /dev/null +++ b/setup.py @@ -0,0 +1,43 @@ +# Always prefer setuptools over distutils +from setuptools import setup, find_packages + +setup( + setup_requires=["pbr"], + pbr=True, + # Specify which Python versions you support. In contrast to the + # 'Programming Language' classifiers above, 'pip install' will check this + # and refuse to install the project if the version does not match. If you + # do not support Python 2, you can simplify this to '>=3.5' or similar, see + # https://packaging.python.org/guides/distributing-packages-using-setuptools/#python-requires + python_requires=">=3.7, <4", + + # For an analysis of "install_requires" vs pip's requirements files see: + # https://packaging.python.org/en/latest/requirements.html + install_requires=[], # Optional + # List additional groups of dependencies here (e.g. development + # dependencies). Users will be able to install these using the "extras" + # syntax, for example: + # + # $ pip install sampleproject[dev] + # + # Similar to `install_requires` above, these must be valid existing + # projects. + extras_require={"dev": []}, # Optional + # If there are data files included in your packages that need to be + # installed, specify them here. + # + # Sometimes you’ll want to use packages that are properly arranged with + # setuptools, but are not published to PyPI. In those cases, you can specify + # a list of one or more dependency_links URLs where the package can + # be downloaded, along with some additional hints, and setuptools + # will find and install the package correctly. + # see https://python-packaging.readthedocs.io/en/latest/dependencies.html#packages-not-on-pypi + # + dependency_links=[], + entry_points={ + 'setuptools.installation': [ + 'eggsecutable = volttron_openadr_ven.agent:main', + ] + } + +) \ No newline at end of file diff --git a/volttron_openadr_ven/__init__.py b/volttron_openadr_ven/__init__.py new file mode 100644 index 0000000..f7557d5 --- /dev/null +++ b/volttron_openadr_ven/__init__.py @@ -0,0 +1,82 @@ +# -*- coding: utf-8 -*- {{{ +# vim: set fenc=utf-8 ft=python sw=4 ts=4 sts=4 et: +# +# Copyright 2020, Battelle Memorial Institute. +# +# Licensed under the Apache License, Version 2.0 (the "License"); +# you may not use this file except in compliance with the License. +# You may obtain a copy of the License at +# +# http://www.apache.org/licenses/LICENSE-2.0 +# +# Unless required by applicable law or agreed to in writing, software +# distributed under the License is distributed on an "AS IS" BASIS, +# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. +# See the License for the specific language governing permissions and +# limitations under the License. +# +# This material was prepared as an account of work sponsored by an agency of +# the United States Government. Neither the United States Government nor the +# United States Department of Energy, nor Battelle, nor any of their +# employees, nor any jurisdiction or organization that has cooperated in the +# development of these materials, makes any warranty, express or +# implied, or assumes any legal liability or responsibility for the accuracy, +# completeness, or usefulness or any information, apparatus, product, +# software, or process disclosed, or represents that its use would not infringe +# privately owned rights. Reference herein to any specific commercial product, +# process, or service by trade name, trademark, manufacturer, or otherwise +# does not necessarily constitute or imply its endorsement, recommendation, or +# favoring by the United States Government or any agency thereof, or +# Battelle Memorial Institute. The views and opinions of authors expressed +# herein do not necessarily state or reflect those of the +# United States Government or any agency thereof. +# +# PACIFIC NORTHWEST NATIONAL LABORATORY operated by +# BATTELLE for the UNITED STATES DEPARTMENT OF ENERGY +# under Contract DE-AC05-76RL01830 +# }}} +""" +This __init__.py file is a shim that allows the agent to run in either the 8.x version of VOLTTRON +or the new modular volttron. The imports are the primary difference and therefore if the +8.x version is loaded but the modular version is only available then an import exception will +be thrown. + +If neither are available then an import exception will be thrown. + +In order to use this, the agent itself should use this import by using + +from . import ( + get_aware_utc_now, + .... + .... +) + +This redirects the correct version of the code to either the modular or 8.x version to import. +""" +try: + # Attempt to import from 8.x version of VOLTTRON if not successful to import then + # attempt to import from modular version of VOLTTRON. + from volttron.platform import jsonapi + from volttron.platform.agent.utils import ( + get_aware_utc_now, + setup_logging, + format_timestamp, + load_config, + isapipe + ) + from volttron.platform.messaging import topics, headers + from volttron.platform.vip.agent import Core, Agent + +except ImportError: + + from volttron.utils import ( + get_aware_utc_now, + jsonapi, + setup_logging, + format_timestamp, + load_config, + isapipe, + ) + + from volttron.client.vip.agent import Agent, Core + from volttron.client.messaging import topics, headers diff --git a/volttron_openadr_ven/agent.py b/volttron_openadr_ven/agent.py index ef2c3e2..1334e10 100644 --- a/volttron_openadr_ven/agent.py +++ b/volttron_openadr_ven/agent.py @@ -35,45 +35,58 @@ # BATTELLE for the UNITED STATES DEPARTMENT OF ENERGY # under Contract DE-AC05-76RL01830 # }}} - import logging import asyncio import sys -import openleadr.enums +import os +import gevent +from pathlib import Path from pprint import pformat -from datetime import datetime +from datetime import timedelta from openleadr.enums import OPT, REPORT_NAME, MEASUREMENTS -from openleadr import OpenADRClient +from openleadr.client import OpenADRClient -from volttron.utils import ( +from . import ( get_aware_utc_now, - jsonapi, setup_logging, - vip_main, format_timestamp, load_config, + isapipe, + jsonapi, + topics, + headers, + Agent, +) + +from volttron_openadr_ven.constants import ( + REQUIRED_KEYS, + VEN_NAME, + VTN_URL, + VIP_ADDRESS, + SERVER_KEY, + AGENT_PUBLIC, + AGENT_SECRET, + DEBUG, + CERT, + KEY, + PASSPHRASE, + VTN_FINGERPRINT, + SHOW_FINGERPRINT, + CA_FILE, + VEN_ID, + DISABLE_SIGNATURE, + OPENADR_CLIENT_TYPE, + IDENTITY, ) -from volttron.client.vip.agent import Agent, Core -from volttron.client.messaging import topics, headers +from .volttron_openadr_client import openadr_clients -setup_logging(level=logging.DEBUG) + +setup_logging() _log = logging.getLogger(__name__) __version__ = "1.0" -VEN_NAME = "ven_name" -VTN_URL = "vtn_url" -DEBUG = "debug" -CERT = "cert" -KEY = "key" -PASSPHRASE = "passphrase" -VTN_FINGERPRINT = "vtn_fingerprint" -SHOW_FINGERPRINT = "show_fingerprint" -CA_FILE = "ca_file" -VEN_ID = "ven_id" -DISABLE_SIGNATURE = "disable_signature" - class OpenADRVenAgent(Agent): """ @@ -88,6 +101,7 @@ def __init__( self, ven_name: str, vtn_url: str, + openadr_client_type: str, debug: bool = False, cert: str = None, key: str = None, @@ -119,6 +133,7 @@ def __init__( certificate. str ven_id: The ID for this VEN. If you leave this blank, a VEN_ID will be assigned by the VTN. bool disable_signature: Whether to disable digital signatures + str openadr_client_type: The type of openadr client to use. Valid client types are the openadr client class names in ~/volttron_openadr_ven/volttron_openadr_client.py """ super(OpenADRVenAgent, self).__init__(enable_web=True, **kwargs) @@ -136,6 +151,7 @@ def __init__( CA_FILE: ca_file, VEN_ID: ven_id, DISABLE_SIGNATURE: disable_signature, + OPENADR_CLIENT_TYPE: openadr_client_type, } # SubSystem/ConfigStore @@ -147,17 +163,24 @@ def __init__( # initialize all attributes of this class self.configure_agent(self.default_config) - def configure_agent(self, config: dict) -> None: + def configure_agent(self, config) -> None: """ Initialize the agent's configuration. Create an OpenADR Client using OpenLeadr. """ - _log.debug(f"Configuring agent with: \n {pformat(config)} ") + _log.info(f"Configuring agent with: \n {pformat(config)} ") # instantiate and add handlers to the OpenADR Client - _log.info("Creating OpenLeadrVen Client...") - self.ven_client = OpenADRClient( - ven_name=config.get(VEN_NAME), - vtn_url=config.get(VTN_URL), + client_type = config.get(OPENADR_CLIENT_TYPE) + openADRClient = openadr_clients().get(client_type) + if openADRClient is None: + msg = f"Invalid client type: {client_type}. Please use valid client types: {list(openadr_clients().keys())}" + _log.debug(msg) + raise KeyError(msg) + + _log.info(f"Creating OpenADRClient type: {client_type}...") + self.ven_client = openADRClient( + config.get(VEN_NAME), + config.get(VTN_URL), debug=config.get(DEBUG), cert=config.get(CERT), key=config.get(KEY), @@ -168,7 +191,11 @@ def configure_agent(self, config: dict) -> None: ven_id=config.get(VEN_ID), disable_signature=config.get(DISABLE_SIGNATURE), ) + _log.info(f"{client_type} successfully created.") + _log.info( + f"Adding capabilities (e.g. handlers, reports) to {client_type}..." + ) # Add event handling capability to the client # if you want to add more handlers on a specific event, you must create a coroutine in this class # and then add it as the second input for 'self.ven_client.add_handler(<some event>, <coroutine>)' @@ -183,7 +210,8 @@ def configure_agent(self, config: dict) -> None: measurement=MEASUREMENTS.VOLTAGE, ) - _log.info("Configuration complete.") + _log.info("Capabilities successfully added.") + gevent.spawn_later(3, self.start_asyncio_loop) def _configure(self, config_name, action, contents: dict) -> None: """The agent's config may have changed. Re-initialize it.""" @@ -191,39 +219,15 @@ def _configure(self, config_name, action, contents: dict) -> None: config.update(contents) self.configure_agent(config) - # ***************** Methods for Managing the Agent on Volttron ******************** - - @Core.receiver("onstart") - def onstart(self, sender) -> None: - """The agent has started.""" - _log.info(f"Sender {sender}") + def start_asyncio_loop(self): _log.info("Starting agent...") loop = asyncio.get_event_loop() loop.create_task(self.ven_client.run()) loop.run_forever() - # # TODO: Identify actions needed to be done before shutdown - # @Core.receiver("onstop") - # def onstop(self, sender, **kwargs): - # """ - # This method is called when the Agent is about to shutdown, but before it disconnects from - # the message bus. - # """ - # pass - # - # # TODO: Identify what, if at all, RPC methods should be available to other Agents - # @RPC.export - # def rpc_method(self, arg1, arg2, kwarg1=None, kwarg2=None): - # """ - # RPC method - # - # May be called from another agent via self.vip.rpc.call - # """ - # pass - # ***************** Methods for Servicing VTN Requests ******************** - async def handle_event(self, event: dict) -> openleadr.enums.OPT: + async def handle_event(self, event) -> OPT: """ Publish event to the Volttron message bus. Return OPT response. This coroutine will be called when there is an event to be handled. @@ -252,7 +256,7 @@ async def handle_event(self, event: dict) -> openleadr.enums.OPT: """ try: _log.debug("Received event. Processing event now...") - signal = event.get("event_signals")[0] # type: ignore[index] + signal = event.get("event_signals")[0] _log.info(f"Event signal:\n {pformat(signal)}") except IndexError as e: _log.debug( @@ -299,14 +303,79 @@ def _json_object(self, obj: dict): """Ensure that an object is valid JSON by dumping it with json_converter and then reloading it.""" obj_string = jsonapi.dumps( obj, - default=lambda x: x.__str__() if isinstance(x, datetime) else None, + default=lambda x: int(x.total_seconds()) + if isinstance(x, timedelta) + else None, ) return jsonapi.loads(obj_string) def main(): """Main method called to start the agent.""" - vip_main(ven_agent, version=__version__) + # TODO: when volttron.utils gets fixed by https://github.com/VOLTTRON/volttron-utils/issues/6, uncomment the line below and remove vip_main_tmp + # vip_main(ven_agent, version=__version__) + vip_main_tmp() + + +def vip_main_tmp(): + config_path = os.environ.get("AGENT_CONFIG") + if not config_path: + # this function borrows code from volttron.utils.commands.vip_main + # it allows the user of this agent to set the certificates so that the remote volttron platform can authenticate this agent + import argparse + + # Instantiate the parser + parser = argparse.ArgumentParser() + parser.add_argument("config_path") + args = parser.parse_args() + config_path = args.config_path + + if isapipe(sys.stdout): + # Hold a reference to the previous file object so it doesn't + # get garbage collected and close the underlying descriptor. + stdout = sys.stdout + sys.stdout = os.fdopen(stdout.fileno(), "w", 1) + + agent = ven_agent(config_path) + + try: + run = agent.run + except AttributeError: + run = agent.core.run + task = gevent.spawn(run) + try: + task.join() + finally: + task.kill() + + +def check_required_key(required_key, key_actual): + if required_key == VEN_NAME and not key_actual: + raise KeyError(f"{VEN_NAME} is required.") + elif required_key == VTN_URL and not key_actual: + raise KeyError( + f"{VTN_URL} is required. Ensure {VTN_URL} is given a URL to the VTN." + ) + elif required_key == VIP_ADDRESS and not key_actual: + raise KeyError( + f"{VIP_ADDRESS} is required. For example, if running volttron instance locally, use tcp://127.0.0.1" + ) + elif required_key == SERVER_KEY and not key_actual: + raise KeyError( + f"{SERVER_KEY} is required. To get the server key from the volttron instance that this agent will be connected to, use the command: vctl auth serverkey" + ) + elif required_key == AGENT_PUBLIC and not key_actual: + raise KeyError( + f"{AGENT_PUBLIC} is required. To generate a public key and associated secret key from the volttron instance that this agent will be connected to, run the command: vctl auth keypair" + ) + elif required_key == AGENT_SECRET and not key_actual: + raise KeyError( + f"{AGENT_SECRET} is required. To generate a public key and associated secret key from the volttron instance that this agent will be connected to, run the command: vctl auth keypair" + ) + elif required_key == OPENADR_CLIENT_TYPE and not key_actual: + raise KeyError( + f"{OPENADR_CLIENT_TYPE} is required. Specify one of the following valid client types: {list(openadr_clients().keys())}" + ) def ven_agent(config_path: str, **kwargs) -> OpenADRVenAgent: @@ -331,27 +400,42 @@ def ven_agent(config_path: str, **kwargs) -> OpenADRVenAgent: if not config: raise Exception("Configuration cannot be empty.") - ven_name = config.get(VEN_NAME) - if not ven_name: - raise Exception(f"{VEN_NAME} is required.") - vtn_url = config.get(VTN_URL) - if not vtn_url: - raise Exception( - f"{VTN_URL} is required. Ensure {VTN_URL} is given a URL to the VTN." - ) + req_keys_actual = {k: "" for k in REQUIRED_KEYS} + for required_key in REQUIRED_KEYS: + key_actual = config.get(required_key) + check_required_key(required_key, key_actual) + req_keys_actual[required_key] = key_actual + + remote_url = ( + f"{req_keys_actual[VIP_ADDRESS]}" + f"?serverkey={req_keys_actual[SERVER_KEY]}" + f"&publickey={req_keys_actual[AGENT_PUBLIC]}" + f"&secretkey={req_keys_actual[AGENT_SECRET]}" + ) + + # optional configurations debug = config.get(DEBUG) + + # keypair paths cert = config.get(CERT) + if cert: + cert = str(Path(cert).expanduser().resolve(strict=True)) key = config.get(KEY) + if key: + key = str(Path(key).expanduser().resolve(strict=True)) + passphrase = config.get(PASSPHRASE) vtn_fingerprint = config.get(VTN_FINGERPRINT) show_fingerprint = config.get(SHOW_FINGERPRINT) ca_file = config.get(CA_FILE) ven_id = config.get(VEN_ID) disable_signature = bool(config.get(DISABLE_SIGNATURE)) + openadr_client_type = req_keys_actual[OPENADR_CLIENT_TYPE] return OpenADRVenAgent( - ven_name, - vtn_url, + req_keys_actual[VEN_NAME], + req_keys_actual[VTN_URL], + openadr_client_type, debug=debug, cert=cert, key=key, @@ -361,6 +445,9 @@ def ven_agent(config_path: str, **kwargs) -> OpenADRVenAgent: ca_file=ca_file, ven_id=ven_id, disable_signature=disable_signature, + identity=IDENTITY, + # TODO: when volttron.utils gets fixed by https://github.com/VOLTTRON/volttron-utils/issues/6, remove the input 'address' + address=remote_url, **kwargs, ) diff --git a/volttron_openadr_ven/config.json b/volttron_openadr_ven/config.json deleted file mode 100644 index acba83b..0000000 --- a/volttron_openadr_ven/config.json +++ /dev/null @@ -1,8 +0,0 @@ -{ - "ven_name": "PNNLVEN", - "vtn_url": "https://eiss2demo.ipkeys.com/oadr2/OpenADR2/Simple/2.0b", - "cert": "secret/", - "key": "secret/", - "debug": true, - "disable_signature": true -} diff --git a/volttron_openadr_ven/constants.py b/volttron_openadr_ven/constants.py new file mode 100644 index 0000000..7588f47 --- /dev/null +++ b/volttron_openadr_ven/constants.py @@ -0,0 +1,28 @@ +# key names for processing config file for OpenADR VEN agent +VEN_NAME = "ven_name" +VTN_URL = "vtn_url" +DEBUG = "debug" +CERT = "cert_path" +KEY = "key_path" +PASSPHRASE = "passphrase" +VTN_FINGERPRINT = "vtn_fingerprint" +SHOW_FINGERPRINT = "show_fingerprint" +CA_FILE = "ca_file" +VEN_ID = "ven_id" +DISABLE_SIGNATURE = "disable_signature" +VIP_ADDRESS = "vip_address" +SERVER_KEY = "server_key" +AGENT_PUBLIC = "agent_public" +AGENT_SECRET = "agent_secret" +OPENADR_CLIENT_TYPE = "openadr_client_type" +REQUIRED_KEYS = [ + VEN_NAME, + VTN_URL, + VIP_ADDRESS, + SERVER_KEY, + AGENT_PUBLIC, + AGENT_SECRET, + OPENADR_CLIENT_TYPE, +] + +IDENTITY = "openadr_ven" diff --git a/volttron_openadr_ven/volttron_openadr_client.py b/volttron_openadr_ven/volttron_openadr_client.py new file mode 100644 index 0000000..377bd91 --- /dev/null +++ b/volttron_openadr_ven/volttron_openadr_client.py @@ -0,0 +1,206 @@ +import logging +import asyncio + +from abc import ABC +from functools import partial +from lxml import etree + +from openleadr.client import OpenADRClient +from openleadr.preflight import preflight_message +from openleadr.messaging import TEMPLATES, SIGNER, _create_replay_protect +from openleadr import utils, enums + +from . import setup_logging + +setup_logging() +logger = logging.getLogger(__name__) + + +""" +The VolttronOpenADR VEN agent uses the python library OpenLEADR https://github.com/openleadr/openleadr-python to create +an OpenADR VEN client. This agent extends the OpenADRClient class from OpenLEADR so that it can have the flexibility +to connect to any implementation of and OpenADR VTN. For example, to connect to an IPKeys VTN that was implemented +on an old OpenADR protocol, the IPKeysDemoVTNOpenADRClient is created so that it can successfully connect to an IPKeys VTN. + +If you have a specific VTN that you want to connect to and require further customization of the VEN client, create your +own OpenADRClient by extending the base class VolttronOpenADRClientBase, updating your client with your business logic, and putting that subclass in this module +""" + + +class OpenADRClientBase(OpenADRClient, ABC): + def __init__(self, ven_name, vtn_url, disable_signature=False, **kwargs): + super().__init__(ven_name, vtn_url, **kwargs) + self.disable_signature = disable_signature + + +class IPKeysClient(OpenADRClientBase, ABC): + def __init__(self, ven_name, vtn_url, disable_signature, **kwargs): + super().__init__(ven_name, vtn_url, disable_signature, **kwargs) + + self._create_message = partial( + self.create_message_ipkeys, + cert=self.cert_path, + key=self.key_path, + passphrase=self.passphrase, + disable_signature=self.disable_signature, + ) + + async def _on_event(self, message): + logger.debug("The VEN received an event") + events = message["events"] + try: + results = [] + + for event in message["events"]: + event_id = event["event_descriptor"]["event_id"] + event_status = event["event_descriptor"]["event_status"] + modification_number = event["event_descriptor"][ + "modification_number" + ] + received_event = utils.find_by( + self.received_events, "event_descriptor.event_id", event_id + ) + + if received_event: + if ( + received_event["event_descriptor"][ + "modification_number" + ] + == modification_number + ): + # Re-submit the same opt type as we already had previously + result = self.responded_events[event_id] + else: + # Replace the event with the fresh copy + utils.pop_by( + self.received_events, + "event_descriptor.event_id", + event_id, + ) + self.received_events.append(event) + # Wait for the result of the on_update_event handler + result = await utils.await_if_required( + self.on_update_event(event) + ) + else: + # Wait for the result of the on_event + self.received_events.append(event) + result = self.on_event(event) + if asyncio.iscoroutine(result): + result = await result + results.append(result) + if ( + event_status + in ( + enums.EVENT_STATUS.COMPLETED, + enums.EVENT_STATUS.CANCELLED, + ) + and event_id in self.responded_events + ): + self.responded_events.pop(event_id) + else: + self.responded_events[event_id] = result + for i, result in enumerate(results): + if ( + result not in ("optIn", "optOut") + and events[i]["response_required"] == "always" + ): + logger.error( + "Your on_event or on_update_event handler must return 'optIn' or 'optOut'; " + f"you supplied {result}. Please fix your on_event handler." + ) + results[i] = "optOut" + except Exception as err: + logger.error( + "Your on_event handler encountered an error. Will Opt Out of the event. " + f"The error was {err.__class__.__name__}: {str(err)}" + ) + results = ["optOut"] * len(events) + + event_responses = [ + { + "response_code": 200, + "response_description": "OK", + "opt_type": results[i], + "request_id": message["request_id"], + "modification_number": events[i]["event_descriptor"][ + "modification_number" + ], + "event_id": events[i]["event_descriptor"]["event_id"], + } + for i, event in enumerate(events) + if event["response_required"] == "always" + and not utils.determine_event_status(event["active_period"]) + == "completed" + ] + + if len(event_responses) > 0: + response = { + "response_code": 200, + "response_description": "OK", + "request_id": message["request_id"], + } + message = self._create_message( + "oadrCreatedEvent", + response=response, + event_responses=event_responses, + ven_id=self.ven_id, + ) + service = "EiEvent" + response_type, response_payload = await self._perform_request( + service, message + ) + logger.info(response_type, response_payload) + else: + logger.info( + "Not sending any event responses, because a response was not required/allowed by the VTN." + ) + + @staticmethod + def create_message_ipkeys( + message_type, + cert=None, + key=None, + passphrase=None, + disable_signature=False, + **message_payload, + ): + """ + Create and optionally sign an OpenADR message. Returns an XML string. + """ + message_payload = preflight_message(message_type, message_payload) + template = TEMPLATES.get_template(f"{message_type}.xml") + signed_object = utils.flatten_xml(template.render(**message_payload)) + envelope = TEMPLATES.get_template("oadrPayload.xml") + if cert and key and not disable_signature: + tree = etree.fromstring(signed_object) + signature_tree = SIGNER.sign( + tree, + key=key, + cert=cert, + passphrase=utils.ensure_bytes(passphrase), + reference_uri="#oadrSignedObject", + signature_properties=_create_replay_protect(), + ) + signature = etree.tostring(signature_tree).decode("utf-8") + else: + signature = None + msg = envelope.render( + template=f"{message_type}", + signature=signature, + signed_object=signed_object, + ) + return msg + + +def openadr_clients(): + clients = {} + work = [OpenADRClient] + while work: + parent = work.pop() + for child in parent.__subclasses__(): + child_name = child.__name__ + if child_name not in clients: + clients[child_name] = child + work.append(child) + return clients

Connect to IPKeys vtn and see what gaps we have with openleadr (see ben's stuff in email) Create Skeleton Agent Structure This skeleton will have volttron agent properties as well as being able to start the async loop and be able to call the openleadr source code https://pypi.org/project/openleadr/

Was finally able to connect to IPKeys demo VTN using OpenLEADR, albeit with some strings attached. I had to modify the OpenLEADR library to create a VEN client that can disable signatures for clients that use TLS and certs. See https://github.com/bonicim/openleadr-python/tree/feature/disable-signature I opened up a PR to OpenLEADR with my changes: https://github.com/OpenLEADR/openleadr-python/pull/98 Since Openleadr has yet to review my PR, I have forked it, made the required changes to work with the Volttron Agent framework and to receive messages from the IPKeys VTN. The forked version used for the agent is at: https://github.com/bonicim/openleadr-python/tree/e306798f9f4af36f0349c10fec4607aaa18d34c8 Additional Notes: * This agent will take the following dependencies: * https://github.com/VOLTTRON/volttron-client/tree/main/volttron/platform * https://github.com/VOLTTRON/volttron-utils * This agent is a pared-down version of OpenADRAgent in volttron core; see https://github.com/VOLTTRON/volttron/blob/develop/services/core/OpenADRVenAgent/openadrven/agent.py#L414 * Agent should have the following methods * publish_event * process_event * process_report * Agent should ideally be a gevent-based agent that can make asyn calls; see example from openleader: https://openleadr.org/docs/#getting-started * Agent should be packaged and able to be installed via wheel on a volttron platform using `vctl install`

2021-12-04T00:40:06

0.0

[]

eclipse-volttron/volttron-openadr-ven

eclipse-volttron__volttron-openadr-ven-13

7b0f5a919c37a8588736c5c85eb52a081685cd57

diff --git a/.gitignore b/.gitignore index b97131e..94cae34 100644 --- a/.gitignore +++ b/.gitignore @@ -129,3 +129,6 @@ dmypy.json .pyre/ .idea/ + +*secret +config_local.json diff --git a/.gitmodules b/.gitmodules new file mode 100644 index 0000000..8dd3851 --- /dev/null +++ b/.gitmodules @@ -0,0 +1,3 @@ +[submodule "openleadr"] + path = openleadr + url = [email protected]:bonicim/openleadr-python.git diff --git a/README.md b/README.md index 0191cf6..933493b 100644 --- a/README.md +++ b/README.md @@ -1,43 +1,85 @@ -# Prerequisites +# OpenADRVen Agent + +OpenADR (Automated Demand Response) is a standard for alerting and responding to the need to adjust electric power +consumption in response to fluctuations in grid demand. + +For further information about OpenADR and this agent, please see the OpenADR documentation in VOLTTRON ReadTheDocs. + +## Agent Configuration + +The only required parameters for this agent are "ven_name" and "vtn_url". Below is an example of a +correct configuration with optional parameters added. + +```jsonpath + { + "ven_name": "PNNLVEN", + "vtn_url": "https://eiss2demo.ipkeys.com/oadr2/OpenADR2/Simple/2.0b", + + # below are optional parameters + + # if you want/need to sign outgoing messages using a public-private key pair, you need to add a directory named "secret" in the top level + # of this repo; then provide the relative path to the cert and key + # below is an example + # test certificates for IP Keys VTN's can be obtained at https://testcerts.kyrio.com/#/ + "cert": "secret/TEST_RSA_VEN_210923215148_certs/TEST_RSA_VEN_210923215148_cert.pem", + "key": "secret/TEST_RSA_VEN_210923215148_certs/TEST_RSA_VEN_210923215148_privkey.pem", + + # other optional parameters + "debug": true, + "disable_signature": true + } +``` + +## Volttron integration + +To ensure that this agent is created within the Volttron framework, ensure that you set the following environmental variable +to the absolute path of the configuration JSON file for this agent. For example, on Linux, use the following example to set +the environmental variable AGENT_CONFIG to a path to your JSON config file: + +`export AGENT_CONFIG /home/usr/path-to-my-config-file` + +# Development + +## Prerequisites * Python >=3.7 * Pip >=20.1 * Poetry >=1.16 -# Development ## Environment setup Note: This repo uses [Poetry](https://python-poetry.org/), a dependency management and packaging tool for Python. If you don't have Poetry installed on your machine, follow [these steps](https://python-poetry.org/docs/#installation) to install it on your machine. -To check if Poetry is installed, run ```poetry --version```. If your receive 'command not found: poetry' error, add the following line to your '~/.bashrc' script: ```export PATH=$PATH:$HOME/.poetry/bin```. +To check if Poetry is installed, run `poetry --version`. If you receive the error 'command not found: poetry', add the following line to your '~/.bashrc' script: ```export PATH=$PATH:$HOME/.poetry/bin```. + +0. Install the required submodules -To setup your development environment, do the following: +```git submodule update --init --recursive``` 1. Create virtual environment -Run the following command on the CLI. By default, poetry creates a virtual environment in {cache-dir}/virtualenvs -({cache-dir}\virtualenvs on Windows). For more info on creating virtual environments through Poetry, see [this](https://python-poetry.org/docs/basic-usage/#using-your-virtual-environment). +By default, poetry creates a virtual environment in {cache-dir}/virtualenvs +({cache-dir}\virtualenvs on Windows). To configure 'poetry' to create the virtualenv inside this project's root +directory, run the following command: + +[```poetry config virtualenvs.in-project true```](https://python-poetry.org/docs/configuration +) + +Then to create the virtual environment itself, run the following command: ```shell -$ poetry shell +poetry shell ``` NOTE: Poetry needs the command 'python' in order to create a new shell. On Linux, if you only have python3 installed, run the following to link 'python' to 'python3': ```shell -$ sudo apt install python-is-python3 +sudo apt install python-is-python3 ``` 2. Install requirements ```shell -(env) $ poetry install -``` - - -3. Install Volttron-specific dependencies for this agent. - -```shell -(env) $ pip install volttron_utils-0.1.1-py3-none-any.whl volttron_client-0.1.2-py3-none-any.whl +poetry install ``` @@ -46,16 +88,21 @@ $ sudo apt install python-is-python3 ## Git Hooks and Pre-commit This repo provides pre-commit hooks to enforce code quality and standardization before pushing any commit to the repo. To use this tool locally, -run the following commands: +run the following command. ```shell -(env) $ pre-commit install +pre-commit install +``` + +To check that pre-commit scripts are installed and properly run, run pre-commit on all the files: -# To check that pre-commit scripts are installed and properly run, run pre-commit on all the files -(env) $ pre-commit run --all +``` +pre-commit run --all ``` -With pre-commit setup, every time you run "git commit", git will check your commit against a set of code linters. For details, see .pre-commit-config.yaml +With pre-commit installed and configured, every time you run `git commit`, git will check your commit against a set of code linters. For details, see .pre-commit-config.yaml + +This project installs the following tools to aid in code standardization and best practices: ## Black @@ -63,14 +110,14 @@ This repo uses [Black](https://pypi.org/project/black/), an opinionated Python c run the following command on your chosen Python file: ```shell -(env) $ black <path to file> +black <path to file> ``` ## MyPy This repo uses [MyPy](https://mypy.readthedocs.io/en/stable/), a static type checker for Python. To use MyPy on a Python file, run the following command: ```shell -(env) $ mypy <path to file> +mypy <path to file> ``` # CI/CD @@ -93,3 +140,23 @@ Test uploading Upload to PyPi ```python -m twin upload dist/*``` + + +# Testing + +## Ven Certificates + +https://testcerts.kyrio.com/#/ + +## Test VTN's + +* IPKeys: https://eiss2demo.ipkeys.com/ui/home/#/login + +# Other + +## Installing dependencies behind a proxy + +If running this behind a PNNL VPN, run the following so Poetry can install all the dependencies: +See: https://leifengblog.net/blog/how-to-use-pip-behind-a-proxy/ + +```export https_proxy=http://proxy01.pnl.gov:3128``` diff --git a/mypy.ini b/mypy.ini index 03aefad..8f59715 100644 --- a/mypy.ini +++ b/mypy.ini @@ -2,3 +2,4 @@ show_error_context = True pretty = True show_column_numbers = True +show_error_codes = True diff --git a/openleadr b/openleadr new file mode 160000 index 0000000..e306798 --- /dev/null +++ b/openleadr @@ -0,0 +1,1 @@ +Subproject commit e306798f9f4af36f0349c10fec4607aaa18d34c8 diff --git a/poetry.lock b/poetry.lock index 26f5d1c..b2f2e1c 100644 --- a/poetry.lock +++ b/poetry.lock @@ -17,17 +17,9 @@ yarl = ">=1.0,<2.0" [package.extras] speedups = ["aiodns", "brotlipy", "cchardet"] -[[package]] -name = "appdirs" -version = "1.4.4" -description = "A small Python module for determining appropriate platform-specific dirs, e.g. a \"user data dir\"." -category = "main" -optional = false -python-versions = "*" - [[package]] name = "apscheduler" -version = "3.7.0" +version = "3.8.0" description = "In-process task scheduler with Cron-like capabilities" category = "main" optional = false @@ -51,6 +43,14 @@ tornado = ["tornado (>=4.3)"] twisted = ["twisted"] zookeeper = ["kazoo"] +[[package]] +name = "asn1crypto" +version = "1.4.0" +description = "Fast ASN.1 parser and serializer with definitions for private keys, public keys, certificates, CRL, OCSP, CMS, PKCS#3, PKCS#7, PKCS#8, PKCS#12, PKCS#5, X.509 and TSP" +category = "main" +optional = false +python-versions = "*" + [[package]] name = "async-timeout" version = "3.0.1" @@ -81,31 +81,48 @@ docs = ["furo", "sphinx", "zope.interface", "sphinx-notfound-page"] tests = ["coverage[toml] (>=5.0.2)", "hypothesis", "pympler", "pytest (>=4.3.0)", "six", "mypy", "pytest-mypy-plugins", "zope.interface"] tests_no_zope = ["coverage[toml] (>=5.0.2)", "hypothesis", "pympler", "pytest (>=4.3.0)", "six", "mypy", "pytest-mypy-plugins"] +[[package]] +name = "backports.entry-points-selectable" +version = "1.1.0" +description = "Compatibility shim providing selectable entry points for older implementations" +category = "main" +optional = false +python-versions = ">=2.7" + +[package.extras] +docs = ["sphinx", "jaraco.packaging (>=8.2)", "rst.linker (>=1.9)"] +testing = ["pytest (>=4.6)", "pytest-flake8", "pytest-cov", "pytest-black (>=0.3.7)", "pytest-mypy", "pytest-checkdocs (>=2.4)", "pytest-enabler (>=1.0.1)"] + [[package]] name = "black" -version = "21.5b2" +version = "21.9b0" description = "The uncompromising code formatter." category = "main" optional = false python-versions = ">=3.6.2" [package.dependencies] -appdirs = "*" click = ">=7.1.2" mypy-extensions = ">=0.4.3" -pathspec = ">=0.8.1,<1" +pathspec = ">=0.9.0,<1" +platformdirs = ">=2" regex = ">=2020.1.8" -toml = ">=0.10.1" +tomli = ">=0.2.6,<2.0.0" +typing-extensions = [ + {version = ">=3.10.0.0", markers = "python_version < \"3.10\""}, + {version = "!=3.10.0.1", markers = "python_version >= \"3.10\""}, +] [package.extras] colorama = ["colorama (>=0.4.3)"] d = ["aiohttp (>=3.6.0)", "aiohttp-cors (>=0.4.0)"] +jupyter = ["ipython (>=7.8.0)", "tokenize-rt (>=3.2.0)"] python2 = ["typed-ast (>=1.4.2)"] uvloop = ["uvloop (>=0.15.2)"] [[package]] name = "certifi" -version = "2021.5.30" +version = "2021.10.8" description = "Python package for providing Mozilla's CA Bundle." category = "main" optional = false @@ -113,7 +130,7 @@ python-versions = "*" [[package]] name = "cffi" -version = "1.14.5" +version = "1.15.0" description = "Foreign Function Interface for Python calling C code." category = "main" optional = false @@ -124,7 +141,7 @@ pycparser = "*" [[package]] name = "cfgv" -version = "3.3.0" +version = "3.3.1" description = "Validate configuration and produce human readable error messages." category = "main" optional = false @@ -138,9 +155,20 @@ category = "main" optional = false python-versions = ">=2.7, !=3.0.*, !=3.1.*, !=3.2.*, !=3.3.*, !=3.4.*" +[[package]] +name = "charset-normalizer" +version = "2.0.7" +description = "The Real First Universal Charset Detector. Open, modern and actively maintained alternative to Chardet." +category = "main" +optional = false +python-versions = ">=3.5.0" + +[package.extras] +unicode_backport = ["unicodedata2"] + [[package]] name = "click" -version = "8.0.1" +version = "8.0.3" description = "Composable command line interface toolkit" category = "main" optional = false @@ -159,26 +187,39 @@ python-versions = ">=2.7, !=3.0.*, !=3.1.*, !=3.2.*, !=3.3.*, !=3.4.*" [[package]] name = "cryptography" -version = "3.4.7" +version = "2.3" description = "cryptography is a package which provides cryptographic recipes and primitives to Python developers." category = "main" optional = false -python-versions = ">=3.6" +python-versions = ">=2.7,!=3.0.*,!=3.1.*,!=3.2.*,!=3.3.*" [package.dependencies] -cffi = ">=1.12" +asn1crypto = ">=0.21.0" +cffi = ">=1.7,<1.11.3 || >1.11.3" +idna = ">=2.1" +six = ">=1.4.1" [package.extras] -docs = ["sphinx (>=1.6.5,!=1.8.0,!=3.1.0,!=3.1.1)", "sphinx-rtd-theme"] -docstest = ["doc8", "pyenchant (>=1.6.11)", "twine (>=1.12.0)", "sphinxcontrib-spelling (>=4.0.1)"] -pep8test = ["black", "flake8", "flake8-import-order", "pep8-naming"] -sdist = ["setuptools-rust (>=0.11.4)"] -ssh = ["bcrypt (>=3.1.5)"] -test = ["pytest (>=6.0)", "pytest-cov", "pytest-subtests", "pytest-xdist", "pretend", "iso8601", "pytz", "hypothesis (>=1.11.4,!=3.79.2)"] +docs = ["sphinx (>=1.6.5)", "sphinx-rtd-theme"] +docstest = ["doc8", "pyenchant (>=1.6.11)", "readme_renderer (>=16.0)", "sphinxcontrib-spelling (>=4.0.1)"] +pep8test = ["flake8", "flake8-import-order", "pep8-naming"] +test = ["pytest (>=3.2.1,!=3.3.0)", "pretend", "iso8601", "pytz", "hypothesis (>=1.11.4)"] + +[[package]] +name = "dateutils" +version = "0.6.12" +description = "Various utilities for working with date and datetime objects" +category = "main" +optional = false +python-versions = "*" + +[package.dependencies] +python-dateutil = "*" +pytz = "*" [[package]] name = "distlib" -version = "0.3.2" +version = "0.3.3" description = "Distribution utilities" category = "main" optional = false @@ -197,11 +238,15 @@ future = ">=0.16,<0.19" [[package]] name = "filelock" -version = "3.0.12" +version = "3.3.1" description = "A platform independent file lock." category = "main" optional = false -python-versions = "*" +python-versions = ">=3.6" + +[package.extras] +docs = ["furo (>=2021.8.17b43)", "sphinx (>=4.1)", "sphinx-autodoc-typehints (>=1.12)"] +testing = ["covdefaults (>=1.2.0)", "coverage (>=4)", "pytest (>=4)", "pytest-cov", "pytest-timeout (>=1.4.2)"] [[package]] name = "future" @@ -211,9 +256,50 @@ category = "main" optional = false python-versions = ">=2.6, !=3.0.*, !=3.1.*, !=3.2.*" +[[package]] +name = "gevent" +version = "20.6.1" +description = "Coroutine-based network library" +category = "main" +optional = false +python-versions = ">=2.7,!=3.0.*,!=3.1.*,!=3.2.*,!=3.3.*,!=3.4.*" + +[package.dependencies] +cffi = {version = ">=1.12.2", markers = "platform_python_implementation == \"CPython\" and sys_platform == \"win32\""} +greenlet = {version = ">=0.4.16", markers = "platform_python_implementation == \"CPython\""} +"zope.event" = "*" +"zope.interface" = "*" + +[package.extras] +dnspython = ["dnspython (>=1.16.0)", "idna"] +docs = ["repoze.sphinx.autointerface", "sphinxcontrib-programoutput"] +monitor = ["psutil (>=5.6.1)", "psutil (==5.6.3)"] +recommended = ["dnspython (>=1.16.0)", "idna", "cffi (>=1.12.2)", "psutil (>=5.6.1)", "psutil (==5.6.3)", "selectors2", "backports.socketpair"] +test = ["dnspython (>=1.16.0)", "idna", "requests", "objgraph", "cffi (>=1.12.2)", "psutil (>=5.6.1)", "psutil (==5.6.3)", "selectors2", "futures", "mock", "backports.socketpair", "contextvars (==2.4)", "coverage (<5.0)", "coveralls (>=1.7.0)"] + +[[package]] +name = "greenlet" +version = "0.4.16" +description = "Lightweight in-process concurrent programming" +category = "main" +optional = false +python-versions = "*" + +[[package]] +name = "grequests" +version = "0.6.0" +description = "Requests + Gevent" +category = "main" +optional = false +python-versions = "*" + +[package.dependencies] +gevent = "*" +requests = "*" + [[package]] name = "identify" -version = "2.2.7" +version = "2.3.0" description = "File identification library for Python" category = "main" optional = false @@ -224,11 +310,11 @@ license = ["editdistance-s"] [[package]] name = "idna" -version = "3.2" +version = "2.10" description = "Internationalized Domain Names in Applications (IDNA)" category = "main" optional = false -python-versions = ">=3.5" +python-versions = ">=2.7, !=3.0.*, !=3.1.*, !=3.2.*, !=3.3.*" [[package]] name = "iniconfig" @@ -240,7 +326,7 @@ python-versions = "*" [[package]] name = "jinja2" -version = "3.0.1" +version = "3.0.2" description = "A very fast and expressive template engine." category = "main" optional = false @@ -276,7 +362,7 @@ python-versions = ">=3.6" [[package]] name = "multidict" -version = "5.1.0" +version = "5.2.0" description = "multidict implementation" category = "main" optional = false @@ -316,11 +402,12 @@ python-versions = "*" [[package]] name = "openleadr" -version = "0.5.23" -description = "Python3 library for building OpenADR Clients (VENs) and Servers (VTNs)" +version = "0.5.24" +description = "" category = "main" optional = false python-versions = ">=3.7.0" +develop = true [package.dependencies] aiohttp = "*" @@ -329,39 +416,56 @@ jinja2 = "*" signxml-openadr = "2.9.1" xmltodict = "*" +[package.source] +type = "directory" +url = "openleadr" + [[package]] name = "packaging" -version = "20.9" +version = "21.0" description = "Core utilities for Python packages" category = "dev" optional = false -python-versions = ">=2.7, !=3.0.*, !=3.1.*, !=3.2.*, !=3.3.*" +python-versions = ">=3.6" [package.dependencies] pyparsing = ">=2.0.2" [[package]] name = "pathspec" -version = "0.8.1" +version = "0.9.0" description = "Utility library for gitignore style pattern matching of file paths." category = "main" optional = false -python-versions = ">=2.7, !=3.0.*, !=3.1.*, !=3.2.*, !=3.3.*, !=3.4.*" +python-versions = "!=3.0.*,!=3.1.*,!=3.2.*,!=3.3.*,!=3.4.*,>=2.7" + +[[package]] +name = "platformdirs" +version = "2.4.0" +description = "A small Python module for determining appropriate platform-specific dirs, e.g. a \"user data dir\"." +category = "main" +optional = false +python-versions = ">=3.6" + +[package.extras] +docs = ["Sphinx (>=4)", "furo (>=2021.7.5b38)", "proselint (>=0.10.2)", "sphinx-autodoc-typehints (>=1.12)"] +test = ["appdirs (==1.4.4)", "pytest (>=6)", "pytest-cov (>=2.7)", "pytest-mock (>=3.6)"] [[package]] name = "pluggy" -version = "0.13.1" +version = "1.0.0" description = "plugin and hook calling mechanisms for python" category = "dev" optional = false -python-versions = ">=2.7, !=3.0.*, !=3.1.*, !=3.2.*, !=3.3.*" +python-versions = ">=3.6" [package.extras] dev = ["pre-commit", "tox"] +testing = ["pytest", "pytest-benchmark"] [[package]] name = "pre-commit" -version = "2.13.0" +version = "2.15.0" description = "A framework for managing and maintaining multi-language pre-commit hooks." category = "main" optional = false @@ -375,11 +479,22 @@ pyyaml = ">=5.1" toml = "*" virtualenv = ">=20.0.8" +[[package]] +name = "psutil" +version = "5.8.0" +description = "Cross-platform lib for process and system monitoring in Python." +category = "main" +optional = false +python-versions = ">=2.6, !=3.0.*, !=3.1.*, !=3.2.*, !=3.3.*" + +[package.extras] +test = ["ipaddress", "mock", "unittest2", "enum34", "pywin32", "wmi"] + [[package]] name = "py" version = "1.10.0" description = "library with cross-python path, ini-parsing, io, code, log facilities" -category = "dev" +category = "main" optional = false python-versions = ">=2.7, !=3.0.*, !=3.1.*, !=3.2.*, !=3.3.*" @@ -393,14 +508,14 @@ python-versions = ">=2.7, !=3.0.*, !=3.1.*, !=3.2.*, !=3.3.*" [[package]] name = "pyopenssl" -version = "19.1.0" +version = "19.0.0" description = "Python wrapper module around the OpenSSL library" category = "main" optional = false python-versions = "*" [package.dependencies] -cryptography = ">=2.8" +cryptography = ">=2.3" six = ">=1.5.2" [package.extras] @@ -417,7 +532,7 @@ python-versions = ">=2.6, !=3.0.*, !=3.1.*, !=3.2.*" [[package]] name = "pytest" -version = "6.2.4" +version = "6.2.5" description = "pytest: simple powerful testing with Python" category = "dev" optional = false @@ -429,16 +544,27 @@ attrs = ">=19.2.0" colorama = {version = "*", markers = "sys_platform == \"win32\""} iniconfig = "*" packaging = "*" -pluggy = ">=0.12,<1.0.0a1" +pluggy = ">=0.12,<2.0" py = ">=1.8.2" toml = "*" [package.extras] testing = ["argcomplete", "hypothesis (>=3.56)", "mock", "nose", "requests", "xmlschema"] +[[package]] +name = "python-dateutil" +version = "2.8.2" +description = "Extensions to the standard Python datetime module" +category = "main" +optional = false +python-versions = "!=3.0.*,!=3.1.*,!=3.2.*,>=2.7" + +[package.dependencies] +six = ">=1.5" + [[package]] name = "pytz" -version = "2021.1" +version = "2021.3" description = "World timezone definitions, modern and historical" category = "main" optional = false @@ -452,14 +578,44 @@ category = "main" optional = false python-versions = ">=2.7, !=3.0.*, !=3.1.*, !=3.2.*, !=3.3.*, !=3.4.*, !=3.5.*" +[[package]] +name = "pyzmq" +version = "22.3.0" +description = "Python bindings for 0MQ" +category = "main" +optional = false +python-versions = ">=3.6" + +[package.dependencies] +cffi = {version = "*", markers = "implementation_name == \"pypy\""} +py = {version = "*", markers = "implementation_name == \"pypy\""} + [[package]] name = "regex" -version = "2021.4.4" +version = "2021.10.21" description = "Alternative regular expression module, to replace re." category = "main" optional = false python-versions = "*" +[[package]] +name = "requests" +version = "2.26.0" +description = "Python HTTP for Humans." +category = "main" +optional = false +python-versions = ">=2.7, !=3.0.*, !=3.1.*, !=3.2.*, !=3.3.*, !=3.4.*, !=3.5.*" + +[package.dependencies] +certifi = ">=2017.4.17" +charset-normalizer = {version = ">=2.0.0,<2.1.0", markers = "python_version >= \"3\""} +idna = {version = ">=2.5,<4", markers = "python_version >= \"3\""} +urllib3 = ">=1.21.1,<1.27" + +[package.extras] +socks = ["PySocks (>=1.5.6,!=1.5.7)", "win-inet-pton"] +use_chardet_on_py3 = ["chardet (>=3.0.2,<5)"] + [[package]] name = "signxml-openadr" version = "2.9.1" @@ -491,6 +647,14 @@ category = "main" optional = false python-versions = ">=2.6, !=3.0.*, !=3.1.*, !=3.2.*" +[[package]] +name = "tomli" +version = "1.2.1" +description = "A lil' TOML parser" +category = "main" +optional = false +python-versions = ">=3.6" + [[package]] name = "typed-ast" version = "1.4.3" @@ -501,7 +665,7 @@ python-versions = "*" [[package]] name = "typing-extensions" -version = "3.10.0.0" +version = "3.10.0.2" description = "Backported and Experimental Type Hints for Python 3.5+" category = "main" optional = false @@ -518,23 +682,96 @@ python-versions = "*" [package.dependencies] pytz = "*" +[[package]] +name = "urllib3" +version = "1.26.7" +description = "HTTP library with thread-safe connection pooling, file post, and more." +category = "main" +optional = false +python-versions = ">=2.7, !=3.0.*, !=3.1.*, !=3.2.*, !=3.3.*, !=3.4.*, <4" + +[package.extras] +brotli = ["brotlipy (>=0.6.0)"] +secure = ["pyOpenSSL (>=0.14)", "cryptography (>=1.3.4)", "idna (>=2.0.0)", "certifi", "ipaddress"] +socks = ["PySocks (>=1.5.6,!=1.5.7,<2.0)"] + [[package]] name = "virtualenv" -version = "20.4.7" +version = "20.8.1" description = "Virtual Python Environment builder" category = "main" optional = false -python-versions = "!=3.0.*,!=3.1.*,!=3.2.*,!=3.3.*,>=2.7" +python-versions = "!=3.0.*,!=3.1.*,!=3.2.*,!=3.3.*,!=3.4.*,>=2.7" [package.dependencies] -appdirs = ">=1.4.3,<2" +"backports.entry-points-selectable" = ">=1.0.4" distlib = ">=0.3.1,<1" filelock = ">=3.0.0,<4" +platformdirs = ">=2,<3" six = ">=1.9.0,<2" [package.extras] docs = ["proselint (>=0.10.2)", "sphinx (>=3)", "sphinx-argparse (>=0.2.5)", "sphinx-rtd-theme (>=0.4.3)", "towncrier (>=19.9.0rc1)"] -testing = ["coverage (>=4)", "coverage-enable-subprocess (>=1)", "flaky (>=3)", "pytest (>=4)", "pytest-env (>=0.6.2)", "pytest-freezegun (>=0.4.1)", "pytest-mock (>=2)", "pytest-randomly (>=1)", "pytest-timeout (>=1)", "packaging (>=20.0)", "xonsh (>=0.9.16)"] +testing = ["coverage (>=4)", "coverage-enable-subprocess (>=1)", "flaky (>=3)", "pytest (>=4)", "pytest-env (>=0.6.2)", "pytest-freezegun (>=0.4.1)", "pytest-mock (>=2)", "pytest-randomly (>=1)", "pytest-timeout (>=1)", "packaging (>=20.0)"] + +[[package]] +name = "volttron-client" +version = "0.3.4" +description = "Client for connecting to a volttron server" +category = "main" +optional = false +python-versions = ">=3.6.2,<4" + +[package.dependencies] +dateutils = ">=0.6,<0.7" +grequests = ">=0.6.0,<0.7.0" +idna = ">=2.5,<3" +requests = ">=2.26.0,<3.0.0" +toml = ">=0.10.2,<0.11.0" +volttron-utils = ">=0.3,<0.4" + +[[package]] +name = "volttron-utils" +version = "0.3.2" +description = "Utility classes and methods for VOLTTRON" +category = "main" +optional = false +python-versions = ">=3.6.2,<4.0" + +[package.dependencies] +cryptography = "2.3" +dateutils = ">=0.6.7,<0.7.0" +gevent = "20.6.1" +greenlet = "0.4.16" +psutil = ">=5.7.0,<6.0.0" +pyOpenSSL = "19.0.0" +PyYAML = ">=5.4.1,<6.0.0" +tzlocal = "2.1" +watchdog-gevent = ">=0.1.1,<0.2.0" +zmq = ">=0.0.0,<0.0.1" + +[[package]] +name = "watchdog" +version = "2.1.6" +description = "Filesystem events monitoring" +category = "main" +optional = false +python-versions = ">=3.6" + +[package.extras] +watchmedo = ["PyYAML (>=3.10)"] + +[[package]] +name = "watchdog-gevent" +version = "0.1.1" +description = "A gevent-based observer for watchdog." +category = "main" +optional = false +python-versions = "*" + +[package.dependencies] +gevent = ">=1.1" +watchdog = ">=0.8" [[package]] name = "xmltodict" @@ -546,7 +783,7 @@ python-versions = ">=2.7, !=3.0.*, !=3.1.*, !=3.2.*, !=3.3.*" [[package]] name = "yarl" -version = "1.6.3" +version = "1.7.0" description = "Yet another URL library" category = "main" optional = false @@ -556,10 +793,46 @@ python-versions = ">=3.6" idna = ">=2.0" multidict = ">=4.0" +[[package]] +name = "zmq" +version = "0.0.0" +description = "You are probably looking for pyzmq." +category = "main" +optional = false +python-versions = "*" + +[package.dependencies] +pyzmq = "*" + +[[package]] +name = "zope.event" +version = "4.5.0" +description = "Very basic event publishing system" +category = "main" +optional = false +python-versions = "*" + +[package.extras] +docs = ["sphinx"] +test = ["zope.testrunner"] + +[[package]] +name = "zope.interface" +version = "5.4.0" +description = "Interfaces for Python" +category = "main" +optional = false +python-versions = ">=2.7, !=3.0.*, !=3.1.*, !=3.2.*, !=3.3.*, !=3.4.*" + +[package.extras] +docs = ["sphinx", "repoze.sphinx.autointerface"] +test = ["coverage (>=5.0.3)", "zope.event", "zope.testing"] +testing = ["coverage (>=5.0.3)", "zope.event", "zope.testing"] + [metadata] lock-version = "1.1" python-versions = "^3.8" -content-hash = "2a1dd13cfcde4f16340cb5425979909af99a26d9e7091cd2163a7e118ca1004b" +content-hash = "05aefd0e959b8da3ef1df6dbd3f3ec194bfe8b7cbc8d49f393a5919cd78c075e" [metadata.files] aiohttp = [ @@ -601,13 +874,13 @@ aiohttp = [ {file = "aiohttp-3.7.4.post0-cp39-cp39-win_amd64.whl", hash = "sha256:02f46fc0e3c5ac58b80d4d56eb0a7c7d97fcef69ace9326289fb9f1955e65cfe"}, {file = "aiohttp-3.7.4.post0.tar.gz", hash = "sha256:493d3299ebe5f5a7c66b9819eacdcfbbaaf1a8e84911ddffcdc48888497afecf"}, ] -appdirs = [ - {file = "appdirs-1.4.4-py2.py3-none-any.whl", hash = "sha256:a841dacd6b99318a741b166adb07e19ee71a274450e68237b4650ca1055ab128"}, - {file = "appdirs-1.4.4.tar.gz", hash = "sha256:7d5d0167b2b1ba821647616af46a749d1c653740dd0d2415100fe26e27afdf41"}, -] apscheduler = [ - {file = "APScheduler-3.7.0-py2.py3-none-any.whl", hash = "sha256:c06cc796d5bb9eb3c4f77727f6223476eb67749e7eea074d1587550702a7fbe3"}, - {file = "APScheduler-3.7.0.tar.gz", hash = "sha256:1cab7f2521e107d07127b042155b632b7a1cd5e02c34be5a28ff62f77c900c6a"}, + {file = "APScheduler-3.8.0-py2.py3-none-any.whl", hash = "sha256:82d6d21b2f0343510d07bb35618333a794653439a265635555b12935647b460a"}, + {file = "APScheduler-3.8.0.tar.gz", hash = "sha256:793b2d37c52ece53e34626619e6142e99b20b59a12155f39e1e6932e324f079d"}, +] +asn1crypto = [ + {file = "asn1crypto-1.4.0-py2.py3-none-any.whl", hash = "sha256:4bcdf33c861c7d40bdcd74d8e4dd7661aac320fcdf40b9a3f95b4ee12fde2fa8"}, + {file = "asn1crypto-1.4.0.tar.gz", hash = "sha256:f4f6e119474e58e04a2b1af817eb585b4fd72bdd89b998624712b5c99be7641c"}, ] async-timeout = [ {file = "async-timeout-3.0.1.tar.gz", hash = "sha256:0c3c816a028d47f659d6ff5c745cb2acf1f966da1fe5c19c77a70282b25f4c5f"}, @@ -621,125 +894,193 @@ attrs = [ {file = "attrs-21.2.0-py2.py3-none-any.whl", hash = "sha256:149e90d6d8ac20db7a955ad60cf0e6881a3f20d37096140088356da6c716b0b1"}, {file = "attrs-21.2.0.tar.gz", hash = "sha256:ef6aaac3ca6cd92904cdd0d83f629a15f18053ec84e6432106f7a4d04ae4f5fb"}, ] +"backports.entry-points-selectable" = [ + {file = "backports.entry_points_selectable-1.1.0-py2.py3-none-any.whl", hash = "sha256:a6d9a871cde5e15b4c4a53e3d43ba890cc6861ec1332c9c2428c92f977192acc"}, + {file = "backports.entry_points_selectable-1.1.0.tar.gz", hash = "sha256:988468260ec1c196dab6ae1149260e2f5472c9110334e5d51adcb77867361f6a"}, +] black = [ - {file = "black-21.5b2-py3-none-any.whl", hash = "sha256:e5cf21ebdffc7a9b29d73912b6a6a9a4df4ce70220d523c21647da2eae0751ef"}, - {file = "black-21.5b2.tar.gz", hash = "sha256:1fc0e0a2c8ae7d269dfcf0c60a89afa299664f3e811395d40b1922dff8f854b5"}, + {file = "black-21.9b0-py3-none-any.whl", hash = "sha256:380f1b5da05e5a1429225676655dddb96f5ae8c75bdf91e53d798871b902a115"}, + {file = "black-21.9b0.tar.gz", hash = "sha256:7de4cfc7eb6b710de325712d40125689101d21d25283eed7e9998722cf10eb91"}, ] certifi = [ - {file = "certifi-2021.5.30-py2.py3-none-any.whl", hash = "sha256:50b1e4f8446b06f41be7dd6338db18e0990601dce795c2b1686458aa7e8fa7d8"}, - {file = "certifi-2021.5.30.tar.gz", hash = "sha256:2bbf76fd432960138b3ef6dda3dde0544f27cbf8546c458e60baf371917ba9ee"}, + {file = "certifi-2021.10.8-py2.py3-none-any.whl", hash = "sha256:d62a0163eb4c2344ac042ab2bdf75399a71a2d8c7d47eac2e2ee91b9d6339569"}, + {file = "certifi-2021.10.8.tar.gz", hash = "sha256:78884e7c1d4b00ce3cea67b44566851c4343c120abd683433ce934a68ea58872"}, ] cffi = [ - {file = "cffi-1.14.5-cp27-cp27m-macosx_10_9_x86_64.whl", hash = "sha256:bb89f306e5da99f4d922728ddcd6f7fcebb3241fc40edebcb7284d7514741991"}, - {file = "cffi-1.14.5-cp27-cp27m-manylinux1_i686.whl", hash = "sha256:34eff4b97f3d982fb93e2831e6750127d1355a923ebaeeb565407b3d2f8d41a1"}, - {file = "cffi-1.14.5-cp27-cp27m-manylinux1_x86_64.whl", hash = "sha256:99cd03ae7988a93dd00bcd9d0b75e1f6c426063d6f03d2f90b89e29b25b82dfa"}, - {file = "cffi-1.14.5-cp27-cp27m-win32.whl", hash = "sha256:65fa59693c62cf06e45ddbb822165394a288edce9e276647f0046e1ec26920f3"}, - {file = "cffi-1.14.5-cp27-cp27m-win_amd64.whl", hash = "sha256:51182f8927c5af975fece87b1b369f722c570fe169f9880764b1ee3bca8347b5"}, - {file = "cffi-1.14.5-cp27-cp27mu-manylinux1_i686.whl", hash = "sha256:43e0b9d9e2c9e5d152946b9c5fe062c151614b262fda2e7b201204de0b99e482"}, - {file = "cffi-1.14.5-cp27-cp27mu-manylinux1_x86_64.whl", hash = "sha256:cbde590d4faaa07c72bf979734738f328d239913ba3e043b1e98fe9a39f8b2b6"}, - {file = "cffi-1.14.5-cp35-cp35m-macosx_10_9_x86_64.whl", hash = "sha256:5de7970188bb46b7bf9858eb6890aad302577a5f6f75091fd7cdd3ef13ef3045"}, - {file = "cffi-1.14.5-cp35-cp35m-manylinux1_i686.whl", hash = "sha256:a465da611f6fa124963b91bf432d960a555563efe4ed1cc403ba5077b15370aa"}, - {file = "cffi-1.14.5-cp35-cp35m-manylinux1_x86_64.whl", hash = "sha256:d42b11d692e11b6634f7613ad8df5d6d5f8875f5d48939520d351007b3c13406"}, - {file = "cffi-1.14.5-cp35-cp35m-win32.whl", hash = "sha256:72d8d3ef52c208ee1c7b2e341f7d71c6fd3157138abf1a95166e6165dd5d4369"}, - {file = "cffi-1.14.5-cp35-cp35m-win_amd64.whl", hash = "sha256:29314480e958fd8aab22e4a58b355b629c59bf5f2ac2492b61e3dc06d8c7a315"}, - {file = "cffi-1.14.5-cp36-cp36m-macosx_10_9_x86_64.whl", hash = "sha256:3d3dd4c9e559eb172ecf00a2a7517e97d1e96de2a5e610bd9b68cea3925b4892"}, - {file = "cffi-1.14.5-cp36-cp36m-manylinux1_i686.whl", hash = "sha256:48e1c69bbacfc3d932221851b39d49e81567a4d4aac3b21258d9c24578280058"}, - {file = "cffi-1.14.5-cp36-cp36m-manylinux1_x86_64.whl", hash = "sha256:69e395c24fc60aad6bb4fa7e583698ea6cc684648e1ffb7fe85e3c1ca131a7d5"}, - {file = "cffi-1.14.5-cp36-cp36m-manylinux2014_aarch64.whl", hash = "sha256:9e93e79c2551ff263400e1e4be085a1210e12073a31c2011dbbda14bda0c6132"}, - {file = "cffi-1.14.5-cp36-cp36m-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:24ec4ff2c5c0c8f9c6b87d5bb53555bf267e1e6f70e52e5a9740d32861d36b6f"}, - {file = "cffi-1.14.5-cp36-cp36m-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:3c3f39fa737542161d8b0d680df2ec249334cd70a8f420f71c9304bd83c3cbed"}, - {file = "cffi-1.14.5-cp36-cp36m-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:681d07b0d1e3c462dd15585ef5e33cb021321588bebd910124ef4f4fb71aef55"}, - {file = "cffi-1.14.5-cp36-cp36m-win32.whl", hash = "sha256:58e3f59d583d413809d60779492342801d6e82fefb89c86a38e040c16883be53"}, - {file = "cffi-1.14.5-cp36-cp36m-win_amd64.whl", hash = "sha256:005a36f41773e148deac64b08f233873a4d0c18b053d37da83f6af4d9087b813"}, - {file = "cffi-1.14.5-cp37-cp37m-macosx_10_9_x86_64.whl", hash = "sha256:2894f2df484ff56d717bead0a5c2abb6b9d2bf26d6960c4604d5c48bbc30ee73"}, - {file = "cffi-1.14.5-cp37-cp37m-manylinux1_i686.whl", hash = "sha256:0857f0ae312d855239a55c81ef453ee8fd24136eaba8e87a2eceba644c0d4c06"}, - {file = "cffi-1.14.5-cp37-cp37m-manylinux1_x86_64.whl", hash = "sha256:cd2868886d547469123fadc46eac7ea5253ea7fcb139f12e1dfc2bbd406427d1"}, - {file = "cffi-1.14.5-cp37-cp37m-manylinux2014_aarch64.whl", hash = "sha256:35f27e6eb43380fa080dccf676dece30bef72e4a67617ffda586641cd4508d49"}, - {file = "cffi-1.14.5-cp37-cp37m-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:06d7cd1abac2ffd92e65c0609661866709b4b2d82dd15f611e602b9b188b0b69"}, - {file = "cffi-1.14.5-cp37-cp37m-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:0f861a89e0043afec2a51fd177a567005847973be86f709bbb044d7f42fc4e05"}, - {file = "cffi-1.14.5-cp37-cp37m-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:cc5a8e069b9ebfa22e26d0e6b97d6f9781302fe7f4f2b8776c3e1daea35f1adc"}, - {file = "cffi-1.14.5-cp37-cp37m-win32.whl", hash = "sha256:9ff227395193126d82e60319a673a037d5de84633f11279e336f9c0f189ecc62"}, - {file = "cffi-1.14.5-cp37-cp37m-win_amd64.whl", hash = "sha256:9cf8022fb8d07a97c178b02327b284521c7708d7c71a9c9c355c178ac4bbd3d4"}, - {file = "cffi-1.14.5-cp38-cp38-macosx_10_9_x86_64.whl", hash = "sha256:8b198cec6c72df5289c05b05b8b0969819783f9418e0409865dac47288d2a053"}, - {file = "cffi-1.14.5-cp38-cp38-manylinux1_i686.whl", hash = "sha256:ad17025d226ee5beec591b52800c11680fca3df50b8b29fe51d882576e039ee0"}, - {file = "cffi-1.14.5-cp38-cp38-manylinux1_x86_64.whl", hash = "sha256:6c97d7350133666fbb5cf4abdc1178c812cb205dc6f41d174a7b0f18fb93337e"}, - {file = "cffi-1.14.5-cp38-cp38-manylinux2014_aarch64.whl", hash = "sha256:8ae6299f6c68de06f136f1f9e69458eae58f1dacf10af5c17353eae03aa0d827"}, - {file = "cffi-1.14.5-cp38-cp38-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:04c468b622ed31d408fea2346bec5bbffba2cc44226302a0de1ade9f5ea3d373"}, - {file = "cffi-1.14.5-cp38-cp38-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:06db6321b7a68b2bd6df96d08a5adadc1fa0e8f419226e25b2a5fbf6ccc7350f"}, - {file = "cffi-1.14.5-cp38-cp38-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:293e7ea41280cb28c6fcaaa0b1aa1f533b8ce060b9e701d78511e1e6c4a1de76"}, - {file = "cffi-1.14.5-cp38-cp38-win32.whl", hash = "sha256:b85eb46a81787c50650f2392b9b4ef23e1f126313b9e0e9013b35c15e4288e2e"}, - {file = "cffi-1.14.5-cp38-cp38-win_amd64.whl", hash = "sha256:1f436816fc868b098b0d63b8920de7d208c90a67212546d02f84fe78a9c26396"}, - {file = "cffi-1.14.5-cp39-cp39-macosx_10_9_x86_64.whl", hash = "sha256:1071534bbbf8cbb31b498d5d9db0f274f2f7a865adca4ae429e147ba40f73dea"}, - {file = "cffi-1.14.5-cp39-cp39-manylinux1_i686.whl", hash = "sha256:9de2e279153a443c656f2defd67769e6d1e4163952b3c622dcea5b08a6405322"}, - {file = "cffi-1.14.5-cp39-cp39-manylinux1_x86_64.whl", hash = "sha256:6e4714cc64f474e4d6e37cfff31a814b509a35cb17de4fb1999907575684479c"}, - {file = "cffi-1.14.5-cp39-cp39-manylinux2014_aarch64.whl", hash = "sha256:158d0d15119b4b7ff6b926536763dc0714313aa59e320ddf787502c70c4d4bee"}, - {file = "cffi-1.14.5-cp39-cp39-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:1bf1ac1984eaa7675ca8d5745a8cb87ef7abecb5592178406e55858d411eadc0"}, - {file = "cffi-1.14.5-cp39-cp39-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:df5052c5d867c1ea0b311fb7c3cd28b19df469c056f7fdcfe88c7473aa63e333"}, - {file = "cffi-1.14.5-cp39-cp39-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:24a570cd11895b60829e941f2613a4f79df1a27344cbbb82164ef2e0116f09c7"}, - {file = "cffi-1.14.5-cp39-cp39-win32.whl", hash = "sha256:afb29c1ba2e5a3736f1c301d9d0abe3ec8b86957d04ddfa9d7a6a42b9367e396"}, - {file = "cffi-1.14.5-cp39-cp39-win_amd64.whl", hash = "sha256:f2d45f97ab6bb54753eab54fffe75aaf3de4ff2341c9daee1987ee1837636f1d"}, - {file = "cffi-1.14.5.tar.gz", hash = "sha256:fd78e5fee591709f32ef6edb9a015b4aa1a5022598e36227500c8f4e02328d9c"}, + {file = "cffi-1.15.0-cp27-cp27m-macosx_10_9_x86_64.whl", hash = "sha256:c2502a1a03b6312837279c8c1bd3ebedf6c12c4228ddbad40912d671ccc8a962"}, + {file = "cffi-1.15.0-cp27-cp27m-manylinux1_i686.whl", hash = "sha256:23cfe892bd5dd8941608f93348c0737e369e51c100d03718f108bf1add7bd6d0"}, + {file = "cffi-1.15.0-cp27-cp27m-manylinux1_x86_64.whl", hash = "sha256:41d45de54cd277a7878919867c0f08b0cf817605e4eb94093e7516505d3c8d14"}, + {file = "cffi-1.15.0-cp27-cp27m-win32.whl", hash = "sha256:4a306fa632e8f0928956a41fa8e1d6243c71e7eb59ffbd165fc0b41e316b2474"}, + {file = "cffi-1.15.0-cp27-cp27m-win_amd64.whl", hash = "sha256:e7022a66d9b55e93e1a845d8c9eba2a1bebd4966cd8bfc25d9cd07d515b33fa6"}, + {file = "cffi-1.15.0-cp27-cp27mu-manylinux1_i686.whl", hash = "sha256:14cd121ea63ecdae71efa69c15c5543a4b5fbcd0bbe2aad864baca0063cecf27"}, + {file = "cffi-1.15.0-cp27-cp27mu-manylinux1_x86_64.whl", hash = "sha256:d4d692a89c5cf08a8557fdeb329b82e7bf609aadfaed6c0d79f5a449a3c7c023"}, + {file = "cffi-1.15.0-cp310-cp310-macosx_10_9_x86_64.whl", hash = "sha256:0104fb5ae2391d46a4cb082abdd5c69ea4eab79d8d44eaaf79f1b1fd806ee4c2"}, + {file = "cffi-1.15.0-cp310-cp310-macosx_11_0_arm64.whl", hash = "sha256:91ec59c33514b7c7559a6acda53bbfe1b283949c34fe7440bcf917f96ac0723e"}, + {file = "cffi-1.15.0-cp310-cp310-manylinux_2_12_i686.manylinux2010_i686.whl", hash = "sha256:f5c7150ad32ba43a07c4479f40241756145a1f03b43480e058cfd862bf5041c7"}, + {file = "cffi-1.15.0-cp310-cp310-manylinux_2_12_x86_64.manylinux2010_x86_64.whl", hash = "sha256:00c878c90cb53ccfaae6b8bc18ad05d2036553e6d9d1d9dbcf323bbe83854ca3"}, + {file = "cffi-1.15.0-cp310-cp310-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:abb9a20a72ac4e0fdb50dae135ba5e77880518e742077ced47eb1499e29a443c"}, + {file = "cffi-1.15.0-cp310-cp310-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:a5263e363c27b653a90078143adb3d076c1a748ec9ecc78ea2fb916f9b861962"}, + {file = "cffi-1.15.0-cp310-cp310-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:f54a64f8b0c8ff0b64d18aa76675262e1700f3995182267998c31ae974fbc382"}, + {file = "cffi-1.15.0-cp310-cp310-win32.whl", hash = "sha256:c21c9e3896c23007803a875460fb786118f0cdd4434359577ea25eb556e34c55"}, + {file = "cffi-1.15.0-cp310-cp310-win_amd64.whl", hash = "sha256:5e069f72d497312b24fcc02073d70cb989045d1c91cbd53979366077959933e0"}, + {file = "cffi-1.15.0-cp36-cp36m-macosx_10_9_x86_64.whl", hash = "sha256:64d4ec9f448dfe041705426000cc13e34e6e5bb13736e9fd62e34a0b0c41566e"}, + {file = "cffi-1.15.0-cp36-cp36m-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:2756c88cbb94231c7a147402476be2c4df2f6078099a6f4a480d239a8817ae39"}, + {file = "cffi-1.15.0-cp36-cp36m-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:3b96a311ac60a3f6be21d2572e46ce67f09abcf4d09344c49274eb9e0bf345fc"}, + {file = "cffi-1.15.0-cp36-cp36m-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:75e4024375654472cc27e91cbe9eaa08567f7fbdf822638be2814ce059f58032"}, + {file = "cffi-1.15.0-cp36-cp36m-manylinux_2_5_i686.manylinux1_i686.whl", hash = "sha256:59888172256cac5629e60e72e86598027aca6bf01fa2465bdb676d37636573e8"}, + {file = "cffi-1.15.0-cp36-cp36m-manylinux_2_5_x86_64.manylinux1_x86_64.whl", hash = "sha256:27c219baf94952ae9d50ec19651a687b826792055353d07648a5695413e0c605"}, + {file = "cffi-1.15.0-cp36-cp36m-win32.whl", hash = "sha256:4958391dbd6249d7ad855b9ca88fae690783a6be9e86df65865058ed81fc860e"}, + {file = "cffi-1.15.0-cp36-cp36m-win_amd64.whl", hash = "sha256:f6f824dc3bce0edab5f427efcfb1d63ee75b6fcb7282900ccaf925be84efb0fc"}, + {file = "cffi-1.15.0-cp37-cp37m-macosx_10_9_x86_64.whl", hash = "sha256:06c48159c1abed75c2e721b1715c379fa3200c7784271b3c46df01383b593636"}, + {file = "cffi-1.15.0-cp37-cp37m-manylinux_2_12_i686.manylinux2010_i686.whl", hash = "sha256:c2051981a968d7de9dd2d7b87bcb9c939c74a34626a6e2f8181455dd49ed69e4"}, + {file = "cffi-1.15.0-cp37-cp37m-manylinux_2_12_x86_64.manylinux2010_x86_64.whl", hash = "sha256:fd8a250edc26254fe5b33be00402e6d287f562b6a5b2152dec302fa15bb3e997"}, + {file = "cffi-1.15.0-cp37-cp37m-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:91d77d2a782be4274da750752bb1650a97bfd8f291022b379bb8e01c66b4e96b"}, + {file = "cffi-1.15.0-cp37-cp37m-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:45db3a33139e9c8f7c09234b5784a5e33d31fd6907800b316decad50af323ff2"}, + {file = "cffi-1.15.0-cp37-cp37m-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:263cc3d821c4ab2213cbe8cd8b355a7f72a8324577dc865ef98487c1aeee2bc7"}, + {file = "cffi-1.15.0-cp37-cp37m-win32.whl", hash = "sha256:17771976e82e9f94976180f76468546834d22a7cc404b17c22df2a2c81db0c66"}, + {file = "cffi-1.15.0-cp37-cp37m-win_amd64.whl", hash = "sha256:3415c89f9204ee60cd09b235810be700e993e343a408693e80ce7f6a40108029"}, + {file = "cffi-1.15.0-cp38-cp38-macosx_10_9_x86_64.whl", hash = "sha256:4238e6dab5d6a8ba812de994bbb0a79bddbdf80994e4ce802b6f6f3142fcc880"}, + {file = "cffi-1.15.0-cp38-cp38-manylinux_2_12_i686.manylinux2010_i686.whl", hash = "sha256:0808014eb713677ec1292301ea4c81ad277b6cdf2fdd90fd540af98c0b101d20"}, + {file = "cffi-1.15.0-cp38-cp38-manylinux_2_12_x86_64.manylinux2010_x86_64.whl", hash = "sha256:57e9ac9ccc3101fac9d6014fba037473e4358ef4e89f8e181f8951a2c0162024"}, + {file = "cffi-1.15.0-cp38-cp38-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:8b6c2ea03845c9f501ed1313e78de148cd3f6cad741a75d43a29b43da27f2e1e"}, + {file = "cffi-1.15.0-cp38-cp38-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:10dffb601ccfb65262a27233ac273d552ddc4d8ae1bf93b21c94b8511bffe728"}, + {file = "cffi-1.15.0-cp38-cp38-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:786902fb9ba7433aae840e0ed609f45c7bcd4e225ebb9c753aa39725bb3e6ad6"}, + {file = "cffi-1.15.0-cp38-cp38-win32.whl", hash = "sha256:da5db4e883f1ce37f55c667e5c0de439df76ac4cb55964655906306918e7363c"}, + {file = "cffi-1.15.0-cp38-cp38-win_amd64.whl", hash = "sha256:181dee03b1170ff1969489acf1c26533710231c58f95534e3edac87fff06c443"}, + {file = "cffi-1.15.0-cp39-cp39-macosx_10_9_x86_64.whl", hash = "sha256:45e8636704eacc432a206ac7345a5d3d2c62d95a507ec70d62f23cd91770482a"}, + {file = "cffi-1.15.0-cp39-cp39-macosx_11_0_arm64.whl", hash = "sha256:31fb708d9d7c3f49a60f04cf5b119aeefe5644daba1cd2a0fe389b674fd1de37"}, + {file = "cffi-1.15.0-cp39-cp39-manylinux_2_12_i686.manylinux2010_i686.whl", hash = "sha256:6dc2737a3674b3e344847c8686cf29e500584ccad76204efea14f451d4cc669a"}, + {file = "cffi-1.15.0-cp39-cp39-manylinux_2_12_x86_64.manylinux2010_x86_64.whl", hash = "sha256:74fdfdbfdc48d3f47148976f49fab3251e550a8720bebc99bf1483f5bfb5db3e"}, + {file = "cffi-1.15.0-cp39-cp39-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:ffaa5c925128e29efbde7301d8ecaf35c8c60ffbcd6a1ffd3a552177c8e5e796"}, + {file = "cffi-1.15.0-cp39-cp39-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:3f7d084648d77af029acb79a0ff49a0ad7e9d09057a9bf46596dac9514dc07df"}, + {file = "cffi-1.15.0-cp39-cp39-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:ef1f279350da2c586a69d32fc8733092fd32cc8ac95139a00377841f59a3f8d8"}, + {file = "cffi-1.15.0-cp39-cp39-win32.whl", hash = "sha256:2a23af14f408d53d5e6cd4e3d9a24ff9e05906ad574822a10563efcef137979a"}, + {file = "cffi-1.15.0-cp39-cp39-win_amd64.whl", hash = "sha256:3773c4d81e6e818df2efbc7dd77325ca0dcb688116050fb2b3011218eda36139"}, + {file = "cffi-1.15.0.tar.gz", hash = "sha256:920f0d66a896c2d99f0adbb391f990a84091179542c205fa53ce5787aff87954"}, ] cfgv = [ - {file = "cfgv-3.3.0-py2.py3-none-any.whl", hash = "sha256:b449c9c6118fe8cca7fa5e00b9ec60ba08145d281d52164230a69211c5d597a1"}, - {file = "cfgv-3.3.0.tar.gz", hash = "sha256:9e600479b3b99e8af981ecdfc80a0296104ee610cab48a5ae4ffd0b668650eb1"}, + {file = "cfgv-3.3.1-py2.py3-none-any.whl", hash = "sha256:c6a0883f3917a037485059700b9e75da2464e6c27051014ad85ba6aaa5884426"}, + {file = "cfgv-3.3.1.tar.gz", hash = "sha256:f5a830efb9ce7a445376bb66ec94c638a9787422f96264c98edc6bdeed8ab736"}, ] chardet = [ {file = "chardet-4.0.0-py2.py3-none-any.whl", hash = "sha256:f864054d66fd9118f2e67044ac8981a54775ec5b67aed0441892edb553d21da5"}, {file = "chardet-4.0.0.tar.gz", hash = "sha256:0d6f53a15db4120f2b08c94f11e7d93d2c911ee118b6b30a04ec3ee8310179fa"}, ] +charset-normalizer = [ + {file = "charset-normalizer-2.0.7.tar.gz", hash = "sha256:e019de665e2bcf9c2b64e2e5aa025fa991da8720daa3c1138cadd2fd1856aed0"}, + {file = "charset_normalizer-2.0.7-py3-none-any.whl", hash = "sha256:f7af805c321bfa1ce6714c51f254e0d5bb5e5834039bc17db7ebe3a4cec9492b"}, +] click = [ - {file = "click-8.0.1-py3-none-any.whl", hash = "sha256:fba402a4a47334742d782209a7c79bc448911afe1149d07bdabdf480b3e2f4b6"}, - {file = "click-8.0.1.tar.gz", hash = "sha256:8c04c11192119b1ef78ea049e0a6f0463e4c48ef00a30160c704337586f3ad7a"}, + {file = "click-8.0.3-py3-none-any.whl", hash = "sha256:353f466495adaeb40b6b5f592f9f91cb22372351c84caeb068132442a4518ef3"}, + {file = "click-8.0.3.tar.gz", hash = "sha256:410e932b050f5eed773c4cda94de75971c89cdb3155a72a0831139a79e5ecb5b"}, ] colorama = [ {file = "colorama-0.4.4-py2.py3-none-any.whl", hash = "sha256:9f47eda37229f68eee03b24b9748937c7dc3868f906e8ba69fbcbdd3bc5dc3e2"}, {file = "colorama-0.4.4.tar.gz", hash = "sha256:5941b2b48a20143d2267e95b1c2a7603ce057ee39fd88e7329b0c292aa16869b"}, ] cryptography = [ - {file = "cryptography-3.4.7-cp36-abi3-macosx_10_10_x86_64.whl", hash = "sha256:3d8427734c781ea5f1b41d6589c293089704d4759e34597dce91014ac125aad1"}, - {file = "cryptography-3.4.7-cp36-abi3-macosx_11_0_arm64.whl", hash = "sha256:8e56e16617872b0957d1c9742a3f94b43533447fd78321514abbe7db216aa250"}, - {file = "cryptography-3.4.7-cp36-abi3-manylinux2010_x86_64.whl", hash = "sha256:37340614f8a5d2fb9aeea67fd159bfe4f5f4ed535b1090ce8ec428b2f15a11f2"}, - {file = "cryptography-3.4.7-cp36-abi3-manylinux2014_aarch64.whl", hash = "sha256:240f5c21aef0b73f40bb9f78d2caff73186700bf1bc6b94285699aff98cc16c6"}, - {file = "cryptography-3.4.7-cp36-abi3-manylinux2014_x86_64.whl", hash = "sha256:1e056c28420c072c5e3cb36e2b23ee55e260cb04eee08f702e0edfec3fb51959"}, - {file = "cryptography-3.4.7-cp36-abi3-win32.whl", hash = "sha256:0f1212a66329c80d68aeeb39b8a16d54ef57071bf22ff4e521657b27372e327d"}, - {file = "cryptography-3.4.7-cp36-abi3-win_amd64.whl", hash = "sha256:de4e5f7f68220d92b7637fc99847475b59154b7a1b3868fb7385337af54ac9ca"}, - {file = "cryptography-3.4.7-pp36-pypy36_pp73-manylinux2010_x86_64.whl", hash = "sha256:26965837447f9c82f1855e0bc8bc4fb910240b6e0d16a664bb722df3b5b06873"}, - {file = "cryptography-3.4.7-pp36-pypy36_pp73-manylinux2014_x86_64.whl", hash = "sha256:eb8cc2afe8b05acbd84a43905832ec78e7b3873fb124ca190f574dca7389a87d"}, - {file = "cryptography-3.4.7-pp37-pypy37_pp73-manylinux2010_x86_64.whl", hash = "sha256:7ec5d3b029f5fa2b179325908b9cd93db28ab7b85bb6c1db56b10e0b54235177"}, - {file = "cryptography-3.4.7-pp37-pypy37_pp73-manylinux2014_x86_64.whl", hash = "sha256:ee77aa129f481be46f8d92a1a7db57269a2f23052d5f2433b4621bb457081cc9"}, - {file = "cryptography-3.4.7.tar.gz", hash = "sha256:3d10de8116d25649631977cb37da6cbdd2d6fa0e0281d014a5b7d337255ca713"}, + {file = "cryptography-2.3-cp27-cp27m-macosx_10_6_intel.whl", hash = "sha256:87d092a7c2a44e5f7414ab02fb4145723ebba411425e1a99773531dd4c0e9b8d"}, + {file = "cryptography-2.3-cp27-cp27m-manylinux1_i686.whl", hash = "sha256:67f7f57eae8dede577f3f7775957f5bec93edd6bdb6ce597bb5b28e1bdf3d4fb"}, + {file = "cryptography-2.3-cp27-cp27m-manylinux1_x86_64.whl", hash = "sha256:27bb401a20a838d6d0ea380f08c6ead3ccd8c9d8a0232dc9adcc0e4994576a66"}, + {file = "cryptography-2.3-cp27-cp27m-win32.whl", hash = "sha256:7ba834564daef87557e7fcd35c3c3183a4147b0b3a57314e53317360b9b201b3"}, + {file = "cryptography-2.3-cp27-cp27m-win_amd64.whl", hash = "sha256:9449f5d4d7c516a6118fa9210c4a00f34384cb1d2028672100ee0c6cce49d7f6"}, + {file = "cryptography-2.3-cp27-cp27mu-manylinux1_i686.whl", hash = "sha256:c3d945b7b577f07a477700f618f46cbc287af3a9222cd73035c6ef527ef2c363"}, + {file = "cryptography-2.3-cp27-cp27mu-manylinux1_x86_64.whl", hash = "sha256:bc2301170986ad82d9349a91eb8884e0e191209c45f5541b16aa7c0cfb135978"}, + {file = "cryptography-2.3-cp34-abi3-macosx_10_6_intel.whl", hash = "sha256:d01dfc5c2b3495184f683574e03c70022674ca9a7be88589c5aba130d835ea90"}, + {file = "cryptography-2.3-cp34-abi3-manylinux1_i686.whl", hash = "sha256:2a35b7570d8f247889784010aac8b384fd2e4a47b33e15c4a60b45a7c1944120"}, + {file = "cryptography-2.3-cp34-abi3-manylinux1_x86_64.whl", hash = "sha256:5eb86f03f9c4f0ac2336ac5431271072ddf7ecc76b338e26366732cfac58aa19"}, + {file = "cryptography-2.3-cp34-cp34m-win32.whl", hash = "sha256:29720c4253263cff9aea64585adbbe85013ba647f6e98367efff9db2d7193ded"}, + {file = "cryptography-2.3-cp34-cp34m-win_amd64.whl", hash = "sha256:82409f5150e529d699e5c33fa8fd85e965104db03bc564f5f4b6a9199e591f7c"}, + {file = "cryptography-2.3-cp35-cp35m-win32.whl", hash = "sha256:7d7f084cbe1fdb82be5a0545062b59b1ad3637bc5a48612ac2eb428ff31b31ea"}, + {file = "cryptography-2.3-cp35-cp35m-win_amd64.whl", hash = "sha256:6ec84edcbc966ae460560a51a90046503ff0b5b66157a9efc61515c68059f6c8"}, + {file = "cryptography-2.3-cp36-cp36m-win32.whl", hash = "sha256:8c56ef989342e42b9fcaba7c74b446f0cc9bed546dd00034fa7ad66fc00307ef"}, + {file = "cryptography-2.3-cp36-cp36m-win_amd64.whl", hash = "sha256:42c531a6a354407f42ee07fda5c2c0dc822cf6d52744949c182f2b295fbd4183"}, + {file = "cryptography-2.3-cp37-cp37m-win32.whl", hash = "sha256:21af753934f2f6d1a10fe8f4c0a64315af209ef6adeaee63ca349797d747d687"}, + {file = "cryptography-2.3-cp37-cp37m-win_amd64.whl", hash = "sha256:cee18beb4c807b5c0b178f4fa2fae03cef9d51821a358c6890f8b23465b7e5d2"}, + {file = "cryptography-2.3.tar.gz", hash = "sha256:c132bab45d4bd0fff1d3fe294d92b0a6eb8404e93337b3127bdec9f21de117e6"}, +] +dateutils = [ + {file = "dateutils-0.6.12-py2.py3-none-any.whl", hash = "sha256:f33b6ab430fa4166e7e9cb8b21ee9f6c9843c48df1a964466f52c79b2a8d53b3"}, + {file = "dateutils-0.6.12.tar.gz", hash = "sha256:03dd90bcb21541bd4eb4b013637e4f1b5f944881c46cc6e4b67a6059e370e3f1"}, ] distlib = [ - {file = "distlib-0.3.2-py2.py3-none-any.whl", hash = "sha256:23e223426b28491b1ced97dc3bbe183027419dfc7982b4fa2f05d5f3ff10711c"}, - {file = "distlib-0.3.2.zip", hash = "sha256:106fef6dc37dd8c0e2c0a60d3fca3e77460a48907f335fa28420463a6f799736"}, + {file = "distlib-0.3.3-py2.py3-none-any.whl", hash = "sha256:c8b54e8454e5bf6237cc84c20e8264c3e991e824ef27e8f1e81049867d861e31"}, + {file = "distlib-0.3.3.zip", hash = "sha256:d982d0751ff6eaaab5e2ec8e691d949ee80eddf01a62eaa96ddb11531fe16b05"}, ] eight = [ {file = "eight-1.0.1-py2.py3-none-any.whl", hash = "sha256:bd6967ec7cf9e61ab18b55751236ed89a8363cfcb69127aace336869615245d5"}, {file = "eight-1.0.1.tar.gz", hash = "sha256:6465970d3c12a1d759151690629b10b50ffb124c9b62e912999ee9233378a562"}, ] filelock = [ - {file = "filelock-3.0.12-py3-none-any.whl", hash = "sha256:929b7d63ec5b7d6b71b0fa5ac14e030b3f70b75747cef1b10da9b879fef15836"}, - {file = "filelock-3.0.12.tar.gz", hash = "sha256:18d82244ee114f543149c66a6e0c14e9c4f8a1044b5cdaadd0f82159d6a6ff59"}, + {file = "filelock-3.3.1-py3-none-any.whl", hash = "sha256:2b5eb3589e7fdda14599e7eb1a50e09b4cc14f34ed98b8ba56d33bfaafcbef2f"}, + {file = "filelock-3.3.1.tar.gz", hash = "sha256:34a9f35f95c441e7b38209775d6e0337f9a3759f3565f6c5798f19618527c76f"}, ] future = [ {file = "future-0.18.2.tar.gz", hash = "sha256:b1bead90b70cf6ec3f0710ae53a525360fa360d306a86583adc6bf83a4db537d"}, ] +gevent = [ + {file = "gevent-20.6.1-cp27-cp27m-macosx_10_15_x86_64.whl", hash = "sha256:12565278b65f958686199fe7d06e3c2aed5f0587be2e66e62e90b2df0a62a6c7"}, + {file = "gevent-20.6.1-cp27-cp27m-manylinux2010_x86_64.whl", hash = "sha256:2fb496f3b4de6591a293a65d04a14b32881d9e4e424defaa7193769489bc7c6d"}, + {file = "gevent-20.6.1-cp27-cp27m-win32.whl", hash = "sha256:edc4c7213f9c4ee88ad5d03a8f24f3d80b4decd09129f4d1eb660d3068906a69"}, + {file = "gevent-20.6.1-cp27-cp27m-win_amd64.whl", hash = "sha256:261413e577a6c8ca41b9fdf5ab5370bd470e3b32e176ad18e786cc45175e3af4"}, + {file = "gevent-20.6.1-cp27-cp27mu-manylinux2010_x86_64.whl", hash = "sha256:9d33866360dfd9457f509aa249746deba14f8790ca3544c0f1317fda773e3495"}, + {file = "gevent-20.6.1-cp35-cp35m-manylinux2010_x86_64.whl", hash = "sha256:f1dc96268bc3357cfb1d490eaf322c29a3649e63b672b718658e1797bd38cd64"}, + {file = "gevent-20.6.1-cp35-cp35m-win32.whl", hash = "sha256:1f5407c045fcdcf16f3c1e032c09e868eab4e2971a13003cdc38944cfef4a661"}, + {file = "gevent-20.6.1-cp35-cp35m-win_amd64.whl", hash = "sha256:2d36ea41da5347e165f8988718dca481f12711d62d999db953101516852d63fa"}, + {file = "gevent-20.6.1-cp36-cp36m-macosx_10_15_x86_64.whl", hash = "sha256:efdb51bcde9d604ab76d11e08c1adc62a5e87944b5435dcfad030d4cbbfe9a0f"}, + {file = "gevent-20.6.1-cp36-cp36m-manylinux2010_x86_64.whl", hash = "sha256:0fd5fbdaf5de55c38c45584489f73f263fbb4e06dd22496e9ba75ff42f469437"}, + {file = "gevent-20.6.1-cp36-cp36m-win32.whl", hash = "sha256:9164c8d0c217cf1a290d549c51a06f1352cd2cc169dac36305f377169cd7828b"}, + {file = "gevent-20.6.1-cp36-cp36m-win_amd64.whl", hash = "sha256:ad47b503b590dd972f791f01539f8806af33dfe854cae1d577846f9689deeecd"}, + {file = "gevent-20.6.1-cp37-cp37m-macosx_10_15_x86_64.whl", hash = "sha256:50af4b6a54d3b2b631a73c5c0100e2366e66c6d48ad9634f3c8fbdc0fa537964"}, + {file = "gevent-20.6.1-cp37-cp37m-manylinux2010_x86_64.whl", hash = "sha256:4070b229638ddba19850e9e648727462c0c8f0dc4277a753ae84faa7ac489712"}, + {file = "gevent-20.6.1-cp37-cp37m-win32.whl", hash = "sha256:7cb2bef61e59da4c4a6a3520b88ce8b7d5c3f1006657bc1355a693490a5d597a"}, + {file = "gevent-20.6.1-cp37-cp37m-win_amd64.whl", hash = "sha256:75fcbe5b50ba3b55a7a0683f58f02ece099714429e2120bfaa62f1d70d9dbfa5"}, + {file = "gevent-20.6.1-cp38-cp38-macosx_10_15_x86_64.whl", hash = "sha256:db00eb55398844ff4a6dc78f4db80a7dd6acb5cdbd5a0de0d6acaddc52e9c82f"}, + {file = "gevent-20.6.1-cp38-cp38-manylinux2010_x86_64.whl", hash = "sha256:4776da6ba4a784d80b6d9b6055929ebc2e09c1d2a5876f38b32c7d6435b59eb6"}, + {file = "gevent-20.6.1-cp38-cp38-win32.whl", hash = "sha256:68cc23bbf79f80f5313f94be0d5679c6862ce5bc45b34eb927472a327798f691"}, + {file = "gevent-20.6.1-cp38-cp38-win_amd64.whl", hash = "sha256:48ab796f200d2730fded2669ca52e36b68a57d114420ef36adf301bc3f593398"}, + {file = "gevent-20.6.1-cp39-cp39-manylinux2010_x86_64.whl", hash = "sha256:a498902a736382406d401d174df66ba99124b064d5e30db0e61e8bf2aeaf7bb1"}, + {file = "gevent-20.6.1-pp27-pypy_73-win32.whl", hash = "sha256:691e8f96b6357657eb6cd4d76affdac6e1648b3be69e3841b0dc331270c3835c"}, + {file = "gevent-20.6.1.tar.gz", hash = "sha256:53fdc591806b1ed3a747a8a27839f7c6ae229155aad309e265682f0f398eb488"}, +] +greenlet = [ + {file = "greenlet-0.4.16-cp27-cp27m-manylinux1_x86_64.whl", hash = "sha256:80cb0380838bf4e48da6adedb0c7cd060c187bb4a75f67a5aa9ec33689b84872"}, + {file = "greenlet-0.4.16-cp27-cp27m-win32.whl", hash = "sha256:df7de669cbf21de4b04a3ffc9920bc8426cab4c61365fa84d79bf97401a8bef7"}, + {file = "greenlet-0.4.16-cp27-cp27m-win_amd64.whl", hash = "sha256:1429dc183b36ec972055e13250d96e174491559433eb3061691b446899b87384"}, + {file = "greenlet-0.4.16-cp27-cp27mu-manylinux1_x86_64.whl", hash = "sha256:5ea034d040e6ab1d2ae04ab05a3f37dbd719c4dee3804b13903d4cc794b1336e"}, + {file = "greenlet-0.4.16-cp35-cp35m-manylinux1_x86_64.whl", hash = "sha256:c196a5394c56352e21cb7224739c6dd0075b69dd56f758505951d1d8d68cf8a9"}, + {file = "greenlet-0.4.16-cp35-cp35m-win32.whl", hash = "sha256:1000038ba0ea9032948e2156a9c15f5686f36945e8f9906e6b8db49f358e7b52"}, + {file = "greenlet-0.4.16-cp35-cp35m-win_amd64.whl", hash = "sha256:1b805231bfb7b2900a16638c3c8b45c694334c811f84463e52451e00c9412691"}, + {file = "greenlet-0.4.16-cp36-cp36m-manylinux1_x86_64.whl", hash = "sha256:e5db19d4a7d41bbeb3dd89b49fc1bc7e6e515b51bbf32589c618655a0ebe0bf0"}, + {file = "greenlet-0.4.16-cp36-cp36m-win32.whl", hash = "sha256:eac2a3f659d5f41d6bbfb6a97733bc7800ea5e906dc873732e00cebb98cec9e4"}, + {file = "greenlet-0.4.16-cp36-cp36m-win_amd64.whl", hash = "sha256:7eed31f4efc8356e200568ba05ad645525f1fbd8674f1e5be61a493e715e3873"}, + {file = "greenlet-0.4.16-cp37-cp37m-manylinux1_x86_64.whl", hash = "sha256:682328aa576ec393c1872615bcb877cf32d800d4a2f150e1a5dc7e56644010b1"}, + {file = "greenlet-0.4.16-cp37-cp37m-win32.whl", hash = "sha256:3a35e33902b2e6079949feed7a2dafa5ac6f019da97bd255842bb22de3c11bf5"}, + {file = "greenlet-0.4.16-cp37-cp37m-win_amd64.whl", hash = "sha256:b0b2a984bbfc543d144d88caad6cc7ff4a71be77102014bd617bd88cfb038727"}, + {file = "greenlet-0.4.16-cp38-cp38-manylinux1_x86_64.whl", hash = "sha256:d83c1d38658b0f81c282b41238092ed89d8f93c6e342224ab73fb39e16848721"}, + {file = "greenlet-0.4.16-cp38-cp38-win32.whl", hash = "sha256:e695ac8c3efe124d998230b219eb51afb6ef10524a50b3c45109c4b77a8a3a92"}, + {file = "greenlet-0.4.16-cp38-cp38-win_amd64.whl", hash = "sha256:133ba06bad4e5f2f8bf6a0ac434e0fd686df749a86b3478903b92ec3a9c0c90b"}, + {file = "greenlet-0.4.16.tar.gz", hash = "sha256:6e06eac722676797e8fce4adb8ad3dc57a1bb3adfb0dd3fdf8306c055a38456c"}, +] +grequests = [ + {file = "grequests-0.6.0-py3-none-any.whl", hash = "sha256:6eff964416021bb1dee4182a56cc2b551bfa42d37820e6aee6f2efa00d43a061"}, + {file = "grequests-0.6.0.tar.gz", hash = "sha256:7dec890c6668e6755a1ea968565535867956639301268394d24df67b478df666"}, +] identify = [ - {file = "identify-2.2.7-py2.py3-none-any.whl", hash = "sha256:92d6ad08eca19ceb17576733759944b94c0761277ddc3acf65e75e57ef190e32"}, - {file = "identify-2.2.7.tar.gz", hash = "sha256:c29e74c3671fe9537715cb695148231d777170ca1498e1c30c675d4ea782afe9"}, + {file = "identify-2.3.0-py2.py3-none-any.whl", hash = "sha256:d1e82c83d063571bb88087676f81261a4eae913c492dafde184067c584bc7c05"}, + {file = "identify-2.3.0.tar.gz", hash = "sha256:fd08c97f23ceee72784081f1ce5125c8f53a02d3f2716dde79a6ab8f1039fea5"}, ] idna = [ - {file = "idna-3.2-py3-none-any.whl", hash = "sha256:14475042e284991034cb48e06f6851428fb14c4dc953acd9be9a5e95c7b6dd7a"}, - {file = "idna-3.2.tar.gz", hash = "sha256:467fbad99067910785144ce333826c71fb0e63a425657295239737f7ecd125f3"}, + {file = "idna-2.10-py2.py3-none-any.whl", hash = "sha256:b97d804b1e9b523befed77c48dacec60e6dcb0b5391d57af6a65a312a90648c0"}, + {file = "idna-2.10.tar.gz", hash = "sha256:b307872f855b18632ce0c21c5e45be78c0ea7ae4c15c828c20788b26921eb3f6"}, ] iniconfig = [ {file = "iniconfig-1.1.1-py2.py3-none-any.whl", hash = "sha256:011e24c64b7f47f6ebd835bb12a743f2fbe9a26d4cecaa7f53bc4f35ee9da8b3"}, {file = "iniconfig-1.1.1.tar.gz", hash = "sha256:bc3af051d7d14b2ee5ef9969666def0cd1a000e121eaea580d4a313df4b37f32"}, ] jinja2 = [ - {file = "Jinja2-3.0.1-py3-none-any.whl", hash = "sha256:1f06f2da51e7b56b8f238affdd6b4e2c61e39598a378cc49345bc1bd42a978a4"}, - {file = "Jinja2-3.0.1.tar.gz", hash = "sha256:703f484b47a6af502e743c9122595cc812b0271f661722403114f71a79d0f5a4"}, + {file = "Jinja2-3.0.2-py3-none-any.whl", hash = "sha256:8569982d3f0889eed11dd620c706d39b60c36d6d25843961f33f77fb6bc6b20c"}, + {file = "Jinja2-3.0.2.tar.gz", hash = "sha256:827a0e32839ab1600d4eb1c4c33ec5a8edfbc5cb42dafa13b81f182f97784b45"}, ] lxml = [ {file = "lxml-4.6.3-cp27-cp27m-macosx_10_9_x86_64.whl", hash = "sha256:df7c53783a46febb0e70f6b05df2ba104610f2fb0d27023409734a3ecbb78fb2"}, @@ -749,6 +1090,8 @@ lxml = [ {file = "lxml-4.6.3-cp27-cp27m-win_amd64.whl", hash = "sha256:8157dadbb09a34a6bd95a50690595e1fa0af1a99445e2744110e3dca7831c4ee"}, {file = "lxml-4.6.3-cp27-cp27mu-manylinux1_i686.whl", hash = "sha256:7728e05c35412ba36d3e9795ae8995e3c86958179c9770e65558ec3fdfd3724f"}, {file = "lxml-4.6.3-cp27-cp27mu-manylinux1_x86_64.whl", hash = "sha256:4bff24dfeea62f2e56f5bab929b4428ae6caba2d1eea0c2d6eb618e30a71e6d4"}, + {file = "lxml-4.6.3-cp310-cp310-manylinux_2_12_i686.manylinux2010_i686.manylinux_2_24_i686.whl", hash = "sha256:64812391546a18896adaa86c77c59a4998f33c24788cadc35789e55b727a37f4"}, + {file = "lxml-4.6.3-cp310-cp310-manylinux_2_17_x86_64.manylinux2014_x86_64.manylinux_2_24_x86_64.whl", hash = "sha256:c1a40c06fd5ba37ad39caa0b3144eb3772e813b5fb5b084198a985431c2f1e8d"}, {file = "lxml-4.6.3-cp35-cp35m-manylinux1_i686.whl", hash = "sha256:74f7d8d439b18fa4c385f3f5dfd11144bb87c1da034a466c5b5577d23a1d9b51"}, {file = "lxml-4.6.3-cp35-cp35m-manylinux1_x86_64.whl", hash = "sha256:f90ba11136bfdd25cae3951af8da2e95121c9b9b93727b1b896e3fa105b2f586"}, {file = "lxml-4.6.3-cp35-cp35m-manylinux2010_i686.whl", hash = "sha256:4c61b3a0db43a1607d6264166b230438f85bfed02e8cff20c22e564d0faff354"}, @@ -790,12 +1133,22 @@ lxml = [ {file = "lxml-4.6.3.tar.gz", hash = "sha256:39b78571b3b30645ac77b95f7c69d1bffc4cf8c3b157c435a34da72e78c82468"}, ] markupsafe = [ + {file = "MarkupSafe-2.0.1-cp310-cp310-macosx_10_9_universal2.whl", hash = "sha256:d8446c54dc28c01e5a2dbac5a25f071f6653e6e40f3a8818e8b45d790fe6ef53"}, + {file = "MarkupSafe-2.0.1-cp310-cp310-macosx_10_9_x86_64.whl", hash = "sha256:36bc903cbb393720fad60fc28c10de6acf10dc6cc883f3e24ee4012371399a38"}, + {file = "MarkupSafe-2.0.1-cp310-cp310-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:2d7d807855b419fc2ed3e631034685db6079889a1f01d5d9dac950f764da3dad"}, + {file = "MarkupSafe-2.0.1-cp310-cp310-manylinux_2_5_i686.manylinux1_i686.manylinux_2_12_i686.manylinux2010_i686.whl", hash = "sha256:add36cb2dbb8b736611303cd3bfcee00afd96471b09cda130da3581cbdc56a6d"}, + {file = "MarkupSafe-2.0.1-cp310-cp310-manylinux_2_5_x86_64.manylinux1_x86_64.manylinux_2_12_x86_64.manylinux2010_x86_64.whl", hash = "sha256:168cd0a3642de83558a5153c8bd34f175a9a6e7f6dc6384b9655d2697312a646"}, + {file = "MarkupSafe-2.0.1-cp310-cp310-win32.whl", hash = "sha256:99df47edb6bda1249d3e80fdabb1dab8c08ef3975f69aed437cb69d0a5de1e28"}, + {file = "MarkupSafe-2.0.1-cp310-cp310-win_amd64.whl", hash = "sha256:e0f138900af21926a02425cf736db95be9f4af72ba1bb21453432a07f6082134"}, {file = "MarkupSafe-2.0.1-cp36-cp36m-macosx_10_9_x86_64.whl", hash = "sha256:f9081981fe268bd86831e5c75f7de206ef275defcb82bc70740ae6dc507aee51"}, {file = "MarkupSafe-2.0.1-cp36-cp36m-manylinux1_i686.whl", hash = "sha256:0955295dd5eec6cb6cc2fe1698f4c6d84af2e92de33fbcac4111913cd100a6ff"}, {file = "MarkupSafe-2.0.1-cp36-cp36m-manylinux1_x86_64.whl", hash = "sha256:0446679737af14f45767963a1a9ef7620189912317d095f2d9ffa183a4d25d2b"}, {file = "MarkupSafe-2.0.1-cp36-cp36m-manylinux2010_i686.whl", hash = "sha256:f826e31d18b516f653fe296d967d700fddad5901ae07c622bb3705955e1faa94"}, {file = "MarkupSafe-2.0.1-cp36-cp36m-manylinux2010_x86_64.whl", hash = "sha256:fa130dd50c57d53368c9d59395cb5526eda596d3ffe36666cd81a44d56e48872"}, {file = "MarkupSafe-2.0.1-cp36-cp36m-manylinux2014_aarch64.whl", hash = "sha256:905fec760bd2fa1388bb5b489ee8ee5f7291d692638ea5f67982d968366bef9f"}, + {file = "MarkupSafe-2.0.1-cp36-cp36m-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:bf5d821ffabf0ef3533c39c518f3357b171a1651c1ff6827325e4489b0e46c3c"}, + {file = "MarkupSafe-2.0.1-cp36-cp36m-manylinux_2_5_i686.manylinux1_i686.manylinux_2_12_i686.manylinux2010_i686.whl", hash = "sha256:0d4b31cc67ab36e3392bbf3862cfbadac3db12bdd8b02a2731f509ed5b829724"}, + {file = "MarkupSafe-2.0.1-cp36-cp36m-manylinux_2_5_x86_64.manylinux1_x86_64.manylinux_2_12_x86_64.manylinux2010_x86_64.whl", hash = "sha256:baa1a4e8f868845af802979fcdbf0bb11f94f1cb7ced4c4b8a351bb60d108145"}, {file = "MarkupSafe-2.0.1-cp36-cp36m-win32.whl", hash = "sha256:6c4ca60fa24e85fe25b912b01e62cb969d69a23a5d5867682dd3e80b5b02581d"}, {file = "MarkupSafe-2.0.1-cp36-cp36m-win_amd64.whl", hash = "sha256:b2f4bf27480f5e5e8ce285a8c8fd176c0b03e93dcc6646477d4630e83440c6a9"}, {file = "MarkupSafe-2.0.1-cp37-cp37m-macosx_10_9_x86_64.whl", hash = "sha256:0717a7390a68be14b8c793ba258e075c6f4ca819f15edfc2a3a027c823718567"}, @@ -804,14 +1157,21 @@ markupsafe = [ {file = "MarkupSafe-2.0.1-cp37-cp37m-manylinux2010_i686.whl", hash = "sha256:d7f9850398e85aba693bb640262d3611788b1f29a79f0c93c565694658f4071f"}, {file = "MarkupSafe-2.0.1-cp37-cp37m-manylinux2010_x86_64.whl", hash = "sha256:6a7fae0dd14cf60ad5ff42baa2e95727c3d81ded453457771d02b7d2b3f9c0c2"}, {file = "MarkupSafe-2.0.1-cp37-cp37m-manylinux2014_aarch64.whl", hash = "sha256:b7f2d075102dc8c794cbde1947378051c4e5180d52d276987b8d28a3bd58c17d"}, + {file = "MarkupSafe-2.0.1-cp37-cp37m-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:e9936f0b261d4df76ad22f8fee3ae83b60d7c3e871292cd42f40b81b70afae85"}, + {file = "MarkupSafe-2.0.1-cp37-cp37m-manylinux_2_5_i686.manylinux1_i686.manylinux_2_12_i686.manylinux2010_i686.whl", hash = "sha256:2a7d351cbd8cfeb19ca00de495e224dea7e7d919659c2841bbb7f420ad03e2d6"}, + {file = "MarkupSafe-2.0.1-cp37-cp37m-manylinux_2_5_x86_64.manylinux1_x86_64.manylinux_2_12_x86_64.manylinux2010_x86_64.whl", hash = "sha256:60bf42e36abfaf9aff1f50f52644b336d4f0a3fd6d8a60ca0d054ac9f713a864"}, {file = "MarkupSafe-2.0.1-cp37-cp37m-win32.whl", hash = "sha256:a30e67a65b53ea0a5e62fe23682cfe22712e01f453b95233b25502f7c61cb415"}, {file = "MarkupSafe-2.0.1-cp37-cp37m-win_amd64.whl", hash = "sha256:611d1ad9a4288cf3e3c16014564df047fe08410e628f89805e475368bd304914"}, + {file = "MarkupSafe-2.0.1-cp38-cp38-macosx_10_9_universal2.whl", hash = "sha256:5bb28c636d87e840583ee3adeb78172efc47c8b26127267f54a9c0ec251d41a9"}, {file = "MarkupSafe-2.0.1-cp38-cp38-macosx_10_9_x86_64.whl", hash = "sha256:be98f628055368795d818ebf93da628541e10b75b41c559fdf36d104c5787066"}, {file = "MarkupSafe-2.0.1-cp38-cp38-manylinux1_i686.whl", hash = "sha256:1d609f577dc6e1aa17d746f8bd3c31aa4d258f4070d61b2aa5c4166c1539de35"}, {file = "MarkupSafe-2.0.1-cp38-cp38-manylinux1_x86_64.whl", hash = "sha256:7d91275b0245b1da4d4cfa07e0faedd5b0812efc15b702576d103293e252af1b"}, {file = "MarkupSafe-2.0.1-cp38-cp38-manylinux2010_i686.whl", hash = "sha256:01a9b8ea66f1658938f65b93a85ebe8bc016e6769611be228d797c9d998dd298"}, {file = "MarkupSafe-2.0.1-cp38-cp38-manylinux2010_x86_64.whl", hash = "sha256:47ab1e7b91c098ab893b828deafa1203de86d0bc6ab587b160f78fe6c4011f75"}, {file = "MarkupSafe-2.0.1-cp38-cp38-manylinux2014_aarch64.whl", hash = "sha256:97383d78eb34da7e1fa37dd273c20ad4320929af65d156e35a5e2d89566d9dfb"}, + {file = "MarkupSafe-2.0.1-cp38-cp38-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:6fcf051089389abe060c9cd7caa212c707e58153afa2c649f00346ce6d260f1b"}, + {file = "MarkupSafe-2.0.1-cp38-cp38-manylinux_2_5_i686.manylinux1_i686.manylinux_2_12_i686.manylinux2010_i686.whl", hash = "sha256:5855f8438a7d1d458206a2466bf82b0f104a3724bf96a1c781ab731e4201731a"}, + {file = "MarkupSafe-2.0.1-cp38-cp38-manylinux_2_5_x86_64.manylinux1_x86_64.manylinux_2_12_x86_64.manylinux2010_x86_64.whl", hash = "sha256:3dd007d54ee88b46be476e293f48c85048603f5f516008bee124ddd891398ed6"}, {file = "MarkupSafe-2.0.1-cp38-cp38-win32.whl", hash = "sha256:023cb26ec21ece8dc3907c0e8320058b2e0cb3c55cf9564da612bc325bed5e64"}, {file = "MarkupSafe-2.0.1-cp38-cp38-win_amd64.whl", hash = "sha256:984d76483eb32f1bcb536dc27e4ad56bba4baa70be32fa87152832cdd9db0833"}, {file = "MarkupSafe-2.0.1-cp39-cp39-macosx_10_9_universal2.whl", hash = "sha256:2ef54abee730b502252bcdf31b10dacb0a416229b72c18b19e24a4509f273d26"}, @@ -821,48 +1181,86 @@ markupsafe = [ {file = "MarkupSafe-2.0.1-cp39-cp39-manylinux2010_i686.whl", hash = "sha256:4efca8f86c54b22348a5467704e3fec767b2db12fc39c6d963168ab1d3fc9135"}, {file = "MarkupSafe-2.0.1-cp39-cp39-manylinux2010_x86_64.whl", hash = "sha256:ab3ef638ace319fa26553db0624c4699e31a28bb2a835c5faca8f8acf6a5a902"}, {file = "MarkupSafe-2.0.1-cp39-cp39-manylinux2014_aarch64.whl", hash = "sha256:f8ba0e8349a38d3001fae7eadded3f6606f0da5d748ee53cc1dab1d6527b9509"}, + {file = "MarkupSafe-2.0.1-cp39-cp39-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:c47adbc92fc1bb2b3274c4b3a43ae0e4573d9fbff4f54cd484555edbf030baf1"}, + {file = "MarkupSafe-2.0.1-cp39-cp39-manylinux_2_5_i686.manylinux1_i686.manylinux_2_12_i686.manylinux2010_i686.whl", hash = "sha256:37205cac2a79194e3750b0af2a5720d95f786a55ce7df90c3af697bfa100eaac"}, + {file = "MarkupSafe-2.0.1-cp39-cp39-manylinux_2_5_x86_64.manylinux1_x86_64.manylinux_2_12_x86_64.manylinux2010_x86_64.whl", hash = "sha256:1f2ade76b9903f39aa442b4aadd2177decb66525062db244b35d71d0ee8599b6"}, {file = "MarkupSafe-2.0.1-cp39-cp39-win32.whl", hash = "sha256:10f82115e21dc0dfec9ab5c0223652f7197feb168c940f3ef61563fc2d6beb74"}, {file = "MarkupSafe-2.0.1-cp39-cp39-win_amd64.whl", hash = "sha256:693ce3f9e70a6cf7d2fb9e6c9d8b204b6b39897a2c4a1aa65728d5ac97dcc1d8"}, {file = "MarkupSafe-2.0.1.tar.gz", hash = "sha256:594c67807fb16238b30c44bdf74f36c02cdf22d1c8cda91ef8a0ed8dabf5620a"}, ] multidict = [ - {file = "multidict-5.1.0-cp36-cp36m-macosx_10_14_x86_64.whl", hash = "sha256:b7993704f1a4b204e71debe6095150d43b2ee6150fa4f44d6d966ec356a8d61f"}, - {file = "multidict-5.1.0-cp36-cp36m-manylinux1_i686.whl", hash = "sha256:9dd6e9b1a913d096ac95d0399bd737e00f2af1e1594a787e00f7975778c8b2bf"}, - {file = "multidict-5.1.0-cp36-cp36m-manylinux2014_aarch64.whl", hash = "sha256:f21756997ad8ef815d8ef3d34edd98804ab5ea337feedcd62fb52d22bf531281"}, - {file = "multidict-5.1.0-cp36-cp36m-manylinux2014_i686.whl", hash = "sha256:1ab820665e67373de5802acae069a6a05567ae234ddb129f31d290fc3d1aa56d"}, - {file = "multidict-5.1.0-cp36-cp36m-manylinux2014_ppc64le.whl", hash = "sha256:9436dc58c123f07b230383083855593550c4d301d2532045a17ccf6eca505f6d"}, - {file = "multidict-5.1.0-cp36-cp36m-manylinux2014_s390x.whl", hash = "sha256:830f57206cc96ed0ccf68304141fec9481a096c4d2e2831f311bde1c404401da"}, - {file = "multidict-5.1.0-cp36-cp36m-manylinux2014_x86_64.whl", hash = "sha256:2e68965192c4ea61fff1b81c14ff712fc7dc15d2bd120602e4a3494ea6584224"}, - {file = "multidict-5.1.0-cp36-cp36m-win32.whl", hash = "sha256:2f1a132f1c88724674271d636e6b7351477c27722f2ed789f719f9e3545a3d26"}, - {file = "multidict-5.1.0-cp36-cp36m-win_amd64.whl", hash = "sha256:3a4f32116f8f72ecf2a29dabfb27b23ab7cdc0ba807e8459e59a93a9be9506f6"}, - {file = "multidict-5.1.0-cp37-cp37m-macosx_10_14_x86_64.whl", hash = "sha256:46c73e09ad374a6d876c599f2328161bcd95e280f84d2060cf57991dec5cfe76"}, - {file = "multidict-5.1.0-cp37-cp37m-manylinux1_i686.whl", hash = "sha256:018132dbd8688c7a69ad89c4a3f39ea2f9f33302ebe567a879da8f4ca73f0d0a"}, - {file = "multidict-5.1.0-cp37-cp37m-manylinux2014_aarch64.whl", hash = "sha256:4b186eb7d6ae7c06eb4392411189469e6a820da81447f46c0072a41c748ab73f"}, - {file = "multidict-5.1.0-cp37-cp37m-manylinux2014_i686.whl", hash = "sha256:3a041b76d13706b7fff23b9fc83117c7b8fe8d5fe9e6be45eee72b9baa75f348"}, - {file = "multidict-5.1.0-cp37-cp37m-manylinux2014_ppc64le.whl", hash = "sha256:051012ccee979b2b06be928a6150d237aec75dd6bf2d1eeeb190baf2b05abc93"}, - {file = "multidict-5.1.0-cp37-cp37m-manylinux2014_s390x.whl", hash = "sha256:6a4d5ce640e37b0efcc8441caeea8f43a06addace2335bd11151bc02d2ee31f9"}, - {file = "multidict-5.1.0-cp37-cp37m-manylinux2014_x86_64.whl", hash = "sha256:5cf3443199b83ed9e955f511b5b241fd3ae004e3cb81c58ec10f4fe47c7dce37"}, - {file = "multidict-5.1.0-cp37-cp37m-win32.whl", hash = "sha256:f200755768dc19c6f4e2b672421e0ebb3dd54c38d5a4f262b872d8cfcc9e93b5"}, - {file = "multidict-5.1.0-cp37-cp37m-win_amd64.whl", hash = "sha256:05c20b68e512166fddba59a918773ba002fdd77800cad9f55b59790030bab632"}, - {file = "multidict-5.1.0-cp38-cp38-macosx_10_14_x86_64.whl", hash = "sha256:54fd1e83a184e19c598d5e70ba508196fd0bbdd676ce159feb412a4a6664f952"}, - {file = "multidict-5.1.0-cp38-cp38-manylinux1_i686.whl", hash = "sha256:0e3c84e6c67eba89c2dbcee08504ba8644ab4284863452450520dad8f1e89b79"}, - {file = "multidict-5.1.0-cp38-cp38-manylinux2014_aarch64.whl", hash = "sha256:dc862056f76443a0db4509116c5cd480fe1b6a2d45512a653f9a855cc0517456"}, - {file = "multidict-5.1.0-cp38-cp38-manylinux2014_i686.whl", hash = "sha256:0e929169f9c090dae0646a011c8b058e5e5fb391466016b39d21745b48817fd7"}, - {file = "multidict-5.1.0-cp38-cp38-manylinux2014_ppc64le.whl", hash = "sha256:d81eddcb12d608cc08081fa88d046c78afb1bf8107e6feab5d43503fea74a635"}, - {file = "multidict-5.1.0-cp38-cp38-manylinux2014_s390x.whl", hash = "sha256:585fd452dd7782130d112f7ddf3473ffdd521414674c33876187e101b588738a"}, - {file = "multidict-5.1.0-cp38-cp38-manylinux2014_x86_64.whl", hash = "sha256:37e5438e1c78931df5d3c0c78ae049092877e5e9c02dd1ff5abb9cf27a5914ea"}, - {file = "multidict-5.1.0-cp38-cp38-win32.whl", hash = "sha256:07b42215124aedecc6083f1ce6b7e5ec5b50047afa701f3442054373a6deb656"}, - {file = "multidict-5.1.0-cp38-cp38-win_amd64.whl", hash = "sha256:929006d3c2d923788ba153ad0de8ed2e5ed39fdbe8e7be21e2f22ed06c6783d3"}, - {file = "multidict-5.1.0-cp39-cp39-macosx_10_14_x86_64.whl", hash = "sha256:b797515be8743b771aa868f83563f789bbd4b236659ba52243b735d80b29ed93"}, - {file = "multidict-5.1.0-cp39-cp39-manylinux1_i686.whl", hash = "sha256:d5c65bdf4484872c4af3150aeebe101ba560dcfb34488d9a8ff8dbcd21079647"}, - {file = "multidict-5.1.0-cp39-cp39-manylinux2014_aarch64.whl", hash = "sha256:b47a43177a5e65b771b80db71e7be76c0ba23cc8aa73eeeb089ed5219cdbe27d"}, - {file = "multidict-5.1.0-cp39-cp39-manylinux2014_i686.whl", hash = "sha256:806068d4f86cb06af37cd65821554f98240a19ce646d3cd24e1c33587f313eb8"}, - {file = "multidict-5.1.0-cp39-cp39-manylinux2014_ppc64le.whl", hash = "sha256:46dd362c2f045095c920162e9307de5ffd0a1bfbba0a6e990b344366f55a30c1"}, - {file = "multidict-5.1.0-cp39-cp39-manylinux2014_s390x.whl", hash = "sha256:ace010325c787c378afd7f7c1ac66b26313b3344628652eacd149bdd23c68841"}, - {file = "multidict-5.1.0-cp39-cp39-manylinux2014_x86_64.whl", hash = "sha256:ecc771ab628ea281517e24fd2c52e8f31c41e66652d07599ad8818abaad38cda"}, - {file = "multidict-5.1.0-cp39-cp39-win32.whl", hash = "sha256:fc13a9524bc18b6fb6e0dbec3533ba0496bbed167c56d0aabefd965584557d80"}, - {file = "multidict-5.1.0-cp39-cp39-win_amd64.whl", hash = "sha256:7df80d07818b385f3129180369079bd6934cf70469f99daaebfac89dca288359"}, - {file = "multidict-5.1.0.tar.gz", hash = "sha256:25b4e5f22d3a37ddf3effc0710ba692cfc792c2b9edfb9c05aefe823256e84d5"}, + {file = "multidict-5.2.0-cp310-cp310-macosx_10_9_universal2.whl", hash = "sha256:3822c5894c72e3b35aae9909bef66ec83e44522faf767c0ad39e0e2de11d3b55"}, + {file = "multidict-5.2.0-cp310-cp310-macosx_10_9_x86_64.whl", hash = "sha256:28e6d883acd8674887d7edc896b91751dc2d8e87fbdca8359591a13872799e4e"}, + {file = "multidict-5.2.0-cp310-cp310-macosx_11_0_arm64.whl", hash = "sha256:b61f85101ef08cbbc37846ac0e43f027f7844f3fade9b7f6dd087178caedeee7"}, + {file = "multidict-5.2.0-cp310-cp310-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:d9b668c065968c5979fe6b6fa6760bb6ab9aeb94b75b73c0a9c1acf6393ac3bf"}, + {file = "multidict-5.2.0-cp310-cp310-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:517d75522b7b18a3385726b54a081afd425d4f41144a5399e5abd97ccafdf36b"}, + {file = "multidict-5.2.0-cp310-cp310-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:1b4ac3ba7a97b35a5ccf34f41b5a8642a01d1e55454b699e5e8e7a99b5a3acf5"}, + {file = "multidict-5.2.0-cp310-cp310-manylinux_2_5_i686.manylinux1_i686.manylinux_2_12_i686.manylinux2010_i686.whl", hash = "sha256:df23c83398715b26ab09574217ca21e14694917a0c857e356fd39e1c64f8283f"}, + {file = "multidict-5.2.0-cp310-cp310-manylinux_2_5_x86_64.manylinux1_x86_64.manylinux_2_12_x86_64.manylinux2010_x86_64.whl", hash = "sha256:e58a9b5cc96e014ddf93c2227cbdeca94b56a7eb77300205d6e4001805391747"}, + {file = "multidict-5.2.0-cp310-cp310-musllinux_1_1_aarch64.whl", hash = "sha256:f76440e480c3b2ca7f843ff8a48dc82446b86ed4930552d736c0bac507498a52"}, + {file = "multidict-5.2.0-cp310-cp310-musllinux_1_1_i686.whl", hash = "sha256:cfde464ca4af42a629648c0b0d79b8f295cf5b695412451716531d6916461628"}, + {file = "multidict-5.2.0-cp310-cp310-musllinux_1_1_ppc64le.whl", hash = "sha256:0fed465af2e0eb6357ba95795d003ac0bdb546305cc2366b1fc8f0ad67cc3fda"}, + {file = "multidict-5.2.0-cp310-cp310-musllinux_1_1_s390x.whl", hash = "sha256:b70913cbf2e14275013be98a06ef4b412329fe7b4f83d64eb70dce8269ed1e1a"}, + {file = "multidict-5.2.0-cp310-cp310-musllinux_1_1_x86_64.whl", hash = "sha256:a5635bcf1b75f0f6ef3c8a1ad07b500104a971e38d3683167b9454cb6465ac86"}, + {file = "multidict-5.2.0-cp310-cp310-win32.whl", hash = "sha256:77f0fb7200cc7dedda7a60912f2059086e29ff67cefbc58d2506638c1a9132d7"}, + {file = "multidict-5.2.0-cp310-cp310-win_amd64.whl", hash = "sha256:9416cf11bcd73c861267e88aea71e9fcc35302b3943e45e1dbb4317f91a4b34f"}, + {file = "multidict-5.2.0-cp36-cp36m-macosx_10_9_x86_64.whl", hash = "sha256:fd77c8f3cba815aa69cb97ee2b2ef385c7c12ada9c734b0f3b32e26bb88bbf1d"}, + {file = "multidict-5.2.0-cp36-cp36m-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:98ec9aea6223adf46999f22e2c0ab6cf33f5914be604a404f658386a8f1fba37"}, + {file = "multidict-5.2.0-cp36-cp36m-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:e5283c0a00f48e8cafcecadebfa0ed1dac8b39e295c7248c44c665c16dc1138b"}, + {file = "multidict-5.2.0-cp36-cp36m-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:5f79c19c6420962eb17c7e48878a03053b7ccd7b69f389d5831c0a4a7f1ac0a1"}, + {file = "multidict-5.2.0-cp36-cp36m-manylinux_2_5_i686.manylinux1_i686.manylinux_2_12_i686.manylinux2010_i686.whl", hash = "sha256:e4a67f1080123de76e4e97a18d10350df6a7182e243312426d508712e99988d4"}, + {file = "multidict-5.2.0-cp36-cp36m-manylinux_2_5_x86_64.manylinux1_x86_64.manylinux_2_12_x86_64.manylinux2010_x86_64.whl", hash = "sha256:94b117e27efd8e08b4046c57461d5a114d26b40824995a2eb58372b94f9fca02"}, + {file = "multidict-5.2.0-cp36-cp36m-musllinux_1_1_aarch64.whl", hash = "sha256:2e77282fd1d677c313ffcaddfec236bf23f273c4fba7cdf198108f5940ae10f5"}, + {file = "multidict-5.2.0-cp36-cp36m-musllinux_1_1_i686.whl", hash = "sha256:116347c63ba049c1ea56e157fa8aa6edaf5e92925c9b64f3da7769bdfa012858"}, + {file = "multidict-5.2.0-cp36-cp36m-musllinux_1_1_ppc64le.whl", hash = "sha256:dc3a866cf6c13d59a01878cd806f219340f3e82eed514485e094321f24900677"}, + {file = "multidict-5.2.0-cp36-cp36m-musllinux_1_1_s390x.whl", hash = "sha256:ac42181292099d91217a82e3fa3ce0e0ddf3a74fd891b7c2b347a7f5aa0edded"}, + {file = "multidict-5.2.0-cp36-cp36m-musllinux_1_1_x86_64.whl", hash = "sha256:f0bb0973f42ffcb5e3537548e0767079420aefd94ba990b61cf7bb8d47f4916d"}, + {file = "multidict-5.2.0-cp36-cp36m-win32.whl", hash = "sha256:ea21d4d5104b4f840b91d9dc8cbc832aba9612121eaba503e54eaab1ad140eb9"}, + {file = "multidict-5.2.0-cp36-cp36m-win_amd64.whl", hash = "sha256:e6453f3cbeb78440747096f239d282cc57a2997a16b5197c9bc839099e1633d0"}, + {file = "multidict-5.2.0-cp37-cp37m-macosx_10_9_x86_64.whl", hash = "sha256:d3def943bfd5f1c47d51fd324df1e806d8da1f8e105cc7f1c76a1daf0f7e17b0"}, + {file = "multidict-5.2.0-cp37-cp37m-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:35591729668a303a02b06e8dba0eb8140c4a1bfd4c4b3209a436a02a5ac1de11"}, + {file = "multidict-5.2.0-cp37-cp37m-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:ce8cacda0b679ebc25624d5de66c705bc53dcc7c6f02a7fb0f3ca5e227d80422"}, + {file = "multidict-5.2.0-cp37-cp37m-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:baf1856fab8212bf35230c019cde7c641887e3fc08cadd39d32a421a30151ea3"}, + {file = "multidict-5.2.0-cp37-cp37m-manylinux_2_5_i686.manylinux1_i686.manylinux_2_12_i686.manylinux2010_i686.whl", hash = "sha256:a43616aec0f0d53c411582c451f5d3e1123a68cc7b3475d6f7d97a626f8ff90d"}, + {file = "multidict-5.2.0-cp37-cp37m-manylinux_2_5_x86_64.manylinux1_x86_64.manylinux_2_12_x86_64.manylinux2010_x86_64.whl", hash = "sha256:25cbd39a9029b409167aa0a20d8a17f502d43f2efebfe9e3ac019fe6796c59ac"}, + {file = "multidict-5.2.0-cp37-cp37m-musllinux_1_1_aarch64.whl", hash = "sha256:0a2cbcfbea6dc776782a444db819c8b78afe4db597211298dd8b2222f73e9cd0"}, + {file = "multidict-5.2.0-cp37-cp37m-musllinux_1_1_i686.whl", hash = "sha256:3d2d7d1fff8e09d99354c04c3fd5b560fb04639fd45926b34e27cfdec678a704"}, + {file = "multidict-5.2.0-cp37-cp37m-musllinux_1_1_ppc64le.whl", hash = "sha256:a37e9a68349f6abe24130846e2f1d2e38f7ddab30b81b754e5a1fde32f782b23"}, + {file = "multidict-5.2.0-cp37-cp37m-musllinux_1_1_s390x.whl", hash = "sha256:637c1896497ff19e1ee27c1c2c2ddaa9f2d134bbb5e0c52254361ea20486418d"}, + {file = "multidict-5.2.0-cp37-cp37m-musllinux_1_1_x86_64.whl", hash = "sha256:9815765f9dcda04921ba467957be543423e5ec6a1136135d84f2ae092c50d87b"}, + {file = "multidict-5.2.0-cp37-cp37m-win32.whl", hash = "sha256:8b911d74acdc1fe2941e59b4f1a278a330e9c34c6c8ca1ee21264c51ec9b67ef"}, + {file = "multidict-5.2.0-cp37-cp37m-win_amd64.whl", hash = "sha256:380b868f55f63d048a25931a1632818f90e4be71d2081c2338fcf656d299949a"}, + {file = "multidict-5.2.0-cp38-cp38-macosx_10_9_universal2.whl", hash = "sha256:e7d81ce5744757d2f05fc41896e3b2ae0458464b14b5a2c1e87a6a9d69aefaa8"}, + {file = "multidict-5.2.0-cp38-cp38-macosx_10_9_x86_64.whl", hash = "sha256:2d1d55cdf706ddc62822d394d1df53573d32a7a07d4f099470d3cb9323b721b6"}, + {file = "multidict-5.2.0-cp38-cp38-macosx_11_0_arm64.whl", hash = "sha256:a4771d0d0ac9d9fe9e24e33bed482a13dfc1256d008d101485fe460359476065"}, + {file = "multidict-5.2.0-cp38-cp38-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:da7d57ea65744d249427793c042094c4016789eb2562576fb831870f9c878d9e"}, + {file = "multidict-5.2.0-cp38-cp38-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:cdd68778f96216596218b4e8882944d24a634d984ee1a5a049b300377878fa7c"}, + {file = "multidict-5.2.0-cp38-cp38-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:ecc99bce8ee42dcad15848c7885197d26841cb24fa2ee6e89d23b8993c871c64"}, + {file = "multidict-5.2.0-cp38-cp38-manylinux_2_5_i686.manylinux1_i686.manylinux_2_12_i686.manylinux2010_i686.whl", hash = "sha256:067150fad08e6f2dd91a650c7a49ba65085303fcc3decbd64a57dc13a2733031"}, + {file = "multidict-5.2.0-cp38-cp38-manylinux_2_5_x86_64.manylinux1_x86_64.manylinux_2_12_x86_64.manylinux2010_x86_64.whl", hash = "sha256:78c106b2b506b4d895ddc801ff509f941119394b89c9115580014127414e6c2d"}, + {file = "multidict-5.2.0-cp38-cp38-musllinux_1_1_aarch64.whl", hash = "sha256:e6c4fa1ec16e01e292315ba76eb1d012c025b99d22896bd14a66628b245e3e01"}, + {file = "multidict-5.2.0-cp38-cp38-musllinux_1_1_i686.whl", hash = "sha256:b227345e4186809d31f22087d0265655114af7cda442ecaf72246275865bebe4"}, + {file = "multidict-5.2.0-cp38-cp38-musllinux_1_1_ppc64le.whl", hash = "sha256:06560fbdcf22c9387100979e65b26fba0816c162b888cb65b845d3def7a54c9b"}, + {file = "multidict-5.2.0-cp38-cp38-musllinux_1_1_s390x.whl", hash = "sha256:7878b61c867fb2df7a95e44b316f88d5a3742390c99dfba6c557a21b30180cac"}, + {file = "multidict-5.2.0-cp38-cp38-musllinux_1_1_x86_64.whl", hash = "sha256:246145bff76cc4b19310f0ad28bd0769b940c2a49fc601b86bfd150cbd72bb22"}, + {file = "multidict-5.2.0-cp38-cp38-win32.whl", hash = "sha256:c30ac9f562106cd9e8071c23949a067b10211917fdcb75b4718cf5775356a940"}, + {file = "multidict-5.2.0-cp38-cp38-win_amd64.whl", hash = "sha256:f19001e790013ed580abfde2a4465388950728861b52f0da73e8e8a9418533c0"}, + {file = "multidict-5.2.0-cp39-cp39-macosx_10_9_universal2.whl", hash = "sha256:c1ff762e2ee126e6f1258650ac641e2b8e1f3d927a925aafcfde943b77a36d24"}, + {file = "multidict-5.2.0-cp39-cp39-macosx_10_9_x86_64.whl", hash = "sha256:bd6c9c50bf2ad3f0448edaa1a3b55b2e6866ef8feca5d8dbec10ec7c94371d21"}, + {file = "multidict-5.2.0-cp39-cp39-macosx_11_0_arm64.whl", hash = "sha256:fc66d4016f6e50ed36fb39cd287a3878ffcebfa90008535c62e0e90a7ab713ae"}, + {file = "multidict-5.2.0-cp39-cp39-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:a9acb76d5f3dd9421874923da2ed1e76041cb51b9337fd7f507edde1d86535d6"}, + {file = "multidict-5.2.0-cp39-cp39-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:dfc924a7e946dd3c6360e50e8f750d51e3ef5395c95dc054bc9eab0f70df4f9c"}, + {file = "multidict-5.2.0-cp39-cp39-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:32fdba7333eb2351fee2596b756d730d62b5827d5e1ab2f84e6cbb287cc67fe0"}, + {file = "multidict-5.2.0-cp39-cp39-manylinux_2_5_i686.manylinux1_i686.manylinux_2_12_i686.manylinux2010_i686.whl", hash = "sha256:b9aad49466b8d828b96b9e3630006234879c8d3e2b0a9d99219b3121bc5cdb17"}, + {file = "multidict-5.2.0-cp39-cp39-manylinux_2_5_x86_64.manylinux1_x86_64.manylinux_2_12_x86_64.manylinux2010_x86_64.whl", hash = "sha256:93de39267c4c676c9ebb2057e98a8138bade0d806aad4d864322eee0803140a0"}, + {file = "multidict-5.2.0-cp39-cp39-musllinux_1_1_aarch64.whl", hash = "sha256:f9bef5cff994ca3026fcc90680e326d1a19df9841c5e3d224076407cc21471a1"}, + {file = "multidict-5.2.0-cp39-cp39-musllinux_1_1_i686.whl", hash = "sha256:5f841c4f14331fd1e36cbf3336ed7be2cb2a8f110ce40ea253e5573387db7621"}, + {file = "multidict-5.2.0-cp39-cp39-musllinux_1_1_ppc64le.whl", hash = "sha256:38ba256ee9b310da6a1a0f013ef4e422fca30a685bcbec86a969bd520504e341"}, + {file = "multidict-5.2.0-cp39-cp39-musllinux_1_1_s390x.whl", hash = "sha256:3bc3b1621b979621cee9f7b09f024ec76ec03cc365e638126a056317470bde1b"}, + {file = "multidict-5.2.0-cp39-cp39-musllinux_1_1_x86_64.whl", hash = "sha256:6ee908c070020d682e9b42c8f621e8bb10c767d04416e2ebe44e37d0f44d9ad5"}, + {file = "multidict-5.2.0-cp39-cp39-win32.whl", hash = "sha256:1c7976cd1c157fa7ba5456ae5d31ccdf1479680dc9b8d8aa28afabc370df42b8"}, + {file = "multidict-5.2.0-cp39-cp39-win_amd64.whl", hash = "sha256:c9631c642e08b9fff1c6255487e62971d8b8e821808ddd013d8ac058087591ac"}, + {file = "multidict-5.2.0.tar.gz", hash = "sha256:0dd1c93edb444b33ba2274b66f63def8a327d607c6c790772f448a53b6ea59ce"}, ] mypy = [ {file = "mypy-0.812-cp35-cp35m-macosx_10_9_x86_64.whl", hash = "sha256:a26f8ec704e5a7423c8824d425086705e381b4f1dfdef6e3a1edab7ba174ec49"}, @@ -896,25 +1294,56 @@ nodeenv = [ {file = "nodeenv-1.6.0-py2.py3-none-any.whl", hash = "sha256:621e6b7076565ddcacd2db0294c0381e01fd28945ab36bcf00f41c5daf63bef7"}, {file = "nodeenv-1.6.0.tar.gz", hash = "sha256:3ef13ff90291ba2a4a7a4ff9a979b63ffdd00a464dbe04acf0ea6471517a4c2b"}, ] -openleadr = [ - {file = "openleadr-0.5.23-py3-none-any.whl", hash = "sha256:863522020f7489d67827c69c1b3d6e75531ff75637238c87ecf69a4b08af439c"}, - {file = "openleadr-0.5.23.tar.gz", hash = "sha256:d6a77576120e51db471fcd1f70df50163363989399a058082e43e5a51a0f7c67"}, -] +openleadr = [] packaging = [ - {file = "packaging-20.9-py2.py3-none-any.whl", hash = "sha256:67714da7f7bc052e064859c05c595155bd1ee9f69f76557e21f051443c20947a"}, - {file = "packaging-20.9.tar.gz", hash = "sha256:5b327ac1320dc863dca72f4514ecc086f31186744b84a230374cc1fd776feae5"}, + {file = "packaging-21.0-py3-none-any.whl", hash = "sha256:c86254f9220d55e31cc94d69bade760f0847da8000def4dfe1c6b872fd14ff14"}, + {file = "packaging-21.0.tar.gz", hash = "sha256:7dc96269f53a4ccec5c0670940a4281106dd0bb343f47b7471f779df49c2fbe7"}, ] pathspec = [ - {file = "pathspec-0.8.1-py2.py3-none-any.whl", hash = "sha256:aa0cb481c4041bf52ffa7b0d8fa6cd3e88a2ca4879c533c9153882ee2556790d"}, - {file = "pathspec-0.8.1.tar.gz", hash = "sha256:86379d6b86d75816baba717e64b1a3a3469deb93bb76d613c9ce79edc5cb68fd"}, + {file = "pathspec-0.9.0-py2.py3-none-any.whl", hash = "sha256:7d15c4ddb0b5c802d161efc417ec1a2558ea2653c2e8ad9c19098201dc1c993a"}, + {file = "pathspec-0.9.0.tar.gz", hash = "sha256:e564499435a2673d586f6b2130bb5b95f04a3ba06f81b8f895b651a3c76aabb1"}, +] +platformdirs = [ + {file = "platformdirs-2.4.0-py3-none-any.whl", hash = "sha256:8868bbe3c3c80d42f20156f22e7131d2fb321f5bc86a2a345375c6481a67021d"}, + {file = "platformdirs-2.4.0.tar.gz", hash = "sha256:367a5e80b3d04d2428ffa76d33f124cf11e8fff2acdaa9b43d545f5c7d661ef2"}, ] pluggy = [ - {file = "pluggy-0.13.1-py2.py3-none-any.whl", hash = "sha256:966c145cd83c96502c3c3868f50408687b38434af77734af1e9ca461a4081d2d"}, - {file = "pluggy-0.13.1.tar.gz", hash = "sha256:15b2acde666561e1298d71b523007ed7364de07029219b604cf808bfa1c765b0"}, + {file = "pluggy-1.0.0-py2.py3-none-any.whl", hash = "sha256:74134bbf457f031a36d68416e1509f34bd5ccc019f0bcc952c7b909d06b37bd3"}, + {file = "pluggy-1.0.0.tar.gz", hash = "sha256:4224373bacce55f955a878bf9cfa763c1e360858e330072059e10bad68531159"}, ] pre-commit = [ - {file = "pre_commit-2.13.0-py2.py3-none-any.whl", hash = "sha256:b679d0fddd5b9d6d98783ae5f10fd0c4c59954f375b70a58cbe1ce9bcf9809a4"}, - {file = "pre_commit-2.13.0.tar.gz", hash = "sha256:764972c60693dc668ba8e86eb29654ec3144501310f7198742a767bec385a378"}, + {file = "pre_commit-2.15.0-py2.py3-none-any.whl", hash = "sha256:a4ed01000afcb484d9eb8d504272e642c4c4099bbad3a6b27e519bd6a3e928a6"}, + {file = "pre_commit-2.15.0.tar.gz", hash = "sha256:3c25add78dbdfb6a28a651780d5c311ac40dd17f160eb3954a0c59da40a505a7"}, +] +psutil = [ + {file = "psutil-5.8.0-cp27-cp27m-macosx_10_9_x86_64.whl", hash = "sha256:0066a82f7b1b37d334e68697faba68e5ad5e858279fd6351c8ca6024e8d6ba64"}, + {file = "psutil-5.8.0-cp27-cp27m-manylinux2010_i686.whl", hash = "sha256:0ae6f386d8d297177fd288be6e8d1afc05966878704dad9847719650e44fc49c"}, + {file = "psutil-5.8.0-cp27-cp27m-manylinux2010_x86_64.whl", hash = "sha256:12d844996d6c2b1d3881cfa6fa201fd635971869a9da945cf6756105af73d2df"}, + {file = "psutil-5.8.0-cp27-cp27mu-manylinux2010_i686.whl", hash = "sha256:02b8292609b1f7fcb34173b25e48d0da8667bc85f81d7476584d889c6e0f2131"}, + {file = "psutil-5.8.0-cp27-cp27mu-manylinux2010_x86_64.whl", hash = "sha256:6ffe81843131ee0ffa02c317186ed1e759a145267d54fdef1bc4ea5f5931ab60"}, + {file = "psutil-5.8.0-cp27-none-win32.whl", hash = "sha256:ea313bb02e5e25224e518e4352af4bf5e062755160f77e4b1767dd5ccb65f876"}, + {file = "psutil-5.8.0-cp27-none-win_amd64.whl", hash = "sha256:5da29e394bdedd9144c7331192e20c1f79283fb03b06e6abd3a8ae45ffecee65"}, + {file = "psutil-5.8.0-cp36-cp36m-macosx_10_9_x86_64.whl", hash = "sha256:74fb2557d1430fff18ff0d72613c5ca30c45cdbfcddd6a5773e9fc1fe9364be8"}, + {file = "psutil-5.8.0-cp36-cp36m-manylinux2010_i686.whl", hash = "sha256:74f2d0be88db96ada78756cb3a3e1b107ce8ab79f65aa885f76d7664e56928f6"}, + {file = "psutil-5.8.0-cp36-cp36m-manylinux2010_x86_64.whl", hash = "sha256:99de3e8739258b3c3e8669cb9757c9a861b2a25ad0955f8e53ac662d66de61ac"}, + {file = "psutil-5.8.0-cp36-cp36m-win32.whl", hash = "sha256:36b3b6c9e2a34b7d7fbae330a85bf72c30b1c827a4366a07443fc4b6270449e2"}, + {file = "psutil-5.8.0-cp36-cp36m-win_amd64.whl", hash = "sha256:52de075468cd394ac98c66f9ca33b2f54ae1d9bff1ef6b67a212ee8f639ec06d"}, + {file = "psutil-5.8.0-cp37-cp37m-macosx_10_9_x86_64.whl", hash = "sha256:c6a5fd10ce6b6344e616cf01cc5b849fa8103fbb5ba507b6b2dee4c11e84c935"}, + {file = "psutil-5.8.0-cp37-cp37m-manylinux2010_i686.whl", hash = "sha256:61f05864b42fedc0771d6d8e49c35f07efd209ade09a5afe6a5059e7bb7bf83d"}, + {file = "psutil-5.8.0-cp37-cp37m-manylinux2010_x86_64.whl", hash = "sha256:0dd4465a039d343925cdc29023bb6960ccf4e74a65ad53e768403746a9207023"}, + {file = "psutil-5.8.0-cp37-cp37m-win32.whl", hash = "sha256:1bff0d07e76114ec24ee32e7f7f8d0c4b0514b3fae93e3d2aaafd65d22502394"}, + {file = "psutil-5.8.0-cp37-cp37m-win_amd64.whl", hash = "sha256:fcc01e900c1d7bee2a37e5d6e4f9194760a93597c97fee89c4ae51701de03563"}, + {file = "psutil-5.8.0-cp38-cp38-macosx_10_9_x86_64.whl", hash = "sha256:6223d07a1ae93f86451d0198a0c361032c4c93ebd4bf6d25e2fb3edfad9571ef"}, + {file = "psutil-5.8.0-cp38-cp38-manylinux2010_i686.whl", hash = "sha256:d225cd8319aa1d3c85bf195c4e07d17d3cd68636b8fc97e6cf198f782f99af28"}, + {file = "psutil-5.8.0-cp38-cp38-manylinux2010_x86_64.whl", hash = "sha256:28ff7c95293ae74bf1ca1a79e8805fcde005c18a122ca983abf676ea3466362b"}, + {file = "psutil-5.8.0-cp38-cp38-win32.whl", hash = "sha256:ce8b867423291cb65cfc6d9c4955ee9bfc1e21fe03bb50e177f2b957f1c2469d"}, + {file = "psutil-5.8.0-cp38-cp38-win_amd64.whl", hash = "sha256:90f31c34d25b1b3ed6c40cdd34ff122b1887a825297c017e4cbd6796dd8b672d"}, + {file = "psutil-5.8.0-cp39-cp39-macosx_10_9_x86_64.whl", hash = "sha256:6323d5d845c2785efb20aded4726636546b26d3b577aded22492908f7c1bdda7"}, + {file = "psutil-5.8.0-cp39-cp39-manylinux2010_i686.whl", hash = "sha256:245b5509968ac0bd179287d91210cd3f37add77dad385ef238b275bad35fa1c4"}, + {file = "psutil-5.8.0-cp39-cp39-manylinux2010_x86_64.whl", hash = "sha256:90d4091c2d30ddd0a03e0b97e6a33a48628469b99585e2ad6bf21f17423b112b"}, + {file = "psutil-5.8.0-cp39-cp39-win32.whl", hash = "sha256:ea372bcc129394485824ae3e3ddabe67dc0b118d262c568b4d2602a7070afdb0"}, + {file = "psutil-5.8.0-cp39-cp39-win_amd64.whl", hash = "sha256:f4634b033faf0d968bb9220dd1c793b897ab7f1189956e1aa9eae752527127d3"}, + {file = "psutil-5.8.0.tar.gz", hash = "sha256:0c9ccb99ab76025f2f0bbecf341d4656e9c1351db8cc8a03ccd62e318ab4b5c6"}, ] py = [ {file = "py-1.10.0-py2.py3-none-any.whl", hash = "sha256:3b80836aa6d1feeaa108e046da6423ab8f6ceda6468545ae8d02d9d58d18818a"}, @@ -925,20 +1354,24 @@ pycparser = [ {file = "pycparser-2.20.tar.gz", hash = "sha256:2d475327684562c3a96cc71adf7dc8c4f0565175cf86b6d7a404ff4c771f15f0"}, ] pyopenssl = [ - {file = "pyOpenSSL-19.1.0-py2.py3-none-any.whl", hash = "sha256:621880965a720b8ece2f1b2f54ea2071966ab00e2970ad2ce11d596102063504"}, - {file = "pyOpenSSL-19.1.0.tar.gz", hash = "sha256:9a24494b2602aaf402be5c9e30a0b82d4a5c67528fe8fb475e3f3bc00dd69507"}, + {file = "pyOpenSSL-19.0.0-py2.py3-none-any.whl", hash = "sha256:c727930ad54b10fc157015014b666f2d8b41f70c0d03e83ab67624fd3dd5d1e6"}, + {file = "pyOpenSSL-19.0.0.tar.gz", hash = "sha256:aeca66338f6de19d1aa46ed634c3b9ae519a64b458f8468aec688e7e3c20f200"}, ] pyparsing = [ {file = "pyparsing-2.4.7-py2.py3-none-any.whl", hash = "sha256:ef9d7589ef3c200abe66653d3f1ab1033c3c419ae9b9bdb1240a85b024efc88b"}, {file = "pyparsing-2.4.7.tar.gz", hash = "sha256:c203ec8783bf771a155b207279b9bccb8dea02d8f0c9e5f8ead507bc3246ecc1"}, ] pytest = [ - {file = "pytest-6.2.4-py3-none-any.whl", hash = "sha256:91ef2131a9bd6be8f76f1f08eac5c5317221d6ad1e143ae03894b862e8976890"}, - {file = "pytest-6.2.4.tar.gz", hash = "sha256:50bcad0a0b9c5a72c8e4e7c9855a3ad496ca6a881a3641b4260605450772c54b"}, + {file = "pytest-6.2.5-py3-none-any.whl", hash = "sha256:7310f8d27bc79ced999e760ca304d69f6ba6c6649c0b60fb0e04a4a77cacc134"}, + {file = "pytest-6.2.5.tar.gz", hash = "sha256:131b36680866a76e6781d13f101efb86cf674ebb9762eb70d3082b6f29889e89"}, +] +python-dateutil = [ + {file = "python-dateutil-2.8.2.tar.gz", hash = "sha256:0123cacc1627ae19ddf3c27a5de5bd67ee4586fbdd6440d9748f8abb483d3e86"}, + {file = "python_dateutil-2.8.2-py2.py3-none-any.whl", hash = "sha256:961d03dc3453ebbc59dbdea9e4e11c5651520a876d0f4db161e8674aae935da9"}, ] pytz = [ - {file = "pytz-2021.1-py2.py3-none-any.whl", hash = "sha256:eb10ce3e7736052ed3623d49975ce333bcd712c7bb19a58b9e2089d4057d0798"}, - {file = "pytz-2021.1.tar.gz", hash = "sha256:83a4a90894bf38e243cf052c8b58f381bfe9a7a483f6a9cab140bc7f702ac4da"}, + {file = "pytz-2021.3-py2.py3-none-any.whl", hash = "sha256:3672058bc3453457b622aab7a1c3bfd5ab0bdae451512f6cf25f64ed37f5b87c"}, + {file = "pytz-2021.3.tar.gz", hash = "sha256:acad2d8b20a1af07d4e4c9d2e9285c5ed9104354062f275f3fcd88dcef4f1326"}, ] pyyaml = [ {file = "PyYAML-5.4.1-cp27-cp27m-macosx_10_9_x86_64.whl", hash = "sha256:3b2b1824fe7112845700f815ff6a489360226a5609b96ec2190a45e62a9fc922"}, @@ -971,48 +1404,86 @@ pyyaml = [ {file = "PyYAML-5.4.1-cp39-cp39-win_amd64.whl", hash = "sha256:c20cfa2d49991c8b4147af39859b167664f2ad4561704ee74c1de03318e898db"}, {file = "PyYAML-5.4.1.tar.gz", hash = "sha256:607774cbba28732bfa802b54baa7484215f530991055bb562efbed5b2f20a45e"}, ] +pyzmq = [ + {file = "pyzmq-22.3.0-cp310-cp310-macosx_10_15_universal2.whl", hash = "sha256:6b217b8f9dfb6628f74b94bdaf9f7408708cb02167d644edca33f38746ca12dd"}, + {file = "pyzmq-22.3.0-cp310-cp310-macosx_10_9_x86_64.whl", hash = "sha256:2841997a0d85b998cbafecb4183caf51fd19c4357075dfd33eb7efea57e4c149"}, + {file = "pyzmq-22.3.0-cp310-cp310-manylinux_2_12_i686.manylinux2010_i686.whl", hash = "sha256:f89468059ebc519a7acde1ee50b779019535db8dcf9b8c162ef669257fef7a93"}, + {file = "pyzmq-22.3.0-cp310-cp310-manylinux_2_12_x86_64.manylinux2010_x86_64.whl", hash = "sha256:ea12133df25e3a6918718fbb9a510c6ee5d3fdd5a346320421aac3882f4feeea"}, + {file = "pyzmq-22.3.0-cp310-cp310-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:76c532fd68b93998aab92356be280deec5de8f8fe59cd28763d2cc8a58747b7f"}, + {file = "pyzmq-22.3.0-cp310-cp310-win32.whl", hash = "sha256:67db33bea0a29d03e6eeec55a8190e033318cee3cbc732ba8fd939617cbf762d"}, + {file = "pyzmq-22.3.0-cp310-cp310-win_amd64.whl", hash = "sha256:7661fc1d5cb73481cf710a1418a4e1e301ed7d5d924f91c67ba84b2a1b89defd"}, + {file = "pyzmq-22.3.0-cp36-cp36m-macosx_10_9_x86_64.whl", hash = "sha256:79244b9e97948eaf38695f4b8e6fc63b14b78cc37f403c6642ba555517ac1268"}, + {file = "pyzmq-22.3.0-cp36-cp36m-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:ab888624ed68930442a3f3b0b921ad7439c51ba122dbc8c386e6487a658e4a4e"}, + {file = "pyzmq-22.3.0-cp36-cp36m-manylinux_2_5_i686.manylinux1_i686.whl", hash = "sha256:18cd854b423fce44951c3a4d3e686bac8f1243d954f579e120a1714096637cc0"}, + {file = "pyzmq-22.3.0-cp36-cp36m-manylinux_2_5_x86_64.manylinux1_x86_64.whl", hash = "sha256:de8df0684398bd74ad160afdc2a118ca28384ac6f5e234eb0508858d8d2d9364"}, + {file = "pyzmq-22.3.0-cp36-cp36m-win32.whl", hash = "sha256:3c1895c95be92600233e476fe283f042e71cf8f0b938aabf21b7aafa62a8dac9"}, + {file = "pyzmq-22.3.0-cp36-cp36m-win_amd64.whl", hash = "sha256:851977788b9caa8ed011f5f643d3ee8653af02c5fc723fa350db5125abf2be7b"}, + {file = "pyzmq-22.3.0-cp37-cp37m-macosx_10_9_x86_64.whl", hash = "sha256:b4ebed0977f92320f6686c96e9e8dd29eed199eb8d066936bac991afc37cbb70"}, + {file = "pyzmq-22.3.0-cp37-cp37m-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:42abddebe2c6a35180ca549fadc7228d23c1e1f76167c5ebc8a936b5804ea2df"}, + {file = "pyzmq-22.3.0-cp37-cp37m-manylinux_2_5_i686.manylinux1_i686.whl", hash = "sha256:c1e41b32d6f7f9c26bc731a8b529ff592f31fc8b6ef2be9fa74abd05c8a342d7"}, + {file = "pyzmq-22.3.0-cp37-cp37m-manylinux_2_5_x86_64.manylinux1_x86_64.whl", hash = "sha256:be4e0f229cf3a71f9ecd633566bd6f80d9fa6afaaff5489492be63fe459ef98c"}, + {file = "pyzmq-22.3.0-cp37-cp37m-win32.whl", hash = "sha256:7c58f598d9fcc52772b89a92d72bf8829c12d09746a6d2c724c5b30076c1f11d"}, + {file = "pyzmq-22.3.0-cp37-cp37m-win_amd64.whl", hash = "sha256:2b97502c16a5ec611cd52410bdfaab264997c627a46b0f98d3f666227fd1ea2d"}, + {file = "pyzmq-22.3.0-cp38-cp38-macosx_10_9_x86_64.whl", hash = "sha256:d728b08448e5ac3e4d886b165385a262883c34b84a7fe1166277fe675e1c197a"}, + {file = "pyzmq-22.3.0-cp38-cp38-manylinux_2_12_i686.manylinux2010_i686.whl", hash = "sha256:480b9931bfb08bf8b094edd4836271d4d6b44150da051547d8c7113bf947a8b0"}, + {file = "pyzmq-22.3.0-cp38-cp38-manylinux_2_12_x86_64.manylinux2010_x86_64.whl", hash = "sha256:7dc09198e4073e6015d9a8ea093fc348d4e59de49382476940c3dd9ae156fba8"}, + {file = "pyzmq-22.3.0-cp38-cp38-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:0ca6cd58f62a2751728016d40082008d3b3412a7f28ddfb4a2f0d3c130f69e74"}, + {file = "pyzmq-22.3.0-cp38-cp38-win32.whl", hash = "sha256:c0f84360dcca3481e8674393bdf931f9f10470988f87311b19d23cda869bb6b7"}, + {file = "pyzmq-22.3.0-cp38-cp38-win_amd64.whl", hash = "sha256:f762442bab706fd874064ca218b33a1d8e40d4938e96c24dafd9b12e28017f45"}, + {file = "pyzmq-22.3.0-cp39-cp39-macosx_10_15_universal2.whl", hash = "sha256:954e73c9cd4d6ae319f1c936ad159072b6d356a92dcbbabfd6e6204b9a79d356"}, + {file = "pyzmq-22.3.0-cp39-cp39-macosx_10_9_x86_64.whl", hash = "sha256:f43b4a2e6218371dd4f41e547bd919ceeb6ebf4abf31a7a0669cd11cd91ea973"}, + {file = "pyzmq-22.3.0-cp39-cp39-manylinux_2_12_i686.manylinux2010_i686.whl", hash = "sha256:acebba1a23fb9d72b42471c3771b6f2f18dcd46df77482612054bd45c07dfa36"}, + {file = "pyzmq-22.3.0-cp39-cp39-manylinux_2_12_x86_64.manylinux2010_x86_64.whl", hash = "sha256:cf98fd7a6c8aaa08dbc699ffae33fd71175696d78028281bc7b832b26f00ca57"}, + {file = "pyzmq-22.3.0-cp39-cp39-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:d072f7dfbdb184f0786d63bda26e8a0882041b1e393fbe98940395f7fab4c5e2"}, + {file = "pyzmq-22.3.0-cp39-cp39-win32.whl", hash = "sha256:e6a02cf7271ee94674a44f4e62aa061d2d049001c844657740e156596298b70b"}, + {file = "pyzmq-22.3.0-cp39-cp39-win_amd64.whl", hash = "sha256:d3dcb5548ead4f1123851a5ced467791f6986d68c656bc63bfff1bf9e36671e2"}, + {file = "pyzmq-22.3.0-pp37-pypy37_pp73-macosx_10_9_x86_64.whl", hash = "sha256:3a4c9886d61d386b2b493377d980f502186cd71d501fffdba52bd2a0880cef4f"}, + {file = "pyzmq-22.3.0-pp37-pypy37_pp73-manylinux_2_12_i686.manylinux2010_i686.whl", hash = "sha256:80e043a89c6cadefd3a0712f8a1322038e819ebe9dbac7eca3bce1721bcb63bf"}, + {file = "pyzmq-22.3.0-pp37-pypy37_pp73-manylinux_2_12_x86_64.manylinux2010_x86_64.whl", hash = "sha256:1621e7a2af72cced1f6ec8ca8ca91d0f76ac236ab2e8828ac8fe909512d566cb"}, + {file = "pyzmq-22.3.0-pp37-pypy37_pp73-win_amd64.whl", hash = "sha256:d6157793719de168b199194f6b6173f0ccd3bf3499e6870fac17086072e39115"}, + {file = "pyzmq-22.3.0.tar.gz", hash = "sha256:8eddc033e716f8c91c6a2112f0a8ebc5e00532b4a6ae1eb0ccc48e027f9c671c"}, +] regex = [ - {file = "regex-2021.4.4-cp36-cp36m-macosx_10_9_x86_64.whl", hash = "sha256:619d71c59a78b84d7f18891fe914446d07edd48dc8328c8e149cbe0929b4e000"}, - {file = "regex-2021.4.4-cp36-cp36m-manylinux1_i686.whl", hash = "sha256:47bf5bf60cf04d72bf6055ae5927a0bd9016096bf3d742fa50d9bf9f45aa0711"}, - {file = "regex-2021.4.4-cp36-cp36m-manylinux1_x86_64.whl", hash = "sha256:281d2fd05555079448537fe108d79eb031b403dac622621c78944c235f3fcf11"}, - {file = "regex-2021.4.4-cp36-cp36m-manylinux2010_i686.whl", hash = "sha256:bd28bc2e3a772acbb07787c6308e00d9626ff89e3bfcdebe87fa5afbfdedf968"}, - {file = "regex-2021.4.4-cp36-cp36m-manylinux2010_x86_64.whl", hash = "sha256:7c2a1af393fcc09e898beba5dd59196edaa3116191cc7257f9224beaed3e1aa0"}, - {file = "regex-2021.4.4-cp36-cp36m-manylinux2014_aarch64.whl", hash = "sha256:c38c71df845e2aabb7fb0b920d11a1b5ac8526005e533a8920aea97efb8ec6a4"}, - {file = "regex-2021.4.4-cp36-cp36m-manylinux2014_i686.whl", hash = "sha256:96fcd1888ab4d03adfc9303a7b3c0bd78c5412b2bfbe76db5b56d9eae004907a"}, - {file = "regex-2021.4.4-cp36-cp36m-manylinux2014_x86_64.whl", hash = "sha256:ade17eb5d643b7fead300a1641e9f45401c98eee23763e9ed66a43f92f20b4a7"}, - {file = "regex-2021.4.4-cp36-cp36m-win32.whl", hash = "sha256:e8e5b509d5c2ff12f8418006d5a90e9436766133b564db0abaec92fd27fcee29"}, - {file = "regex-2021.4.4-cp36-cp36m-win_amd64.whl", hash = "sha256:11d773d75fa650cd36f68d7ca936e3c7afaae41b863b8c387a22aaa78d3c5c79"}, - {file = "regex-2021.4.4-cp37-cp37m-macosx_10_9_x86_64.whl", hash = "sha256:d3029c340cfbb3ac0a71798100ccc13b97dddf373a4ae56b6a72cf70dfd53bc8"}, - {file = "regex-2021.4.4-cp37-cp37m-manylinux1_i686.whl", hash = "sha256:18c071c3eb09c30a264879f0d310d37fe5d3a3111662438889ae2eb6fc570c31"}, - {file = "regex-2021.4.4-cp37-cp37m-manylinux1_x86_64.whl", hash = "sha256:4c557a7b470908b1712fe27fb1ef20772b78079808c87d20a90d051660b1d69a"}, - {file = "regex-2021.4.4-cp37-cp37m-manylinux2010_i686.whl", hash = "sha256:01afaf2ec48e196ba91b37451aa353cb7eda77efe518e481707e0515025f0cd5"}, - {file = "regex-2021.4.4-cp37-cp37m-manylinux2010_x86_64.whl", hash = "sha256:3a9cd17e6e5c7eb328517969e0cb0c3d31fd329298dd0c04af99ebf42e904f82"}, - {file = "regex-2021.4.4-cp37-cp37m-manylinux2014_aarch64.whl", hash = "sha256:90f11ff637fe8798933fb29f5ae1148c978cccb0452005bf4c69e13db951e765"}, - {file = "regex-2021.4.4-cp37-cp37m-manylinux2014_i686.whl", hash = "sha256:919859aa909429fb5aa9cf8807f6045592c85ef56fdd30a9a3747e513db2536e"}, - {file = "regex-2021.4.4-cp37-cp37m-manylinux2014_x86_64.whl", hash = "sha256:339456e7d8c06dd36a22e451d58ef72cef293112b559010db3d054d5560ef439"}, - {file = "regex-2021.4.4-cp37-cp37m-win32.whl", hash = "sha256:67bdb9702427ceddc6ef3dc382455e90f785af4c13d495f9626861763ee13f9d"}, - {file = "regex-2021.4.4-cp37-cp37m-win_amd64.whl", hash = "sha256:32e65442138b7b76dd8173ffa2cf67356b7bc1768851dded39a7a13bf9223da3"}, - {file = "regex-2021.4.4-cp38-cp38-macosx_10_9_x86_64.whl", hash = "sha256:1e1c20e29358165242928c2de1482fb2cf4ea54a6a6dea2bd7a0e0d8ee321500"}, - {file = "regex-2021.4.4-cp38-cp38-manylinux1_i686.whl", hash = "sha256:314d66636c494ed9c148a42731b3834496cc9a2c4251b1661e40936814542b14"}, - {file = "regex-2021.4.4-cp38-cp38-manylinux1_x86_64.whl", hash = "sha256:6d1b01031dedf2503631d0903cb563743f397ccaf6607a5e3b19a3d76fc10480"}, - {file = "regex-2021.4.4-cp38-cp38-manylinux2010_i686.whl", hash = "sha256:741a9647fcf2e45f3a1cf0e24f5e17febf3efe8d4ba1281dcc3aa0459ef424dc"}, - {file = "regex-2021.4.4-cp38-cp38-manylinux2010_x86_64.whl", hash = "sha256:4c46e22a0933dd783467cf32b3516299fb98cfebd895817d685130cc50cd1093"}, - {file = "regex-2021.4.4-cp38-cp38-manylinux2014_aarch64.whl", hash = "sha256:e512d8ef5ad7b898cdb2d8ee1cb09a8339e4f8be706d27eaa180c2f177248a10"}, - {file = "regex-2021.4.4-cp38-cp38-manylinux2014_i686.whl", hash = "sha256:980d7be47c84979d9136328d882f67ec5e50008681d94ecc8afa8a65ed1f4a6f"}, - {file = "regex-2021.4.4-cp38-cp38-manylinux2014_x86_64.whl", hash = "sha256:ce15b6d103daff8e9fee13cf7f0add05245a05d866e73926c358e871221eae87"}, - {file = "regex-2021.4.4-cp38-cp38-win32.whl", hash = "sha256:a91aa8619b23b79bcbeb37abe286f2f408d2f2d6f29a17237afda55bb54e7aac"}, - {file = "regex-2021.4.4-cp38-cp38-win_amd64.whl", hash = "sha256:c0502c0fadef0d23b128605d69b58edb2c681c25d44574fc673b0e52dce71ee2"}, - {file = "regex-2021.4.4-cp39-cp39-macosx_10_9_x86_64.whl", hash = "sha256:598585c9f0af8374c28edd609eb291b5726d7cbce16be6a8b95aa074d252ee17"}, - {file = "regex-2021.4.4-cp39-cp39-manylinux1_i686.whl", hash = "sha256:ee54ff27bf0afaf4c3b3a62bcd016c12c3fdb4ec4f413391a90bd38bc3624605"}, - {file = "regex-2021.4.4-cp39-cp39-manylinux1_x86_64.whl", hash = "sha256:7d9884d86dd4dd489e981d94a65cd30d6f07203d90e98f6f657f05170f6324c9"}, - {file = "regex-2021.4.4-cp39-cp39-manylinux2010_i686.whl", hash = "sha256:bf5824bfac591ddb2c1f0a5f4ab72da28994548c708d2191e3b87dd207eb3ad7"}, - {file = "regex-2021.4.4-cp39-cp39-manylinux2010_x86_64.whl", hash = "sha256:563085e55b0d4fb8f746f6a335893bda5c2cef43b2f0258fe1020ab1dd874df8"}, - {file = "regex-2021.4.4-cp39-cp39-manylinux2014_aarch64.whl", hash = "sha256:b9c3db21af35e3b3c05764461b262d6f05bbca08a71a7849fd79d47ba7bc33ed"}, - {file = "regex-2021.4.4-cp39-cp39-manylinux2014_i686.whl", hash = "sha256:3916d08be28a1149fb97f7728fca1f7c15d309a9f9682d89d79db75d5e52091c"}, - {file = "regex-2021.4.4-cp39-cp39-manylinux2014_x86_64.whl", hash = "sha256:fd45ff9293d9274c5008a2054ecef86a9bfe819a67c7be1afb65e69b405b3042"}, - {file = "regex-2021.4.4-cp39-cp39-win32.whl", hash = "sha256:fa4537fb4a98fe8fde99626e4681cc644bdcf2a795038533f9f711513a862ae6"}, - {file = "regex-2021.4.4-cp39-cp39-win_amd64.whl", hash = "sha256:97f29f57d5b84e73fbaf99ab3e26134e6687348e95ef6b48cfd2c06807005a07"}, - {file = "regex-2021.4.4.tar.gz", hash = "sha256:52ba3d3f9b942c49d7e4bc105bb28551c44065f139a65062ab7912bef10c9afb"}, + {file = "regex-2021.10.21-cp310-cp310-macosx_10_9_x86_64.whl", hash = "sha256:edff4e31d159672a7b9d70164b21289e4b53b239ce1dc945bf9643d266537573"}, + {file = "regex-2021.10.21-cp310-cp310-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:d6432daf42f2c487b357e1aa0bdc43193f050ff53a3188bfab20b88202b53027"}, + {file = "regex-2021.10.21-cp310-cp310-manylinux_2_5_i686.manylinux1_i686.manylinux_2_12_i686.manylinux2010_i686.whl", hash = "sha256:468de52dd3f20187ab5ca4fd265c1bea61a5346baef01ad0333a5e89fa9fad29"}, + {file = "regex-2021.10.21-cp310-cp310-manylinux_2_5_i686.manylinux1_i686.manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:c5a2ac760f2fc13a1c58131ec217779911890899ce1a0a63c9409bd23fecde6f"}, + {file = "regex-2021.10.21-cp310-cp310-manylinux_2_5_x86_64.manylinux1_x86_64.manylinux_2_12_x86_64.manylinux2010_x86_64.whl", hash = "sha256:ad1fedca001fefc3030d1e9022b038af429e58dc06a7e9c55e40bd1f834582ec"}, + {file = "regex-2021.10.21-cp310-cp310-win32.whl", hash = "sha256:9c613d797a3790f6b12e78a61e1cd29df7fc88135218467cf8b0891353292b9c"}, + {file = "regex-2021.10.21-cp310-cp310-win_amd64.whl", hash = "sha256:678d9a4ce79e1eaa4ebe88bc9769df52919eb30c597576a0deba1f3cf2360e65"}, + {file = "regex-2021.10.21-cp36-cp36m-macosx_10_9_x86_64.whl", hash = "sha256:2044174af237bb9c56ecc07294cf38623ee379e8dca14b01e970f8b015c71917"}, + {file = "regex-2021.10.21-cp36-cp36m-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:98743a2d827a135bf3390452be18d95839b947a099734d53c17e09a64fc09480"}, + {file = "regex-2021.10.21-cp36-cp36m-manylinux_2_5_i686.manylinux1_i686.manylinux_2_12_i686.manylinux2010_i686.whl", hash = "sha256:f1b23304855303bd97b5954edab63b8ddd56c91c41c6d4eba408228c0bae95f3"}, + {file = "regex-2021.10.21-cp36-cp36m-manylinux_2_5_i686.manylinux1_i686.manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:19c4fd59747236423016ccd89b9a6485d958bf1aa7a8a902a6ba28029107a87f"}, + {file = "regex-2021.10.21-cp36-cp36m-manylinux_2_5_x86_64.manylinux1_x86_64.manylinux_2_12_x86_64.manylinux2010_x86_64.whl", hash = "sha256:130a002fa386c976615a2f6d6dff0fcc25da24858994a36b14d2e3129dce7de2"}, + {file = "regex-2021.10.21-cp36-cp36m-win32.whl", hash = "sha256:8bd83d9b8ee125350cd666b55294f4bc9993c4f0d9b1be9344a318d0762e94cc"}, + {file = "regex-2021.10.21-cp36-cp36m-win_amd64.whl", hash = "sha256:98fe0e1b07a314f0a86dc58af4e717c379d48a403eddd8d966ab9b8bf91ce164"}, + {file = "regex-2021.10.21-cp37-cp37m-macosx_10_9_x86_64.whl", hash = "sha256:ded4748c7be6f31fb207387ee83a3a0f625e700defe32f268cb1d350ed6e4a66"}, + {file = "regex-2021.10.21-cp37-cp37m-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:f3da121de36a9ead0f32b44ea720ee8c87edbb59dca6bb980d18377d84ad58a3"}, + {file = "regex-2021.10.21-cp37-cp37m-manylinux_2_5_i686.manylinux1_i686.manylinux_2_12_i686.manylinux2010_i686.whl", hash = "sha256:b9dfba513eae785e3d868803f5a7e21a032cb2b038fa4a1ea7ec691037426ad3"}, + {file = "regex-2021.10.21-cp37-cp37m-manylinux_2_5_i686.manylinux1_i686.manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:2ff91696888755e96230138355cbe8ce2965d930d967d6cff7c636082d038c78"}, + {file = "regex-2021.10.21-cp37-cp37m-manylinux_2_5_x86_64.manylinux1_x86_64.manylinux_2_12_x86_64.manylinux2010_x86_64.whl", hash = "sha256:0f82de529d7595011a40573cc0f27422e59cafa94943b64a4d17d966d75f2c01"}, + {file = "regex-2021.10.21-cp37-cp37m-win32.whl", hash = "sha256:164e51ace4d00f07c519f85ec2209e8faaeab18bc77be6b35685c18d4ac1c22a"}, + {file = "regex-2021.10.21-cp37-cp37m-win_amd64.whl", hash = "sha256:e39eafa854e469d7225066c806c76b9a0acba5ff5ce36c82c0224b75e24888f2"}, + {file = "regex-2021.10.21-cp38-cp38-macosx_10_9_x86_64.whl", hash = "sha256:740a28580520b099b804776db1e919360fcbf30a734a14c5985d5e39a39e7237"}, + {file = "regex-2021.10.21-cp38-cp38-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:de7dbf72ae80f06e79444ff9614fb5e3a7956645d513b0e12d1bbe6f3ccebd11"}, + {file = "regex-2021.10.21-cp38-cp38-manylinux_2_5_i686.manylinux1_i686.manylinux_2_12_i686.manylinux2010_i686.whl", hash = "sha256:dc1a9bedf389bf3d3627a4d2b21cbdc5fe5e0f029d1f465972f4437833dcc946"}, + {file = "regex-2021.10.21-cp38-cp38-manylinux_2_5_i686.manylinux1_i686.manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:9cd14f22425beecf727f6dbdf5c893e46ecbc5ff16197c16a6f38a9066f2d4d5"}, + {file = "regex-2021.10.21-cp38-cp38-manylinux_2_5_x86_64.manylinux1_x86_64.manylinux_2_12_x86_64.manylinux2010_x86_64.whl", hash = "sha256:5b75a3db3aab0bfa51b6af3f820760779d360eb79f59e32c88c7fba648990b4f"}, + {file = "regex-2021.10.21-cp38-cp38-win32.whl", hash = "sha256:f68c71aabb10b1352a06515e25a425a703ba85660ae04cf074da5eb91c0af5e5"}, + {file = "regex-2021.10.21-cp38-cp38-win_amd64.whl", hash = "sha256:c0f49f1f03be3e4a5faaadc35db7afa2b83a871943b889f9f7bba56e0e2e8bd5"}, + {file = "regex-2021.10.21-cp39-cp39-macosx_10_9_x86_64.whl", hash = "sha256:201890fdc8a65396cfb6aa4493201353b2a6378e27d2de65234446f8329233cb"}, + {file = "regex-2021.10.21-cp39-cp39-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:fd1bfc6b7347de9f0ae1fb6f9080426bed6a9ca55b5766fa4fdf7b3a29ccae9c"}, + {file = "regex-2021.10.21-cp39-cp39-manylinux_2_5_i686.manylinux1_i686.manylinux_2_12_i686.manylinux2010_i686.whl", hash = "sha256:72a0b98d41c4508ed23a96eef41090f78630b44ba746e28cd621ecbe961e0a16"}, + {file = "regex-2021.10.21-cp39-cp39-manylinux_2_5_i686.manylinux1_i686.manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:3b5a0660a63b0703380758a7141b96cc1c1a13dee2b8e9c280a2522962fd12af"}, + {file = "regex-2021.10.21-cp39-cp39-manylinux_2_5_x86_64.manylinux1_x86_64.manylinux_2_12_x86_64.manylinux2010_x86_64.whl", hash = "sha256:f82d3adde46ac9188db3aa7e6e1690865ebb6448d245df5a3ea22284f70d9e46"}, + {file = "regex-2021.10.21-cp39-cp39-win32.whl", hash = "sha256:bc4637390235f1e3e2fcdd3e904ca0b42aa655ae28a78072248b2992b4ad4c08"}, + {file = "regex-2021.10.21-cp39-cp39-win_amd64.whl", hash = "sha256:74d03c256cf0aed81997e87be8e24297b5792c9718f3a735f5055ddfad392f06"}, + {file = "regex-2021.10.21.tar.gz", hash = "sha256:4832736b3f24617e63dc919ce8c4215680ba94250a5d9e710fcc0c5f457b5028"}, +] +requests = [ + {file = "requests-2.26.0-py2.py3-none-any.whl", hash = "sha256:6c1246513ecd5ecd4528a0906f910e8f0f9c6b8ec72030dc9fd154dc1a6efd24"}, + {file = "requests-2.26.0.tar.gz", hash = "sha256:b8aa58f8cf793ffd8782d3d8cb19e66ef36f7aba4353eec859e74678b01b07a7"}, ] signxml-openadr = [ {file = "signxml-openadr-2.9.1.tar.gz", hash = "sha256:93bf41824247a36fa09d8ac23435f5675aaa9e1e58fcb3809a549fba9a905c23"}, @@ -1026,6 +1497,10 @@ toml = [ {file = "toml-0.10.2-py2.py3-none-any.whl", hash = "sha256:806143ae5bfb6a3c6e736a764057db0e6a0e05e338b5630894a5f779cabb4f9b"}, {file = "toml-0.10.2.tar.gz", hash = "sha256:b3bda1d108d5dd99f4a20d24d9c348e91c4db7ab1b749200bded2f839ccbe68f"}, ] +tomli = [ + {file = "tomli-1.2.1-py3-none-any.whl", hash = "sha256:8dd0e9524d6f386271a36b41dbf6c57d8e32fd96fd22b6584679dc569d20899f"}, + {file = "tomli-1.2.1.tar.gz", hash = "sha256:a5b75cb6f3968abb47af1b40c1819dc519ea82bcc065776a866e8d74c5ca9442"}, +] typed-ast = [ {file = "typed_ast-1.4.3-cp35-cp35m-manylinux1_i686.whl", hash = "sha256:2068531575a125b87a41802130fa7e29f26c09a2833fea68d9a40cf33902eba6"}, {file = "typed_ast-1.4.3-cp35-cp35m-manylinux1_x86_64.whl", hash = "sha256:c907f561b1e83e93fad565bac5ba9c22d96a54e7ea0267c708bffe863cbe4075"}, @@ -1059,58 +1534,195 @@ typed-ast = [ {file = "typed_ast-1.4.3.tar.gz", hash = "sha256:fb1bbeac803adea29cedd70781399c99138358c26d05fcbd23c13016b7f5ec65"}, ] typing-extensions = [ - {file = "typing_extensions-3.10.0.0-py2-none-any.whl", hash = "sha256:0ac0f89795dd19de6b97debb0c6af1c70987fd80a2d62d1958f7e56fcc31b497"}, - {file = "typing_extensions-3.10.0.0-py3-none-any.whl", hash = "sha256:779383f6086d90c99ae41cf0ff39aac8a7937a9283ce0a414e5dd782f4c94a84"}, - {file = "typing_extensions-3.10.0.0.tar.gz", hash = "sha256:50b6f157849174217d0656f99dc82fe932884fb250826c18350e159ec6cdf342"}, + {file = "typing_extensions-3.10.0.2-py2-none-any.whl", hash = "sha256:d8226d10bc02a29bcc81df19a26e56a9647f8b0a6d4a83924139f4a8b01f17b7"}, + {file = "typing_extensions-3.10.0.2-py3-none-any.whl", hash = "sha256:f1d25edafde516b146ecd0613dabcc61409817af4766fbbcfb8d1ad4ec441a34"}, + {file = "typing_extensions-3.10.0.2.tar.gz", hash = "sha256:49f75d16ff11f1cd258e1b988ccff82a3ca5570217d7ad8c5f48205dd99a677e"}, ] tzlocal = [ {file = "tzlocal-2.1-py2.py3-none-any.whl", hash = "sha256:e2cb6c6b5b604af38597403e9852872d7f534962ae2954c7f35efcb1ccacf4a4"}, {file = "tzlocal-2.1.tar.gz", hash = "sha256:643c97c5294aedc737780a49d9df30889321cbe1204eac2c2ec6134035a92e44"}, ] +urllib3 = [ + {file = "urllib3-1.26.7-py2.py3-none-any.whl", hash = "sha256:c4fdf4019605b6e5423637e01bc9fe4daef873709a7973e195ceba0a62bbc844"}, + {file = "urllib3-1.26.7.tar.gz", hash = "sha256:4987c65554f7a2dbf30c18fd48778ef124af6fab771a377103da0585e2336ece"}, +] virtualenv = [ - {file = "virtualenv-20.4.7-py2.py3-none-any.whl", hash = "sha256:2b0126166ea7c9c3661f5b8e06773d28f83322de7a3ff7d06f0aed18c9de6a76"}, - {file = "virtualenv-20.4.7.tar.gz", hash = "sha256:14fdf849f80dbb29a4eb6caa9875d476ee2a5cf76a5f5415fa2f1606010ab467"}, + {file = "virtualenv-20.8.1-py2.py3-none-any.whl", hash = "sha256:10062e34c204b5e4ec5f62e6ef2473f8ba76513a9a617e873f1f8fb4a519d300"}, + {file = "virtualenv-20.8.1.tar.gz", hash = "sha256:bcc17f0b3a29670dd777d6f0755a4c04f28815395bca279cdcb213b97199a6b8"}, +] +volttron-client = [ + {file = "volttron-client-0.3.4.tar.gz", hash = "sha256:a426efd8c5d10c5f574b78501082c3ac76714903c29d4fe9c8c9d335abc03d47"}, + {file = "volttron_client-0.3.4-py3-none-any.whl", hash = "sha256:60ba299b75431e5d249ff0cf871e1978ab75e65814a28b6252a5f7612f1ec658"}, +] +volttron-utils = [ + {file = "volttron-utils-0.3.2.tar.gz", hash = "sha256:abb35d335ba60ecc06e9e1bbcbb5dd0b3dbc9445e58a366dfaa6071e4ce5ce80"}, + {file = "volttron_utils-0.3.2-py3-none-any.whl", hash = "sha256:240e6f84ef164d7740b0c035b01efbd00c194c1b5b12bb3b73fe28386fff6f5f"}, +] +watchdog = [ + {file = "watchdog-2.1.6-cp310-cp310-macosx_10_9_universal2.whl", hash = "sha256:9693f35162dc6208d10b10ddf0458cc09ad70c30ba689d9206e02cd836ce28a3"}, + {file = "watchdog-2.1.6-cp310-cp310-macosx_10_9_x86_64.whl", hash = "sha256:aba5c812f8ee8a3ff3be51887ca2d55fb8e268439ed44110d3846e4229eb0e8b"}, + {file = "watchdog-2.1.6-cp310-cp310-macosx_11_0_arm64.whl", hash = "sha256:4ae38bf8ba6f39d5b83f78661273216e7db5b00f08be7592062cb1fc8b8ba542"}, + {file = "watchdog-2.1.6-cp36-cp36m-macosx_10_9_x86_64.whl", hash = "sha256:ad6f1796e37db2223d2a3f302f586f74c72c630b48a9872c1e7ae8e92e0ab669"}, + {file = "watchdog-2.1.6-cp37-cp37m-macosx_10_9_x86_64.whl", hash = "sha256:922a69fa533cb0c793b483becaaa0845f655151e7256ec73630a1b2e9ebcb660"}, + {file = "watchdog-2.1.6-cp38-cp38-macosx_10_9_universal2.whl", hash = "sha256:b2fcf9402fde2672545b139694284dc3b665fd1be660d73eca6805197ef776a3"}, + {file = "watchdog-2.1.6-cp38-cp38-macosx_10_9_x86_64.whl", hash = "sha256:3386b367e950a11b0568062b70cc026c6f645428a698d33d39e013aaeda4cc04"}, + {file = "watchdog-2.1.6-cp38-cp38-macosx_11_0_arm64.whl", hash = "sha256:8f1c00aa35f504197561060ca4c21d3cc079ba29cf6dd2fe61024c70160c990b"}, + {file = "watchdog-2.1.6-cp39-cp39-macosx_10_9_universal2.whl", hash = "sha256:b52b88021b9541a60531142b0a451baca08d28b74a723d0c99b13c8c8d48d604"}, + {file = "watchdog-2.1.6-cp39-cp39-macosx_10_9_x86_64.whl", hash = "sha256:8047da932432aa32c515ec1447ea79ce578d0559362ca3605f8e9568f844e3c6"}, + {file = "watchdog-2.1.6-cp39-cp39-macosx_11_0_arm64.whl", hash = "sha256:e92c2d33858c8f560671b448205a268096e17870dcf60a9bb3ac7bfbafb7f5f9"}, + {file = "watchdog-2.1.6-pp37-pypy37_pp73-macosx_10_9_x86_64.whl", hash = "sha256:b7d336912853d7b77f9b2c24eeed6a5065d0a0cc0d3b6a5a45ad6d1d05fb8cd8"}, + {file = "watchdog-2.1.6-py3-none-manylinux2014_aarch64.whl", hash = "sha256:cca7741c0fcc765568350cb139e92b7f9f3c9a08c4f32591d18ab0a6ac9e71b6"}, + {file = "watchdog-2.1.6-py3-none-manylinux2014_armv7l.whl", hash = "sha256:25fb5240b195d17de949588628fdf93032ebf163524ef08933db0ea1f99bd685"}, + {file = "watchdog-2.1.6-py3-none-manylinux2014_i686.whl", hash = "sha256:be9be735f827820a06340dff2ddea1fb7234561fa5e6300a62fe7f54d40546a0"}, + {file = "watchdog-2.1.6-py3-none-manylinux2014_ppc64.whl", hash = "sha256:d0d19fb2441947b58fbf91336638c2b9f4cc98e05e1045404d7a4cb7cddc7a65"}, + {file = "watchdog-2.1.6-py3-none-manylinux2014_ppc64le.whl", hash = "sha256:3becdb380d8916c873ad512f1701f8a92ce79ec6978ffde92919fd18d41da7fb"}, + {file = "watchdog-2.1.6-py3-none-manylinux2014_s390x.whl", hash = "sha256:ae67501c95606072aafa865b6ed47343ac6484472a2f95490ba151f6347acfc2"}, + {file = "watchdog-2.1.6-py3-none-manylinux2014_x86_64.whl", hash = "sha256:e0f30db709c939cabf64a6dc5babb276e6d823fd84464ab916f9b9ba5623ca15"}, + {file = "watchdog-2.1.6-py3-none-win32.whl", hash = "sha256:e02794ac791662a5eafc6ffeaf9bcc149035a0e48eb0a9d40a8feb4622605a3d"}, + {file = "watchdog-2.1.6-py3-none-win_amd64.whl", hash = "sha256:bd9ba4f332cf57b2c1f698be0728c020399ef3040577cde2939f2e045b39c1e5"}, + {file = "watchdog-2.1.6-py3-none-win_ia64.whl", hash = "sha256:a0f1c7edf116a12f7245be06120b1852275f9506a7d90227648b250755a03923"}, + {file = "watchdog-2.1.6.tar.gz", hash = "sha256:a36e75df6c767cbf46f61a91c70b3ba71811dfa0aca4a324d9407a06a8b7a2e7"}, +] +watchdog-gevent = [ + {file = "watchdog_gevent-0.1.1-py3-none-any.whl", hash = "sha256:dcebd07668b472790ad0e7a7b40ad8c365e156c40cf54ffd5499c65c65c4f66f"}, + {file = "watchdog_gevent-0.1.1.tar.gz", hash = "sha256:71dced5ec451b3deafd1dc73baaf52a7ce4d54d9a1eb606f9370235125f2424a"}, ] xmltodict = [ {file = "xmltodict-0.12.0-py2.py3-none-any.whl", hash = "sha256:8bbcb45cc982f48b2ca8fe7e7827c5d792f217ecf1792626f808bf41c3b86051"}, {file = "xmltodict-0.12.0.tar.gz", hash = "sha256:50d8c638ed7ecb88d90561beedbf720c9b4e851a9fa6c47ebd64e99d166d8a21"}, ] yarl = [ - {file = "yarl-1.6.3-cp36-cp36m-macosx_10_14_x86_64.whl", hash = "sha256:0355a701b3998dcd832d0dc47cc5dedf3874f966ac7f870e0f3a6788d802d434"}, - {file = "yarl-1.6.3-cp36-cp36m-manylinux1_i686.whl", hash = "sha256:bafb450deef6861815ed579c7a6113a879a6ef58aed4c3a4be54400ae8871478"}, - {file = "yarl-1.6.3-cp36-cp36m-manylinux2014_aarch64.whl", hash = "sha256:547f7665ad50fa8563150ed079f8e805e63dd85def6674c97efd78eed6c224a6"}, - {file = "yarl-1.6.3-cp36-cp36m-manylinux2014_i686.whl", hash = "sha256:63f90b20ca654b3ecc7a8d62c03ffa46999595f0167d6450fa8383bab252987e"}, - {file = "yarl-1.6.3-cp36-cp36m-manylinux2014_ppc64le.whl", hash = "sha256:97b5bdc450d63c3ba30a127d018b866ea94e65655efaf889ebeabc20f7d12406"}, - {file = "yarl-1.6.3-cp36-cp36m-manylinux2014_s390x.whl", hash = "sha256:d8d07d102f17b68966e2de0e07bfd6e139c7c02ef06d3a0f8d2f0f055e13bb76"}, - {file = "yarl-1.6.3-cp36-cp36m-manylinux2014_x86_64.whl", hash = "sha256:15263c3b0b47968c1d90daa89f21fcc889bb4b1aac5555580d74565de6836366"}, - {file = "yarl-1.6.3-cp36-cp36m-win32.whl", hash = "sha256:b5dfc9a40c198334f4f3f55880ecf910adebdcb2a0b9a9c23c9345faa9185721"}, - {file = "yarl-1.6.3-cp36-cp36m-win_amd64.whl", hash = "sha256:b2e9a456c121e26d13c29251f8267541bd75e6a1ccf9e859179701c36a078643"}, - {file = "yarl-1.6.3-cp37-cp37m-macosx_10_14_x86_64.whl", hash = "sha256:ce3beb46a72d9f2190f9e1027886bfc513702d748047b548b05dab7dfb584d2e"}, - {file = "yarl-1.6.3-cp37-cp37m-manylinux1_i686.whl", hash = "sha256:2ce4c621d21326a4a5500c25031e102af589edb50c09b321049e388b3934eec3"}, - {file = "yarl-1.6.3-cp37-cp37m-manylinux2014_aarch64.whl", hash = "sha256:d26608cf178efb8faa5ff0f2d2e77c208f471c5a3709e577a7b3fd0445703ac8"}, - {file = "yarl-1.6.3-cp37-cp37m-manylinux2014_i686.whl", hash = "sha256:4c5bcfc3ed226bf6419f7a33982fb4b8ec2e45785a0561eb99274ebbf09fdd6a"}, - {file = "yarl-1.6.3-cp37-cp37m-manylinux2014_ppc64le.whl", hash = "sha256:4736eaee5626db8d9cda9eb5282028cc834e2aeb194e0d8b50217d707e98bb5c"}, - {file = "yarl-1.6.3-cp37-cp37m-manylinux2014_s390x.whl", hash = "sha256:68dc568889b1c13f1e4745c96b931cc94fdd0defe92a72c2b8ce01091b22e35f"}, - {file = "yarl-1.6.3-cp37-cp37m-manylinux2014_x86_64.whl", hash = "sha256:7356644cbed76119d0b6bd32ffba704d30d747e0c217109d7979a7bc36c4d970"}, - {file = "yarl-1.6.3-cp37-cp37m-win32.whl", hash = "sha256:00d7ad91b6583602eb9c1d085a2cf281ada267e9a197e8b7cae487dadbfa293e"}, - {file = "yarl-1.6.3-cp37-cp37m-win_amd64.whl", hash = "sha256:69ee97c71fee1f63d04c945f56d5d726483c4762845400a6795a3b75d56b6c50"}, - {file = "yarl-1.6.3-cp38-cp38-macosx_10_14_x86_64.whl", hash = "sha256:e46fba844f4895b36f4c398c5af062a9808d1f26b2999c58909517384d5deda2"}, - {file = "yarl-1.6.3-cp38-cp38-manylinux1_i686.whl", hash = "sha256:31ede6e8c4329fb81c86706ba8f6bf661a924b53ba191b27aa5fcee5714d18ec"}, - {file = "yarl-1.6.3-cp38-cp38-manylinux2014_aarch64.whl", hash = "sha256:fcbb48a93e8699eae920f8d92f7160c03567b421bc17362a9ffbbd706a816f71"}, - {file = "yarl-1.6.3-cp38-cp38-manylinux2014_i686.whl", hash = "sha256:72a660bdd24497e3e84f5519e57a9ee9220b6f3ac4d45056961bf22838ce20cc"}, - {file = "yarl-1.6.3-cp38-cp38-manylinux2014_ppc64le.whl", hash = "sha256:324ba3d3c6fee56e2e0b0d09bf5c73824b9f08234339d2b788af65e60040c959"}, - {file = "yarl-1.6.3-cp38-cp38-manylinux2014_s390x.whl", hash = "sha256:e6b5460dc5ad42ad2b36cca524491dfcaffbfd9c8df50508bddc354e787b8dc2"}, - {file = "yarl-1.6.3-cp38-cp38-manylinux2014_x86_64.whl", hash = "sha256:6d6283d8e0631b617edf0fd726353cb76630b83a089a40933043894e7f6721e2"}, - {file = "yarl-1.6.3-cp38-cp38-win32.whl", hash = "sha256:9ede61b0854e267fd565e7527e2f2eb3ef8858b301319be0604177690e1a3896"}, - {file = "yarl-1.6.3-cp38-cp38-win_amd64.whl", hash = "sha256:f0b059678fd549c66b89bed03efcabb009075bd131c248ecdf087bdb6faba24a"}, - {file = "yarl-1.6.3-cp39-cp39-macosx_10_14_x86_64.whl", hash = "sha256:329412812ecfc94a57cd37c9d547579510a9e83c516bc069470db5f75684629e"}, - {file = "yarl-1.6.3-cp39-cp39-manylinux1_i686.whl", hash = "sha256:c49ff66d479d38ab863c50f7bb27dee97c6627c5fe60697de15529da9c3de724"}, - {file = "yarl-1.6.3-cp39-cp39-manylinux2014_aarch64.whl", hash = "sha256:f040bcc6725c821a4c0665f3aa96a4d0805a7aaf2caf266d256b8ed71b9f041c"}, - {file = "yarl-1.6.3-cp39-cp39-manylinux2014_i686.whl", hash = "sha256:d5c32c82990e4ac4d8150fd7652b972216b204de4e83a122546dce571c1bdf25"}, - {file = "yarl-1.6.3-cp39-cp39-manylinux2014_ppc64le.whl", hash = "sha256:d597767fcd2c3dc49d6eea360c458b65643d1e4dbed91361cf5e36e53c1f8c96"}, - {file = "yarl-1.6.3-cp39-cp39-manylinux2014_s390x.whl", hash = "sha256:8aa3decd5e0e852dc68335abf5478a518b41bf2ab2f330fe44916399efedfae0"}, - {file = "yarl-1.6.3-cp39-cp39-manylinux2014_x86_64.whl", hash = "sha256:73494d5b71099ae8cb8754f1df131c11d433b387efab7b51849e7e1e851f07a4"}, - {file = "yarl-1.6.3-cp39-cp39-win32.whl", hash = "sha256:5b883e458058f8d6099e4420f0cc2567989032b5f34b271c0827de9f1079a424"}, - {file = "yarl-1.6.3-cp39-cp39-win_amd64.whl", hash = "sha256:4953fb0b4fdb7e08b2f3b3be80a00d28c5c8a2056bb066169de00e6501b986b6"}, - {file = "yarl-1.6.3.tar.gz", hash = "sha256:8a9066529240171b68893d60dca86a763eae2139dd42f42106b03cf4b426bf10"}, + {file = "yarl-1.7.0-cp310-cp310-macosx_10_9_universal2.whl", hash = "sha256:e35d8230e4b08d86ea65c32450533b906a8267a87b873f2954adeaecede85169"}, + {file = "yarl-1.7.0-cp310-cp310-macosx_10_9_x86_64.whl", hash = "sha256:eb4b3f277880c314e47720b4b6bb2c85114ab3c04c5442c9bc7006b3787904d8"}, + {file = "yarl-1.7.0-cp310-cp310-macosx_11_0_arm64.whl", hash = "sha256:c7015dcedb91d90a138eebdc7e432aec8966e0147ab2a55f2df27b1904fa7291"}, + {file = "yarl-1.7.0-cp310-cp310-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:bb3e478175e15e00d659fb0354a6a8db71a7811a2a5052aed98048bc972e5d2b"}, + {file = "yarl-1.7.0-cp310-cp310-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:8b8c409aa3a7966647e7c1c524846b362a6bcbbe120bf8a176431f940d2b9a2e"}, + {file = "yarl-1.7.0-cp310-cp310-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:b22ea41c7e98170474a01e3eded1377d46b2dfaef45888a0005c683eaaa49285"}, + {file = "yarl-1.7.0-cp310-cp310-manylinux_2_5_i686.manylinux1_i686.manylinux_2_12_i686.manylinux2010_i686.whl", hash = "sha256:a7dfc46add4cfe5578013dbc4127893edc69fe19132d2836ff2f6e49edc5ecd6"}, + {file = "yarl-1.7.0-cp310-cp310-manylinux_2_5_x86_64.manylinux1_x86_64.manylinux_2_12_x86_64.manylinux2010_x86_64.whl", hash = "sha256:82ff6f85f67500a4f74885d81659cd270eb24dfe692fe44e622b8a2fd57e7279"}, + {file = "yarl-1.7.0-cp310-cp310-musllinux_1_1_aarch64.whl", hash = "sha256:f3cd2158b2ed0fb25c6811adfdcc47224efe075f2d68a750071dacc03a7a66e4"}, + {file = "yarl-1.7.0-cp310-cp310-musllinux_1_1_i686.whl", hash = "sha256:59c0f13f9592820c51280d1cf811294d753e4a18baf90f0139d1dc93d4b6fc5f"}, + {file = "yarl-1.7.0-cp310-cp310-musllinux_1_1_ppc64le.whl", hash = "sha256:7f7655ad83d1a8afa48435a449bf2f3009293da1604f5dd95b5ddcf5f673bd69"}, + {file = "yarl-1.7.0-cp310-cp310-musllinux_1_1_s390x.whl", hash = "sha256:aa9f0d9b62d15182341b3e9816582f46182cab91c1a57b2d308b9a3c4e2c4f78"}, + {file = "yarl-1.7.0-cp310-cp310-musllinux_1_1_x86_64.whl", hash = "sha256:fdd1b90c225a653b1bd1c0cae8edf1957892b9a09c8bf7ee6321eeb8208eac0f"}, + {file = "yarl-1.7.0-cp310-cp310-win32.whl", hash = "sha256:7c8d0bb76eabc5299db203e952ec55f8f4c53f08e0df4285aac8c92bd9e12675"}, + {file = "yarl-1.7.0-cp310-cp310-win_amd64.whl", hash = "sha256:622a36fa779efb4ff9eff5fe52730ff17521431379851a31e040958fc251670c"}, + {file = "yarl-1.7.0-cp36-cp36m-macosx_10_9_x86_64.whl", hash = "sha256:3d461b7a8e139b9e4b41f62eb417ffa0b98d1c46d4caf14c845e6a3b349c0bb1"}, + {file = "yarl-1.7.0-cp36-cp36m-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:81cfacdd1e40bc931b5519499342efa388d24d262c30a3d31187bfa04f4a7001"}, + {file = "yarl-1.7.0-cp36-cp36m-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:821b978f2152be7695d4331ef0621d207aedf9bbd591ba23a63412a3efc29a01"}, + {file = "yarl-1.7.0-cp36-cp36m-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:b64bd24c8c9a487f4a12260dc26732bf41028816dbf0c458f17864fbebdb3131"}, + {file = "yarl-1.7.0-cp36-cp36m-manylinux_2_5_i686.manylinux1_i686.manylinux_2_12_i686.manylinux2010_i686.whl", hash = "sha256:98c9ddb92b60a83c21be42c776d3d9d5ec632a762a094c41bda37b7dfbd2cd83"}, + {file = "yarl-1.7.0-cp36-cp36m-manylinux_2_5_x86_64.manylinux1_x86_64.manylinux_2_12_x86_64.manylinux2010_x86_64.whl", hash = "sha256:a532d75ca74431c053a88a802e161fb3d651b8bf5821a3440bc3616e38754583"}, + {file = "yarl-1.7.0-cp36-cp36m-musllinux_1_1_aarch64.whl", hash = "sha256:053e09817eafb892e94e172d05406c1b3a22a93bc68f6eff5198363a3d764459"}, + {file = "yarl-1.7.0-cp36-cp36m-musllinux_1_1_i686.whl", hash = "sha256:98c51f02d542945d306c8e934aa2c1e66ba5e9c1c86b5bf37f3a51c8a747067e"}, + {file = "yarl-1.7.0-cp36-cp36m-musllinux_1_1_ppc64le.whl", hash = "sha256:15ec41a5a5fdb7bace6d7b16701f9440007a82734f69127c0fbf6d87e10f4a1e"}, + {file = "yarl-1.7.0-cp36-cp36m-musllinux_1_1_s390x.whl", hash = "sha256:a7f08819dba1e1255d6991ed37448a1bf4b1352c004bcd899b9da0c47958513d"}, + {file = "yarl-1.7.0-cp36-cp36m-musllinux_1_1_x86_64.whl", hash = "sha256:8e3ffab21db0542ffd1887f3b9575ddd58961f2cf61429cb6458afc00c4581e0"}, + {file = "yarl-1.7.0-cp36-cp36m-win32.whl", hash = "sha256:50127634f519b2956005891507e3aa4ac345f66a7ea7bbc2d7dcba7401f41898"}, + {file = "yarl-1.7.0-cp36-cp36m-win_amd64.whl", hash = "sha256:36ec44f15193f6d5288d42ebb8e751b967ebdfb72d6830983838d45ab18edb4f"}, + {file = "yarl-1.7.0-cp37-cp37m-macosx_10_9_x86_64.whl", hash = "sha256:ec1b5a25a25c880c976d0bb3d107def085bb08dbb3db7f4442e0a2b980359d24"}, + {file = "yarl-1.7.0-cp37-cp37m-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:b36f5a63c891f813c6f04ef19675b382efc190fd5ce7e10ab19386d2548bca06"}, + {file = "yarl-1.7.0-cp37-cp37m-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:38173b8c3a29945e7ecade9a3f6ff39581eee8201338ee6a2c8882db5df3e806"}, + {file = "yarl-1.7.0-cp37-cp37m-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:8ba402f32184f0b405fb281b93bd0d8ab7e3257735b57b62a6ed2e94cdf4fe50"}, + {file = "yarl-1.7.0-cp37-cp37m-manylinux_2_5_i686.manylinux1_i686.manylinux_2_12_i686.manylinux2010_i686.whl", hash = "sha256:be52bc5208d767cdd8308a9e93059b3b36d1e048fecbea0e0346d0d24a76adc0"}, + {file = "yarl-1.7.0-cp37-cp37m-manylinux_2_5_x86_64.manylinux1_x86_64.manylinux_2_12_x86_64.manylinux2010_x86_64.whl", hash = "sha256:08c2044a956f4ef30405f2f433ce77f1f57c2c773bf81ae43201917831044d5a"}, + {file = "yarl-1.7.0-cp37-cp37m-musllinux_1_1_aarch64.whl", hash = "sha256:484d61c047c45670ef5967653a1d0783e232c54bf9dd786a7737036828fa8d54"}, + {file = "yarl-1.7.0-cp37-cp37m-musllinux_1_1_i686.whl", hash = "sha256:b7de92a4af85cfcaf4081f8aa6165b1d63ee5de150af3ee85f954145f93105a7"}, + {file = "yarl-1.7.0-cp37-cp37m-musllinux_1_1_ppc64le.whl", hash = "sha256:376e41775aab79c5575534924a386c8e0f1a5d91db69fc6133fd27a489bcaf10"}, + {file = "yarl-1.7.0-cp37-cp37m-musllinux_1_1_s390x.whl", hash = "sha256:8a8b10d0e7bac154f959b709fcea593cda527b234119311eb950096653816a86"}, + {file = "yarl-1.7.0-cp37-cp37m-musllinux_1_1_x86_64.whl", hash = "sha256:f46cd4c43e6175030e2a56def8f1d83b64e6706eeb2bb9ab0ef4756f65eab23f"}, + {file = "yarl-1.7.0-cp37-cp37m-win32.whl", hash = "sha256:b28cfb46140efe1a6092b8c5c4994a1fe70dc83c38fbcea4992401e0c6fb9cce"}, + {file = "yarl-1.7.0-cp37-cp37m-win_amd64.whl", hash = "sha256:9624154ec9c02a776802da1086eed7f5034bd1971977f5146233869c2ac80297"}, + {file = "yarl-1.7.0-cp38-cp38-macosx_10_9_universal2.whl", hash = "sha256:69945d13e1bbf81784a9bc48824feb9cd66491e6a503d4e83f6cd7c7cc861361"}, + {file = "yarl-1.7.0-cp38-cp38-macosx_10_9_x86_64.whl", hash = "sha256:46a742ed9e363bd01be64160ce7520e92e11989bd4cb224403cfd31c101cc83d"}, + {file = "yarl-1.7.0-cp38-cp38-macosx_11_0_arm64.whl", hash = "sha256:cb4ff1ac7cb4500f43581b3f4cbd627d702143aa6be1fdc1fa3ebffaf4dc1be5"}, + {file = "yarl-1.7.0-cp38-cp38-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:3ad51e17cd65ea3debb0e10f0120cf8dd987c741fe423ed2285087368090b33d"}, + {file = "yarl-1.7.0-cp38-cp38-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:7e37786ea89a5d3ffbbf318ea9790926f8dfda83858544f128553c347ad143c6"}, + {file = "yarl-1.7.0-cp38-cp38-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:c63c1e208f800daad71715786bfeb1cecdc595d87e2e9b1cd234fd6e597fd71d"}, + {file = "yarl-1.7.0-cp38-cp38-manylinux_2_5_i686.manylinux1_i686.manylinux_2_12_i686.manylinux2010_i686.whl", hash = "sha256:91cbe24300c11835ef186436363352b3257db7af165e0a767f4f17aa25761388"}, + {file = "yarl-1.7.0-cp38-cp38-manylinux_2_5_x86_64.manylinux1_x86_64.manylinux_2_12_x86_64.manylinux2010_x86_64.whl", hash = "sha256:e510dbec7c59d32eaa61ffa48173d5e3d7170a67f4a03e8f5e2e9e3971aca622"}, + {file = "yarl-1.7.0-cp38-cp38-musllinux_1_1_aarch64.whl", hash = "sha256:3def6e681cc02397e5d8141ee97b41d02932b2bcf0fb34532ad62855eab7c60e"}, + {file = "yarl-1.7.0-cp38-cp38-musllinux_1_1_i686.whl", hash = "sha256:263c81b94e6431942b27f6f671fa62f430a0a5c14bb255f2ab69eeb9b2b66ff7"}, + {file = "yarl-1.7.0-cp38-cp38-musllinux_1_1_ppc64le.whl", hash = "sha256:e78c91faefe88d601ddd16e3882918dbde20577a2438e2320f8239c8b7507b8f"}, + {file = "yarl-1.7.0-cp38-cp38-musllinux_1_1_s390x.whl", hash = "sha256:22b2430c49713bfb2f0a0dd4a8d7aab218b28476ba86fd1c78ad8899462cbcf2"}, + {file = "yarl-1.7.0-cp38-cp38-musllinux_1_1_x86_64.whl", hash = "sha256:2e7ad9db939082f5d0b9269cfd92c025cb8f2fbbb1f1b9dc5a393c639db5bd92"}, + {file = "yarl-1.7.0-cp38-cp38-win32.whl", hash = "sha256:3a31e4a8dcb1beaf167b7e7af61b88cb961b220db8d3ba1c839723630e57eef7"}, + {file = "yarl-1.7.0-cp38-cp38-win_amd64.whl", hash = "sha256:d579957439933d752358c6a300c93110f84aae67b63dd0c19dde6ecbf4056f6b"}, + {file = "yarl-1.7.0-cp39-cp39-macosx_10_9_universal2.whl", hash = "sha256:87721b549505a546eb003252185103b5ec8147de6d3ad3714d148a5a67b6fe53"}, + {file = "yarl-1.7.0-cp39-cp39-macosx_10_9_x86_64.whl", hash = "sha256:a1fa866fa24d9f4108f9e58ea8a2135655419885cdb443e36b39a346e1181532"}, + {file = "yarl-1.7.0-cp39-cp39-macosx_11_0_arm64.whl", hash = "sha256:1d3b8449dfedfe94eaff2b77954258b09b24949f6818dfa444b05dbb05ae1b7e"}, + {file = "yarl-1.7.0-cp39-cp39-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:db2372e350794ce8b9f810feb094c606b7e0e4aa6807141ac4fadfe5ddd75bb0"}, + {file = "yarl-1.7.0-cp39-cp39-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:a06d9d0b9a97fa99b84fee71d9dd11e69e21ac8a27229089f07b5e5e50e8d63c"}, + {file = "yarl-1.7.0-cp39-cp39-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:a3455c2456d6307bcfa80bc1157b8603f7d93573291f5bdc7144489ca0df4628"}, + {file = "yarl-1.7.0-cp39-cp39-manylinux_2_5_i686.manylinux1_i686.manylinux_2_12_i686.manylinux2010_i686.whl", hash = "sha256:d30d67e3486aea61bb2cbf7cf81385364c2e4f7ce7469a76ed72af76a5cdfe6b"}, + {file = "yarl-1.7.0-cp39-cp39-manylinux_2_5_x86_64.manylinux1_x86_64.manylinux_2_12_x86_64.manylinux2010_x86_64.whl", hash = "sha256:c18a4b286e8d780c3a40c31d7b79836aa93b720f71d5743f20c08b7e049ca073"}, + {file = "yarl-1.7.0-cp39-cp39-musllinux_1_1_aarch64.whl", hash = "sha256:d54c925396e7891666cabc0199366ca55b27d003393465acef63fd29b8b7aa92"}, + {file = "yarl-1.7.0-cp39-cp39-musllinux_1_1_i686.whl", hash = "sha256:64773840952de17851a1c7346ad7f71688c77e74248d1f0bc230e96680f84028"}, + {file = "yarl-1.7.0-cp39-cp39-musllinux_1_1_ppc64le.whl", hash = "sha256:acbf1756d9dc7cd0ae943d883be72e84e04396f6c2ff93a6ddeca929d562039f"}, + {file = "yarl-1.7.0-cp39-cp39-musllinux_1_1_s390x.whl", hash = "sha256:2e48f27936aa838939c798f466c851ba4ae79e347e8dfce43b009c64b930df12"}, + {file = "yarl-1.7.0-cp39-cp39-musllinux_1_1_x86_64.whl", hash = "sha256:1beef4734ca1ad40a9d8c6b20a76ab46e3a2ed09f38561f01e4aa2ea82cafcef"}, + {file = "yarl-1.7.0-cp39-cp39-win32.whl", hash = "sha256:8ee78c9a5f3c642219d4607680a4693b59239c27a3aa608b64ef79ddc9698039"}, + {file = "yarl-1.7.0-cp39-cp39-win_amd64.whl", hash = "sha256:d750503682605088a14d29a4701548c15c510da4f13c8b17409c4097d5b04c52"}, + {file = "yarl-1.7.0.tar.gz", hash = "sha256:8e7ebaf62e19c2feb097ffb7c94deb0f0c9fab52590784c8cd679d30ab009162"}, +] +zmq = [ + {file = "zmq-0.0.0.tar.gz", hash = "sha256:6b1a1de53338646e8c8405803cffb659e8eb7bb02fff4c9be62a7acfac8370c9"}, + {file = "zmq-0.0.0.zip", hash = "sha256:21cfc6be254c9bc25e4dabb8a3b2006a4227966b7b39a637426084c8dc6901f7"}, +] +"zope.event" = [ + {file = "zope.event-4.5.0-py2.py3-none-any.whl", hash = "sha256:2666401939cdaa5f4e0c08cf7f20c9b21423b95e88f4675b1443973bdb080c42"}, + {file = "zope.event-4.5.0.tar.gz", hash = "sha256:5e76517f5b9b119acf37ca8819781db6c16ea433f7e2062c4afc2b6fbedb1330"}, +] +"zope.interface" = [ + {file = "zope.interface-5.4.0-cp27-cp27m-macosx_10_14_x86_64.whl", hash = "sha256:7df1e1c05304f26faa49fa752a8c690126cf98b40b91d54e6e9cc3b7d6ffe8b7"}, + {file = "zope.interface-5.4.0-cp27-cp27m-manylinux1_i686.whl", hash = "sha256:2c98384b254b37ce50eddd55db8d381a5c53b4c10ee66e1e7fe749824f894021"}, + {file = "zope.interface-5.4.0-cp27-cp27m-manylinux1_x86_64.whl", hash = "sha256:08f9636e99a9d5410181ba0729e0408d3d8748026ea938f3b970a0249daa8192"}, + {file = "zope.interface-5.4.0-cp27-cp27m-manylinux2010_i686.whl", hash = "sha256:0ea1d73b7c9dcbc5080bb8aaffb776f1c68e807767069b9ccdd06f27a161914a"}, + {file = "zope.interface-5.4.0-cp27-cp27m-manylinux2010_x86_64.whl", hash = "sha256:273f158fabc5ea33cbc936da0ab3d4ba80ede5351babc4f577d768e057651531"}, + {file = "zope.interface-5.4.0-cp27-cp27m-win32.whl", hash = "sha256:a1e6e96217a0f72e2b8629e271e1b280c6fa3fe6e59fa8f6701bec14e3354325"}, + {file = "zope.interface-5.4.0-cp27-cp27m-win_amd64.whl", hash = "sha256:877473e675fdcc113c138813a5dd440da0769a2d81f4d86614e5d62b69497155"}, + {file = "zope.interface-5.4.0-cp27-cp27mu-manylinux1_i686.whl", hash = "sha256:f7ee479e96f7ee350db1cf24afa5685a5899e2b34992fb99e1f7c1b0b758d263"}, + {file = "zope.interface-5.4.0-cp27-cp27mu-manylinux1_x86_64.whl", hash = "sha256:b0297b1e05fd128d26cc2460c810d42e205d16d76799526dfa8c8ccd50e74959"}, + {file = "zope.interface-5.4.0-cp27-cp27mu-manylinux2010_i686.whl", hash = "sha256:af310ec8335016b5e52cae60cda4a4f2a60a788cbb949a4fbea13d441aa5a09e"}, + {file = "zope.interface-5.4.0-cp27-cp27mu-manylinux2010_x86_64.whl", hash = "sha256:9a9845c4c6bb56e508651f005c4aeb0404e518c6f000d5a1123ab077ab769f5c"}, + {file = "zope.interface-5.4.0-cp35-cp35m-manylinux1_i686.whl", hash = "sha256:0b465ae0962d49c68aa9733ba92a001b2a0933c317780435f00be7ecb959c702"}, + {file = "zope.interface-5.4.0-cp35-cp35m-manylinux1_x86_64.whl", hash = "sha256:5dd9ca406499444f4c8299f803d4a14edf7890ecc595c8b1c7115c2342cadc5f"}, + {file = "zope.interface-5.4.0-cp35-cp35m-manylinux2010_i686.whl", hash = "sha256:469e2407e0fe9880ac690a3666f03eb4c3c444411a5a5fddfdabc5d184a79f05"}, + {file = "zope.interface-5.4.0-cp35-cp35m-manylinux2010_x86_64.whl", hash = "sha256:52de7fc6c21b419078008f697fd4103dbc763288b1406b4562554bd47514c004"}, + {file = "zope.interface-5.4.0-cp35-cp35m-manylinux2014_aarch64.whl", hash = "sha256:3dd4952748521205697bc2802e4afac5ed4b02909bb799ba1fe239f77fd4e117"}, + {file = "zope.interface-5.4.0-cp35-cp35m-win32.whl", hash = "sha256:dd93ea5c0c7f3e25335ab7d22a507b1dc43976e1345508f845efc573d3d779d8"}, + {file = "zope.interface-5.4.0-cp35-cp35m-win_amd64.whl", hash = "sha256:3748fac0d0f6a304e674955ab1365d515993b3a0a865e16a11ec9d86fb307f63"}, + {file = "zope.interface-5.4.0-cp36-cp36m-macosx_10_14_x86_64.whl", hash = "sha256:66c0061c91b3b9cf542131148ef7ecbecb2690d48d1612ec386de9d36766058f"}, + {file = "zope.interface-5.4.0-cp36-cp36m-manylinux1_i686.whl", hash = "sha256:d0c1bc2fa9a7285719e5678584f6b92572a5b639d0e471bb8d4b650a1a910920"}, + {file = "zope.interface-5.4.0-cp36-cp36m-manylinux1_x86_64.whl", hash = "sha256:2876246527c91e101184f63ccd1d716ec9c46519cc5f3d5375a3351c46467c46"}, + {file = "zope.interface-5.4.0-cp36-cp36m-manylinux2010_i686.whl", hash = "sha256:334701327f37c47fa628fc8b8d28c7d7730ce7daaf4bda1efb741679c2b087fc"}, + {file = "zope.interface-5.4.0-cp36-cp36m-manylinux2010_x86_64.whl", hash = "sha256:71aace0c42d53abe6fc7f726c5d3b60d90f3c5c055a447950ad6ea9cec2e37d9"}, + {file = "zope.interface-5.4.0-cp36-cp36m-manylinux2014_aarch64.whl", hash = "sha256:5bb3489b4558e49ad2c5118137cfeaf59434f9737fa9c5deefc72d22c23822e2"}, + {file = "zope.interface-5.4.0-cp36-cp36m-win32.whl", hash = "sha256:1c0e316c9add0db48a5b703833881351444398b04111188069a26a61cfb4df78"}, + {file = "zope.interface-5.4.0-cp36-cp36m-win_amd64.whl", hash = "sha256:6f0c02cbb9691b7c91d5009108f975f8ffeab5dff8f26d62e21c493060eff2a1"}, + {file = "zope.interface-5.4.0-cp37-cp37m-macosx_10_14_x86_64.whl", hash = "sha256:7d97a4306898b05404a0dcdc32d9709b7d8832c0c542b861d9a826301719794e"}, + {file = "zope.interface-5.4.0-cp37-cp37m-manylinux1_i686.whl", hash = "sha256:867a5ad16892bf20e6c4ea2aab1971f45645ff3102ad29bd84c86027fa99997b"}, + {file = "zope.interface-5.4.0-cp37-cp37m-manylinux1_x86_64.whl", hash = "sha256:5f931a1c21dfa7a9c573ec1f50a31135ccce84e32507c54e1ea404894c5eb96f"}, + {file = "zope.interface-5.4.0-cp37-cp37m-manylinux2010_i686.whl", hash = "sha256:194d0bcb1374ac3e1e023961610dc8f2c78a0f5f634d0c737691e215569e640d"}, + {file = "zope.interface-5.4.0-cp37-cp37m-manylinux2010_x86_64.whl", hash = "sha256:8270252effc60b9642b423189a2fe90eb6b59e87cbee54549db3f5562ff8d1b8"}, + {file = "zope.interface-5.4.0-cp37-cp37m-manylinux2014_aarch64.whl", hash = "sha256:15e7d1f7a6ee16572e21e3576d2012b2778cbacf75eb4b7400be37455f5ca8bf"}, + {file = "zope.interface-5.4.0-cp37-cp37m-win32.whl", hash = "sha256:8892f89999ffd992208754851e5a052f6b5db70a1e3f7d54b17c5211e37a98c7"}, + {file = "zope.interface-5.4.0-cp37-cp37m-win_amd64.whl", hash = "sha256:2e5a26f16503be6c826abca904e45f1a44ff275fdb7e9d1b75c10671c26f8b94"}, + {file = "zope.interface-5.4.0-cp38-cp38-macosx_10_14_x86_64.whl", hash = "sha256:0f91b5b948686659a8e28b728ff5e74b1be6bf40cb04704453617e5f1e945ef3"}, + {file = "zope.interface-5.4.0-cp38-cp38-manylinux1_i686.whl", hash = "sha256:4de4bc9b6d35c5af65b454d3e9bc98c50eb3960d5a3762c9438df57427134b8e"}, + {file = "zope.interface-5.4.0-cp38-cp38-manylinux1_x86_64.whl", hash = "sha256:bf68f4b2b6683e52bec69273562df15af352e5ed25d1b6641e7efddc5951d1a7"}, + {file = "zope.interface-5.4.0-cp38-cp38-manylinux2010_i686.whl", hash = "sha256:63b82bb63de7c821428d513607e84c6d97d58afd1fe2eb645030bdc185440120"}, + {file = "zope.interface-5.4.0-cp38-cp38-manylinux2010_x86_64.whl", hash = "sha256:db1fa631737dab9fa0b37f3979d8d2631e348c3b4e8325d6873c2541d0ae5a48"}, + {file = "zope.interface-5.4.0-cp38-cp38-manylinux2014_aarch64.whl", hash = "sha256:f44e517131a98f7a76696a7b21b164bcb85291cee106a23beccce454e1f433a4"}, + {file = "zope.interface-5.4.0-cp38-cp38-win32.whl", hash = "sha256:a9506a7e80bcf6eacfff7f804c0ad5350c8c95b9010e4356a4b36f5322f09abb"}, + {file = "zope.interface-5.4.0-cp38-cp38-win_amd64.whl", hash = "sha256:3c02411a3b62668200910090a0dff17c0b25aaa36145082a5a6adf08fa281e54"}, + {file = "zope.interface-5.4.0-cp39-cp39-macosx_10_14_x86_64.whl", hash = "sha256:0cee5187b60ed26d56eb2960136288ce91bcf61e2a9405660d271d1f122a69a4"}, + {file = "zope.interface-5.4.0-cp39-cp39-manylinux1_i686.whl", hash = "sha256:a8156e6a7f5e2a0ff0c5b21d6bcb45145efece1909efcbbbf48c56f8da68221d"}, + {file = "zope.interface-5.4.0-cp39-cp39-manylinux1_x86_64.whl", hash = "sha256:205e40ccde0f37496904572035deea747390a8b7dc65146d30b96e2dd1359a83"}, + {file = "zope.interface-5.4.0-cp39-cp39-manylinux2010_i686.whl", hash = "sha256:3f24df7124c323fceb53ff6168da70dbfbae1442b4f3da439cd441681f54fe25"}, + {file = "zope.interface-5.4.0-cp39-cp39-manylinux2010_x86_64.whl", hash = "sha256:5208ebd5152e040640518a77827bdfcc73773a15a33d6644015b763b9c9febc1"}, + {file = "zope.interface-5.4.0-cp39-cp39-manylinux2014_aarch64.whl", hash = "sha256:17776ecd3a1fdd2b2cd5373e5ef8b307162f581c693575ec62e7c5399d80794c"}, + {file = "zope.interface-5.4.0-cp39-cp39-win32.whl", hash = "sha256:d4d9d6c1a455d4babd320203b918ccc7fcbefe308615c521062bc2ba1aa4d26e"}, + {file = "zope.interface-5.4.0-cp39-cp39-win_amd64.whl", hash = "sha256:0cba8477e300d64a11a9789ed40ee8932b59f9ee05f85276dbb4b59acee5dd09"}, + {file = "zope.interface-5.4.0.tar.gz", hash = "sha256:5dba5f530fec3f0988d83b78cc591b58c0b6eb8431a85edd1569a0539a8a5a0e"}, ] diff --git a/pyproject.toml b/pyproject.toml index 0343e3c..3f39fdf 100644 --- a/pyproject.toml +++ b/pyproject.toml @@ -6,10 +6,13 @@ authors = ["Mark Bonicillo <[email protected]>"] [tool.poetry.dependencies] python = "^3.8" -openleadr = "^0.5.23" black = "^21.5b2" mypy = "^0.812" pre-commit = "^2.13.0" +volttron-client = "^0.3.4" + +# local dependencies +openleadr= {path="./openleadr", develop=true} [tool.poetry.dev-dependencies] pytest = "^6.2.4" @@ -17,3 +20,19 @@ pytest = "^6.2.4" [build-system] requires = ["poetry-core>=1.0.0"] build-backend = "poetry.core.masonry.api" + +[tool.black] +line-length = 79 +include = '\.pyi?$' +exclude = ''' +/( + \.git + | \.hg + | \.mypy_cache + | \.tox + | \.venv + | _build + | buck-out + | build + | dist +)/''' diff --git a/volttron_client-0.1.2-py3-none-any.whl b/volttron_client-0.1.2-py3-none-any.whl deleted file mode 100644 index 870e627..0000000 Binary files a/volttron_client-0.1.2-py3-none-any.whl and /dev/null differ diff --git a/volttron_openadr_ven/agent.py b/volttron_openadr_ven/agent.py index c79e9de..ef2c3e2 100644 --- a/volttron_openadr_ven/agent.py +++ b/volttron_openadr_ven/agent.py @@ -42,22 +42,38 @@ import openleadr.enums from pprint import pformat -from typing import Dict from datetime import datetime -from openleadr.enums import OPT -from openleadr.objects import Event +from openleadr.enums import OPT, REPORT_NAME, MEASUREMENTS from openleadr import OpenADRClient -from volttron.platform.agent import utils -from volttron.platform.vip.agent import Agent, Core -from volttron.platform.messaging import topics, headers -from volttron.utils import jsonapi - -utils.setup_logging() +from volttron.utils import ( + get_aware_utc_now, + jsonapi, + setup_logging, + vip_main, + format_timestamp, + load_config, +) +from volttron.client.vip.agent import Agent, Core +from volttron.client.messaging import topics, headers + +setup_logging(level=logging.DEBUG) _log = logging.getLogger(__name__) __version__ = "1.0" +VEN_NAME = "ven_name" +VTN_URL = "vtn_url" +DEBUG = "debug" +CERT = "cert" +KEY = "key" +PASSPHRASE = "passphrase" +VTN_FINGERPRINT = "vtn_fingerprint" +SHOW_FINGERPRINT = "show_fingerprint" +CA_FILE = "ca_file" +VEN_ID = "ven_id" +DISABLE_SIGNATURE = "disable_signature" + class OpenADRVenAgent(Agent): """ @@ -72,106 +88,116 @@ def __init__( self, ven_name: str, vtn_url: str, - debug: bool, - cert: str, - key: str, - fingerprint: str, - show_fingerprint: str, - ca_file: str, + debug: bool = False, + cert: str = None, + key: str = None, + passphrase: str = None, + vtn_fingerprint: str = None, + show_fingerprint: str = None, + ca_file: str = None, + ven_id: str = None, + disable_signature: bool = None, **kwargs, ) -> None: + """ + Initialize the agent's configuration. Create an OpenADR Client using OpenLeadr. + Configuration parameters (see config for a sample config file): + + str ven_name: The name for this VEN + str vtn_url: The URL of the VTN (Server) to connect to + bool debug: Whether or not to print debugging messages + str cert: The path to a PEM-formatted Certificate file to use + for signing messages. + str key: The path to a PEM-formatted Private Key file to use + for signing messages. + str passphrase: The passphrase for the Private Key + str vtn_fingerprint: The fingerprint for the VTN's certificate to + verify incomnig messages + str show_fingerprint: Whether to print your own fingerprint + on startup. Defaults to True. + str ca_file: The path to the PEM-formatted CA file for validating the VTN server's + certificate. + str ven_id: The ID for this VEN. If you leave this blank, a VEN_ID will be assigned by the VTN. + bool disable_signature: Whether to disable digital signatures + """ super(OpenADRVenAgent, self).__init__(enable_web=True, **kwargs) - self.ven_name = None - self.vtn_url = None - self.debug = None - self.cert = None - self.key = None - self.fingerprint = None - self.show_fingerprint = None - self.ca_file = None - self.ven_id = None + # client will be initialized in configure_agent() + self.ven_client: OpenADRClient self.default_config = { - "ven_name": ven_name, - "vtn_url": vtn_url, - "debug": debug, - "cert": cert, - "key": key, - "fingerprint": fingerprint, - "show_fingerprint": show_fingerprint, - "ca_file": ca_file, + VEN_NAME: ven_name, + VTN_URL: vtn_url, + DEBUG: debug, + CERT: cert, + KEY: key, + PASSPHRASE: passphrase, + VTN_FINGERPRINT: vtn_fingerprint, + SHOW_FINGERPRINT: show_fingerprint, + CA_FILE: ca_file, + VEN_ID: ven_id, + DISABLE_SIGNATURE: disable_signature, } + + # SubSystem/ConfigStore self.vip.config.set_default("config", self.default_config) self.vip.config.subscribe( self._configure, actions=["NEW", "UPDATE"], pattern="config" ) - # properties of the this class will be initialized in initialize_config - # self.ven_client is an instance of OpenLeadr's OpenADRClient, an implementation of OpenADR; - # OpenADRClient will be instantiated in initialize_config - self.initialize_config(self.default_config) + # initialize all attributes of this class + self.configure_agent(self.default_config) - def initialize_config(self, config: Dict) -> None: + def configure_agent(self, config: dict) -> None: """ Initialize the agent's configuration. Create an OpenADR Client using OpenLeadr. - Configuration parameters (see config for a sample config file): - - str ven_name: The name for this VEN - str vtn_url: The URL of the VTN (Server) to connect to - bool debug: Whether or not to print debugging messages - str cert: The path to a PEM-formatted Certificate file to use - for signing messages. - str key: The path to a PEM-formatted Private Key file to use - for signing messages. - str fingerprint: The fingerprint for the VTN's certificate to - verify incomnig messages - str show_fingerprint: Whether to print your own fingerprint - on startup. Defaults to True. - str ca_file: The path to the PEM-formatted CA file for validating the VTN server's - certificate. """ - _log.debug("Configuring agent") - self.ven_name = config.get("ven_name") - self.vtn_url = config.get("vtn_url") - self.debug = config.get("debug", False) - self.cert = config.get("cert") - self.key = config.get("key") - self.fingerprint = config.get("fingerprint") - self.show_fingerprint = config.get("show_fingerprint", True) - self.ca_file = config.get("ca_file") - - _log.info("Configuration parameters:") - _log.info(pformat(config)) + _log.debug(f"Configuring agent with: \n {pformat(config)} ") # instantiate and add handlers to the OpenADR Client + _log.info("Creating OpenLeadrVen Client...") self.ven_client = OpenADRClient( - ven_name=self.ven_name, - vtn_url=self.vtn_url, - debug=self.debug, - cert=self.cert, - key=self.key, - show_fingerprint=self.show_fingerprint, - ca_file=self.ca_file, + ven_name=config.get(VEN_NAME), + vtn_url=config.get(VTN_URL), + debug=config.get(DEBUG), + cert=config.get(CERT), + key=config.get(KEY), + passphrase=config.get(PASSPHRASE), + vtn_fingerprint=config.get(VTN_FINGERPRINT), + show_fingerprint=config.get(SHOW_FINGERPRINT), + ca_file=config.get(CA_FILE), + ven_id=config.get(VEN_ID), + disable_signature=config.get(DISABLE_SIGNATURE), ) - # if you want to add more handlers on a specific event, you must create a coroutine - # in this class and then add it as the second input for 'self.ven_client.add_handler(<some event>, <coroutine>)' + # Add event handling capability to the client + # if you want to add more handlers on a specific event, you must create a coroutine in this class + # and then add it as the second input for 'self.ven_client.add_handler(<some event>, <coroutine>)' self.ven_client.add_handler("on_event", self.handle_event) - def _configure(self, contents: Dict) -> None: + # Add the report capability to the client + # the following is a report to be sent to a IPKeys test VTN + self.ven_client.add_report( + callback=self.collect_report_value, + report_name=REPORT_NAME.TELEMETRY_USAGE, + resource_id="device001", + measurement=MEASUREMENTS.VOLTAGE, + ) + + _log.info("Configuration complete.") + + def _configure(self, config_name, action, contents: dict) -> None: """The agent's config may have changed. Re-initialize it.""" config = self.default_config.copy() config.update(contents) - self.initialize_config(config) + self.configure_agent(config) # ***************** Methods for Managing the Agent on Volttron ******************** @Core.receiver("onstart") - def onstart(self) -> None: + def onstart(self, sender) -> None: """The agent has started.""" - _log.debug("Starting agent") - - # Running the ven_client in the Python AsyncIO Event Loop + _log.info(f"Sender {sender}") + _log.info("Starting agent...") loop = asyncio.get_event_loop() loop.create_task(self.ven_client.run()) loop.run_forever() @@ -197,13 +223,12 @@ def onstart(self) -> None: # ***************** Methods for Servicing VTN Requests ******************** - async def handle_event(self, event: openleadr.objects.Event) -> openleadr.enums.OPT: + async def handle_event(self, event: dict) -> openleadr.enums.OPT: """ Publish event to the Volttron message bus. Return OPT response. This coroutine will be called when there is an event to be handled. - :param event: Event - For more info on shape of 'event', see https://openleadr.org/docs/client.html#dealing-with-events and https://openleadr.org/docs/api/openleadr.html#openleadr.objects.Event + :param event: dict Example of event: { 'event_id': '123786-129831', @@ -225,51 +250,63 @@ async def handle_event(self, event: openleadr.objects.Event) -> openleadr.enums. } :return: openleadr.enums.OPT """ - _log.info(f"Event: {pformat(event)}") + try: + _log.debug("Received event. Processing event now...") + signal = event.get("event_signals")[0] # type: ignore[index] + _log.info(f"Event signal:\n {pformat(signal)}") + except IndexError as e: + _log.debug( + f"Event signals is empty. {e} \n Showing whole event: {pformat(event)}" + ) + pass + self.publish_event(event) + return OPT.OPT_IN + async def collect_report_value(self): + # This callback is called when you need to collect a value for your Report + # below is dummy code; replace with your business logic + return 1.23 + # ***************** VOLTTRON Pub/Sub Requests ******************** - def publish_event(self, event: openleadr.objects.Event) -> None: + def publish_event(self, event: dict) -> None: """Publish an event to the Volttron message bus. When an event is created/updated, it is published to the VOLTTRON bus with a topic that includes 'openadr/event_update'. - :param event: + :param event: dict :return: None """ # OADR rule 6: If testEvent is present and != "false", handle the event as a test event. - if event.event_descriptor.test_event != "false": - _log.debug(f"Suppressing publication of test event {event}") - return - - _log.debug("Publishing event {}".format(event)) + try: + if event["event_descriptor"]["test_event"]: + _log.debug("Suppressing publication of test event") + return + except KeyError as e: + _log.debug(f"Key error: {e}") + pass + + _log.debug(f"Publishing real/non-test event \n {pformat(event)}") self.vip.pubsub.publish( peer="pubsub", - topic=f"{topics.OPENADR_EVENT}/{self.ven_id}", - headers={ - headers.TIMESTAMP: utils.format_timestamp(utils.get_aware_utc_now()) - }, - message=self._json_object(event.__dict__), + topic=f"{topics.OPENADR_EVENT}/{self.ven_client.ven_name}", + headers={headers.TIMESTAMP: format_timestamp(get_aware_utc_now())}, + message=self._json_object(event), ) + return # ***************** Helper methods ******************** - def _json_object(self, obj: Dict): + def _json_object(self, obj: dict): """Ensure that an object is valid JSON by dumping it with json_converter and then reloading it.""" obj_string = jsonapi.dumps( - obj, default=lambda x: x.__str__() if isinstance(x, datetime) else None + obj, + default=lambda x: x.__str__() if isinstance(x, datetime) else None, ) return jsonapi.loads(obj_string) def main(): """Main method called to start the agent.""" - utils.vip_main(ven_agent, version=__version__) - - -if __name__ == "__main__": - try: - sys.exit(main()) - except KeyboardInterrupt: - pass + vip_main(ven_agent, version=__version__) def ven_agent(config_path: str, **kwargs) -> OpenADRVenAgent: @@ -280,10 +317,13 @@ def ven_agent(config_path: str, **kwargs) -> OpenADRVenAgent: :param config_path: (str) Path to a configuration file. - :returns: OpenADRVenAgent instance + :returns: OpenADRVenAgent """ try: - config = utils.load_config(config_path) + config = load_config(config_path) + except NameError as err: + _log.exception(err) + raise except Exception as err: _log.error("Error loading configuration: {}".format(err)) config = {} @@ -291,31 +331,42 @@ def ven_agent(config_path: str, **kwargs) -> OpenADRVenAgent: if not config: raise Exception("Configuration cannot be empty.") - ven_name = config.get("ven_name") + ven_name = config.get(VEN_NAME) if not ven_name: - raise Exception("ven_name is required.") - vtn_url = config.get("vtn_url") + raise Exception(f"{VEN_NAME} is required.") + vtn_url = config.get(VTN_URL) if not vtn_url: raise Exception( - "vtn_url is required. Ensure vtn_url is given a URL to the VTN." + f"{VTN_URL} is required. Ensure {VTN_URL} is given a URL to the VTN." ) - - # below are optional configurations - debug = config.get("debug") - cert = config.get("cert") - key = config.get("key") - fingerprint = config.get("fingerprint") - show_fingerprint = config.get("show_fingerprint") - ca_file = config.get("ca_file") + debug = config.get(DEBUG) + cert = config.get(CERT) + key = config.get(KEY) + passphrase = config.get(PASSPHRASE) + vtn_fingerprint = config.get(VTN_FINGERPRINT) + show_fingerprint = config.get(SHOW_FINGERPRINT) + ca_file = config.get(CA_FILE) + ven_id = config.get(VEN_ID) + disable_signature = bool(config.get(DISABLE_SIGNATURE)) return OpenADRVenAgent( ven_name, vtn_url, - debug, - cert, - key, - fingerprint, - show_fingerprint, - ca_file, + debug=debug, + cert=cert, + key=key, + passphrase=passphrase, + vtn_fingerprint=vtn_fingerprint, + show_fingerprint=show_fingerprint, + ca_file=ca_file, + ven_id=ven_id, + disable_signature=disable_signature, **kwargs, ) + + +if __name__ == "__main__": + try: + sys.exit(main()) + except KeyboardInterrupt: + pass diff --git a/volttron_openadr_ven/config.json b/volttron_openadr_ven/config.json new file mode 100644 index 0000000..acba83b --- /dev/null +++ b/volttron_openadr_ven/config.json @@ -0,0 +1,8 @@ +{ + "ven_name": "PNNLVEN", + "vtn_url": "https://eiss2demo.ipkeys.com/oadr2/OpenADR2/Simple/2.0b", + "cert": "secret/", + "key": "secret/", + "debug": true, + "disable_signature": true +} diff --git a/volttron_utils-0.1.1-py3-none-any.whl b/volttron_utils-0.1.1-py3-none-any.whl deleted file mode 100644 index 6512925..0000000 Binary files a/volttron_utils-0.1.1-py3-none-any.whl and /dev/null differ

Work with openleadr in a script with certificates Verify working with openleadr allows the security and usage we are going to need. https://pypi.org/project/openleadr/

2021-10-21T21:44:33

0.0

[]

eclipse-volttron/volttron-openadr-ven

eclipse-volttron__volttron-openadr-ven-8

a3f3883b84c59f609ea9d5f81f17e3343f678f2c

diff --git a/.gitignore b/.gitignore index b6e4761..b97131e 100644 --- a/.gitignore +++ b/.gitignore @@ -127,3 +127,5 @@ dmypy.json # Pyre type checker .pyre/ + +.idea/ diff --git a/.pre-commit-config.yaml b/.pre-commit-config.yaml new file mode 100644 index 0000000..218ffbf --- /dev/null +++ b/.pre-commit-config.yaml @@ -0,0 +1,15 @@ +repos: +- repo: https://github.com/pre-commit/pre-commit-hooks + rev: v2.3.0 + hooks: + - id: check-yaml + - id: end-of-file-fixer + - id: trailing-whitespace +- repo: https://github.com/psf/black + rev: 19.3b0 + hooks: + - id: black +- repo: https://github.com/pre-commit/mirrors-mypy + rev: v0.812 + hooks: + - id: mypy diff --git a/README.md b/README.md index d59628d..0191cf6 100644 --- a/README.md +++ b/README.md @@ -1,1 +1,95 @@ -# volttron-openadr-ven \ No newline at end of file +# Prerequisites + +* Python >=3.7 +* Pip >=20.1 +* Poetry >=1.16 + +# Development + +## Environment setup + +Note: This repo uses [Poetry](https://python-poetry.org/), a dependency management and packaging tool for Python. If you don't have Poetry installed on your machine, follow [these steps](https://python-poetry.org/docs/#installation) to install it on your machine. +To check if Poetry is installed, run ```poetry --version```. If your receive 'command not found: poetry' error, add the following line to your '~/.bashrc' script: ```export PATH=$PATH:$HOME/.poetry/bin```. + +To setup your development environment, do the following: + +1. Create virtual environment + +Run the following command on the CLI. By default, poetry creates a virtual environment in {cache-dir}/virtualenvs +({cache-dir}\virtualenvs on Windows). For more info on creating virtual environments through Poetry, see [this](https://python-poetry.org/docs/basic-usage/#using-your-virtual-environment). + +```shell +$ poetry shell +``` + +NOTE: Poetry needs the command 'python' in order to create a new shell. On Linux, if you only have python3 installed, run the following to link 'python' to 'python3': +```shell +$ sudo apt install python-is-python3 +``` + +2. Install requirements + +```shell +(env) $ poetry install +``` + + +3. Install Volttron-specific dependencies for this agent. + +```shell +(env) $ pip install volttron_utils-0.1.1-py3-none-any.whl volttron_client-0.1.2-py3-none-any.whl +``` + + +## Code Quality and Standardization + +## Git Hooks and Pre-commit + +This repo provides pre-commit hooks to enforce code quality and standardization before pushing any commit to the repo. To use this tool locally, +run the following commands: + +```shell +(env) $ pre-commit install + +# To check that pre-commit scripts are installed and properly run, run pre-commit on all the files +(env) $ pre-commit run --all +``` + +With pre-commit setup, every time you run "git commit", git will check your commit against a set of code linters. For details, see .pre-commit-config.yaml + +## Black + +This repo uses [Black](https://pypi.org/project/black/), an opinionated Python code formatter. To use Black to check and automagically reformat your code, +run the following command on your chosen Python file: + +```shell +(env) $ black <path to file> +``` + +## MyPy + +This repo uses [MyPy](https://mypy.readthedocs.io/en/stable/), a static type checker for Python. To use MyPy on a Python file, run the following command: +```shell +(env) $ mypy <path to file> +``` + +# CI/CD + +TODO: + +# Building Wheel and Publishing to PyPi + +TODO: + +Build wheel +```python setup.py -q sdist bdist_wheel``` + +Check wheel +```check-wheel-contents dist/*.whl``` +```tar tzf <>``` + +Test uploading +```twine upload --repository-url https://test.pypi.org/legacy/ dist/*``` + +Upload to PyPi +```python -m twin upload dist/*``` diff --git a/mypy.ini b/mypy.ini new file mode 100644 index 0000000..03aefad --- /dev/null +++ b/mypy.ini @@ -0,0 +1,4 @@ +[mypy] +show_error_context = True +pretty = True +show_column_numbers = True diff --git a/poetry.lock b/poetry.lock new file mode 100644 index 0000000..26f5d1c --- /dev/null +++ b/poetry.lock @@ -0,0 +1,1116 @@ +[[package]] +name = "aiohttp" +version = "3.7.4.post0" +description = "Async http client/server framework (asyncio)" +category = "main" +optional = false +python-versions = ">=3.6" + +[package.dependencies] +async-timeout = ">=3.0,<4.0" +attrs = ">=17.3.0" +chardet = ">=2.0,<5.0" +multidict = ">=4.5,<7.0" +typing-extensions = ">=3.6.5" +yarl = ">=1.0,<2.0" + +[package.extras] +speedups = ["aiodns", "brotlipy", "cchardet"] + +[[package]] +name = "appdirs" +version = "1.4.4" +description = "A small Python module for determining appropriate platform-specific dirs, e.g. a \"user data dir\"." +category = "main" +optional = false +python-versions = "*" + +[[package]] +name = "apscheduler" +version = "3.7.0" +description = "In-process task scheduler with Cron-like capabilities" +category = "main" +optional = false +python-versions = ">=2.7, !=3.0.*, !=3.1.*, !=3.2.*, !=3.3.*, !=3.4.*, <4" + +[package.dependencies] +pytz = "*" +six = ">=1.4.0" +tzlocal = ">=2.0,<3.0" + +[package.extras] +asyncio = ["trollius"] +doc = ["sphinx", "sphinx-rtd-theme"] +gevent = ["gevent"] +mongodb = ["pymongo (>=3.0)"] +redis = ["redis (>=3.0)"] +rethinkdb = ["rethinkdb (>=2.4.0)"] +sqlalchemy = ["sqlalchemy (>=0.8)"] +testing = ["pytest (<6)", "pytest-cov", "pytest-tornado5", "mock", "pytest-asyncio (<0.6)", "pytest-asyncio"] +tornado = ["tornado (>=4.3)"] +twisted = ["twisted"] +zookeeper = ["kazoo"] + +[[package]] +name = "async-timeout" +version = "3.0.1" +description = "Timeout context manager for asyncio programs" +category = "main" +optional = false +python-versions = ">=3.5.3" + +[[package]] +name = "atomicwrites" +version = "1.4.0" +description = "Atomic file writes." +category = "dev" +optional = false +python-versions = ">=2.7, !=3.0.*, !=3.1.*, !=3.2.*, !=3.3.*" + +[[package]] +name = "attrs" +version = "21.2.0" +description = "Classes Without Boilerplate" +category = "main" +optional = false +python-versions = ">=2.7, !=3.0.*, !=3.1.*, !=3.2.*, !=3.3.*, !=3.4.*" + +[package.extras] +dev = ["coverage[toml] (>=5.0.2)", "hypothesis", "pympler", "pytest (>=4.3.0)", "six", "mypy", "pytest-mypy-plugins", "zope.interface", "furo", "sphinx", "sphinx-notfound-page", "pre-commit"] +docs = ["furo", "sphinx", "zope.interface", "sphinx-notfound-page"] +tests = ["coverage[toml] (>=5.0.2)", "hypothesis", "pympler", "pytest (>=4.3.0)", "six", "mypy", "pytest-mypy-plugins", "zope.interface"] +tests_no_zope = ["coverage[toml] (>=5.0.2)", "hypothesis", "pympler", "pytest (>=4.3.0)", "six", "mypy", "pytest-mypy-plugins"] + +[[package]] +name = "black" +version = "21.5b2" +description = "The uncompromising code formatter." +category = "main" +optional = false +python-versions = ">=3.6.2" + +[package.dependencies] +appdirs = "*" +click = ">=7.1.2" +mypy-extensions = ">=0.4.3" +pathspec = ">=0.8.1,<1" +regex = ">=2020.1.8" +toml = ">=0.10.1" + +[package.extras] +colorama = ["colorama (>=0.4.3)"] +d = ["aiohttp (>=3.6.0)", "aiohttp-cors (>=0.4.0)"] +python2 = ["typed-ast (>=1.4.2)"] +uvloop = ["uvloop (>=0.15.2)"] + +[[package]] +name = "certifi" +version = "2021.5.30" +description = "Python package for providing Mozilla's CA Bundle." +category = "main" +optional = false +python-versions = "*" + +[[package]] +name = "cffi" +version = "1.14.5" +description = "Foreign Function Interface for Python calling C code." +category = "main" +optional = false +python-versions = "*" + +[package.dependencies] +pycparser = "*" + +[[package]] +name = "cfgv" +version = "3.3.0" +description = "Validate configuration and produce human readable error messages." +category = "main" +optional = false +python-versions = ">=3.6.1" + +[[package]] +name = "chardet" +version = "4.0.0" +description = "Universal encoding detector for Python 2 and 3" +category = "main" +optional = false +python-versions = ">=2.7, !=3.0.*, !=3.1.*, !=3.2.*, !=3.3.*, !=3.4.*" + +[[package]] +name = "click" +version = "8.0.1" +description = "Composable command line interface toolkit" +category = "main" +optional = false +python-versions = ">=3.6" + +[package.dependencies] +colorama = {version = "*", markers = "platform_system == \"Windows\""} + +[[package]] +name = "colorama" +version = "0.4.4" +description = "Cross-platform colored terminal text." +category = "main" +optional = false +python-versions = ">=2.7, !=3.0.*, !=3.1.*, !=3.2.*, !=3.3.*, !=3.4.*" + +[[package]] +name = "cryptography" +version = "3.4.7" +description = "cryptography is a package which provides cryptographic recipes and primitives to Python developers." +category = "main" +optional = false +python-versions = ">=3.6" + +[package.dependencies] +cffi = ">=1.12" + +[package.extras] +docs = ["sphinx (>=1.6.5,!=1.8.0,!=3.1.0,!=3.1.1)", "sphinx-rtd-theme"] +docstest = ["doc8", "pyenchant (>=1.6.11)", "twine (>=1.12.0)", "sphinxcontrib-spelling (>=4.0.1)"] +pep8test = ["black", "flake8", "flake8-import-order", "pep8-naming"] +sdist = ["setuptools-rust (>=0.11.4)"] +ssh = ["bcrypt (>=3.1.5)"] +test = ["pytest (>=6.0)", "pytest-cov", "pytest-subtests", "pytest-xdist", "pretend", "iso8601", "pytz", "hypothesis (>=1.11.4,!=3.79.2)"] + +[[package]] +name = "distlib" +version = "0.3.2" +description = "Distribution utilities" +category = "main" +optional = false +python-versions = "*" + +[[package]] +name = "eight" +version = "1.0.1" +description = "Python 2 to the power of 3. A lightweight porting helper library." +category = "main" +optional = false +python-versions = "*" + +[package.dependencies] +future = ">=0.16,<0.19" + +[[package]] +name = "filelock" +version = "3.0.12" +description = "A platform independent file lock." +category = "main" +optional = false +python-versions = "*" + +[[package]] +name = "future" +version = "0.18.2" +description = "Clean single-source support for Python 3 and 2" +category = "main" +optional = false +python-versions = ">=2.6, !=3.0.*, !=3.1.*, !=3.2.*" + +[[package]] +name = "identify" +version = "2.2.7" +description = "File identification library for Python" +category = "main" +optional = false +python-versions = ">=3.6.1" + +[package.extras] +license = ["editdistance-s"] + +[[package]] +name = "idna" +version = "3.2" +description = "Internationalized Domain Names in Applications (IDNA)" +category = "main" +optional = false +python-versions = ">=3.5" + +[[package]] +name = "iniconfig" +version = "1.1.1" +description = "iniconfig: brain-dead simple config-ini parsing" +category = "dev" +optional = false +python-versions = "*" + +[[package]] +name = "jinja2" +version = "3.0.1" +description = "A very fast and expressive template engine." +category = "main" +optional = false +python-versions = ">=3.6" + +[package.dependencies] +MarkupSafe = ">=2.0" + +[package.extras] +i18n = ["Babel (>=2.7)"] + +[[package]] +name = "lxml" +version = "4.6.3" +description = "Powerful and Pythonic XML processing library combining libxml2/libxslt with the ElementTree API." +category = "main" +optional = false +python-versions = ">=2.7, !=3.0.*, !=3.1.*, !=3.2.*, !=3.3.*, != 3.4.*" + +[package.extras] +cssselect = ["cssselect (>=0.7)"] +html5 = ["html5lib"] +htmlsoup = ["beautifulsoup4"] +source = ["Cython (>=0.29.7)"] + +[[package]] +name = "markupsafe" +version = "2.0.1" +description = "Safely add untrusted strings to HTML/XML markup." +category = "main" +optional = false +python-versions = ">=3.6" + +[[package]] +name = "multidict" +version = "5.1.0" +description = "multidict implementation" +category = "main" +optional = false +python-versions = ">=3.6" + +[[package]] +name = "mypy" +version = "0.812" +description = "Optional static typing for Python" +category = "main" +optional = false +python-versions = ">=3.5" + +[package.dependencies] +mypy-extensions = ">=0.4.3,<0.5.0" +typed-ast = ">=1.4.0,<1.5.0" +typing-extensions = ">=3.7.4" + +[package.extras] +dmypy = ["psutil (>=4.0)"] + +[[package]] +name = "mypy-extensions" +version = "0.4.3" +description = "Experimental type system extensions for programs checked with the mypy typechecker." +category = "main" +optional = false +python-versions = "*" + +[[package]] +name = "nodeenv" +version = "1.6.0" +description = "Node.js virtual environment builder" +category = "main" +optional = false +python-versions = "*" + +[[package]] +name = "openleadr" +version = "0.5.23" +description = "Python3 library for building OpenADR Clients (VENs) and Servers (VTNs)" +category = "main" +optional = false +python-versions = ">=3.7.0" + +[package.dependencies] +aiohttp = "*" +apscheduler = "*" +jinja2 = "*" +signxml-openadr = "2.9.1" +xmltodict = "*" + +[[package]] +name = "packaging" +version = "20.9" +description = "Core utilities for Python packages" +category = "dev" +optional = false +python-versions = ">=2.7, !=3.0.*, !=3.1.*, !=3.2.*, !=3.3.*" + +[package.dependencies] +pyparsing = ">=2.0.2" + +[[package]] +name = "pathspec" +version = "0.8.1" +description = "Utility library for gitignore style pattern matching of file paths." +category = "main" +optional = false +python-versions = ">=2.7, !=3.0.*, !=3.1.*, !=3.2.*, !=3.3.*, !=3.4.*" + +[[package]] +name = "pluggy" +version = "0.13.1" +description = "plugin and hook calling mechanisms for python" +category = "dev" +optional = false +python-versions = ">=2.7, !=3.0.*, !=3.1.*, !=3.2.*, !=3.3.*" + +[package.extras] +dev = ["pre-commit", "tox"] + +[[package]] +name = "pre-commit" +version = "2.13.0" +description = "A framework for managing and maintaining multi-language pre-commit hooks." +category = "main" +optional = false +python-versions = ">=3.6.1" + +[package.dependencies] +cfgv = ">=2.0.0" +identify = ">=1.0.0" +nodeenv = ">=0.11.1" +pyyaml = ">=5.1" +toml = "*" +virtualenv = ">=20.0.8" + +[[package]] +name = "py" +version = "1.10.0" +description = "library with cross-python path, ini-parsing, io, code, log facilities" +category = "dev" +optional = false +python-versions = ">=2.7, !=3.0.*, !=3.1.*, !=3.2.*, !=3.3.*" + +[[package]] +name = "pycparser" +version = "2.20" +description = "C parser in Python" +category = "main" +optional = false +python-versions = ">=2.7, !=3.0.*, !=3.1.*, !=3.2.*, !=3.3.*" + +[[package]] +name = "pyopenssl" +version = "19.1.0" +description = "Python wrapper module around the OpenSSL library" +category = "main" +optional = false +python-versions = "*" + +[package.dependencies] +cryptography = ">=2.8" +six = ">=1.5.2" + +[package.extras] +docs = ["sphinx", "sphinx-rtd-theme"] +test = ["flaky", "pretend", "pytest (>=3.0.1)"] + +[[package]] +name = "pyparsing" +version = "2.4.7" +description = "Python parsing module" +category = "dev" +optional = false +python-versions = ">=2.6, !=3.0.*, !=3.1.*, !=3.2.*" + +[[package]] +name = "pytest" +version = "6.2.4" +description = "pytest: simple powerful testing with Python" +category = "dev" +optional = false +python-versions = ">=3.6" + +[package.dependencies] +atomicwrites = {version = ">=1.0", markers = "sys_platform == \"win32\""} +attrs = ">=19.2.0" +colorama = {version = "*", markers = "sys_platform == \"win32\""} +iniconfig = "*" +packaging = "*" +pluggy = ">=0.12,<1.0.0a1" +py = ">=1.8.2" +toml = "*" + +[package.extras] +testing = ["argcomplete", "hypothesis (>=3.56)", "mock", "nose", "requests", "xmlschema"] + +[[package]] +name = "pytz" +version = "2021.1" +description = "World timezone definitions, modern and historical" +category = "main" +optional = false +python-versions = "*" + +[[package]] +name = "pyyaml" +version = "5.4.1" +description = "YAML parser and emitter for Python" +category = "main" +optional = false +python-versions = ">=2.7, !=3.0.*, !=3.1.*, !=3.2.*, !=3.3.*, !=3.4.*, !=3.5.*" + +[[package]] +name = "regex" +version = "2021.4.4" +description = "Alternative regular expression module, to replace re." +category = "main" +optional = false +python-versions = "*" + +[[package]] +name = "signxml-openadr" +version = "2.9.1" +description = "Python XML Signature library" +category = "main" +optional = false +python-versions = "*" + +[package.dependencies] +certifi = ">=2018.1.18" +cryptography = ">=2.1.4,<4" +eight = ">=0.4.2,<2" +lxml = ">=4.2.1,<5" +pyOpenSSL = ">=17.5.0,<20" + +[[package]] +name = "six" +version = "1.16.0" +description = "Python 2 and 3 compatibility utilities" +category = "main" +optional = false +python-versions = ">=2.7, !=3.0.*, !=3.1.*, !=3.2.*" + +[[package]] +name = "toml" +version = "0.10.2" +description = "Python Library for Tom's Obvious, Minimal Language" +category = "main" +optional = false +python-versions = ">=2.6, !=3.0.*, !=3.1.*, !=3.2.*" + +[[package]] +name = "typed-ast" +version = "1.4.3" +description = "a fork of Python 2 and 3 ast modules with type comment support" +category = "main" +optional = false +python-versions = "*" + +[[package]] +name = "typing-extensions" +version = "3.10.0.0" +description = "Backported and Experimental Type Hints for Python 3.5+" +category = "main" +optional = false +python-versions = "*" + +[[package]] +name = "tzlocal" +version = "2.1" +description = "tzinfo object for the local timezone" +category = "main" +optional = false +python-versions = "*" + +[package.dependencies] +pytz = "*" + +[[package]] +name = "virtualenv" +version = "20.4.7" +description = "Virtual Python Environment builder" +category = "main" +optional = false +python-versions = "!=3.0.*,!=3.1.*,!=3.2.*,!=3.3.*,>=2.7" + +[package.dependencies] +appdirs = ">=1.4.3,<2" +distlib = ">=0.3.1,<1" +filelock = ">=3.0.0,<4" +six = ">=1.9.0,<2" + +[package.extras] +docs = ["proselint (>=0.10.2)", "sphinx (>=3)", "sphinx-argparse (>=0.2.5)", "sphinx-rtd-theme (>=0.4.3)", "towncrier (>=19.9.0rc1)"] +testing = ["coverage (>=4)", "coverage-enable-subprocess (>=1)", "flaky (>=3)", "pytest (>=4)", "pytest-env (>=0.6.2)", "pytest-freezegun (>=0.4.1)", "pytest-mock (>=2)", "pytest-randomly (>=1)", "pytest-timeout (>=1)", "packaging (>=20.0)", "xonsh (>=0.9.16)"] + +[[package]] +name = "xmltodict" +version = "0.12.0" +description = "Makes working with XML feel like you are working with JSON" +category = "main" +optional = false +python-versions = ">=2.7, !=3.0.*, !=3.1.*, !=3.2.*, !=3.3.*" + +[[package]] +name = "yarl" +version = "1.6.3" +description = "Yet another URL library" +category = "main" +optional = false +python-versions = ">=3.6" + +[package.dependencies] +idna = ">=2.0" +multidict = ">=4.0" + +[metadata] +lock-version = "1.1" +python-versions = "^3.8" +content-hash = "2a1dd13cfcde4f16340cb5425979909af99a26d9e7091cd2163a7e118ca1004b" + +[metadata.files] +aiohttp = [ + {file = "aiohttp-3.7.4.post0-cp36-cp36m-macosx_10_14_x86_64.whl", hash = "sha256:3cf75f7cdc2397ed4442594b935a11ed5569961333d49b7539ea741be2cc79d5"}, + {file = "aiohttp-3.7.4.post0-cp36-cp36m-manylinux1_i686.whl", hash = "sha256:4b302b45040890cea949ad092479e01ba25911a15e648429c7c5aae9650c67a8"}, + {file = "aiohttp-3.7.4.post0-cp36-cp36m-manylinux2014_aarch64.whl", hash = "sha256:fe60131d21b31fd1a14bd43e6bb88256f69dfc3188b3a89d736d6c71ed43ec95"}, + {file = "aiohttp-3.7.4.post0-cp36-cp36m-manylinux2014_i686.whl", hash = "sha256:393f389841e8f2dfc86f774ad22f00923fdee66d238af89b70ea314c4aefd290"}, + {file = "aiohttp-3.7.4.post0-cp36-cp36m-manylinux2014_ppc64le.whl", hash = "sha256:c6e9dcb4cb338d91a73f178d866d051efe7c62a7166653a91e7d9fb18274058f"}, + {file = "aiohttp-3.7.4.post0-cp36-cp36m-manylinux2014_s390x.whl", hash = "sha256:5df68496d19f849921f05f14f31bd6ef53ad4b00245da3195048c69934521809"}, + {file = "aiohttp-3.7.4.post0-cp36-cp36m-manylinux2014_x86_64.whl", hash = "sha256:0563c1b3826945eecd62186f3f5c7d31abb7391fedc893b7e2b26303b5a9f3fe"}, + {file = "aiohttp-3.7.4.post0-cp36-cp36m-win32.whl", hash = "sha256:3d78619672183be860b96ed96f533046ec97ca067fd46ac1f6a09cd9b7484287"}, + {file = "aiohttp-3.7.4.post0-cp36-cp36m-win_amd64.whl", hash = "sha256:f705e12750171c0ab4ef2a3c76b9a4024a62c4103e3a55dd6f99265b9bc6fcfc"}, + {file = "aiohttp-3.7.4.post0-cp37-cp37m-macosx_10_14_x86_64.whl", hash = "sha256:230a8f7e24298dea47659251abc0fd8b3c4e38a664c59d4b89cca7f6c09c9e87"}, + {file = "aiohttp-3.7.4.post0-cp37-cp37m-manylinux1_i686.whl", hash = "sha256:2e19413bf84934d651344783c9f5e22dee452e251cfd220ebadbed2d9931dbf0"}, + {file = "aiohttp-3.7.4.post0-cp37-cp37m-manylinux2014_aarch64.whl", hash = "sha256:e4b2b334e68b18ac9817d828ba44d8fcb391f6acb398bcc5062b14b2cbeac970"}, + {file = "aiohttp-3.7.4.post0-cp37-cp37m-manylinux2014_i686.whl", hash = "sha256:d012ad7911653a906425d8473a1465caa9f8dea7fcf07b6d870397b774ea7c0f"}, + {file = "aiohttp-3.7.4.post0-cp37-cp37m-manylinux2014_ppc64le.whl", hash = "sha256:40eced07f07a9e60e825554a31f923e8d3997cfc7fb31dbc1328c70826e04cde"}, + {file = "aiohttp-3.7.4.post0-cp37-cp37m-manylinux2014_s390x.whl", hash = "sha256:209b4a8ee987eccc91e2bd3ac36adee0e53a5970b8ac52c273f7f8fd4872c94c"}, + {file = "aiohttp-3.7.4.post0-cp37-cp37m-manylinux2014_x86_64.whl", hash = "sha256:14762875b22d0055f05d12abc7f7d61d5fd4fe4642ce1a249abdf8c700bf1fd8"}, + {file = "aiohttp-3.7.4.post0-cp37-cp37m-win32.whl", hash = "sha256:7615dab56bb07bff74bc865307aeb89a8bfd9941d2ef9d817b9436da3a0ea54f"}, + {file = "aiohttp-3.7.4.post0-cp37-cp37m-win_amd64.whl", hash = "sha256:d9e13b33afd39ddeb377eff2c1c4f00544e191e1d1dee5b6c51ddee8ea6f0cf5"}, + {file = "aiohttp-3.7.4.post0-cp38-cp38-macosx_10_14_x86_64.whl", hash = "sha256:547da6cacac20666422d4882cfcd51298d45f7ccb60a04ec27424d2f36ba3eaf"}, + {file = "aiohttp-3.7.4.post0-cp38-cp38-manylinux1_i686.whl", hash = "sha256:af9aa9ef5ba1fd5b8c948bb11f44891968ab30356d65fd0cc6707d989cd521df"}, + {file = "aiohttp-3.7.4.post0-cp38-cp38-manylinux2014_aarch64.whl", hash = "sha256:64322071e046020e8797117b3658b9c2f80e3267daec409b350b6a7a05041213"}, + {file = "aiohttp-3.7.4.post0-cp38-cp38-manylinux2014_i686.whl", hash = "sha256:bb437315738aa441251214dad17428cafda9cdc9729499f1d6001748e1d432f4"}, + {file = "aiohttp-3.7.4.post0-cp38-cp38-manylinux2014_ppc64le.whl", hash = "sha256:e54962802d4b8b18b6207d4a927032826af39395a3bd9196a5af43fc4e60b009"}, + {file = "aiohttp-3.7.4.post0-cp38-cp38-manylinux2014_s390x.whl", hash = "sha256:a00bb73540af068ca7390e636c01cbc4f644961896fa9363154ff43fd37af2f5"}, + {file = "aiohttp-3.7.4.post0-cp38-cp38-manylinux2014_x86_64.whl", hash = "sha256:79ebfc238612123a713a457d92afb4096e2148be17df6c50fb9bf7a81c2f8013"}, + {file = "aiohttp-3.7.4.post0-cp38-cp38-win32.whl", hash = "sha256:515dfef7f869a0feb2afee66b957cc7bbe9ad0cdee45aec7fdc623f4ecd4fb16"}, + {file = "aiohttp-3.7.4.post0-cp38-cp38-win_amd64.whl", hash = "sha256:114b281e4d68302a324dd33abb04778e8557d88947875cbf4e842c2c01a030c5"}, + {file = "aiohttp-3.7.4.post0-cp39-cp39-macosx_10_14_x86_64.whl", hash = "sha256:7b18b97cf8ee5452fa5f4e3af95d01d84d86d32c5e2bfa260cf041749d66360b"}, + {file = "aiohttp-3.7.4.post0-cp39-cp39-manylinux1_i686.whl", hash = "sha256:15492a6368d985b76a2a5fdd2166cddfea5d24e69eefed4630cbaae5c81d89bd"}, + {file = "aiohttp-3.7.4.post0-cp39-cp39-manylinux2014_aarch64.whl", hash = "sha256:bdb230b4943891321e06fc7def63c7aace16095be7d9cf3b1e01be2f10fba439"}, + {file = "aiohttp-3.7.4.post0-cp39-cp39-manylinux2014_i686.whl", hash = "sha256:cffe3ab27871bc3ea47df5d8f7013945712c46a3cc5a95b6bee15887f1675c22"}, + {file = "aiohttp-3.7.4.post0-cp39-cp39-manylinux2014_ppc64le.whl", hash = "sha256:f881853d2643a29e643609da57b96d5f9c9b93f62429dcc1cbb413c7d07f0e1a"}, + {file = "aiohttp-3.7.4.post0-cp39-cp39-manylinux2014_s390x.whl", hash = "sha256:a5ca29ee66f8343ed336816c553e82d6cade48a3ad702b9ffa6125d187e2dedb"}, + {file = "aiohttp-3.7.4.post0-cp39-cp39-manylinux2014_x86_64.whl", hash = "sha256:17c073de315745a1510393a96e680d20af8e67e324f70b42accbd4cb3315c9fb"}, + {file = "aiohttp-3.7.4.post0-cp39-cp39-win32.whl", hash = "sha256:932bb1ea39a54e9ea27fc9232163059a0b8855256f4052e776357ad9add6f1c9"}, + {file = "aiohttp-3.7.4.post0-cp39-cp39-win_amd64.whl", hash = "sha256:02f46fc0e3c5ac58b80d4d56eb0a7c7d97fcef69ace9326289fb9f1955e65cfe"}, + {file = "aiohttp-3.7.4.post0.tar.gz", hash = "sha256:493d3299ebe5f5a7c66b9819eacdcfbbaaf1a8e84911ddffcdc48888497afecf"}, +] +appdirs = [ + {file = "appdirs-1.4.4-py2.py3-none-any.whl", hash = "sha256:a841dacd6b99318a741b166adb07e19ee71a274450e68237b4650ca1055ab128"}, + {file = "appdirs-1.4.4.tar.gz", hash = "sha256:7d5d0167b2b1ba821647616af46a749d1c653740dd0d2415100fe26e27afdf41"}, +] +apscheduler = [ + {file = "APScheduler-3.7.0-py2.py3-none-any.whl", hash = "sha256:c06cc796d5bb9eb3c4f77727f6223476eb67749e7eea074d1587550702a7fbe3"}, + {file = "APScheduler-3.7.0.tar.gz", hash = "sha256:1cab7f2521e107d07127b042155b632b7a1cd5e02c34be5a28ff62f77c900c6a"}, +] +async-timeout = [ + {file = "async-timeout-3.0.1.tar.gz", hash = "sha256:0c3c816a028d47f659d6ff5c745cb2acf1f966da1fe5c19c77a70282b25f4c5f"}, + {file = "async_timeout-3.0.1-py3-none-any.whl", hash = "sha256:4291ca197d287d274d0b6cb5d6f8f8f82d434ed288f962539ff18cc9012f9ea3"}, +] +atomicwrites = [ + {file = "atomicwrites-1.4.0-py2.py3-none-any.whl", hash = "sha256:6d1784dea7c0c8d4a5172b6c620f40b6e4cbfdf96d783691f2e1302a7b88e197"}, + {file = "atomicwrites-1.4.0.tar.gz", hash = "sha256:ae70396ad1a434f9c7046fd2dd196fc04b12f9e91ffb859164193be8b6168a7a"}, +] +attrs = [ + {file = "attrs-21.2.0-py2.py3-none-any.whl", hash = "sha256:149e90d6d8ac20db7a955ad60cf0e6881a3f20d37096140088356da6c716b0b1"}, + {file = "attrs-21.2.0.tar.gz", hash = "sha256:ef6aaac3ca6cd92904cdd0d83f629a15f18053ec84e6432106f7a4d04ae4f5fb"}, +] +black = [ + {file = "black-21.5b2-py3-none-any.whl", hash = "sha256:e5cf21ebdffc7a9b29d73912b6a6a9a4df4ce70220d523c21647da2eae0751ef"}, + {file = "black-21.5b2.tar.gz", hash = "sha256:1fc0e0a2c8ae7d269dfcf0c60a89afa299664f3e811395d40b1922dff8f854b5"}, +] +certifi = [ + {file = "certifi-2021.5.30-py2.py3-none-any.whl", hash = "sha256:50b1e4f8446b06f41be7dd6338db18e0990601dce795c2b1686458aa7e8fa7d8"}, + {file = "certifi-2021.5.30.tar.gz", hash = "sha256:2bbf76fd432960138b3ef6dda3dde0544f27cbf8546c458e60baf371917ba9ee"}, +] +cffi = [ + {file = "cffi-1.14.5-cp27-cp27m-macosx_10_9_x86_64.whl", hash = "sha256:bb89f306e5da99f4d922728ddcd6f7fcebb3241fc40edebcb7284d7514741991"}, + {file = "cffi-1.14.5-cp27-cp27m-manylinux1_i686.whl", hash = "sha256:34eff4b97f3d982fb93e2831e6750127d1355a923ebaeeb565407b3d2f8d41a1"}, + {file = "cffi-1.14.5-cp27-cp27m-manylinux1_x86_64.whl", hash = "sha256:99cd03ae7988a93dd00bcd9d0b75e1f6c426063d6f03d2f90b89e29b25b82dfa"}, + {file = "cffi-1.14.5-cp27-cp27m-win32.whl", hash = "sha256:65fa59693c62cf06e45ddbb822165394a288edce9e276647f0046e1ec26920f3"}, + {file = "cffi-1.14.5-cp27-cp27m-win_amd64.whl", hash = "sha256:51182f8927c5af975fece87b1b369f722c570fe169f9880764b1ee3bca8347b5"}, + {file = "cffi-1.14.5-cp27-cp27mu-manylinux1_i686.whl", hash = "sha256:43e0b9d9e2c9e5d152946b9c5fe062c151614b262fda2e7b201204de0b99e482"}, + {file = "cffi-1.14.5-cp27-cp27mu-manylinux1_x86_64.whl", hash = "sha256:cbde590d4faaa07c72bf979734738f328d239913ba3e043b1e98fe9a39f8b2b6"}, + {file = "cffi-1.14.5-cp35-cp35m-macosx_10_9_x86_64.whl", hash = "sha256:5de7970188bb46b7bf9858eb6890aad302577a5f6f75091fd7cdd3ef13ef3045"}, + {file = "cffi-1.14.5-cp35-cp35m-manylinux1_i686.whl", hash = "sha256:a465da611f6fa124963b91bf432d960a555563efe4ed1cc403ba5077b15370aa"}, + {file = "cffi-1.14.5-cp35-cp35m-manylinux1_x86_64.whl", hash = "sha256:d42b11d692e11b6634f7613ad8df5d6d5f8875f5d48939520d351007b3c13406"}, + {file = "cffi-1.14.5-cp35-cp35m-win32.whl", hash = "sha256:72d8d3ef52c208ee1c7b2e341f7d71c6fd3157138abf1a95166e6165dd5d4369"}, + {file = "cffi-1.14.5-cp35-cp35m-win_amd64.whl", hash = "sha256:29314480e958fd8aab22e4a58b355b629c59bf5f2ac2492b61e3dc06d8c7a315"}, + {file = "cffi-1.14.5-cp36-cp36m-macosx_10_9_x86_64.whl", hash = "sha256:3d3dd4c9e559eb172ecf00a2a7517e97d1e96de2a5e610bd9b68cea3925b4892"}, + {file = "cffi-1.14.5-cp36-cp36m-manylinux1_i686.whl", hash = "sha256:48e1c69bbacfc3d932221851b39d49e81567a4d4aac3b21258d9c24578280058"}, + {file = "cffi-1.14.5-cp36-cp36m-manylinux1_x86_64.whl", hash = "sha256:69e395c24fc60aad6bb4fa7e583698ea6cc684648e1ffb7fe85e3c1ca131a7d5"}, + {file = "cffi-1.14.5-cp36-cp36m-manylinux2014_aarch64.whl", hash = "sha256:9e93e79c2551ff263400e1e4be085a1210e12073a31c2011dbbda14bda0c6132"}, + {file = "cffi-1.14.5-cp36-cp36m-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:24ec4ff2c5c0c8f9c6b87d5bb53555bf267e1e6f70e52e5a9740d32861d36b6f"}, + {file = "cffi-1.14.5-cp36-cp36m-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:3c3f39fa737542161d8b0d680df2ec249334cd70a8f420f71c9304bd83c3cbed"}, + {file = "cffi-1.14.5-cp36-cp36m-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:681d07b0d1e3c462dd15585ef5e33cb021321588bebd910124ef4f4fb71aef55"}, + {file = "cffi-1.14.5-cp36-cp36m-win32.whl", hash = "sha256:58e3f59d583d413809d60779492342801d6e82fefb89c86a38e040c16883be53"}, + {file = "cffi-1.14.5-cp36-cp36m-win_amd64.whl", hash = "sha256:005a36f41773e148deac64b08f233873a4d0c18b053d37da83f6af4d9087b813"}, + {file = "cffi-1.14.5-cp37-cp37m-macosx_10_9_x86_64.whl", hash = "sha256:2894f2df484ff56d717bead0a5c2abb6b9d2bf26d6960c4604d5c48bbc30ee73"}, + {file = "cffi-1.14.5-cp37-cp37m-manylinux1_i686.whl", hash = "sha256:0857f0ae312d855239a55c81ef453ee8fd24136eaba8e87a2eceba644c0d4c06"}, + {file = "cffi-1.14.5-cp37-cp37m-manylinux1_x86_64.whl", hash = "sha256:cd2868886d547469123fadc46eac7ea5253ea7fcb139f12e1dfc2bbd406427d1"}, + {file = "cffi-1.14.5-cp37-cp37m-manylinux2014_aarch64.whl", hash = "sha256:35f27e6eb43380fa080dccf676dece30bef72e4a67617ffda586641cd4508d49"}, + {file = "cffi-1.14.5-cp37-cp37m-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:06d7cd1abac2ffd92e65c0609661866709b4b2d82dd15f611e602b9b188b0b69"}, + {file = "cffi-1.14.5-cp37-cp37m-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:0f861a89e0043afec2a51fd177a567005847973be86f709bbb044d7f42fc4e05"}, + {file = "cffi-1.14.5-cp37-cp37m-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:cc5a8e069b9ebfa22e26d0e6b97d6f9781302fe7f4f2b8776c3e1daea35f1adc"}, + {file = "cffi-1.14.5-cp37-cp37m-win32.whl", hash = "sha256:9ff227395193126d82e60319a673a037d5de84633f11279e336f9c0f189ecc62"}, + {file = "cffi-1.14.5-cp37-cp37m-win_amd64.whl", hash = "sha256:9cf8022fb8d07a97c178b02327b284521c7708d7c71a9c9c355c178ac4bbd3d4"}, + {file = "cffi-1.14.5-cp38-cp38-macosx_10_9_x86_64.whl", hash = "sha256:8b198cec6c72df5289c05b05b8b0969819783f9418e0409865dac47288d2a053"}, + {file = "cffi-1.14.5-cp38-cp38-manylinux1_i686.whl", hash = "sha256:ad17025d226ee5beec591b52800c11680fca3df50b8b29fe51d882576e039ee0"}, + {file = "cffi-1.14.5-cp38-cp38-manylinux1_x86_64.whl", hash = "sha256:6c97d7350133666fbb5cf4abdc1178c812cb205dc6f41d174a7b0f18fb93337e"}, + {file = "cffi-1.14.5-cp38-cp38-manylinux2014_aarch64.whl", hash = "sha256:8ae6299f6c68de06f136f1f9e69458eae58f1dacf10af5c17353eae03aa0d827"}, + {file = "cffi-1.14.5-cp38-cp38-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:04c468b622ed31d408fea2346bec5bbffba2cc44226302a0de1ade9f5ea3d373"}, + {file = "cffi-1.14.5-cp38-cp38-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:06db6321b7a68b2bd6df96d08a5adadc1fa0e8f419226e25b2a5fbf6ccc7350f"}, + {file = "cffi-1.14.5-cp38-cp38-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:293e7ea41280cb28c6fcaaa0b1aa1f533b8ce060b9e701d78511e1e6c4a1de76"}, + {file = "cffi-1.14.5-cp38-cp38-win32.whl", hash = "sha256:b85eb46a81787c50650f2392b9b4ef23e1f126313b9e0e9013b35c15e4288e2e"}, + {file = "cffi-1.14.5-cp38-cp38-win_amd64.whl", hash = "sha256:1f436816fc868b098b0d63b8920de7d208c90a67212546d02f84fe78a9c26396"}, + {file = "cffi-1.14.5-cp39-cp39-macosx_10_9_x86_64.whl", hash = "sha256:1071534bbbf8cbb31b498d5d9db0f274f2f7a865adca4ae429e147ba40f73dea"}, + {file = "cffi-1.14.5-cp39-cp39-manylinux1_i686.whl", hash = "sha256:9de2e279153a443c656f2defd67769e6d1e4163952b3c622dcea5b08a6405322"}, + {file = "cffi-1.14.5-cp39-cp39-manylinux1_x86_64.whl", hash = "sha256:6e4714cc64f474e4d6e37cfff31a814b509a35cb17de4fb1999907575684479c"}, + {file = "cffi-1.14.5-cp39-cp39-manylinux2014_aarch64.whl", hash = "sha256:158d0d15119b4b7ff6b926536763dc0714313aa59e320ddf787502c70c4d4bee"}, + {file = "cffi-1.14.5-cp39-cp39-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:1bf1ac1984eaa7675ca8d5745a8cb87ef7abecb5592178406e55858d411eadc0"}, + {file = "cffi-1.14.5-cp39-cp39-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:df5052c5d867c1ea0b311fb7c3cd28b19df469c056f7fdcfe88c7473aa63e333"}, + {file = "cffi-1.14.5-cp39-cp39-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:24a570cd11895b60829e941f2613a4f79df1a27344cbbb82164ef2e0116f09c7"}, + {file = "cffi-1.14.5-cp39-cp39-win32.whl", hash = "sha256:afb29c1ba2e5a3736f1c301d9d0abe3ec8b86957d04ddfa9d7a6a42b9367e396"}, + {file = "cffi-1.14.5-cp39-cp39-win_amd64.whl", hash = "sha256:f2d45f97ab6bb54753eab54fffe75aaf3de4ff2341c9daee1987ee1837636f1d"}, + {file = "cffi-1.14.5.tar.gz", hash = "sha256:fd78e5fee591709f32ef6edb9a015b4aa1a5022598e36227500c8f4e02328d9c"}, +] +cfgv = [ + {file = "cfgv-3.3.0-py2.py3-none-any.whl", hash = "sha256:b449c9c6118fe8cca7fa5e00b9ec60ba08145d281d52164230a69211c5d597a1"}, + {file = "cfgv-3.3.0.tar.gz", hash = "sha256:9e600479b3b99e8af981ecdfc80a0296104ee610cab48a5ae4ffd0b668650eb1"}, +] +chardet = [ + {file = "chardet-4.0.0-py2.py3-none-any.whl", hash = "sha256:f864054d66fd9118f2e67044ac8981a54775ec5b67aed0441892edb553d21da5"}, + {file = "chardet-4.0.0.tar.gz", hash = "sha256:0d6f53a15db4120f2b08c94f11e7d93d2c911ee118b6b30a04ec3ee8310179fa"}, +] +click = [ + {file = "click-8.0.1-py3-none-any.whl", hash = "sha256:fba402a4a47334742d782209a7c79bc448911afe1149d07bdabdf480b3e2f4b6"}, + {file = "click-8.0.1.tar.gz", hash = "sha256:8c04c11192119b1ef78ea049e0a6f0463e4c48ef00a30160c704337586f3ad7a"}, +] +colorama = [ + {file = "colorama-0.4.4-py2.py3-none-any.whl", hash = "sha256:9f47eda37229f68eee03b24b9748937c7dc3868f906e8ba69fbcbdd3bc5dc3e2"}, + {file = "colorama-0.4.4.tar.gz", hash = "sha256:5941b2b48a20143d2267e95b1c2a7603ce057ee39fd88e7329b0c292aa16869b"}, +] +cryptography = [ + {file = "cryptography-3.4.7-cp36-abi3-macosx_10_10_x86_64.whl", hash = "sha256:3d8427734c781ea5f1b41d6589c293089704d4759e34597dce91014ac125aad1"}, + {file = "cryptography-3.4.7-cp36-abi3-macosx_11_0_arm64.whl", hash = "sha256:8e56e16617872b0957d1c9742a3f94b43533447fd78321514abbe7db216aa250"}, + {file = "cryptography-3.4.7-cp36-abi3-manylinux2010_x86_64.whl", hash = "sha256:37340614f8a5d2fb9aeea67fd159bfe4f5f4ed535b1090ce8ec428b2f15a11f2"}, + {file = "cryptography-3.4.7-cp36-abi3-manylinux2014_aarch64.whl", hash = "sha256:240f5c21aef0b73f40bb9f78d2caff73186700bf1bc6b94285699aff98cc16c6"}, + {file = "cryptography-3.4.7-cp36-abi3-manylinux2014_x86_64.whl", hash = "sha256:1e056c28420c072c5e3cb36e2b23ee55e260cb04eee08f702e0edfec3fb51959"}, + {file = "cryptography-3.4.7-cp36-abi3-win32.whl", hash = "sha256:0f1212a66329c80d68aeeb39b8a16d54ef57071bf22ff4e521657b27372e327d"}, + {file = "cryptography-3.4.7-cp36-abi3-win_amd64.whl", hash = "sha256:de4e5f7f68220d92b7637fc99847475b59154b7a1b3868fb7385337af54ac9ca"}, + {file = "cryptography-3.4.7-pp36-pypy36_pp73-manylinux2010_x86_64.whl", hash = "sha256:26965837447f9c82f1855e0bc8bc4fb910240b6e0d16a664bb722df3b5b06873"}, + {file = "cryptography-3.4.7-pp36-pypy36_pp73-manylinux2014_x86_64.whl", hash = "sha256:eb8cc2afe8b05acbd84a43905832ec78e7b3873fb124ca190f574dca7389a87d"}, + {file = "cryptography-3.4.7-pp37-pypy37_pp73-manylinux2010_x86_64.whl", hash = "sha256:7ec5d3b029f5fa2b179325908b9cd93db28ab7b85bb6c1db56b10e0b54235177"}, + {file = "cryptography-3.4.7-pp37-pypy37_pp73-manylinux2014_x86_64.whl", hash = "sha256:ee77aa129f481be46f8d92a1a7db57269a2f23052d5f2433b4621bb457081cc9"}, + {file = "cryptography-3.4.7.tar.gz", hash = "sha256:3d10de8116d25649631977cb37da6cbdd2d6fa0e0281d014a5b7d337255ca713"}, +] +distlib = [ + {file = "distlib-0.3.2-py2.py3-none-any.whl", hash = "sha256:23e223426b28491b1ced97dc3bbe183027419dfc7982b4fa2f05d5f3ff10711c"}, + {file = "distlib-0.3.2.zip", hash = "sha256:106fef6dc37dd8c0e2c0a60d3fca3e77460a48907f335fa28420463a6f799736"}, +] +eight = [ + {file = "eight-1.0.1-py2.py3-none-any.whl", hash = "sha256:bd6967ec7cf9e61ab18b55751236ed89a8363cfcb69127aace336869615245d5"}, + {file = "eight-1.0.1.tar.gz", hash = "sha256:6465970d3c12a1d759151690629b10b50ffb124c9b62e912999ee9233378a562"}, +] +filelock = [ + {file = "filelock-3.0.12-py3-none-any.whl", hash = "sha256:929b7d63ec5b7d6b71b0fa5ac14e030b3f70b75747cef1b10da9b879fef15836"}, + {file = "filelock-3.0.12.tar.gz", hash = "sha256:18d82244ee114f543149c66a6e0c14e9c4f8a1044b5cdaadd0f82159d6a6ff59"}, +] +future = [ + {file = "future-0.18.2.tar.gz", hash = "sha256:b1bead90b70cf6ec3f0710ae53a525360fa360d306a86583adc6bf83a4db537d"}, +] +identify = [ + {file = "identify-2.2.7-py2.py3-none-any.whl", hash = "sha256:92d6ad08eca19ceb17576733759944b94c0761277ddc3acf65e75e57ef190e32"}, + {file = "identify-2.2.7.tar.gz", hash = "sha256:c29e74c3671fe9537715cb695148231d777170ca1498e1c30c675d4ea782afe9"}, +] +idna = [ + {file = "idna-3.2-py3-none-any.whl", hash = "sha256:14475042e284991034cb48e06f6851428fb14c4dc953acd9be9a5e95c7b6dd7a"}, + {file = "idna-3.2.tar.gz", hash = "sha256:467fbad99067910785144ce333826c71fb0e63a425657295239737f7ecd125f3"}, +] +iniconfig = [ + {file = "iniconfig-1.1.1-py2.py3-none-any.whl", hash = "sha256:011e24c64b7f47f6ebd835bb12a743f2fbe9a26d4cecaa7f53bc4f35ee9da8b3"}, + {file = "iniconfig-1.1.1.tar.gz", hash = "sha256:bc3af051d7d14b2ee5ef9969666def0cd1a000e121eaea580d4a313df4b37f32"}, +] +jinja2 = [ + {file = "Jinja2-3.0.1-py3-none-any.whl", hash = "sha256:1f06f2da51e7b56b8f238affdd6b4e2c61e39598a378cc49345bc1bd42a978a4"}, + {file = "Jinja2-3.0.1.tar.gz", hash = "sha256:703f484b47a6af502e743c9122595cc812b0271f661722403114f71a79d0f5a4"}, +] +lxml = [ + {file = "lxml-4.6.3-cp27-cp27m-macosx_10_9_x86_64.whl", hash = "sha256:df7c53783a46febb0e70f6b05df2ba104610f2fb0d27023409734a3ecbb78fb2"}, + {file = "lxml-4.6.3-cp27-cp27m-manylinux1_i686.whl", hash = "sha256:1b7584d421d254ab86d4f0b13ec662a9014397678a7c4265a02a6d7c2b18a75f"}, + {file = "lxml-4.6.3-cp27-cp27m-manylinux1_x86_64.whl", hash = "sha256:079f3ae844f38982d156efce585bc540c16a926d4436712cf4baee0cce487a3d"}, + {file = "lxml-4.6.3-cp27-cp27m-win32.whl", hash = "sha256:bc4313cbeb0e7a416a488d72f9680fffffc645f8a838bd2193809881c67dd106"}, + {file = "lxml-4.6.3-cp27-cp27m-win_amd64.whl", hash = "sha256:8157dadbb09a34a6bd95a50690595e1fa0af1a99445e2744110e3dca7831c4ee"}, + {file = "lxml-4.6.3-cp27-cp27mu-manylinux1_i686.whl", hash = "sha256:7728e05c35412ba36d3e9795ae8995e3c86958179c9770e65558ec3fdfd3724f"}, + {file = "lxml-4.6.3-cp27-cp27mu-manylinux1_x86_64.whl", hash = "sha256:4bff24dfeea62f2e56f5bab929b4428ae6caba2d1eea0c2d6eb618e30a71e6d4"}, + {file = "lxml-4.6.3-cp35-cp35m-manylinux1_i686.whl", hash = "sha256:74f7d8d439b18fa4c385f3f5dfd11144bb87c1da034a466c5b5577d23a1d9b51"}, + {file = "lxml-4.6.3-cp35-cp35m-manylinux1_x86_64.whl", hash = "sha256:f90ba11136bfdd25cae3951af8da2e95121c9b9b93727b1b896e3fa105b2f586"}, + {file = "lxml-4.6.3-cp35-cp35m-manylinux2010_i686.whl", hash = "sha256:4c61b3a0db43a1607d6264166b230438f85bfed02e8cff20c22e564d0faff354"}, + {file = "lxml-4.6.3-cp35-cp35m-manylinux2014_x86_64.whl", hash = "sha256:5c8c163396cc0df3fd151b927e74f6e4acd67160d6c33304e805b84293351d16"}, + {file = "lxml-4.6.3-cp35-cp35m-win32.whl", hash = "sha256:f2380a6376dfa090227b663f9678150ef27543483055cc327555fb592c5967e2"}, + {file = "lxml-4.6.3-cp35-cp35m-win_amd64.whl", hash = "sha256:c4f05c5a7c49d2fb70223d0d5bcfbe474cf928310ac9fa6a7c6dddc831d0b1d4"}, + {file = "lxml-4.6.3-cp36-cp36m-macosx_10_9_x86_64.whl", hash = "sha256:d2e35d7bf1c1ac8c538f88d26b396e73dd81440d59c1ef8522e1ea77b345ede4"}, + {file = "lxml-4.6.3-cp36-cp36m-manylinux1_i686.whl", hash = "sha256:289e9ca1a9287f08daaf796d96e06cb2bc2958891d7911ac7cae1c5f9e1e0ee3"}, + {file = "lxml-4.6.3-cp36-cp36m-manylinux1_x86_64.whl", hash = "sha256:bccbfc27563652de7dc9bdc595cb25e90b59c5f8e23e806ed0fd623755b6565d"}, + {file = "lxml-4.6.3-cp36-cp36m-manylinux2010_i686.whl", hash = "sha256:d916d31fd85b2f78c76400d625076d9124de3e4bda8b016d25a050cc7d603f24"}, + {file = "lxml-4.6.3-cp36-cp36m-manylinux2014_aarch64.whl", hash = "sha256:820628b7b3135403540202e60551e741f9b6d3304371712521be939470b454ec"}, + {file = "lxml-4.6.3-cp36-cp36m-manylinux2014_x86_64.whl", hash = "sha256:c47ff7e0a36d4efac9fd692cfa33fbd0636674c102e9e8d9b26e1b93a94e7617"}, + {file = "lxml-4.6.3-cp36-cp36m-win32.whl", hash = "sha256:5a0a14e264069c03e46f926be0d8919f4105c1623d620e7ec0e612a2e9bf1c04"}, + {file = "lxml-4.6.3-cp36-cp36m-win_amd64.whl", hash = "sha256:92e821e43ad382332eade6812e298dc9701c75fe289f2a2d39c7960b43d1e92a"}, + {file = "lxml-4.6.3-cp37-cp37m-macosx_10_9_x86_64.whl", hash = "sha256:efd7a09678fd8b53117f6bae4fa3825e0a22b03ef0a932e070c0bdbb3a35e654"}, + {file = "lxml-4.6.3-cp37-cp37m-manylinux1_i686.whl", hash = "sha256:efac139c3f0bf4f0939f9375af4b02c5ad83a622de52d6dfa8e438e8e01d0eb0"}, + {file = "lxml-4.6.3-cp37-cp37m-manylinux1_x86_64.whl", hash = "sha256:0fbcf5565ac01dff87cbfc0ff323515c823081c5777a9fc7703ff58388c258c3"}, + {file = "lxml-4.6.3-cp37-cp37m-manylinux2010_i686.whl", hash = "sha256:36108c73739985979bf302006527cf8a20515ce444ba916281d1c43938b8bb96"}, + {file = "lxml-4.6.3-cp37-cp37m-manylinux2014_aarch64.whl", hash = "sha256:122fba10466c7bd4178b07dba427aa516286b846b2cbd6f6169141917283aae2"}, + {file = "lxml-4.6.3-cp37-cp37m-manylinux2014_x86_64.whl", hash = "sha256:cdaf11d2bd275bf391b5308f86731e5194a21af45fbaaaf1d9e8147b9160ea92"}, + {file = "lxml-4.6.3-cp37-cp37m-win32.whl", hash = "sha256:3439c71103ef0e904ea0a1901611863e51f50b5cd5e8654a151740fde5e1cade"}, + {file = "lxml-4.6.3-cp37-cp37m-win_amd64.whl", hash = "sha256:4289728b5e2000a4ad4ab8da6e1db2e093c63c08bdc0414799ee776a3f78da4b"}, + {file = "lxml-4.6.3-cp38-cp38-macosx_10_9_x86_64.whl", hash = "sha256:b007cbb845b28db4fb8b6a5cdcbf65bacb16a8bd328b53cbc0698688a68e1caa"}, + {file = "lxml-4.6.3-cp38-cp38-manylinux1_i686.whl", hash = "sha256:76fa7b1362d19f8fbd3e75fe2fb7c79359b0af8747e6f7141c338f0bee2f871a"}, + {file = "lxml-4.6.3-cp38-cp38-manylinux1_x86_64.whl", hash = "sha256:26e761ab5b07adf5f555ee82fb4bfc35bf93750499c6c7614bd64d12aaa67927"}, + {file = "lxml-4.6.3-cp38-cp38-manylinux2010_i686.whl", hash = "sha256:e1cbd3f19a61e27e011e02f9600837b921ac661f0c40560eefb366e4e4fb275e"}, + {file = "lxml-4.6.3-cp38-cp38-manylinux2014_aarch64.whl", hash = "sha256:66e575c62792c3f9ca47cb8b6fab9e35bab91360c783d1606f758761810c9791"}, + {file = "lxml-4.6.3-cp38-cp38-manylinux2014_x86_64.whl", hash = "sha256:1b38116b6e628118dea5b2186ee6820ab138dbb1e24a13e478490c7db2f326ae"}, + {file = "lxml-4.6.3-cp38-cp38-win32.whl", hash = "sha256:89b8b22a5ff72d89d48d0e62abb14340d9e99fd637d046c27b8b257a01ffbe28"}, + {file = "lxml-4.6.3-cp38-cp38-win_amd64.whl", hash = "sha256:2a9d50e69aac3ebee695424f7dbd7b8c6d6eb7de2a2eb6b0f6c7db6aa41e02b7"}, + {file = "lxml-4.6.3-cp39-cp39-macosx_10_9_x86_64.whl", hash = "sha256:ce256aaa50f6cc9a649c51be3cd4ff142d67295bfc4f490c9134d0f9f6d58ef0"}, + {file = "lxml-4.6.3-cp39-cp39-manylinux1_i686.whl", hash = "sha256:7610b8c31688f0b1be0ef882889817939490a36d0ee880ea562a4e1399c447a1"}, + {file = "lxml-4.6.3-cp39-cp39-manylinux1_x86_64.whl", hash = "sha256:f8380c03e45cf09f8557bdaa41e1fa7c81f3ae22828e1db470ab2a6c96d8bc23"}, + {file = "lxml-4.6.3-cp39-cp39-manylinux2010_i686.whl", hash = "sha256:3082c518be8e97324390614dacd041bb1358c882d77108ca1957ba47738d9d59"}, + {file = "lxml-4.6.3-cp39-cp39-manylinux2014_aarch64.whl", hash = "sha256:884ab9b29feaca361f7f88d811b1eea9bfca36cf3da27768d28ad45c3ee6f969"}, + {file = "lxml-4.6.3-cp39-cp39-manylinux2014_x86_64.whl", hash = "sha256:6f12e1427285008fd32a6025e38e977d44d6382cf28e7201ed10d6c1698d2a9a"}, + {file = "lxml-4.6.3-cp39-cp39-win32.whl", hash = "sha256:33bb934a044cf32157c12bfcfbb6649807da20aa92c062ef51903415c704704f"}, + {file = "lxml-4.6.3-cp39-cp39-win_amd64.whl", hash = "sha256:542d454665a3e277f76954418124d67516c5f88e51a900365ed54a9806122b83"}, + {file = "lxml-4.6.3.tar.gz", hash = "sha256:39b78571b3b30645ac77b95f7c69d1bffc4cf8c3b157c435a34da72e78c82468"}, +] +markupsafe = [ + {file = "MarkupSafe-2.0.1-cp36-cp36m-macosx_10_9_x86_64.whl", hash = "sha256:f9081981fe268bd86831e5c75f7de206ef275defcb82bc70740ae6dc507aee51"}, + {file = "MarkupSafe-2.0.1-cp36-cp36m-manylinux1_i686.whl", hash = "sha256:0955295dd5eec6cb6cc2fe1698f4c6d84af2e92de33fbcac4111913cd100a6ff"}, + {file = "MarkupSafe-2.0.1-cp36-cp36m-manylinux1_x86_64.whl", hash = "sha256:0446679737af14f45767963a1a9ef7620189912317d095f2d9ffa183a4d25d2b"}, + {file = "MarkupSafe-2.0.1-cp36-cp36m-manylinux2010_i686.whl", hash = "sha256:f826e31d18b516f653fe296d967d700fddad5901ae07c622bb3705955e1faa94"}, + {file = "MarkupSafe-2.0.1-cp36-cp36m-manylinux2010_x86_64.whl", hash = "sha256:fa130dd50c57d53368c9d59395cb5526eda596d3ffe36666cd81a44d56e48872"}, + {file = "MarkupSafe-2.0.1-cp36-cp36m-manylinux2014_aarch64.whl", hash = "sha256:905fec760bd2fa1388bb5b489ee8ee5f7291d692638ea5f67982d968366bef9f"}, + {file = "MarkupSafe-2.0.1-cp36-cp36m-win32.whl", hash = "sha256:6c4ca60fa24e85fe25b912b01e62cb969d69a23a5d5867682dd3e80b5b02581d"}, + {file = "MarkupSafe-2.0.1-cp36-cp36m-win_amd64.whl", hash = "sha256:b2f4bf27480f5e5e8ce285a8c8fd176c0b03e93dcc6646477d4630e83440c6a9"}, + {file = "MarkupSafe-2.0.1-cp37-cp37m-macosx_10_9_x86_64.whl", hash = "sha256:0717a7390a68be14b8c793ba258e075c6f4ca819f15edfc2a3a027c823718567"}, + {file = "MarkupSafe-2.0.1-cp37-cp37m-manylinux1_i686.whl", hash = "sha256:6557b31b5e2c9ddf0de32a691f2312a32f77cd7681d8af66c2692efdbef84c18"}, + {file = "MarkupSafe-2.0.1-cp37-cp37m-manylinux1_x86_64.whl", hash = "sha256:49e3ceeabbfb9d66c3aef5af3a60cc43b85c33df25ce03d0031a608b0a8b2e3f"}, + {file = "MarkupSafe-2.0.1-cp37-cp37m-manylinux2010_i686.whl", hash = "sha256:d7f9850398e85aba693bb640262d3611788b1f29a79f0c93c565694658f4071f"}, + {file = "MarkupSafe-2.0.1-cp37-cp37m-manylinux2010_x86_64.whl", hash = "sha256:6a7fae0dd14cf60ad5ff42baa2e95727c3d81ded453457771d02b7d2b3f9c0c2"}, + {file = "MarkupSafe-2.0.1-cp37-cp37m-manylinux2014_aarch64.whl", hash = "sha256:b7f2d075102dc8c794cbde1947378051c4e5180d52d276987b8d28a3bd58c17d"}, + {file = "MarkupSafe-2.0.1-cp37-cp37m-win32.whl", hash = "sha256:a30e67a65b53ea0a5e62fe23682cfe22712e01f453b95233b25502f7c61cb415"}, + {file = "MarkupSafe-2.0.1-cp37-cp37m-win_amd64.whl", hash = "sha256:611d1ad9a4288cf3e3c16014564df047fe08410e628f89805e475368bd304914"}, + {file = "MarkupSafe-2.0.1-cp38-cp38-macosx_10_9_x86_64.whl", hash = "sha256:be98f628055368795d818ebf93da628541e10b75b41c559fdf36d104c5787066"}, + {file = "MarkupSafe-2.0.1-cp38-cp38-manylinux1_i686.whl", hash = "sha256:1d609f577dc6e1aa17d746f8bd3c31aa4d258f4070d61b2aa5c4166c1539de35"}, + {file = "MarkupSafe-2.0.1-cp38-cp38-manylinux1_x86_64.whl", hash = "sha256:7d91275b0245b1da4d4cfa07e0faedd5b0812efc15b702576d103293e252af1b"}, + {file = "MarkupSafe-2.0.1-cp38-cp38-manylinux2010_i686.whl", hash = "sha256:01a9b8ea66f1658938f65b93a85ebe8bc016e6769611be228d797c9d998dd298"}, + {file = "MarkupSafe-2.0.1-cp38-cp38-manylinux2010_x86_64.whl", hash = "sha256:47ab1e7b91c098ab893b828deafa1203de86d0bc6ab587b160f78fe6c4011f75"}, + {file = "MarkupSafe-2.0.1-cp38-cp38-manylinux2014_aarch64.whl", hash = "sha256:97383d78eb34da7e1fa37dd273c20ad4320929af65d156e35a5e2d89566d9dfb"}, + {file = "MarkupSafe-2.0.1-cp38-cp38-win32.whl", hash = "sha256:023cb26ec21ece8dc3907c0e8320058b2e0cb3c55cf9564da612bc325bed5e64"}, + {file = "MarkupSafe-2.0.1-cp38-cp38-win_amd64.whl", hash = "sha256:984d76483eb32f1bcb536dc27e4ad56bba4baa70be32fa87152832cdd9db0833"}, + {file = "MarkupSafe-2.0.1-cp39-cp39-macosx_10_9_universal2.whl", hash = "sha256:2ef54abee730b502252bcdf31b10dacb0a416229b72c18b19e24a4509f273d26"}, + {file = "MarkupSafe-2.0.1-cp39-cp39-macosx_10_9_x86_64.whl", hash = "sha256:3c112550557578c26af18a1ccc9e090bfe03832ae994343cfdacd287db6a6ae7"}, + {file = "MarkupSafe-2.0.1-cp39-cp39-manylinux1_i686.whl", hash = "sha256:53edb4da6925ad13c07b6d26c2a852bd81e364f95301c66e930ab2aef5b5ddd8"}, + {file = "MarkupSafe-2.0.1-cp39-cp39-manylinux1_x86_64.whl", hash = "sha256:f5653a225f31e113b152e56f154ccbe59eeb1c7487b39b9d9f9cdb58e6c79dc5"}, + {file = "MarkupSafe-2.0.1-cp39-cp39-manylinux2010_i686.whl", hash = "sha256:4efca8f86c54b22348a5467704e3fec767b2db12fc39c6d963168ab1d3fc9135"}, + {file = "MarkupSafe-2.0.1-cp39-cp39-manylinux2010_x86_64.whl", hash = "sha256:ab3ef638ace319fa26553db0624c4699e31a28bb2a835c5faca8f8acf6a5a902"}, + {file = "MarkupSafe-2.0.1-cp39-cp39-manylinux2014_aarch64.whl", hash = "sha256:f8ba0e8349a38d3001fae7eadded3f6606f0da5d748ee53cc1dab1d6527b9509"}, + {file = "MarkupSafe-2.0.1-cp39-cp39-win32.whl", hash = "sha256:10f82115e21dc0dfec9ab5c0223652f7197feb168c940f3ef61563fc2d6beb74"}, + {file = "MarkupSafe-2.0.1-cp39-cp39-win_amd64.whl", hash = "sha256:693ce3f9e70a6cf7d2fb9e6c9d8b204b6b39897a2c4a1aa65728d5ac97dcc1d8"}, + {file = "MarkupSafe-2.0.1.tar.gz", hash = "sha256:594c67807fb16238b30c44bdf74f36c02cdf22d1c8cda91ef8a0ed8dabf5620a"}, +] +multidict = [ + {file = "multidict-5.1.0-cp36-cp36m-macosx_10_14_x86_64.whl", hash = "sha256:b7993704f1a4b204e71debe6095150d43b2ee6150fa4f44d6d966ec356a8d61f"}, + {file = "multidict-5.1.0-cp36-cp36m-manylinux1_i686.whl", hash = "sha256:9dd6e9b1a913d096ac95d0399bd737e00f2af1e1594a787e00f7975778c8b2bf"}, + {file = "multidict-5.1.0-cp36-cp36m-manylinux2014_aarch64.whl", hash = "sha256:f21756997ad8ef815d8ef3d34edd98804ab5ea337feedcd62fb52d22bf531281"}, + {file = "multidict-5.1.0-cp36-cp36m-manylinux2014_i686.whl", hash = "sha256:1ab820665e67373de5802acae069a6a05567ae234ddb129f31d290fc3d1aa56d"}, + {file = "multidict-5.1.0-cp36-cp36m-manylinux2014_ppc64le.whl", hash = "sha256:9436dc58c123f07b230383083855593550c4d301d2532045a17ccf6eca505f6d"}, + {file = "multidict-5.1.0-cp36-cp36m-manylinux2014_s390x.whl", hash = "sha256:830f57206cc96ed0ccf68304141fec9481a096c4d2e2831f311bde1c404401da"}, + {file = "multidict-5.1.0-cp36-cp36m-manylinux2014_x86_64.whl", hash = "sha256:2e68965192c4ea61fff1b81c14ff712fc7dc15d2bd120602e4a3494ea6584224"}, + {file = "multidict-5.1.0-cp36-cp36m-win32.whl", hash = "sha256:2f1a132f1c88724674271d636e6b7351477c27722f2ed789f719f9e3545a3d26"}, + {file = "multidict-5.1.0-cp36-cp36m-win_amd64.whl", hash = "sha256:3a4f32116f8f72ecf2a29dabfb27b23ab7cdc0ba807e8459e59a93a9be9506f6"}, + {file = "multidict-5.1.0-cp37-cp37m-macosx_10_14_x86_64.whl", hash = "sha256:46c73e09ad374a6d876c599f2328161bcd95e280f84d2060cf57991dec5cfe76"}, + {file = "multidict-5.1.0-cp37-cp37m-manylinux1_i686.whl", hash = "sha256:018132dbd8688c7a69ad89c4a3f39ea2f9f33302ebe567a879da8f4ca73f0d0a"}, + {file = "multidict-5.1.0-cp37-cp37m-manylinux2014_aarch64.whl", hash = "sha256:4b186eb7d6ae7c06eb4392411189469e6a820da81447f46c0072a41c748ab73f"}, + {file = "multidict-5.1.0-cp37-cp37m-manylinux2014_i686.whl", hash = "sha256:3a041b76d13706b7fff23b9fc83117c7b8fe8d5fe9e6be45eee72b9baa75f348"}, + {file = "multidict-5.1.0-cp37-cp37m-manylinux2014_ppc64le.whl", hash = "sha256:051012ccee979b2b06be928a6150d237aec75dd6bf2d1eeeb190baf2b05abc93"}, + {file = "multidict-5.1.0-cp37-cp37m-manylinux2014_s390x.whl", hash = "sha256:6a4d5ce640e37b0efcc8441caeea8f43a06addace2335bd11151bc02d2ee31f9"}, + {file = "multidict-5.1.0-cp37-cp37m-manylinux2014_x86_64.whl", hash = "sha256:5cf3443199b83ed9e955f511b5b241fd3ae004e3cb81c58ec10f4fe47c7dce37"}, + {file = "multidict-5.1.0-cp37-cp37m-win32.whl", hash = "sha256:f200755768dc19c6f4e2b672421e0ebb3dd54c38d5a4f262b872d8cfcc9e93b5"}, + {file = "multidict-5.1.0-cp37-cp37m-win_amd64.whl", hash = "sha256:05c20b68e512166fddba59a918773ba002fdd77800cad9f55b59790030bab632"}, + {file = "multidict-5.1.0-cp38-cp38-macosx_10_14_x86_64.whl", hash = "sha256:54fd1e83a184e19c598d5e70ba508196fd0bbdd676ce159feb412a4a6664f952"}, + {file = "multidict-5.1.0-cp38-cp38-manylinux1_i686.whl", hash = "sha256:0e3c84e6c67eba89c2dbcee08504ba8644ab4284863452450520dad8f1e89b79"}, + {file = "multidict-5.1.0-cp38-cp38-manylinux2014_aarch64.whl", hash = "sha256:dc862056f76443a0db4509116c5cd480fe1b6a2d45512a653f9a855cc0517456"}, + {file = "multidict-5.1.0-cp38-cp38-manylinux2014_i686.whl", hash = "sha256:0e929169f9c090dae0646a011c8b058e5e5fb391466016b39d21745b48817fd7"}, + {file = "multidict-5.1.0-cp38-cp38-manylinux2014_ppc64le.whl", hash = "sha256:d81eddcb12d608cc08081fa88d046c78afb1bf8107e6feab5d43503fea74a635"}, + {file = "multidict-5.1.0-cp38-cp38-manylinux2014_s390x.whl", hash = "sha256:585fd452dd7782130d112f7ddf3473ffdd521414674c33876187e101b588738a"}, + {file = "multidict-5.1.0-cp38-cp38-manylinux2014_x86_64.whl", hash = "sha256:37e5438e1c78931df5d3c0c78ae049092877e5e9c02dd1ff5abb9cf27a5914ea"}, + {file = "multidict-5.1.0-cp38-cp38-win32.whl", hash = "sha256:07b42215124aedecc6083f1ce6b7e5ec5b50047afa701f3442054373a6deb656"}, + {file = "multidict-5.1.0-cp38-cp38-win_amd64.whl", hash = "sha256:929006d3c2d923788ba153ad0de8ed2e5ed39fdbe8e7be21e2f22ed06c6783d3"}, + {file = "multidict-5.1.0-cp39-cp39-macosx_10_14_x86_64.whl", hash = "sha256:b797515be8743b771aa868f83563f789bbd4b236659ba52243b735d80b29ed93"}, + {file = "multidict-5.1.0-cp39-cp39-manylinux1_i686.whl", hash = "sha256:d5c65bdf4484872c4af3150aeebe101ba560dcfb34488d9a8ff8dbcd21079647"}, + {file = "multidict-5.1.0-cp39-cp39-manylinux2014_aarch64.whl", hash = "sha256:b47a43177a5e65b771b80db71e7be76c0ba23cc8aa73eeeb089ed5219cdbe27d"}, + {file = "multidict-5.1.0-cp39-cp39-manylinux2014_i686.whl", hash = "sha256:806068d4f86cb06af37cd65821554f98240a19ce646d3cd24e1c33587f313eb8"}, + {file = "multidict-5.1.0-cp39-cp39-manylinux2014_ppc64le.whl", hash = "sha256:46dd362c2f045095c920162e9307de5ffd0a1bfbba0a6e990b344366f55a30c1"}, + {file = "multidict-5.1.0-cp39-cp39-manylinux2014_s390x.whl", hash = "sha256:ace010325c787c378afd7f7c1ac66b26313b3344628652eacd149bdd23c68841"}, + {file = "multidict-5.1.0-cp39-cp39-manylinux2014_x86_64.whl", hash = "sha256:ecc771ab628ea281517e24fd2c52e8f31c41e66652d07599ad8818abaad38cda"}, + {file = "multidict-5.1.0-cp39-cp39-win32.whl", hash = "sha256:fc13a9524bc18b6fb6e0dbec3533ba0496bbed167c56d0aabefd965584557d80"}, + {file = "multidict-5.1.0-cp39-cp39-win_amd64.whl", hash = "sha256:7df80d07818b385f3129180369079bd6934cf70469f99daaebfac89dca288359"}, + {file = "multidict-5.1.0.tar.gz", hash = "sha256:25b4e5f22d3a37ddf3effc0710ba692cfc792c2b9edfb9c05aefe823256e84d5"}, +] +mypy = [ + {file = "mypy-0.812-cp35-cp35m-macosx_10_9_x86_64.whl", hash = "sha256:a26f8ec704e5a7423c8824d425086705e381b4f1dfdef6e3a1edab7ba174ec49"}, + {file = "mypy-0.812-cp35-cp35m-manylinux1_x86_64.whl", hash = "sha256:28fb5479c494b1bab244620685e2eb3c3f988d71fd5d64cc753195e8ed53df7c"}, + {file = "mypy-0.812-cp35-cp35m-manylinux2010_x86_64.whl", hash = "sha256:9743c91088d396c1a5a3c9978354b61b0382b4e3c440ce83cf77994a43e8c521"}, + {file = "mypy-0.812-cp35-cp35m-win_amd64.whl", hash = "sha256:d7da2e1d5f558c37d6e8c1246f1aec1e7349e4913d8fb3cb289a35de573fe2eb"}, + {file = "mypy-0.812-cp36-cp36m-macosx_10_9_x86_64.whl", hash = "sha256:4eec37370483331d13514c3f55f446fc5248d6373e7029a29ecb7b7494851e7a"}, + {file = "mypy-0.812-cp36-cp36m-manylinux1_x86_64.whl", hash = "sha256:d65cc1df038ef55a99e617431f0553cd77763869eebdf9042403e16089fe746c"}, + {file = "mypy-0.812-cp36-cp36m-manylinux2010_x86_64.whl", hash = "sha256:61a3d5b97955422964be6b3baf05ff2ce7f26f52c85dd88db11d5e03e146a3a6"}, + {file = "mypy-0.812-cp36-cp36m-win_amd64.whl", hash = "sha256:25adde9b862f8f9aac9d2d11971f226bd4c8fbaa89fb76bdadb267ef22d10064"}, + {file = "mypy-0.812-cp37-cp37m-macosx_10_9_x86_64.whl", hash = "sha256:552a815579aa1e995f39fd05dde6cd378e191b063f031f2acfe73ce9fb7f9e56"}, + {file = "mypy-0.812-cp37-cp37m-manylinux1_x86_64.whl", hash = "sha256:499c798053cdebcaa916eef8cd733e5584b5909f789de856b482cd7d069bdad8"}, + {file = "mypy-0.812-cp37-cp37m-manylinux2010_x86_64.whl", hash = "sha256:5873888fff1c7cf5b71efbe80e0e73153fe9212fafdf8e44adfe4c20ec9f82d7"}, + {file = "mypy-0.812-cp37-cp37m-win_amd64.whl", hash = "sha256:9f94aac67a2045ec719ffe6111df543bac7874cee01f41928f6969756e030564"}, + {file = "mypy-0.812-cp38-cp38-macosx_10_9_x86_64.whl", hash = "sha256:d23e0ea196702d918b60c8288561e722bf437d82cb7ef2edcd98cfa38905d506"}, + {file = "mypy-0.812-cp38-cp38-manylinux1_x86_64.whl", hash = "sha256:674e822aa665b9fd75130c6c5f5ed9564a38c6cea6a6432ce47eafb68ee578c5"}, + {file = "mypy-0.812-cp38-cp38-manylinux2010_x86_64.whl", hash = "sha256:abf7e0c3cf117c44d9285cc6128856106183938c68fd4944763003decdcfeb66"}, + {file = "mypy-0.812-cp38-cp38-win_amd64.whl", hash = "sha256:0d0a87c0e7e3a9becdfbe936c981d32e5ee0ccda3e0f07e1ef2c3d1a817cf73e"}, + {file = "mypy-0.812-cp39-cp39-macosx_10_9_x86_64.whl", hash = "sha256:7ce3175801d0ae5fdfa79b4f0cfed08807af4d075b402b7e294e6aa72af9aa2a"}, + {file = "mypy-0.812-cp39-cp39-manylinux1_x86_64.whl", hash = "sha256:b09669bcda124e83708f34a94606e01b614fa71931d356c1f1a5297ba11f110a"}, + {file = "mypy-0.812-cp39-cp39-manylinux2010_x86_64.whl", hash = "sha256:33f159443db0829d16f0a8d83d94df3109bb6dd801975fe86bacb9bf71628e97"}, + {file = "mypy-0.812-cp39-cp39-win_amd64.whl", hash = "sha256:3f2aca7f68580dc2508289c729bd49ee929a436208d2b2b6aab15745a70a57df"}, + {file = "mypy-0.812-py3-none-any.whl", hash = "sha256:2f9b3407c58347a452fc0736861593e105139b905cca7d097e413453a1d650b4"}, + {file = "mypy-0.812.tar.gz", hash = "sha256:cd07039aa5df222037005b08fbbfd69b3ab0b0bd7a07d7906de75ae52c4e3119"}, +] +mypy-extensions = [ + {file = "mypy_extensions-0.4.3-py2.py3-none-any.whl", hash = "sha256:090fedd75945a69ae91ce1303b5824f428daf5a028d2f6ab8a299250a846f15d"}, + {file = "mypy_extensions-0.4.3.tar.gz", hash = "sha256:2d82818f5bb3e369420cb3c4060a7970edba416647068eb4c5343488a6c604a8"}, +] +nodeenv = [ + {file = "nodeenv-1.6.0-py2.py3-none-any.whl", hash = "sha256:621e6b7076565ddcacd2db0294c0381e01fd28945ab36bcf00f41c5daf63bef7"}, + {file = "nodeenv-1.6.0.tar.gz", hash = "sha256:3ef13ff90291ba2a4a7a4ff9a979b63ffdd00a464dbe04acf0ea6471517a4c2b"}, +] +openleadr = [ + {file = "openleadr-0.5.23-py3-none-any.whl", hash = "sha256:863522020f7489d67827c69c1b3d6e75531ff75637238c87ecf69a4b08af439c"}, + {file = "openleadr-0.5.23.tar.gz", hash = "sha256:d6a77576120e51db471fcd1f70df50163363989399a058082e43e5a51a0f7c67"}, +] +packaging = [ + {file = "packaging-20.9-py2.py3-none-any.whl", hash = "sha256:67714da7f7bc052e064859c05c595155bd1ee9f69f76557e21f051443c20947a"}, + {file = "packaging-20.9.tar.gz", hash = "sha256:5b327ac1320dc863dca72f4514ecc086f31186744b84a230374cc1fd776feae5"}, +] +pathspec = [ + {file = "pathspec-0.8.1-py2.py3-none-any.whl", hash = "sha256:aa0cb481c4041bf52ffa7b0d8fa6cd3e88a2ca4879c533c9153882ee2556790d"}, + {file = "pathspec-0.8.1.tar.gz", hash = "sha256:86379d6b86d75816baba717e64b1a3a3469deb93bb76d613c9ce79edc5cb68fd"}, +] +pluggy = [ + {file = "pluggy-0.13.1-py2.py3-none-any.whl", hash = "sha256:966c145cd83c96502c3c3868f50408687b38434af77734af1e9ca461a4081d2d"}, + {file = "pluggy-0.13.1.tar.gz", hash = "sha256:15b2acde666561e1298d71b523007ed7364de07029219b604cf808bfa1c765b0"}, +] +pre-commit = [ + {file = "pre_commit-2.13.0-py2.py3-none-any.whl", hash = "sha256:b679d0fddd5b9d6d98783ae5f10fd0c4c59954f375b70a58cbe1ce9bcf9809a4"}, + {file = "pre_commit-2.13.0.tar.gz", hash = "sha256:764972c60693dc668ba8e86eb29654ec3144501310f7198742a767bec385a378"}, +] +py = [ + {file = "py-1.10.0-py2.py3-none-any.whl", hash = "sha256:3b80836aa6d1feeaa108e046da6423ab8f6ceda6468545ae8d02d9d58d18818a"}, + {file = "py-1.10.0.tar.gz", hash = "sha256:21b81bda15b66ef5e1a777a21c4dcd9c20ad3efd0b3f817e7a809035269e1bd3"}, +] +pycparser = [ + {file = "pycparser-2.20-py2.py3-none-any.whl", hash = "sha256:7582ad22678f0fcd81102833f60ef8d0e57288b6b5fb00323d101be910e35705"}, + {file = "pycparser-2.20.tar.gz", hash = "sha256:2d475327684562c3a96cc71adf7dc8c4f0565175cf86b6d7a404ff4c771f15f0"}, +] +pyopenssl = [ + {file = "pyOpenSSL-19.1.0-py2.py3-none-any.whl", hash = "sha256:621880965a720b8ece2f1b2f54ea2071966ab00e2970ad2ce11d596102063504"}, + {file = "pyOpenSSL-19.1.0.tar.gz", hash = "sha256:9a24494b2602aaf402be5c9e30a0b82d4a5c67528fe8fb475e3f3bc00dd69507"}, +] +pyparsing = [ + {file = "pyparsing-2.4.7-py2.py3-none-any.whl", hash = "sha256:ef9d7589ef3c200abe66653d3f1ab1033c3c419ae9b9bdb1240a85b024efc88b"}, + {file = "pyparsing-2.4.7.tar.gz", hash = "sha256:c203ec8783bf771a155b207279b9bccb8dea02d8f0c9e5f8ead507bc3246ecc1"}, +] +pytest = [ + {file = "pytest-6.2.4-py3-none-any.whl", hash = "sha256:91ef2131a9bd6be8f76f1f08eac5c5317221d6ad1e143ae03894b862e8976890"}, + {file = "pytest-6.2.4.tar.gz", hash = "sha256:50bcad0a0b9c5a72c8e4e7c9855a3ad496ca6a881a3641b4260605450772c54b"}, +] +pytz = [ + {file = "pytz-2021.1-py2.py3-none-any.whl", hash = "sha256:eb10ce3e7736052ed3623d49975ce333bcd712c7bb19a58b9e2089d4057d0798"}, + {file = "pytz-2021.1.tar.gz", hash = "sha256:83a4a90894bf38e243cf052c8b58f381bfe9a7a483f6a9cab140bc7f702ac4da"}, +] +pyyaml = [ + {file = "PyYAML-5.4.1-cp27-cp27m-macosx_10_9_x86_64.whl", hash = "sha256:3b2b1824fe7112845700f815ff6a489360226a5609b96ec2190a45e62a9fc922"}, + {file = "PyYAML-5.4.1-cp27-cp27m-win32.whl", hash = "sha256:129def1b7c1bf22faffd67b8f3724645203b79d8f4cc81f674654d9902cb4393"}, + {file = "PyYAML-5.4.1-cp27-cp27m-win_amd64.whl", hash = "sha256:4465124ef1b18d9ace298060f4eccc64b0850899ac4ac53294547536533800c8"}, + {file = "PyYAML-5.4.1-cp27-cp27mu-manylinux1_x86_64.whl", hash = "sha256:bb4191dfc9306777bc594117aee052446b3fa88737cd13b7188d0e7aa8162185"}, + {file = "PyYAML-5.4.1-cp36-cp36m-macosx_10_9_x86_64.whl", hash = "sha256:6c78645d400265a062508ae399b60b8c167bf003db364ecb26dcab2bda048253"}, + {file = "PyYAML-5.4.1-cp36-cp36m-manylinux1_x86_64.whl", hash = "sha256:4e0583d24c881e14342eaf4ec5fbc97f934b999a6828693a99157fde912540cc"}, + {file = "PyYAML-5.4.1-cp36-cp36m-manylinux2014_aarch64.whl", hash = "sha256:72a01f726a9c7851ca9bfad6fd09ca4e090a023c00945ea05ba1638c09dc3347"}, + {file = "PyYAML-5.4.1-cp36-cp36m-manylinux2014_s390x.whl", hash = "sha256:895f61ef02e8fed38159bb70f7e100e00f471eae2bc838cd0f4ebb21e28f8541"}, + {file = "PyYAML-5.4.1-cp36-cp36m-win32.whl", hash = "sha256:3bd0e463264cf257d1ffd2e40223b197271046d09dadf73a0fe82b9c1fc385a5"}, + {file = "PyYAML-5.4.1-cp36-cp36m-win_amd64.whl", hash = "sha256:e4fac90784481d221a8e4b1162afa7c47ed953be40d31ab4629ae917510051df"}, + {file = "PyYAML-5.4.1-cp37-cp37m-macosx_10_9_x86_64.whl", hash = "sha256:5accb17103e43963b80e6f837831f38d314a0495500067cb25afab2e8d7a4018"}, + {file = "PyYAML-5.4.1-cp37-cp37m-manylinux1_x86_64.whl", hash = "sha256:e1d4970ea66be07ae37a3c2e48b5ec63f7ba6804bdddfdbd3cfd954d25a82e63"}, + {file = "PyYAML-5.4.1-cp37-cp37m-manylinux2014_aarch64.whl", hash = "sha256:cb333c16912324fd5f769fff6bc5de372e9e7a202247b48870bc251ed40239aa"}, + {file = "PyYAML-5.4.1-cp37-cp37m-manylinux2014_s390x.whl", hash = "sha256:fe69978f3f768926cfa37b867e3843918e012cf83f680806599ddce33c2c68b0"}, + {file = "PyYAML-5.4.1-cp37-cp37m-win32.whl", hash = "sha256:dd5de0646207f053eb0d6c74ae45ba98c3395a571a2891858e87df7c9b9bd51b"}, + {file = "PyYAML-5.4.1-cp37-cp37m-win_amd64.whl", hash = "sha256:08682f6b72c722394747bddaf0aa62277e02557c0fd1c42cb853016a38f8dedf"}, + {file = "PyYAML-5.4.1-cp38-cp38-macosx_10_9_x86_64.whl", hash = "sha256:d2d9808ea7b4af864f35ea216be506ecec180628aced0704e34aca0b040ffe46"}, + {file = "PyYAML-5.4.1-cp38-cp38-manylinux1_x86_64.whl", hash = "sha256:8c1be557ee92a20f184922c7b6424e8ab6691788e6d86137c5d93c1a6ec1b8fb"}, + {file = "PyYAML-5.4.1-cp38-cp38-manylinux2014_aarch64.whl", hash = "sha256:fd7f6999a8070df521b6384004ef42833b9bd62cfee11a09bda1079b4b704247"}, + {file = "PyYAML-5.4.1-cp38-cp38-manylinux2014_s390x.whl", hash = "sha256:bfb51918d4ff3d77c1c856a9699f8492c612cde32fd3bcd344af9be34999bfdc"}, + {file = "PyYAML-5.4.1-cp38-cp38-win32.whl", hash = "sha256:fa5ae20527d8e831e8230cbffd9f8fe952815b2b7dae6ffec25318803a7528fc"}, + {file = "PyYAML-5.4.1-cp38-cp38-win_amd64.whl", hash = "sha256:0f5f5786c0e09baddcd8b4b45f20a7b5d61a7e7e99846e3c799b05c7c53fa696"}, + {file = "PyYAML-5.4.1-cp39-cp39-macosx_10_9_x86_64.whl", hash = "sha256:294db365efa064d00b8d1ef65d8ea2c3426ac366c0c4368d930bf1c5fb497f77"}, + {file = "PyYAML-5.4.1-cp39-cp39-manylinux1_x86_64.whl", hash = "sha256:74c1485f7707cf707a7aef42ef6322b8f97921bd89be2ab6317fd782c2d53183"}, + {file = "PyYAML-5.4.1-cp39-cp39-manylinux2014_aarch64.whl", hash = "sha256:d483ad4e639292c90170eb6f7783ad19490e7a8defb3e46f97dfe4bacae89122"}, + {file = "PyYAML-5.4.1-cp39-cp39-manylinux2014_s390x.whl", hash = "sha256:fdc842473cd33f45ff6bce46aea678a54e3d21f1b61a7750ce3c498eedfe25d6"}, + {file = "PyYAML-5.4.1-cp39-cp39-win32.whl", hash = "sha256:49d4cdd9065b9b6e206d0595fee27a96b5dd22618e7520c33204a4a3239d5b10"}, + {file = "PyYAML-5.4.1-cp39-cp39-win_amd64.whl", hash = "sha256:c20cfa2d49991c8b4147af39859b167664f2ad4561704ee74c1de03318e898db"}, + {file = "PyYAML-5.4.1.tar.gz", hash = "sha256:607774cbba28732bfa802b54baa7484215f530991055bb562efbed5b2f20a45e"}, +] +regex = [ + {file = "regex-2021.4.4-cp36-cp36m-macosx_10_9_x86_64.whl", hash = "sha256:619d71c59a78b84d7f18891fe914446d07edd48dc8328c8e149cbe0929b4e000"}, + {file = "regex-2021.4.4-cp36-cp36m-manylinux1_i686.whl", hash = "sha256:47bf5bf60cf04d72bf6055ae5927a0bd9016096bf3d742fa50d9bf9f45aa0711"}, + {file = "regex-2021.4.4-cp36-cp36m-manylinux1_x86_64.whl", hash = "sha256:281d2fd05555079448537fe108d79eb031b403dac622621c78944c235f3fcf11"}, + {file = "regex-2021.4.4-cp36-cp36m-manylinux2010_i686.whl", hash = "sha256:bd28bc2e3a772acbb07787c6308e00d9626ff89e3bfcdebe87fa5afbfdedf968"}, + {file = "regex-2021.4.4-cp36-cp36m-manylinux2010_x86_64.whl", hash = "sha256:7c2a1af393fcc09e898beba5dd59196edaa3116191cc7257f9224beaed3e1aa0"}, + {file = "regex-2021.4.4-cp36-cp36m-manylinux2014_aarch64.whl", hash = "sha256:c38c71df845e2aabb7fb0b920d11a1b5ac8526005e533a8920aea97efb8ec6a4"}, + {file = "regex-2021.4.4-cp36-cp36m-manylinux2014_i686.whl", hash = "sha256:96fcd1888ab4d03adfc9303a7b3c0bd78c5412b2bfbe76db5b56d9eae004907a"}, + {file = "regex-2021.4.4-cp36-cp36m-manylinux2014_x86_64.whl", hash = "sha256:ade17eb5d643b7fead300a1641e9f45401c98eee23763e9ed66a43f92f20b4a7"}, + {file = "regex-2021.4.4-cp36-cp36m-win32.whl", hash = "sha256:e8e5b509d5c2ff12f8418006d5a90e9436766133b564db0abaec92fd27fcee29"}, + {file = "regex-2021.4.4-cp36-cp36m-win_amd64.whl", hash = "sha256:11d773d75fa650cd36f68d7ca936e3c7afaae41b863b8c387a22aaa78d3c5c79"}, + {file = "regex-2021.4.4-cp37-cp37m-macosx_10_9_x86_64.whl", hash = "sha256:d3029c340cfbb3ac0a71798100ccc13b97dddf373a4ae56b6a72cf70dfd53bc8"}, + {file = "regex-2021.4.4-cp37-cp37m-manylinux1_i686.whl", hash = "sha256:18c071c3eb09c30a264879f0d310d37fe5d3a3111662438889ae2eb6fc570c31"}, + {file = "regex-2021.4.4-cp37-cp37m-manylinux1_x86_64.whl", hash = "sha256:4c557a7b470908b1712fe27fb1ef20772b78079808c87d20a90d051660b1d69a"}, + {file = "regex-2021.4.4-cp37-cp37m-manylinux2010_i686.whl", hash = "sha256:01afaf2ec48e196ba91b37451aa353cb7eda77efe518e481707e0515025f0cd5"}, + {file = "regex-2021.4.4-cp37-cp37m-manylinux2010_x86_64.whl", hash = "sha256:3a9cd17e6e5c7eb328517969e0cb0c3d31fd329298dd0c04af99ebf42e904f82"}, + {file = "regex-2021.4.4-cp37-cp37m-manylinux2014_aarch64.whl", hash = "sha256:90f11ff637fe8798933fb29f5ae1148c978cccb0452005bf4c69e13db951e765"}, + {file = "regex-2021.4.4-cp37-cp37m-manylinux2014_i686.whl", hash = "sha256:919859aa909429fb5aa9cf8807f6045592c85ef56fdd30a9a3747e513db2536e"}, + {file = "regex-2021.4.4-cp37-cp37m-manylinux2014_x86_64.whl", hash = "sha256:339456e7d8c06dd36a22e451d58ef72cef293112b559010db3d054d5560ef439"}, + {file = "regex-2021.4.4-cp37-cp37m-win32.whl", hash = "sha256:67bdb9702427ceddc6ef3dc382455e90f785af4c13d495f9626861763ee13f9d"}, + {file = "regex-2021.4.4-cp37-cp37m-win_amd64.whl", hash = "sha256:32e65442138b7b76dd8173ffa2cf67356b7bc1768851dded39a7a13bf9223da3"}, + {file = "regex-2021.4.4-cp38-cp38-macosx_10_9_x86_64.whl", hash = "sha256:1e1c20e29358165242928c2de1482fb2cf4ea54a6a6dea2bd7a0e0d8ee321500"}, + {file = "regex-2021.4.4-cp38-cp38-manylinux1_i686.whl", hash = "sha256:314d66636c494ed9c148a42731b3834496cc9a2c4251b1661e40936814542b14"}, + {file = "regex-2021.4.4-cp38-cp38-manylinux1_x86_64.whl", hash = "sha256:6d1b01031dedf2503631d0903cb563743f397ccaf6607a5e3b19a3d76fc10480"}, + {file = "regex-2021.4.4-cp38-cp38-manylinux2010_i686.whl", hash = "sha256:741a9647fcf2e45f3a1cf0e24f5e17febf3efe8d4ba1281dcc3aa0459ef424dc"}, + {file = "regex-2021.4.4-cp38-cp38-manylinux2010_x86_64.whl", hash = "sha256:4c46e22a0933dd783467cf32b3516299fb98cfebd895817d685130cc50cd1093"}, + {file = "regex-2021.4.4-cp38-cp38-manylinux2014_aarch64.whl", hash = "sha256:e512d8ef5ad7b898cdb2d8ee1cb09a8339e4f8be706d27eaa180c2f177248a10"}, + {file = "regex-2021.4.4-cp38-cp38-manylinux2014_i686.whl", hash = "sha256:980d7be47c84979d9136328d882f67ec5e50008681d94ecc8afa8a65ed1f4a6f"}, + {file = "regex-2021.4.4-cp38-cp38-manylinux2014_x86_64.whl", hash = "sha256:ce15b6d103daff8e9fee13cf7f0add05245a05d866e73926c358e871221eae87"}, + {file = "regex-2021.4.4-cp38-cp38-win32.whl", hash = "sha256:a91aa8619b23b79bcbeb37abe286f2f408d2f2d6f29a17237afda55bb54e7aac"}, + {file = "regex-2021.4.4-cp38-cp38-win_amd64.whl", hash = "sha256:c0502c0fadef0d23b128605d69b58edb2c681c25d44574fc673b0e52dce71ee2"}, + {file = "regex-2021.4.4-cp39-cp39-macosx_10_9_x86_64.whl", hash = "sha256:598585c9f0af8374c28edd609eb291b5726d7cbce16be6a8b95aa074d252ee17"}, + {file = "regex-2021.4.4-cp39-cp39-manylinux1_i686.whl", hash = "sha256:ee54ff27bf0afaf4c3b3a62bcd016c12c3fdb4ec4f413391a90bd38bc3624605"}, + {file = "regex-2021.4.4-cp39-cp39-manylinux1_x86_64.whl", hash = "sha256:7d9884d86dd4dd489e981d94a65cd30d6f07203d90e98f6f657f05170f6324c9"}, + {file = "regex-2021.4.4-cp39-cp39-manylinux2010_i686.whl", hash = "sha256:bf5824bfac591ddb2c1f0a5f4ab72da28994548c708d2191e3b87dd207eb3ad7"}, + {file = "regex-2021.4.4-cp39-cp39-manylinux2010_x86_64.whl", hash = "sha256:563085e55b0d4fb8f746f6a335893bda5c2cef43b2f0258fe1020ab1dd874df8"}, + {file = "regex-2021.4.4-cp39-cp39-manylinux2014_aarch64.whl", hash = "sha256:b9c3db21af35e3b3c05764461b262d6f05bbca08a71a7849fd79d47ba7bc33ed"}, + {file = "regex-2021.4.4-cp39-cp39-manylinux2014_i686.whl", hash = "sha256:3916d08be28a1149fb97f7728fca1f7c15d309a9f9682d89d79db75d5e52091c"}, + {file = "regex-2021.4.4-cp39-cp39-manylinux2014_x86_64.whl", hash = "sha256:fd45ff9293d9274c5008a2054ecef86a9bfe819a67c7be1afb65e69b405b3042"}, + {file = "regex-2021.4.4-cp39-cp39-win32.whl", hash = "sha256:fa4537fb4a98fe8fde99626e4681cc644bdcf2a795038533f9f711513a862ae6"}, + {file = "regex-2021.4.4-cp39-cp39-win_amd64.whl", hash = "sha256:97f29f57d5b84e73fbaf99ab3e26134e6687348e95ef6b48cfd2c06807005a07"}, + {file = "regex-2021.4.4.tar.gz", hash = "sha256:52ba3d3f9b942c49d7e4bc105bb28551c44065f139a65062ab7912bef10c9afb"}, +] +signxml-openadr = [ + {file = "signxml-openadr-2.9.1.tar.gz", hash = "sha256:93bf41824247a36fa09d8ac23435f5675aaa9e1e58fcb3809a549fba9a905c23"}, + {file = "signxml_openadr-2.9.1-py2.py3-none-any.whl", hash = "sha256:da393d73f2f9739b3dcc6b7cbf3c32e22406370ce714cc490ec303d1db822c48"}, +] +six = [ + {file = "six-1.16.0-py2.py3-none-any.whl", hash = "sha256:8abb2f1d86890a2dfb989f9a77cfcfd3e47c2a354b01111771326f8aa26e0254"}, + {file = "six-1.16.0.tar.gz", hash = "sha256:1e61c37477a1626458e36f7b1d82aa5c9b094fa4802892072e49de9c60c4c926"}, +] +toml = [ + {file = "toml-0.10.2-py2.py3-none-any.whl", hash = "sha256:806143ae5bfb6a3c6e736a764057db0e6a0e05e338b5630894a5f779cabb4f9b"}, + {file = "toml-0.10.2.tar.gz", hash = "sha256:b3bda1d108d5dd99f4a20d24d9c348e91c4db7ab1b749200bded2f839ccbe68f"}, +] +typed-ast = [ + {file = "typed_ast-1.4.3-cp35-cp35m-manylinux1_i686.whl", hash = "sha256:2068531575a125b87a41802130fa7e29f26c09a2833fea68d9a40cf33902eba6"}, + {file = "typed_ast-1.4.3-cp35-cp35m-manylinux1_x86_64.whl", hash = "sha256:c907f561b1e83e93fad565bac5ba9c22d96a54e7ea0267c708bffe863cbe4075"}, + {file = "typed_ast-1.4.3-cp35-cp35m-manylinux2014_aarch64.whl", hash = "sha256:1b3ead4a96c9101bef08f9f7d1217c096f31667617b58de957f690c92378b528"}, + {file = "typed_ast-1.4.3-cp35-cp35m-win32.whl", hash = "sha256:dde816ca9dac1d9c01dd504ea5967821606f02e510438120091b84e852367428"}, + {file = "typed_ast-1.4.3-cp35-cp35m-win_amd64.whl", hash = "sha256:777a26c84bea6cd934422ac2e3b78863a37017618b6e5c08f92ef69853e765d3"}, + {file = "typed_ast-1.4.3-cp36-cp36m-macosx_10_9_x86_64.whl", hash = "sha256:f8afcf15cc511ada719a88e013cec87c11aff7b91f019295eb4530f96fe5ef2f"}, + {file = "typed_ast-1.4.3-cp36-cp36m-manylinux1_i686.whl", hash = "sha256:52b1eb8c83f178ab787f3a4283f68258525f8d70f778a2f6dd54d3b5e5fb4341"}, + {file = "typed_ast-1.4.3-cp36-cp36m-manylinux1_x86_64.whl", hash = "sha256:01ae5f73431d21eead5015997ab41afa53aa1fbe252f9da060be5dad2c730ace"}, + {file = "typed_ast-1.4.3-cp36-cp36m-manylinux2014_aarch64.whl", hash = "sha256:c190f0899e9f9f8b6b7863debfb739abcb21a5c054f911ca3596d12b8a4c4c7f"}, + {file = "typed_ast-1.4.3-cp36-cp36m-win32.whl", hash = "sha256:398e44cd480f4d2b7ee8d98385ca104e35c81525dd98c519acff1b79bdaac363"}, + {file = "typed_ast-1.4.3-cp36-cp36m-win_amd64.whl", hash = "sha256:bff6ad71c81b3bba8fa35f0f1921fb24ff4476235a6e94a26ada2e54370e6da7"}, + {file = "typed_ast-1.4.3-cp37-cp37m-macosx_10_9_x86_64.whl", hash = "sha256:0fb71b8c643187d7492c1f8352f2c15b4c4af3f6338f21681d3681b3dc31a266"}, + {file = "typed_ast-1.4.3-cp37-cp37m-manylinux1_i686.whl", hash = "sha256:760ad187b1041a154f0e4d0f6aae3e40fdb51d6de16e5c99aedadd9246450e9e"}, + {file = "typed_ast-1.4.3-cp37-cp37m-manylinux1_x86_64.whl", hash = "sha256:5feca99c17af94057417d744607b82dd0a664fd5e4ca98061480fd8b14b18d04"}, + {file = "typed_ast-1.4.3-cp37-cp37m-manylinux2014_aarch64.whl", hash = "sha256:95431a26309a21874005845c21118c83991c63ea800dd44843e42a916aec5899"}, + {file = "typed_ast-1.4.3-cp37-cp37m-win32.whl", hash = "sha256:aee0c1256be6c07bd3e1263ff920c325b59849dc95392a05f258bb9b259cf39c"}, + {file = "typed_ast-1.4.3-cp37-cp37m-win_amd64.whl", hash = "sha256:9ad2c92ec681e02baf81fdfa056fe0d818645efa9af1f1cd5fd6f1bd2bdfd805"}, + {file = "typed_ast-1.4.3-cp38-cp38-macosx_10_9_x86_64.whl", hash = "sha256:b36b4f3920103a25e1d5d024d155c504080959582b928e91cb608a65c3a49e1a"}, + {file = "typed_ast-1.4.3-cp38-cp38-manylinux1_i686.whl", hash = "sha256:067a74454df670dcaa4e59349a2e5c81e567d8d65458d480a5b3dfecec08c5ff"}, + {file = "typed_ast-1.4.3-cp38-cp38-manylinux1_x86_64.whl", hash = "sha256:7538e495704e2ccda9b234b82423a4038f324f3a10c43bc088a1636180f11a41"}, + {file = "typed_ast-1.4.3-cp38-cp38-manylinux2014_aarch64.whl", hash = "sha256:af3d4a73793725138d6b334d9d247ce7e5f084d96284ed23f22ee626a7b88e39"}, + {file = "typed_ast-1.4.3-cp38-cp38-win32.whl", hash = "sha256:f2362f3cb0f3172c42938946dbc5b7843c2a28aec307c49100c8b38764eb6927"}, + {file = "typed_ast-1.4.3-cp38-cp38-win_amd64.whl", hash = "sha256:dd4a21253f42b8d2b48410cb31fe501d32f8b9fbeb1f55063ad102fe9c425e40"}, + {file = "typed_ast-1.4.3-cp39-cp39-macosx_10_9_x86_64.whl", hash = "sha256:f328adcfebed9f11301eaedfa48e15bdece9b519fb27e6a8c01aa52a17ec31b3"}, + {file = "typed_ast-1.4.3-cp39-cp39-manylinux1_i686.whl", hash = "sha256:2c726c276d09fc5c414693a2de063f521052d9ea7c240ce553316f70656c84d4"}, + {file = "typed_ast-1.4.3-cp39-cp39-manylinux1_x86_64.whl", hash = "sha256:cae53c389825d3b46fb37538441f75d6aecc4174f615d048321b716df2757fb0"}, + {file = "typed_ast-1.4.3-cp39-cp39-manylinux2014_aarch64.whl", hash = "sha256:b9574c6f03f685070d859e75c7f9eeca02d6933273b5e69572e5ff9d5e3931c3"}, + {file = "typed_ast-1.4.3-cp39-cp39-win32.whl", hash = "sha256:209596a4ec71d990d71d5e0d312ac935d86930e6eecff6ccc7007fe54d703808"}, + {file = "typed_ast-1.4.3-cp39-cp39-win_amd64.whl", hash = "sha256:9c6d1a54552b5330bc657b7ef0eae25d00ba7ffe85d9ea8ae6540d2197a3788c"}, + {file = "typed_ast-1.4.3.tar.gz", hash = "sha256:fb1bbeac803adea29cedd70781399c99138358c26d05fcbd23c13016b7f5ec65"}, +] +typing-extensions = [ + {file = "typing_extensions-3.10.0.0-py2-none-any.whl", hash = "sha256:0ac0f89795dd19de6b97debb0c6af1c70987fd80a2d62d1958f7e56fcc31b497"}, + {file = "typing_extensions-3.10.0.0-py3-none-any.whl", hash = "sha256:779383f6086d90c99ae41cf0ff39aac8a7937a9283ce0a414e5dd782f4c94a84"}, + {file = "typing_extensions-3.10.0.0.tar.gz", hash = "sha256:50b6f157849174217d0656f99dc82fe932884fb250826c18350e159ec6cdf342"}, +] +tzlocal = [ + {file = "tzlocal-2.1-py2.py3-none-any.whl", hash = "sha256:e2cb6c6b5b604af38597403e9852872d7f534962ae2954c7f35efcb1ccacf4a4"}, + {file = "tzlocal-2.1.tar.gz", hash = "sha256:643c97c5294aedc737780a49d9df30889321cbe1204eac2c2ec6134035a92e44"}, +] +virtualenv = [ + {file = "virtualenv-20.4.7-py2.py3-none-any.whl", hash = "sha256:2b0126166ea7c9c3661f5b8e06773d28f83322de7a3ff7d06f0aed18c9de6a76"}, + {file = "virtualenv-20.4.7.tar.gz", hash = "sha256:14fdf849f80dbb29a4eb6caa9875d476ee2a5cf76a5f5415fa2f1606010ab467"}, +] +xmltodict = [ + {file = "xmltodict-0.12.0-py2.py3-none-any.whl", hash = "sha256:8bbcb45cc982f48b2ca8fe7e7827c5d792f217ecf1792626f808bf41c3b86051"}, + {file = "xmltodict-0.12.0.tar.gz", hash = "sha256:50d8c638ed7ecb88d90561beedbf720c9b4e851a9fa6c47ebd64e99d166d8a21"}, +] +yarl = [ + {file = "yarl-1.6.3-cp36-cp36m-macosx_10_14_x86_64.whl", hash = "sha256:0355a701b3998dcd832d0dc47cc5dedf3874f966ac7f870e0f3a6788d802d434"}, + {file = "yarl-1.6.3-cp36-cp36m-manylinux1_i686.whl", hash = "sha256:bafb450deef6861815ed579c7a6113a879a6ef58aed4c3a4be54400ae8871478"}, + {file = "yarl-1.6.3-cp36-cp36m-manylinux2014_aarch64.whl", hash = "sha256:547f7665ad50fa8563150ed079f8e805e63dd85def6674c97efd78eed6c224a6"}, + {file = "yarl-1.6.3-cp36-cp36m-manylinux2014_i686.whl", hash = "sha256:63f90b20ca654b3ecc7a8d62c03ffa46999595f0167d6450fa8383bab252987e"}, + {file = "yarl-1.6.3-cp36-cp36m-manylinux2014_ppc64le.whl", hash = "sha256:97b5bdc450d63c3ba30a127d018b866ea94e65655efaf889ebeabc20f7d12406"}, + {file = "yarl-1.6.3-cp36-cp36m-manylinux2014_s390x.whl", hash = "sha256:d8d07d102f17b68966e2de0e07bfd6e139c7c02ef06d3a0f8d2f0f055e13bb76"}, + {file = "yarl-1.6.3-cp36-cp36m-manylinux2014_x86_64.whl", hash = "sha256:15263c3b0b47968c1d90daa89f21fcc889bb4b1aac5555580d74565de6836366"}, + {file = "yarl-1.6.3-cp36-cp36m-win32.whl", hash = "sha256:b5dfc9a40c198334f4f3f55880ecf910adebdcb2a0b9a9c23c9345faa9185721"}, + {file = "yarl-1.6.3-cp36-cp36m-win_amd64.whl", hash = "sha256:b2e9a456c121e26d13c29251f8267541bd75e6a1ccf9e859179701c36a078643"}, + {file = "yarl-1.6.3-cp37-cp37m-macosx_10_14_x86_64.whl", hash = "sha256:ce3beb46a72d9f2190f9e1027886bfc513702d748047b548b05dab7dfb584d2e"}, + {file = "yarl-1.6.3-cp37-cp37m-manylinux1_i686.whl", hash = "sha256:2ce4c621d21326a4a5500c25031e102af589edb50c09b321049e388b3934eec3"}, + {file = "yarl-1.6.3-cp37-cp37m-manylinux2014_aarch64.whl", hash = "sha256:d26608cf178efb8faa5ff0f2d2e77c208f471c5a3709e577a7b3fd0445703ac8"}, + {file = "yarl-1.6.3-cp37-cp37m-manylinux2014_i686.whl", hash = "sha256:4c5bcfc3ed226bf6419f7a33982fb4b8ec2e45785a0561eb99274ebbf09fdd6a"}, + {file = "yarl-1.6.3-cp37-cp37m-manylinux2014_ppc64le.whl", hash = "sha256:4736eaee5626db8d9cda9eb5282028cc834e2aeb194e0d8b50217d707e98bb5c"}, + {file = "yarl-1.6.3-cp37-cp37m-manylinux2014_s390x.whl", hash = "sha256:68dc568889b1c13f1e4745c96b931cc94fdd0defe92a72c2b8ce01091b22e35f"}, + {file = "yarl-1.6.3-cp37-cp37m-manylinux2014_x86_64.whl", hash = "sha256:7356644cbed76119d0b6bd32ffba704d30d747e0c217109d7979a7bc36c4d970"}, + {file = "yarl-1.6.3-cp37-cp37m-win32.whl", hash = "sha256:00d7ad91b6583602eb9c1d085a2cf281ada267e9a197e8b7cae487dadbfa293e"}, + {file = "yarl-1.6.3-cp37-cp37m-win_amd64.whl", hash = "sha256:69ee97c71fee1f63d04c945f56d5d726483c4762845400a6795a3b75d56b6c50"}, + {file = "yarl-1.6.3-cp38-cp38-macosx_10_14_x86_64.whl", hash = "sha256:e46fba844f4895b36f4c398c5af062a9808d1f26b2999c58909517384d5deda2"}, + {file = "yarl-1.6.3-cp38-cp38-manylinux1_i686.whl", hash = "sha256:31ede6e8c4329fb81c86706ba8f6bf661a924b53ba191b27aa5fcee5714d18ec"}, + {file = "yarl-1.6.3-cp38-cp38-manylinux2014_aarch64.whl", hash = "sha256:fcbb48a93e8699eae920f8d92f7160c03567b421bc17362a9ffbbd706a816f71"}, + {file = "yarl-1.6.3-cp38-cp38-manylinux2014_i686.whl", hash = "sha256:72a660bdd24497e3e84f5519e57a9ee9220b6f3ac4d45056961bf22838ce20cc"}, + {file = "yarl-1.6.3-cp38-cp38-manylinux2014_ppc64le.whl", hash = "sha256:324ba3d3c6fee56e2e0b0d09bf5c73824b9f08234339d2b788af65e60040c959"}, + {file = "yarl-1.6.3-cp38-cp38-manylinux2014_s390x.whl", hash = "sha256:e6b5460dc5ad42ad2b36cca524491dfcaffbfd9c8df50508bddc354e787b8dc2"}, + {file = "yarl-1.6.3-cp38-cp38-manylinux2014_x86_64.whl", hash = "sha256:6d6283d8e0631b617edf0fd726353cb76630b83a089a40933043894e7f6721e2"}, + {file = "yarl-1.6.3-cp38-cp38-win32.whl", hash = "sha256:9ede61b0854e267fd565e7527e2f2eb3ef8858b301319be0604177690e1a3896"}, + {file = "yarl-1.6.3-cp38-cp38-win_amd64.whl", hash = "sha256:f0b059678fd549c66b89bed03efcabb009075bd131c248ecdf087bdb6faba24a"}, + {file = "yarl-1.6.3-cp39-cp39-macosx_10_14_x86_64.whl", hash = "sha256:329412812ecfc94a57cd37c9d547579510a9e83c516bc069470db5f75684629e"}, + {file = "yarl-1.6.3-cp39-cp39-manylinux1_i686.whl", hash = "sha256:c49ff66d479d38ab863c50f7bb27dee97c6627c5fe60697de15529da9c3de724"}, + {file = "yarl-1.6.3-cp39-cp39-manylinux2014_aarch64.whl", hash = "sha256:f040bcc6725c821a4c0665f3aa96a4d0805a7aaf2caf266d256b8ed71b9f041c"}, + {file = "yarl-1.6.3-cp39-cp39-manylinux2014_i686.whl", hash = "sha256:d5c32c82990e4ac4d8150fd7652b972216b204de4e83a122546dce571c1bdf25"}, + {file = "yarl-1.6.3-cp39-cp39-manylinux2014_ppc64le.whl", hash = "sha256:d597767fcd2c3dc49d6eea360c458b65643d1e4dbed91361cf5e36e53c1f8c96"}, + {file = "yarl-1.6.3-cp39-cp39-manylinux2014_s390x.whl", hash = "sha256:8aa3decd5e0e852dc68335abf5478a518b41bf2ab2f330fe44916399efedfae0"}, + {file = "yarl-1.6.3-cp39-cp39-manylinux2014_x86_64.whl", hash = "sha256:73494d5b71099ae8cb8754f1df131c11d433b387efab7b51849e7e1e851f07a4"}, + {file = "yarl-1.6.3-cp39-cp39-win32.whl", hash = "sha256:5b883e458058f8d6099e4420f0cc2567989032b5f34b271c0827de9f1079a424"}, + {file = "yarl-1.6.3-cp39-cp39-win_amd64.whl", hash = "sha256:4953fb0b4fdb7e08b2f3b3be80a00d28c5c8a2056bb066169de00e6501b986b6"}, + {file = "yarl-1.6.3.tar.gz", hash = "sha256:8a9066529240171b68893d60dca86a763eae2139dd42f42106b03cf4b426bf10"}, +] diff --git a/pyproject.toml b/pyproject.toml new file mode 100644 index 0000000..0343e3c --- /dev/null +++ b/pyproject.toml @@ -0,0 +1,19 @@ +[tool.poetry] +name = "volttron-openadr-ven" +version = "0.1.0" +description = "A Volttron agent that acts as a Virtual End Node (VEN) within the OpenADR 2.0b specification." +authors = ["Mark Bonicillo <[email protected]>"] + +[tool.poetry.dependencies] +python = "^3.8" +openleadr = "^0.5.23" +black = "^21.5b2" +mypy = "^0.812" +pre-commit = "^2.13.0" + +[tool.poetry.dev-dependencies] +pytest = "^6.2.4" + +[build-system] +requires = ["poetry-core>=1.0.0"] +build-backend = "poetry.core.masonry.api" diff --git a/volttron_client-0.1.2-py3-none-any.whl b/volttron_client-0.1.2-py3-none-any.whl new file mode 100644 index 0000000..870e627 Binary files /dev/null and b/volttron_client-0.1.2-py3-none-any.whl differ diff --git a/volttron_openadr_ven/agent.py b/volttron_openadr_ven/agent.py new file mode 100644 index 0000000..c79e9de --- /dev/null +++ b/volttron_openadr_ven/agent.py @@ -0,0 +1,321 @@ +# -*- coding: utf-8 -*- {{{ +# vim: set fenc=utf-8 ft=python sw=4 ts=4 sts=4 et: +# +# Copyright 2020, Battelle Memorial Institute. +# +# Licensed under the Apache License, Version 2.0 (the "License"); +# you may not use this file except in compliance with the License. +# You may obtain a copy of the License at +# +# http://www.apache.org/licenses/LICENSE-2.0 +# +# Unless required by applicable law or agreed to in writing, software +# distributed under the License is distributed on an "AS IS" BASIS, +# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. +# See the License for the specific language governing permissions and +# limitations under the License. +# +# This material was prepared as an account of work sponsored by an agency of +# the United States Government. Neither the United States Government nor the +# United States Department of Energy, nor Battelle, nor any of their +# employees, nor any jurisdiction or organization that has cooperated in the +# development of these materials, makes any warranty, express or +# implied, or assumes any legal liability or responsibility for the accuracy, +# completeness, or usefulness or any information, apparatus, product, +# software, or process disclosed, or represents that its use would not infringe +# privately owned rights. Reference herein to any specific commercial product, +# process, or service by trade name, trademark, manufacturer, or otherwise +# does not necessarily constitute or imply its endorsement, recommendation, or +# favoring by the United States Government or any agency thereof, or +# Battelle Memorial Institute. The views and opinions of authors expressed +# herein do not necessarily state or reflect those of the +# United States Government or any agency thereof. +# +# PACIFIC NORTHWEST NATIONAL LABORATORY operated by +# BATTELLE for the UNITED STATES DEPARTMENT OF ENERGY +# under Contract DE-AC05-76RL01830 +# }}} + +import logging +import asyncio +import sys +import openleadr.enums + +from pprint import pformat +from typing import Dict +from datetime import datetime + +from openleadr.enums import OPT +from openleadr.objects import Event +from openleadr import OpenADRClient + +from volttron.platform.agent import utils +from volttron.platform.vip.agent import Agent, Core +from volttron.platform.messaging import topics, headers +from volttron.utils import jsonapi + +utils.setup_logging() +_log = logging.getLogger(__name__) +__version__ = "1.0" + + +class OpenADRVenAgent(Agent): + """ + OpenADR (Automated Demand Response) is a standard for alerting and responding + to the need to adjust electric power consumption in response to fluctuations in grid demand. + OpenADR communications are conducted between Virtual Top Nodes (VTNs) and Virtual End Nodes (VENs). + This agent is a wrapper around OpenLEADR, an open-source implementation of OpenADR 2.0.b for both servers, VTN, and clients, VEN. + This agent creates an instance of OpenLEADR's VEN client, which is used to communicated with a VTN. + """ + + def __init__( + self, + ven_name: str, + vtn_url: str, + debug: bool, + cert: str, + key: str, + fingerprint: str, + show_fingerprint: str, + ca_file: str, + **kwargs, + ) -> None: + super(OpenADRVenAgent, self).__init__(enable_web=True, **kwargs) + self.ven_name = None + self.vtn_url = None + self.debug = None + self.cert = None + self.key = None + self.fingerprint = None + self.show_fingerprint = None + self.ca_file = None + self.ven_id = None + + self.default_config = { + "ven_name": ven_name, + "vtn_url": vtn_url, + "debug": debug, + "cert": cert, + "key": key, + "fingerprint": fingerprint, + "show_fingerprint": show_fingerprint, + "ca_file": ca_file, + } + self.vip.config.set_default("config", self.default_config) + self.vip.config.subscribe( + self._configure, actions=["NEW", "UPDATE"], pattern="config" + ) + + # properties of the this class will be initialized in initialize_config + # self.ven_client is an instance of OpenLeadr's OpenADRClient, an implementation of OpenADR; + # OpenADRClient will be instantiated in initialize_config + self.initialize_config(self.default_config) + + def initialize_config(self, config: Dict) -> None: + """ + Initialize the agent's configuration. Create an OpenADR Client using OpenLeadr. + Configuration parameters (see config for a sample config file): + + str ven_name: The name for this VEN + str vtn_url: The URL of the VTN (Server) to connect to + bool debug: Whether or not to print debugging messages + str cert: The path to a PEM-formatted Certificate file to use + for signing messages. + str key: The path to a PEM-formatted Private Key file to use + for signing messages. + str fingerprint: The fingerprint for the VTN's certificate to + verify incomnig messages + str show_fingerprint: Whether to print your own fingerprint + on startup. Defaults to True. + str ca_file: The path to the PEM-formatted CA file for validating the VTN server's + certificate. + """ + _log.debug("Configuring agent") + self.ven_name = config.get("ven_name") + self.vtn_url = config.get("vtn_url") + self.debug = config.get("debug", False) + self.cert = config.get("cert") + self.key = config.get("key") + self.fingerprint = config.get("fingerprint") + self.show_fingerprint = config.get("show_fingerprint", True) + self.ca_file = config.get("ca_file") + + _log.info("Configuration parameters:") + _log.info(pformat(config)) + + # instantiate and add handlers to the OpenADR Client + self.ven_client = OpenADRClient( + ven_name=self.ven_name, + vtn_url=self.vtn_url, + debug=self.debug, + cert=self.cert, + key=self.key, + show_fingerprint=self.show_fingerprint, + ca_file=self.ca_file, + ) + + # if you want to add more handlers on a specific event, you must create a coroutine + # in this class and then add it as the second input for 'self.ven_client.add_handler(<some event>, <coroutine>)' + self.ven_client.add_handler("on_event", self.handle_event) + + def _configure(self, contents: Dict) -> None: + """The agent's config may have changed. Re-initialize it.""" + config = self.default_config.copy() + config.update(contents) + self.initialize_config(config) + + # ***************** Methods for Managing the Agent on Volttron ******************** + + @Core.receiver("onstart") + def onstart(self) -> None: + """The agent has started.""" + _log.debug("Starting agent") + + # Running the ven_client in the Python AsyncIO Event Loop + loop = asyncio.get_event_loop() + loop.create_task(self.ven_client.run()) + loop.run_forever() + + # # TODO: Identify actions needed to be done before shutdown + # @Core.receiver("onstop") + # def onstop(self, sender, **kwargs): + # """ + # This method is called when the Agent is about to shutdown, but before it disconnects from + # the message bus. + # """ + # pass + # + # # TODO: Identify what, if at all, RPC methods should be available to other Agents + # @RPC.export + # def rpc_method(self, arg1, arg2, kwarg1=None, kwarg2=None): + # """ + # RPC method + # + # May be called from another agent via self.vip.rpc.call + # """ + # pass + + # ***************** Methods for Servicing VTN Requests ******************** + + async def handle_event(self, event: openleadr.objects.Event) -> openleadr.enums.OPT: + """ + Publish event to the Volttron message bus. Return OPT response. + This coroutine will be called when there is an event to be handled. + + :param event: Event + For more info on shape of 'event', see https://openleadr.org/docs/client.html#dealing-with-events and https://openleadr.org/docs/api/openleadr.html#openleadr.objects.Event + Example of event: + { + 'event_id': '123786-129831', + 'active_period': {'dtstart': datetime.datetime(2020, 1, 1, 12, 0, 0, tzinfo=timezone.utc), + 'duration': datetime.timedelta(minutes=30)} + 'event_signals': [{'signal_name': 'simple', + 'signal_type': 'level', + 'intervals': [{'dtstart': datetime.datetime(2020, 1, 1, 12, 0, 0, tzinfo=timezone.utc), + 'duration': datetime.timedelta(minutes=10), + 'signal_payload': 1}, + {'dtstart': datetime.datetime(2020, 1, 1, 12, 10, 0, tzinfo=timezone.utc), + 'duration': datetime.timedelta(minutes=10), + 'signal_payload': 0}, + {'dtstart': datetime.datetime(2020, 1, 1, 12, 20, 0, tzinfo=timezone.utc), + 'duration': datetime.timedelta(minutes=10), + 'signal_payload': 1}], + 'targets': [{'resource_id': 'Device001'}], + 'targets_by_type': {'resource_id': ['Device001']} + } + :return: openleadr.enums.OPT + """ + _log.info(f"Event: {pformat(event)}") + self.publish_event(event) + return OPT.OPT_IN + + # ***************** VOLTTRON Pub/Sub Requests ******************** + def publish_event(self, event: openleadr.objects.Event) -> None: + """Publish an event to the Volttron message bus. When an event is created/updated, it is published to the VOLTTRON bus with a topic that includes 'openadr/event_update'. + :param event: + :return: None + """ + # OADR rule 6: If testEvent is present and != "false", handle the event as a test event. + if event.event_descriptor.test_event != "false": + _log.debug(f"Suppressing publication of test event {event}") + return + + _log.debug("Publishing event {}".format(event)) + self.vip.pubsub.publish( + peer="pubsub", + topic=f"{topics.OPENADR_EVENT}/{self.ven_id}", + headers={ + headers.TIMESTAMP: utils.format_timestamp(utils.get_aware_utc_now()) + }, + message=self._json_object(event.__dict__), + ) + return + + # ***************** Helper methods ******************** + def _json_object(self, obj: Dict): + """Ensure that an object is valid JSON by dumping it with json_converter and then reloading it.""" + obj_string = jsonapi.dumps( + obj, default=lambda x: x.__str__() if isinstance(x, datetime) else None + ) + return jsonapi.loads(obj_string) + + +def main(): + """Main method called to start the agent.""" + utils.vip_main(ven_agent, version=__version__) + + +if __name__ == "__main__": + try: + sys.exit(main()) + except KeyboardInterrupt: + pass + + +def ven_agent(config_path: str, **kwargs) -> OpenADRVenAgent: + """ + Parse the OpenADRVenAgent configuration file and return an instance of + the agent that has been created using that configuration. + See initialize_config() method documentation for a description of each configurable parameter. + + :param config_path: (str) Path to a configuration file. + + :returns: OpenADRVenAgent instance + """ + try: + config = utils.load_config(config_path) + except Exception as err: + _log.error("Error loading configuration: {}".format(err)) + config = {} + + if not config: + raise Exception("Configuration cannot be empty.") + + ven_name = config.get("ven_name") + if not ven_name: + raise Exception("ven_name is required.") + vtn_url = config.get("vtn_url") + if not vtn_url: + raise Exception( + "vtn_url is required. Ensure vtn_url is given a URL to the VTN." + ) + + # below are optional configurations + debug = config.get("debug") + cert = config.get("cert") + key = config.get("key") + fingerprint = config.get("fingerprint") + show_fingerprint = config.get("show_fingerprint") + ca_file = config.get("ca_file") + + return OpenADRVenAgent( + ven_name, + vtn_url, + debug, + cert, + key, + fingerprint, + show_fingerprint, + ca_file, + **kwargs, + ) diff --git a/volttron_utils-0.1.1-py3-none-any.whl b/volttron_utils-0.1.1-py3-none-any.whl new file mode 100644 index 0000000..6512925 Binary files /dev/null and b/volttron_utils-0.1.1-py3-none-any.whl differ

Create Skeleton Agent Structure This skeleton will have volttron agent properties as well as being able to start the async loop and be able to call the openleadr source code https://pypi.org/project/openleadr/

Additional Notes: * This agent will take the following dependencies: * https://github.com/VOLTTRON/volttron-client/tree/main/volttron/platform * https://github.com/VOLTTRON/volttron-utils * This agent is a pared-down version of OpenADRAgent in volttron core; see https://github.com/VOLTTRON/volttron/blob/develop/services/core/OpenADRVenAgent/openadrven/agent.py#L414 * Agent should have the following methods * publish_event * process_event * process_report * Agent should ideally be a gevent-based agent that can make asyn calls; see example from openleader: https://openleadr.org/docs/#getting-started * Agent should be packaged and able to be installed via wheel on a volttron platform using `vctl install`

2021-05-30T19:53:10

0.0

[]

octue/django-gcp

octue__django-gcp-40

99d1932865eb9bb8ea765bbab57c5c6e3b9416e4

diff --git a/.devcontainer/devcontainer.json b/.devcontainer/devcontainer.json index 0bfdd36..89c9842 100644 --- a/.devcontainer/devcontainer.json +++ b/.devcontainer/devcontainer.json @@ -71,7 +71,8 @@ "thebarkman.vscode-djaneiro", "trond-snekvik.simple-rst", "ms-azuretools.vscode-docker", - "ryanluker.vscode-coverage-gutters" + "ryanluker.vscode-coverage-gutters", + "ms-python.black-formatter" ] } }, diff --git a/.pre-commit-config.yaml b/.pre-commit-config.yaml index eada270..395bc36 100644 --- a/.pre-commit-config.yaml +++ b/.pre-commit-config.yaml @@ -20,7 +20,7 @@ repos: - id: check-json5 - repo: https://github.com/psf/black - rev: 22.3.0 + rev: 22.10.0 hooks: - id: black args: ['--line-length', '120'] diff --git a/django_gcp/__init__.py b/django_gcp/__init__.py index 3dd48a2..d73c30f 100644 --- a/django_gcp/__init__.py +++ b/django_gcp/__init__.py @@ -1,1 +1,1 @@ -default_app_config = "django_gcp.apps.DjangoGcpAppConfig" +default_app_config = "django_gcp.apps.DjangoGCPAppConfig" diff --git a/django_gcp/storage/fields.py b/django_gcp/storage/fields.py index 76dd6f0..7aec214 100644 --- a/django_gcp/storage/fields.py +++ b/django_gcp/storage/fields.py @@ -59,8 +59,8 @@ class BlobField(models.JSONField): We do this to support a clear and maintainable implementation. :param ingress_to: A string defining the path within the bucket to which direct uploads - will be ingressed, if temporary_ingress=True. These uploaded files will be moved to their - ultimate path in the store on save of the model. + will be ingressed. These uploaded files will be moved to their ultimate path in the store on save of the model. + :param accept_mimetype: A string passed to widgets, suitable for use in the `accept` attribute of an html filepicker. This will allow you to narrow down, eg to `image/*` or an even more specific mimetype. No validation is done at the field level that objects actually are of this @@ -77,7 +77,7 @@ class BlobField(models.JSONField): :param max_size_bytes: The maximum size in bytes of files that can be uploaded. - :overwrite_mode: One of `OVERWRITE_MODES` determining the circumstances under which overwrite + :param overwrite_mode: One of `OVERWRITE_MODES` determining the circumstances under which overwrite is allowed. overwrite mode behaves as follows: - never: Never allows overwrite - update: Only when updating an object @@ -86,6 +86,10 @@ class BlobField(models.JSONField): - add-versioned: Only when adding an object to a versioned bucket (again, for completeness) - add-update: Always allow (ie when adding or updating an object) - add-update-versioned: When adding or updating an object to a versioned bucket + + :param on_change: A callable that will be executed on change of the field value. This will be called + on commit of the transaction (ie once any file upload is ingressed to its final location) and allows you, + for example, to dispatch a worker task to further process the uploaded blob. """ description = _("A GCP Cloud Storage object") @@ -99,6 +103,7 @@ def __init__( accept_mimetype=DEFAULT_ACCEPT_MIMETYPE, max_size_bytes=DEFAULT_MAX_SIZE_BYTES, overwrite_mode=DEFAULT_OVERWRITE_MODE, + on_change=None, **kwargs, ): self._versioning_enabled = None @@ -112,6 +117,7 @@ def __init__( self.store_key = store_key self.accept_mimetype = accept_mimetype self.max_size_bytes = max_size_bytes + self.on_change = on_change kwargs["default"] = kwargs.pop("default", None) kwargs["help_text"] = kwargs.pop("help_text", "GCP cloud storage object") @@ -144,6 +150,7 @@ def check(self, **kwargs): *self._check_get_destination_path(), *self._check_ingress_to(), *self._check_overwrite_mode(), + *self._check_on_change(), ] def deconstruct(self): @@ -153,6 +160,7 @@ def deconstruct(self): kwargs["accept_mimetype"] = self.accept_mimetype kwargs["overwrite_mode"] = self.overwrite_mode kwargs["get_destination_path"] = self.get_destination_path + kwargs["on_change"] = self.on_change return name, path, args, kwargs def formfield(self, **kwargs): @@ -172,7 +180,11 @@ def formfield(self, **kwargs): @property def override_blobfield_value(self): """Shortcut to access the GCP_STORAGE_OVERRIDE_BLOBFIELD_VALUE setting""" - return getattr(settings, "GCP_STORAGE_OVERRIDE_BLOBFIELD_VALUE", DEFAULT_OVERRIDE_BLOBFIELD_VALUE) + return getattr( + settings, + "GCP_STORAGE_OVERRIDE_BLOBFIELD_VALUE", + DEFAULT_OVERRIDE_BLOBFIELD_VALUE, + ) def pre_save(self, model_instance, add): """Return field's value just before saving.""" @@ -184,7 +196,12 @@ def pre_save(self, model_instance, add): # explicitly for the purpose of data migration and manipulation. You should never allow an untrusted # client to set paths directly, because knowing the path of a pre-existing object allows you to assume # access to it. Tip: You can use django's override_settings context manager to set this temporarily. - if getattr(settings, "GCP_STORAGE_OVERRIDE_BLOBFIELD_VALUE", DEFAULT_OVERRIDE_BLOBFIELD_VALUE): + # Note that you'll have to execute any on_changed + if getattr( + settings, + "GCP_STORAGE_OVERRIDE_BLOBFIELD_VALUE", + DEFAULT_OVERRIDE_BLOBFIELD_VALUE, + ): logger.warning( "Overriding %s value to %s", self.__class__.__name__, @@ -206,8 +223,15 @@ def pre_save(self, model_instance, add): elif adding_blank or updating_valid_to_blank: new_value = None - elif adding_valid or updating_blank_to_valid or updating_valid_to_valid: + # Trigger the on_change callback at the end of the commit when we know the + # database transaction will work + def on_commit_blank(): + if self.on_change is not None: + self.on_change(new_value, instance=model_instance) + + transaction.on_commit(on_commit_blank) + elif adding_valid or updating_blank_to_valid or updating_valid_to_valid: new_value = {} new_value["path"], allow_overwrite = self.get_destination_path( instance=model_instance, @@ -228,8 +252,8 @@ def pre_save(self, model_instance, add): # capture the dual edge cases of the file not moving correctly, and the database # row not saving (eg due to validation errors in other model fields). # https://stackoverflow.com/questions/33180727/trigering-post-save-signal-only-after-transaction-has-completed - transaction.on_commit( - lambda: copy_blob( + def on_commit_valid(): + copy_blob( self.storage.bucket, value["_tmp_path"], self.storage.bucket, @@ -238,9 +262,15 @@ def pre_save(self, model_instance, add): overwrite=allow_overwrite, attributes=value.get("attributes", None), ) - ) + if self.on_change is not None: + self.on_change(new_value, instance=model_instance) + + transaction.on_commit(on_commit_valid) + logger.info( - "Registered move of %s to %s to happen on transaction commit", value["_tmp_path"], new_value["path"] + "Registered move of %s to %s to happen on transaction commit", + value["_tmp_path"], + new_value["path"], ) else: @@ -407,6 +437,17 @@ def _check_get_destination_path(self): ] return [] + def _check_on_change(self): + if self.on_change is not None and not callable(self.on_change): + return [ + checks.Error( + f"'on_change' argument in {self.__class__.__name__} must be or a callable function, or None", + obj=self, + id="fields.E201", + ) + ] + return [] + def _get_allow_overwrite(self, add): """Return a boolean determining if overwrite should be allowed for this operation""" mode_map = { diff --git a/django_gcp/storage/forms.py b/django_gcp/storage/forms.py index 6748e67..a16a0dc 100644 --- a/django_gcp/storage/forms.py +++ b/django_gcp/storage/forms.py @@ -12,59 +12,3 @@ class CloudObjectFormField(JSONField): empty_value = None widget = CloudObjectWidget - - -# from django.forms import FileField -# from django.core.files.uploadedfile import UploadedFile -# from django.core.exceptions import ValidationError - -# from .direct_upload_widgets import DirectUploadWidget -# class DirectUploadedFile(UploadedFile): -# """Represents a file that was uploaded direct to bucket""" - - -# class DirectUploadFormField(FileField): -# """Overrides the usual forms.FileField with a widget that handles direct uploads""" - -# widget = DirectUploadWidget - -# def to_python(self, data): -# """Override the FileField to_python method to accept data as a string file name""" - -# if data in self.empty_values: -# return None - -# # A file was attached as an indirect upload (the normal way) -# if isinstance(data, UploadedFile): -# try: -# file_name = data.name -# file_size = data.size -# except AttributeError: -# # pylint: disable=raise-missing-from -# raise ValidationError(self.error_messages["invalid"], code="invalid") -# # pylint: enable=raise-missing-from - -# if not self.allow_empty_file and not file_size: -# raise ValidationError(self.error_messages["empty"], code="empty") - -# # A file was uploaded direct-to-bucket, data only contains its name -# elif isinstance(data, str): -# file_name = data -# file_size = None -# data = DirectUploadedFile(name=file_name, size=None) -# if not self.allow_empty_file: -# logger.warning( -# "`allow_empty_file` is set but file has been directly uploaded. Cannot validate emptiness of file." -# ) - -# else: -# raise ValueError(f"Unknown data of class {data.__class__}: {data}") -# print("FILE NAME", file_name) -# if self.max_length is not None and len(file_name) > self.max_length: -# params = {"max": self.max_length, "length": len(file_name)} -# raise ValidationError(self.error_messages["max_length"], code="max_length", params=params) - -# if not file_name: -# raise ValidationError(self.error_messages["invalid"], code="invalid") - -# return data diff --git a/django_gcp/storage/gcloud.py b/django_gcp/storage/gcloud.py index 251a153..108586b 100644 --- a/django_gcp/storage/gcloud.py +++ b/django_gcp/storage/gcloud.py @@ -1,6 +1,7 @@ import logging import mimetypes from tempfile import SpooledTemporaryFile +from django.conf import settings from django.core.exceptions import SuspiciousOperation from django.core.files.base import File from django.core.files.storage import Storage @@ -12,7 +13,7 @@ from .compress import CompressedFileMixin, CompressStorageMixin from .settings import StorageSettings -from .utils import clean_name, get_available_overwrite_name, safe_join, setting, to_bytes +from .utils import clean_name, get_available_overwrite_name, safe_join, to_bytes CONTENT_ENCODING = "content_encoding" @@ -44,7 +45,7 @@ def _get_file(self): self._file = SpooledTemporaryFile( max_size=self._storage.settings.max_memory_size, suffix=".GSStorageFile", - dir=setting("FILE_UPLOAD_TEMP_DIR"), + dir=getattr(settings, "FILE_UPLOAD_TEMP_DIR"), ) if "r" in self._mode: self._is_dirty = False @@ -286,7 +287,7 @@ def get_modified_time(self, name): name = self._normalize_name(clean_name(name)) blob = self._get_blob(name) updated = blob.updated - return updated if setting("USE_TZ") else timezone.make_naive(updated) + return updated if getattr(settings, "USE_TZ") else timezone.make_naive(updated) def get_created_time(self, name): """ @@ -296,7 +297,7 @@ def get_created_time(self, name): name = self._normalize_name(clean_name(name)) blob = self._get_blob(name) created = blob.time_created - return created if setting("USE_TZ") else timezone.make_naive(created) + return created if getattr(settings, "USE_TZ") else timezone.make_naive(created) def url(self, name): """ diff --git a/django_gcp/storage/utils.py b/django_gcp/storage/utils.py index 8420b54..e36d141 100644 --- a/django_gcp/storage/utils.py +++ b/django_gcp/storage/utils.py @@ -1,6 +1,5 @@ import os import posixpath -from django.conf import settings from django.core.exceptions import SuspiciousFileOperation from django.utils.encoding import force_bytes @@ -13,37 +12,24 @@ def to_bytes(content): return force_bytes(content) -def setting(name, default=None): - """ - Helper function to get a Django setting by name. If setting doesn't exists - it will return a default. - - :param name: Name of setting - :type name: str - :param default: Value if setting is unfound - :returns: Setting's value - """ - return getattr(settings, name, default) - - def clean_name(name): """ Cleans the name so that Windows style paths work """ # Normalize Windows style paths - clean_name = posixpath.normpath(name).replace("\\", "/") + cleaned = posixpath.normpath(name).replace("\\", "/") # os.path.normpath() can strip trailing slashes so we implement # a workaround here. - if name.endswith("/") and not clean_name.endswith("/"): + if name.endswith("/") and not cleaned.endswith("/"): # Add a trailing slash as it was stripped. - clean_name = clean_name + "/" + cleaned = cleaned + "/" # Given an empty string, os.path.normpath() will return ., which we don't want - if clean_name == ".": - clean_name = "" + if cleaned == ".": + cleaned = "" - return clean_name + return cleaned def safe_join(base, *paths): @@ -77,12 +63,13 @@ def safe_join(base, *paths): # the base path is /. base_path_len = len(base_path) if not final_path.startswith(base_path) or final_path[base_path_len] != "/": - raise ValueError("the joined path is located outside of the base path" " component") + raise ValueError("the joined path is located outside of the base path component") return final_path.lstrip("/") def get_available_overwrite_name(name, max_length): + """Truncate a filename to obey max_length limit""" if max_length is None or len(name) <= max_length: return name @@ -94,8 +81,6 @@ def get_available_overwrite_name(name, max_length): file_root = file_root[:-truncation] if not file_root: raise SuspiciousFileOperation( - 'Storage tried to truncate away entire filename "%s". ' - "Please make sure that the corresponding file field " - 'allows sufficient "max_length".' % name + f'Storage tried to truncate away entire filename "{name}". Please make sure that the corresponding file field allows sufficient "max_length".' ) - return os.path.join(dir_name, "{}{}".format(file_root, file_ext)) + return os.path.join(dir_name, f"{file_root}{file_ext}") diff --git a/pyproject.toml b/pyproject.toml index 317cb20..f03918f 100644 --- a/pyproject.toml +++ b/pyproject.toml @@ -1,6 +1,6 @@ [tool.poetry] name = "django-gcp" -version = "0.9.3" +version = "0.10.0" description = "Utilities to run Django on Google Cloud Platform" authors = ["Tom Clark"] license = "MIT" diff --git a/tox.ini b/tox.ini index 85e4686..4842010 100644 --- a/tox.ini +++ b/tox.ini @@ -5,7 +5,7 @@ isolated_build = True [testenv] whitelist_externals = poetry setenv = - DJANGO_SETTINGS_MODULE=tests.server.settings + DJANGO_SETTINGS_MODULE=tests.server.settings_ci PYTHONPATH = {toxinidir}:{toxinidir}/django_gcp sqlite: DATABASE_ENGINE=sqlite postgres: DATABASE_ENGINE=postgres

Incorrect Default App Config ## Bug report ### What is the current behavior? The `default_app_config` value in `__init__.py` is incorrecly set to `django_gcp.apps.DjangoGcpAppConfig` when the class name is `DjangoGCPAppConfig`. This results in the following (largely confusing) error: ``` Exception in thread django-main-thread: Traceback (most recent call last): File "/usr/local/lib/python3.9/threading.py", line 980, in _bootstrap_inner self.run() File "/usr/local/lib/python3.9/threading.py", line 917, in run self._target(*self._args, **self._kwargs) File "/usr/local/lib/python3.9/site-packages/django/utils/autoreload.py", line 64, in wrapper fn(*args, **kwargs) File "/usr/local/lib/python3.9/site-packages/django/core/management/commands/runserver.py", line 110, in inner_run autoreload.raise_last_exception() File "/usr/local/lib/python3.9/site-packages/django/utils/autoreload.py", line 87, in raise_last_exception raise _exception[1] File "/usr/local/lib/python3.9/site-packages/django/core/management/__init__.py", line 375, in execute autoreload.check_errors(django.setup)() File "/usr/local/lib/python3.9/site-packages/django/utils/autoreload.py", line 64, in wrapper fn(*args, **kwargs) File "/usr/local/lib/python3.9/site-packages/django/__init__.py", line 24, in setup apps.populate(settings.INSTALLED_APPS) File "/usr/local/lib/python3.9/site-packages/django/apps/registry.py", line 91, in populate app_config = AppConfig.create(entry) File "/usr/local/lib/python3.9/site-packages/django/apps/config.py", line 228, in create if not issubclass(app_config_class, AppConfig): TypeError: issubclass() arg 1 must be a class ``` ### Workaround Set the full path in your django installed apps: ```py INSTALLED_APPS = [ # ... "django_gcp.apps.DjangoGCPAppConfig", # ... ] ``` ### Proposed Solution Fix the naming.

2023-06-12T16:25:50

0.0

[]

scikit-hep/cuda-histogram

scikit-hep__cuda-histogram-13

3a747b6fb7cda2e5d1ffe41016b95fab58e78255

diff --git a/README.md b/README.md index 6aa53dc..5ff37b5 100644 --- a/README.md +++ b/README.md @@ -13,10 +13,10 @@ [![Conda latest release][conda-version]][conda-link] [![LICENSE][license-badge]][license-link] [![Scikit-HEP][sk-badge]][sk-link] -`cuda-histogram` is a histogram filling, transformation, and plotting package -for GPUs. The package follows [UHI](https://uhi.readthedocs.io) and keeps its -API similar to [boost-histogram](https://github.com/scikit-hep/boost-histogram) -and [hist](https://github.com/scikit-hep/hist). +`cuda-histogram` is a histogram filling package for GPUs. The package follows +[UHI](https://uhi.readthedocs.io) and keeps its API similar to +[boost-histogram](https://github.com/scikit-hep/boost-histogram) and +[hist](https://github.com/scikit-hep/hist). The package is under active development at the moment. diff --git a/docs/api/cuda_histogram.plot.rst b/docs/api/cuda_histogram.plot.rst deleted file mode 100644 index bc175e5..0000000 --- a/docs/api/cuda_histogram.plot.rst +++ /dev/null @@ -1,8 +0,0 @@ -cuda_histogram.plot module -========================== - -.. automodule:: cuda_histogram.plot - :members: - :undoc-members: - :show-inheritance: - :private-members: diff --git a/docs/api/cuda_histogram.rst b/docs/api/cuda_histogram.rst index 65f032a..dd7d1ad 100644 --- a/docs/api/cuda_histogram.rst +++ b/docs/api/cuda_histogram.rst @@ -8,4 +8,3 @@ Submodules :maxdepth: 3 cuda_histogram.hist_tools - cuda_histogram.plot diff --git a/noxfile.py b/noxfile.py index a3108fe..8936386 100644 --- a/noxfile.py +++ b/noxfile.py @@ -30,7 +30,7 @@ def tests(session: nox.Session) -> None: """ Run the unit and regular tests. """ - session.install(".[test,dev,plot,gpu]") + session.install(".[test,dev]") session.run("pytest", *session.posargs) diff --git a/pyproject.toml b/pyproject.toml index 5b91e5b..35cb87e 100644 --- a/pyproject.toml +++ b/pyproject.toml @@ -41,15 +41,6 @@ dependencies = [ "boost-histogram", ] [project.optional-dependencies] -plot = [ - "matplotlib>=3", - "mplhep>=0.1.18", - "bokeh!=3.0.*,>=2.4.2", -] -test = [ - "uproot>=5.3.0,!=5.3.3,!=5.3.4,!=5.3.5", - "coffea>=2024.6.0" -] dev = [ "pre-commit", "pytest>=6", diff --git a/src/cuda_histogram/__init__.py b/src/cuda_histogram/__init__.py index e0e1068..768e57f 100644 --- a/src/cuda_histogram/__init__.py +++ b/src/cuda_histogram/__init__.py @@ -1,5 +1,5 @@ """ -`cuda-histogram` is a histogram filling, transformation, and plotting package for GPUs. +`cuda-histogram` is a histogram filling package for GPUs. The package follows `UHI <https://uhi.readthedocs.io>`__ and keeps its API similar to `boost-histogram <https://github.com/scikit-hep/boost-histogram>`__ @@ -9,30 +9,14 @@ from __future__ import annotations from cuda_histogram.hist_tools import Bin, Cat, Hist, Interval, StringBin -from cuda_histogram.plot import ( - clopper_pearson_interval, - normal_interval, - plot1d, - plot2d, - plotgrid, - plotratio, - poisson_interval, -) from ._version import version as __version__ -__all__ = [ +__all__: list[str] = [ "Hist", "Bin", "Interval", "Cat", "StringBin", - "poisson_interval", - "clopper_pearson_interval", - "normal_interval", - "plot1d", - "plotratio", - "plot2d", - "plotgrid", "__version__", ] diff --git a/src/cuda_histogram/plot.py b/src/cuda_histogram/plot.py deleted file mode 100644 index 6184817..0000000 --- a/src/cuda_histogram/plot.py +++ /dev/null @@ -1,1065 +0,0 @@ -from __future__ import annotations - -import numbers -import warnings - -import numpy as np -import scipy.stats - -from .hist_tools import DenseAxis, Interval, SparseAxis, overflow_behavior - -# Plotting is always terrible -# Let's try our best to follow matplotlib idioms -# https://matplotlib.org/tutorials/introductory/usage.html#coding-styles - -_coverage1sd = scipy.stats.norm.cdf(1) - scipy.stats.norm.cdf(-1) - - -def poisson_interval(sumw, sumw2, coverage=_coverage1sd): - """Frequentist coverage interval for Poisson-distributed observations - - Parameters - ---------- - sumw : np.ndarray - Sum of weights vector - sumw2 : np.ndarray - Sum weights squared vector - coverage : float, optional - Central coverage interval, defaults to 68% - - Calculates the so-called 'Garwood' interval, - c.f. https://www.ine.pt/revstat/pdf/rs120203.pdf or - http://ms.mcmaster.ca/peter/s743/poissonalpha.html - For weighted data, this approximates the observed count by ``sumw**2/sumw2``, which - effectively scales the unweighted poisson interval by the average weight. - This may not be the optimal solution: see https://arxiv.org/pdf/1309.1287.pdf for a proper treatment. - When a bin is zero, the scale of the nearest nonzero bin is substituted to scale the nominal upper bound. - If all bins zero, a warning is generated and interval is set to ``sumw``. - """ - scale = np.empty_like(sumw) - scale[sumw != 0] = sumw2[sumw != 0] / sumw[sumw != 0] - if np.sum(sumw == 0) > 0: - missing = np.where(sumw == 0) - available = np.nonzero(sumw) - if len(available[0]) == 0: - warnings.warn( - "All sumw are zero! Cannot compute meaningful error bars", - RuntimeWarning, - ) - return np.vstack([sumw, sumw]) - nearest = sum( - [np.subtract.outer(d, d0) ** 2 for d, d0 in zip(available, missing)] - ).argmin(axis=0) - argnearest = tuple(dim[nearest] for dim in available) - scale[missing] = scale[argnearest] - counts = sumw / scale - lo = scale * scipy.stats.chi2.ppf((1 - coverage) / 2, 2 * counts) / 2.0 - hi = scale * scipy.stats.chi2.ppf((1 + coverage) / 2, 2 * (counts + 1)) / 2.0 - interval = np.array([lo, hi]) - interval[interval == np.nan] = 0.0 # chi2.ppf produces nan for counts=0 - return interval - - -def clopper_pearson_interval(num, denom, coverage=_coverage1sd): - """Compute Clopper-Pearson coverage interval for a binomial distribution - - Parameters - ---------- - num : np.ndarray - Numerator, or number of successes, vectorized - denom : np.ndarray - Denominator or number of trials, vectorized - coverage : float, optional - Central coverage interval, defaults to 68% - - c.f. http://en.wikipedia.org/wiki/Binomial_proportion_confidence_interval - """ - if np.any(num > denom): - raise ValueError( - "Found numerator larger than denominator while calculating binomial uncertainty" - ) - lo = scipy.stats.beta.ppf((1 - coverage) / 2, num, denom - num + 1) - hi = scipy.stats.beta.ppf((1 + coverage) / 2, num + 1, denom - num) - interval = np.array([lo, hi]) - interval[:, num == 0.0] = 0.0 - interval[1, num == denom] = 1.0 - return interval - - -def normal_interval(pw, tw, pw2, tw2, coverage=_coverage1sd): - """Compute errors based on the expansion of pass/(pass + fail), possibly weighted - - Parameters - ---------- - pw : np.ndarray - Numerator, or number of (weighted) successes, vectorized - tw : np.ndarray - Denominator or number of (weighted) trials, vectorized - pw2 : np.ndarray - Numerator sum of weights squared, vectorized - tw2 : np.ndarray - Denominator sum of weights squared, vectorized - coverage : float, optional - Central coverage interval, defaults to 68% - - c.f. https://root.cern.ch/doc/master/TEfficiency_8cxx_source.html#l02515 - """ - - eff = pw / tw - - variance = (pw2 * (1 - 2 * eff) + tw2 * eff**2) / (tw**2) - sigma = np.sqrt(variance) - - prob = 0.5 * (1 - coverage) - delta = np.zeros_like(sigma) - delta[sigma != 0] = scipy.stats.norm.ppf(prob, scale=sigma[sigma != 0]) - - lo = eff - np.minimum(eff + delta, np.ones_like(eff)) - hi = np.maximum(eff - delta, np.zeros_like(eff)) - eff - - return np.array([lo, hi]) - - -def plot1d( - hist, - ax=None, - clear=True, - overlay=None, - stack=False, - overflow="none", - line_opts=None, - fill_opts=None, - error_opts=None, - legend_opts=None, - overlay_overflow="none", - density=False, - binwnorm=None, - order=None, -): - """Create a 1D plot from a 1D or 2D `Hist` object - - Parameters - ---------- - hist : Hist - Histogram with maximum of two dimensions - ax : matplotlib.axes.Axes, optional - Axes object (if None, one is created) - clear : bool, optional - Whether to clear Axes before drawing (if passed); if False, this function will skip drawing the legend - overlay : str, optional - In the case that ``hist`` is 2D, specify the axis of hist to overlay (remaining axis will be x axis) - stack : bool, optional - Whether to stack or overlay non-axis dimension (if it exists) - order : list, optional - How to order when stacking. Take a list of identifiers. - overflow : str, optional - If overflow behavior is not 'none', extra bins will be drawn on either end of the nominal - axis range, to represent the contents of the overflow bins. See `Hist.sum` documentation - for a description of the options. - line_opts : dict, optional - A dictionary of options to pass to the matplotlib - `ax.step <https://matplotlib.org/3.1.1/api/_as_gen/matplotlib.pyplot.step.html>`_ call - internal to this function. Leave blank for defaults. - fill_opts : dict, optional - A dictionary of options to pass to the matplotlib - `ax.fill_between <https://matplotlib.org/3.1.1/api/_as_gen/matplotlib.axes.Axes.fill_between.html>`_ call - internal to this function. Leave blank for defaults. - error_opts : dict, optional - A dictionary of options to pass to the matplotlib - `ax.errorbar <https://matplotlib.org/3.1.1/api/_as_gen/matplotlib.pyplot.errorbar.html>`_ call - internal to this function. Leave blank for defaults. Some special options are interpreted by - this function and not passed to matplotlib: 'emarker' (default: '') specifies the marker type - to place at cap of the errorbar. - legend_opts : dict, optional - A dictionary of options to pass to the matplotlib - `ax.legend <https://matplotlib.org/3.1.1/api/_as_gen/matplotlib.pyplot.legend.html>`_ call - internal to this function. Leave blank for defaults. - overlay_overflow : str, optional - If overflow behavior is not 'none', extra bins in the overlay axis will be overlaid or stacked, - to represent the contents of the overflow bins. See `Hist.sum` documentation for a description of the options. - density : bool, optional - If true, convert sum weights to probability density (i.e. integrates to 1 over domain of axis) - (Note: this option conflicts with ``binwnorm``) - binwnorm : float, optional - If true, convert sum weights to bin-width-normalized, with unit equal to supplied value (usually you want to specify 1.) - - - Returns - ------- - ax : matplotlib.axes.Axes - A matplotlib `Axes <https://matplotlib.org/3.1.1/api/axes_api.html>`_ object - """ - import matplotlib.pyplot as plt - import mplhep as hep - - if legend_opts is None: - legend_opts = {} - - if ax is None: - ax = plt.gca() - else: - if not isinstance(ax, plt.Axes): - raise ValueError("ax must be a matplotlib Axes object") - if clear: - ax.clear() - if hist.dim() > 2: - raise ValueError( - "plot1d() can only support up to two dimensions (one for axis, one to stack or overlay)" - ) - if overlay is None and hist.sparse_dim() == 1 and hist.dense_dim() == 1: - overlay = hist.sparse_axes()[0].name - elif overlay is None and hist.dim() > 1: - raise ValueError( - "plot1d() can only support one dimension without an overlay axis chosen" - ) - if density and binwnorm is not None: - raise ValueError("Cannot use density and binwnorm at the same time!") - if binwnorm is not None and not isinstance(binwnorm, numbers.Number): - raise ValueError( - f"Bin width normalization not a number, but a {binwnorm.__class__!r}" - ) - if line_opts is None and fill_opts is None and error_opts is None: - if stack: - fill_opts = {} - else: - line_opts = {} - error_opts = {} - - axis = hist.axes()[0] - if overlay is not None: - overlay = hist.axis(overlay) - if axis == overlay: - axis = hist.axes()[1] - if isinstance(axis, SparseAxis): - raise NotImplementedError("Plot a sparse axis (e.g. bar chart)") - elif isinstance(axis, DenseAxis): - ax.set_xlabel(axis.label) - ax.set_ylabel(hist.label) - edges = axis.edges(overflow=overflow).get() - if order is None: - identifiers = ( - hist.identifiers(overlay, overflow=overlay_overflow) - if overlay is not None - else [None] - ) - else: - identifiers = order - plot_info = { - "identifier": identifiers, - "label": list(map(str, identifiers)), - "sumw": [], - "sumw2": [], - } - for i, identifier in enumerate(identifiers): - if identifier is None: - sumw, sumw2 = hist.values(sumw2=True, overflow=overflow)[()] - elif isinstance(overlay, SparseAxis): - sumw, sumw2 = hist.integrate(overlay, identifier).values( - sumw2=True, overflow=overflow - )[()] - else: - sumw, sumw2 = hist.values(sumw2=True, overflow="allnan")[()] - the_slice = ( - i if overflow_behavior(overlay_overflow).start is None else i + 1, - overflow_behavior(overflow), - ) - if hist._idense(overlay) == 1: - the_slice = (the_slice[1], the_slice[0]) - sumw = sumw[the_slice] - sumw2 = sumw2[the_slice] - plot_info["sumw"].append(sumw.get()) - plot_info["sumw2"].append(sumw2.get()) - - def w2err(sumw, sumw2): - err = [] - for a, b in zip(sumw, sumw2): - err.append(np.abs(poisson_interval(a, b) - a)) - return err - - kwargs = None - if line_opts is not None and error_opts is None: - _error = None - else: - _error = w2err(plot_info["sumw"], plot_info["sumw2"]) - if fill_opts is not None: - histtype = "fill" - kwargs = fill_opts - elif error_opts is not None and line_opts is None: - histtype = "errorbar" - kwargs = error_opts - else: - histtype = "step" - kwargs = line_opts - if kwargs is None: - kwargs = {} - - hep.histplot( - plot_info["sumw"], - edges, - label=plot_info["label"], - yerr=_error, - histtype=histtype, - ax=ax, - density=density, - binwnorm=binwnorm, - stack=stack, - **kwargs, - ) - - if stack and error_opts is not None: - stack_sumw = np.sum(plot_info["sumw"], axis=0) - stack_sumw2 = np.sum(plot_info["sumw2"], axis=0) - err = poisson_interval(stack_sumw, stack_sumw2) - if binwnorm is not None: - err *= binwnorm / np.diff(edges)[None, :] - opts = { - "step": "post", - "label": "Sum unc.", - "hatch": "///", - "facecolor": "none", - "edgecolor": (0, 0, 0, 0.5), - "linewidth": 0, - } - opts.update(error_opts) - ax.fill_between( - x=edges, - y1=np.r_[err[0, :], err[0, -1]], - y2=np.r_[err[1, :], err[1, -1]], - **opts, - ) - - if legend_opts is not None: - _label = overlay.label if overlay is not None else "" - ax.legend(title=_label, **legend_opts) - else: - ax.legend(title=_label) - ax.autoscale(axis="x", tight=True) - ax.set_ylim(0, None) - - return ax - - -def plotratio( - num, - denom, - ax=None, - clear=True, - overflow="none", - xerr=False, - error_opts=None, - denom_fill_opts=None, - guide_opts=None, - unc="clopper-pearson", - label=None, -): - """Create a ratio plot, dividing two compatible histograms - - Parameters - ---------- - num : Hist - Numerator, a single-axis histogram - denom : Hist - Denominator, a single-axis histogram - ax : matplotlib.axes.Axes, optional - Axes object (if None, one is created) - clear : bool, optional - Whether to clear Axes before drawing (if passed); if False, this function will skip drawing the legend - overflow : str, optional - If overflow behavior is not 'none', extra bins will be drawn on either end of the nominal - axis range, to represent the contents of the overflow bins. See `Hist.sum` documentation - for a description of the options. - xerr: bool, optional - If true, then error bars are drawn for x-axis to indicate the size of the bin. - error_opts : dict, optional - A dictionary of options to pass to the matplotlib - `ax.errorbar <https://matplotlib.org/3.1.1/api/_as_gen/matplotlib.pyplot.errorbar.html>`_ call - internal to this function. Leave blank for defaults. Some special options are interpreted by - this function and not passed to matplotlib: 'emarker' (default: '') specifies the marker type - to place at cap of the errorbar. - denom_fill_opts : dict, optional - A dictionary of options to pass to the matplotlib - `ax.fill_between <https://matplotlib.org/3.1.1/api/_as_gen/matplotlib.axes.Axes.fill_between.html>`_ call - internal to this function, filling the denominator uncertainty band. Leave blank for defaults. - guide_opts : dict, optional - A dictionary of options to pass to the matplotlib - `ax.axhline <https://matplotlib.org/3.1.1/api/_as_gen/matplotlib.axes.Axes.axhline.html>`_ call - internal to this function, to plot a horizontal guide line at ratio of 1. Leave blank for defaults. - unc : str, optional - Uncertainty calculation option: 'clopper-pearson' interval for efficiencies; 'poisson-ratio' interval - for ratio of poisson distributions; 'num' poisson interval of numerator scaled by denominator value - (common for data/mc, for better or worse). - label : str, optional - Associate a label to this entry (note: y axis label set by ``num.label``) - - Returns - ------- - ax : matplotlib.axes.Axes - A matplotlib `Axes <https://matplotlib.org/3.1.1/api/axes_api.html>`_ object - """ - import matplotlib.pyplot as plt - - if ax is None: - fig, ax = plt.subplots(1, 1) - else: - if not isinstance(ax, plt.Axes): - raise ValueError("ax must be a matplotlib Axes object") - if clear: - ax.clear() - if not num.compatible(denom): - raise ValueError( - "numerator and denominator histograms have incompatible axis definitions" - ) - if num.dim() > 1: - raise ValueError("plotratio() can only support one-dimensional histograms") - if error_opts is None and denom_fill_opts is None and guide_opts is None: - error_opts = {} - denom_fill_opts = {} - - axis = num.axes()[0] - if isinstance(axis, SparseAxis): - raise NotImplementedError("Ratio for sparse axes (labeled axis with errorbars)") - elif isinstance(axis, DenseAxis): - ax.set_xlabel(axis.label) - ax.set_ylabel(num.label) - edges = axis.edges(overflow=overflow).get() - centers = axis.centers(overflow=overflow).get() - ranges = (edges[1:] - edges[:-1]) / 2 if xerr else None - - sumw_num, sumw2_num = num.values(sumw2=True, overflow=overflow)[()] - sumw_denom, sumw2_denom = denom.values(sumw2=True, overflow=overflow)[()] - sumw_num, sumw2_num = sumw_num.get(), sumw2_num.get() - sumw_denom, sumw2_denom = sumw_denom.get(), sumw2_denom.get() - - rsumw = sumw_num / sumw_denom - if unc == "clopper-pearson": - rsumw_err = np.abs(clopper_pearson_interval(sumw_num, sumw_denom) - rsumw) - elif unc == "poisson-ratio": - # poisson ratio n/m is equivalent to binomial n/(n+m) - rsumw_err = np.abs( - clopper_pearson_interval(sumw_num, sumw_num + sumw_denom) - rsumw - ) - elif unc == "num": - rsumw_err = np.abs( - poisson_interval(rsumw, sumw2_num / sumw_denom**2) - rsumw - ) - elif unc == "normal": - rsumw_err = np.abs( - normal_interval(sumw_num, sumw_denom, sumw2_num, sumw2_denom) - ) - else: - raise ValueError(f"Unrecognized uncertainty option: {unc!r}") - - if error_opts is not None: - opts = {"label": label, "linestyle": "none"} - opts.update(error_opts) - emarker = opts.pop("emarker", "") - errbar = ax.errorbar( - x=centers, y=rsumw, xerr=ranges, yerr=rsumw_err, **opts - ) - plt.setp(errbar[1], "marker", emarker) - if denom_fill_opts is not None: - unity = np.ones_like(sumw_denom) - denom_unc = poisson_interval(unity, sumw2_denom / sumw_denom**2) - opts = {"step": "post", "facecolor": (0, 0, 0, 0.3), "linewidth": 0} - opts.update(denom_fill_opts) - ax.fill_between( - edges, - np.r_[denom_unc[0], denom_unc[0, -1]], - np.r_[denom_unc[1], denom_unc[1, -1]], - **opts, - ) - if guide_opts is not None: - opts = {"linestyle": "--", "color": (0, 0, 0, 0.5), "linewidth": 1} - opts.update(guide_opts) - ax.axhline(1.0, **opts) - - if clear: - ax.autoscale(axis="x", tight=True) - ax.set_ylim(0, None) - - return ax - - -def plot2d( - hist, - xaxis, - ax=None, - clear=True, - xoverflow="none", - yoverflow="none", - patch_opts=None, - text_opts=None, - density=False, - binwnorm=None, -): - """Create a 2D plot from a 2D `Hist` object - - Parameters - ---------- - hist : Hist - Histogram with two dimensions - xaxis : str or Axis - Which of the two dimensions to use as an x axis - ax : matplotlib.axes.Axes, optional - Axes object (if None, one is created) - clear : bool, optional - Whether to clear Axes before drawing (if passed); if False, this function will skip drawing the legend - xoverflow : str, optional - If overflow behavior is not 'none', extra bins will be drawn on either end of the nominal x - axis range, to represent the contents of the overflow bins. See `Hist.sum` documentation - for a description of the options. - yoverflow : str, optional - Similar to ``xoverflow`` - patch_opts : dict, optional - Options passed to the matplotlib `pcolormesh <https://matplotlib.org/api/_as_gen/matplotlib.pyplot.pcolormesh.html>`_ - call internal to this function, to plot a rectangular grid of patches colored according to the bin values. - Leave empty for defaults. - text_opts : dict, optional - Options passed to the matplotlib `text <https://matplotlib.org/api/text_api.html#matplotlib.text.Text>`_ - call internal to this function, to place a text label at each bin center with the bin value. Special - options interpreted by this function and not passed to matplotlib: 'format': printf-style float format - , default '%.2g'. - density : bool, optional - If true, convert sum weights to probability density (i.e. integrates to 1 over domain of axis) - (Note: this option conflicts with ``binwnorm``) - binwnorm : float, optional - If true, convert sum weights to bin-width-normalized, with unit equal to supplied value (usually you want to specify 1.) - - Returns - ------- - ax : matplotlib.axes.Axes - A matplotlib `Axes <https://matplotlib.org/3.1.1/api/axes_api.html>`_ object - """ - import matplotlib.pyplot as plt - - if ax is None: - fig, ax = plt.subplots(1, 1) - else: - if not isinstance(ax, plt.Axes): - raise ValueError("ax must be a matplotlib Axes object") - if clear: - ax.clear() - fig = ax.figure - if hist.dim() != 2: - raise ValueError("plot2d() can only support exactly two dimensions") - if density and binwnorm is not None: - raise ValueError("Cannot use density and binwnorm at the same time!") - if binwnorm is not None and not isinstance(binwnorm, numbers.Number): - raise ValueError( - f"Bin width normalization not a number, but a {binwnorm.__class__!r}" - ) - if patch_opts is None and text_opts is None: - patch_opts = {} - - xaxis = hist.axis(xaxis) - yaxis = hist.axes()[1] - transpose = False - if yaxis == xaxis: - yaxis = hist.axes()[0] - transpose = True - if isinstance(xaxis, SparseAxis) or isinstance(yaxis, SparseAxis): - raise NotImplementedError("Plot a sparse axis (e.g. bar chart or labeled bins)") - else: - xedges = xaxis.edges(overflow=xoverflow).get() - yedges = yaxis.edges(overflow=yoverflow).get() - sumw, sumw2 = hist.values(sumw2=True, overflow="allnan")[()] - sumw, sumw2 = sumw.get(), sumw2.get() - if transpose: - sumw = sumw.T - sumw2 = sumw2.T - # no support for different overflow behavior per axis, do it ourselves - sumw = sumw[overflow_behavior(xoverflow), overflow_behavior(yoverflow)] - sumw2 = sumw2[overflow_behavior(xoverflow), overflow_behavior(yoverflow)] - if (density or binwnorm is not None) and np.sum(sumw) > 0: - overallnorm = np.sum(sumw) * binwnorm if binwnorm is not None else 1.0 - areas = np.multiply.outer(np.diff(xedges), np.diff(yedges)) - binnorms = overallnorm / (areas * np.sum(sumw)) - sumw = sumw * binnorms - sumw2 = sumw2 * binnorms**2 - - if patch_opts is not None: - opts = {"cmap": "viridis"} - opts.update(patch_opts) - pc = ax.pcolormesh(xedges, yedges, sumw.T, **opts) - ax.add_collection(pc) - if clear: - fig.colorbar(pc, ax=ax, label=hist.label) - if text_opts is not None: - for ix, xcenter in enumerate(xaxis.centers()): - for iy, ycenter in enumerate(yaxis.centers()): - opts = { - "horizontalalignment": "center", - "verticalalignment": "center", - } - if patch_opts is not None: - opts["color"] = ( - "black" if pc.norm(sumw[ix, iy]) > 0.5 else "lightgrey" - ) - opts.update(text_opts) - txtformat = opts.pop("format", r"%.2g") - ax.text(xcenter, ycenter, txtformat % sumw[ix, iy], **opts) - - if clear: - ax.set_xlabel(xaxis.label) - ax.set_ylabel(yaxis.label) - ax.set_xlim(xedges[0], xedges[-1]) - ax.set_ylim(yedges[0], yedges[-1]) - - return ax - - -def plotgrid( - h, - figure=None, - row=None, - col=None, - overlay=None, - row_overflow="none", - col_overflow="none", - **plot_opts, -): - """Create a grid of plots, enumerating identifiers on up to 3 axes - - Parameters - ---------- - h : Hist - A histogram with up to 3 axes - figure : matplotlib.figure.Figure, optional - If omitted, a new figure is created. Otherwise, the axes will be redrawn on this existing figure. - row : str - Name of row axis - col : str - Name of column axis - overlay : str - name of overlay axis - row_overflow : str, optional - If overflow behavior is not 'none', extra bins will be drawn on either end of the nominal x - axis range, to represent the contents of the overflow bins. See `Hist.sum` documentation - for a description of the options. - col_overflow : str, optional - Similar to ``row_overflow`` - ``**plot_opts`` : kwargs - The remaining axis of the histogram, after removing any of ``row,col,overlay`` specified, - will be the plot axis, with ``plot_opts`` passed to the `plot1d` call. - - Returns - ------- - axes : np.ndarray - An array of matplotlib `Axes <https://matplotlib.org/3.1.1/api/axes_api.html>`_ objects - """ - import matplotlib.pyplot as plt - - haxes = {ax.name for ax in h.axes()} - nrow, ncol = 1, 1 - if row: - row_identifiers = h.identifiers(row, overflow=row_overflow) - nrow = len(row_identifiers) - haxes.remove(row) - if col: - col_identifiers = h.identifiers(col, overflow=col_overflow) - ncol = len(col_identifiers) - haxes.remove(col) - if overlay: - haxes.remove(overlay) - if len(haxes) > 1: - raise ValueError( - "More than one dimension left: {}".format(",".join(ax for ax in haxes)) - ) - elif len(haxes) == 0: - raise ValueError(f"Not enough dimensions available in {h!r}") - - figsize = plt.rcParams["figure.figsize"] - figsize = figsize[0] * max(ncol, 1), figsize[1] * max(nrow, 1) - if figure is None: - fig, axes = plt.subplots( - nrow, ncol, figsize=figsize, squeeze=False, sharex=True, sharey=True - ) - else: - fig = figure - shape = (0, 0) - lastax = fig.get_children()[-1] - if isinstance(lastax, plt.Axes): - shape = lastax.rowNum + 1, lastax.colNum + 1 - if shape[0] == nrow and shape[1] == ncol: - axes = np.array(fig.axes).reshape(shape) - else: - fig.clear() - # fig.set_size_inches(figsize) - axes = fig.subplots(nrow, ncol, squeeze=False, sharex=True, sharey=True) - - for icol in range(ncol): - hcol = h - coltitle = None - if col: - vcol = col_identifiers[icol] - hcol = h.integrate(col, vcol) - coltitle = str(vcol) - if isinstance(vcol, Interval) and vcol.label is None: - coltitle = f"{h.axis(col).label} ∈ {coltitle}" - for irow in range(nrow): - ax = axes[irow, icol] - hplot = hcol - rowtitle = None - if row: - vrow = row_identifiers[irow] - hplot = hcol.integrate(row, vrow) - rowtitle = str(vrow) - if isinstance(vrow, Interval) and vrow.label is None: - rowtitle = f"{h.axis(row).label} ∈ {rowtitle}" - - plot1d(hplot, ax=ax, overlay=overlay, **plot_opts) - if row is not None and col is not None: - ax.set_title(f"{rowtitle}, {coltitle}") - elif row is not None: - ax.set_title(rowtitle) - elif col is not None: - ax.set_title(coltitle) - - for ax in axes.flatten(): - ax.autoscale(axis="y") - ax.set_ylim(0, None) - - return axes - - -def isnotebook(): - try: - shell = get_ipython().__class__.__name__ - if shell == "ZMQInteractiveShell": - return True # Jupyter notebook or qtconsole - elif shell == "TerminalInteractiveShell": - return False # Terminal running IPython - else: - return False # Other type (?) - except NameError: - return False - - -def bokeh_plot(histo, jup_url="http://127.0.0.1:8889"): - if not isnotebook(): - raise NotImplementedError("Only usable in jupyter notebook") - import bokeh.plotting.figure as bk_figure - from bokeh import palettes - - # init bokeh - from bokeh.application import Application - from bokeh.application.handlers import FunctionHandler - from bokeh.core.validation import silence - from bokeh.core.validation.warnings import EMPTY_LAYOUT - from bokeh.io import ( - output_notebook, # enables plot interface in J notebook - show, - ) - from bokeh.layouts import column, row - from bokeh.models import ColumnDataSource - from bokeh.models.widgets import ( - CheckboxButtonGroup, - Div, - RadioButtonGroup, - RangeSlider, - ) - - silence(EMPTY_LAYOUT, True) - - output_notebook() - - # Set up widgets - cfg_labels = ["Ghost"] - wi_config = CheckboxButtonGroup(labels=cfg_labels, active=[0]) - wi_dense_select = RadioButtonGroup( - labels=[ax.name for ax in histo.dense_axes()], active=0 - ) - wi_sparse_select = RadioButtonGroup( - labels=[ax.name for ax in histo.sparse_axes()], active=0 - ) - - # Dense widgets - sliders = {} - for ax in histo.dense_axes(): - edge_vals = (histo.axis(ax.name).edges()[0], histo.axis(ax.name).edges()[-1]) - _smallest_bin = np.min(np.diff(histo.axis(ax.name).edges())) - sliders[ax.name] = RangeSlider( - title=ax.name, - value=edge_vals, - start=edge_vals[0], - end=edge_vals[1], - step=_smallest_bin, - name=ax.name, - ) - - # Cat widgets - togglers = {} - for ax in histo.sparse_axes(): - togglers[ax.name] = CheckboxButtonGroup( - labels=[i.name for i in ax.identifiers()], active=[0], name=ax.name - ) - - # Toggles for all widgets - configers = {} - for ax in histo.sparse_axes(): - configers[ax.name] = CheckboxButtonGroup( - labels=["Display", "Ghost"], active=[0, 1], name=ax.name - ) - for ax in histo.dense_axes(): - configers[ax.name] = CheckboxButtonGroup( - labels=["Display"], active=[0], name=ax.name - ) - - # Figure - fig = bk_figure( - title="1D Projection", - plot_width=500, - plot_height=500, - min_border=20, - toolbar_location=None, - ) - fig.yaxis.axis_label = "N" - fig.xaxis.axis_label = "Quantity" - - # Iterate over possible overlays - _max_idents = 0 # Max number of simultaneou histograms - for ax in histo.sparse_axes(): - _max_idents = max(_max_idents, len([i.name for i in ax.identifiers()])) - - # Data source list - sources = [] - sources_ghost = [] - for _ in range(_max_idents): - sources.append( - ColumnDataSource({"left": [], "right": [], "top": [], "bottom": []}) - ) - sources_ghost.append( - ColumnDataSource({"left": [], "right": [], "top": [], "bottom": []}) - ) - - # Hist list - hists = [] - hists_ghost = [] - for i in range(_max_idents): - if _max_idents < 10: - _color = palettes.Category10[min(max(3, _max_idents), 10)][i] - else: - _color = palettes.magma(_max_idents)[i] - hists.append( - fig.quad( - left="left", - right="right", - top="top", - bottom="bottom", - source=sources[i], - alpha=0.9, - color=_color, - ) - ) - hists_ghost.append( - fig.quad( - left="left", - right="right", - top="top", - bottom="bottom", - source=sources_ghost[i], - alpha=0.05, - color=_color, - ) - ) - - def update_data(attrname, old, new): # noqa: ARG001 - sparse_active = wi_sparse_select.active - sparse_name = [ax.name for ax in histo.sparse_axes()][sparse_active] - sparse_other = [ax.name for ax in histo.sparse_axes() if ax.name != sparse_name] - - dense_active = wi_dense_select.active - dense_name = [ax.name for ax in histo.dense_axes()][dense_active] - dense_other = [ax.name for ax in histo.dense_axes() if ax.name != dense_name] - - # Apply cuts in projections - _h = histo.copy() - for proj_ax in sparse_other: - _idents = histo.axis(proj_ax).identifiers() - _labels = [ident.name for ident in _idents] - if 0 in configers[proj_ax].active: - _h = _h.integrate( - proj_ax, [_labels[i] for i in togglers[proj_ax].active] - ) - else: - _h = _h.integrate(proj_ax) - - for proj_ax in dense_other: - _h = _h.integrate( - proj_ax, slice(sliders[proj_ax].value[0], sliders[proj_ax].value[1]) - ) - - for cat_ix in range(_max_idents): - # Update histo for each toggled overlay - if cat_ix in togglers[sparse_name].active: - cat_value = histo.axis(sparse_name).identifiers()[cat_ix] - h1d = _h.integrate(sparse_name, cat_value) - - # Get shown histogram - values = h1d.project(dense_name).values() - if values != {}: - h = values[()] - bins = h1d.axis(dense_name).edges() - - # Apply cuts on shown axis - bin_los = bins[:-1][bins[:-1] > sliders[dense_name].value[0]] - bin_his = bins[1:][bins[1:] < sliders[dense_name].value[1]] - new_bins = np.intersect1d(bin_los, bin_his) - bin_ixs = np.searchsorted(bins, new_bins)[:-1] - h = h[bin_ixs] - - sources[cat_ix].data = { - "left": new_bins[:-1], - "right": new_bins[1:], - "top": h, - "bottom": np.zeros_like(h), - } - else: - sources[cat_ix].data = { - "left": [], - "right": [], - "top": [], - "bottom": [], - } - - # Add ghosts - if 0 in wi_config.active: - h1d = histo.integrate(sparse_name, cat_value) - for proj_ax in sparse_other: - _idents = histo.axis(proj_ax).identifiers() - _labels = [ident.name for ident in _idents] - if 1 not in configers[proj_ax].active: - h1d = h1d.integrate( - proj_ax, [_labels[i] for i in togglers[proj_ax].active] - ) - else: - h1d = h1d.integrate(proj_ax) - values = h1d.project(dense_name).values() - if values != {}: - h = h1d.project(dense_name).values()[()] - bins = h1d.axis(dense_name).edges() - sources_ghost[cat_ix].data = { - "left": bins[:-1], - "right": bins[1:], - "top": h, - "bottom": np.zeros_like(h), - } - else: - sources_ghost[cat_ix].data = { - "left": [], - "right": [], - "top": [], - "bottom": [], - } - else: - sources[cat_ix].data = { - "left": [], - "right": [], - "top": [], - "bottom": [], - } - sources_ghost[cat_ix].data = { - "left": [], - "right": [], - "top": [], - "bottom": [], - } - - # Cosmetics - fig.xaxis.axis_label = dense_name - - for _, slider in sliders.items(): - slider.on_change("value", update_data) - for _, toggler in togglers.items(): - toggler.on_change("active", update_data) - for _, configurer in configers.items(): - configurer.on_change("active", update_data) - # Button - for w in [wi_dense_select, wi_sparse_select, wi_config]: - w.on_change("active", update_data) - - from bokeh.models.widgets import Panel, Tabs - - layout = row( - fig, - column( - Div( - text="Overlay Axis:", - style={"font-size": "100%", "color": "black"}, - ), - wi_sparse_select, - Div( - text="Plot Axis:", style={"font-size": "100%", "color": "black"} - ), - wi_dense_select, - Div( - text="Categorical Cuts:", - style={"font-size": "100%", "color": "black"}, - ), - *[toggler for name, toggler in togglers.items()], - Div( - text="Dense Cuts:", style={"font-size": "100%", "color": "black"} - ), - *[slider for name, slider in sliders.items()], - ), - ) - - # Config prep - incl_lists = [[], [], []] - for i, key in enumerate(list(configers.keys())): - incl_lists[i // max(5, len(list(configers.keys())) / 3)].append( - Div( - text=f"{key}:", - style={"font-size": "70%", "color": "black"}, - ) - ) - incl_lists[i // max(5, len(list(configers.keys())) / 3)].append(configers[key]) - - layout_cfgs = column( - row( - column( - Div( - text="Configs:", - style={"font-size": "100%", "color": "black"}, - ), - wi_config, - ) - ), - Div( - text="Axis togglers:", style={"font-size": "100%", "color": "black"} - ), - row( - column(incl_lists[0]), - column(incl_lists[1]), - column(incl_lists[2]), - ), - ) - - # Update active buttons - def update_layout(attrname, old, new): # noqa: ARG001 - active_axes = [None] - for name, wi in configers.items(): - if 0 in wi.active: - active_axes.append(name) - for child in layout.children[1].children: - if child.name not in active_axes: - child.visible = False - else: - child.visible = True - - for _, configurer in configers.items(): - configurer.on_change("active", update_layout) - - tab1 = Panel(child=layout, title="Projection") - tab2 = Panel(child=layout_cfgs, title="Configs") - tabs = Tabs(tabs=[tab1, tab2]) - - def modify_doc(doc): - doc.add_root(row(tabs, width=800)) - doc.title = "Sliders" - - handler = FunctionHandler(modify_doc) - app = Application(handler) - - show(app, notebook_url=jup_url) - update_data("", "", "")

Figure out testing on GPUs (failing tests in GH Actions)

2024-07-22T17:07:59

0.0

[]

RudolfCardinal/pythonlib

RudolfCardinal__pythonlib-16

15f5c245129befef8537506b742273a017f66a75

diff --git a/.gitignore b/.gitignore index fab9b37..c02d158 100644 --- a/.gitignore +++ b/.gitignore @@ -9,3 +9,6 @@ private_testing/* # Silver searcher .ignore + +# pytest coverage report +.coverage diff --git a/cardinal_pythonlib/pdf.py b/cardinal_pythonlib/pdf.py index 7ae502f..40aefc1 100644 --- a/cardinal_pythonlib/pdf.py +++ b/cardinal_pythonlib/pdf.py @@ -36,12 +36,13 @@ import tempfile from typing import Any, Dict, Iterable, Union -from cardinal_pythonlib.logs import get_brace_style_log_with_null_handler - # noinspection PyProtectedMember -from PyPDF2 import PdfFileMerger, PdfFileReader, PdfFileWriter +from pypdf import PdfMerger, PdfReader, PdfWriter from semantic_version import Version +from cardinal_pythonlib.logs import get_brace_style_log_with_null_handler + + # ============================================================================= # Conditional/optional imports # ============================================================================= @@ -171,13 +172,13 @@ def __init__( self.filename = filename def add_to_writer( - self, writer: PdfFileWriter, start_recto: bool = True + self, writer: PdfWriter, start_recto: bool = True ) -> None: """ Add the PDF described by this class to a PDF writer. Args: - writer: a :class:`PyPDF2.PdfFileWriter` + writer: a :class:`pypdf.PdfWriter` start_recto: start a new right-hand page? """ @@ -191,10 +192,10 @@ def add_to_writer( ) append_memory_pdf_to_writer(pdf, writer, start_recto=start_recto) elif self.is_filename: - if start_recto and writer.getNumPages() % 2 != 0: - writer.addBlankPage() - writer.appendPagesFromReader( - PdfFileReader(open(self.filename, "rb")) + if start_recto and len(writer.pages) % 2 != 0: + writer.add_blank_page() + writer.append_pages_from_reader( + PdfReader(open(self.filename, "rb")) ) else: raise AssertionError("PdfPlan: shouldn't get here!") @@ -547,9 +548,9 @@ def make_pdf_on_disk_from_html( return result -def pdf_from_writer(writer: Union[PdfFileWriter, PdfFileMerger]) -> bytes: +def pdf_from_writer(writer: Union[PdfWriter, PdfMerger]) -> bytes: """ - Extracts a PDF (as binary data) from a PyPDF2 writer or merger object. + Extracts a PDF (as binary data) from a pypdf writer or merger object. """ memfile = io.BytesIO() writer.write(memfile) @@ -577,18 +578,18 @@ def serve_pdf_to_stdout(pdf: bytes) -> None: sys.stdout.buffer.write(pdf) -def make_pdf_writer() -> PdfFileWriter: +def make_pdf_writer() -> PdfWriter: """ - Creates and returns a PyPDF2 writer. + Creates and returns a pypdf writer. """ - return PdfFileWriter() + return PdfWriter() def append_memory_pdf_to_writer( - input_pdf: bytes, writer: PdfFileWriter, start_recto: bool = True + input_pdf: bytes, writer: PdfWriter, start_recto: bool = True ) -> None: """ - Appends a PDF (as bytes in memory) to a PyPDF2 writer. + Appends a PDF (as bytes in memory) to a pypdf writer. Args: input_pdf: the PDF, as ``bytes`` @@ -597,16 +598,16 @@ def append_memory_pdf_to_writer( """ if not input_pdf: return - if start_recto and writer.getNumPages() % 2 != 0: - writer.addBlankPage() + if start_recto and len(writer.pages) % 2 != 0: + writer.add_blank_page() # ... suitable for double-sided printing infile = io.BytesIO(input_pdf) - reader = PdfFileReader(infile) - for page_num in range(reader.numPages): - writer.addPage(reader.getPage(page_num)) + reader = PdfReader(infile) + for page in reader.pages: + writer.add_page(page) -def append_pdf(input_pdf: bytes, output_writer: PdfFileWriter): +def append_pdf(input_pdf: bytes, output_writer: PdfWriter): """ Appends a PDF to a pyPDF writer. Legacy interface. """ @@ -630,13 +631,13 @@ def append_pdf(input_pdf: bytes, output_writer: PdfFileWriter): # def append_disk_pdf_to_writer(filename, writer): # """Appends a PDF from disk to a pyPDF writer.""" -# if writer.getNumPages() % 2 != 0: -# writer.addBlankPage() +# if len(writer.pages) % 2 != 0: +# writer.add_blank_page() # # ... keeps final result suitable for double-sided printing # with open(filename, mode='rb') as infile: -# reader = PdfFileReader(infile) -# for page_num in range(reader.numPages): -# writer.addPage(reader.getPage(page_num)) +# reader = PdfReader(infile) +# for page in reader.pages: +# writer.add_page(page) def get_concatenated_pdf_from_disk( @@ -656,17 +657,17 @@ def get_concatenated_pdf_from_disk( # https://stackoverflow.com/questions/17104926/pypdf-merging-multiple-pdf-files-into-one-pdf # noqa # https://en.wikipedia.org/wiki/Recto_and_verso if start_recto: - writer = PdfFileWriter() + writer = PdfWriter() for filename in filenames: if filename: - if writer.getNumPages() % 2 != 0: - writer.addBlankPage() - writer.appendPagesFromReader( - PdfFileReader(open(filename, "rb")) + if len(writer.pages) % 2 != 0: + writer.add_blank_page() + writer.append_pages_from_reader( + PdfReader(open(filename, "rb")) ) return pdf_from_writer(writer) else: - merger = PdfFileMerger() + merger = PdfMerger() for filename in filenames: if filename: merger.append(open(filename, "rb")) @@ -687,7 +688,7 @@ def get_concatenated_pdf_in_memory( concatenated PDF, as ``bytes`` """ - writer = PdfFileWriter() + writer = PdfWriter() for pdfplan in pdf_plans: pdfplan.add_to_writer(writer, start_recto=start_recto) return pdf_from_writer(writer) diff --git a/docs/docs_requirements.txt b/docs/docs_requirements.txt index 2ed1df9..bcf81ad 100644 --- a/docs/docs_requirements.txt +++ b/docs/docs_requirements.txt @@ -7,6 +7,7 @@ dogpile.cache==0.9.2 # CRATE is on 3.2 Django<4.0 libChEBIpy +pdfkit pyramid==1.10.8 pytest # CamCOPS and CRATE both on this version diff --git a/docs/source/autodoc/_index.rst b/docs/source/autodoc/_index.rst index d2e2706..888322b 100644 --- a/docs/source/autodoc/_index.rst +++ b/docs/source/autodoc/_index.rst @@ -174,6 +174,7 @@ Automatic documentation of source code tests/dogpile_cache_tests.py.rst tests/interval_tests.py.rst tests/lists_tests.py.rst + tests/pdf_tests.py.rst tests/rate_limiting_tests.py.rst tests/rounding_tests.py.rst tests/rpm_tests.py.rst diff --git a/docs/source/changelog.rst b/docs/source/changelog.rst index e746c93..7b924a6 100644 --- a/docs/source/changelog.rst +++ b/docs/source/changelog.rst @@ -768,3 +768,6 @@ Quick links: ``wkhtmltopdf_options``; this prevents a bug where calls using e.g. a temporary file as footer HTML then make the next call, with no footer, fail (because the footer filename was written back to the dict). + +- ``pdf.py`` updated to use pypdf instead of PyPDF2, which is no longer + supported. diff --git a/setup.py b/setup.py index 35aebf3..d86604a 100644 --- a/setup.py +++ b/setup.py @@ -61,7 +61,7 @@ "psutil", "pygments", "pyparsing", - "PyPDF2", + "pypdf>=3.1.0", "python-dateutil", # "import dateutil" "scipy", "semantic-version",

PyPDF2 is no longer maintained According to https://pypi.org/project/PyPDF2/ : "NOTE: The PyPDF2 project is going back to its roots. PyPDF2==3.0.X will be the last version of PyPDF2. Development will continue with [pypdf==3.1.0](https://pypi.org/project/pyPdf/)." There is a migration guide at https://pypdf.readthedocs.io/en/latest/user/migration-1-to-2.html I don't know if there are changes between 2 and 3.

There is also a security advisory for 1.27.5, the version currently used by CRATE.

2023-01-27T17:05:48

0.0

[]

RudolfCardinal/pythonlib

RudolfCardinal__pythonlib-14

6e0e850b2c0372dcb120ebbce7c182dba2fe94e1

diff --git a/.github/scripts/python_checks.sh b/.github/scripts/python_checks.sh new file mode 100755 index 0000000..0f13bf9 --- /dev/null +++ b/.github/scripts/python_checks.sh @@ -0,0 +1,17 @@ +#!/usr/bin/env bash + +# Run from .github/workflows/python_checks.yml + +set -eux -o pipefail + +python3 -m venv "${HOME}/venv" +source "${HOME}/venv/bin/activate" +python -m site +python -m pip install -U pip setuptools + +echo installing pip packages +python -m pip install -e . +python -m pip install pre-commit + +echo running pre-commit checks +pre-commit run --all-files diff --git a/.github/workflows/python_checks.yml b/.github/workflows/python_checks.yml new file mode 100644 index 0000000..384b7d4 --- /dev/null +++ b/.github/workflows/python_checks.yml @@ -0,0 +1,23 @@ +--- +# yamllint disable rule:line-length +name: Python checks +# yamllint disable-line rule:truthy +on: + push: + paths: + - '**.py' + - .github/workflows/python-checks.yml + - .github/scripts/python_checks.sh +jobs: + python-checks: + runs-on: ubuntu-latest + strategy: + matrix: + python-version: ["3.7", "3.8", "3.9", "3.10"] + steps: + - uses: actions/checkout@v3 + - uses: actions/setup-python@v4 + with: + python-version: ${{ matrix.python-version }} + - name: Python checks + run: ${GITHUB_WORKSPACE}/.github/scripts/python_checks.sh diff --git a/.pre-commit-config.yaml b/.pre-commit-config.yaml new file mode 100644 index 0000000..fc00319 --- /dev/null +++ b/.pre-commit-config.yaml @@ -0,0 +1,20 @@ +# See https://pre-commit.com for more information +# See https://pre-commit.com/hooks.html for more hooks +repos: +- repo: https://github.com/pre-commit/pre-commit-hooks + rev: v3.2.0 + hooks: + - id: check-added-large-files + - id: check-json + - id: check-yaml + - id: check-xml +- repo: https://github.com/psf/black + rev: 22.8.0 + hooks: + - id: black + args: + - --line-length=79 +- repo: https://github.com/PyCQA/flake8 + rev: 5.0.4 + hooks: + - id: flake8 diff --git a/cardinal_pythonlib/__init__.py b/cardinal_pythonlib/__init__.py index 1e14c7a..4730460 100644 --- a/cardinal_pythonlib/__init__.py +++ b/cardinal_pythonlib/__init__.py @@ -29,4 +29,6 @@ # Create cardinal_pythonlib.__version__: # noinspection PyPep8Naming -from cardinal_pythonlib.version_string import VERSION_STRING as __version__ +from cardinal_pythonlib.version_string import ( # noqa: F401 + VERSION_STRING as __version__, +) diff --git a/cardinal_pythonlib/athena_ohdsi.py b/cardinal_pythonlib/athena_ohdsi.py index 38843c7..4801368 100644 --- a/cardinal_pythonlib/athena_ohdsi.py +++ b/cardinal_pythonlib/athena_ohdsi.py @@ -388,42 +388,42 @@ def gen_rows() -> Generator[AthenaConceptRow, None, None]: yield concept def filter_vocab( - concepts_: Iterable[AthenaConceptRow] + concepts_: Iterable[AthenaConceptRow], ) -> Generator[AthenaConceptRow, None, None]: for concept in concepts_: if concept.vocabulary_id in vocabulary_ids: yield concept def filter_code( - concepts_: Iterable[AthenaConceptRow] + concepts_: Iterable[AthenaConceptRow], ) -> Generator[AthenaConceptRow, None, None]: for concept in concepts_: if concept.concept_code in concept_codes: yield concept def filter_id( - concepts_: Iterable[AthenaConceptRow] + concepts_: Iterable[AthenaConceptRow], ) -> Generator[AthenaConceptRow, None, None]: for concept in concepts_: if concept.concept_id in concept_ids: yield concept def filter_not_vocab( - concepts_: Iterable[AthenaConceptRow] + concepts_: Iterable[AthenaConceptRow], ) -> Generator[AthenaConceptRow, None, None]: for concept in concepts_: if concept.vocabulary_id not in not_vocabulary_ids: yield concept def filter_not_code( - concepts_: Iterable[AthenaConceptRow] + concepts_: Iterable[AthenaConceptRow], ) -> Generator[AthenaConceptRow, None, None]: for concept in concepts_: if concept.concept_code not in not_concept_codes: yield concept def filter_not_id( - concepts_: Iterable[AthenaConceptRow] + concepts_: Iterable[AthenaConceptRow], ) -> Generator[AthenaConceptRow, None, None]: for concept in concepts_: if concept.concept_id not in not_concept_ids: @@ -526,42 +526,42 @@ def gen_rows() -> Generator[AthenaConceptRelationshipRow, None, None]: yield rel def filter_rel( - rels: Iterable[AthenaConceptRelationshipRow] + rels: Iterable[AthenaConceptRelationshipRow], ) -> Generator[AthenaConceptRelationshipRow, None, None]: for rel in rels: if rel.relationship_id in relationship_id_values: yield rel def filter_c1( - rels: Iterable[AthenaConceptRelationshipRow] + rels: Iterable[AthenaConceptRelationshipRow], ) -> Generator[AthenaConceptRelationshipRow, None, None]: for rel in rels: if rel.concept_id_1 in concept_id_1_values: yield rel def filter_c2( - rels: Iterable[AthenaConceptRelationshipRow] + rels: Iterable[AthenaConceptRelationshipRow], ) -> Generator[AthenaConceptRelationshipRow, None, None]: for rel in rels: if rel.concept_id_2 in concept_id_2_values: yield rel def filter_not_rel( - rels: Iterable[AthenaConceptRelationshipRow] + rels: Iterable[AthenaConceptRelationshipRow], ) -> Generator[AthenaConceptRelationshipRow, None, None]: for rel in rels: if rel.relationship_id not in not_relationship_id_values: yield rel def filter_not_c1( - rels: Iterable[AthenaConceptRelationshipRow] + rels: Iterable[AthenaConceptRelationshipRow], ) -> Generator[AthenaConceptRelationshipRow, None, None]: for rel in rels: if rel.concept_id_1 not in not_concept_id_1_values: yield rel def filter_not_c2( - rels: Iterable[AthenaConceptRelationshipRow] + rels: Iterable[AthenaConceptRelationshipRow], ) -> Generator[AthenaConceptRelationshipRow, None, None]: for rel in rels: if rel.concept_id_2 not in not_concept_id_2_values: diff --git a/cardinal_pythonlib/bulk_email/main.py b/cardinal_pythonlib/bulk_email/main.py index 32afb24..f4a1699 100644 --- a/cardinal_pythonlib/bulk_email/main.py +++ b/cardinal_pythonlib/bulk_email/main.py @@ -164,7 +164,7 @@ def add_job( # ------------------------------------------------------------------------- log.info(f"Reading recipients from: {recipients_filename}") - recipient_addresses = set() # type: Set[str] + recipient_addresses: Set[str] = set() recipients_failed = False for recipient_addr in gen_lines_without_comments(recipients_filename): if not is_email_valid(recipient_addr): @@ -175,7 +175,7 @@ def add_job( if recipients_failed: raise ValueError("Errors in recipient addresses") - recipients = [] # type: List[Recipient] + recipients: List[Recipient] = [] for recipient_addr in sorted(recipient_addresses): recipient = Recipient.fetch_or_add(recipient_addr, session) session.add(recipient) @@ -434,7 +434,9 @@ def main() -> None: "--recipients_file", type=str, required=True, - help="Name of a file containing one recipient e-mail address per line.", + help=( + "Name of a file containing one recipient e-mail address per line." + ), ) parser_add_job.add_argument( "--from_addr", type=str, required=True, help="From address." diff --git a/cardinal_pythonlib/bulk_email/models.py b/cardinal_pythonlib/bulk_email/models.py index 8ee0565..51f38ce 100644 --- a/cardinal_pythonlib/bulk_email/models.py +++ b/cardinal_pythonlib/bulk_email/models.py @@ -153,7 +153,7 @@ class Config(Base): # https://dev.mysql.com/doc/refman/8.0/en/create-table-check-constraints.html # noqa primary_key=True, autoincrement=True, - comment=f"Primary key.", + comment="Primary key.", ) database_creation_datetime = Column( DateTime, comment="Date/time this database was created." diff --git a/cardinal_pythonlib/colander_utils.py b/cardinal_pythonlib/colander_utils.py index 909a704..924584e 100644 --- a/cardinal_pythonlib/colander_utils.py +++ b/cardinal_pythonlib/colander_utils.py @@ -468,8 +468,8 @@ def schema_type() -> SchemaType: OptionalPendulumNode = ( - OptionalPendulumNodeLocalTZ -) # synonym for back-compatibility + OptionalPendulumNodeLocalTZ # synonym for back-compatibility +) class OptionalPendulumNodeUTC(SchemaNode): diff --git a/cardinal_pythonlib/datetimefunc.py b/cardinal_pythonlib/datetimefunc.py index 95e58d9..f6cb499 100644 --- a/cardinal_pythonlib/datetimefunc.py +++ b/cardinal_pythonlib/datetimefunc.py @@ -63,8 +63,6 @@ from pendulum.tz import local_timezone from pendulum.tz.timezone import Timezone -from cardinal_pythonlib.logs import main_only_quicksetup_rootlogger - if Arrow is not None: PotentialDatetimeType = Union[ None, datetime.datetime, datetime.date, DateTime, str, Arrow @@ -651,8 +649,8 @@ def duration_to_iso( if permit_years_months: # Watch out here. Before Pendulum 2.1.1, d.total_seconds() ignored # year/month components. But from Pendulum 2.1.1, it incorporates - # year/month information with the assumption that a year is 365 days and - # a month is 30 days, which is perhaps a bit iffy. + # year/month information with the assumption that a year is 365 days + # and a month is 30 days, which is perhaps a bit iffy. y = d.years m = d.months s = get_pendulum_duration_nonyear_nonmonth_seconds(d) @@ -676,7 +674,7 @@ def duration_to_iso( def truncate_date_to_first_of_month( - dt: Optional[DateLikeType] + dt: Optional[DateLikeType], ) -> Optional[DateLikeType]: """ Change the day to the first of the month. diff --git a/cardinal_pythonlib/debugging.py b/cardinal_pythonlib/debugging.py index 0dd1ce2..7f916f6 100644 --- a/cardinal_pythonlib/debugging.py +++ b/cardinal_pythonlib/debugging.py @@ -28,7 +28,6 @@ import ctypes import inspect -from inspect import FrameInfo import logging import pdb import sys @@ -54,7 +53,7 @@ def pdb_run(func: Callable, *args: Any, **kwargs: Any) -> Any: # noinspection PyBroadException try: return func(*args, **kwargs) - except: # nopep8 + except Exception: type_, value, tb = sys.exc_info() traceback.print_exc() pdb.post_mortem(tb) diff --git a/cardinal_pythonlib/django/fields/isodatetimetz.py b/cardinal_pythonlib/django/fields/isodatetimetz.py index a7a8b96..572aec6 100644 --- a/cardinal_pythonlib/django/fields/isodatetimetz.py +++ b/cardinal_pythonlib/django/fields/isodatetimetz.py @@ -56,7 +56,7 @@ def iso_string_to_python_datetime( - isostring: str + isostring: str, ) -> Optional[datetime.datetime]: """ Takes an ISO-8601 string and returns a ``datetime``. @@ -67,7 +67,7 @@ def iso_string_to_python_datetime( def python_utc_datetime_to_sqlite_strftime_string( - value: datetime.datetime + value: datetime.datetime, ) -> str: """ Converts a Python datetime to a string literal compatible with SQLite, @@ -246,8 +246,8 @@ def get_prep_value(self, value): Convert Python value to database value for QUERYING. We query with UTC, so this function converts datetime values to UTC. - Calls to this function are followed by calls to ``get_db_prep_value()``, - which is for backend-specific conversions. + Calls to this function are followed by calls to + ``get_db_prep_value()``, which is for backend-specific conversions. """ log.debug("get_prep_value: {}, {}", value, type(value)) if not value: diff --git a/cardinal_pythonlib/django/function_cache.py b/cardinal_pythonlib/django/function_cache.py index f663b19..a994ecb 100644 --- a/cardinal_pythonlib/django/function_cache.py +++ b/cardinal_pythonlib/django/function_cache.py @@ -87,8 +87,8 @@ def make_cache_key(call_signature: str, debug_cache: bool = False) -> str: is ``'a.b.c'`` a ``TableId``, or is it a ``ColumnId`` with no ``'db'`` field? - We could use ``repr()``: sometimes that gives us helpful things that - could in principle be passed to ``eval()``, in which case ``repr()`` would - be fine, but sometimes it doesn't, and gives unhelpful things like + could in principle be passed to ``eval()``, in which case ``repr()`` + would be fine, but sometimes it doesn't, and gives unhelpful things like ``'<__main__.Thing object at 0x7ff3093ebda0>'``. - However, if something encodes to JSON, that representation should be reversible and thus contain the right sort of information. diff --git a/cardinal_pythonlib/django/tools/generate_new_django_secret_key.py b/cardinal_pythonlib/django/tools/generate_new_django_secret_key.py index f3b0026..96759d4 100644 --- a/cardinal_pythonlib/django/tools/generate_new_django_secret_key.py +++ b/cardinal_pythonlib/django/tools/generate_new_django_secret_key.py @@ -35,10 +35,10 @@ def main() -> None: """ Generates a new Django secret key and prints it to stdout. """ - chars = 'abcdefghijklmnopqrstuvwxyz0123456789!@#$%^&*(-_=+)' # valid chars + chars = "abcdefghijklmnopqrstuvwxyz0123456789!@#$%^&*(-_=+)" # valid chars key = get_random_string(50, chars) print(key) -if __name__ == '__main__': +if __name__ == "__main__": main() diff --git a/cardinal_pythonlib/email/sendmail.py b/cardinal_pythonlib/email/sendmail.py index 891c50d..d90c73c 100755 --- a/cardinal_pythonlib/email/sendmail.py +++ b/cardinal_pythonlib/email/sendmail.py @@ -66,21 +66,24 @@ # Make e-mail message # ============================================================================= -def make_email(from_addr: str, - date: str = None, - sender: str = "", - reply_to: Union[str, List[str]] = "", - to: Union[str, List[str]] = "", - cc: Union[str, List[str]] = "", - bcc: Union[str, List[str]] = "", - subject: str = "", - body: str = "", - content_type: str = CONTENT_TYPE_TEXT, - charset: str = UTF8, - attachment_filenames: Sequence[str] = None, - attachment_binaries: Sequence[bytes] = None, - attachment_binary_filenames: Sequence[str] = None, - verbose: bool = False) -> email.mime.multipart.MIMEMultipart: + +def make_email( + from_addr: str, + date: str = None, + sender: str = "", + reply_to: Union[str, List[str]] = "", + to: Union[str, List[str]] = "", + cc: Union[str, List[str]] = "", + bcc: Union[str, List[str]] = "", + subject: str = "", + body: str = "", + content_type: str = CONTENT_TYPE_TEXT, + charset: str = UTF8, + attachment_filenames: Sequence[str] = None, + attachment_binaries: Sequence[bytes] = None, + attachment_binary_filenames: Sequence[str] = None, + verbose: bool = False, +) -> email.mime.multipart.MIMEMultipart: """ Makes an e-mail message. @@ -129,6 +132,7 @@ def make_email(from_addr: str, :exc:`AssertionError`, :exc:`ValueError` """ + def _csv_list_to_list(x: str) -> List[str]: stripped = [item.strip() for item in x.split(COMMA)] return [item for item in stripped if item] @@ -137,9 +141,9 @@ def _assert_nocomma(x: Union[str, List[str]]) -> None: if isinstance(x, str): x = [x] for _addr in x: - assert COMMA not in _addr, ( - f"Commas not allowed in e-mail addresses: {_addr!r}" - ) + assert ( + COMMA not in _addr + ), f"Commas not allowed in e-mail addresses: {_addr!r}" # ------------------------------------------------------------------------- # Arguments @@ -151,9 +155,9 @@ def _assert_nocomma(x: Union[str, List[str]]) -> None: f"(for Python sendmail, not RFC 2822); was {from_addr!r}" ) _assert_nocomma(from_addr) - assert isinstance(sender, str), ( - f"'Sender:' can only be a single address; was {sender!r}" - ) + assert isinstance( + sender, str + ), f"'Sender:' can only be a single address; was {sender!r}" _assert_nocomma(sender) if isinstance(reply_to, str): reply_to = [reply_to] if reply_to else [] # type: List[str] @@ -169,11 +173,13 @@ def _assert_nocomma(x: Union[str, List[str]]) -> None: _assert_nocomma(cc) _assert_nocomma(bcc) attachment_filenames = attachment_filenames or [] # type: List[str] - assert all(attachment_filenames), ( - f"Missing attachment filenames: {attachment_filenames!r}" - ) + assert all( + attachment_filenames + ), f"Missing attachment filenames: {attachment_filenames!r}" attachment_binaries = attachment_binaries or [] # type: List[bytes] - attachment_binary_filenames = attachment_binary_filenames or [] # type: List[str] # noqa + attachment_binary_filenames = ( + attachment_binary_filenames or [] + ) # type: List[str] # noqa assert len(attachment_binaries) == len(attachment_binary_filenames), ( "If you specify attachment_binaries or attachment_binary_filenames, " "they must be iterables of the same length." @@ -229,8 +235,8 @@ def _assert_nocomma(x: Union[str, List[str]]) -> None: part.set_payload(open(f, "rb").read()) email.encoders.encode_base64(part) part.add_header( - 'Content-Disposition', - 'attachment; filename="%s"' % os.path.basename(f) + "Content-Disposition", + 'attachment; filename="%s"' % os.path.basename(f), ) msg.attach(part) if attachment_binaries: @@ -238,8 +244,10 @@ def _assert_nocomma(x: Union[str, List[str]]) -> None: # Binary attachments, which have a notional filename # ----------------------------------------------------------------- if verbose: - log.debug("attachment_binary_filenames: {}", - attachment_binary_filenames) + log.debug( + "attachment_binary_filenames: {}", + attachment_binary_filenames, + ) for i in range(len(attachment_binaries)): blob = attachment_binaries[i] filename = attachment_binary_filenames[i] @@ -247,8 +255,9 @@ def _assert_nocomma(x: Union[str, List[str]]) -> None: part.set_payload(blob) email.encoders.encode_base64(part) part.add_header( - 'Content-Disposition', - 'attachment; filename="%s"' % filename) + "Content-Disposition", + 'attachment; filename="%s"' % filename, + ) msg.attach(part) except Exception as e: raise ValueError(f"send_email: Failed to attach files: {e}") @@ -260,15 +269,18 @@ def _assert_nocomma(x: Union[str, List[str]]) -> None: # Send message # ============================================================================= -def send_msg(from_addr: str, - to_addrs: Union[str, List[str]], - host: str, - user: str, - password: str, - port: int = None, - use_tls: bool = True, - msg: email.mime.multipart.MIMEMultipart = None, - msg_string: str = None) -> None: + +def send_msg( + from_addr: str, + to_addrs: Union[str, List[str]], + host: str, + user: str, + password: str, + port: int = None, + use_tls: bool = True, + msg: email.mime.multipart.MIMEMultipart = None, + msg_string: str = None, +) -> None: """ Sends a pre-built e-mail message. @@ -303,7 +315,8 @@ def send_msg(from_addr: str, # e.g. ConnectionRefusedError when creating the socket # SMTPException is a subclass of OSError since 3.4 raise RuntimeError( - f"send_msg: Failed to connect to host {host}, port {port}: {e}") + f"send_msg: Failed to connect to host {host}, port {port}: {e}" + ) try: session.ehlo() except smtplib.SMTPException as e: @@ -322,7 +335,8 @@ def send_msg(from_addr: str, session.login(user, password) except smtplib.SMTPException as e: raise RuntimeError( - f"send_msg: Failed to login as user {user}: {e}") + f"send_msg: Failed to login as user {user}: {e}" + ) else: log.debug("Not using SMTP AUTH; no user specified") # For systems with... lax... security requirements @@ -341,26 +355,29 @@ def send_msg(from_addr: str, # Send e-mail # ============================================================================= -def send_email(from_addr: str, - host: str, - user: str, - password: str, - port: int = None, - use_tls: bool = True, - date: str = None, - sender: str = "", - reply_to: Union[str, List[str]] = "", - to: Union[str, List[str]] = "", - cc: Union[str, List[str]] = "", - bcc: Union[str, List[str]] = "", - subject: str = "", - body: str = "", - content_type: str = CONTENT_TYPE_TEXT, - charset: str = UTF8, - attachment_filenames: Sequence[str] = None, - attachment_binaries: Sequence[bytes] = None, - attachment_binary_filenames: Sequence[str] = None, - verbose: bool = False) -> Tuple[bool, str]: + +def send_email( + from_addr: str, + host: str, + user: str, + password: str, + port: int = None, + use_tls: bool = True, + date: str = None, + sender: str = "", + reply_to: Union[str, List[str]] = "", + to: Union[str, List[str]] = "", + cc: Union[str, List[str]] = "", + bcc: Union[str, List[str]] = "", + subject: str = "", + body: str = "", + content_type: str = CONTENT_TYPE_TEXT, + charset: str = UTF8, + attachment_filenames: Sequence[str] = None, + attachment_binaries: Sequence[bytes] = None, + attachment_binary_filenames: Sequence[str] = None, + verbose: bool = False, +) -> Tuple[bool, str]: """ Sends an e-mail in text/html format using SMTP via TLS. @@ -546,6 +563,7 @@ def get_email_domain(email_: str) -> str: # Parse command line # ============================================================================= + def main() -> NoReturn: """ Command-line processor. See ``--help`` for details. @@ -553,29 +571,25 @@ def main() -> NoReturn: logging.basicConfig() log.setLevel(logging.DEBUG) parser = argparse.ArgumentParser( - description="Send an e-mail from the command line.") - parser.add_argument("sender", action="store", - help="Sender's e-mail address") - parser.add_argument("host", action="store", - help="SMTP server hostname") - parser.add_argument("user", action="store", - help="SMTP username") - parser.add_argument("password", action="store", - help="SMTP password") - parser.add_argument("recipient", action="append", - help="Recipient e-mail address(es)") - parser.add_argument("subject", action="store", - help="Message subject") - parser.add_argument("body", action="store", - help="Message body") - parser.add_argument("--attach", nargs="*", - help="Filename(s) to attach") - parser.add_argument("--tls", action="store_false", - help="Use TLS connection security") - parser.add_argument("--verbose", action="store_true", - help="Be verbose") - parser.add_argument("-h --help", action="help", - help="Prints this help") + description="Send an e-mail from the command line." + ) + parser.add_argument( + "sender", action="store", help="Sender's e-mail address" + ) + parser.add_argument("host", action="store", help="SMTP server hostname") + parser.add_argument("user", action="store", help="SMTP username") + parser.add_argument("password", action="store", help="SMTP password") + parser.add_argument( + "recipient", action="append", help="Recipient e-mail address(es)" + ) + parser.add_argument("subject", action="store", help="Message subject") + parser.add_argument("body", action="store", help="Message body") + parser.add_argument("--attach", nargs="*", help="Filename(s) to attach") + parser.add_argument( + "--tls", action="store_false", help="Use TLS connection security" + ) + parser.add_argument("--verbose", action="store_true", help="Be verbose") + parser.add_argument("-h --help", action="help", help="Prints this help") args = parser.parse_args() (result, msg) = send_email( from_addr=args.sender, @@ -597,5 +611,5 @@ def main() -> NoReturn: sys.exit(0 if result else 1) -if __name__ == '__main__': +if __name__ == "__main__": main() diff --git a/cardinal_pythonlib/extract_text.py b/cardinal_pythonlib/extract_text.py index 04d0784..1209ce7 100755 --- a/cardinal_pythonlib/extract_text.py +++ b/cardinal_pythonlib/extract_text.py @@ -1574,13 +1574,13 @@ def main() -> None: help="Use semi-plain format (re e.g. table layouts)", ) parser.add_argument( - "--width", type=int, default=DEFAULT_WIDTH, help=f"Word wrapping width" + "--width", type=int, default=DEFAULT_WIDTH, help="Word wrapping width" ) parser.add_argument( "--min-col-width", type=int, default=DEFAULT_MIN_COL_WIDTH, - help=f"Minimum column width for tables", + help="Minimum column width for tables", ) args = parser.parse_args() if args.availability: diff --git a/cardinal_pythonlib/file_io.py b/cardinal_pythonlib/file_io.py index 3ac558f..3bbc5f2 100644 --- a/cardinal_pythonlib/file_io.py +++ b/cardinal_pythonlib/file_io.py @@ -219,7 +219,7 @@ def gen_lines_without_comments( def gen_textfiles_from_filenames( - filenames: Iterable[str] + filenames: Iterable[str], ) -> Generator[TextIO, None, None]: """ Generates file-like objects from a list of filenames. @@ -237,7 +237,7 @@ def gen_textfiles_from_filenames( def gen_lines_from_textfiles( - files: Iterable[TextIO] + files: Iterable[TextIO], ) -> Generator[str, None, None]: """ Generates lines from file-like objects. diff --git a/cardinal_pythonlib/getch.py b/cardinal_pythonlib/getch.py index 973fe0c..fb3befd 100644 --- a/cardinal_pythonlib/getch.py +++ b/cardinal_pythonlib/getch.py @@ -41,7 +41,7 @@ except ImportError: msvcrt = None import termios # Unix only - import tty # requires termios, so Unix only + import tty # requires termios, so Unix only # noqa: F401 # ============================================================================= diff --git a/cardinal_pythonlib/hash.py b/cardinal_pythonlib/hash.py index 5362e85..105214a 100644 --- a/cardinal_pythonlib/hash.py +++ b/cardinal_pythonlib/hash.py @@ -73,7 +73,7 @@ # https://docs.python.org/3/library/platform.html#platform.architecture -IS_64_BIT = sys.maxsize > 2 ** 32 +IS_64_BIT = sys.maxsize > 2**32 TIMING_HASH = "hash" @@ -485,7 +485,7 @@ def murmur3_64(data: Union[bytes, bytearray], seed: int = 19820125) -> int: m = 0xC6A4A7935BD1E995 r = 47 - mask = 2 ** 64 - 1 + mask = 2**64 - 1 length = len(data) diff --git a/cardinal_pythonlib/interval.py b/cardinal_pythonlib/interval.py index b5d1294..4372044 100644 --- a/cardinal_pythonlib/interval.py +++ b/cardinal_pythonlib/interval.py @@ -59,9 +59,9 @@ datetime.datetime.strptime(x, "%Y-%m-%d").date() for x in [ # https://www.gov.uk/bank-holidays - # All bank holiday dates vary, even the date-based ones; e.g. if Christmas - # Day is a Sunday, then the Christmas Day substitute bank holiday is Tue 27 - # Dec, after the Boxing Day Monday bank holiday. + # All bank holiday dates vary, even the date-based ones; e.g. if + # Christmas Day is a Sunday, then the Christmas Day substitute bank + # holiday is Tue 27 Dec, after the Boxing Day Monday bank holiday. # 2014 "2014-01-01", # New Year's Day "2014-04-18", # Good Friday diff --git a/cardinal_pythonlib/lists.py b/cardinal_pythonlib/lists.py index fc7046a..38823a1 100644 --- a/cardinal_pythonlib/lists.py +++ b/cardinal_pythonlib/lists.py @@ -186,13 +186,13 @@ def count_bool(blist: Iterable[Any]) -> int: def delete_elements_by_index(x: List[Any], indices: Sequence[int]) -> None: """ Deletes (in place) objects from a list, by (zero-based) index values. - + Args: x: list to modify indices: zero-based index values at which to delete elements of x - + Note: - If you pass duplicate values, unexpected results will occur. - If you pass out-of-bound indices, :exc:`IndexError` will be raised. diff --git a/cardinal_pythonlib/logs.py b/cardinal_pythonlib/logs.py index d6fd5e5..2704492 100644 --- a/cardinal_pythonlib/logs.py +++ b/cardinal_pythonlib/logs.py @@ -593,7 +593,7 @@ def emit(self, record: logging.LogRecord) -> None: try: html = self.format(record) self.logfunction(html) - except: # nopep8 + except Exception: self.handleError(record) diff --git a/cardinal_pythonlib/maths_numpy.py b/cardinal_pythonlib/maths_numpy.py index f26c748..1c46358 100644 --- a/cardinal_pythonlib/maths_numpy.py +++ b/cardinal_pythonlib/maths_numpy.py @@ -229,7 +229,7 @@ def _test_softmax() -> None: [1, 1, 1, 1, 1.01], [1, 2, 3, 4, 5], [1, 2, 3, 4, 1000], - [1, 2, 3, 4, 5.0 ** 400], + [1, 2, 3, 4, 5.0**400], ] betas = [0, 0.5, 1, 2, 10, 100] for x in arrays: diff --git a/cardinal_pythonlib/module_version.py b/cardinal_pythonlib/module_version.py index 9e21988..ea9a67b 100644 --- a/cardinal_pythonlib/module_version.py +++ b/cardinal_pythonlib/module_version.py @@ -40,7 +40,7 @@ from semantic_version import Version # noinspection PyUnresolvedReferences -import cardinal_pythonlib.ensure_test_executed_correctly +import cardinal_pythonlib.ensure_test_executed_correctly # noqa: F401 if sys.version_info > (3, 8): # Python 3.8 or higher diff --git a/cardinal_pythonlib/nhs.py b/cardinal_pythonlib/nhs.py index 31c7a3d..dfd690b 100644 --- a/cardinal_pythonlib/nhs.py +++ b/cardinal_pythonlib/nhs.py @@ -163,7 +163,7 @@ def test_nhs_rng(n: int = 100) -> None: def generate_nhs_number_from_first_9_digits( - first9digits: str + first9digits: str, ) -> Optional[int]: """ Returns a valid NHS number, as an ``int``, given the first 9 digits. diff --git a/cardinal_pythonlib/openxml/find_recovered_openxml.py b/cardinal_pythonlib/openxml/find_recovered_openxml.py index ff5f193..deca194 100644 --- a/cardinal_pythonlib/openxml/find_recovered_openxml.py +++ b/cardinal_pythonlib/openxml/find_recovered_openxml.py @@ -123,7 +123,8 @@ (SOURCE_STDOUT, " or ENTER (try reading this split again): ", "q\n"), ( SOURCE_STDERR, - "zip: malloc.c:2394: sysmalloc: Assertion `(old_top == initial_top (av) " + "zip: malloc.c:2394: sysmalloc: " + "Assertion `(old_top == initial_top (av) " "&& old_size == 0) || ((unsigned long) (old_size) >= MINSIZE && " "prev_inuse (old_top) && ((unsigned long) old_end & (pagesize - 1)) " "== 0)' failed.", @@ -546,7 +547,9 @@ def main() -> None: "filename": filename, "filetypes": filetypes, "move_to": args.move_to, - "delete_if_not_specified_file_type": args.delete_if_not_specified_file_type, + "delete_if_not_specified_file_type": ( + args.delete_if_not_specified_file_type + ), "show_zip_output": args.show_zip_output, } # log.critical("start") diff --git a/cardinal_pythonlib/platformfunc.py b/cardinal_pythonlib/platformfunc.py index 816d952..a7e0acd 100644 --- a/cardinal_pythonlib/platformfunc.py +++ b/cardinal_pythonlib/platformfunc.py @@ -48,39 +48,6 @@ def fix_windows_utf8_output() -> None: # Python 3 only now, so nothing to do return - # if six.PY3: - # return - # reload_module(sys) - # # noinspection PyUnresolvedReferences - # sys.setdefaultencoding('utf-8') - # # print sys.getdefaultencoding() - # - # if sys.platform == 'win32': - # try: - # import win32console - # except ImportError: - # win32console = None - # print( - # "Python Win32 Extensions module is required.\n " - # "You can download it from " - # "https://sourceforge.net/projects/pywin32/ " - # "(x86 and x64 builds are available)\n") - # exit(-1) - # # win32console implementation of SetConsoleCP does not return a value - # # CP_UTF8 = 65001 - # win32console.SetConsoleCP(65001) - # if win32console.GetConsoleCP() != 65001: - # raise RuntimeError("Cannot set console codepage to 65001 (UTF-8)") - # win32console.SetConsoleOutputCP(65001) - # if win32console.GetConsoleOutputCP() != 65001: - # raise RuntimeError("Cannot set console output codepage to 65001 " - # "(UTF-8)") - # - # sys.stdout = codecs.getwriter('utf8')(sys.stdout) - # sys.stderr = codecs.getwriter('utf8')(sys.stderr) - # # CHECK: does that modify the "global" sys.stdout? - # # You can't use "global sys.stdout"; that raises an error - def test_windows_utf8_output() -> None: """ @@ -262,7 +229,7 @@ def handle_line(line: str) -> Tuple[str, str]: # RNC: as function lines = gen_lines() # RNC # consume whatever output occurs until the tag is reached - consume(itertools.takewhile(lambda l: tag not in l, lines)) + consume(itertools.takewhile(lambda line: tag not in line, lines)) # ... RNC: note that itertools.takewhile() generates values not matching # the condition, but then consumes the condition value itself. So the # tag's already gone. Example: diff --git a/cardinal_pythonlib/probability.py b/cardinal_pythonlib/probability.py index 9f9efcc..19f32b6 100644 --- a/cardinal_pythonlib/probability.py +++ b/cardinal_pythonlib/probability.py @@ -87,7 +87,7 @@ def ln(x: float) -> float: # return math_ln(x) if x != 0 else MINUS_INFINITY timeit.timeit('(ln(x) for x in range(1, 100))', number=10000) # 0.012411762960255146 - + # try... except ValueError... timeit.timeit('(ln(x) for x in range(1, 100))', number=10000) # 0.007783170789480209 diff --git a/cardinal_pythonlib/process.py b/cardinal_pythonlib/process.py index 95b9e7b..bfbac32 100644 --- a/cardinal_pythonlib/process.py +++ b/cardinal_pythonlib/process.py @@ -26,6 +26,7 @@ """ +import os import shlex import subprocess import sys @@ -81,7 +82,7 @@ def get_pipe_series_output( # than the maximum specified. # print commands - processes = [] # type: List[subprocess.Popen] + processes: List[subprocess.Popen] = [] for i in range(len(commands)): if i == 0: # first processes processes.append( diff --git a/cardinal_pythonlib/psychiatry/drugs.py b/cardinal_pythonlib/psychiatry/drugs.py index ce1b9e5..0cffc75 100644 --- a/cardinal_pythonlib/psychiatry/drugs.py +++ b/cardinal_pythonlib/psychiatry/drugs.py @@ -560,8 +560,8 @@ def regex_to_sql_like( Args: regex_text: regular expression text to work with single_wildcard: SQL single wildcard, typically an underscore - zero_or_more_wildcard: SQL "zero/one/many" wildcard, probably always - a percent symbol + zero_or_more_wildcard: SQL "zero/one/many" wildcard, probably + always a percent symbol Returns: string for an SQL string literal @@ -1468,7 +1468,7 @@ def all_drugs_where( print([d.generic_name for d in non_ssri_antidep]) conventional_antidep = all_drugs_where(conventional_antidepressant=True) print([d.generic_name for d in conventional_antidep]) - """ + """ # noqa: E501 matching_drugs = [] # type: List[Drug] for drug in DRUGS: if drug.category_not_drug and not include_categories: diff --git a/cardinal_pythonlib/psychiatry/simhelpers.py b/cardinal_pythonlib/psychiatry/simhelpers.py index 4f2311d..46ade62 100644 --- a/cardinal_pythonlib/psychiatry/simhelpers.py +++ b/cardinal_pythonlib/psychiatry/simhelpers.py @@ -58,7 +58,8 @@ def _central_value(values: Union[List, Tuple], name: str = "?") -> Any: """ if not isinstance(values, (list, tuple)): raise ValueError( - f"Parameter {name!r} should have been list/tuple but was {values!r}" + f"Parameter {name!r} should have been list/tuple but was " + f"{values!r}" ) n = len(values) if n == 0: @@ -96,13 +97,13 @@ def gen_params_around_centre( Collins, 2019, PMID 31769410, Box 2/Figure 1B), because simulating a subject that chooses based on calculated probabilities introduces random noise. You will want to simulate a range of parameter combinations. - + You might want to explore the entire parameter space (in which case, see :func:`cardinal_pythonlib.iterhelp.product_dict`), but that may be computationally unfeasible -- creating the simulations takes time and energy, and analysing them to recover those parameters takes time and energy. - + An alternative strategy is to take central values of all parameters, and then vary one parameter at a time about those central values. For example, if you have n = 6 parameters with k = 5 possible values each, the diff --git a/cardinal_pythonlib/psychiatry/timeline.py b/cardinal_pythonlib/psychiatry/timeline.py index 4f4783e..97265cf 100755 --- a/cardinal_pythonlib/psychiatry/timeline.py +++ b/cardinal_pythonlib/psychiatry/timeline.py @@ -223,7 +223,7 @@ # reticulate_1.11.1-9000 [see sessionInfo()]... # ... yes, that fixes it. -""" +""" # noqa: E501 # ============================================================================= # Imports diff --git a/cardinal_pythonlib/psychiatry/treatment_resistant_depression.py b/cardinal_pythonlib/psychiatry/treatment_resistant_depression.py index 5396349..0a58508 100755 --- a/cardinal_pythonlib/psychiatry/treatment_resistant_depression.py +++ b/cardinal_pythonlib/psychiatry/treatment_resistant_depression.py @@ -467,7 +467,7 @@ def _get_patient_result(_patient_id: str) -> Optional[DataFrame]: def test_two_antidepressant_episodes( - n_sets: int = TEST_PATIENT_MULTIPLE + n_sets: int = TEST_PATIENT_MULTIPLE, ) -> None: fluox = "fluoxetine" cital = "citalopram" diff --git a/cardinal_pythonlib/pyramid/requests.py b/cardinal_pythonlib/pyramid/requests.py index 8afdefb..d444d3c 100644 --- a/cardinal_pythonlib/pyramid/requests.py +++ b/cardinal_pythonlib/pyramid/requests.py @@ -63,7 +63,7 @@ def gen_accept_encoding_definitions( - accept_encoding: str + accept_encoding: str, ) -> Generator[str, None, None]: """ For a given HTTP ``Accept-Encoding`` field value, generate encoding @@ -150,7 +150,7 @@ def gen_content_encodings(request: Request) -> Generator[str, None, None]: def gen_content_encodings_reversed( - request: Request + request: Request, ) -> Generator[str, None, None]: """ Generates content encodings in reverse order -- that is, in the order diff --git a/cardinal_pythonlib/rate_limiting.py b/cardinal_pythonlib/rate_limiting.py index 8fc733c..0752c29 100644 --- a/cardinal_pythonlib/rate_limiting.py +++ b/cardinal_pythonlib/rate_limiting.py @@ -30,8 +30,6 @@ from time import perf_counter, sleep from typing import Any, Callable, Optional, Union -from cardinal_pythonlib.logs import main_only_quicksetup_rootlogger - log = logging.getLogger(__name__) FuncType = Callable[..., Any] diff --git a/cardinal_pythonlib/rnc_db.py b/cardinal_pythonlib/rnc_db.py index 9d9f554..0eaf3f2 100644 --- a/cardinal_pythonlib/rnc_db.py +++ b/cardinal_pythonlib/rnc_db.py @@ -188,7 +188,7 @@ ) from cardinal_pythonlib.logs import get_brace_style_log_with_null_handler -from cardinal_pythonlib.sql.validation import * +from cardinal_pythonlib.sql.validation import split_long_sqltype # 1. An ODBC driver try: @@ -959,7 +959,7 @@ def get_sql_select_all_non_pk_fields_by_pk( table: str, fieldlist: Sequence[str], delims: Tuple[str, str] = ("", "") ) -> str: """Returns SQL: - SELECT [all but the first field] WHERE [the first field] = ? + SELECT [all but the first field] WHERE [the first field] = ? """ return ( "SELECT " @@ -979,7 +979,7 @@ def get_sql_select_all_fields_by_key( delims: Tuple[str, str] = ("", ""), ) -> str: """Returns SQL: - SELECT [all fields in the fieldlist] WHERE [keyname] = ? + SELECT [all fields in the fieldlist] WHERE [keyname] = ? """ return ( "SELECT " @@ -1010,7 +1010,8 @@ def get_sql_insert( def get_sql_insert_or_update( table: str, fieldlist: Sequence[str], delims: Tuple[str, str] = ("", "") ) -> str: - """Returns ?-marked SQL for an INSERT-or-if-duplicate-key-UPDATE statement. + """ + Returns ?-marked SQL for an INSERT-or-if-duplicate-key-UPDATE statement. """ # https://stackoverflow.com/questions/4205181 return """ @@ -1104,8 +1105,8 @@ def debug_object(obj: T) -> str: """Prints key/value pairs for an object's dictionary.""" pairs = [] for k, v in vars(obj).items(): - pairs.append(u"{}={}".format(k, v)) - return u", ".join(pairs) + pairs.append("{}={}".format(k, v)) + return ", ".join(pairs) def dump_database_object(obj: T, fieldlist: Iterable[str]) -> None: @@ -1113,7 +1114,7 @@ def dump_database_object(obj: T, fieldlist: Iterable[str]) -> None: log.info(_LINE_EQUALS) log.info("DUMP OF: {}", obj) for f in fieldlist: - log.info(u"{f}: {v}", f=f, v=getattr(obj, f)) + log.info("{f}: {v}", f=f, v=getattr(obj, f)) log.info(_LINE_EQUALS) @@ -1138,8 +1139,8 @@ def create_object_from_list( **kwargs ) -> T: """ - Create an object by instantiating ``cls(*args, **kwargs)`` and assigning the - values in ``valuelist`` to the fields in ``fieldlist``. + Create an object by instantiating ``cls(*args, **kwargs)`` and assigning + the values in ``valuelist`` to the fields in ``fieldlist``. If ``construct_with_pk`` is ``True``, initialize with ``cls(valuelist[0], *args, **kwargs)`` @@ -1293,7 +1294,7 @@ def reconfigure_jaydebeapi() -> None: log.warning("Old jaydebeapi version") # noinspection PyUnresolvedReferences,PyProtectedMember converters = jaydebeapi.dbapi2._DEFAULT_CONVERTERS - except: # nopep8 + except Exception: raise AssertionError( "Don't know how to hook into this version of JayDeBeApi" ) @@ -1499,7 +1500,7 @@ def get_database( autocommit=autocommit, # if False, need to commit ) return db - except: # nopep8 + except Exception: if securely: raise NoDatabaseError( "Problem opening or reading from database {}; details " @@ -1517,7 +1518,7 @@ def get_database_from_configparser( dbc = DatabaseConfig(parser, section) db = dbc.get_database(securely=securely) return db - except: # nopep8 + except Exception: if securely: raise NoDatabaseError( "Problem opening or reading from database {}; details " @@ -1573,8 +1574,8 @@ def connect( use_unicode: bool = True, ) -> bool: """ - engine: access, mysql, sqlserver - interface: mysql, odbc, jdbc + engine: access, mysql, sqlserver + interface: mysql, odbc, jdbc """ # Catch all exceptions, so the error-catcher never shows a password. @@ -1811,8 +1812,10 @@ def _connect( dr=driver, s=host, db=database, u=user ) ) - connectstring = "DRIVER={};SERVER={};DATABASE={};UID={};PWD={}".format( - driver, host, database, user, password + connectstring = ( + "DRIVER={};SERVER={};DATABASE={};UID={};PWD={}".format( + driver, host, database, user, password + ) ) # noinspection PyUnresolvedReferences self.db = pyodbc.connect(connectstring, unicode_results=True) @@ -2117,7 +2120,7 @@ def insert_record( new_pk = get_pk_of_last_insert(cursor) log.debug("Record inserted.") return new_pk - except: # nopep8 + except Exception: log.exception("insert_record: Failed to insert record.") raise @@ -2165,7 +2168,7 @@ def insert_record_by_dict( new_pk = get_pk_of_last_insert(cursor) log.debug("Record inserted.") return new_pk - except: # nopep8 + except Exception: log.exception("insert_record_by_dict: Failed to insert record.") raise @@ -2191,7 +2194,7 @@ def insert_multiple_records( # http://www.python.org/dev/peps/pep-0249/ log.debug("Records inserted.") return cursor.rowcount - except: # nopep8 + except Exception: log.exception("insert_multiple_records: Failed to insert records.") raise @@ -2203,7 +2206,7 @@ def db_exec_with_cursor(self, cursor, sql: str, *args) -> int: debug_sql(sql, args) cursor.execute(sql, args) return cursor.rowcount - except: # nopep8 + except Exception: log.exception("db_exec_with_cursor: SQL was: " + sql) raise # MySQLdb: @@ -2233,7 +2236,7 @@ def db_exec_literal(self, sql: str) -> int: try: cursor.execute(sql) return cursor.rowcount - except: # nopep8 + except Exception: log.exception("db_exec_literal: SQL was: " + sql) raise @@ -2265,7 +2268,7 @@ def fetchone(self, sql: str, *args) -> Optional[Sequence[Any]]: self.db_exec_with_cursor(cursor, sql, *args) try: return cursor.fetchone() - except: # nopep8 + except Exception: log.exception("fetchone: SQL was: " + sql) raise @@ -2277,7 +2280,7 @@ def fetchall(self, sql: str, *args) -> Sequence[Sequence[Any]]: try: rows = cursor.fetchall() return rows - except: # nopep8 + except Exception: log.exception("fetchall: SQL was: " + sql) raise @@ -2291,7 +2294,7 @@ def gen_fetchall(self, sql: str, *args) -> Iterator[Sequence[Any]]: while row is not None: yield row row = cursor.fetchone() - except: # nopep8 + except Exception: log.exception("gen_fetchall: SQL was: " + sql) raise @@ -2305,7 +2308,7 @@ def gen_fetchfirst(self, sql: str, *args) -> Iterator[Any]: while row is not None: yield row[0] row = cursor.fetchone() - except: # nopep8 + except Exception: log.exception("gen_fetchfirst: SQL was: " + sql) raise @@ -2320,7 +2323,7 @@ def fetchall_with_fieldnames( rows = cursor.fetchall() fieldnames = [i[0] for i in cursor.description] return rows, fieldnames - except: # nopep8 + except Exception: log.exception("fetchall_with_fieldnames: SQL was: " + sql) raise @@ -2337,7 +2340,7 @@ def fetchall_as_dictlist(self, sql: str, *args) -> List[Dict[str, Any]]: for r in rows: dictlist.append(dict(zip(fieldnames, r))) return dictlist - except: # nopep8 + except Exception: log.exception("fetchall_as_dictlist: SQL was: " + sql) raise @@ -2353,7 +2356,7 @@ def fetch_fieldnames(self, sql: str, *args) -> List[str]: self.db_exec_with_cursor(cursor, sql, *args) try: return [i[0] for i in cursor.description] - except: # nopep8 + except Exception: log.exception("fetch_fieldnames: SQL was: " + sql) raise @@ -2696,7 +2699,7 @@ def create_fulltext_index( @staticmethod def fieldnames_from_fieldspeclist( - fieldspeclist: FIELDSPECLIST_TYPE + fieldspeclist: FIELDSPECLIST_TYPE, ) -> List[str]: """Returns fieldnames from a field specification list.""" return [x["name"] for x in fieldspeclist] diff --git a/cardinal_pythonlib/snomed.py b/cardinal_pythonlib/snomed.py index adf8185..7d2063b 100644 --- a/cardinal_pythonlib/snomed.py +++ b/cardinal_pythonlib/snomed.py @@ -195,7 +195,7 @@ def test(s): ret.append(r"\\") elif ord(c) < 32: # two-digit hex format, e.g. \x1F for ASCII 31 - ret.append(fr"\x{ord(c):02X}") + ret.append(rf"\x{ord(c):02X}") else: ret.append(c) ret.append(dquote) diff --git a/cardinal_pythonlib/sphinxtools.py b/cardinal_pythonlib/sphinxtools.py index fd6a2d4..3d196cb 100644 --- a/cardinal_pythonlib/sphinxtools.py +++ b/cardinal_pythonlib/sphinxtools.py @@ -47,15 +47,18 @@ sep, splitext, ) -from typing import Dict, Iterable, List, NoReturn, Union +from typing import Dict, Iterable, List, NoReturn, TYPE_CHECKING, Union from cardinal_pythonlib.fileops import mkdir_p, relative_filename_within_dir from cardinal_pythonlib.logs import BraceStyleAdapter from cardinal_pythonlib.reprfunc import auto_repr -from pygments.lexer import Lexer from pygments.lexers import get_lexer_for_filename from pygments.util import ClassNotFound +if TYPE_CHECKING: + from pygments.lexer import Lexer + + log = BraceStyleAdapter(logging.getLogger(__name__)) @@ -172,6 +175,7 @@ def filename_matches_glob(filename: str, globtext: str) -> bool: # Otherwise: return False + # ============================================================================= # FileToAutodocument # ============================================================================= @@ -771,7 +775,7 @@ def add_source_files( source_filename ), method=method, - source_rst_title_style_python=source_rst_title_style_python, + source_rst_title_style_python=source_rst_title_style_python, # noqa: E501 pygments_language_override=pygments_language_override, ) ) diff --git a/cardinal_pythonlib/spreadsheets.py b/cardinal_pythonlib/spreadsheets.py index 5558d12..b72743e 100644 --- a/cardinal_pythonlib/spreadsheets.py +++ b/cardinal_pythonlib/spreadsheets.py @@ -892,8 +892,8 @@ def __init__(self, sheetholder: SheetHolder, row: int) -> None: self.sheetholder = sheetholder self.row = row # zero-based index of our row self._next_col = ( - 0 - ) # zero-based column index of the next column to read # noqa + 0 # zero-based column index of the next column to read # noqa + ) # ------------------------------------------------------------------------- # Information diff --git a/cardinal_pythonlib/sql/literals.py b/cardinal_pythonlib/sql/literals.py index ef6b8dc..beece64 100644 --- a/cardinal_pythonlib/sql/literals.py +++ b/cardinal_pythonlib/sql/literals.py @@ -42,8 +42,8 @@ def sql_string_literal(text: str) -> str: """ - Transforms text into its ANSI SQL-quoted version, e.g. (in Python ``repr()`` - format): + Transforms text into its ANSI SQL-quoted version, e.g. (in Python + ``repr()`` format): .. code-block:: none diff --git a/cardinal_pythonlib/sql/sql_grammar_mssql.py b/cardinal_pythonlib/sql/sql_grammar_mssql.py index 4bcf1ba..c7f8b44 100644 --- a/cardinal_pythonlib/sql/sql_grammar_mssql.py +++ b/cardinal_pythonlib/sql/sql_grammar_mssql.py @@ -393,7 +393,8 @@ class SqlGrammarMSSQLServer(SqlGrammar): opAssoc.LEFT, ), ((BETWEEN, AND), TERNARY_OP, opAssoc.LEFT), - # CASE handled above (hoping precedence is not too much of a problem) + # CASE handled above (hoping precedence is not too much of a + # problem) (NOT, UNARY_OP, opAssoc.RIGHT), (AND | "&&" | OR | "||" | ":=", BINARY_OP, opAssoc.LEFT), ], diff --git a/cardinal_pythonlib/sql/sql_grammar_mysql.py b/cardinal_pythonlib/sql/sql_grammar_mysql.py index 7467c71..8d9e1c0 100644 --- a/cardinal_pythonlib/sql/sql_grammar_mysql.py +++ b/cardinal_pythonlib/sql/sql_grammar_mysql.py @@ -380,8 +380,9 @@ class SqlGrammarMySQL(SqlGrammar): argument_list = ( delimitedList(expr).setName("arglist").setParseAction(", ".join) ) - # ... we don't care about sub-parsing the argument list, so use combine=True - # or setParseAction: https://stackoverflow.com/questions/37926516 + # ... we don't care about sub-parsing the argument list, so use + # combine=True or setParseAction: + # https://stackoverflow.com/questions/37926516 function_call = Combine(function_name + LPAR) + argument_list + RPAR # http://dev.mysql.com/doc/refman/5.7/en/partitioning-selection.html @@ -516,7 +517,8 @@ class SqlGrammarMySQL(SqlGrammar): opAssoc.LEFT, ), ((BETWEEN, AND), TERNARY_OP, opAssoc.LEFT), - # CASE handled above (hoping precedence is not too much of a problem) + # CASE handled above (hoping precedence is not too much of a + # problem) (NOT, UNARY_OP, opAssoc.RIGHT), (AND | "&&" | XOR | OR | "||" | ":=", BINARY_OP, opAssoc.LEFT), ], @@ -577,7 +579,7 @@ class SqlGrammarMySQL(SqlGrammar): | BIT_OR | BIT_XOR | COUNT - | GROUP_CONCAT # also: special handling for COUNT(DISTINCT ...), see below + | GROUP_CONCAT # also: special handling for COUNT(DISTINCT ...), see below # noqa: E501 | MAX | MIN | STD diff --git a/cardinal_pythonlib/sqlalchemy/alembic_func.py b/cardinal_pythonlib/sqlalchemy/alembic_func.py index da7356d..7d54dac 100644 --- a/cardinal_pythonlib/sqlalchemy/alembic_func.py +++ b/cardinal_pythonlib/sqlalchemy/alembic_func.py @@ -289,51 +289,51 @@ def create_database_migration_numbered_style( n_sequence_chars: int = 4, ) -> None: """ - Create a new Alembic migration script. + Create a new Alembic migration script. - Alembic compares the **state of the database** to the **state of the - metadata**, and generates a migration that brings the former up to the - latter. (It does **not** compare the most recent revision to the current - metadata, so make sure your database is up to date with the most recent - revision before running this!) + Alembic compares the **state of the database** to the **state of the + metadata**, and generates a migration that brings the former up to the + latter. (It does **not** compare the most recent revision to the current + metadata, so make sure your database is up to date with the most recent + revision before running this!) - You **must check** that the autogenerated revisions are sensible. + You **must check** that the autogenerated revisions are sensible. - How does it know where to look for the database? + How does it know where to look for the database? - 1. This function changes into the directory of the Alembic ``.ini`` - file and calls the external program + 1. This function changes into the directory of the Alembic ``.ini`` + file and calls the external program - .. code-block:: bash + .. code-block:: bash - alembic -c ALEMBIC_INI_FILE revision --autogenerate -m MESSAGE --rev-id REVISION_ID + alembic -c ALEMBIC_INI_FILE revision --autogenerate -m MESSAGE --rev-id REVISION_ID - 2. The Alembic ``.ini`` file points (via the ``script_location`` - variable) to a directory containing your ``env.py``. Alembic loads - this script. + 2. The Alembic ``.ini`` file points (via the ``script_location`` + variable) to a directory containing your ``env.py``. Alembic loads + this script. - 3. That script typically works out the database URL and calls further - into the Alembic code. + 3. That script typically works out the database URL and calls further + into the Alembic code. - See https://alembic.zzzcomputing.com/en/latest/autogenerate.html. + See https://alembic.zzzcomputing.com/en/latest/autogenerate.html. - Regarding filenames: the default ``n_sequence_chars`` of 4 is like Django - and gives files with names like + Regarding filenames: the default ``n_sequence_chars`` of 4 is like Django + and gives files with names like - .. code-block:: none + .. code-block:: none - 0001_x.py, 0002_y.py, ... + 0001_x.py, 0002_y.py, ... - NOTE THAT TO USE A NON-STANDARD ALEMBIC VERSION TABLE, YOU MUST SPECIFY - THAT IN YOUR ``env.py`` (see e.g. CamCOPS). + NOTE THAT TO USE A NON-STANDARD ALEMBIC VERSION TABLE, YOU MUST SPECIFY + THAT IN YOUR ``env.py`` (see e.g. CamCOPS). - Args: - alembic_ini_file: filename of Alembic ``alembic.ini`` file - alembic_versions_dir: directory in which you keep your Python scripts, - one per Alembic revision - message: message to be associated with this revision - n_sequence_chars: number of numerical sequence characters to use in the - filename/revision (see above). + Args: + alembic_ini_file: filename of Alembic ``alembic.ini`` file + alembic_versions_dir: directory in which you keep your Python scripts, + one per Alembic revision + message: message to be associated with this revision + n_sequence_chars: number of numerical sequence characters to use in the + filename/revision (see above). """ # noqa file_regex = r"\d{" + str(n_sequence_chars) + r"}_\S*\.py$" diff --git a/cardinal_pythonlib/sqlalchemy/logs.py b/cardinal_pythonlib/sqlalchemy/logs.py index 67496a9..4a41f30 100644 --- a/cardinal_pythonlib/sqlalchemy/logs.py +++ b/cardinal_pythonlib/sqlalchemy/logs.py @@ -44,9 +44,9 @@ def pre_disable_sqlalchemy_extra_echo_log() -> None: sqlalchemy.engine.base.Engine.__init__() sqlalchemy.log.instance_logger() sqlalchemy.log.InstanceLogger.__init__() - # ... which checks that the logger has no handlers and if not calls: + # ... which checks that the logger has no handlers and if not calls: sqlalchemy.log._add_default_handler() - # ... which adds a handler to sys.stdout + # ... which adds a handler to sys.stdout """ log = logging.getLogger("sqlalchemy.engine.base.Engine") log.addHandler(logging.NullHandler()) diff --git a/cardinal_pythonlib/sqlalchemy/orm_inspect.py b/cardinal_pythonlib/sqlalchemy/orm_inspect.py index a422408..293e656 100644 --- a/cardinal_pythonlib/sqlalchemy/orm_inspect.py +++ b/cardinal_pythonlib/sqlalchemy/orm_inspect.py @@ -96,7 +96,7 @@ class type. This function ensures that such classes are converted to """ if isinstance(coltype, TypeEngine): return coltype - return coltype() # type: TypeEngine + return coltype() # ============================================================================= @@ -627,7 +627,7 @@ def get_pk_attrnames(obj) -> List[str]: def gen_columns_for_uninstrumented_class( - cls: Type + cls: Type, ) -> Generator[Tuple[str, Column], None, None]: """ Generate ``(attr_name, Column)`` tuples from an UNINSTRUMENTED class, i.e. @@ -675,7 +675,7 @@ def colname_to_attrname_dict(cls) -> Dict[str, str]: def gen_relationships( - obj + obj, ) -> Generator[Tuple[str, RelationshipProperty, Type], None, None]: """ Yields tuples of ``(attrname, RelationshipProperty, related_class)`` diff --git a/cardinal_pythonlib/sqlalchemy/schema.py b/cardinal_pythonlib/sqlalchemy/schema.py index 5dc3e07..5f9b9b7 100644 --- a/cardinal_pythonlib/sqlalchemy/schema.py +++ b/cardinal_pythonlib/sqlalchemy/schema.py @@ -120,8 +120,8 @@ class SqlaColumnInspectionInfo(object): """ Class to represent information from inspecting a database column. - A clearer way of getting information than the plain ``dict`` that SQLAlchemy - uses. + A clearer way of getting information than the plain ``dict`` that + SQLAlchemy uses. """ def __init__(self, sqla_info_dict: Dict[str, Any]) -> None: @@ -333,9 +333,8 @@ def mssql_get_pk_index_name( AND s.name = :schemaname """ ).bindparams(tablename=tablename, schemaname=schemaname) - with contextlib.closing( - engine.execute(query) - ) as result: # type: ResultProxy + with contextlib.closing(engine.execute(query)) as result: + result: ResultProxy row = result.fetchone() return row[0] if row else "" @@ -849,7 +848,7 @@ def get_sqla_coltype_from_dialect_str( except TypeError: return cls() - except: + except Exception: # noinspection PyUnresolvedReferences raise ValueError( f"Failed to convert SQL type {coltype!r} in dialect " diff --git a/cardinal_pythonlib/tools/explore_clang_format_config.py b/cardinal_pythonlib/tools/explore_clang_format_config.py index 5fdf657..92c7f28 100644 --- a/cardinal_pythonlib/tools/explore_clang_format_config.py +++ b/cardinal_pythonlib/tools/explore_clang_format_config.py @@ -81,7 +81,7 @@ def clang_format( subprocess.check_call( [ clang_format_exe, - f"-style=file", # read ./.clang-format; see above + "-style=file", # read ./.clang-format; see above "-i", # edit in place dest, ] @@ -144,7 +144,7 @@ def explore_clang_format( log.info(f"Source = {source_filenames}") _ = monitor_diff([working_src, current_cpp, last_cpp], meld_exe) watcher = FileWatcher([config_filename] + source_filenames) - log.info(f"Watching config/source files; press Ctrl-C to abort") + log.info("Watching config/source files; press Ctrl-C to abort") while True: time.sleep(sleep_time_s) changed_filenames = watcher.changed() @@ -186,24 +186,30 @@ def main() -> None: ) parser.add_argument( "config", - help=f"YAML clang-format config file. Create one with e.g.: " - f"clang-format --style=WebKit --dump-config > myconfig.yaml", + help=( + "YAML clang-format config file. Create one with e.g.: " + "clang-format --style=WebKit --dump-config > myconfig.yaml" + ), ) parser.add_argument( "source", nargs="+", - help=f"C++ source (.cpp, .h) file(s) to play with (if you specify " - f"several, they will be concatenated)", + help=( + "C++ source (.cpp, .h) file(s) to play with (if you specify " + "several, they will be concatenated)" + ), ) parser.add_argument( "--clangformat", - help=f"clang-format executable (see https://apt.llvm.org/ and " - f"consider: 'sudo apt install clang-format-14')", + help=( + "clang-format executable (see https://apt.llvm.org/ and " + "consider: 'sudo apt install clang-format-14')" + ), default=shutil.which("clang-format"), ) parser.add_argument( "--meld", - help=f"meld executable (consider: 'sudo apt install meld')", + help="meld executable (consider: 'sudo apt install meld')", default=shutil.which("meld"), ) args = parser.parse_args() diff --git a/cardinal_pythonlib/ui.py b/cardinal_pythonlib/ui.py index 0059e0b..91f9bcd 100644 --- a/cardinal_pythonlib/ui.py +++ b/cardinal_pythonlib/ui.py @@ -30,7 +30,10 @@ # For back compatibility (these functions used to be in this file): # noinspection PyUnresolvedReferences -from cardinal_pythonlib.ui_commandline import ask_user, ask_user_password +from cardinal_pythonlib.ui_commandline import ( # noqa: F401 + ask_user, + ask_user_password, +) try: import tkinter diff --git a/cardinal_pythonlib/wsgi/request_logging_mw.py b/cardinal_pythonlib/wsgi/request_logging_mw.py index 101a4e2..8854ad5 100644 --- a/cardinal_pythonlib/wsgi/request_logging_mw.py +++ b/cardinal_pythonlib/wsgi/request_logging_mw.py @@ -114,14 +114,16 @@ def __call__( ) except KeyError: forwarded_for = "" - request_details = '{remote}{fwd}: "{method} {path}{qmark}{query} {proto}"'.format( - remote=environ.get(WsgiEnvVar.REMOTE_ADDR, ""), - fwd=forwarded_for, - method=environ.get(WsgiEnvVar.REQUEST_METHOD, ""), - path=environ.get(WsgiEnvVar.PATH_INFO, ""), - qmark="?" if query_string else "", - query=query_string, - proto=environ.get(WsgiEnvVar.SERVER_PROTOCOL, ""), + request_details = ( + '{remote}{fwd}: "{method} {path}{qmark}{query} {proto}"'.format( + remote=environ.get(WsgiEnvVar.REMOTE_ADDR, ""), + fwd=forwarded_for, + method=environ.get(WsgiEnvVar.REQUEST_METHOD, ""), + path=environ.get(WsgiEnvVar.PATH_INFO, ""), + qmark="?" if query_string else "", + query=query_string, + proto=environ.get(WsgiEnvVar.SERVER_PROTOCOL, ""), + ) ) msg_parts = [] # type: List[str] if self.show_request_immediately: diff --git a/docs/create_all_autodocs.py b/docs/create_all_autodocs.py index afd3d52..1bfca4b 100755 --- a/docs/create_all_autodocs.py +++ b/docs/create_all_autodocs.py @@ -75,8 +75,11 @@ def make_autodoc(make: bool, destroy_first: bool) -> None: log.info("Deleting directory {!r}", AUTODOC_DIR) rmtree(AUTODOC_DIR) else: - log.warning("Would delete directory {!r} (not doing so as in mock " - "mode)", AUTODOC_DIR) + log.warning( + "Would delete directory {!r} (not doing so as in mock " + "mode)", + AUTODOC_DIR, + ) top_idx = AutodocIndex( autodoc_rst_root_dir=AUTODOC_DIR, highest_code_dir=CODE_ROOT_DIR, @@ -100,22 +103,24 @@ def make_autodoc(make: bool, destroy_first: bool) -> None: def main() -> None: parser = argparse.ArgumentParser() parser.add_argument( - "--make", action="store_true", - help="Do things! Otherwise will just show its intent.") - parser.add_argument( - "--destroy_first", action="store_true", - help="Destroy all existing autodocs first") + "--make", + action="store_true", + help="Do things! Otherwise will just show its intent.", + ) parser.add_argument( - "--verbose", action="store_true", - help="Be verbose") + "--destroy_first", + action="store_true", + help="Destroy all existing autodocs first", + ) + parser.add_argument("--verbose", action="store_true", help="Be verbose") args = parser.parse_args() main_only_quicksetup_rootlogger( - level=logging.DEBUG if args.verbose else logging.INFO) + level=logging.DEBUG if args.verbose else logging.INFO + ) - make_autodoc(make=args.make, - destroy_first=args.destroy_first) + make_autodoc(make=args.make, destroy_first=args.destroy_first) -if __name__ == '__main__': +if __name__ == "__main__": main() diff --git a/docs/rebuild_docs.py b/docs/rebuild_docs.py index 5073420..4b37cbb 100755 --- a/docs/rebuild_docs.py +++ b/docs/rebuild_docs.py @@ -28,6 +28,7 @@ import shutil import subprocess import sys + if sys.version_info[0] < 3: raise AssertionError("Need Python 3") @@ -37,7 +38,7 @@ DEST_DIRS = [] -if __name__ == '__main__': +if __name__ == "__main__": # Remove anything old for destdir in [BUILD_HTML_DIR] + DEST_DIRS: print(f"Deleting directory {destdir!r}") diff --git a/docs/source/conf.py b/docs/source/conf.py index 4813d02..3a34f55 100644 --- a/docs/source/conf.py +++ b/docs/source/conf.py @@ -18,6 +18,7 @@ try: # noinspection PyUnresolvedReferences import django + # noinspection PyUnresolvedReferences from django.conf import settings except ImportError: @@ -29,10 +30,10 @@ # -- Project information ----------------------------------------------------- -project = 'cardinal_pythonlib' +project = "cardinal_pythonlib" # noinspection PyShadowingBuiltins -copyright = '2009-2020, Rudolf Cardinal' -author = 'Rudolf Cardinal' +copyright = "2009-2020, Rudolf Cardinal" +author = "Rudolf Cardinal" # The short X.Y version version = VERSION_STRING @@ -50,26 +51,26 @@ # extensions coming with Sphinx (named 'sphinx.ext.*') or your custom # ones. extensions = [ - 'sphinx.ext.autodoc', - 'sphinx.ext.doctest', - 'sphinx.ext.todo', - 'sphinx.ext.imgmath', - 'sphinx.ext.viewcode', - 'sphinx.ext.napoleon', - 'sphinx_paramlinks', + "sphinx.ext.autodoc", + "sphinx.ext.doctest", + "sphinx.ext.todo", + "sphinx.ext.imgmath", + "sphinx.ext.viewcode", + "sphinx.ext.napoleon", + "sphinx_paramlinks", ] # Add any paths that contain templates here, relative to this directory. -templates_path = ['_templates'] +templates_path = ["_templates"] # The suffix(es) of source filenames. # You can specify multiple suffix as a list of string: # # source_suffix = ['.rst', '.md'] -source_suffix = '.rst' +source_suffix = ".rst" # The master toctree document. -master_doc = 'index' +master_doc = "index" # The language for content autogenerated by Sphinx. Refer to documentation # for a list of supported languages. @@ -84,7 +85,7 @@ exclude_patterns = [] # The name of the Pygments (syntax highlighting) style to use. -pygments_style = 'sphinx' +pygments_style = "sphinx" # -- Options for HTML output ------------------------------------------------- @@ -93,7 +94,7 @@ # a list of builtin themes. # html_theme = 'classic' -html_theme = 'sphinx_rtd_theme' # pip install sphinx_rtd_theme +html_theme = "sphinx_rtd_theme" # pip install sphinx_rtd_theme # Theme options are theme-specific and customize the look and feel of a theme # further. For a list of options available for each theme, see the @@ -120,7 +121,7 @@ # -- Options for HTMLHelp output --------------------------------------------- # Output file base name for HTML help builder. -htmlhelp_basename = 'cardinal_pythonlibdoc' +htmlhelp_basename = "cardinal_pythonlibdoc" # -- Options for LaTeX output ------------------------------------------------ @@ -129,15 +130,12 @@ # The paper size ('letterpaper' or 'a4paper'). # # 'papersize': 'letterpaper', - # The font size ('10pt', '11pt' or '12pt'). # # 'pointsize': '10pt', - # Additional stuff for the LaTeX preamble. # # 'preamble': '', - # Latex figure (float) alignment # # 'figure_align': 'htbp', @@ -147,8 +145,13 @@ # (source start file, target name, title, # author, documentclass [howto, manual, or own class]). latex_documents = [ - (master_doc, 'cardinal_pythonlib.tex', 'cardinal\\_pythonlib Documentation', - 'Rudolf Cardinal', 'manual'), + ( + master_doc, + "cardinal_pythonlib.tex", + "cardinal\\_pythonlib Documentation", + "Rudolf Cardinal", + "manual", + ), ] @@ -157,8 +160,13 @@ # One entry per manual page. List of tuples # (source start file, name, description, authors, manual section). man_pages = [ - (master_doc, 'cardinal_pythonlib', 'cardinal_pythonlib Documentation', - [author], 1) + ( + master_doc, + "cardinal_pythonlib", + "cardinal_pythonlib Documentation", + [author], + 1, + ) ] @@ -168,9 +176,15 @@ # (source start file, target name, title, author, # dir menu entry, description, category) texinfo_documents = [ - (master_doc, 'cardinal_pythonlib', 'cardinal_pythonlib Documentation', - author, 'cardinal_pythonlib', 'One line description of project.', - 'Miscellaneous'), + ( + master_doc, + "cardinal_pythonlib", + "cardinal_pythonlib Documentation", + author, + "cardinal_pythonlib", + "One line description of project.", + "Miscellaneous", + ), ] @@ -184,7 +198,7 @@ # -- Options for autodoc extension ------------------------------------------- -autoclass_content = 'both' +autoclass_content = "both" # https://stackoverflow.com/questions/5599254/how-to-use-sphinxs-autodoc-to-document-a-classs-init-self-method # noqa diff --git a/setup.cfg b/setup.cfg index a40f10d..7985ef3 100644 --- a/setup.cfg +++ b/setup.cfg @@ -7,3 +7,18 @@ filterwarnings = testpaths = cardinal_pythonlib + +[flake8] +linters=pycodestyle,pyflakes +max-line-length=79 +# Not compatible with Black and not PEP8 apparently +# E203: Whitespace before ':' +extend-ignore = E203 + +[mypy] +# MyPy is a static type checker. It will not execute the code! +# pip install mypy +no_strict_optional = True +allow_redefinition = True +disallow_untyped_defs = True +disallow_incomplete_defs = True

Code style Use Black, as per https://github.com/RudolfCardinal/crate

2023-01-24T18:59:16

0.0

[]

soda-inria/sklearn-numba-dpex

soda-inria__sklearn-numba-dpex-67

3d306c12ccd7fc55b7f0305dca61ec6acb8d9918

diff --git a/.github/workflows/run_code_format_checks.yaml b/.github/workflows/run_code_format_checks.yaml index b2c67d7..5c5eada 100644 --- a/.github/workflows/run_code_format_checks.yaml +++ b/.github/workflows/run_code_format_checks.yaml @@ -18,4 +18,4 @@ jobs: pre-commit install - name: Run checks - run: pre-commit run + run: pre-commit run --all-files diff --git a/README.md b/README.md index 4a496d8..59fc676 100644 --- a/README.md +++ b/README.md @@ -9,13 +9,13 @@ workable state at the moment. This requires working with the following branch of scikit-learn which is itself not yet in a working state at the moment: -- `wip-engines` branch on https://github.com/ogrisel/scikit-learn +- `wip-engines` branch on https://github.com/ogrisel/scikit-learn ## List of Included Engines - `sklearn.cluster.KMeans` for the standard LLoyd's algorithm on dense data arrays. The k-means++ initialization scheme is not implemented yet but this is planned. - + No implementation of k-means for sparse input data or for the "Elkan" algorithm is included at the moment. @@ -201,7 +201,7 @@ To run the `scikit-learn` tests with the `sklearn_numba_dpex` engine you can run pytest --sklearn-engine-provider sklearn_numba_dpex --pyargs sklearn.cluster.tests.test_k_means ``` -(change the `--pyargs` option accordingly to select other test suites). +(change the `--pyargs` option accordingly to select other test suites). The `--sklearn-engine-provider sklearn_numba_dpex` option offered by the sklearn pytest plugin will automatically activate the `sklearn_numba_dpex` engine for all tests. diff --git a/benchmark/kmeans.py b/benchmark/kmeans.py index 10c964f..957493a 100644 --- a/benchmark/kmeans.py +++ b/benchmark/kmeans.py @@ -131,7 +131,8 @@ def _check_same_fit(self, estimator, name, max_iter, assert_allclose): if n_iter != max_iter: raise RuntimeError( f"The estimator {name} only ran {n_iter} iterations instead of " - f"max_iter={max_iter} iterations. " + runtime_error_message + f"max_iter={max_iter} iterations. " + + runtime_error_message ) if not assert_allclose: diff --git a/sklearn_numba_dpex/common/kernels.py b/sklearn_numba_dpex/common/kernels.py index 0b801dd..a491585 100644 --- a/sklearn_numba_dpex/common/kernels.py +++ b/sklearn_numba_dpex/common/kernels.py @@ -20,13 +20,12 @@ import math from functools import lru_cache -import numpy as np import dpctl.tensor as dpt import numba_dpex as dpex +import numpy as np from sklearn_numba_dpex.common._utils import check_power_of_2 - zero_idx = np.int64(0) @@ -114,7 +113,7 @@ def half_l2_norm( data, # IN (size0, size1) result, # OUT (size1,) ): - # fmt: on + # fmt: on col_idx = dpex.get_global_id(zero_idx) if col_idx >= size1: @@ -265,7 +264,7 @@ def partial_sum_reduction( summands, # IN (n_rows, sum_axis_size) result, # OUT (n_rows, math.ceil(size / (2 * work_group_size),) ): - # fmt: on + # fmt: on # NB: This kernel only perform a partial reduction group_id = dpex.get_group_id(zero_idx) local_work_id = dpex.get_local_id(zero_idx) @@ -294,7 +293,11 @@ def partial_sum_reduction( # We discard half of the remaining active work items at each iteration current_n_work_items = current_n_work_items // two_as_a_long if local_work_id < current_n_work_items: - add_cols_inplace(local_values, local_work_id + current_n_work_items, local_work_id) + add_cols_inplace( + local_values, + local_work_id + current_n_work_items, + local_work_id, + ) dpex.barrier(dpex.CLK_LOCAL_MEM_FENCE) @@ -320,8 +323,8 @@ def partial_sum_reduction( # NB: here memory for partial results is allocated ahead of time and will only # be garbage collected when the instance of `sum_reduction` is garbage # collected. Thus it can be more efficient to re-use a same instance of - # `sum_reduction` (e.g within iterations of a loop) since it avoid deallocation - # and reallocation every time. + # `sum_reduction` (e.g within iterations of a loop) since it avoid + # deallocation and reallocation every time. result = dpt.empty(result_shape, dtype=dtype, device=device) kernels_and_empty_tensors_pairs.append((kernel, result)) @@ -337,7 +340,8 @@ def sum_reduction(summands): @lru_cache def make_argmin_reduction_1d_kernel(size, work_group_size, device, dtype): - """Implement 1d argmin with the same strategy than for make_sum_reduction_2d_axis1_kernel.""" + """Implement 1d argmin with the same strategy than for + make_sum_reduction_2d_axis1_kernel.""" check_power_of_2(work_group_size) # Number of iteration in each execution of the kernel: @@ -361,14 +365,16 @@ def partial_argmin_reduction( # / (2 * work_group_size),) # )) ): - # fmt: on + # fmt: on group_id = dpex.get_group_id(zero_idx) local_work_id = dpex.get_local_id(zero_idx) first_work_id = local_work_id == zero_idx previous_result_size = previous_result.shape[zero_idx] has_previous_result = previous_result_size > one_idx - current_size = previous_result_size if has_previous_result else values.shape[zero_idx] + current_size = ( + previous_result_size if has_previous_result else values.shape[zero_idx] + ) local_argmin = dpex.local.array(work_group_size, dtype=np.int32) local_values = dpex.local.array(work_group_size, dtype=dtype) @@ -395,7 +401,7 @@ def partial_argmin_reduction( y = values[y_idx] if x < y or (x == y and x_idx < y_idx): local_argmin[local_work_id] = x_idx - local_values[local_work_id] = x + local_values[local_work_id] = x else: local_argmin[local_work_id] = y_idx local_values[local_work_id] = y @@ -408,10 +414,10 @@ def partial_argmin_reduction( local_x_idx = local_work_id local_y_idx = local_work_id + current_n_work_items - x= local_values[local_x_idx] - y= local_values[local_y_idx] + x = local_values[local_x_idx] + y = local_values[local_y_idx] - if x> y: + if x > y: local_values[local_x_idx] = y local_argmin[local_x_idx] = local_argmin[local_y_idx] diff --git a/sklearn_numba_dpex/common/random.py b/sklearn_numba_dpex/common/random.py index 36c63a2..6379568 100644 --- a/sklearn_numba_dpex/common/random.py +++ b/sklearn_numba_dpex/common/random.py @@ -1,20 +1,19 @@ -import warnings import random +import warnings from functools import lru_cache -from numba import float32, float64, uint32, int64, uint64 -import numpy as np - import dpctl import dpctl.tensor as dpt import numba_dpex as dpex +import numpy as np +from numba import float32, float64, int64, uint32, uint64 # This code is largely inspired from the numba.cuda.random module and the # numba/cuda/random.py where it's defined (v<0.57), and by the implementation of the # same algorithm in the package `randomgen`. # numba.cuda.random: https://github.com/numba/numba/blob/0.56.3/numba/cuda/random.py -# randomgen: https://github.com/bashtage/randomgen/blob/v1.23.1/randomgen/xoroshiro128.pyx +# randomgen: https://github.com/bashtage/randomgen/blob/v1.23.1/randomgen/xoroshiro128.pyx # noqa # NB1: we implement xoroshiro128++ rather than just xoroshiro128+, which is preferred. # Reference resource about PRNG: https://prng.di.unimi.it/ @@ -284,7 +283,7 @@ def _xoroshiro128pp_jump(states, state_idx): # result = s0 + s1 # s1 ^= s0 - # states[state_idx, zero_idx] = _rotl(s0, next_rot_p_1) ^ s1 ^ (s1 << next_rot_p_2) + # states[state_idx, zero_idx] = _rotl(s0, next_rot_p_1) ^ s1 ^ (s1 << next_rot_p_2) # noqa # states[state_idx, one_idx] = _rotl(s1, next_rot_p_3) # return result @@ -300,8 +299,7 @@ def _xoroshiro128pp_jump(states, state_idx): from_cpu_to_device = True except dpctl.SyclDeviceCreationError: warnings.warn( - "No CPU found, falling back random initiatlization to default " - "device." + "No CPU found, falling back random initiatlization to default device." ) states = dpt.empty( diff --git a/sklearn_numba_dpex/kmeans/drivers.py b/sklearn_numba_dpex/kmeans/drivers.py index 74434fb..c794d90 100644 --- a/sklearn_numba_dpex/kmeans/drivers.py +++ b/sklearn_numba_dpex/kmeans/drivers.py @@ -1,37 +1,35 @@ -import warnings import math +import warnings -import numpy as np import dpctl import dpctl.tensor as dpt import dpnp - -from sklearn_numba_dpex.common.random import ( - get_random_raw, - create_xoroshiro128pp_states, -) +import numpy as np from sklearn_numba_dpex.common.kernels import ( - make_initialize_to_zeros_2d_kernel, - make_initialize_to_zeros_3d_kernel, + make_argmin_reduction_1d_kernel, make_broadcast_division_1d_2d_kernel, make_half_l2_norm_2d_axis0_kernel, + make_initialize_to_zeros_2d_kernel, + make_initialize_to_zeros_3d_kernel, make_sum_reduction_2d_axis1_kernel, - make_argmin_reduction_1d_kernel, ) - +from sklearn_numba_dpex.common.random import ( + create_xoroshiro128pp_states, + get_random_raw, +) from sklearn_numba_dpex.kmeans.kernels import ( - make_lloyd_single_step_fixed_window_kernel, + make_centroid_shifts_kernel, make_compute_euclidean_distances_fixed_window_kernel, - make_label_assignment_fixed_window_kernel, make_compute_inertia_kernel, make_kmeansplusplus_init_kernel, - make_sample_center_candidates_kernel, make_kmeansplusplus_single_step_fixed_window_kernel, + make_label_assignment_fixed_window_kernel, + make_lloyd_single_step_fixed_window_kernel, + make_reduce_centroid_data_kernel, make_relocate_empty_clusters_kernel, + make_sample_center_candidates_kernel, make_select_samples_far_from_centroid_kernel, - make_centroid_shifts_kernel, - make_reduce_centroid_data_kernel, ) @@ -67,7 +65,7 @@ def lloyd( return_assignments=bool(verbose), sub_group_size=sub_group_size, global_mem_cache_size=global_mem_cache_size, - centroids_private_copies_max_cache_occupancy=centroids_private_copies_max_cache_occupancy, + centroids_private_copies_max_cache_occupancy=centroids_private_copies_max_cache_occupancy, # noqa work_group_size=max_work_group_size, dtype=compute_dtype, ) diff --git a/sklearn_numba_dpex/kmeans/engine.py b/sklearn_numba_dpex/kmeans/engine.py index 41338e5..ed82ff2 100644 --- a/sklearn_numba_dpex/kmeans/engine.py +++ b/sklearn_numba_dpex/kmeans/engine.py @@ -1,14 +1,13 @@ import warnings +from typing import Any, Dict -import numpy as np import dpctl import dpctl.tensor as dpt - +import numpy as np from sklearn.cluster._kmeans import KMeansCythonEngine +from sklearn.exceptions import DataConversionWarning, NotSupportedByEngineError -from sklearn.exceptions import NotSupportedByEngineError, DataConversionWarning - -from .drivers import lloyd, get_labels_inertia, get_euclidean_distances, kmeans_plusplus +from .drivers import get_euclidean_distances, get_labels_inertia, kmeans_plusplus, lloyd class _IgnoreSampleWeight: @@ -47,7 +46,7 @@ class KMeansEngine(KMeansCythonEngine): """ - _CONFIG = dict() + _CONFIG: Dict[str, Any] = dict() def __init__(self, estimator): self.device = dpctl.SyclDevice(self._CONFIG.get("device")) @@ -75,13 +74,15 @@ def prepare_fit(self, X, y=None, sample_weight=None): ) except Exception as e: raise NotSupportedByEngineError( - "The sklearn_nunmba_dpex engine for KMeans does not support the format of the inputed data." + "The sklearn_nunmba_dpex engine for KMeans does not support the format" + " of the inputed data." ) from e algorithm = estimator.algorithm if algorithm not in ("lloyd", "auto", "full"): raise NotSupportedByEngineError( - f"The sklearn_nunmba_dpex engine for KMeans only support the Lloyd algorithm, {algorithm} is not supported." + "The sklearn_nunmba_dpex engine for KMeans only support the Lloyd" + f" algorithm, {algorithm} is not supported." ) self.sample_weight = sample_weight @@ -226,15 +227,15 @@ def _set_dtype(self, X, sample_weight, cluster_centers): if (compute_dtype != np.float32) and (compute_dtype != np.float64): text = ( f"KMeans has been set to compute with type {compute_dtype} but only " - f"the types float32 and float64 are supported. The computations and " - f"outputs will default back to float32 type." + "the types float32 and float64 are supported. The computations and " + "outputs will default back to float32 type." ) output_dtype = compute_dtype = np.float32 elif (compute_dtype == np.float64) and not self.device.has_aspect_fp64: text = ( f"KMeans is set to compute with type {compute_dtype} but this type is " f"not supported by the device {self.device.name}. The computations " - f"will default back to float32 type." + "will default back to float32 type." ) compute_dtype = np.float32 @@ -244,8 +245,8 @@ def _set_dtype(self, X, sample_weight, cluster_centers): if copy: text += ( f" A copy of the data casted to type {compute_dtype} will be created. " - f"To save memory and suppress this warning, ensure that the dtype of " - f"the input data matches the dtype required for computations." + "To save memory and suppress this warning, ensure that the dtype of " + "the input data matches the dtype required for computations." ) warnings.warn(text, DataConversionWarning) # TODO: instead of triggering a copy on the host side, we could use the @@ -262,7 +263,7 @@ def _set_dtype(self, X, sample_weight, cluster_centers): f"The centers have been passed with type {cluster_centers_dtype} but " f"type {compute_dtype} is expected. A copy will be created with the " f"correct type {compute_dtype}. Ensure that the centers are passed " - f"with the correct dtype to save memory and suppress this warning.", + "with the correct dtype to save memory and suppress this warning.", DataConversionWarning, ) cluster_centers = cluster_centers.astype(compute_dtype) @@ -274,7 +275,7 @@ def _set_dtype(self, X, sample_weight, cluster_centers): f"sample_weight has been passed with type {sample_weight.dtype} but " f"type {compute_dtype} is expected. A copy will be created with the " f"correct type {compute_dtype}. Ensure that sample_weight is passed " - f"with the correct dtype to save memory and suppress this warning.", + "with the correct dtype to save memory and suppress this warning.", DataConversionWarning, ) sample_weight = sample_weight.astype(compute_dtype) diff --git a/sklearn_numba_dpex/kmeans/kernels/__init__.py b/sklearn_numba_dpex/kmeans/kernels/__init__.py index b0444d3..c4949d2 100644 --- a/sklearn_numba_dpex/kmeans/kernels/__init__.py +++ b/sklearn_numba_dpex/kmeans/kernels/__init__.py @@ -1,22 +1,21 @@ -from .lloyd_single_step import make_lloyd_single_step_fixed_window_kernel from .compute_euclidean_distances import ( make_compute_euclidean_distances_fixed_window_kernel, ) -from .compute_labels import make_label_assignment_fixed_window_kernel from .compute_inertia import make_compute_inertia_kernel +from .compute_labels import make_label_assignment_fixed_window_kernel from .kmeans_plusplus import ( make_kmeansplusplus_init_kernel, - make_sample_center_candidates_kernel, make_kmeansplusplus_single_step_fixed_window_kernel, + make_sample_center_candidates_kernel, ) +from .lloyd_single_step import make_lloyd_single_step_fixed_window_kernel from .utils import ( - make_relocate_empty_clusters_kernel, - make_select_samples_far_from_centroid_kernel, make_centroid_shifts_kernel, make_reduce_centroid_data_kernel, + make_relocate_empty_clusters_kernel, + make_select_samples_far_from_centroid_kernel, ) - __all__ = ( "make_lloyd_single_step_fixed_window_kernel", "make_compute_euclidean_distances_fixed_window_kernel", diff --git a/sklearn_numba_dpex/kmeans/kernels/_base_kmeans_kernel_funcs.py b/sklearn_numba_dpex/kmeans/kernels/_base_kmeans_kernel_funcs.py index f990484..ce6d0bc 100644 --- a/sklearn_numba_dpex/kmeans/kernels/_base_kmeans_kernel_funcs.py +++ b/sklearn_numba_dpex/kmeans/kernels/_base_kmeans_kernel_funcs.py @@ -210,7 +210,7 @@ def _initialize_window_of_centroids( window_of_centroids_half_l2_norms, # OUT results, # OUT ): - # fmt: on + # fmt: on _initialize_results(results) # The first `window_n_centroids` work items cooperate on loading the @@ -240,18 +240,18 @@ def _load_window_of_centroids_and_features( current_centroids_t, # IN centroids_window, # OUT ): - # fmt: on + # fmt: on # The work items in the work group cooperatively load the values in shared # memory. At each iteration, the work item loads one value and adjacent work # items load adjacent values. loading_feature_idx = first_feature_idx + window_loading_feature_offset - + if (loading_feature_idx < n_features) and ( loading_centroid_idx < n_clusters ): - value = current_centroids_t[loading_feature_idx, loading_centroid_idx] + value = current_centroids_t[loading_feature_idx, loading_centroid_idx] else: - value = zero + value = zero centroids_window[ window_loading_feature_offset, window_loading_centroid_idx @@ -276,7 +276,7 @@ def _accumulate_sum_of_ops( centroids_window, # IN result, # OUT ): - # fmt: on + # fmt: on for window_feature_idx in range(window_n_features): feature_idx = window_feature_idx + first_feature_idx @@ -289,7 +289,9 @@ def _accumulate_sum_of_ops( # For this given feature, loop on all centroids in the current # window and accumulate the partial results for window_centroid_idx in range(window_n_centroids): - centroid_value = centroids_window[window_feature_idx, window_centroid_idx] + centroid_value = ( + centroids_window[window_feature_idx, window_centroid_idx] + ) if accumulate_dot_product: result[window_centroid_idx] += centroid_value * X_value else: @@ -351,7 +353,7 @@ def update_closest_centroid( window_of_centroids_half_l2_norms, # IN dot_products, # IN ): - # fmt: on + # fmt: on for i in range(window_n_centroids): current_sample_pseudo_inertia = ( window_of_centroids_half_l2_norms[i] - dot_products[i] diff --git a/sklearn_numba_dpex/kmeans/kernels/compute_euclidean_distances.py b/sklearn_numba_dpex/kmeans/kernels/compute_euclidean_distances.py index 9cb8cc5..f8980fd 100644 --- a/sklearn_numba_dpex/kmeans/kernels/compute_euclidean_distances.py +++ b/sklearn_numba_dpex/kmeans/kernels/compute_euclidean_distances.py @@ -1,8 +1,8 @@ import math from functools import lru_cache -import numpy as np import numba_dpex as dpex +import numpy as np from ._base_kmeans_kernel_funcs import make_pairwise_ops_base_kernel_funcs @@ -60,7 +60,7 @@ def compute_distances( current_centroids_t, # IN READ-ONLY (n_features, n_clusters) euclidean_distances_t, # OUT (n_clusters, n_samples) ): - # fmt: on + # fmt: on sample_idx = dpex.get_global_id(zero_idx) local_work_id = dpex.get_local_id(zero_idx) @@ -112,10 +112,12 @@ def compute_distances( for i in range(window_n_centroids): centroid_idx = first_centroid_idx + i if centroid_idx < n_clusters: - euclidean_distances_t[first_centroid_idx + i, sample_idx] = math.sqrt(sq_distances[i]) + euclidean_distances_t[first_centroid_idx + i, sample_idx] = ( + math.sqrt(sq_distances[i]) + ) first_centroid_idx += window_n_centroids - + dpex.barrier(dpex.CLK_LOCAL_MEM_FENCE) global_size = (math.ceil(n_samples / work_group_size)) * (work_group_size) diff --git a/sklearn_numba_dpex/kmeans/kernels/compute_inertia.py b/sklearn_numba_dpex/kmeans/kernels/compute_inertia.py index 52244cb..59184dc 100644 --- a/sklearn_numba_dpex/kmeans/kernels/compute_inertia.py +++ b/sklearn_numba_dpex/kmeans/kernels/compute_inertia.py @@ -1,8 +1,8 @@ import math from functools import lru_cache -import numpy as np import numba_dpex as dpex +import numpy as np @lru_cache @@ -20,7 +20,7 @@ def compute_inertia( assignments_idx, # IN READ-ONLY (n_samples,) per_sample_inertia, # OUT (n_samples,) ): - # fmt: on + # fmt: on sample_idx = dpex.get_global_id(zero_idx) diff --git a/sklearn_numba_dpex/kmeans/kernels/compute_labels.py b/sklearn_numba_dpex/kmeans/kernels/compute_labels.py index 059a2cf..f1bdba1 100644 --- a/sklearn_numba_dpex/kmeans/kernels/compute_labels.py +++ b/sklearn_numba_dpex/kmeans/kernels/compute_labels.py @@ -1,8 +1,8 @@ import math from functools import lru_cache -import numpy as np import numba_dpex as dpex +import numpy as np from ._base_kmeans_kernel_funcs import ( make_pairwise_ops_base_kernel_funcs, @@ -68,7 +68,7 @@ def assignment( centroids_half_l2_norm, # IN (n_clusters,) assignments_idx, # OUT (n_samples,) ): - # fmt: on + # fmt: on sample_idx = dpex.get_global_id(zero_idx) local_work_id = dpex.get_local_id(zero_idx) @@ -125,7 +125,7 @@ def assignment( is_last_feature_window, is_last_centroid_window ) - + first_feature_idx += centroids_window_height dpex.barrier(dpex.CLK_LOCAL_MEM_FENCE) @@ -140,7 +140,7 @@ def assignment( ) first_centroid_idx += window_n_centroids - + dpex.barrier(dpex.CLK_LOCAL_MEM_FENCE) # No update step, only store min_idx in the output array diff --git a/sklearn_numba_dpex/kmeans/kernels/kmeans_plusplus.py b/sklearn_numba_dpex/kmeans/kernels/kmeans_plusplus.py index 498858b..e922611 100644 --- a/sklearn_numba_dpex/kmeans/kernels/kmeans_plusplus.py +++ b/sklearn_numba_dpex/kmeans/kernels/kmeans_plusplus.py @@ -1,13 +1,13 @@ import math from functools import lru_cache -import numpy as np import numba_dpex as dpex - -from ._base_kmeans_kernel_funcs import make_pairwise_ops_base_kernel_funcs +import numpy as np from sklearn_numba_dpex.common.random import make_rand_uniform_kernel_func +from ._base_kmeans_kernel_funcs import make_pairwise_ops_base_kernel_funcs + # NB: refer to the definition of the main lloyd function for a more comprehensive # inline commenting of the kernel. @@ -32,7 +32,7 @@ def kmeansplusplus_init( center_indices, # OUT (n_clusters,) closest_dist_sq, # OUT (n_samples,) ): - # fmt: on + # fmt: on sample_idx = dpex.get_global_id(zero_idx) if sample_idx >= n_samples: return @@ -81,7 +81,7 @@ def sample_center_candidates( random_state, # INOUT (n_local_trials, 2) candidates_id, # OUT (n_local_trials,) ): - # fmt: on + # fmt: on local_trial_idx = dpex.get_global_id(zero_idx) if local_trial_idx >= n_local_trials: return @@ -90,12 +90,15 @@ def sample_center_candidates( cumulative_potential = zero_init candidate_id = one_decr - while (random_value > cumulative_potential) and (candidate_id < max_candidate_id): + while ( + (random_value > cumulative_potential) and + (candidate_id < max_candidate_id) + ): candidate_id += one_incr cumulative_potential += closest_dist_sq[candidate_id] candidates_id[local_trial_idx] = candidate_id - global_size = (math.ceil(n_local_trials / work_group_size)) * (work_group_size) + global_size = (math.ceil(n_local_trials / work_group_size)) * work_group_size return sample_center_candidates[global_size, work_group_size] @@ -151,7 +154,7 @@ def kmeansplusplus_single_step( closest_dist_sq, # IN (n_samples,) sq_distances_t, # OUT (n_candidates, n_samples) ): - # fmt: on + # fmt: on sample_idx = dpex.get_global_id(zero_idx) local_work_id = dpex.get_local_id(zero_idx) @@ -211,8 +214,11 @@ def kmeansplusplus_single_step( if candidate_idx < n_candidates: sq_distance_i = min( sq_distances[i] * sample_weight_, - closest_dist_sq_) - sq_distances_t[first_candidate_idx + i, sample_idx] = sq_distance_i + closest_dist_sq_ + ) + sq_distances_t[first_candidate_idx + i, sample_idx] = ( + sq_distance_i + ) first_candidate_idx += window_n_candidates diff --git a/sklearn_numba_dpex/kmeans/kernels/lloyd_single_step.py b/sklearn_numba_dpex/kmeans/kernels/lloyd_single_step.py index 55dd295..17caf70 100644 --- a/sklearn_numba_dpex/kmeans/kernels/lloyd_single_step.py +++ b/sklearn_numba_dpex/kmeans/kernels/lloyd_single_step.py @@ -1,8 +1,8 @@ import math from functools import lru_cache -import numpy as np import numba_dpex as dpex +import numpy as np from ._base_kmeans_kernel_funcs import ( make_pairwise_ops_base_kernel_funcs, @@ -129,7 +129,7 @@ def fused_lloyd_single_step( new_centroids_t_private_copies, # OUT (n_private_copies, n_features, n_clusters) # noqa cluster_sizes_private_copies, # OUT (n_private_copies, n_clusters) # noqa ): - # fmt: on + # fmt: on """One full iteration of LLoyd's k-means. The kernel is meant to be invoked on a 1D grid spanning the data samples of @@ -252,7 +252,6 @@ def fused_lloyd_single_step( # operations in the current iteration to be over for all work items. dpex.barrier(dpex.CLK_LOCAL_MEM_FENCE) - # End of inner loop. The pseudo inertia is now computed for all centroids # in the window, we can coalesce it to the accumulation of the min pseudo # inertia for the current sample. @@ -267,13 +266,11 @@ def fused_lloyd_single_step( first_centroid_idx += window_n_centroids - # When the next iteration starts work items will overwrite shared memory # with new values, so before that we must wait for all reading # operations in the current iteration to be over for all work items. dpex.barrier(dpex.CLK_LOCAL_MEM_FENCE) - # End of outer loop. By now min_idx and min_sample_pseudo_inertia # contains the expected values. @@ -300,7 +297,7 @@ def fused_lloyd_single_step( # which we mitigate with a strategy of "privatization" for reducing the # probability of collisions. The array of centroids is duplicated in global # memory as many time as possible and each sub-group of work items of size - # `sub_group_size` is assigned to a different duplicata and update the values + # `sub_group_size` is assigned to a different duplicata and update the values # of this single duplicata. # The resulting copies of centroids updates will then need to be reduced to a diff --git a/sklearn_numba_dpex/kmeans/kernels/utils.py b/sklearn_numba_dpex/kmeans/kernels/utils.py index ea33a12..35d4ba9 100644 --- a/sklearn_numba_dpex/kmeans/kernels/utils.py +++ b/sklearn_numba_dpex/kmeans/kernels/utils.py @@ -1,9 +1,8 @@ import math from functools import lru_cache -import numpy as np import numba_dpex as dpex - +import numpy as np zero_idx = np.int64(0) @@ -30,8 +29,8 @@ def relocate_empty_clusters( per_sample_inertia, # INOUT (n_samples,) centroids_t, # INOUT (n_features, n_clusters) cluster_sizes # INOUT (n_clusters,) - ): - # fmt: on + ): + # fmt: on """NB: because of how the array was created, samples_far_from_center values are located between (n_selected_gt_threshold - relocated_idx) and (relocated_idx) indices. @@ -39,22 +38,25 @@ def relocate_empty_clusters( group_idx = dpex.get_group_id(zero_idx) item_idx = dpex.get_local_id(zero_idx) relocated_idx = group_idx // n_work_groups_for_cluster - feature_idx = ((group_idx % n_work_groups_for_cluster) * work_group_size) + item_idx + feature_idx = ( + ((group_idx % n_work_groups_for_cluster) * work_group_size) + item_idx + ) if feature_idx >= n_features: return relocated_cluster_idx = empty_clusters_list[relocated_idx] - new_location_X_idx = samples_far_from_center[n_selected_gt_threshold - relocated_idx] + new_location_X_idx = ( + samples_far_from_center[n_selected_gt_threshold - relocated_idx] + ) new_location_previous_assignment = assignments_idx[new_location_X_idx] new_centroid_value = X_t[feature_idx, new_location_X_idx] new_location_weight = sample_weight[new_location_X_idx] X_centroid_addend = new_centroid_value * new_location_weight - - # Cancel the contribution to new_location_previous_assignment, the previously - # updated centroids of the sample. This contribution will be assigned to the + # Cancel the contribution to new_location_previous_assignment, the previously + # updated centroids of the sample. This contribution will be assigned to the # centroid of the clusters that relocates to this sample. dpex.atomic.sub( centroids_t, @@ -66,7 +68,7 @@ def relocate_empty_clusters( # sample the cluster has been relocated to. centroids_t[feature_idx, relocated_cluster_idx] = X_centroid_addend - # Likewise, we update the weights in the clusters. This is done once using + # Likewise, we update the weights in the clusters. This is done once using # `feature_idx`'s value. if feature_idx == zero_idx: per_sample_inertia[new_location_X_idx] = zero @@ -98,8 +100,8 @@ def select_samples_far_from_centroid( selected_samples_idx, # OUT (n_samples,) n_selected_gt_threshold, # OUT (1,) n_selected_eq_threshold, # OUT (1,) - ): - # fmt: on + ): + # fmt: on """This kernel writes in selected_samples_idx the idx of the top n_selected items in distance_to_centroid with highest values. @@ -122,8 +124,10 @@ def select_samples_far_from_centroid( n_selected_gt_threshold_ = n_selected_gt_threshold[zero_idx] n_selected_eq_threshold_ = n_selected_eq_threshold[zero_idx] - if ((n_selected_gt_threshold_ == max_n_selected_gt_threshold) - and (n_selected_eq_threshold_ >= min_n_selected_eq_threshold)): + if ( + (n_selected_gt_threshold_ == max_n_selected_gt_threshold) and + (n_selected_eq_threshold_ >= min_n_selected_eq_threshold) + ): return distance_to_centroid_ = distance_to_centroid[sample_idx] @@ -166,7 +170,7 @@ def centroid_shifts( new_centroids_t, # IN (n_features, n_clusters) centroid_shifts, # OUT (n_clusters,) ): - # fmt: on + # fmt: on sample_idx = dpex.get_global_id(zero_idx) if sample_idx >= n_clusters: @@ -213,7 +217,7 @@ def reduce_centroid_data( empty_clusters_list, # OUT (n_clusters,) n_empty_clusters, # OUT (1,) ): - # fmt: on + # fmt: on group_idx = dpex.get_group_id(zero_idx) item_idx = dpex.get_local_id(zero_idx)

Code formatting pipeline doesn't seem to work Example of output: https://github.com/soda-inria/sklearn-numba-dpex/actions/runs/3541364642/jobs/5945541586 Configuration needed to point to the folders that should be scanned ?

Indeed, the checked aren't run on the CI. Adding the `--all-files` option force it to runs everywhere: ```bash pre-commit run --all-files ``` I will do a PR.

2022-11-28T10:49:45

0.0

[]

soda-inria/sklearn-numba-dpex

soda-inria__sklearn-numba-dpex-25

3d28986e375cb47d981629cd6fac47e911336275

diff --git a/benchmark/ext_helpers/daal4py.py b/benchmark/ext_helpers/daal4py.py index 7dde6f0..a0b1721 100644 --- a/benchmark/ext_helpers/daal4py.py +++ b/benchmark/ext_helpers/daal4py.py @@ -22,6 +22,9 @@ from daal4py.sklearn.cluster._k_means_0_23 import KMeans +from sklearn.exceptions import NotSupportedByEngineError + + # TODO: instead of relying on monkey patching, find a way to register a distinct entry # point that can load a distinct engine outside of setup.py (impossible ?) @@ -37,6 +40,11 @@ def daal4py_kmeans_single(self, X, sample_weight, centers_init): f"but got `estimator._algorithm = {self.estimator._algorithm}` instead " ) + if sample_weight is not None and any(sample_weight != sample_weight[0]): + raise NotSupportedByEngineError( + "Non unary sample_weight is not supported by daal4py." + ) + est = KMeans( n_clusters=centers_init.shape[0], init=centers_init, diff --git a/benchmark/kmeans.py b/benchmark/kmeans.py index 295a6dd..d729e19 100644 --- a/benchmark/kmeans.py +++ b/benchmark/kmeans.py @@ -1,9 +1,11 @@ from time import perf_counter +from inspect import signature import numpy as np import sklearn from numpy.testing import assert_array_equal from sklearn.cluster import KMeans +from sklearn.exceptions import NotSupportedByEngineError # TODO: maybe this class could be abstracted to use the same strategy with other @@ -35,12 +37,19 @@ def __init__( max_iter, skip_slow=False, run_consistency_checks=True, + **data_initialization_kwargs, ): ( self.X, - self.sample_weight, + self.sample_weight_, self.centers_init, - ) = data_initialization_fn() + ) = data_initialization_fn(**data_initialization_kwargs) + + self.sample_weight = data_initialization_kwargs.get( + "sample_weight", + signature(data_initialization_fn).parameters["sample_weight"].default, + ) + self.max_iter = max_iter self.run_consistency_checks = run_consistency_checks self.results = None @@ -67,7 +76,7 @@ def timeit(self, name, engine_provider=None, is_slow=False): centers_init = self.centers_init.copy() X = self.X.copy() sample_weight = ( - None if self.sample_weight is None else self.sample_weight.copy() + None if self.sample_weight_ is None else self.sample_weight_.copy() ) est_kwargs = dict( @@ -85,14 +94,22 @@ def timeit(self, name, engine_provider=None, is_slow=False): estimator = KMeans(**est_kwargs) - print(f"Running {name} with parameters max_iter={max_iter} tol={tol} ...") + print( + f"Running {name} with parameters sample_weight={self.sample_weight} " + f"n_clusters={n_clusters} data_shape={X.shape} max_iter={max_iter} " + "tol={tol} ..." + ) with sklearn.config_context(engine_provider=engine_provider): # dry run, to be fair for JIT implementations - KMeans(**est_kwargs).set_params(max_iter=1).fit( - X, sample_weight=sample_weight - ) + try: + KMeans(**est_kwargs).set_params(max_iter=1).fit( + X, sample_weight=sample_weight + ) + except NotSupportedByEngineError as e: + print((repr(e) + "\n")) + return t0 = perf_counter() estimator.fit(X, sample_weight=sample_weight) @@ -154,6 +171,7 @@ def _check_same_fit(self, estimator, name, max_iter, assert_allclose): # TODO: expose CLI args. random_state = 123 + sample_weight = "random" # None, "unary", or "random" n_clusters = 127 max_iter = 100 tol = 0 @@ -163,7 +181,11 @@ def _check_same_fit(self, estimator, name, max_iter, assert_allclose): # close results but with significant differences for a few elements. run_consistency_checks = False - def benchmark_data_initialization(random_state=random_state, n_clusters=n_clusters): + def benchmark_data_initialization( + random_state, + n_clusters, + sample_weight, + ): X, _ = fetch_openml(name="spoken-arabic-digit", return_X_y=True, parser="auto") X = X.astype(dtype) scaler_x = MinMaxScaler() @@ -174,10 +196,29 @@ def benchmark_data_initialization(random_state=random_state, n_clusters=n_cluste # X = np.vstack([X for _ in range(20)]) rng = default_rng(random_state) init = np.array(rng.choice(X, n_clusters, replace=False), dtype=np.float32) - return X, None, init + + if sample_weight is None: + pass + elif sample_weight == "unary": + sample_weight = np.ones(len(X), dtype=dtype) + elif sample_weight == "random": + sample_weight = rng.random(size=len(X)).astype(dtype) + else: + raise ValueError( + 'Expected sample_weight in {None, "unary", "random"}, got' + f" {sample_weight} instead." + ) + + return X, sample_weight, init kmeans_timer = KMeansLloydTimeit( - benchmark_data_initialization, max_iter, skip_slow, run_consistency_checks + benchmark_data_initialization, + max_iter, + skip_slow, + run_consistency_checks, + random_state=random_state, + n_clusters=n_clusters, + sample_weight=sample_weight, ) kmeans_timer.timeit( @@ -199,10 +240,12 @@ def benchmark_data_initialization(random_state=random_state, n_clusters=n_cluste kmeans_timer.timeit( name="daal4py lloyd GPU", engine_provider="sklearn_numba_dpex", - is_sloww=True, + is_slow=True, ) - for multiplier in [1, 2, 4, 8]: + work_group_size_multipliers = [1, 2, 4, 8] + + for multiplier in work_group_size_multipliers: with override_attr_context( KMeansEngine, diff --git a/sklearn_numba_dpex/kmeans/drivers.py b/sklearn_numba_dpex/kmeans/drivers.py index b11adb6..227e50d 100644 --- a/sklearn_numba_dpex/kmeans/drivers.py +++ b/sklearn_numba_dpex/kmeans/drivers.py @@ -16,10 +16,8 @@ make_compute_euclidean_distances_fixed_window_kernel, make_centroid_shifts_kernel, make_reduce_centroid_data_kernel, - make_initialize_to_zeros_1d_kernel, make_initialize_to_zeros_2d_kernel, make_initialize_to_zeros_3d_kernel, - make_copyto_2d_kernel, make_broadcast_division_1d_2d_kernel, make_half_l2_norm_2d_axis0_kernel, make_sum_reduction_1d_kernel, @@ -224,6 +222,7 @@ def _lloyd( ( X, + sample_weight, cluster_centers, compute_dtype, output_dtype, @@ -231,7 +230,7 @@ def _lloyd( n_features, n_samples, n_clusters, - ) = self._check_inputs(X, centers_init, sample_weight) + ) = self._check_inputs(X, sample_weight, centers_init) # Create a set of kernels ( @@ -241,6 +240,7 @@ def _lloyd( n_samples, n_features, n_clusters, + return_inertia=bool(verbose), preferred_work_group_size_multiple=self.preferred_work_group_size_multiple, global_mem_cache_size=self.global_mem_cache_size, centroids_window_width_multiplier=self.centroids_window_width_multiplier, @@ -261,10 +261,6 @@ def _lloyd( dtype=compute_dtype, ) - reset_per_sample_inertia_kernel = make_initialize_to_zeros_1d_kernel( - size=n_samples, work_group_size=work_group_size, dtype=compute_dtype - ) - reset_cluster_sizes_private_copies_kernel = make_initialize_to_zeros_2d_kernel( size0=n_centroids_private_copies, size1=n_clusters, @@ -280,10 +276,6 @@ def _lloyd( dtype=compute_dtype, ) - copyto_kernel = make_copyto_2d_kernel( - size0=n_features, size1=n_clusters, work_group_size=work_group_size - ) - broadcast_division_kernel = make_broadcast_division_1d_2d_kernel( size0=n_features, size1=n_clusters, @@ -326,11 +318,15 @@ def _lloyd( dtype=compute_dtype, ) - X_t, centroids_t = self._load_transposed_data_to_device(X, cluster_centers) + X_t, sample_weight, centroids_t = self._load_transposed_data_to_device( + X, + sample_weight, + cluster_centers, + ) # Allocate the necessary memory in the device global memory new_centroids_t = dpctl.tensor.empty_like(centroids_t, device=self.device) - best_centroids_t = dpctl.tensor.empty_like(new_centroids_t, device=self.device) + centroids_half_l2_norm = dpctl.tensor.empty( n_clusters, dtype=compute_dtype, device=self.device ) @@ -340,13 +336,12 @@ def _lloyd( centroid_shifts = dpctl.tensor.empty( n_clusters, dtype=compute_dtype, device=self.device ) - per_sample_inertia = per_sample_pseudo_inertia = dpctl.tensor.empty( + per_sample_inertia = dpctl.tensor.empty( n_samples, dtype=compute_dtype, device=self.device ) assignments_idx = dpctl.tensor.empty( n_samples, dtype=np.uint32, device=self.device ) - new_centroids_t_private_copies = dpctl.tensor.empty( (n_centroids_private_copies, n_features, n_clusters), dtype=compute_dtype, @@ -367,25 +362,12 @@ def _lloyd( # The loop n_iteration = 0 centroid_shifts_sum = np.inf - best_pseudo_inertia = np.inf - - copyto_kernel(centroids_t, best_centroids_t) # TODO: Investigate possible speedup with a custom dpctl queue with a custom # DAG of events and a final single "wait" - # TODO: should we conform to sklearn vanilla kmeans strategy and avoid - # computing an extra centroid update in the last iteration that will not be - # used ? - # It's not a significant difference anyway since when performance matters - # usually we have n_iteration > 100 - # TODO: there's an implicit choice in this code that will zero-out a centroid - # if its cluster is empty. It is not a good choice (a better choice would be to - # replace the centroid with a point in the dataset that is far for its closest - # centroid) - while (n_iteration <= max_iter) and (centroid_shifts_sum > tol): + while (n_iteration < max_iter) and (centroid_shifts_sum > tol): half_l2_norm_kernel(centroids_t, centroids_half_l2_norm) - reset_per_sample_inertia_kernel(per_sample_pseudo_inertia) reset_cluster_sizes_private_copies_kernel(cluster_sizes_private_copies) reset_centroids_private_copies_kernel(new_centroids_t_private_copies) nb_empty_clusters[0] = np.int32(0) @@ -393,9 +375,10 @@ def _lloyd( # TODO: implement special case where only one copy is needed fused_lloyd_fixed_window_single_step_kernel( X_t, + sample_weight, centroids_t, centroids_half_l2_norm, - per_sample_pseudo_inertia, + per_sample_inertia, new_centroids_t_private_copies, cluster_sizes_private_copies, ) @@ -426,17 +409,34 @@ def _lloyd( reduce_centroid_shifts_kernel(centroid_shifts) ) - pseudo_inertia, *_ = dpctl.tensor.asnumpy( - reduce_inertia_kernel(per_sample_pseudo_inertia) - ) - - # TODO: should we drop this check ? - # (it should not be needed because we have theoritical guarantee that - # inertia decreases, if it doesn't it should only be because of rounding - # errors) - if pseudo_inertia < best_pseudo_inertia: - best_pseudo_inertia = pseudo_inertia - copyto_kernel(centroids_t, best_centroids_t) + if verbose: + # ???: verbosity comes at the cost of performance since it triggers + # computing exact inertia at each iteration. Shouldn't this be + # documented ? + inertia, *_ = dpctl.tensor.asnumpy( + reduce_inertia_kernel(per_sample_inertia) + ) + print(f"Iteration {n_iteration}, inertia {inertia}") + + # ???: unlike sklearn, sklearn_intelex checks that pseudo_inertia decreases + # and keep an additional copy of centroids that is updated only if the + # value of the pseudo_inertia is smaller than all the past values. + # + # (the check should not be needed because we have theoritical guarantee + # that inertia decreases, if it doesn't it should only be because of + # rounding errors ?) + # + # To this purpose this code could be inserted here: + # + # if pseudo_inertia < best_pseudo_inertia: + # best_pseudo_inertia = pseudo_inertia + # copyto_kernel(centroids_t, best_centroids_t) + # + # Note that what that is saved as "best centroid" is the array before the + # update, to which the pseudo inertia that is computed at this iteration + # refers. For this reason, this strategy is not compatible with sklearn + # unit tests, that consider new_centroids_t (the array after the update) + # to be the best centroids at each iteration. centroids_t, new_centroids_t = ( new_centroids_t, @@ -445,14 +445,25 @@ def _lloyd( n_iteration += 1 + if verbose: + converged_at = n_iteration - 1 + if centroid_shifts_sum == 0: # NB: possible if tol = 0 + print(f"Converged at iteration {converged_at}: strict convergence.") + + elif centroid_shifts_sum <= tol: + print( + f"Converged at iteration {converged_at}: center shift " + f"{centroid_shifts_sum} within tolerance {tol}." + ) + # Finally, run an assignment kernel to compute the assignments to the best # centroids found, along with the exact inertia. - half_l2_norm_kernel(best_centroids_t, centroids_half_l2_norm) - reset_per_sample_inertia_kernel(per_sample_inertia) + half_l2_norm_kernel(centroids_t, centroids_half_l2_norm) fixed_window_assignment_kernel( X_t, - best_centroids_t, + sample_weight, + centroids_t, centroids_half_l2_norm, per_sample_inertia, assignments_idx, @@ -466,8 +477,8 @@ def _lloyd( return ( assignments_idx, inertia, - best_centroids_t.T, - n_iteration - 1, # substract 1 to conform with vanilla sklearn count + centroids_t.T, + n_iteration, output_dtype, ) @@ -477,13 +488,12 @@ def get_labels( sample_weight, centers, ): - labels = self._get_labels(X, sample_weight, centers) + labels = self._get_labels(X, centers) return dpctl.tensor.asnumpy(labels).astype(np.int32) def _get_labels( self, X, - sample_weight, centers, ): """This call is expected to accept the same inputs than @@ -492,6 +502,7 @@ def _get_labels( """ ( X, + _, centers, compute_dtype, _, @@ -499,7 +510,11 @@ def _get_labels( n_features, n_samples, n_clusters, - ) = self._check_inputs(X, centers, sample_weight) + ) = self._check_inputs( + X, + sample_weight=_IgnoreSampleWeight, + cluster_centers=centers, + ) # Create a set of kernels label_assignment_fixed_window_kernel = make_label_assignment_fixed_window_kernel( @@ -520,7 +535,7 @@ def _get_labels( dtype=compute_dtype, ) - X_t, centroids_t = self._load_transposed_data_to_device(X, centers) + X_t, _, centroids_t = self._load_transposed_data_to_device(X, None, centers) centroids_half_l2_norm = dpctl.tensor.empty( n_clusters, dtype=compute_dtype, device=self.device @@ -552,6 +567,7 @@ def get_inertia( """ ( X, + sample_weight, centers, compute_dtype, output_dtype, @@ -559,7 +575,7 @@ def get_inertia( n_features, n_samples, n_clusters, - ) = self._check_inputs(X, centers, sample_weight) + ) = self._check_inputs(X, sample_weight, centers) # Create a set of kernels compute_inertia_fixed_window_kernel = make_compute_inertia_fixed_window_kernel( @@ -587,7 +603,9 @@ def get_inertia( dtype=compute_dtype, ) - X_t, centroids_t = self._load_transposed_data_to_device(X, centers) + X_t, sample_weight, centroids_t = self._load_transposed_data_to_device( + X, sample_weight, centers + ) centroids_half_l2_norm = dpctl.tensor.empty( n_clusters, dtype=compute_dtype, device=self.device @@ -600,6 +618,7 @@ def get_inertia( compute_inertia_fixed_window_kernel( X_t, + sample_weight, centroids_t, centroids_half_l2_norm, per_sample_inertia, @@ -622,9 +641,9 @@ def get_euclidean_distances(self, X, Y): return dpctl.tensor.asnumpy(euclidean_distances).astype(output_dtype) def _get_euclidean_distances(self, X, Y): - # Input validation ( X, + _, Y, compute_dtype, output_dtype, @@ -632,7 +651,11 @@ def _get_euclidean_distances(self, X, Y): n_features, n_samples, n_clusters, - ) = self._check_inputs(X, Y, sample_weight=None) + ) = self._check_inputs( + X, + sample_weight=_IgnoreSampleWeight, + cluster_centers=Y, + ) label_assignment_fixed_window_kernel = make_compute_euclidean_distances_fixed_window_kernel( n_samples, @@ -652,7 +675,7 @@ def _get_euclidean_distances(self, X, Y): dtype=compute_dtype, ) - X_t, Y_t = self._load_transposed_data_to_device(X, Y) + X_t, _, Y_t = self._load_transposed_data_to_device(X, None, Y) Y_half_l2_norm = dpctl.tensor.empty( n_clusters, dtype=compute_dtype, device=self.device @@ -671,7 +694,7 @@ def _get_euclidean_distances(self, X, Y): ) return euclidean_distances_t.T, output_dtype() - def _set_dtype(self, X, centers_init): + def _set_dtype(self, X, sample_weight, centers_init): input_dtype = output_dtype = np.dtype(X.dtype).type compute_dtype = np.dtype(self.dtype or input_dtype).type copy = True @@ -701,7 +724,7 @@ def _set_dtype(self, X, centers_init): if copy: text += ( f" A copy of the data casted to type {compute_dtype} will be created. " - f"To save memory and prevent this warning, ensure that the dtype of " + f"To save memory and suppress this warning, ensure that the dtype of " f"the input data matches the dtype required for computations." ) warnings.warn(text, DataConversionWarning) @@ -718,26 +741,37 @@ def _set_dtype(self, X, centers_init): f"The centers have been passed with type {centers_init_dtype} but " f"type {compute_dtype} is expected. A copy will be created with the " f"correct type {compute_dtype}. Ensure that the centers are passed " - f"with the correct dtype to save memory and disable this warning.", + f"with the correct dtype to save memory and suppress this warning.", DataConversionWarning, ) centers_init = centers_init.astype(compute_dtype) - return X, centers_init, compute_dtype, output_dtype - - def _check_inputs(self, X, cluster_centers, sample_weight): - if (sample_weight is not None) and ((sample_weight != sample_weight[0]).any()): - raise NotSupportedByEngineError( - "The sklearn_numba_dpex engine for KMeans does not support " - "non-uniform sample weights at this time, but it is a planned " - "feature. More information about the progress toward support is " - "available on the issue tracker at: " - "https://github.com/soda-inria/sklearn-numba-dpex/issues/22 ." + if (sample_weight is not _IgnoreSampleWeight) and ( + sample_weight.dtype != compute_dtype + ): + warnings.warn( + f"sample_weight has been passed with type {sample_weight.dtype} but " + f"type {compute_dtype} is expected. A copy will be created with the " + f"correct type {compute_dtype}. Ensure that sample_weight is passed " + f"with the correct dtype to save memory and suppress this warning.", + DataConversionWarning, ) + sample_weight = sample_weight.astype(compute_dtype) - X, cluster_centers, compute_dtype, output_dtype = self._set_dtype( - X, cluster_centers - ) + return X, sample_weight, centers_init, compute_dtype, output_dtype + + def _check_inputs(self, X, sample_weight, cluster_centers): + + if sample_weight is None: + sample_weight = np.ones(len(X), dtype=(self.dtype or X.dtype)) + + ( + X, + sample_weight, + cluster_centers, + compute_dtype, + output_dtype, + ) = self._set_dtype(X, sample_weight, cluster_centers) work_group_size = ( self.work_group_size_multiplier * self.preferred_work_group_size_multiple @@ -748,6 +782,7 @@ def _check_inputs(self, X, cluster_centers, sample_weight): n_clusters = cluster_centers.shape[0] return ( X, + sample_weight, cluster_centers, compute_dtype, output_dtype, @@ -757,7 +792,7 @@ def _check_inputs(self, X, cluster_centers, sample_weight): n_clusters, ) - def _load_transposed_data_to_device(self, X, cluster_centers): + def _load_transposed_data_to_device(self, X, sample_weight, cluster_centers): # Transfer the input data to device memory, # TODO: let the user pass directly dpctl.tensor or dpnp arrays to avoid copies. if self.X_layout == "C": @@ -777,4 +812,16 @@ def _load_transposed_data_to_device(self, X, cluster_centers): raise ValueError( f"Expected X_layout to be equal to 'C' or 'F', but got {self.X_layout} ." ) - return X_t, dpctl.tensor.from_numpy(cluster_centers.T, device=self.device) + + sample_weight = dpctl.tensor.from_numpy(sample_weight, device=self.device) + cluster_centers = dpctl.tensor.from_numpy(cluster_centers.T, device=self.device) + + return ( + X_t, + sample_weight, + cluster_centers, + ) + + +class _IgnoreSampleWeight: + pass diff --git a/sklearn_numba_dpex/kmeans/kernels/__init__.py b/sklearn_numba_dpex/kmeans/kernels/__init__.py index b049ebc..329ea58 100644 --- a/sklearn_numba_dpex/kmeans/kernels/__init__.py +++ b/sklearn_numba_dpex/kmeans/kernels/__init__.py @@ -8,10 +8,8 @@ from .utils import ( make_centroid_shifts_kernel, make_reduce_centroid_data_kernel, - make_initialize_to_zeros_1d_kernel, make_initialize_to_zeros_2d_kernel, make_initialize_to_zeros_3d_kernel, - make_copyto_2d_kernel, make_broadcast_division_1d_2d_kernel, make_half_l2_norm_2d_axis0_kernel, make_sum_reduction_1d_kernel, @@ -26,10 +24,8 @@ "make_compute_labels_inertia_fixed_window_kernel", "make_centroid_shifts_kernel", "make_reduce_centroid_data_kernel", - "make_initialize_to_zeros_1d_kernel", "make_initialize_to_zeros_2d_kernel", "make_initialize_to_zeros_3d_kernel", - "make_copyto_2d_kernel", "make_broadcast_division_1d_2d_kernel", "make_half_l2_norm_2d_axis0_kernel", "make_sum_reduction_1d_kernel", diff --git a/sklearn_numba_dpex/kmeans/kernels/compute_euclidean_distances.py b/sklearn_numba_dpex/kmeans/kernels/compute_euclidean_distances.py index 31b482c..96cc1ef 100644 --- a/sklearn_numba_dpex/kmeans/kernels/compute_euclidean_distances.py +++ b/sklearn_numba_dpex/kmeans/kernels/compute_euclidean_distances.py @@ -5,7 +5,6 @@ from ._base_kmeans_kernel_funcs import ( _make_initialize_window_kernel_funcs, - _make_update_closest_centroid_kernel_func, _make_accumulate_dot_products_kernel_func, ) diff --git a/sklearn_numba_dpex/kmeans/kernels/compute_inertia.py b/sklearn_numba_dpex/kmeans/kernels/compute_inertia.py index f12b81e..e9db5fc 100644 --- a/sklearn_numba_dpex/kmeans/kernels/compute_inertia.py +++ b/sklearn_numba_dpex/kmeans/kernels/compute_inertia.py @@ -75,6 +75,7 @@ def make_compute_inertia_fixed_window_kernel( # fmt: off def compute_inertia( X_t, # IN READ-ONLY (n_features, n_samples) + sample_weight, # IN READ-ONLY (n_features,) centroids_t, # IN READ-ONLY (n_features, n_clusters) centroids_half_l2_norm, # IN (n_clusters,) per_sample_inertia, # OUT (n_samples,) @@ -159,7 +160,7 @@ def compute_inertia( if sample_idx >= n_samples: return - per_sample_inertia[sample_idx] = X_l2_norm + (two * min_sample_pseudo_inertia) + per_sample_inertia[sample_idx] = X_l2_norm + (two * min_sample_pseudo_inertia) * sample_weight[sample_idx] global_size = (math.ceil(n_samples / work_group_size)) * (work_group_size) return compute_inertia[global_size, work_group_size] diff --git a/sklearn_numba_dpex/kmeans/kernels/compute_labels_inertia.py b/sklearn_numba_dpex/kmeans/kernels/compute_labels_inertia.py index d086da4..03404c1 100644 --- a/sklearn_numba_dpex/kmeans/kernels/compute_labels_inertia.py +++ b/sklearn_numba_dpex/kmeans/kernels/compute_labels_inertia.py @@ -74,6 +74,7 @@ def make_compute_labels_inertia_fixed_window_kernel( # fmt: off def labels_inertia( X_t, # IN READ-ONLY (n_features, n_samples) + sample_weight, # IN READ-ONLY (n_features,) centroids_t, # IN (n_features, n_clusters) centroids_half_l2_norm, # IN (n_clusters,) per_sample_inertia, # OUT (n_samples,) @@ -171,7 +172,7 @@ def labels_inertia( return assignments_idx[sample_idx] = min_idx - per_sample_inertia[sample_idx] = X_l2_norm + (two * min_sample_pseudo_inertia) + per_sample_inertia[sample_idx] = (X_l2_norm + (two * min_sample_pseudo_inertia)) * sample_weight[sample_idx] global_size = (math.ceil(n_samples / work_group_size)) * (work_group_size) return labels_inertia[global_size, work_group_size] diff --git a/sklearn_numba_dpex/kmeans/kernels/lloyd_single_step.py b/sklearn_numba_dpex/kmeans/kernels/lloyd_single_step.py index 1e70ca8..3895531 100644 --- a/sklearn_numba_dpex/kmeans/kernels/lloyd_single_step.py +++ b/sklearn_numba_dpex/kmeans/kernels/lloyd_single_step.py @@ -34,6 +34,7 @@ def make_lloyd_single_step_fixed_window_kernel( n_samples, n_features, n_clusters, + return_inertia, preferred_work_group_size_multiple, global_mem_cache_size, centroids_window_width_multiplier, @@ -71,6 +72,19 @@ def make_lloyd_single_step_fixed_window_kernel( dtype=dtype, ) + if return_inertia: + _accumulate_dot_products_and_X_l2_norm = ( + _make_accumulate_dot_products_kernel_func( + n_samples, + n_features, + centroids_window_height, + window_n_centroids, + with_X_l2_norm=True, + dtype=dtype, + ) + ) + two = dtype(2) + n_windows_per_feature = math.ceil(n_clusters / window_n_centroids) n_windows_per_centroid = math.ceil(n_features / centroids_window_height) @@ -86,7 +100,6 @@ def make_lloyd_single_step_fixed_window_kernel( // n_cluster_bytes ) - one = dtype(1.0) inf = dtype(math.inf) # TODO: currently, constant memory is not supported by numba_dpex, but for read-only @@ -104,9 +117,10 @@ def make_lloyd_single_step_fixed_window_kernel( # fmt: off def fused_lloyd_single_step( X_t, # IN READ-ONLY (n_features, n_samples) + sample_weight, # IN READ-ONLY (n_features,) current_centroids_t, # IN (n_features, n_clusters) centroids_half_l2_norm, # IN (n_clusters,) - per_sample_pseudo_inertia, # OUT (n_samples,) + per_sample_inertia, # OUT (n_samples,) new_centroids_t_private_copies, # OUT (n_private_copies, n_features, n_clusters) # noqa cluster_sizes_private_copies, # OUT (n_private_copies, n_clusters) # noqa ): @@ -213,13 +227,22 @@ def fused_lloyd_single_step( # before going forward dpex.barrier(dpex.CLK_LOCAL_MEM_FENCE) - _accumulate_dot_products( - sample_idx, - first_feature_idx, - X_t, - centroids_window, - dot_products, - ) + if return_inertia and _0 == 0: + X_l2_norm = _accumulate_dot_products_and_X_l2_norm( + sample_idx, + first_feature_idx, + X_t, + centroids_window, + dot_products, + ) + else: + _accumulate_dot_products( + sample_idx, + first_feature_idx, + X_t, + centroids_window, + dot_products, + ) # When the next iteration starts work items will overwrite shared memory # with new values, so before that we must wait for all reading @@ -249,6 +272,20 @@ def fused_lloyd_single_step( # End of outer loop. By now min_idx and min_sample_pseudo_inertia # contains the expected values. + # NB: this check can't be moved at the top at the kernel, because if a work item + # exits early with a `return` it will never reach the barriers, thus causing a + # deadlock. Early returns are only possible when there are no barriers within + # the remaining set of instructions for the running kernel. + if sample_idx >= n_samples: + return + + weight = sample_weight[sample_idx] + + if return_inertia: + inertia = min_sample_pseudo_inertia + inertia = (X_l2_norm + (two * min_sample_pseudo_inertia)) * weight + per_sample_inertia[sample_idx] = inertia + # STEP 2: update centroids. # Each work item updates n_features values in global memory for the centroid @@ -278,15 +315,6 @@ def fused_lloyd_single_step( # The privatization strategy also applies to the updates of the centroid # counts. - # NB: this check can't be moved at the top at the kernel, because if a work item - # exits early with a `return` it will never reach the barriers, thus causing a - # deadlock. Early returns are only possible when there are no barriers within - # the remaining set of instructions for the running kernel. - if sample_idx >= n_samples: - return - - per_sample_pseudo_inertia[sample_idx] = min_sample_pseudo_inertia - # each work item is assigned an array of centroids in a round robin manner privatization_idx = ( sample_idx // preferred_work_group_size_multiple @@ -295,14 +323,14 @@ def fused_lloyd_single_step( dpex.atomic.add( cluster_sizes_private_copies, (privatization_idx, min_idx), - one, + weight ) for feature_idx in range(n_features): dpex.atomic.add( new_centroids_t_private_copies, (privatization_idx, feature_idx, min_idx), - X_t[feature_idx, sample_idx], + X_t[feature_idx, sample_idx] * weight, ) global_size = (math.ceil(n_samples / work_group_size)) * (work_group_size) return ( diff --git a/sklearn_numba_dpex/kmeans/kernels/utils.py b/sklearn_numba_dpex/kmeans/kernels/utils.py index 7a2bbe5..81134d0 100644 --- a/sklearn_numba_dpex/kmeans/kernels/utils.py +++ b/sklearn_numba_dpex/kmeans/kernels/utils.py @@ -121,24 +121,6 @@ def reduce_centroid_data( return reduce_centroid_data[global_size, work_group_size] -@lru_cache -def make_initialize_to_zeros_1d_kernel(size, work_group_size, dtype): - - global_size = math.ceil(size / work_group_size) * work_group_size - zero = dtype(0) - - @dpex.kernel - def initialize_to_zeros(data): - item_idx = dpex.get_global_id(0) - - if item_idx >= size: - return - - data[item_idx] = zero - - return initialize_to_zeros[global_size, work_group_size] - - @lru_cache def make_initialize_to_zeros_2d_kernel(size0, size1, work_group_size, dtype): @@ -186,25 +168,6 @@ def initialize_to_zeros(data): return initialize_to_zeros[global_size, work_group_size] -@lru_cache -def make_copyto_2d_kernel(size0, size1, work_group_size): - global_size = math.ceil(size1 / work_group_size) * work_group_size - - # Optimized for C-contiguous array and for - # size1 >> preferred_work_group_size_multiple - @dpex.kernel - def copyto_kernel(input_array, target_array): - col_idx = dpex.get_global_id(0) - - if col_idx >= size1: - return - - for row_idx in range(size0): - target_array[row_idx, col_idx] = input_array[row_idx, col_idx] - - return copyto_kernel[global_size, work_group_size] - - @lru_cache def make_broadcast_division_1d_2d_kernel(size0, size1, work_group_size): global_size = math.ceil(size1 / work_group_size) * work_group_size

The KMeans engine does not support non-uniform sample_weight Trying to use the KMeans engine in `sklearn_numba_dpex` currently results in a `NotSupportedByEngineError` exception.

2022-09-29T16:15:29

0.0

[]

erikrose/parsimonious

erikrose__parsimonious-211

69d8ab8ddfcaf67a35a3cc0f8738a5b0d47e3689

diff --git a/README.rst b/README.rst index afa55a6..fab1b16 100644 --- a/README.rst +++ b/README.rst @@ -443,6 +443,7 @@ Version History =============== (Next release) + * Improve error message in left-recursive rules. (lucaswiman) * Add support for range ``{min,max}`` repetition expressions (righthandabacus) * Fix bug in ``*`` and ``+`` for token grammars (lucaswiman) * Add support for grammars on bytestrings (lucaswiman) diff --git a/parsimonious/exceptions.py b/parsimonious/exceptions.py index d0f9375..a03a549 100644 --- a/parsimonious/exceptions.py +++ b/parsimonious/exceptions.py @@ -1,3 +1,4 @@ +from textwrap import dedent from parsimonious.utils import StrAndRepr @@ -44,6 +45,20 @@ def column(self): return self.pos + 1 +class LeftRecursionError(ParseError): + def __str__(self): + rule_name = self.expr.name if self.expr.name else str(self.expr) + window = self.text[self.pos:self.pos + 20] + return dedent(f""" + Left recursion in rule {rule_name!r} at {window!r} (line {self.line()}, column {self.column()}). + + Parsimonious is a packrat parser, so it can't handle left recursion. + See https://en.wikipedia.org/wiki/Parsing_expression_grammar#Indirect_left_recursion + for how to rewrite your grammar into a rule that does not use left-recursion. + """ + ).strip() + + class IncompleteParseError(ParseError): """A call to ``parse()`` matched a whole Expression but did not consume the entire text.""" diff --git a/parsimonious/expressions.py b/parsimonious/expressions.py index 9200365..f93f2c6 100644 --- a/parsimonious/expressions.py +++ b/parsimonious/expressions.py @@ -10,7 +10,7 @@ from inspect import getfullargspec, isfunction, ismethod, ismethoddescriptor import regex as re -from parsimonious.exceptions import ParseError, IncompleteParseError +from parsimonious.exceptions import ParseError, IncompleteParseError, LeftRecursionError from parsimonious.nodes import Node, RegexNode from parsimonious.utils import StrAndRepr @@ -96,6 +96,9 @@ def _as_rhs(self): return AdHocExpression(name=rule_name) +IN_PROGRESS = object() + + class Expression(StrAndRepr): """A thing that can be matched against a piece of text""" @@ -186,10 +189,10 @@ def match_core(self, text, pos, cache, error): node = expr_cache[pos] else: # TODO: Set default value to prevent infinite recursion in left-recursive rules. - node = expr_cache[pos] = self._uncached_match(text, - pos, - cache, - error) + expr_cache[pos] = IN_PROGRESS # Mark as in progress + node = expr_cache[pos] = self._uncached_match(text, pos, cache, error) + if node is IN_PROGRESS: + raise LeftRecursionError(text, pos=-1, expr=self) # Record progress for error reporting: if node is None and pos >= error.pos and (

Don't allow recursion if no characters are being used up. I was trying to write left associative operators in grammar, but an issue that kept coming up was infinite recursion. This could be solved by making every operator right associative, because the parser would cut out the option of just going down the same node over and over. Making it so that the parser can only visit one node twice if characters have been used up would completely solve this issue.

After looking through the code I found that there is already a todo in the code. `expressions.py` line 183. This would be really nice to get solved. @joranmulderij could you post the example that's causing the infinite recursion, or a simplified version of it? That would make addressing the bug much easier because it can be turned into a test case. Thanks! Here is a pretty minimal example of what would cause this issue. This is a stripped down version of my grammar. ``` from parsimonious.grammar import Grammar language_grammar = r""" expression = operator_expression / non_operator_expression non_operator_expression = number_expression operator_expression = expression "+" non_operator_expression number_expression = ~"[0-9]+" """ grammar = Grammar(language_grammar) grammar.parse('1 + 2') ``` When this script is run, infinite recursion occurs.

2022-05-07T20:59:50

0.0

[]

erikrose/parsimonious

erikrose__parsimonious-203

aa032364ffac8b0c1893f9c29dfdca7dd70298f6

diff --git a/README.rst b/README.rst index a6c8e07..80a41bf 100644 --- a/README.rst +++ b/README.rst @@ -282,6 +282,15 @@ Syntax Reference ``(things)`` Parentheses are used for grouping, like in every other language. + +``thing{n}`` Exactly ``n`` repetitions of ``thing``. + +``thing{n,m}`` Between ``n`` and ``m`` repititions (inclusive.) + +``thing{,m}`` At most ``m`` repetitions of ``thing``. + +``thing{n,}`` At least ``n`` repetitions of ``thing``. + ==================== ======================================================== .. _flags: https://docs.python.org/3/howto/regex.html#compilation @@ -434,6 +443,8 @@ Version History =============== (Next release) + * Add support for range ``{min,max}`` repetition expressions (righthandabacus) + * Fix bug in ``*`` and ``+`` for token grammars (lucaswiman) * Add support for grammars on bytestrings (lucaswiman) .. warning:: diff --git a/parsimonious/exceptions.py b/parsimonious/exceptions.py index b51c336..29cdef9 100644 --- a/parsimonious/exceptions.py +++ b/parsimonious/exceptions.py @@ -30,14 +30,17 @@ def line(self): match.""" # This is a method rather than a property in case we ever wanted to # pass in which line endings we want to use. - return self.text.count('\n', 0, self.pos) + 1 + if isinstance(self.text, list): # TokenGrammar + return None + else: + return self.text.count('\n', 0, self.pos) + 1 def column(self): """Return the 1-based column where the expression ceased to match.""" # We choose 1-based because that's what Python does with SyntaxErrors. try: return self.pos - self.text.rindex('\n', 0, self.pos) - except ValueError: + except (ValueError, AttributeError): return self.pos + 1 diff --git a/parsimonious/expressions.py b/parsimonious/expressions.py index 0679527..bf10656 100644 --- a/parsimonious/expressions.py +++ b/parsimonious/expressions.py @@ -375,105 +375,71 @@ class Lookahead(Compound): """An expression which consumes nothing, even if its contained expression succeeds""" - # TODO: Merge this and Not for better cache hit ratios and less code. - # Downside: pretty-printed grammars might be spelled differently than what - # went in. That doesn't bother me. + __slots__ = ['negativity'] - def _uncached_match(self, text, pos, cache, error): - node = self.members[0].match_core(text, pos, cache, error) - if node is not None: - return Node(self, text, pos, pos) - - def _as_rhs(self): - return u'&%s' % self._unicode_members()[0] + def __init__(self, member, *, negative=False, **kwargs): + super(Lookahead, self).__init__(member, **kwargs) + self.negativity = bool(negative) - -class Not(Compound): - """An expression that succeeds only if the expression within it doesn't - - In any case, it never consumes any characters; it's a negative lookahead. - - """ def _uncached_match(self, text, pos, cache, error): - # FWIW, the implementation in Parsing Techniques in Figure 15.29 does - # not bother to cache NOTs directly. node = self.members[0].match_core(text, pos, cache, error) - if node is None: + if (node is None) == self.negativity: # negative lookahead == match only if not found return Node(self, text, pos, pos) def _as_rhs(self): - # TODO: Make sure this parenthesizes the member properly if it's an OR - # or AND. - return u'!%s' % self._unicode_members()[0] + return u'%s%s' % ('!' if self.negativity else '&', self._unicode_members()[0]) +def Not(term): + return Lookahead(term, negative=True) # Quantifiers. None of these is strictly necessary, but they're darn handy. -class Optional(Compound): - """An expression that succeeds whether or not the contained one does - - If the contained expression succeeds, it goes ahead and consumes what it - consumes. Otherwise, it consumes nothing. - - """ - def _uncached_match(self, text, pos, cache, error): - node = self.members[0].match_core(text, pos, cache, error) - return (Node(self, text, pos, pos) if node is None else - Node(self, text, pos, node.end, children=[node])) - - def _as_rhs(self): - return u'%s?' % self._unicode_members()[0] +class Quantifier(Compound): + """An expression wrapper like the */+/?/{n,m} quantifier in regexes.""" + __slots__ = ['min', 'max'] -# TODO: Merge with OneOrMore. -class ZeroOrMore(Compound): - """An expression wrapper like the * quantifier in regexes.""" - - def _uncached_match(self, text, pos, cache, error): - new_pos = pos - children = [] - while True: - node = self.members[0].match_core(text, new_pos, cache, error) - if node is None or not (node.end - node.start): - # Node was None or 0 length. 0 would otherwise loop infinitely. - return Node(self, text, pos, new_pos, children) - children.append(node) - new_pos += node.end - node.start - - def _as_rhs(self): - return u'%s*' % self._unicode_members()[0] - - -class OneOrMore(Compound): - """An expression wrapper like the + quantifier in regexes. - - You can also pass in an alternate minimum to make this behave like "2 or - more", "3 or more", etc. - - """ - __slots__ = ['min'] - - # TODO: Add max. It should probably succeed if there are more than the max - # --just not consume them. - - def __init__(self, member, name='', min=1): - super(OneOrMore, self).__init__(member, name=name) + def __init__(self, member, *, min=0, max=float('inf'), name='', **kwargs): + super(Quantifier, self).__init__(member, name=name, **kwargs) self.min = min + self.max = max def _uncached_match(self, text, pos, cache, error): new_pos = pos children = [] - while True: + size = len(text) + while new_pos < size and len(children) < self.max: node = self.members[0].match_core(text, new_pos, cache, error) if node is None: - break + break # no more matches children.append(node) length = node.end - node.start - if length == 0: # Don't loop infinitely. + if len(children) >= self.min and length == 0: # Don't loop infinitely break new_pos += length if len(children) >= self.min: return Node(self, text, pos, new_pos, children) def _as_rhs(self): - return u'%s+' % self._unicode_members()[0] + if self.min == 0 and self.max == 1: + qualifier = '?' + elif self.min == 0 and self.max == float('inf'): + qualifier = '*' + elif self.min == 1 and self.max == float('inf'): + qualifier = '+' + elif self.max == float('inf'): + qualifier = '{%d,}' % self.min + elif self.min == 0: + qualifier = '{,%d}' % self.max + else: + qualifier = '{%d,%d}' % (self.min, self.max) + return '%s%s' % (self._unicode_members()[0], qualifier) + +def ZeroOrMore(member, name=''): + return Quantifier(member, name=name, min=0, max=float('inf')) + +def OneOrMore(member, name='', min=1): + return Quantifier(member, name=name, min=min, max=float('inf')) + +def Optional(member, name=''): + return Quantifier(member, name=name, min=0, max=1) diff --git a/parsimonious/grammar.py b/parsimonious/grammar.py index 617f4d1..e38f1d0 100644 --- a/parsimonious/grammar.py +++ b/parsimonious/grammar.py @@ -10,7 +10,7 @@ from parsimonious.exceptions import BadGrammar, UndefinedLabel from parsimonious.expressions import (Literal, Regex, Sequence, OneOf, - Lookahead, Optional, ZeroOrMore, OneOrMore, Not, TokenMatcher, + Lookahead, Quantifier, Optional, ZeroOrMore, OneOrMore, Not, TokenMatcher, expression, is_callable) from parsimonious.nodes import NodeVisitor from parsimonious.utils import evaluate_string @@ -241,7 +241,7 @@ def _expressions_from_rules(self, rule_syntax, custom_rules): atom = reference / literal / regex / parenthesized regex = "~" spaceless_literal ~"[ilmsuxa]*"i _ parenthesized = "(" _ expression ")" _ - quantifier = ~"[*+?]" _ + quantifier = ~r"[*+?]|\{\d*,\d+\}|\{\d+,\d*\}|\{\d+\}" _ reference = label !equals # A subsequent equal sign is the only thing that distinguishes a label @@ -305,7 +305,17 @@ def visit_quantifier(self, node, quantifier): def visit_quantified(self, node, quantified): atom, quantifier = quantified - return self.quantifier_classes[quantifier.text](atom) + try: + return self.quantifier_classes[quantifier.text](atom) + except KeyError: + # This should pass: assert re.full_match("\{(\d*)(,(\d*))?\}", quantifier) + quantifier = quantifier.text[1:-1].split(",") + if len(quantifier) == 1: + min_match = max_match = int(quantifier[0]) + else: + min_match = int(quantifier[0]) if quantifier[0] else 0 + max_match = int(quantifier[1]) if quantifier[1] else float('inf') + return Quantifier(atom, min=min_match, max=max_match) def visit_lookahead_term(self, node, lookahead_term): ampersand, term, _ = lookahead_term

Range and some-of operators ### Range operators ~~Stolen~~ Adapted from POSIX regexes: ``` foo = bar{2} # expands into foo = bar bar foo = bar{2,4} # expands into foo = bar bar bar? bar? # this one isn't part of POSIX, but it's a fairly obvious extension foo = bar{,3} # expands into foo = bar? bar? bar? ``` Further, suffixing the upper bound with `=` prevents more instances of `bar` from existing: ``` foo = bar{,3=} # expands into foo = bar? bar? bar? !bar ``` This might be worth making the default? I'm not sure. It shouldn't be usable without the prefixed comma, since that sends conflicting messages about whether the number is a lower or upper bound. ### Some-of operators I have no idea what this syntax should look like. This is just what first came to mind. ``` foo = ~(bar baz bluh){2} # expands into foo = (bar baz) / (bar bluh) / (baz bluh) foo = ~(bar baz bluh){2,3} # expands into foo = (bar baz) / (bar bluh) / (baz bluh) / (bar baz bluh) ``` There isn't much of a point to a lower bound of 0, as that would be equivalent to chaining `?`, but it might as well be allowed anyway for simplicity. Note that reordering the parts isn't permitted here; it would be preferable to allow that, but the resulting implicit alternatives would be confusing: ``` foo = ~(bar baz bluh){2} # could expand into foo = (bar baz) / (bar bluh) / (baz bar) / (baz bluh) / (bluh bar) / (bluh baz) # but what does that parse things as? we just don't know # (it's still well-defined, of course, just weird) ``` If we introduce syntax to do this, then a lower bound of 0 does become meaningful. And, again, `=` is possible: ``` foo = ~(bar baz bluh){,2=} # expands into foo = ((bar baz) / (bar bluh) / (baz bluh)) (!bar !baz !bluh) ``

[Lojban's machine grammar]( http://lojban.org/publications/cll/cll_v1.1_xhtml-chapter-chunks/chapter-grammars.htm) uses `&` for my `~(...){}`. This doesn't allow for the `{}` minimum-maximum addition, but it's a lead to follow, if nothing else.

2022-04-11T06:25:42

0.0

[]

atsb/nose-py3

atsb__nose-py3-5

069782a1bbec38236d5ee376e3c76ed364754269

diff --git a/nose/case.py b/nose/case.py index abef702..35f61ed 100644 --- a/nose/case.py +++ b/nose/case.py @@ -333,7 +333,7 @@ def __init__(self, method, test=None, arg=tuple(), descriptor=None): if isfunction(method): raise ValueError( "Unbound methods must be wrapped using pyversion.unbound_method before passing to MethodTestCase") - self.cls = method.im_class + self.cls = method.__self__.__class__ self.inst = self.cls() if self.test is None: method_name = self.method.__name__ diff --git a/nose/config.py b/nose/config.py index bcff992..fb15e9e 100644 --- a/nose/config.py +++ b/nose/config.py @@ -636,8 +636,8 @@ def __nonzero__(self): def user_config_files(): """Return path to any existing user config files """ - return filter(os.path.exists, - map(os.path.expanduser, config_files)) + return list(filter(os.path.exists, + map(os.path.expanduser, config_files))) def all_config_files(): diff --git a/nose/core.py b/nose/core.py index 0989236..e03d49c 100644 --- a/nose/core.py +++ b/nose/core.py @@ -224,32 +224,26 @@ def add_option(self, *arg, **kw): v = self.config.verbosity self.config.plugins.sort() for p in self.config.plugins: - print - "Plugin %s" % p.name + print("Plugin %s" % p.name) if v >= 2: - print - " score: %s" % p.score - print - '\n'.join(textwrap.wrap(p.help().strip(), - initial_indent=' ', - subsequent_indent=' ')) + print(" score: %s" % p.score) + print('\n'.join(textwrap.wrap(p.help().strip(), + initial_indent=' ', + subsequent_indent=' '))) if v >= 3: parser = DummyParser() p.addOptions(parser) if len(parser.options): - print - print - " Options:" + print() + print(" Options:") for opts, help in parser.options: - print - ' %s' % (', '.join(opts)) + print(' %s' % (', '.join(opts))) if help: - print - '\n'.join( + print('\n'.join( textwrap.wrap(help.strip(), initial_indent=' ', - subsequent_indent=' ')) - print + subsequent_indent=' '))) + print() def usage(cls): import nose diff --git a/nose/ext/dtcompat.py b/nose/ext/dtcompat.py index 8bebb61..5b73c7f 100644 --- a/nose/ext/dtcompat.py +++ b/nose/ext/dtcompat.py @@ -9,7 +9,7 @@ # # Modified for inclusion in nose to provide support for DocFileTest # -# - all doctests removed from module (they fail under 2.3 and 2.5) +# - all doctests removed from module (they fail under 2.3 and 2.5) # - now handles the $py.class extension when ran under Jython r"""Module doctest -- a framework for running examples in docstrings. @@ -905,8 +905,7 @@ def _find(self, tests, obj, name, module, source_lines, globs, seen): add them to `tests`. """ if self._verbose: - print - 'Finding tests in %s' % name + print('Finding tests in %s' % name) # If we've already processed this object, then ignore it. if id(obj) in seen: diff --git a/nose/failure.py b/nose/failure.py index ecef903..d955d9c 100644 --- a/nose/failure.py +++ b/nose/failure.py @@ -15,7 +15,7 @@ class Failure(unittest.TestCase): A Failure case is placed in a test suite to indicate the presence of a test that could not be loaded or executed. A common example is a test module that fails to import. - + """ __test__ = False # do not collect @@ -37,7 +37,7 @@ def address(self): def runTest(self): if self.tb is not None: if is_base_exception(self.exc_val): - raise self.exc_val(None).with_traceback(self.tb) + raise self.exc_class(None).with_traceback(self.tb) raise self.exc_class(self.exc_val).with_traceback(self.tb) else: raise self.exc_class(self.exc_val) diff --git a/nose/loader.py b/nose/loader.py index 605e76e..f57fb9b 100644 --- a/nose/loader.py +++ b/nose/loader.py @@ -112,7 +112,7 @@ def wanted(attr, cls=testCaseClass, sel=self.selector): return False return sel.wantMethod(item) - cases = filter(wanted, dir(testCaseClass)) + cases = list(filter(wanted, dir(testCaseClass))) # add runTest if nothing else picked if not cases and hasattr(testCaseClass, 'runTest'): @@ -225,7 +225,7 @@ def loadTestsFromFile(self, filename): # Plugins can yield False to indicate that they were # unable to load tests from a file, but it was not an # error -- the file just had no tests to load. - tests = filter(None, tests) + tests = list(filter(None, tests)) return self.suiteClass(tests) else: # Nothing was able to even try to load from this file @@ -276,8 +276,8 @@ def loadTestsFromGeneratorMethod(self, generator, cls): """ # convert the unbound generator method # into a bound method so it can be called below - if hasattr(generator, 'im_class'): - cls = generator.im_class + if hasattr(generator, '__self__'): + cls = generator.__self__.__class__ inst = cls() method = generator.__name__ generator = getattr(inst, method) @@ -330,10 +330,12 @@ def loadTestsFromModule(self, module, path=None, discovered=False): test_classes.append(test) elif isfunction(test) and self.selector.wantFunction(test): test_funcs.append(test) + test_classes = sort_list(test_classes, lambda x: x.__name__) + tests.extend(self.makeTest(t, parent=module) for t in test_classes) + test_funcs = sort_list(test_funcs, func_lineno) - tests = map(lambda t: self.makeTest(t, parent=module), - test_classes + test_funcs) + tests.extend(self.makeTest(t, parent=module) for t in test_funcs) # Now, descend into packages # FIXME can or should this be lazy? diff --git a/nose/plugins/attrib.py b/nose/plugins/attrib.py index 217f1e5..5ab0c63 100644 --- a/nose/plugins/attrib.py +++ b/nose/plugins/attrib.py @@ -125,7 +125,7 @@ def wrap_ob(ob): def get_method_attr(method, cls, attr_name, default=False): - """Look up an attribute on a method/ function. + """Look up an attribute on a method/ function. If the attribute isn't found there, looking it up in the method's class, if any. """ @@ -140,7 +140,7 @@ def get_method_attr(method, cls, attr_name, default=False): class ContextHelper: """Object that can act as context dictionary for eval and looks up - names as attributes on a method/ function and its class. + names as attributes on a method/ function and its class. """ def __init__(self, method, cls): @@ -283,7 +283,7 @@ def wantMethod(self, method): """Accept the method if its attributes match. """ try: - cls = method.im_class + cls = method.__self__.__class__ except AttributeError: return False return self.validateAttrib(method, cls) diff --git a/nose/plugins/xunit.py b/nose/plugins/xunit.py index 7500620..1d51e2f 100644 --- a/nose/plugins/xunit.py +++ b/nose/plugins/xunit.py @@ -55,7 +55,7 @@ from io import StringIO -from numpy import unicode +from numpy.core import unicode from nose.plugins.base import Plugin from nose.pyversion import force_unicode, format_exception diff --git a/nose/pyversion.py b/nose/pyversion.py index a241d73..c7de0ae 100644 --- a/nose/pyversion.py +++ b/nose/pyversion.py @@ -6,12 +6,12 @@ import sys import traceback -from numpy import unicode +from numpy.core import unicode import nose.util -__all__ = ['make_instancemethod', 'cmp_to_key', 'sort_list', 'ClassType', - 'TypeType', 'UNICODE_STRINGS', 'unbound_method', 'ismethod', +__all__ = ['make_instancemethod', 'cmp_to_key', 'sort_list', + 'UNICODE_STRINGS', 'unbound_method', 'ismethod', 'bytes_', 'is_base_exception', 'force_unicode', 'exc_to_unicode', 'format_exception', 'isgenerator'] @@ -61,13 +61,6 @@ def sort_list(l, key, reverse=False): return sorted_list -# In Python 3.x, all objects are "new style" objects descended from 'type', and -# thus types.ClassType and types.TypeType don't exist anymore. For -# compatibility, we make sure they still work. -ClassType = type -TypeType = type - - # The following emulates the behavior (we need) of an 'unbound method' under # Python 3.x (namely, the ability to have a class associated with a function # definition so that things can do stuff based on its associated class) @@ -78,8 +71,6 @@ def __init__(self, cls, func): self.__dict__ = func.__dict__.copy() self._func = func self.__self__ = UnboundSelf(cls) - if sys.version_info < (3, 0): - self.im_class = cls self.__doc__ = getattr(func, '__doc__', None) def address(self): diff --git a/nose/suite.py b/nose/suite.py index 9ee91d5..5dfd961 100644 --- a/nose/suite.py +++ b/nose/suite.py @@ -436,9 +436,7 @@ def ancestry(self, context): # Methods include reference to module they are defined in, we # don't want that, instead want the module the class is in now # (classes are re-ancestored elsewhere). - if hasattr(context, 'im_class'): - context = context.im_class - elif hasattr(context, '__self__'): + if hasattr(context, '__self__'): context = context.__self__.__class__ if hasattr(context, '__module__'): ancestors = context.__module__.split('.') diff --git a/nose/util.py b/nose/util.py index 0eb3d2d..b5f325f 100644 --- a/nose/util.py +++ b/nose/util.py @@ -10,14 +10,13 @@ import types import unittest -from numpy import unicode +from numpy.core import unicode -from nose.pyversion import ClassType, TypeType, isgenerator +from nose.pyversion import isgenerator log = logging.getLogger('nose') ident_re = re.compile(r'^[A-Za-z_][A-Za-z0-9_.]*$') -class_types = (ClassType, TypeType) skip_pattern = r"(?:\.svn)|(?:[^.]+\.py[co])|(?:.*~)|(?:.*\$py\.class)|(?:__pycache__)" set = set @@ -157,7 +156,7 @@ def isclass(obj): for objects that can't be subclasses of anything. """ obj_type = type(obj) - return obj_type in class_types or issubclass(obj_type, type) + return obj_type is type or issubclass(obj_type, type) # backwards compat (issue #64) @@ -398,7 +397,7 @@ def test_address(test): file = getattr(test, '__file__', None) module = getattr(test, '__name__', None) return src(file), module, call - if t == types.FunctionType or issubclass(t, type) or t == types.ClassType: + if t == types.FunctionType or t is type or issubclass(t, type): module = getattr(test, '__module__', None) if module is not None: m = sys.modules[module] @@ -408,7 +407,7 @@ def test_address(test): call = getattr(test, '__name__', None) return src(file), module, call if t == types.MethodType: - cls_adr = test_address(test.im_class) + cls_adr = test_address(test.__self__.__class__) return (src(cls_adr[0]), cls_adr[1], "%s.%s" % (cls_adr[2], test.__name__)) # handle unittest.TestCase instances @@ -449,15 +448,16 @@ def try_run(obj, names): if type(obj) == types.ModuleType: # py.test compatibility if isinstance(func, types.FunctionType): - args, varargs, varkw, defaults = \ - inspect.getfullargspec(func) + spec = inspect.getfullargspec(func) + args, varargs, varkw, defaults = spec[:4] + else: # Not a function. If it's callable, call it anyway if hasattr(func, '__call__') and not inspect.ismethod(func): func = func.__call__ try: - args, varargs, varkw, defaults = \ - inspect.getfullargspec(func) + spec = inspect.getfullargspec(func) + args, varargs, varkw, defaults = spec[:4] args.pop(0) # pop the self off except TypeError: raise TypeError("Attribute %s of %r is not a python "

AttributeError since upgrade from 1.4.0 -> 1.4.3 Hi, I get the exception below when upgrading nose-py3 from 1.4.0 to 1.4.3: ``` (daproject) [dauser@desk004 django]$ python --version Python 3.10.3 (daproject) [dauser@desk004 django]$ pip freeze | grep -i nose django-nose==1.4.7 nose-py3==1.4.3 (daproject) [dauser@desk004 django]$ nosetests positions --verbose --nocapture --verbosity=2 Traceback (most recent call last): File "/home/dauser/projects/daproject/venv/daproject/bin/nosetests", line 8, in <module> sys.exit(run_exit()) File "/home/dauser/projects/daproject/venv/daproject/lib/python3.10/site-packages/nose/core.py", line 117, in __init__ unittest.TestProgram.__init__( File "/home/dauser/installed/fedora_f35/python3.10.3/usr/local/lib/python3.10/unittest/main.py", line 100, in __init__ self.parseArgs(argv) File "/home/dauser/projects/daproject/venv/daproject/lib/python3.10/site-packages/nose/core.py", line 178, in parseArgs self.createTests() File "/home/dauser/projects/daproject/venv/daproject/lib/python3.10/site-packages/nose/core.py", line 192, in createTests self.test = self.testLoader.loadTestsFromNames(self.testNames) File "/home/dauser/projects/daproject/venv/daproject/lib/python3.10/site-packages/nose/loader.py", line 484, in loadTestsFromNames return unittest.TestLoader.loadTestsFromNames(self, names, module) File "/home/dauser/installed/fedora_f35/python3.10.3/usr/local/lib/python3.10/unittest/loader.py", line 220, in loadTestsFromNames suites = [self.loadTestsFromName(name, module) for name in names] File "/home/dauser/installed/fedora_f35/python3.10.3/usr/local/lib/python3.10/unittest/loader.py", line 220, in <listcomp> suites = [self.loadTestsFromName(name, module) for name in names] File "/home/dauser/projects/daproject/venv/daproject/lib/python3.10/site-packages/nose/loader.py", line 434, in loadTestsFromName return self.loadTestsFromModule( File "/home/dauser/projects/daproject/venv/daproject/lib/python3.10/site-packages/nose/loader.py", line 357, in loadTestsFromModule tests.extend(self.loadTestsFromDir(module_path)) AttributeError: 'map' object has no attribute 'extend' ```

Hmmm this is really strange. I gave 1.4.3 a good bashing on my system before I uploaded the package and had no issues with it not running tests. I am assuming that these tests worked previously on 1.4.0? > I am assuming that these tests worked previously on 1.4.0? Yes, rolling back to 1.4.0 removes this exception for me. To be precise: I usually run this using django's testing facility by running `python manage.py test positions --keepdb --verbose --nocapture` The first line of output of that statement is `nosetests positions --verbose --nocapture --verbosity=2` Using 1.4.0 it then proceeds as expected, whereas using 1.4.3 leads to the above exception with a slightly longer stacktrace: ``` (daproject) [dauser@desk004 django]$ python manage.py test positions --keepdb --verbose --nocapture nosetests positions --verbose --nocapture --verbosity=2 Traceback (most recent call last): File "/home/dauser/projects/daproject/src/django/manage.py", line 10, in <module> execute_from_command_line(sys.argv) File "/home/dauser/projects/daproject/venv/daproject/lib/python3.10/site-packages/django/core/management/__init__.py", line 446, in execute_from_command_line utility.execute() File "/home/dauser/projects/daproject/venv/daproject/lib/python3.10/site-packages/django/core/management/__init__.py", line 440, in execute self.fetch_command(subcommand).run_from_argv(self.argv) File "/home/dauser/projects/daproject/venv/daproject/lib/python3.10/site-packages/django/core/management/commands/test.py", line 24, in run_from_argv super().run_from_argv(argv) File "/home/dauser/projects/daproject/venv/daproject/lib/python3.10/site-packages/django/core/management/base.py", line 414, in run_from_argv self.execute(*args, **cmd_options) File "/home/dauser/projects/daproject/venv/daproject/lib/python3.10/site-packages/django/core/management/base.py", line 460, in execute output = self.handle(*args, **options) File "/home/dauser/projects/daproject/venv/daproject/lib/python3.10/site-packages/django/core/management/commands/test.py", line 68, in handle failures = test_runner.run_tests(test_labels) File "/home/dauser/projects/daproject/venv/daproject/lib/python3.10/site-packages/django_nose/runner.py", line 308, in run_tests result = self.run_suite(nose_argv) File "/home/dauser/projects/daproject/venv/daproject/lib/python3.10/site-packages/django_nose/runner.py", line 244, in run_suite nose.core.TestProgram(argv=nose_argv, exit=False, File "/home/dauser/projects/daproject/venv/daproject/lib/python3.10/site-packages/nose/core.py", line 117, in __init__ unittest.TestProgram.__init__( File "/home/dauser/installed/fedora_f35/python3.10.3/usr/local/lib/python3.10/unittest/main.py", line 100, in __init__ self.parseArgs(argv) File "/home/dauser/projects/daproject/venv/daproject/lib/python3.10/site-packages/nose/core.py", line 178, in parseArgs self.createTests() File "/home/dauser/projects/daproject/venv/daproject/lib/python3.10/site-packages/nose/core.py", line 192, in createTests self.test = self.testLoader.loadTestsFromNames(self.testNames) File "/home/dauser/projects/daproject/venv/daproject/lib/python3.10/site-packages/nose/loader.py", line 484, in loadTestsFromNames return unittest.TestLoader.loadTestsFromNames(self, names, module) File "/home/dauser/installed/fedora_f35/python3.10.3/usr/local/lib/python3.10/unittest/loader.py", line 220, in loadTestsFromNames suites = [self.loadTestsFromName(name, module) for name in names] File "/home/dauser/installed/fedora_f35/python3.10.3/usr/local/lib/python3.10/unittest/loader.py", line 220, in <listcomp> suites = [self.loadTestsFromName(name, module) for name in names] File "/home/dauser/projects/daproject/venv/daproject/lib/python3.10/site-packages/nose/loader.py", line 434, in loadTestsFromName return self.loadTestsFromModule( File "/home/dauser/projects/daproject/venv/daproject/lib/python3.10/site-packages/nose/loader.py", line 357, in loadTestsFromModule tests.extend(self.loadTestsFromDir(module_path)) AttributeError: 'map' object has no attribute 'extend' ``` If just for fun I then execute `nosetests positions --verbose --nocapture --verbosity=2` without using `manage.py` that comes with django, I get the result in the original description above. Using 1.4.0 without django leads to some other errors that seem to be django related: | | 1.4.0 | 1.4.3| | --------------- | --------------- | --------------- | |django | OK | error1 above | |nosetest only | error2 | error1 above | The error1 prototype seems to be the following: ``` $ python Python 3.10.3 (main, Mar 22 2022, 13:00:21) [GCC 11.2.1 20220127 (Red Hat 11.2.1-9)] on linux Type "help", "copyright", "credits" or "license" for more information. >>> a = map(lambda x: x + 1, [2,3]) >>> a.extend Traceback (most recent call last): File "<stdin>", line 1, in <module> AttributeError: 'map' object has no attribute 'extend' ``` Which presumably could arise from line 335 in loader.py, where it says ``` tests = map(lambda t: self.makeTest(t, parent=module), test_classes + test_funcs) ``` followed by `tests.extend(self.loadTestsFromDir(module_path))` at line 357, which maybe expects `tests` to be a `list` rather than a `map` Replacing line 335 with ``` tests = list(map(lambda t: self.makeTest(t, parent=module), test_classes + test_funcs)) ``` seems to remove this exception, but then I get another one: ``` ====================================================================== ERROR: Failure: AttributeError ('function' object has no attribute 'im_class') ---------------------------------------------------------------------- Traceback (most recent call last): File "/home/dauser/projects/daproject/venv/daproject/lib/python3.10/site-packages/nose/failure.py", line 40, in runTest raise self.exc_val(None).with_traceback(self.tb) TypeError: --------------------- >> end captured logging << --------------------- ``` I'll stay with 1.4.0 for the time being ;-) I just stumbled upon the same *AttributeError* with the latest 1.4.3 release. Pinning the version to 1.4.0 seems to work fine. As I only need this package to migrate some old legacy Python 2 code written with *nose* tests to Python 3 which will not use *nose* any more, this is no big deal here and therefore I did no further research on this topic. I've looked into this quite a bit but have not found the reason. It works fine for my projects on three separate installations, however this could be because I wipe the previous version and install fresh. There were quite a few changes from 1.4.0 to 1.4.3 and upgrading from that could indeed cause issues. In my case, I explicitly installed the *nose-py3* package in a clean CI environment here on GitHub Actions, so in theory there should not be such a conflict. Adding on the comment from @odmi-frijt, the offending *map* call is limited to `not discovered or self.selector.wantModule(module)`, so you might run into a different path inside your projects and never call *map* there. If you want to further debug this: My test repository is a fork of https://github.com/jwilk-archive/ocrodjvu, which I tested with exactly the code of the base repository. This code is still Python2-only, but this should not be the problem, as I got the aforementioned error when simply doing `python -m pip install nose-py3 && make test` while on Python 3. Thanks! I'll get on that

2022-12-23T13:22:08

0.0

[]

scrapinghub/dateparser

scrapinghub__dateparser-1224

f65936408e24ede5a4e3c05f624323bfc474c585

diff --git a/dateparser/parser.py b/dateparser/parser.py index 16751188b..40aa8ff7b 100644 --- a/dateparser/parser.py +++ b/dateparser/parser.py @@ -598,10 +598,13 @@ def _correct_for_month(self, dateobj): relative_base_month = ( relative_base.month if hasattr(relative_base, "month") else relative_base ) - if getattr(self, "_token_month", None) or relative_base_month: + + if getattr(self, "_token_month", None): return dateobj - dateobj = set_correct_month_from_settings(dateobj, self.settings) + dateobj = set_correct_month_from_settings( + dateobj, self.settings, relative_base_month + ) return dateobj @classmethod @@ -613,11 +616,13 @@ def parse(cls, datestring, settings, tz=None): # correction for past, future if applicable dateobj = po._correct_for_time_frame(dateobj, tz) + # correction for preference of month: beginning, current, end + # must happen before day so that day is derived from the correct month + dateobj = po._correct_for_month(dateobj) + # correction for preference of day: beginning, current, end dateobj = po._correct_for_day(dateobj) - # correction for preference of month: beginning, current, end - dateobj = po._correct_for_month(dateobj) period = po._get_period() return dateobj, period

Parsing using `{"PREFER_DAY_OF_MONTH": "last", "PREFER_MONTH_OF_YEAR": "last"}` doesn't use last day of the last month Hi, I noticed that when parsing a date with missing month and day of month with `{"PREFER_DAY_OF_MONTH": "last", "PREFER_MONTH_OF_YEAR": "last"}` settings, we default to the last day of the current month instead of the last day of the last month. For example, when running this code in February, it will return `2022/12/28` instead of `2022/12/31`, as the last day of feb is 28. ```python >>> import dateparser >>> dateparser.parse("2022", settings={"PREFER_DAY_OF_MONTH": "last", "PREFER_MONTH_OF_YEAR": "last"}) datetime.datetime(2022, 12, 28, 0, 0) ```

+1; this bit me because it correctly returned the 31st when I wrote my tests in March, then started failing in April :( I would guess it should resolve the month before the day

2024-04-02T15:56:26

0.0

[]

scrapinghub/dateparser

scrapinghub__dateparser-1219

1d4b05875d2ee35987607f8e33efa4c8df74dfb2

diff --git a/dateparser/data/date_translation_data/en.py b/dateparser/data/date_translation_data/en.py index b70975e1b..1409dcadb 100644 --- a/dateparser/data/date_translation_data/en.py +++ b/dateparser/data/date_translation_data/en.py @@ -800,7 +800,7 @@ "(?:12\\s+)?midnight": "00:00" }, { - "(\\d+[.,]?\\d*)h(\\d+[.,]?\\d*)m?": "\\1:\\2" + "(\\d+[.,]?\\d*)h(\\d+[.,]?\\d*)": "\\1:\\2" }, { "(?<=from\\s+)now": "in" diff --git a/dateparser_data/supplementary_language_data/date_translation_data/en.yaml b/dateparser_data/supplementary_language_data/date_translation_data/en.yaml index d103bc577..38b31297d 100644 --- a/dateparser_data/supplementary_language_data/date_translation_data/en.yaml +++ b/dateparser_data/supplementary_language_data/date_translation_data/en.yaml @@ -67,7 +67,7 @@ simplifications: - a: '1' - (?:12\s+)?noon: '12:00' - (?:12\s+)?midnight: '00:00' - - (\d+[.,]?\d*)h(\d+[.,]?\d*)m?: \1:\2 + - (\d+[.,]?\d*)h(\d+[.,]?\d*): \1:\2 - (?<=from\s+)now: in - less than 1 minute ago: 45 second ago - (\d+[.,]?\d*) (decade|year|month|week|day|hour|minute|second)s? later: in \1 \2

Formats like "1h20m" should be treated as relative but are actually absolute When parsing a string like `"1h20m"`, it should be treated as a relative date; i.e. the same as it would treat the string `"1 hour and 20 minutes"`. However this is not the actual behaviour. ```py >>> from datetime import datetime >>> from dateparser import parse >>> datetime.now() datetime.datetime(2021, 10, 28, 18, 17, 11, 563442) >>> parse("1h20m") datetime.datetime(2021, 10, 28, 1, 20) >>> parse("in 1h20m") datetime.datetime(2021, 10, 28, 1, 20) >>> parse("in 1 hour and 20 minutes") datetime.datetime(2021, 10, 28, 19, 37, 34, 355178) ```

I agree that “in 1h20m” needs fixing, I’m less sure about the one without the “in”. I feel like for the interpretation of that one we should have a setting that could indicate if it should be interpreted as time in the future, time in the past, or absolute time (which I would keep as default). Even with the setting to prefer in the future it still treats "1h20m" as absolute. Also, I feel it would definitely make the most sense for "1h20m" and "1 hour 20 minutes" to give the same results as one is a literal abbreviation of the other. Note that a space makes a difference: ```py >>> parse("1h20m") datetime.datetime(2021, 10, 28, 1, 20) >>> parse("1h 20m") datetime.datetime(2021, 10, 28, 16, 46, 56, 916241) ``` That is really counter intuitive. I think it would make most sense to make all those as relative. A possible workaround for the time being: ```py string = re.sub(r"([dhms])(\d)", r"\1 \2", "1h20m") parse(string) ``` This will add a space which will cause it to be treated relatively

2024-02-15T20:36:12

0.0

[]

scrapinghub/dateparser

scrapinghub__dateparser-1183

3955530ad9a910bad5a463dc986fe383bb185f87

diff --git a/dateparser/parser.py b/dateparser/parser.py index 297e73f3b..dcd8a194d 100644 --- a/dateparser/parser.py +++ b/dateparser/parser.py @@ -563,18 +563,15 @@ def _correct_for_time_frame(self, dateobj, tz): # Convert dateobj to utc time to compare with self.now try: tz = tz or get_timezone_from_tz_string(self.settings.TIMEZONE) + tz_offset = tz.utcoffset(dateobj) except pytz.UnknownTimeZoneError: - tz = None + tz_offset = timedelta(hours=0) - if tz: - dateobj_time = (dateobj - tz.utcoffset(dateobj)).time() - else: - dateobj_time = dateobj.time() if "past" in self.settings.PREFER_DATES_FROM: - if self.now.time() < dateobj_time: + if self.now < dateobj - tz_offset: dateobj = dateobj + timedelta(days=-1) if "future" in self.settings.PREFER_DATES_FROM: - if self.now.time() > dateobj_time: + if self.now > dateobj - tz_offset: dateobj = dateobj + timedelta(days=1) # Reset dateobj to the original value, thus removing any offset awareness that may

Returned datetime skips a day I reported in #1092 that the datetime is returned correctly only when supplying the timezone in the argument but not via settings, which is still the behaviour as of 1.1.7. As of 1.1.8, both options give the same result again, however the date is wrong. See an example below. Note this is a different example than the one provided in #1092, the original example now works correctly on 1.1.8. This bug seems to occur when the time is already the next day in UTC. **Versions:** - Ubuntu 22.04.1 LTS - Python 3.10.6 - dateparser (1.1.8) - python-dateutil (2.8.2) - pytz (2022.6) - regex (2022.3.2) - tzlocal (4.2) **Minimum working example** ``` #!/usr/bin/env python3 import dateparser import datetime parse_settings={ 'PREFER_DATES_FROM': 'future', 'TO_TIMEZONE': 'etc/utc', 'RETURN_AS_TIMEZONE_AWARE': False, 'RELATIVE_BASE': datetime.datetime(2023, 5, 21, 15, 0) } time = dateparser.parse('7pm EDT', settings=parse_settings) print(time) # 2023-05-21 23:00:00 time = dateparser.parse('9pm EDT', settings=parse_settings) print(time) # 2023-05-23 01:00:00 parse_settings['TIMEZONE'] = 'america/new_york' time = dateparser.parse('7pm', settings=parse_settings) print(time) # 2023-05-21 23:00:00 time = dateparser.parse('9pm', settings=parse_settings) print(time) # 2023-05-23 01:00:00 ```

2023-09-14T02:39:40

0.0

[]

scrapinghub/dateparser

scrapinghub__dateparser-1129

7cd3b80369ea85aaf9b544991dfc7e2e122b8b08

diff --git a/dateparser/data/date_translation_data/ru.py b/dateparser/data/date_translation_data/ru.py index eade5dc92..1561b02b2 100644 --- a/dateparser/data/date_translation_data/ru.py +++ b/dateparser/data/date_translation_data/ru.py @@ -431,6 +431,12 @@ }, { "((?<=(через|спустя|в течение)\\s+)года?|(?<=[^\\d]\\s+|^)год(?=(\\s+назад)))": "1 год" + }, + { + "(\\d{3,}1)\\s*год\\s*$": "\\1" + }, + { + "(\\d*[02-9])\\s*год\\b": "\\1" } ] } diff --git a/dateparser_data/supplementary_language_data/date_translation_data/ru.yaml b/dateparser_data/supplementary_language_data/date_translation_data/ru.yaml index b6004c57e..3112ffa1c 100644 --- a/dateparser_data/supplementary_language_data/date_translation_data/ru.yaml +++ b/dateparser_data/supplementary_language_data/date_translation_data/ru.yaml @@ -103,3 +103,5 @@ simplifications: - ((?<=(через|спустя|в течение)\s+)недел[юи]|(?<=[^\d]\s+|^)неделю(?=(\s+назад))): 1 неделю - ((?<=(через|спустя|в течение)\s+)месяца?|(?<=[^\d]\s+|^)месяц(?=(\s+назад))): 1 месяц - ((?<=(через|спустя|в течение)\s+)года?|(?<=[^\d]\s+|^)год(?=(\s+назад))): 1 год + - (\d{3,}1)\s*год\s*$: \1 + - (\d*[02-9])\s*год\b: \1

possible fix of 'ago' problem in Russian '_2000 год_' in Russian is '_year 2000_' Before was parsed as '_2000 years ago_' ``` parse('2000 год') datetime.datetime(17, 8, 3, 18, 44, 48, 222615) ``` After ``` parse('2000 год') datetime.datetime(2000, 8, 3, 0, 0) ```

# [Codecov](https://codecov.io/gh/scrapinghub/dateparser/pull/340?src=pr&el=h1) Report > Merging [#340](https://codecov.io/gh/scrapinghub/dateparser/pull/340?src=pr&el=desc) into [master](https://codecov.io/gh/scrapinghub/dateparser/commit/53084340d11f442b1a681fb13f023b28a879857c?src=pr&el=desc) will **increase** coverage by `<.01%`. > The diff coverage is `100%`. [![Impacted file tree graph](https://codecov.io/gh/scrapinghub/dateparser/pull/340/graphs/tree.svg?height=150&width=650&token=XbZaFXS3m4&src=pr)](https://codecov.io/gh/scrapinghub/dateparser/pull/340?src=pr&el=tree) ```diff @@ Coverage Diff @@ ## master #340 +/- ## ========================================== + Coverage 97.61% 97.61% +<.01% ========================================== Files 20 20 Lines 1674 1677 +3 ========================================== + Hits 1634 1637 +3 Misses 40 40 ``` | [Impacted Files](https://codecov.io/gh/scrapinghub/dateparser/pull/340?src=pr&el=tree) | Coverage Δ | | |---|---|---| | [dateparser/freshness\_date\_parser.py](https://codecov.io/gh/scrapinghub/dateparser/pull/340/diff?src=pr&el=tree#diff-ZGF0ZXBhcnNlci9mcmVzaG5lc3NfZGF0ZV9wYXJzZXIucHk=) | `98.97% <100%> (+0.03%)` | :arrow_up: | ------ [Continue to review full report at Codecov](https://codecov.io/gh/scrapinghub/dateparser/pull/340?src=pr&el=continue). > **Legend** - [Click here to learn more](https://docs.codecov.io/docs/codecov-delta) > `Δ = absolute <relative> (impact)`, `ø = not affected`, `? = missing data` > Powered by [Codecov](https://codecov.io/gh/scrapinghub/dateparser/pull/340?src=pr&el=footer). Last update [5308434...3cf105a](https://codecov.io/gh/scrapinghub/dateparser/pull/340?src=pr&el=lastupdated). Read the [comment docs](https://docs.codecov.io/docs/pull-request-comments). Please, see https://dateparser.readthedocs.io/en/latest/contributing.html#guidelines-for-editing-translation-data The reason behind this is that he `relative-time` parser is applied before the `absolute-time` parser... so it could be fixed by using the `PARSERS` settings, either deleting the `relative-time` parser or putting it in the last position. i.e.: ``` dateparser.parse('2000 год', settings={'PARSERS': ['timestamp', 'custom-formats', 'absolute-time', 'base-formats', 'relative-time']}) ``` Your approach is not good because we can't hardcode that regex there and would have side effects for other languages. However, the issue is still valid. We could probably add a regex into the ru.yaml file. Maybe a simplification as `- (\d+)\s*год: \1` would work, however, I can't speak Russian and I don't know if we should add something more, as it could be that with other following or preceding words (as "ago" or "in") this simplification shouldn't be applied.

2023-01-24T15:49:19

0.0

[]

scrapinghub/dateparser

scrapinghub__dateparser-1121

78dc29866aca496642742f6c7aeffbc1bb508043

diff --git a/dateparser/data/date_translation_data/en.py b/dateparser/data/date_translation_data/en.py index 5c5b18acf..a90dfed21 100644 --- a/dateparser/data/date_translation_data/en.py +++ b/dateparser/data/date_translation_data/en.py @@ -93,7 +93,6 @@ "month": [ "mo", "month", - "mon", "months" ], "week": [ diff --git a/dateparser_data/supplementary_language_data/date_translation_data/en.yaml b/dateparser_data/supplementary_language_data/date_translation_data/en.yaml index 6885c5562..a2736f213 100644 --- a/dateparser_data/supplementary_language_data/date_translation_data/en.yaml +++ b/dateparser_data/supplementary_language_data/date_translation_data/en.yaml @@ -16,7 +16,6 @@ year: - y - years month: - - mon - months week: - weeks

Is not parsing properly the Last-Modified header value when is a Monday Hi, Just in case, in the previous version we were able to parse the following 'Mon, 05 Dec 2022 19:43:03 GMT' without any problem and now is returning None. This is happening only when is Monday. It could be related to this commit: https://github.com/scrapinghub/dateparser/commit/89a5b29517a358e2784cc818687b7b6bed343159 ? Thank you! Best, Nicolas.

I have confirmed https://github.com/scrapinghub/dateparser/commit/89a5b29517a358e2784cc818687b7b6bed343159 as the cause.

2023-01-10T15:51:06

0.0

[]

scrapinghub/dateparser

scrapinghub__dateparser-1103

a51d9bd31c9c28f43bf069f87b42e0cd3454001e

diff --git a/dateparser/data/date_translation_data/en.py b/dateparser/data/date_translation_data/en.py index df187f684..5c5b18acf 100644 --- a/dateparser/data/date_translation_data/en.py +++ b/dateparser/data/date_translation_data/en.py @@ -87,11 +87,13 @@ "year": [ "year", "yr", + "y", "years" ], "month": [ "mo", "month", + "mon", "months" ], "week": [ @@ -101,6 +103,7 @@ ], "day": [ "day", + "d", "days" ], "hour": [ diff --git a/dateparser_data/supplementary_language_data/date_translation_data/en.yaml b/dateparser_data/supplementary_language_data/date_translation_data/en.yaml index 69ef6c515..6885c5562 100644 --- a/dateparser_data/supplementary_language_data/date_translation_data/en.yaml +++ b/dateparser_data/supplementary_language_data/date_translation_data/en.yaml @@ -13,12 +13,15 @@ decade: - decade - decades year: + - y - years month: + - mon - months week: - weeks day: + - d - days hour: - h

Cannot parse short days ago. Exp 1d ago Hi, I cannot parse some type of short days ago. Some examples - `1d ago` - `1mon ago` - `2y ago` ```python from dateparser import parse parse('1mon ago') >> None ```

another issue I find connected to this one: other languages sometimes uses 'ago' and for french the dateparser cannot parse it. ```python from dateparser import parse parse('4 hours ago') >> datetime.datetime(2022, 5, 31, 8, 25, 46, 392093) ``` ```python from dateparser import parse parse('4 heures ago') >> None ``` ```python from dateparser import parse parse('il y a 4 heures') >> datetime.datetime(2022, 5, 31, 8, 29, 7, 442826) ``` I guess it should be really easy to fix this issue, unless the short days as @hoangchunghien presented are a feature dateparser still doesn't includes

2022-11-24T13:55:50

0.0

[]

scrapinghub/dateparser

scrapinghub__dateparser-1084

78dc29866aca496642742f6c7aeffbc1bb508043

diff --git a/dateparser/languages/locale.py b/dateparser/languages/locale.py index a64c7e79b..e3b43b6be 100644 --- a/dateparser/languages/locale.py +++ b/dateparser/languages/locale.py @@ -263,7 +263,7 @@ def _sentence_split(self, string, settings): splitters_dict = {1: r'[\.!?;…\r\n]+(?:\s|$)*', # most European, Tagalog, Hebrew, Georgian, # Indonesian, Vietnamese - 2: r'(?:[¡¿]+|[\.!?;…\r\n]+(?:\s|$))+', # Spanish + 2: r'[\.!?;…\r\n]+(\s*[¡¿]*|$)|[¡¿]+', # Spanish 3: r'[|!?;\r\n]+(?:\s|$)+', # Hindi and Bangla 4: r'[。…‥\.!?？！;\r\n]+(?:\s|$)+', # Japanese and Chinese 5: r'[\r\n]+', # Thai @@ -275,9 +275,7 @@ def _sentence_split(self, string, settings): split_reg = abbreviation_string + splitters_dict[self.info['sentence_splitter_group']] sentences = re.split(split_reg, string) - for i in sentences: - if not i: - sentences.remove(i) + sentences = filter(None, sentences) return sentences def _simplify_split_align(self, original, settings):

SECURITY: bad regex pattern in 'dateparser/languages/locale.py' will cause 'ReDos' security problem. #### Problem description i found one bad regex pattern in 'dateparser/languages/locale.py' ``` r'(?:[¡¿]+|[\.!?;…\r\n]+(?:\s|$))+', # Spanish ``` those pattern will cause 'ReDos' security problem, proof of code like below ``` import re p = re.compile(r'(?:[¡¿]+|[\.!?;…\r\n]+(?:\s|$))+') re.findall(p, "?"*100000+"x") ``` run the above code, cpu utilization will be 100% in a very long period. more detail about 'ReDos' please see [owasp](https://owasp.org/www-community/attacks/Regular_expression_Denial_of_Service_-_ReDoS). #### effect of this security problem some api will call the pattern，like below ``` from dateparser.search import search_dates search_dates("?"*100000+"x") ```

Hi @leveryd I tried running `%time search_dates("x"*100000+"x")` (note that the first character is an "x") and I got: ``` CPU times: user 3min 27s, sys: 6.04 s, total: 3min 33s Wall time: 3 min 38s ``` The CPU was at 99% :sweat_smile: but it didn't hang :stuck_out_tongue:. Then I tried with your example, changing the first character to "?": `%time search_dates("?"*100000+"x")`. And this is what I found: ``` CPU times: user 9min 23s, sys 9.61 s, total: 9 min 32s Wall time: 9min 37s ``` So it seems that it's definitely having some impact, but it's not so critical as I initially thought. After this I introduced the mentioned regex to this reDoS checker: http://redos-checker.surge.sh/ and it wasn't detected as a vulnerable regex (though it could be wrong). Before finishing doing my tests, I tried this: `%time search_dates("?"*10000+"x")` (4 zeroes) and I got: ``` Wall time: 4.54 s ``` and then I tried switching from "?" to "x" (`%time search_dates("x"*10000+"x")`) and I got this: ``` Wall time: 20 s ``` That's definitely suspicious, and I'm not sure if it's a vulnerable regex or not... I would appreciate any help from people with more experience with reDoS. regexploit also reports this as exponential: ``` Redos(starriness=11, prefix_sequence=SEQ{ [27:'] }, redos_sequence=SEQ{ [21:!,2026,0a,0d,2e:.,3b:;,3f:?]{1+}{1+} [27:'] }, repeated_character=[21:!,2026,0a,0d,2e:.,3b:;,3f:?], killer=None) Worst-case complexity: 11 ⭐⭐⭐⭐⭐⭐⭐⭐⭐⭐ (exponential) Repeated character: [21:!,2026,0a,0d,2e:.,3b:;,3f:?] Example: '\'' + '!' * 3456 ``` btw: snyk reports this as vulnerability of this library: https://security.snyk.io/vuln/SNYK-PYTHON-DATEPARSER-1063229 Unfortunately my organization's new security policies have bumped into that SNYK vulnerability where we have a couple of projects that use this library. I won't claim to be a regexpert, but it seems like a quick fix might be to limit the quantifiers to a range one could "reasonably" expect as a sentence separator. E.g., instead of: `(?:[¡¿]+|[\.!?;…\r\n]+(?:\s|$))+` Something like: `(?:[¡¿]{1,50}|[\.!?;…\r\n]{1,50}(?:\s|$))+` (I can't imagine anyone using more than 50 punctuation marks to end a sentence but just having that limit seems to cut the processing time significantly) Would that be a satisfactory resolution to get the SNYK report closed? Also, regexploit no longer seems to report the original regex as vulnerable, I'm not sure what changed there. Any updates on that?

2022-10-12T10:31:07

0.0

[]

scrapinghub/dateparser

scrapinghub__dateparser-1028

4490ca6de44f2e9771e2017e0695f7d9f2d7207e

diff --git a/dateparser/date.py b/dateparser/date.py index ac8283ba2..1a91764e4 100644 --- a/dateparser/date.py +++ b/dateparser/date.py @@ -35,7 +35,7 @@ RE_SANITIZE_SKIP = re.compile(r'\t|\n|\r|\u00bb|,\s\u0432\b|\u200e|\xb7|\u200f|\u064e|\u064f', flags=re.M) RE_SANITIZE_RUSSIAN = re.compile(r'([\W\d])\u0433\.', flags=re.I | re.U) -RE_SANITIZE_PERIOD = re.compile(r'(?<=\D+)\.', flags=re.U) +RE_SANITIZE_PERIOD = re.compile(r'(?<=[^0-9\s])\.', flags=re.U) RE_SANITIZE_ON = re.compile(r'^.*?on:\s+(.*)') RE_SANITIZE_APOSTROPHE = re.compile('|'.join(APOSTROPHE_LOOK_ALIKE_CHARS)) diff --git a/dateparser/parser.py b/dateparser/parser.py index e2df66b03..30536cab7 100644 --- a/dateparser/parser.py +++ b/dateparser/parser.py @@ -223,7 +223,7 @@ class _parser: def __init__(self, tokens, settings): self.settings = settings - self.tokens = list(tokens) + self.tokens = [(t[0].strip(), t[1]) for t in list(tokens)] self.filtered_tokens = [(t[0], t[1], i) for i, t in enumerate(self.tokens) if t[1] <= 1] self.unset_tokens = []

Regression dates with dots and spaces Since upgrading from `0.7.6` to `1.0.0` or `1.1.0`, dates with dots and spaces are not parsed correctly: ```python from datetime import datetime from dateparser import parse # both assert pass in `0.7.6` but not in `1.0.0` or `1.1.0` assert parse("26. 10.21") == datetime(2021, 10, 26) assert parse("26. 10.21", settings={"STRICT_PARSING": True}) == datetime(2021, 10, 26) ``` Is this intended behavior or a bug with the new version? This might be related to https://github.com/scrapinghub/dateparser/pull/741 Thanks!

2021-12-18T16:25:51

0.0

[]

scrapinghub/dateparser

scrapinghub__dateparser-1005

8af4f296a41e5ca9b2e3cde7deccb26a0652b70a

diff --git a/dateparser/data/date_translation_data/en.py b/dateparser/data/date_translation_data/en.py index 62e2649c2..a156bf899 100644 --- a/dateparser/data/date_translation_data/en.py +++ b/dateparser/data/date_translation_data/en.py @@ -126,7 +126,8 @@ ], "relative-type": { "0 day ago": [ - "today" + "today", + "till date" ], "0 hour ago": [ "this hour" diff --git a/dateparser_data/supplementary_language_data/date_translation_data/en.yaml b/dateparser_data/supplementary_language_data/date_translation_data/en.yaml index 600cf325c..b21e70b14 100644 --- a/dateparser_data/supplementary_language_data/date_translation_data/en.yaml +++ b/dateparser_data/supplementary_language_data/date_translation_data/en.yaml @@ -47,6 +47,8 @@ relative-type: - this decade in 1 decade: - next decade + 0 day ago: + - till date relative-type-regex: in \1 decade:

Is the string 'tilldate' supported in the parser? Hi! Is the string `tilldate`/`till date` supported? This is what i've tried: ``` >>> from dateparser import parse, search >>> parsed_date = parse('tilldate') >>> print(parsed_date) None >>> parsed_date = parse('till date') >>> print(parsed_date) None >>> parsed_date = search.search_dates('tilldate') >>> print(parsed_date) None >>> parsed_date = search.search_dates('till date') >>> print(parsed_date) None ``` It returns `None` but i was expecting it to return `datetime.datetime(2021, 1, 6, 0, 0)` instead. Thanks!

Looks like a valid expression: https://english.stackexchange.com/a/146771/87187 Not sure if we support any of the alternative expressions, but I guess it makes sense to eventually support it. Hi @Gallaecio, thanks for a really quick response. Good to know that the expression would be supported in future. Is the project open for contributions regarding this? Cheers! I _think_ so, specially if it’s doable by modifying the corresponding YAML file. See https://dateparser.readthedocs.io/en/latest/contributing.html#guidelines-for-editing-translation-data

2021-10-23T16:29:27

0.0

[]

oeg-upm/yatter

oeg-upm__yatter-85

7421fc9fc07f38cbcc2a2730a3110bcb6c747c14

diff --git a/VERSION b/VERSION index e25d8d9..ea710ab 100644 --- a/VERSION +++ b/VERSION @@ -1,1 +1,1 @@ -1.1.5 +1.2 \ No newline at end of file diff --git a/src/yatter/__init__.py b/src/yatter/__init__.py index f586a89..e32272d 100644 --- a/src/yatter/__init__.py +++ b/src/yatter/__init__.py @@ -17,13 +17,14 @@ def translate(yarrrml_data, mapping_format=RML_URI): rml_mapping.extend(generate_database_connections(yarrrml_data, list_initial_sources)) rml_mapping.extend(add_logical_targets(yarrrml_data)) rml_mapping.extend(add_functions(yarrrml_data)) + external_refs = [] try: mappings, mapping_format = get_non_asserted_mappings(yarrrml_data, mapping_format) for mapping in yarrrml_data.get(YARRRML_MAPPINGS): if mapping_format == R2RML_URI: source_list = add_table(yarrrml_data, mapping, list_initial_sources) else: - source_list = add_source(yarrrml_data, mapping, list_initial_sources) + source_list, external_refs = add_source(yarrrml_data, mapping, list_initial_sources) subject_list = add_subject(yarrrml_data, mapping, mapping_format) pred = add_predicate_object_maps(yarrrml_data, mapping, mapping_format) it = 0 @@ -37,6 +38,9 @@ def translate(yarrrml_data, mapping_format=RML_URI): rml_mapping[len(rml_mapping) - 1] = rml_mapping[len(rml_mapping) - 1][:-2] rml_mapping.append(".\n\n\n") it = it + 1 + external_refs = list(dict.fromkeys(external_refs)) + for ref in external_refs: + rml_mapping.append(ref) logger.info("RML content is created!") rml_mapping_string = "".join(rml_mapping) diff --git a/src/yatter/constants.py b/src/yatter/constants.py index c5aad0a..6f363dd 100644 --- a/src/yatter/constants.py +++ b/src/yatter/constants.py @@ -21,6 +21,7 @@ VOID_URI = 'http://rdfs.org/ns/void#' FNML_URI = 'http://semweb.mmlab.be/ns/fnml#' GREL_URI = 'http://users.ugent.be/~bjdmeest/function/grel.ttl#' +SD_URI = 'https://w3id.org/okn/o/sd/' RML_PREFIX = '@prefix' RML_BASE = '@base' @@ -28,6 +29,7 @@ RML_LOGICAL_SOURCE = 'rml:logicalSource' RML_SOURCE = 'rml:source' RML_REFERENCE_FORMULATION = 'rml:referenceFormulation' +RML_REFERENCE_FORMULATION_CLASS = 'rml:ReferenceFormulation' RML_ITERATOR = 'rml:iterator' RML_REFERENCE = 'rml:reference' RML_LANGUAGE_MAP = 'rml:languageMap' @@ -103,7 +105,16 @@ D2RQ_USER = 'd2rq:username' D2RQ_PASS = 'd2rq:password' - +############################################################################## +############################# SD CONSTANTS ########################### +############################################################################## +SD_DATASET_SPEC = 'sd:DatasetSpecification' +SD_NAME = 'sd:name' +SD_HAS_DATA_TRANSFORMATION = 'sd:hasDataTransformation' +SD_HAS_SOFTWARE_REQUIREMENTS = 'sd:hasSoftwareRequirements' +SD_HAS_SOURCE_CODE= 'sd:hasSourceCode' +SD_PROGRAMMING_LANGUAGE = 'sd:programmingLanguage' +KG4DI_DEFINED_BY = 'kg4di:definedBy' ############################################################################## ############################# YARRRML CONSTANTS ########################### ############################################################################## @@ -122,6 +133,11 @@ YARRRML_USERNAME = 'username' YARRRML_PASSWORD = 'password' +YARRRML_STRUCTURE_DEFINER = 'structureDefiner' +YARRRML_SOFTWARE_SPECIFICATION = 'softwareSpecification' +YARRRML_PROGRAMMING_LANGUAGE = 'programmingLanguage' +YARRRML_SOFTWARE_REQUIREMENTS = 'softwareRequirements' + YARRRML_MAPPINGS = 'mappings' # used for mappings in conditions and mappings main key YARRRML_MAPPING = 'mapping' YARRRRL_MAPPINGS_SHORTCUT = 'm' diff --git a/src/yatter/predicateobject.py b/src/yatter/predicateobject.py index bc2434b..0e2a059 100644 --- a/src/yatter/predicateobject.py +++ b/src/yatter/predicateobject.py @@ -288,7 +288,7 @@ def ref_mapping(data, mapping, om, yarrrml_key, ref_type_property, mapping_forma else: if mapping_format == STAR_URI: object = STAR_OBJECT - source_list = add_source(data, mapping_join, list_initial_sources) + source_list, external_refs = add_source(data, mapping_join, list_initial_sources) number_joins_rml = len(subject_list) * len(source_list) for i in range(number_joins_rml): diff --git a/src/yatter/source.py b/src/yatter/source.py index f5fb009..222acd5 100644 --- a/src/yatter/source.py +++ b/src/yatter/source.py @@ -30,6 +30,7 @@ def add_source(data, mapping, list_initial_sources): source_template = "\t" + RML_LOGICAL_SOURCE + " [\n\t\ta " + RML_LOGICAL_SOURCE_CLASS + \ ";\n\t\t" + RML_SOURCE + " " final_list = [] + external_references_list = [] sources = get_sources(data, mapping) for source in sources: db_identifier = mapping @@ -40,13 +41,24 @@ def add_source(data, mapping, list_initial_sources): if YARRRML_ACCESS in source: if YARRRML_QUERY in source: final_list.append(source_template + database_source(mapping, source, db_identifier)) + elif YARRRML_STRUCTURE_DEFINER in source: + source, external_references = add_in_memory_source(mapping,source) + final_list.append(source_template + source) + if external_references is not None: + external_references_list.append(external_references) else: final_list.append(source_template + add_source_full(mapping, source)) elif type(source) is list: - final_list.append(source_template + add_source_simplified(mapping, source)) + if "$(" in source[0]: + source, external_references = add_in_memory_source(mapping, source) + final_list.append(source_template + source) + if external_references is not None: + external_references_list.append(external_references) + else: + final_list.append(source_template + add_source_simplified(mapping, source)) else: raise Exception("ERROR: source " + source + " in mapping " + mapping + " not valid") - return final_list + return final_list, external_references_list def add_table(data, mapping, list_initial_sources): @@ -105,6 +117,53 @@ def add_source_simplified(mapping, source): + source[1] + "\";\n\t];\n" return source_rdf +def add_in_memory_source(mapping, source): + external_reference_formulation = None + if type(source) is list: + source = extend_in_memory(source) + source_rdf = "[\n\t\t\ta " + SD_DATASET_SPEC + ";\n\t\t\t" + + access = str(source.get(YARRRML_ACCESS)).replace("$(","").replace(")","") + source_rdf += SD_NAME + " \"" + access + "\";\n" + + if YARRRML_SOFTWARE_SPECIFICATION in source: + source_rdf += "\t\t\t" + SD_HAS_DATA_TRANSFORMATION + "[\n\t\t\t\t" + + if YARRRML_SOFTWARE_REQUIREMENTS in source.get(YARRRML_SOFTWARE_SPECIFICATION): + software_requirements = str(source.get(YARRRML_SOFTWARE_SPECIFICATION)[YARRRML_SOFTWARE_REQUIREMENTS]) + source_rdf += SD_HAS_SOFTWARE_REQUIREMENTS + " \""+ software_requirements +"\";\n\t\t\t\t" + + if YARRRML_PROGRAMMING_LANGUAGE in source.get(YARRRML_SOFTWARE_SPECIFICATION): + programming_language = str(source.get(YARRRML_SOFTWARE_SPECIFICATION)[YARRRML_PROGRAMMING_LANGUAGE]) + source_rdf += SD_HAS_SOURCE_CODE +"[\n\t\t\t\t\t" + SD_PROGRAMMING_LANGUAGE + " \"" + programming_language + "\";" + source_rdf +="\n\t\t\t\t];\n" + + source_rdf += "\t\t\t];\n" + source_rdf += "\t\t];\n" + + if YARRRML_ITERATOR in source: + source_rdf += "\t\t" + RML_ITERATOR + " \"" + source.get(YARRRML_ITERATOR) + "\";\n" + + if YARRRML_REFERENCE_FORMULATION in source: + reference_formulation = str(source.get(YARRRML_REFERENCE_FORMULATION)) + source_rdf += "\t\t" + RML_REFERENCE_FORMULATION + " ql:"+ reference_formulation + ";\n" + external_reference_formulation = "ql:" + reference_formulation + " a " + RML_REFERENCE_FORMULATION_CLASS +";\n" + external_reference_formulation +="\t" + KG4DI_DEFINED_BY +" \""+source.get(YARRRML_STRUCTURE_DEFINER) +"\"." + source_rdf += "\t];\n" + + + + return source_rdf, external_reference_formulation + +def extend_in_memory(source): + features = source[0].split("~") + access = features[0] + defined_by = features[1].split("-")[0] + reference_formulation = features[1].split("-")[1] + extended_source = {"access": access, "structureDefiner": defined_by, "referenceFormulation": reference_formulation} + if len(source) == 2: + extended_source["iterator"] = source[1] + return extended_source def add_source_full(mapping, source): source_rdf = ""

In-memory YARRRML to RML support (new-feature) Support of in-memory extension YARRRML syntax to RML.

Test-cases for testing the solution are added by @TheRazorace in #67

2024-10-29T10:46:57

0.0

[]

oeg-upm/yatter

oeg-upm__yatter-84

08a6e3db303122671e102ba560e12b67ce5568bc

diff --git a/src/yatter/__main__.py b/src/yatter/__main__.py index 4b71d8c..5cb9e92 100644 --- a/src/yatter/__main__.py +++ b/src/yatter/__main__.py @@ -5,8 +5,8 @@ from . import translate, inverse_translation, merge_mappings from .constants import * -def write_results(mapping): +def write_results(mapping): if type(mapping) is str: output_file = open(args.output_mapping_path, "w") output_file.write(mapping) @@ -18,6 +18,7 @@ def write_results(mapping): yaml.default_flow_style = False yaml.dump(mapping, f) + def parse_inputs(): input_format = RML_URI yaml = YAML(typ='safe', pure=True) @@ -25,7 +26,8 @@ def parse_inputs(): if args.input_mapping_path.endswith('.yml') or args.input_mapping_path.endswith('.yaml'): with open(args.input_mapping_path) as f: input_data = yaml.load(f) - elif args.input_mapping_path.endswith('.ttl') or args.input_mapping_path.endswith('.rml') or args.input_mapping_path.endswith('.r2rml'): + elif args.input_mapping_path.endswith('.ttl') or args.input_mapping_path.endswith( + '.rml') or args.input_mapping_path.endswith('.r2rml'): input_data = Graph() input_data.parse(args.input_mapping_path, format="turtle") else: diff --git a/src/yatter/predicateobject.py b/src/yatter/predicateobject.py index 7ddb90f..bc2434b 100644 --- a/src/yatter/predicateobject.py +++ b/src/yatter/predicateobject.py @@ -372,15 +372,13 @@ def add_inverse_pom(mapping_id, rdf_mapping, classes, prefixes): logger.error("ERROR: There is POM without predicate map defined") raise Exception("Review your mapping " + str(mapping_id)) - prefix = list({i for i in prefixes if predicate.startswith(prefixes[i])}) - if not predicate.startswith("http") and "{" not in predicate: + if not predicate.startswith("http"): predicate = '$(' + predicate + ')' - elif "{" in predicate and "}" in predicate: + elif predicate.startswith("http") and "{" not in predicate: + predicate = find_prefixes(predicate, prefixes) + else: predicate = predicate.replace('{', '$(').replace('}', ')') - elif prefix: - predicate = tm['predicateValue'].toPython().replace(prefixes[prefix[0]], prefix[0] + ":") - predicate = find_prefixes(predicate,prefixes) if tm['parentTriplesMap']: if tm['child']: diff --git a/src/yatter/subject.py b/src/yatter/subject.py index 3b28037..9934c85 100644 --- a/src/yatter/subject.py +++ b/src/yatter/subject.py @@ -54,7 +54,7 @@ def add_subject(data, mapping, mapping_format): if YARRRML_TARGETS in individual_subject: subject_termmap = subject_termmap[0:-3]+"\t"+RML_LOGICAL_TARGET+" <"+individual_subject[YARRRML_TARGETS]+">\n\t];\n" - if YARRRML_TYPE in individual_subject: + if isinstance(individual_subject, dict) and YARRRML_TYPE in individual_subject: if individual_subject.get(YARRRML_TYPE) == YARRRML_BLANK: subject_termmap = subject_termmap[0:-3] + "\t" + R2RML_TERMTYPE + " " + R2RML_BLANK_NODE +"\n\t];\n"

Package inconsistency between versions. Error: 'str' object has no attribute 'get' in yatter 1.1.4 The following YARRRML excerpt does work with Yatter v1.1.0 (Python 3.9), but does not work with Yatter v1.1.4 (tested with Python 3.9 and Python 3.12), even though the syntax is the same and seems correct, according to documentation. Error: The YARRRML mapping has not been translated. 'str' object has no attribute 'get' ``` prefixes: mlso: http://w3id.org/mlso/ mls: http://www.w3.org/ns/mls# rdfs: http://www.w3.org/2000/01/rdf-schema# sources: evaluations_meta: ['/evaluations_meta.json~jsonpath', '$.*'] mappings: evaluationMeasureDatatype: sources: - evaluations_meta s: http://w3id.org/rexpek-kg/thermal_uc_v1/evaluationMeasure/$(name)/characteristic/datatype po: - [a, mlso:EvaluationMeasureCharacteristic] - [rdfs:label, Datatype, rdfs:Literal] - [mls:hasValue, $(datatype), rdfs:Literal] ``` Package inconsistency between versions. Error: 'str' object has no attribute 'get' in yatter 1.1.4 The following YARRRML excerpt does work with Yatter v1.1.0 (Python 3.9), but does not work with Yatter v1.1.4 (tested with Python 3.9 and Python 3.12), even though the syntax is the same and seems correct, according to documentation. Error: The YARRRML mapping has not been translated. 'str' object has no attribute 'get' ``` prefixes: mlso: http://w3id.org/mlso/ mls: http://www.w3.org/ns/mls# rdfs: http://www.w3.org/2000/01/rdf-schema# sources: evaluations_meta: ['/evaluations_meta.json~jsonpath', '$.*'] mappings: evaluationMeasureDatatype: sources: - evaluations_meta s: http://w3id.org/rexpek-kg/thermal_uc_v1/evaluationMeasure/$(name)/characteristic/datatype po: - [a, mlso:EvaluationMeasureCharacteristic] - [rdfs:label, Datatype, rdfs:Literal] - [mls:hasValue, $(datatype), rdfs:Literal] ```

2024-08-29T11:20:26

0.0

[]

oeg-upm/yatter

oeg-upm__yatter-73

75b4e3ae0cb850f3fa261c1ac99d0584ea06892d

diff --git a/VERSION b/VERSION index 524cb55..45a1b3f 100644 --- a/VERSION +++ b/VERSION @@ -1,1 +1,1 @@ -1.1.1 +1.1.2 diff --git a/src/yatter/mapping.py b/src/yatter/mapping.py index 1edb0a9..4a0fa71 100644 --- a/src/yatter/mapping.py +++ b/src/yatter/mapping.py @@ -1,7 +1,6 @@ from .import * prefixes = {} - def add_mapping(mapping, mappings, it): map_template = "<" + mapping + "_" + str(it) + "> a " if mappings[mapping]: @@ -99,7 +98,7 @@ def add_inverse_prefix(rdf_mapping): def get_non_asserted_mappings(yarrrml_data, mapping_format): - mappings = dict.fromkeys(list(yarrrml_data.get(YARRRML_MAPPINGS).keys())) + mappings = dict.fromkeys(list(yarrrml_data.get(YARRRML_MAPPINGS).keys())) for mapping in yarrrml_data.get(YARRRML_MAPPINGS): keys = yarrrml_data.get(YARRRML_MAPPINGS).get(mapping).keys() for key in keys: diff --git a/src/yatter/predicateobject.py b/src/yatter/predicateobject.py index d8fa5e8..c2d9358 100644 --- a/src/yatter/predicateobject.py +++ b/src/yatter/predicateobject.py @@ -150,7 +150,7 @@ def add_predicate_object(data, mapping, predicate_object, mapping_format=RML_URI execution = True template += generate_rml_termmap(R2RML_PREDICATE, R2RML_PREDICATE_CLASS, pm_value,"\t\t\t") if execution: - template = template.replace(R2RML_CONSTANT + " " + pm_value, RML_EXECUTION + " <" + pm_value + ">") + template = template.replace(R2RML_CONSTANT + " \"" + pm_value + "\"", RML_EXECUTION + " <" + pm_value + ">") if YARRRML_TARGETS in pm: template = template[0:-3] + "\t" + RML_LOGICAL_TARGET + " <" + pm[YARRRML_TARGETS] + ">\n\t\t];\n" @@ -225,7 +225,7 @@ def add_predicate_object(data, mapping, predicate_object, mapping_format=RML_URI object_value, "\t\t\t", mapping_format) elif YARRRML_FUNCTION in om: template += generate_rml_termmap(R2RML_OBJECT, R2RML_OBJECT_CLASS, om[YARRRML_FUNCTION], "\t\t\t", mapping_format) - template = template.replace(R2RML_CONSTANT+" "+om[YARRRML_FUNCTION], RML_EXECUTION + " <" + om.get( + template = template.replace(R2RML_CONSTANT+" \""+om[YARRRML_FUNCTION]+ "\"", RML_EXECUTION + " <" + om.get( YARRRML_FUNCTION) + ">") else: template += generate_rml_termmap(R2RML_OBJECT, R2RML_OBJECT_CLASS, diff --git a/src/yatter/subject.py b/src/yatter/subject.py index d77691d..3b28037 100644 --- a/src/yatter/subject.py +++ b/src/yatter/subject.py @@ -48,7 +48,7 @@ def add_subject(data, mapping, mapping_format): subject_termmap = generate_rml_termmap(R2RML_SUBJECT, R2RML_SUBJECT_CLASS, subject_value, "\t\t", mapping_format) if YARRRML_FUNCTION in individual_subject: - subject_termmap = subject_termmap.replace(R2RML_CONSTANT+" "+subject_value, RML_EXECUTION + " <" + individual_subject.get( + subject_termmap = subject_termmap.replace(R2RML_CONSTANT+" \""+subject_value + "\"", RML_EXECUTION + " <" + individual_subject.get( YARRRML_FUNCTION) + ">") if YARRRML_TARGETS in individual_subject: diff --git a/src/yatter/termmap.py b/src/yatter/termmap.py index ee2334d..8693fe1 100644 --- a/src/yatter/termmap.py +++ b/src/yatter/termmap.py @@ -19,11 +19,16 @@ def get_termmap_type(text, mapping_format): ## Generates a TermMap (subject, predicate, object) based on the property, class and the text def generate_rml_termmap(rml_property, rml_class, text, indentation, mapping_format=RML_URI): - template = indentation[0:-1] + rml_property + " [\n"+indentation+"a " + rml_class + ";\n" + indentation + from .mapping import prefixes + template = indentation[0:-1] + rml_property + " [\n" + indentation + "a " + rml_class + ";\n" + indentation term_map = get_termmap_type(text, mapping_format) if term_map == R2RML_TEMPLATE: text = generate_rml_template(text) text = text.replace('"', r'\"') + if ":" in text: + text_prefix_split = text.split(":") + if text_prefix_split[0] in prefixes: + text = prefixes[text_prefix_split[0]] + text_prefix_split[1] elif term_map == RML_REFERENCE or term_map == R2RML_COLUMN: text = text.replace("$(", "").replace(")", "") text = text.replace('"', r'\"') @@ -41,7 +46,10 @@ def generate_rml_termmap(rml_property, rml_class, text, indentation, mapping_for if text.startswith("http"): template += term_map + " <" + text + ">;\n" + indentation[0:-1] + "];\n" else: - template += term_map + " " + text + ";\n"+indentation[0:-1]+"];\n" + if ":" in text or "<" in text: + template += term_map + " " + text + ";\n"+indentation[0:-1]+"];\n" + else: + template += term_map + " \"" + text + "\";\n" + indentation[0:-1] + "];\n" return template

Resolve prefixes in the generation of SubjectMap **Describe the bug** Considering a SubjectMap `s: ex:person/$(firstname)` and a prefix `ex`, the prefix is not resolved when generating the SubjectMap. **To Reproduce** Use this mapping file ``` prefixes: ex: "http://example.com/" mappings: person: sources: - ['persons.json~jsonpath', '$.persons[*]'] s: ex:person/$(firstname) po: - [a, foaf:Person] ``` **Expected behavior** In the SubjectMap generated the `ex:` is resolved as `http://example.com/`. ``` rr:subjectMap [ a rr:SubjectMap; rr:template "http://example.com/person/{firstname}"; ]; ``` problem translating functions? For the following YARRRML document: ``` prefixes: ex: "http://www.semanticweb.org/tijnd/" ontology: "http://www.semanticweb.org/tijnd/ontology/" schema: "https://schema.org/" idlab-fn: "http://example.com/idlab/function/" grel: "http://users.ugent.be/~bjdmeest/function/grel.ttl#" owl: "http://www.w3.org/2002/07/owl#" rdf: "http://www.w3.org/1999/02/22-rdf-syntax-ns#" rdfs: "http://www.w3.org/2000/01/rdf-schema#" xsd: "http://www.w3.org/2001/XMLSchema#" sources: imdb: access: imdb_video_game_rating_modified.csv referenceFormulation: csv mappings: director: sources: - imdb s: ex:Person/$(title) #Not really the correct subject, only used to simply test it po: - [a, schema:Person] - p: schema:name o: function: grel:string_split parameters: - [grel:p_string_sep, ", "] - [grel:valueParameter, $(directors)] ``` Yatter has problems using the function, mainly with the ", " it gives an error "objectlist expected." The same YARRRML document is properly translated with Matey. The dataset considered for this task is this: https://www.kaggle.com/datasets/nyagami/video-game-ratings-from-imdb (if it matters)

Sorry for the late fix, but now prefixes are substituted in any TermMap with reference and prefixes. This is already solved! Thanks :)

2023-12-19T12:19:13

0.0

[]

oeg-upm/yatter

oeg-upm__yatter-70

da6e9c02e810619c58b95071a1be3e6122a90625

diff --git a/VERSION b/VERSION index 9084fa2..524cb55 100644 --- a/VERSION +++ b/VERSION @@ -1,1 +1,1 @@ -1.1.0 +1.1.1 diff --git a/src/yatter/constants.py b/src/yatter/constants.py index df7387a..08d2c9e 100644 --- a/src/yatter/constants.py +++ b/src/yatter/constants.py @@ -81,6 +81,7 @@ R2RML_TERMTYPE = 'rr:termType' R2RML_LANGUAGE = 'rr:language' R2RML_IRI = 'rr:IRI' +R2RML_BLANK_NODE = 'rr:BlankNode' R2RML_LITERAL = 'rr:Literal' R2RML_REFOBJECT_CLASS = 'rr:RefObjectMap' R2RML_PARENT_TRIPLESMAP = 'rr:parentTriplesMap' @@ -92,6 +93,7 @@ R2RML_TABLE_NAME = 'rr:tableName' R2RML_COLUMN = 'rr:column' + ############################################################################## ############################# D2RQ CONSTANTS ########################### ############################################################################## @@ -154,6 +156,7 @@ YARRRML_IRI = '~iri' YARRRML_LANG = '~lang' +YARRRML_BLANK = 'blank' YARRRML_QUOTED = 'quoted' YARRRML_NON_ASSERTED = 'quotedNonAsserted' diff --git a/src/yatter/predicateobject.py b/src/yatter/predicateobject.py index 4972923..d8fa5e8 100644 --- a/src/yatter/predicateobject.py +++ b/src/yatter/predicateobject.py @@ -156,11 +156,15 @@ def add_predicate_object(data, mapping, predicate_object, mapping_format=RML_URI for om in object_list: iri = False + blank = False if type(om) == list: object_value = om[0] if YARRRML_IRI in om[0]: - object_value = om[0].split(YARRRML_IRI)[0] + object_value = om[0].split("~")[0] iri = True + if YARRRML_BLANK in om[0]: + object_value = om[0].split("~")[0] + blank = True if mapping_format == STAR_URI: template += generate_rml_termmap(STAR_OBJECT, R2RML_OBJECT_CLASS, object_value, "\t\t\t", mapping_format) @@ -195,6 +199,9 @@ def add_predicate_object(data, mapping, predicate_object, mapping_format=RML_URI if iri: template = template[0:len(template) - 5] + "\t\t\t" + R2RML_TERMTYPE + " " \ + R2RML_IRI + "\n\t\t];\n" + if blank: + template = template[0:len(template) - 5] + "\t\t\t" + R2RML_TERMTYPE + " " \ + + R2RML_BLANK_NODE + "\n\t\t];\n" elif YARRRML_MAPPING in om or YARRRML_NON_ASSERTED in om or YARRRML_QUOTED in om: if YARRRML_MAPPING in om: template += ref_mapping(data, mapping, om, YARRRML_MAPPING, R2RML_PARENT_TRIPLESMAP, mapping_format) @@ -236,6 +243,9 @@ def add_predicate_object(data, mapping, predicate_object, mapping_format=RML_URI elif om.get(YARRRML_TYPE) == "literal": template = template[0:len(template) - 5] + "\t\t\t" + R2RML_TERMTYPE + " " \ + R2RML_LITERAL + "\n\t\t];\n" + elif om.get(YARRRML_TYPE) == YARRRML_BLANK: + template = template[0:len(template) - 5] + "\t\t\t" + R2RML_TERMTYPE + " " \ + + R2RML_BLANK_NODE + "\n\t\t];\n" if YARRRML_TARGETS in om: template = template[0:len(template) - 5] + "\t\t\t" + RML_LOGICAL_TARGET + " <"+ om.get(YARRRML_TARGETS) + ">\n\t\t];\n" diff --git a/src/yatter/subject.py b/src/yatter/subject.py index cb92703..d77691d 100644 --- a/src/yatter/subject.py +++ b/src/yatter/subject.py @@ -54,6 +54,10 @@ def add_subject(data, mapping, mapping_format): if YARRRML_TARGETS in individual_subject: subject_termmap = subject_termmap[0:-3]+"\t"+RML_LOGICAL_TARGET+" <"+individual_subject[YARRRML_TARGETS]+">\n\t];\n" + if YARRRML_TYPE in individual_subject: + if individual_subject.get(YARRRML_TYPE) == YARRRML_BLANK: + subject_termmap = subject_termmap[0:-3] + "\t" + R2RML_TERMTYPE + " " + R2RML_BLANK_NODE +"\n\t];\n" + rml_subjects.append(subject_termmap) if YARRRML_GRAPHS in data.get(YARRRML_MAPPINGS).get(mapping): diff --git a/src/yatter/termmap.py b/src/yatter/termmap.py index a522707..ee2334d 100644 --- a/src/yatter/termmap.py +++ b/src/yatter/termmap.py @@ -18,8 +18,8 @@ def get_termmap_type(text, mapping_format): ## Generates a TermMap (subject, predicate, object) based on the property, class and the text -def generate_rml_termmap(rml_property, rml_class, text, identation, mapping_format=RML_URI): - template = identation[0:-1] + rml_property + " [\n"+identation+"a " + rml_class + ";\n" + identation +def generate_rml_termmap(rml_property, rml_class, text, indentation, mapping_format=RML_URI): + template = indentation[0:-1] + rml_property + " [\n"+indentation+"a " + rml_class + ";\n" + indentation term_map = get_termmap_type(text, mapping_format) if term_map == R2RML_TEMPLATE: text = generate_rml_template(text) @@ -32,16 +32,16 @@ def generate_rml_termmap(rml_property, rml_class, text, identation, mapping_form if term_map == STAR_QUOTED: if 'quoted' in text: - template += term_map + " <" + text[YARRRML_QUOTED] + "_0>;\n" + identation[0:-1] + "];\n" + template += term_map + " <" + text[YARRRML_QUOTED] + "_0>;\n" + indentation[0:-1] + "];\n" else: - template += term_map + " <" + text[YARRRML_NON_ASSERTED] + "_0>;\n" + identation[0:-1] + "];\n" + template += term_map + " <" + text[YARRRML_NON_ASSERTED] + "_0>;\n" + indentation[0:-1] + "];\n" elif term_map != "rr:constant": - template += term_map + " \"" + text + "\";\n"+identation[0:-1]+"];\n" + template += term_map + " \"" + text + "\";\n"+indentation[0:-1]+"];\n" else: if text.startswith("http"): - template += term_map + " <" + text + ">;\n" + identation[0:-1] + "];\n" + template += term_map + " <" + text + ">;\n" + indentation[0:-1] + "];\n" else: - template += term_map + " " + text + ";\n"+identation[0:-1]+"];\n" + template += term_map + " " + text + ";\n"+indentation[0:-1]+"];\n" return template

Blank Nodes in YARRRML trigger TypeError: argument of type 'NoneType' is not iterable **Describe the bug** Following the [YARRRML subjects spec](https://rml.io/yarrrml/spec/#subjects), I use blank nodes with `s: null` . This is not supported by yatter **To Reproduce** Steps to reproduce the behavior (and resources): with the files below, run the yatter library `python3 -m yatter -i yarrrml/yatter-test.yml -o rml/yatter-test_rml.ttl` yatter-test.yml ``` # Define namespace prefixes for commonly used ontologies prefixes: exp: "http://example.org/participant/" ex: "http://example.org/#" rdf: "http://www.w3.org/1999/02/22-rdf-syntax-ns#" rdfs: "http://www.w3.org/2000/01/rdf-schema#" xsd: "http://www.w3.org/2001/XMLSchema#" dcterms: "http://purl.org/dc/terms/" locn: "http://www.w3.org/ns/locn#" grel: "http://users.ugent.be/~bjdmeest/function/grel.ttl#" skos: "http://www.w3.org/2004/02/skos/core#" wegwijs: "https://data.vlaanderen.be/id/organisatie/" adms: "http://www.w3.org/ns/adms#" # Define the mapping rules mappings: # Mapping for 'participants' data participants: # Define sources to extract data from JSON sources: - ['yatter-test.csv~csv'] # Define subject mapping s: exp:$(participantUuid) # Define predicate-object mappings po: - [a, ex:Participant] - [dcterms:created, $(created), xsd:dateTime] - [dcterms:modified, $(modified), xsd:dateTime] - p: adms:identifier o: type: blank mapping: identificator condition: function: equal parameters: - [str1, $(participantUuid), s] - [str2, $(participantUuid), o] # Mapping for 'identificator' data identificator: sources: - ['yatter-test.csv~csv'] # Define subject as null, indicating a blank node will be generated s: null # Define predicate-object mappings po: - [a, adms:Identifier] - [skos:notation, exp:$(participantUuid), xsd:anyURI] - [dcterms:creator, wegwijs:OVO026559~iri] ``` yatter-test.csv ``` participantUuid,created,modified,birthYear,placeHolderStatSect 493b78b8-efa7-45b8-af5f-ca08ec90179e,2023-06-06T15:17:32.954+0000,2023-06-06T15:19:14.358+0000,1964,Wilrijk ea9ba26f-e283-42a2-b57a-55ea3f5571ff,2022-12-28T14:46:03.682+0000,2022-12-28T14:48:23.17+0000,1957,Marcinelle e0f5d6b0-984d-4192-b00f-8a07d2189712,2023-06-06T17:36:44.39+0000,2023-06-06T17:36:44.396+0000,1963,Zingem 20a15803-ca0e-42f9-9abe-93dc5d593a31,2022-12-16T11:39:39.481+0000,2022-12-16T11:43:58.073+0000,1951,Gent ``` **Current behavior** TypeError stacktrace ```2023-12-05 10:44:58,430 | INFO: Translating YARRRML mapping to [R2]RML Traceback (most recent call last): File "<frozen runpy>", line 198, in _run_module_as_main File "<frozen runpy>", line 88, in _run_code File "/Library/Frameworks/Python.framework/Versions/3.11/lib/python3.11/site-packages/yatter/__main__.py", line 68, in <module> mapping_content = translate(mapping_data, mapping_format) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "/Library/Frameworks/Python.framework/Versions/3.11/lib/python3.11/site-packages/yatter/__init__.py", line 15, in translate list_initial_sources = get_initial_sources(yarrrml_data) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "/Library/Frameworks/Python.framework/Versions/3.11/lib/python3.11/site-packages/yatter/source.py", line 9, in get_initial_sources if YARRRML_SOURCES in data: ^^^^^^^^^^^^^^^^^^^^^^^ TypeError: argument of type 'NoneType' is not iterable ``` **Expected behavior** an RML file is produced like the one by YARRRML's matey: [output.rml.ttl.txt](https://github.com/oeg-upm/yatter/files/13558196/output.rml.ttl.txt) output turtle [output (1).txt](https://github.com/oeg-upm/yatter/files/13558264/output.1.txt)

Hi @Stiksels, As there is no consensus in the W3C CG about supporting BN without data references, I would not consider implementing this feature for now. However, I can provide support for BN with data reference on the subject. As there is no example in the spec, I will do it similarly as it's indicated for ObjectMaps, as both (OM and SM) are term maps. It will be something like: ```yaml identificator: sources: - ['yatter-test.csv~csv'] # Define subject as null, indicating a blank node will be generated s: value: $(participantUuid) # could be template, reference or constant type: blank # Define predicate-object mappings po: - [a, adms:Identifier] - [skos:notation, exp:$(participantUuid), xsd:anyURI] - [dcterms:creator, wegwijs:OVO026559~iri] ```

2023-12-05T12:18:50

0.0

[]

oeg-upm/yatter

oeg-upm__yatter-45

eee599dcaf6c31abe8af7b5fa641ce9566764fda

diff --git a/README.md b/README.md index 6b3d172..218a8b3 100644 --- a/README.md +++ b/README.md @@ -17,35 +17,62 @@ We are working on including the following features which are not yet implemented pip install yarrrml-translator ``` -## Execution +## Execution from CLI + +### From YARRRML to \[R2\]RML To execute from command line run the following: ```bash -python3 -m yarrrml_translator -i path_to_input_yarrrml.yml -o path_to_output.ttl [-f R2RML] +python3 -m yarrrml_translator -i path_to_input_yarrrml.yml -o path_to_rdf_mapping.ttl [-f R2RML] ``` +`-f R2RML` is an optional parameter for translating input YARRRML into R2RML +### From \[R2\]RML to YARRRML +To execute from command line run the following: +```bash +python3 -m yarrrml_translator -i path_to_input_rdf_mapping.ttl -o path_to_output_yarrrml.yml [-f R2RML] +``` `-f R2RML` is an optional parameter for translating input YARRRML into R2RML +### Merging mappings based on TriplesMap id +To execute from command line run the following (keys of each TriplesMap should be disjoint): +```bash +python3 -m yarrrml_translator -m yarrrml_mapping1.yaml yarrrml_mapping2.yaml [..] -o path_to_output_yarrrml.yml +``` + +## Execution as a library + If you want to include the module in your implementation: -- for translating **RML mappings**: +- for translating **YARRRML mapping** to **RML mappings** (and inverse): ```python import yarrrml_translator import yaml - +# YARRRML to RML rml_content = yarrrml_translator.translate(yaml.safe_load(open("path-to-yarrrml"))) +# RML to YARRRML +yarrrml_content = yarrrml_translator.inverse_translate("rdf_mapping_content") ``` - -- for translating **R2RML mappings**: +- for translating **YARRRML mappings** to **R2RML mappings** (and inverse): ```python import yarrrml_translator import yaml R2RML_URI = 'http://www.w3.org/ns/r2rml#' +# YARRRML to R2RML rml_content = yarrrml_translator.translate(yaml.safe_load(open("path-to-yarrrml")), mapping_format=R2RML_URI) +# R2RML to YARRRML +yarrrml_content = yarrrml_translator.inverse_translate("rdf_mapping_content", mapping_format=R2RML_URI) +``` +- for merging TriplesMap based on id: +```python +import yarrrml_translator +list_yarrrml_mappings = ["content_mapping_yarrrml1", "content_mapping_yarrrml1"] +yarrrml_content = yarrrml_translator.merge_mappings(list_yarrrml_mappings) ``` ## Authors -Ontology Engineering Group - Data Integration: +Ontology Engineering Group: - [David Chaves-Fraga](mailto:[email protected]) +- Marino González García (Final bachelor thesis - Systematic Testing) - Luis López Piñero (Final bachelor thesis - v0.1) diff --git a/VERSION b/VERSION index 39e898a..aec258d 100644 --- a/VERSION +++ b/VERSION @@ -1,1 +1,1 @@ -0.7.1 +0.8 diff --git a/requirements.txt b/requirements.txt index 750f28b..be81d1b 100644 --- a/requirements.txt +++ b/requirements.txt @@ -1,3 +1,3 @@ -pyyaml >= 6.0 +ruamel.yaml >= 0.17.21 rdflib >= 6.2.0 coloredlogs >= 15.0.1 \ No newline at end of file diff --git a/yarrrml-translator/yarrrml_translator/__init__.py b/yarrrml-translator/yarrrml_translator/__init__.py index 5a9a17a..1fc05d6 100644 --- a/yarrrml-translator/yarrrml_translator/__init__.py +++ b/yarrrml-translator/yarrrml_translator/__init__.py @@ -1,26 +1,28 @@ from .constants import * -from .mapping import add_prefix, add_mapping, add_inverse_prefix, get_non_asserted_mappings +from .mapping import add_prefix, add_mapping, add_inverse_prefix, get_non_asserted_mappings, merge_mapping_section_by_key from .source import get_initial_sources, add_source, generate_database_connections, add_table, add_inverse_source from .subject import add_subject, add_inverse_subject from .predicateobject import add_predicate_object_maps, add_inverse_pom import rdflib -import yaml - +import ruamel.yaml as yaml +from ruamel.yaml import YAML +from ruamel.yaml.compat import StringIO +import sys def translate(yarrrml_data, mapping_format=RML_URI): logger.info("Translating YARRRML mapping to [R2]RML") list_initial_sources = get_initial_sources(yarrrml_data) rml_mapping = [add_prefix(yarrrml_data)] - rml_mapping.extend(generate_database_connections(yarrrml_data)) + rml_mapping.extend(generate_database_connections(yarrrml_data, list_initial_sources)) try: - mappings = get_non_asserted_mappings(yarrrml_data, dict.fromkeys(list(yarrrml_data.get(YARRRML_MAPPINGS).keys()))) + mappings, mapping_format = get_non_asserted_mappings(yarrrml_data, dict.fromkeys(list(yarrrml_data.get(YARRRML_MAPPINGS).keys())), mapping_format) for mapping in yarrrml_data.get(YARRRML_MAPPINGS): if mapping_format == R2RML_URI: source_list = add_table(yarrrml_data, mapping, list_initial_sources) else: source_list = add_source(yarrrml_data, mapping, list_initial_sources) - subject_list = add_subject(yarrrml_data, mapping) + subject_list = add_subject(yarrrml_data, mapping, mapping_format) pred = add_predicate_object_maps(yarrrml_data, mapping, mapping_format) it = 0 for source in source_list: @@ -55,22 +57,38 @@ def translate(yarrrml_data, mapping_format=RML_URI): def inverse_translation(rdf_mapping, mapping_format=RML_URI): - yarrrml_mapping = {'prefixes': [], 'mappings': {}} + yarrrml_mapping = {YARRRML_PREFIXES: {}, YARRRML_MAPPINGS: {}} rdf_mapping.bind('rml', rdflib.term.URIRef(RML_URI)) rdf_mapping.bind('rr', rdflib.term.URIRef(R2RML_URI)) rdf_mapping.bind('ql', rdflib.term.URIRef(QL_URI)) - yarrrml_mapping['prefixes'] = add_inverse_prefix(rdf_mapping) + yarrrml_mapping[YARRRML_PREFIXES] = add_inverse_prefix(rdf_mapping) query = f'SELECT ?triplesMap WHERE {{ ?triplesMap {RDF_TYPE} {R2RML_TRIPLES_MAP} . }} ' triples_map = [tm[rdflib.Variable('triplesMap')] for tm in rdf_mapping.query(query).bindings] for tm in triples_map: tm_name = tm.split("/")[-1] - yarrrml_tm = {'sources': [add_inverse_source(tm, rdf_mapping, mapping_format)]} - yarrrml_tm['s'], classes = add_inverse_subject(tm, rdf_mapping) - yarrrml_tm['po'] = add_inverse_pom(tm, rdf_mapping, classes, yarrrml_mapping['prefixes']) - yarrrml_mapping['mappings'][tm_name] = yarrrml_tm + yarrrml_tm = {YARRRML_SOURCE: add_inverse_source(tm, rdf_mapping, mapping_format)} + yarrrml_tm[YARRRML_SHORTCUT_SUBJECTS], classes = add_inverse_subject(tm, rdf_mapping) + yarrrml_tm[YARRRML_SHORTCUT_PREDICATEOBJECT] = add_inverse_pom(tm, rdf_mapping, classes, yarrrml_mapping[YARRRML_PREFIXES]) + yarrrml_mapping[YARRRML_MAPPINGS][tm_name] = yarrrml_tm - string_content = str(yaml.dump(yarrrml_mapping, default_flow_style=None, sort_keys=False)).replace("'\"", - '"').replace( - "\"'", ' " ').replace('\'', '') + string_content = yaml.dump(yarrrml_mapping) return string_content + +def merge_mappings(yarrrrml_list): + combined_mapping = {YARRRML_MAPPINGS:{}} + prefixes_list = [] + for mapping in yarrrrml_list: + prefixes_list.append(mapping[YARRRML_PREFIXES]) + combined_mapping = combined_mapping | merge_mapping_section_by_key(YARRRML_PREFIXES, prefixes_list) + + triples_map_ids = yarrrrml_list[0][YARRRML_MAPPINGS].keys() + for individual_id in triples_map_ids: + mapping_content_list = [] + for mapping in yarrrrml_list: + if individual_id in mapping[YARRRML_MAPPINGS]: + mapping_content_list.append(mapping[YARRRML_MAPPINGS][individual_id]) + combined_mapping[YARRRML_MAPPINGS] = combined_mapping[YARRRML_MAPPINGS] | merge_mapping_section_by_key(individual_id, mapping_content_list) + + string_content = yaml.dump(combined_mapping) + return string_content \ No newline at end of file diff --git a/yarrrml-translator/yarrrml_translator/__main__.py b/yarrrml-translator/yarrrml_translator/__main__.py index 3947bc3..f9c3721 100644 --- a/yarrrml-translator/yarrrml_translator/__main__.py +++ b/yarrrml-translator/yarrrml_translator/__main__.py @@ -1,8 +1,8 @@ import sys -import yaml +from ruamel.yaml import YAML import argparse from rdflib import Graph -from . import translate, inverse_translation +from . import translate, inverse_translation, merge_mappings from .constants import * @@ -18,9 +18,9 @@ def parse_inputs(): if args.input_mapping_path and args.output_mapping_path: if args.input_mapping_path.endswith('.yml') or args.input_mapping_path.endswith('.yaml'): with open(args.input_mapping_path) as f: - input_data = yaml.safe_load(f) - elif args.input_mapping_path.endswith('.ttl') or args.input_mapping_path.endswith( - '.rml') or args.input_mapping_path.endswith('.r2rml'): + yaml = YAML(typ='safe', pure=True) + input_data = yaml.load(f) + elif args.input_mapping_path.endswith('.ttl') or args.input_mapping_path.endswith('.rml') or args.input_mapping_path.endswith('.r2rml'): input_data = Graph() input_data.parse(args.input_mapping_path, format="turtle") else: @@ -30,7 +30,11 @@ def parse_inputs(): if args.format is not None and args.format == "R2RML": input_format = R2RML_URI - + elif args.merge_mapping: + input_data = [] + for mapping in args.merge_mapping: + with open(mapping) as f: + input_data.append(yaml.safe_load(f)) else: sys.tracebacklimit = 0 logger.error("No correct arguments, run python3 -m yarrrml_translator -h to see the help)") @@ -41,8 +45,9 @@ def parse_inputs(): def define_args(): parser = argparse.ArgumentParser() - parser.add_argument("-i", "--input_mapping_path", required=True, help="Input mapping path in YARRRML or RML") - parser.add_argument("-o", "--output_mapping_path", required=True, help="Output path for the generated mapping") + parser.add_argument("-i", "--input_mapping_path", required=False, help="Input mapping path in YARRRML or RML") + parser.add_argument("-o", "--output_mapping_path", required=False, help="Output path for the generated mapping") + parser.add_argument("-m", "--merge_mapping", nargs="*", required=False, help="List of mappings to be merged") parser.add_argument("-f", "--format", required=False, help="Mapping input format. Values: R2RML or RML (RML by default)") return parser @@ -53,6 +58,8 @@ def define_args(): mapping_format, mapping_data = parse_inputs() if type(mapping_data) is Graph: mapping_content = inverse_translation(mapping_data, mapping_format) + elif args.merge_mapping: + mapping_content = merge_mappings(mapping_data) else: mapping_content = translate(mapping_data, mapping_format) write_results(mapping_content) diff --git a/yarrrml-translator/yarrrml_translator/constants.py b/yarrrml-translator/yarrrml_translator/constants.py index cb25adc..3ebd6be 100644 --- a/yarrrml-translator/yarrrml_translator/constants.py +++ b/yarrrml-translator/yarrrml_translator/constants.py @@ -24,13 +24,18 @@ RML_REFERENCE_FORMULATION = 'rml:referenceFormulation' RML_ITERATOR = 'rml:iterator' RML_REFERENCE = 'rml:reference' - +RML_LANGUAGE_MAP = 'rml:languageMap' +RML_LANGUAGE_MAP_CLASS = 'rml:LanguageMap' +RML_DATATYPE_MAP = 'rml:datatypeMap' +RML_DATATYPE_MAP_CLASS = 'rml:DatatypeMap' +RML_QUERY = 'rml:query' STAR_CLASS = 'rml:StarMap' STAR_NON_ASSERTED_CLASS = 'rml:NonAssertedTriplesMap' STAR_QUOTED = 'rml:quotedTriplesMap' STAR_SUBJECT = 'rml:subjectMap' STAR_OBJECT = 'rml:objectMap' +STAR_URI = 'https://w3id.org/kg-construct/rml-star' R2RML_TEMPLATE = 'rr:template' @@ -55,6 +60,7 @@ R2RML_TERMTYPE = 'rr:termType' R2RML_LANGUAGE = 'rr:language' R2RML_IRI = 'rr:IRI' +R2RML_LITERAL = 'rr:Literal' R2RML_REFOBJECT_CLASS = 'rr:RefObjectMap' R2RML_PARENT_TRIPLESMAP = 'rr:parentTriplesMap' R2RML_JOIN_CONITION = 'rr:joinCondition' diff --git a/yarrrml-translator/yarrrml_translator/mapping.py b/yarrrml-translator/yarrrml_translator/mapping.py index 2ed12db..43dfc8e 100644 --- a/yarrrml-translator/yarrrml_translator/mapping.py +++ b/yarrrml-translator/yarrrml_translator/mapping.py @@ -2,7 +2,7 @@ def add_mapping(mapping, mappings, it): - map_template = "<#" + mapping + "_" + str(it) + "> a " + map_template = "<" + mapping + "_" + str(it) + "> a " if mappings[mapping]: if mappings[mapping] == "non_asserted": map_template += STAR_NON_ASSERTED_CLASS + ", " @@ -71,15 +71,30 @@ def add_inverse_prefix(rdf_mapping): return prefixes -def get_non_asserted_mappings(yarrrml_data, mappings): +def get_non_asserted_mappings(yarrrml_data, mappings, mapping_format): for mapping in yarrrml_data.get(YARRRML_MAPPINGS): keys = yarrrml_data.get(YARRRML_MAPPINGS).get(mapping).keys() for key in keys: - if type(yarrrml_data.get(YARRRML_MAPPINGS).get(mapping).get(key)) is list: - values_list = yarrrml_data.get(YARRRML_MAPPINGS).get(mapping).get(key) - for value in values_list: - if YARRRML_NON_ASSERTED in value: - mappings[value[YARRRML_NON_ASSERTED]] = "non_asserted" - elif 'o' in value and YARRRML_NON_ASSERTED in value['o'][0]: - mappings[value['o'][0][YARRRML_NON_ASSERTED]] = "non_asserted" - return mappings + if type(yarrrml_data.get(YARRRML_MAPPINGS).get(mapping).get(key)) is dict: + values = yarrrml_data.get(YARRRML_MAPPINGS).get(mapping).get(key) + if YARRRML_NON_ASSERTED in values: + mappings[values[YARRRML_NON_ASSERTED]] = "non_asserted" + mapping_format = STAR_URI + elif YARRRML_QUOTED in values: + mapping_format = STAR_URI + if type(yarrrml_data.get(YARRRML_MAPPINGS).get(mapping).get(key)) is list and key==YARRRML_SHORTCUT_PREDICATEOBJECT: + for value in yarrrml_data.get(YARRRML_MAPPINGS).get(mapping).get(key): + if type(value) is dict and YARRRML_SHORTCUT_OBJECT in value: + if YARRRML_NON_ASSERTED in value[YARRRML_SHORTCUT_OBJECT]: + mappings[value[YARRRML_SHORTCUT_OBJECT][YARRRML_NON_ASSERTED]] = "non_asserted" + mapping_format = STAR_URI + elif YARRRML_QUOTED in value[YARRRML_SHORTCUT_OBJECT]: + mapping_format = STAR_URI + + return mappings, mapping_format + +def merge_mapping_section_by_key(key,yarrrml_list): + output = {key:{}} + for yarrrml_mapping in yarrrml_list: + output[key] = output[key] | yarrrml_mapping + return output \ No newline at end of file diff --git a/yarrrml-translator/yarrrml_translator/predicateobject.py b/yarrrml-translator/yarrrml_translator/predicateobject.py index 7c3a94d..03735bd 100644 --- a/yarrrml-translator/yarrrml_translator/predicateobject.py +++ b/yarrrml-translator/yarrrml_translator/predicateobject.py @@ -75,7 +75,17 @@ def get_object_list(predicate_object, object_access): for i in range(len(object_maps)): object_list.append([object_maps[i], predicate_object[2]]) else: - object_list.extend(object_maps) + for object in object_maps: + if YARRRML_LANGUAGE in object: + object_list.append([object[YARRRML_VALUE],object[YARRRML_LANGUAGE]+"~lang"]) + elif YARRRML_DATATYPE in object: + object_list.append([object[YARRRML_VALUE], object[YARRRML_DATATYPE]]) + elif YARRRML_TYPE in object: + object_list.append([object[YARRRML_VALUE]+"~"+object[YARRRML_TYPE]]) + elif YARRRML_VALUE in object: + object_list.append([object[YARRRML_VALUE]]) + else: + object_list.append([object]) else: if object_access is None and len(predicate_object) == 3: object_list.append([object_maps, predicate_object[2]]) @@ -124,16 +134,28 @@ def add_predicate_object(data, mapping, predicate_object, mapping_format=RML_URI if YARRRML_IRI in om[0]: object_value = om[0].split(YARRRML_IRI)[0] iri = True - template += generate_rml_termmap(R2RML_OBJECT, R2RML_OBJECT_CLASS, + if mapping_format == STAR_URI: + template += generate_rml_termmap(STAR_OBJECT, R2RML_OBJECT_CLASS, + object_value, "\t\t\t", mapping_format) + else: + template += generate_rml_termmap(R2RML_OBJECT, R2RML_OBJECT_CLASS, object_value, "\t\t\t", mapping_format) if len(om) == 2: types = check_type(om[1]) if types != "error": if types == YARRRML_LANGUAGE: - template = template[0:len(template) - 5] + "\t\t\t" + R2RML_LANGUAGE + " \"" \ + if "$(" in om[1]: + template = template[0:len(template) - 5] + generate_rml_termmap(RML_LANGUAGE_MAP, RML_LANGUAGE_MAP_CLASS, + om[1].replace("~lang",""), "\t\t\t\t", mapping_format) + "\n\t\t];\n" + else: + template = template[0:len(template) - 5] + "\t\t\t" + R2RML_LANGUAGE + " \"" \ + om[1].replace(YARRRML_LANG, "") + "\"\n\t\t];\n" elif types == YARRRML_DATATYPE: - template = template[0:len(template) - 5] + "\t\t\t" + R2RML_DATATYPE + " " \ + if "$(" in om[1]: + template = template[0:len(template) - 5] + generate_rml_termmap(RML_DATATYPE_MAP, RML_DATATYPE_MAP_CLASS, + om[1], "\t\t\t\t", mapping_format) + "\n\t\t];\n" + else: + template = template[0:len(template) - 5] + "\t\t\t" + R2RML_DATATYPE + " " \ + om[1] + "\n\t\t];\n" if iri: template = template[0:len(template) - 5] + "\t\t\t" + R2RML_TERMTYPE + " " \ @@ -142,9 +164,13 @@ def add_predicate_object(data, mapping, predicate_object, mapping_format=RML_URI if YARRRML_MAPPING in om: template += ref_mapping(data, mapping, om, YARRRML_MAPPING, R2RML_PARENT_TRIPLESMAP, mapping_format) elif YARRRML_NON_ASSERTED in om: - template += ref_mapping(data, mapping, om, YARRRML_NON_ASSERTED, STAR_QUOTED, mapping_format) + if YARRRML_CONDITION in om: + template += ref_mapping(data, mapping, om, YARRRML_NON_ASSERTED, STAR_QUOTED, mapping_format) + else: + template += generate_rml_termmap(STAR_OBJECT, STAR_CLASS,om, "\t\t\t", mapping_format) else: - template += ref_mapping(data, mapping, om, YARRRML_QUOTED, STAR_QUOTED, mapping_format) + template += ref_mapping(data, mapping, om, YARRRML_QUOTED, STAR_QUOTED, mapping_format) + # This could be removed else: if YARRRML_VALUE in om: object_value = om.get(YARRRML_VALUE) @@ -153,18 +179,28 @@ def add_predicate_object(data, mapping, predicate_object, mapping_format=RML_URI if YARRRML_IRI in om: object_value = om.split(YARRRML_IRI)[0] iri = True - template += generate_rml_termmap(R2RML_OBJECT, R2RML_OBJECT_CLASS, - object_value, "\t\t\t", mapping_format) - if type(om) == list and YARRRML_DATATYPE in om: + if mapping_format == STAR_URI: + template += generate_rml_termmap(STAR_OBJECT, R2RML_OBJECT_CLASS, + object_value, "\t\t\t", mapping_format) + else: + template += generate_rml_termmap(R2RML_OBJECT, R2RML_OBJECT_CLASS, + object_value, "\t\t\t", mapping_format) + if YARRRML_DATATYPE in om: template = template[0:len(template) - 5] + "\t\t\t" + R2RML_DATATYPE + " " \ + om.get(YARRRML_DATATYPE) + "\n\t\t];\n" - if (type(om) == list and YARRRML_TYPE in om and om.get(YARRRML_TYPE) == "iri") or iri: - template = template[0:len(template) - 5] + "\t\t\t" + R2RML_TERMTYPE + " " \ - + R2RML_IRI + "\n\t\t];\n" - if type(om) == list and YARRRML_LANGUAGE in om: + if YARRRML_LANGUAGE in om: template = template[0:len(template) - 5] + "\t\t\t" + R2RML_LANGUAGE + " \"" \ + om.get(YARRRML_LANGUAGE) + "\"\n\t\t];\n" + if YARRRML_TYPE in om: + if om.get(YARRRML_TYPE) == "iri": + iri = True + elif om.get(YARRRML_TYPE) == "literal": + template = template[0:len(template) - 5] + "\t\t\t" + R2RML_TERMTYPE + " " \ + + R2RML_LITERAL + "\n\t\t];\n" + if iri: + template = template[0:len(template) - 5] + "\t\t\t" + R2RML_TERMTYPE + " " \ + + R2RML_IRI + "\n\t\t];\n" for graph in graph_list: template += generate_rml_termmap(R2RML_GRAPH, R2RML_GRAPH_CLASS, graph, "\t\t\t") @@ -180,7 +216,7 @@ def ref_mapping(data, mapping, om, yarrrml_key, ref_type_property, mapping_forma mapping_join = om.get(yarrrml_key) if mapping_join in list_mappings: - subject_list = add_subject(data, mapping_join) + subject_list = add_subject(data, mapping_join, mapping_format) list_initial_sources = get_initial_sources(data) if mapping_format == R2RML_URI: source_list = add_table(data, mapping_join, list_initial_sources) @@ -191,7 +227,7 @@ def ref_mapping(data, mapping, om, yarrrml_key, ref_type_property, mapping_forma for i in range(number_joins_rml): template += "\t\t" + R2RML_OBJECT + \ " [ \n\t\t\ta " + R2RML_REFOBJECT_CLASS + \ - ";\n\t\t\t" + ref_type_property + " <#" + mapping_join + "_" + str(i) + ">;\n" + ";\n\t\t\t" + ref_type_property + " <" + mapping_join + "_" + str(i) + ">;\n" if YARRRML_CONDITION in om: conditions = om.get(YARRRML_CONDITION) if type(conditions) is not list: @@ -225,8 +261,8 @@ def add_inverse_pom(tm, rdf_mapping, classes, prefixes): for c in classes: yarrrml_poms.append(['rdf:type', c]) - query = f'SELECT ?predicate ?predicateValue ?object ?objectValue ?termtype ?datatype ' \ - f'?parentTriplesMap ?child ?parent' \ + query = f'SELECT ?predicate ?predicateValue ?object ?objectValue ?termtype ?datatype ?datatypeMapValue ' \ + f'?language ?languageMapValue ?parentTriplesMap ?child ?parent' \ f' WHERE {{ ' \ f'<{tm}> {R2RML_PREDICATE_OBJECT_MAP} ?predicateObjectMap . ' \ f'?predicateObjectMap {R2RML_PREDICATE}|{R2RML_SHORTCUT_PREDICATE} ?predicate .' \ @@ -235,7 +271,14 @@ def add_inverse_pom(tm, rdf_mapping, classes, prefixes): f'OPTIONAL {{ ' \ f'?object {R2RML_TEMPLATE}|{R2RML_COLUMN}|{R2RML_CONSTANT}|{RML_REFERENCE} ?objectValue .' \ f'OPTIONAL {{ ?object {R2RML_TERMTYPE} ?termtype . }}' \ - f'OPTIONAL {{ ?object {R2RML_DATATYPE} ?datatype .}} }} ' \ + f'OPTIONAL {{ ?object {R2RML_DATATYPE} ?datatype .}} ' \ + f'OPTIONAL {{ ' \ + f' ?object {RML_DATATYPE_MAP} ?datatypeMap .' \ + f' ?datatypeMap {R2RML_TEMPLATE}|{R2RML_CONSTANT}|{RML_REFERENCE} ?datatypeMapValue .}} ' \ + f'OPTIONAL {{ ?object {R2RML_LANGUAGE} ?language .}} ' \ + f'OPTIONAL {{ ' \ + f' ?object {RML_LANGUAGE_MAP} ?languageMap .' \ + f' ?languageMap {R2RML_TEMPLATE}|{R2RML_CONSTANT}|{RML_REFERENCE} ?languageMapValue .}} }} ' \ f'OPTIONAL {{ ?object {R2RML_PARENT_TRIPLESMAP} ?parentTriplesMap .' \ f'?object {R2RML_JOIN_CONITION} ?join_condition .' \ f'?join_condition {R2RML_CHILD} ?child .' \ @@ -262,13 +305,19 @@ def add_inverse_pom(tm, rdf_mapping, classes, prefixes): else: yarrrml_pom.append(predicate) if tm['objectValue']: # we have extended objectMap version - object = tm['objectValue'].replace('{', "$(").replace('}', ')') + object = tm['objectValue'].replace('{', '$(').replace('}', ')') if tm['termtype']: if tm['termtype'] == rdflib.URIRef(R2RML_IRI): object = object + '~iri' yarrrml_pom.append(object) if tm['datatype']: yarrrml_pom.append(tm['datatype']) + elif tm['datatypeMapValue']: + yarrrml_pom.append(tm['datatypeMapValue'].replace('{', '$(').replace('}', ')')) + if tm['language']: + yarrrml_pom.append(tm['language']+"~lang") + elif tm['languageMapValue']: + yarrrml_pom.append(tm['languageMapValue'].replace('{', '$(').replace('}', ')')+"~lang") elif tm['object'] and not tm['parentTriplesMap']: # we have object shortcut yarrrml_pom.append(tm['object'].replace('{', "$(").replace('}', ')')) diff --git a/yarrrml-translator/yarrrml_translator/source.py b/yarrrml-translator/yarrrml_translator/source.py index 6d55204..a39095b 100644 --- a/yarrrml-translator/yarrrml_translator/source.py +++ b/yarrrml-translator/yarrrml_translator/source.py @@ -31,13 +31,14 @@ def add_source(data, mapping, list_initial_sources): final_list = [] sources = get_sources(data, mapping) for source in sources: - + db_identifier = mapping if source in list_initial_sources: + db_identifier = source source = data.get(YARRRML_SOURCES).get(source) if YARRRML_ACCESS in source: if YARRRML_QUERY in source: - final_list.append(source_template + database_source(mapping, source)) + final_list.append(source_template + database_source(mapping, source, db_identifier)) else: final_list.append(source_template + add_source_full(mapping, source)) elif type(source) is list: @@ -55,11 +56,13 @@ def add_table(data, mapping, list_initial_sources): sources = get_sources(data, mapping) for source in sources: sql_version = False + db_identifier = mapping if source in list_initial_sources: + db_identifier = source source = data.get(YARRRML_SOURCES).get(source) if YARRRML_ACCESS in source and YARRRML_QUERY in source: - r2rml_access = database_source(mapping, source) + r2rml_access = database_source(mapping, source, db_identifier) sql_version = True elif YARRRML_QUERY in source: r2rml_access = R2RML_SQL_QUERY + " \"" + source.get(YARRRML_QUERY).replace("\n", " ").replace("\"", @@ -112,8 +115,7 @@ def add_source_full(mapping, source): reference_formulation = str(source.get(YARRRML_REFERENCE_FORMULATION)) format_from_reference = switch_in_reference_formulation(reference_formulation.lower()) ref_formulation_rml = reference_formulation.replace("jsonpath", "JSONPath").replace("csv", "CSV").replace( - "xpath", - "XPath") + "xpath","XPath") if extension != format_from_reference or format_from_reference == "ERROR": raise Exception("ERROR: not referenceFormulation found or mismatch between the format and " "referenceFormulation in source " + access + "in mapping " + mapping) @@ -139,24 +141,20 @@ def add_source_full(mapping, source): return source_rdf -def database_source(mapping, source): +def database_source(mapping, source, db_identifier): source_rdf = "" if YARRRML_ACCESS in source: if YARRRML_CREDENTIALS in source: if YARRRML_TYPE in source: - type = source.get(YARRRML_TYPE) - access = source.get(YARRRML_ACCESS) - username = source.get(YARRRML_CREDENTIALS).get(YARRRML_USERNAME) - password = source.get(YARRRML_CREDENTIALS).get(YARRRML_PASSWORD) - hash_datasource = abs(hash(access + type + username + password)) - source_rdf += "<#DataSource_" + str(hash_datasource) + ">;\n\t\t" + source_rdf += "<DataSource_" + str(db_identifier) + ">;\n\t\t" if YARRRML_QUERY in source: - source_rdf += R2RML_SQL_QUERY + " \"" + source.get(YARRRML_QUERY) + "\"" + source_rdf += RML_QUERY + " \"" + source.get(YARRRML_QUERY).replace("\n", " ").replace("\"", + "\\\"") + "\"" elif YARRRML_TABLE in source: source_rdf += R2RML_TABLE_NAME + " \"" + source.get(YARRRML_TABLE) + "\"" if YARRRML_REFERENCE_FORMULATION in source: source_rdf += ";\n\t\t" + RML_REFERENCE_FORMULATION + " ql:" \ - + switch_in_reference_formulation(source.get(YARRRML_REFERENCE_FORMULATION)) + + switch_in_reference_formulation(source.get(YARRRML_REFERENCE_FORMULATION)).upper() if YARRRML_QUERY_FORMULATION in source: source_rdf += ";\n\t\t" + R2RML_SQL_VERSION + " rr:" + source.get(YARRRML_QUERY_FORMULATION).upper() else: @@ -187,35 +185,43 @@ def switch_in_reference_formulation(value): return switcher -def generate_database_connections(data): +def generate_database_connections(data, list_initial_sources): database = [] - hash_ids = [] + sources_ids = set() for mapping in data.get(YARRRML_MAPPINGS): sources = get_sources(data, mapping) for source in sources: - if YARRRML_QUERY in source and YARRRML_ACCESS in source: - type = source.get(YARRRML_TYPE) - if type == "mysql": - driver = "com.mysql.jdbc.Driver" - elif type == "postgresql": - driver = "org.postgresql.Driver" - elif type == "sqlserver": - driver = "com.microsoft.sqlserver.jdbc.SQLServerDriver" - else: - driver = None - access = source.get(YARRRML_ACCESS) - username = source.get(YARRRML_CREDENTIALS).get(YARRRML_USERNAME) - password = source.get(YARRRML_CREDENTIALS).get(YARRRML_PASSWORD) - hash_datasource = abs(hash(access + type + username + password)) - if not hash_datasource in hash_ids: - hash_ids.append(hash_datasource) + db_identifier = mapping + external = False + number_external_sources = len(sources_ids) + if source in list_initial_sources: + external = True + db_identifier = source + sources_ids.add(source) + source = data.get(YARRRML_SOURCES).get(source) + if (external and len(sources_ids) > number_external_sources) or not external: + if YARRRML_QUERY in source and YARRRML_ACCESS in source: + db_type = source.get(YARRRML_TYPE) + if db_type == "mysql": + driver = "com.mysql.jdbc.Driver" + elif db_type == "postgresql": + driver = "org.postgresql.Driver" + elif db_type == "sqlserver": + driver = "com.microsoft.sqlserver.jdbc.SQLServerDriver" + else: + driver = None + access = source.get(YARRRML_ACCESS) + username = source.get(YARRRML_CREDENTIALS).get(YARRRML_USERNAME) + password = source.get(YARRRML_CREDENTIALS).get(YARRRML_PASSWORD) + + if driver is None: - database.append("<#DataSource_" + str(hash_datasource) + "> a " + D2RQ_DATABASE_CLASS + ";\n\t" + database.append("<DataSource_" + str(db_identifier) + "> a " + D2RQ_DATABASE_CLASS + ";\n\t" + D2RQ_DSN + " \"" + access + "\";\n\t" + D2RQ_USER + " \"" + username + "\";\n\t" + D2RQ_PASS + " \"" + password + "\".\n\n") else: - database.append("<#DataSource_" + str(hash_datasource) + "> a " + D2RQ_DATABASE_CLASS + ";\n\t" + database.append("<DataSource_" + str(db_identifier) + "> a " + D2RQ_DATABASE_CLASS + ";\n\t" + D2RQ_DSN + " \"" + access + "\";\n\t" + D2RQ_DRIVER + " \"" + driver + "\";\n\t" + D2RQ_USER + " \"" + username + "\";\n\t" diff --git a/yarrrml-translator/yarrrml_translator/subject.py b/yarrrml-translator/yarrrml_translator/subject.py index dda6430..4be5f33 100644 --- a/yarrrml-translator/yarrrml_translator/subject.py +++ b/yarrrml-translator/yarrrml_translator/subject.py @@ -4,7 +4,7 @@ ## Adds a subject or set of subjects to the given TriplesMap -def add_subject(data, mapping): +def add_subject(data, mapping, mapping_format): rml_subjects = [] only_one = False @@ -26,9 +26,19 @@ def add_subject(data, mapping): for individual_subject in subject: if YARRRML_QUOTED in individual_subject or YARRRML_NON_ASSERTED in individual_subject: + if YARRRML_CONDITION in individual_subject: + from .predicateobject import ref_mapping + if YARRRML_NON_ASSERTED in individual_subject: + subject_termmap = ref_mapping(data, mapping, individual_subject, YARRRML_NON_ASSERTED, STAR_QUOTED, mapping_format) + else: + subject_termmap = ref_mapping(data, mapping, individual_subject, YARRRML_QUOTED, STAR_QUOTED, mapping_format) + subject_termmap=subject_termmap.replace(R2RML_REFOBJECT_CLASS,STAR_CLASS).replace(R2RML_OBJECT,STAR_SUBJECT).replace("\t\t\t\t","\t\t\t\t\t").replace("\t\t","\t") + else: + subject_termmap = generate_rml_termmap(STAR_SUBJECT, STAR_CLASS, individual_subject, "\t\t") + elif mapping_format == STAR_URI: subject_termmap = generate_rml_termmap(STAR_SUBJECT, R2RML_SUBJECT_CLASS, individual_subject, "\t\t") else: - subject_termmap = generate_rml_termmap(R2RML_SUBJECT, R2RML_SUBJECT_CLASS, individual_subject, "\t\t") + subject_termmap = generate_rml_termmap(R2RML_SUBJECT, R2RML_SUBJECT_CLASS, individual_subject, "\t\t", mapping_format) rml_subjects.append(subject_termmap) if YARRRML_GRAPHS in data.get(YARRRML_MAPPINGS).get(mapping): diff --git a/yarrrml-translator/yarrrml_translator/termmap.py b/yarrrml-translator/yarrrml_translator/termmap.py index 6406cbf..6d4ab76 100644 --- a/yarrrml-translator/yarrrml_translator/termmap.py +++ b/yarrrml-translator/yarrrml_translator/termmap.py @@ -30,9 +30,9 @@ def generate_rml_termmap(rml_property, rml_class, text, identation, mapping_form if term_map == STAR_QUOTED: if 'quoted' in text: - template += term_map + " <#" + text[YARRRML_QUOTED] + "_0>;\n" + identation[0:-1] + "];\n" + template += term_map + " <" + text[YARRRML_QUOTED] + "_0>;\n" + identation[0:-1] + "];\n" else: - template += term_map + " <#" + text[YARRRML_NON_ASSERTED] + "_0>;\n" + identation[0:-1] + "];\n" + template += term_map + " <" + text[YARRRML_NON_ASSERTED] + "_0>;\n" + identation[0:-1] + "];\n" elif term_map != "rr:constant": template += term_map + " \"" + text + "\";\n"+identation[0:-1]+"];\n" else:

Mapping-pattern merge for PPDS Given a set of non-compliant mapping rules in YARRRML (e.g. TMs without subject), merging them based on the mapping id and validate they are correct tfg test case development this pull request intends to merge the test cases that have been developed in my repository to the translator's main repository

2023-01-12T16:45:12

0.0

[]

oeg-upm/yatter

oeg-upm__yatter-38

7b3d0d20afd322ce82de37afe7e784459352249d

diff --git a/.github/workflows/python-publish-dev.yml b/.github/workflows/python-publish-dev.yml deleted file mode 100644 index 330a33d..0000000 --- a/.github/workflows/python-publish-dev.yml +++ /dev/null @@ -1,32 +0,0 @@ -# This workflows will upload a Python Package using Twine when a release is created -# For more information see: https://help.github.com/en/actions/language-and-framework-guides/using-python-with-github-actions#publishing-to-package-registries - -name: Upload Python Package (development version) - -on: - push: - branch: master - -jobs: - deploy: - - runs-on: ubuntu-latest - - steps: - - uses: actions/checkout@v2 - - name: Set up Python - uses: actions/setup-python@v2 - with: - python-version: '3.x' - - name: Install dependencies - run: | - python -m pip install --upgrade pip - pip install setuptools wheel twine - - name: Build and publish - env: - TWINE_USERNAME: ${{ secrets.PYPI_USERNAME }} - TWINE_PASSWORD: ${{ secrets.PYPI_PASSWORD }} - run: | - cd pretty_yarrrml2rml - python setup.py -k dev sdist bdist_wheel - twine upload dist/* diff --git a/.github/workflows/python-publish-rel.yml b/.github/workflows/python-publish-rel.yml index 99d2aa2..574ace4 100644 --- a/.github/workflows/python-publish-rel.yml +++ b/.github/workflows/python-publish-rel.yml @@ -1,7 +1,7 @@ # This workflows will upload a Python Package using Twine when a release is created # For more information see: https://help.github.com/en/actions/language-and-framework-guides/using-python-with-github-actions#publishing-to-package-registries -name: Upload Python Package (release) +name: Upload Python Package on: release: @@ -27,6 +27,6 @@ jobs: TWINE_USERNAME: ${{ secrets.PYPI_USERNAME }} TWINE_PASSWORD: ${{ secrets.PYPI_PASSWORD }} run: | - cd pretty_yarrrml2rml + cd yarrrml-translator python setup.py -k rel sdist bdist_wheel twine upload dist/* diff --git a/.gitignore b/.gitignore index 0aa167d..994284d 100644 --- a/.gitignore +++ b/.gitignore @@ -1,4 +1,5 @@ .DS_Store lib/__pychache__ .idea -venv \ No newline at end of file +venv +.pytest_cache \ No newline at end of file diff --git a/README.md b/README.md index 171ce80..6b3d172 100644 --- a/README.md +++ b/README.md @@ -1,26 +1,46 @@ -# Pretty-yarrrml2rml -The tool translates mapping rules from YARRRML in a pretty turtle-based serialization of RML. The translation is based -on [RML]((https://rml.io/specs/rml/ "RML Spec") and [YARRRML]((https://rml.io/yarrrml/spec/ "YARRRML Spec") specifications. +# yarrrml-translator -## Limitations: -We are working on inlcuding the following features which are not yet implemented: -- Translation to [R2RML](https://www.w3.org/TR/r2rml) -- [Named graphs](https://www.w3.org/TR/r2rml/#named-graphs) from R2RML -- Functions included using the [FnO](https://fno.io/) Ontology +![GitHub](https://img.shields.io/github/license/oeg-upm/yarrrml-translator?style=flat) +[![DOI](https://zenodo.org/badge/DOI/10.5281/zenodo.7024501.svg)](https://doi.org/10.5281/zenodo.7024501) +[![PyPI](https://img.shields.io/pypi/v/yarrrml-translator?style=flat)](https://pypi.org/project/yarrrml-translator) +![GitHub Release Date](https://img.shields.io/github/release-date/oeg-upm/yarrrml-translator) + +The tool translates mapping rules from YARRRML in a turtle-based serialization of RML or R2RML. The translation is based +on the [RML](https://rml.io/specs/rml/) and [YARRRML](https://rml.io/yarrrml/spec/) specifications. + +## Limitations +We are working on including the following features which are not yet implemented: +- Functions using the [FnO](https://fno.io/) Ontology + +## Installation: +``` +pip install yarrrml-translator +``` ## Execution -In order to execute the code run you can download de PyPi module: +To execute from command line run the following: ```bash -python3 -m pip install pretty-yarrrml2rml -python3 -m pretty_yarrrml2rml -i path_to_input_yarrrml.yml -o path_to_output_rml.rml +python3 -m yarrrml_translator -i path_to_input_yarrrml.yml -o path_to_output.ttl [-f R2RML] ``` +`-f R2RML` is an optional parameter for translating input YARRRML into R2RML + If you want to include the module in your implementation: +- for translating **RML mappings**: +```python +import yarrrml_translator +import yaml + +rml_content = yarrrml_translator.translate(yaml.safe_load(open("path-to-yarrrml"))) +``` + +- for translating **R2RML mappings**: ```python -import pretty_yarrrml2rml +import yarrrml_translator import yaml -rml_content = pretty_yarrrml2rml.translate(yaml.safe_load(open("path-to-yarrrml"))) +R2RML_URI = 'http://www.w3.org/ns/r2rml#' +rml_content = yarrrml_translator.translate(yaml.safe_load(open("path-to-yarrrml")), mapping_format=R2RML_URI) ``` ## Authors diff --git a/VERSION b/VERSION index ee6cdce..39e898a 100644 --- a/VERSION +++ b/VERSION @@ -1,1 +1,1 @@ -0.6.1 +0.7.1 diff --git a/requirements.txt b/requirements.txt index 71e1245..5e28d05 100644 --- a/requirements.txt +++ b/requirements.txt @@ -1,2 +1,3 @@ -pyyaml == 6.0 -rdflib == 6.2.0 +pyyaml >= 6.0 +rdflib >= 6.2.0 +coloredlogs >= 15.0.1 diff --git a/pretty_yarrrml2rml/setup.py b/yarrrml-translator/setup.py similarity index 72% rename from pretty_yarrrml2rml/setup.py rename to yarrrml-translator/setup.py index c44bae0..856696b 100644 --- a/pretty_yarrrml2rml/setup.py +++ b/yarrrml-translator/setup.py @@ -16,25 +16,26 @@ with open("../VERSION", "r") as fh: v = fh.read().replace("\n", "") - if parsed.kind == "rel": - vers_taged = v - else: - vers_taged = v+".dev"+v_time + vers_taged = v with open("../requirements.txt") as r: requirements = list(filter(None, r.read().split("\n")[0:])) setuptools.setup( - name="pretty_yarrrml2rml", + name="yarrrml_translator", version=vers_taged, author="David Chaves-Fraga", author_email="[email protected]", license="Apache 2.0", - description="The tool translates mapping rules in RML from YARRRML serialization to RDF turtle in a pretty and interpreatable way for humans. The translation is based on RML and YARRRML specifications.", + description="A translator from YARRRML to RML mappings.", long_description=long_description, long_description_content_type="text/markdown", - url="https://github.com/oeg-upm/pretty-yarrrml2rml", + url="https://github.com/oeg-upm/yarrrml-translator", + project_urls={ + 'Source code': 'https://github.com/oeg-upm/yarrrml-translator', + 'Issue tracker': 'https://github.com/oeg-upm/yarrrml-translator/issues', + }, include_package_data=True, packages=setuptools.find_packages(), classifiers=[ @@ -42,11 +43,11 @@ "License :: OSI Approved :: Apache Software License", "Operating System :: OS Independent", "Environment :: Console", - "Intended Audience :: Science/Research", + 'Intended Audience :: Information Technology', "Topic :: Utilities", "Topic :: Software Development :: Pre-processors", "Topic :: Scientific/Engineering :: Interface Engine/Protocol Translator" ], install_requires=requirements, - python_requires='>=3.6', + python_requires='>=3.7', ) diff --git a/pretty_yarrrml2rml/pretty_yarrrml2rml/__init__.py b/yarrrml-translator/yarrrml_translator/__init__.py similarity index 54% rename from pretty_yarrrml2rml/pretty_yarrrml2rml/__init__.py rename to yarrrml-translator/yarrrml_translator/__init__.py index b89db22..b8a2452 100644 --- a/pretty_yarrrml2rml/pretty_yarrrml2rml/__init__.py +++ b/yarrrml-translator/yarrrml_translator/__init__.py @@ -1,22 +1,30 @@ +import logging, coloredlogs from .constants import * from .mapping import add_prefix, add_mapping -from .source import get_initial_sources, add_source, generate_database_connections +from .source import get_initial_sources, add_source, generate_database_connections, add_table from .subject import add_subject from .predicateobject import add_predicate_object_maps -import os from rdflib import Graph -def translate(yarrrml_data): - print("------------------------START TRANSLATING YARRRML TO RML-------------------------------") +def translate(yarrrml_data, mapping_format=RML_URI): + # configure logging with colors + logger = logging.getLogger(__name__) + coloredlogs.install(level='DEBUG', fmt='%(asctime)s,%(msecs)03d | %(levelname)s: %(message)s') + + logger.info("Translating YARRRML mapping to [R2]RML") + list_initial_sources = get_initial_sources(yarrrml_data) rml_mapping = [add_prefix(yarrrml_data)] rml_mapping.extend(generate_database_connections(yarrrml_data)) try: for mapping in yarrrml_data.get(YARRRML_MAPPINGS): - source_list = add_source(yarrrml_data, mapping, list_initial_sources) + if mapping_format == R2RML_URI: + source_list = add_table(yarrrml_data, mapping, list_initial_sources) + else: + source_list = add_source(yarrrml_data, mapping, list_initial_sources) subject_list = add_subject(yarrrml_data, mapping) - pred = add_predicate_object_maps(yarrrml_data, mapping) + pred = add_predicate_object_maps(yarrrml_data, mapping, mapping_format) it = 0 for source in source_list: for subject in subject_list: @@ -29,28 +37,21 @@ def translate(yarrrml_data): rml_mapping.append(".\n\n\n") it = it + 1 - print("RML content successfully created!\nStarting the validation with RDFLib....") + logger.info("RML content is created!") rml_mapping_string = "".join(rml_mapping) try: - rml_output_file = open("tmp.nt", "w") - rml_output_file.write(rml_mapping_string) - rml_output_file.close() graph = Graph() - graph.parse("tmp.nt", format="turtle") - if os.path.exists("tmp.nt"): - os.remove("tmp.nt") - print("The mapping in RML was successfully validated.") + graph.parse(data=rml_mapping_string, format="turtle") + logger.info("Mapping has been syntactically validated.") except Exception as e: - print("------------------------ERROR-------------------------------") - print("Validating the RDF generated: " + str(e)) - if os.path.exists("tmp.nt"): - os.remove("tmp.nt") + logger.error("ERROR: There is a syntactic error in the generated mapping") + logger.error(str(e)) return None except Exception as e: - print("------------------------ERROR-------------------------------") - print("RML content not generated: " + str(e)) + logger.error("ERROR: The YARRRML mapping has not been translated") + logger.error(str(e)) return None - print("------------------------END TRANSLATION-------------------------------") + logger.info("Translation has finished successfully.") - return rml_mapping_string + return rml_mapping_string \ No newline at end of file diff --git a/pretty_yarrrml2rml/pretty_yarrrml2rml/__main__.py b/yarrrml-translator/yarrrml_translator/__main__.py similarity index 61% rename from pretty_yarrrml2rml/pretty_yarrrml2rml/__main__.py rename to yarrrml-translator/yarrrml_translator/__main__.py index 23ea966..4680033 100644 --- a/pretty_yarrrml2rml/pretty_yarrrml2rml/__main__.py +++ b/yarrrml-translator/yarrrml_translator/__main__.py @@ -1,42 +1,45 @@ -import sys -import yaml -from . import translate - - -def write_results(rml_mapping): - if len(sys.argv) == 1: - rml_output_path = input("Name the path for the output file:") - else: - rml_output_path = sys.argv[4] - - rml_output_file = open(rml_output_path, "w") - rml_output_file.write(rml_mapping) - rml_output_file.close() - - print("Output mapping saved") - - -def run_parsing_system_inputs(): - if len(sys.argv) == 1: - file = input("Introduce the .yml file you want to convert to RML") - with open(file) as f: - yaml_data = yaml.safe_load(f) - - elif len(sys.argv) == 5 and sys.argv[1] == "-i" and sys.argv[3] == "-o": - with open(sys.argv[2]) as f: - yaml_data = yaml.safe_load(f) - else: - sys.tracebacklimit = 0 - raise Exception( - "\n####################################\nERROR: Wrong argument input. You can:" - "\n-Use no arguments\n-Use arguments (in this order): -i yarrrml.yml -o mapping.rml.ttl" - "\n####################################\n") - - return yaml_data - - -yarrrml_data = run_parsing_system_inputs() - -rml_content = translate(yarrrml_data) - -write_results(rml_content) +import sys +import yaml +from . import translate +from .constants import RML_URI, R2RML_URI + + +def write_results(rml_mapping): + if len(sys.argv) == 1: + rml_output_path = input("Name the path for the output file:") + else: + rml_output_path = sys.argv[sys.argv.index("-o") + 1] + + rml_output_file = open(rml_output_path, "w") + rml_output_file.write(rml_mapping) + rml_output_file.close() + + +def run_parsing_system_inputs(): + mapping_format = RML_URI + if len(sys.argv) == 1: + file = input("Introduce the .yml file you want to convert to RML") + with open(file) as f: + yaml_data = yaml.safe_load(f) + + elif (len(sys.argv) == 5 or len(sys.argv) == 7) and "-i" in sys.argv and "-o" in sys.argv: + with open(sys.argv[sys.argv.index("-i") + 1]) as f: + yaml_data = yaml.safe_load(f) + if "-f" in sys.argv: + if sys.argv[sys.argv.index("-f") + 1] == "R2RML": + mapping_format = R2RML_URI + else: + sys.tracebacklimit = 0 + raise Exception( + "\n####################################\nERROR: Wrong argument input. You can:" + "\n-Use no arguments\n-Use arguments (in this order): -i yarrrml.yml -o mapping.rml.ttl" + "\n####################################\n") + + return mapping_format, yaml_data + + +mapping_format, yarrrml_data = run_parsing_system_inputs() + +rml_content = translate(yarrrml_data, mapping_format) + +write_results(rml_content) diff --git a/pretty_yarrrml2rml/pretty_yarrrml2rml/constants.py b/yarrrml-translator/yarrrml_translator/constants.py similarity index 93% rename from pretty_yarrrml2rml/pretty_yarrrml2rml/constants.py rename to yarrrml-translator/yarrrml_translator/constants.py index 5095eb4..db075a6 100644 --- a/pretty_yarrrml2rml/pretty_yarrrml2rml/constants.py +++ b/yarrrml-translator/yarrrml_translator/constants.py @@ -13,6 +13,7 @@ EXAMPLE_URI = 'http://example.com/ns#' RDFS_URI = 'http://www.w3.org/2000/01/rdf-schema#' XSD_URI = 'http://www.w3.org/2001/XMLSchema#' +FOAF_URI ='http://xmlns.com/foaf/0.1/' RDF_TYPE = 'rdf:type' RML_PREFIX = '@prefix' @@ -51,6 +52,10 @@ R2RML_JOIN_CONITION = 'rr:joinCondition' R2RML_CHILD = 'rr:child' R2RML_PARENT = 'rr:parent' +R2RML_LOGICAL_TABLE_CLASS = 'rr:LogicalTable' +R2RML_LOGICAL_TABLE = 'rr:logicalTable' +R2RML_TABLE_NAME = 'rr:tableName' +R2RML_COLUMN = 'rr:column' ############################################################################## ############################# D2RQ CONSTANTS ########################### @@ -70,9 +75,11 @@ YARRRML_MAPPINGS = 'mappings' YARRRML_SOURCES = 'sources' YARRRML_SOURCE = 'source' +YARRRML_TABLE = 'table' YARRRML_ACCESS = 'access' YARRRML_QUERY = 'query' YARRRML_REFERENCE_FORMULATION = 'referenceFormulation' +YARRRML_QUERY_FORMULATION = 'queryFormulation' YARRRML_ITERATOR = 'iterator' YARRRML_CREDENTIALS = 'credentials' YARRRML_TYPE = 'type' diff --git a/pretty_yarrrml2rml/pretty_yarrrml2rml/mapping.py b/yarrrml-translator/yarrrml_translator/mapping.py similarity index 94% rename from pretty_yarrrml2rml/pretty_yarrrml2rml/mapping.py rename to yarrrml-translator/yarrrml_translator/mapping.py index c0c5da3..ab85c3e 100644 --- a/pretty_yarrrml2rml/pretty_yarrrml2rml/mapping.py +++ b/yarrrml-translator/yarrrml_translator/mapping.py @@ -30,6 +30,8 @@ def add_prefix(data): template.append(RML_PREFIX + " ql: <" + QL_URI + ">.\n") if "d2rq" not in common_prefixes: template.append(RML_PREFIX + " d2rq: <" + D2RQ_URI + ">.\n") + if "foaf" not in common_prefixes: + template.append(RML_PREFIX + " foaf: <" + FOAF_URI + ">.\n") if "base" not in common_prefixes: template.append(RML_BASE + " <" + EXAMPLE_URI + ">.\n") template.append("\n\n") diff --git a/pretty_yarrrml2rml/pretty_yarrrml2rml/predicateobject.py b/yarrrml-translator/yarrrml_translator/predicateobject.py similarity index 91% rename from pretty_yarrrml2rml/pretty_yarrrml2rml/predicateobject.py rename to yarrrml-translator/yarrrml_translator/predicateobject.py index 73dfe13..476cde3 100644 --- a/pretty_yarrrml2rml/pretty_yarrrml2rml/predicateobject.py +++ b/yarrrml-translator/yarrrml_translator/predicateobject.py @@ -46,15 +46,15 @@ def get_graph_access(predicate_object_map): return graph_access -def add_predicate_object_maps(data, mapping): +def add_predicate_object_maps(data, mapping, mapping_format): po_template = "" pom_text = "\t" + R2RML_PREDICATE_OBJECT_MAP + " [\n" mapping_data = data.get(YARRRML_MAPPINGS).get(mapping) for predicate_object_map in mapping_data.get(get_predicate_object_access(mapping_data)): if type(predicate_object_map) is list: - po_template += pom_text + add_predicate_object(data, mapping, predicate_object_map) + "\n" + po_template += pom_text + add_predicate_object(data, mapping, predicate_object_map, mapping_format) + "\n" else: - po_template += pom_text + add_predicate_object(data, mapping, predicate_object_map, + po_template += pom_text + add_predicate_object(data, mapping, predicate_object_map, mapping_format, get_predicate_access(predicate_object_map), get_object_access(predicate_object_map), get_graph_access(predicate_object_map)) + "\n" @@ -102,7 +102,7 @@ def get_graph_list(predicate_object,graph_access): else: graphs = [] return graphs -def add_predicate_object(data, mapping, predicate_object, predicate_access=None, object_access=None, graph_access=None): +def add_predicate_object(data, mapping, predicate_object, mapping_format=RML_URI, predicate_access=None, object_access=None, graph_access=None): template = "" predicate_list = get_predicate_list(predicate_object, predicate_access) object_list = get_object_list(predicate_object, object_access) @@ -119,7 +119,7 @@ def add_predicate_object(data, mapping, predicate_object, predicate_access=None, object_value = om[0].split(YARRRML_IRI)[0] iri = True template += generate_rml_termmap(R2RML_OBJECT, R2RML_OBJECT_CLASS, - object_value, "\t\t\t") + object_value, "\t\t\t", mapping_format) if len(om) == 2: types = check_type(om[1]) if types != "error": @@ -133,7 +133,7 @@ def add_predicate_object(data, mapping, predicate_object, predicate_access=None, template = template[0:len(template) - 5] + "\t\t\t" + R2RML_TERMTYPE + " " \ + R2RML_IRI + "\n\t\t];\n" elif YARRRML_MAPPING in om: - template += join_mapping(data, mapping, om) + template += join_mapping(data, mapping, om, mapping_format) else: if YARRRML_VALUE in om: object_value = om.get(YARRRML_VALUE) @@ -143,7 +143,7 @@ def add_predicate_object(data, mapping, predicate_object, predicate_access=None, object_value = om.split(YARRRML_IRI)[0] iri = True template += generate_rml_termmap(R2RML_OBJECT, R2RML_OBJECT_CLASS, - object_value, "\t\t\t") + object_value, "\t\t\t", mapping_format) if type(om) == list and YARRRML_DATATYPE in om: template = template[0:len(template) - 5] + "\t\t\t" + R2RML_DATATYPE + " " \ + om.get(YARRRML_DATATYPE) + "\n\t\t];\n" @@ -161,7 +161,7 @@ def add_predicate_object(data, mapping, predicate_object, predicate_access=None, return template + "\t];" -def join_mapping(data, mapping, om): +def join_mapping(data, mapping, om, mapping_format): list_mappings = [] template = "" for mappings in data.get(YARRRML_MAPPINGS): @@ -172,7 +172,10 @@ def join_mapping(data, mapping, om): if mapping_join in list_mappings: subject_list = add_subject(data, mapping_join) list_initial_sources = get_initial_sources(data) - source_list = add_source(data, mapping_join, list_initial_sources) + if mapping_format == R2RML_URI: + source_list = add_table(data, mapping_join, list_initial_sources) + else: + source_list = add_source(data, mapping_join, list_initial_sources) number_joins_rml = len(subject_list) * len(source_list) for i in range(number_joins_rml): @@ -195,7 +198,7 @@ def join_mapping(data, mapping, om): ";\n\t\t\t\t" + R2RML_PARENT + " " + parent + ";\n\t\t\t]; \n" else: - raise Exception("Error: more than two parameters in join condition in mapping1 " + mapping) + raise Exception("Error: more than two parameters in join condition in mapping " + mapping) template += "\t\t];\n" else: template += "\n\t\t]\n" diff --git a/pretty_yarrrml2rml/pretty_yarrrml2rml/source.py b/yarrrml-translator/yarrrml_translator/source.py similarity index 76% rename from pretty_yarrrml2rml/pretty_yarrrml2rml/source.py rename to yarrrml-translator/yarrrml_translator/source.py index 2157742..d852550 100644 --- a/pretty_yarrrml2rml/pretty_yarrrml2rml/source.py +++ b/yarrrml-translator/yarrrml_translator/source.py @@ -15,11 +15,12 @@ def get_sources(data, mapping): sources = data.get(YARRRML_MAPPINGS).get(mapping).get(YARRRML_SOURCES) elif YARRRML_SOURCE in data.get(YARRRML_MAPPINGS).get(mapping): sources = data.get(YARRRML_MAPPINGS).get(mapping).get(YARRRML_SOURCE) - if type(sources) is not list: - sources = [sources] else: raise Exception("ERROR: sources not defined in mapping " + mapping) + if type(sources) is not list: + sources = [sources] + return sources @@ -45,12 +46,42 @@ def add_source(data, mapping, list_initial_sources): return final_list +def add_table(data, mapping, list_initial_sources): + table_template = "\t" + R2RML_LOGICAL_TABLE + " [\n\t\ta " + R2RML_LOGICAL_TABLE_CLASS + \ + ";\n\t\t" + + final_list = [] + sources = get_sources(data, mapping) + for source in sources: + sql_version = False + if source in list_initial_sources: + source = data.get(YARRRML_SOURCES).get(source) + + if YARRRML_ACCESS in source and YARRRML_QUERY in source: + r2rml_access = database_source(mapping, source) + sql_version = True + elif YARRRML_QUERY in source: + r2rml_access = R2RML_SQL_QUERY + " \"" + source.get(YARRRML_QUERY).replace("\n", " ").replace("\"", "\\\"") + "\"" + elif YARRRML_TABLE in source: + r2rml_access = R2RML_TABLE_NAME + " \"" + source.get(YARRRML_TABLE) + "\"" + else: + raise Exception("ERROR: table or query is not provided in " + source + " of mapping " + mapping) + if not sql_version: + if YARRRML_QUERY_FORMULATION in source: + r2rml_access += ";\n\t\t" + R2RML_SQL_VERSION + " rr:" + source.get(YARRRML_QUERY_FORMULATION).upper() + else: + r2rml_access += ";\n\t\t" + R2RML_SQL_VERSION + " rr:SQL2008" + r2rml_access += "\n\t];\n" + final_list.append(table_template + r2rml_access) + return final_list + + def add_source_simplified(mapping, source): source_rdf = "" file_path = re.sub("~.*", "", source[0]) reference_formulation = source[0].split('~')[1] source_extension = file_path.split('.')[1] - ref_formulation_rml = reference_formulation.replace("json", "JSON").replace("csv", "CSV").replace("xpath", "XPath") + ref_formulation_rml = reference_formulation.replace("jsonpath", "JSONPath").replace("csv", "CSV").replace("xpath", "XPath") if switch_in_reference_formulation(reference_formulation) != source_extension: raise Exception( "ERROR: mismatch extension and referenceFormulation in source " + source + " in mapping " + mapping) @@ -77,7 +108,7 @@ def add_source_full(mapping, source): if YARRRML_REFERENCE_FORMULATION in source: reference_formulation = str(source.get(YARRRML_REFERENCE_FORMULATION)) format_from_reference = switch_in_reference_formulation(reference_formulation.lower()) - ref_formulation_rml = reference_formulation.replace("json", "JSON").replace("csv", "CSV").replace("xpath", + ref_formulation_rml = reference_formulation.replace("jsonpath", "JSONPath").replace("csv", "CSV").replace("xpath", "XPath") if extension != format_from_reference or format_from_reference == "ERROR": raise Exception("ERROR: not referenceFormulation found or mismatch between the format and " @@ -114,15 +145,19 @@ def database_source(mapping, source): username = source.get(YARRRML_CREDENTIALS).get(YARRRML_USERNAME) password = source.get(YARRRML_CREDENTIALS).get(YARRRML_PASSWORD) hash_datasource = abs(hash(access + type + username + password)) - source_rdf += "<#DataSource_" + str(hash_datasource) + ">;\n\t\t" \ - + R2RML_SQL_VERSION + " rr:SQL2008;\n\t\t" \ - + R2RML_SQL_QUERY + " \"" + source.get(YARRRML_QUERY) + "\";\n\t\t" + source_rdf += "<#DataSource_" + str(hash_datasource) + ">;\n\t\t" + if YARRRML_QUERY in source: + source_rdf += R2RML_SQL_QUERY + " \"" + source.get(YARRRML_QUERY) + "\"" + elif YARRRML_TABLE in source: + source_rdf += R2RML_TABLE_NAME + " \"" + source.get(YARRRML_TABLE) + "\"" if YARRRML_REFERENCE_FORMULATION in source: - source_rdf += "\t\t" + RML_REFERENCE_FORMULATION + " ql:" \ - + switch_in_reference_formulation( - source.get(YARRRML_REFERENCE_FORMULATION)) + "\n\t];\n" + source_rdf += ";\n\t\t" + RML_REFERENCE_FORMULATION + " ql:" \ + + switch_in_reference_formulation(source.get(YARRRML_REFERENCE_FORMULATION)) + if YARRRML_QUERY_FORMULATION in source: + source_rdf += ";\n\t\t" + R2RML_SQL_VERSION + " rr:" + source.get(YARRRML_QUERY_FORMULATION).upper() else: - source_rdf += "\n\t];\n" + source_rdf += ";\n\t\t" + R2RML_SQL_VERSION + " rr:SQL2008" + source_rdf += "\n\t];\n" else: raise Exception("ERROR: no credentials to get access to source in mapping " + mapping) else: @@ -154,7 +189,7 @@ def generate_database_connections(data): for mapping in data.get(YARRRML_MAPPINGS): sources = get_sources(data, mapping) for source in sources: - if YARRRML_ACCESS and YARRRML_QUERY in source: + if YARRRML_QUERY in source and YARRRML_ACCESS in source: type = source.get(YARRRML_TYPE) if type == "mysql": driver = "com.mysql.jdbc.Driver" diff --git a/pretty_yarrrml2rml/pretty_yarrrml2rml/subject.py b/yarrrml-translator/yarrrml_translator/subject.py similarity index 100% rename from pretty_yarrrml2rml/pretty_yarrrml2rml/subject.py rename to yarrrml-translator/yarrrml_translator/subject.py diff --git a/pretty_yarrrml2rml/pretty_yarrrml2rml/termmap.py b/yarrrml-translator/yarrrml_translator/termmap.py similarity index 67% rename from pretty_yarrrml2rml/pretty_yarrrml2rml/termmap.py rename to yarrrml-translator/yarrrml_translator/termmap.py index b188276..2f2b1ad 100644 --- a/pretty_yarrrml2rml/pretty_yarrrml2rml/termmap.py +++ b/yarrrml-translator/yarrrml_translator/termmap.py @@ -2,10 +2,13 @@ ## return the type of TermMap based on the input text -def get_termmap_type(text): +def get_termmap_type(text, mapping_format): if "$(" in text and ")" in text: if text[0] == "$" and text[len(text) - 1] == ")" and text.count("$(") == 1: - return RML_REFERENCE + if mapping_format == R2RML_URI: + return R2RML_COLUMN + else: + return RML_REFERENCE else: return R2RML_TEMPLATE else: @@ -13,13 +16,13 @@ def get_termmap_type(text): ## Generates a TermMap (subject, predicate, object) based on the property, class and the text -def generate_rml_termmap(rml_property, rml_class, text, identation): +def generate_rml_termmap(rml_property, rml_class, text, identation, mapping_format=RML_URI): template = identation[0:-1] + rml_property + " [\n"+identation+"a " + rml_class + ";\n" + identation - term_map = get_termmap_type(text) + term_map = get_termmap_type(text, mapping_format) if term_map == R2RML_TEMPLATE: text = text.replace("$(", "{") text = text.replace(")", "}") - elif term_map == RML_REFERENCE: + elif term_map == RML_REFERENCE or term_map == R2RML_COLUMN: text = text.replace("$(", "") text = text.replace(")", "") elif term_map == R2RML_CONSTANT and text == "a": @@ -27,7 +30,10 @@ def generate_rml_termmap(rml_property, rml_class, text, identation): if term_map != "rr:constant": template += term_map + " \"" + text + "\";\n"+identation[0:-1]+"];\n" else: - template += term_map + " " + text + ";\n"+identation[0:-1]+"];\n" + if text.startswith("http"): + template += term_map + " <" + text + ">;\n" + identation[0:-1] + "];\n" + else: + template += term_map + " " + text + ";\n"+identation[0:-1]+"];\n" return template

Fixed a typo in read me Typo in rml:referenceFormulation ql:JSONPath; Hi, When we convert to RML file, after the source at rml:referenceFormulation we have `rml:referenceFormulation ql:JSONpath;` where `p` is small in `ql:JSONpath` Please fix it to `ql"JSONPath` because `P` has to be capital. Thanks

2022-10-27T07:57:07

0.0

[]

oeg-upm/yatter

oeg-upm__yatter-22

074c3153142d14e02247d31cbd44fae5b9d1f01d

diff --git a/.gitignore b/.gitignore new file mode 100644 index 0000000..d63758b --- /dev/null +++ b/.gitignore @@ -0,0 +1,3 @@ +.DS_Store +lib/__pychache__ +.idea \ No newline at end of file diff --git a/lib/__pycache__/mapping.cpython-38.pyc b/lib/__pycache__/mapping.cpython-38.pyc deleted file mode 100644 index 81ec0f4..0000000 Binary files a/lib/__pycache__/mapping.cpython-38.pyc and /dev/null differ diff --git a/lib/__pycache__/predicateobject.cpython-38.pyc b/lib/__pycache__/predicateobject.cpython-38.pyc deleted file mode 100644 index c8fe2f8..0000000 Binary files a/lib/__pycache__/predicateobject.cpython-38.pyc and /dev/null differ diff --git a/lib/__pycache__/source.cpython-38.pyc b/lib/__pycache__/source.cpython-38.pyc deleted file mode 100644 index 2a6d436..0000000 Binary files a/lib/__pycache__/source.cpython-38.pyc and /dev/null differ diff --git a/lib/__pycache__/subject.cpython-38.pyc b/lib/__pycache__/subject.cpython-38.pyc deleted file mode 100644 index 7f0afd7..0000000 Binary files a/lib/__pycache__/subject.cpython-38.pyc and /dev/null differ diff --git a/lib/constants.py b/lib/constants.py new file mode 100644 index 0000000..77d6aa7 --- /dev/null +++ b/lib/constants.py @@ -0,0 +1,105 @@ + + + +############################################################################## +############################# RML CONSTANTS ############################### +############################################################################## + +RML_URI = 'http://semweb.mmlab.be/ns/rml#' +R2RML_URI = 'http://www.w3.org/ns/r2rml#' +RDF_URI = 'http://www.w3.org/1999/02/22-rdf-syntax-ns#' +RDF_TYPE = 'rdf:type' + +RML_PREFIX = '@prefix' +RML_TRIPLES_MAP = 'rr:TriplesMap' + + +RML_LOGICAL_SOURCE = 'rml:LogicalSource' +RML_LOGICAL_SOURCE_PROPERTY = 'rml:logicalSource' +RML_SOURCE = 'rml:source' +RML_REFERENCE_FORMULATION = 'rml:referenceFormulation' +RML_ITERATOR = 'rml:iterator' +RML_REFERENCE = 'rml:reference' + + + + +R2RML_TEMPLATE = 'rr:template' +R2RML_CONSTANT = 'rr:constant' +R2RML_SUBJECT_PROPERTY = 'rr:subjectMap' +R2RML_SUBJECT_CLASS = 'rr:SubjectMap' +R2RML_SQL_VERSION = 'rr:sqlVersion' +R2RML_SQL_QUERY = 'rr:sqlQuery' +R2RML_PREDICATE_OBJECT_MAP = 'rr:predicateObjectMap' +R2RML_PREDICATE_OBJECT_MAP_CLASS = 'rr:PredicateObjectMap' +R2RML_SHORTCUT_PREDICATE = 'rr:predicate' +R2RML_PREDICATE = 'rr:predicateMap' +R2RML_PREDICATE_CLASS = 'rr:PredicateMap' +R2RML_SHORTCUT_OBJECT = 'rr:object' +R2RML_OBJECT = 'rr:objectMap' +R2RML_OBJECT_CLASS = 'rr:ObjectMap' +R2RML_DATATYPE = 'rr:datatype' +R2RML_TERMTYPE = 'rr:termType' +R2RML_LANGUAGE = 'rr:language' +R2RML_IRI = 'rr:IRI' +R2RML_REFOBJECT_CLASS = 'rr:RefObjectMap' +R2RML_PARENT_TRIPLESMAP = 'rr:parentTriplesMap' +R2RML_JOIN_CONITION = 'rr:joinCondition' +R2RML_CHILD = 'rr:child' +R2RML_PARENT = 'rr:parent' + +############################################################################## +############################# D2RQ CONSTANTS ########################### +############################################################################## +D2RQ_DATABASE_CLASS = 'd2rq:Database' +D2RQ_DSN = 'd2rq:jdbcDSN' +D2RQ_DRIVER = 'd2rq:jdbcDriver' +D2RQ_USER = 'd2rq:username' +D2RQ_PASS = 'd2rq:password' + + +############################################################################## +############################# YARRRML CONSTANTS ########################### +############################################################################## + +YARRRML_PREFIXES = 'prefixes' +YARRRML_MAPPINGS = 'mappings' + +YARRRML_SOURCES = 'sources' +YARRRML_SOURCE = 'source' +YARRRML_ACCESS = 'access' +YARRRML_QUERY = 'query' +YARRRML_REFERENCE_FORMULATION = 'referenceFormulation' +YARRRML_ITERATOR = 'iterator' +YARRRML_CREDENTIALS = 'credentials' +YARRRML_TYPE = 'type' +YARRRML_USERNAME = 'username' +YARRRML_PASSWORD = 'password' + +YARRRML_SHORTCUT_SUBJECTS = 's' +YARRRML_SUBJECT = 'subject' +YARRRML_SUBJECTS = 'subjects' + + +YARRRML_PREDICATEOBJECT = 'predicateobjects' +YARRRML_SHORTCUT_PREDICATEOBJECT = 'po' + +YARRRML_PREDICATES = 'predicates' +YARRRML_SHORTCUT_PREDICATES = 'p' +YARRRML_OBJECT = 'objects' +YARRRML_SHORTCUT_OBJECT = 'o' +YARRRML_VALUE = 'value' +YARRRML_MAPPING = 'mapping' +YARRRML_DATATYPE = 'datatype' +YARRRML_LANGUAGE = 'language' + +YARRRML_CONDITION = 'condition' +YARRRML_PARAMETERS = 'parameters' + +YARRRML_IRI = '~iri' +YARRRML_LANG = '~lang' + + + + + diff --git a/lib/main.py b/lib/main.py index 5a54367..7f47606 100644 --- a/lib/main.py +++ b/lib/main.py @@ -1,80 +1,90 @@ import yaml import sys -import os -import mapping as mappingmod -import source as sourcemod -import subject as subjectmod -import predicateobject as predicateobjectmod +from rdflib import Graph +import mapping as mapping_mod +import source as source_mod +import subject as subject_mod +import predicateobject as predicate_object_mod -if __name__ =="__main__": - if len(sys.argv)==1: - print("Introduce the .yml file you want to convert to RML") - file = input() - #file="../test/gtfs-csv.yml" - print("\n") - #../test/example1/rml/mapping.yml - with open(file) as f: - print("Name for your file:") - end=input() - #end="prueba" - if(end[:-4]!=".rml"): - nuevo_fich=open(end+".rml","a") - else: - nuevo_fich=open(end,"a") - data = yaml.safe_load(f) - yaml_stream = True - - elif len(sys.argv)==5 and sys.argv[1]=="-m" and sys.argv[3]=="-o": - file=sys.argv[2] + +def run_parsing_system_inputs(): + if len(sys.argv) == 1: + file = input("Introduce the .yml file you want to convert to RML") with open(file) as f: - end=sys.argv[4] - if(end[:-4]!=".rml"): - nuevo_fich=open(end+".rml","a") - else: - nuevo_fich=open(end,"a") - data = yaml.safe_load(f) - yaml_stream = True + yaml_data = yaml.safe_load(f) + + elif len(sys.argv) == 5 and sys.argv[1] == "-m" and sys.argv[3] == "-o": + with open(sys.argv[2]) as f: + yaml_data = yaml.safe_load(f) + else: + sys.tracebacklimit = 0 + raise Exception( + "\n####################################\nERROR: Wrong argument input. You can:" + "\n-Use no arguments\n-Use arguments (in this order): -m yarrrml.yml -o mapping.rml.ttl" + "\n####################################\n") + return yaml_data + + +def open_inputs(yarrml_path): + with open(yarrml_path) as f: + yaml_data = yaml.safe_load(f) + return yaml_data + + +def write_results(rml_content): + if len(sys.argv) == 1: + rml_output_path = input("Name the path for the output file:") else: - sys.tracebacklimit=0 - raise Exception("\n####################################\nERROR: Wrong argument input. You can:\n-Use no arguments\n-Use arguments: -m YARRRMLfile -o RMLFile\n####################################\n") + rml_output_path = sys.argv[4] + rml_output_file = open(rml_output_path, "a") + rml_output_file.write(rml_content) + rml_output_file.close() - print("------------------------START RML-------------------------------") - mapping_list=[] - list_initial_sources= sourcemod.getInitialSources(data) - final="" - final = mappingmod.addPrefix(data) + print("Validating the generated RDF file with RDFLib") try: - for mapp in data.get("mappings"): - subject_list=[] - subject_list = subjectmod.addSubject(data, mapp) - source_list=[] - source_list=sourcemod.addSource(data,mapp,list_initial_sources) - pred =predicateobjectmod.addPredicateObject(data,mapp) + graph = Graph() + graph.parse(rml_output_path, format="turtle") + except Exception as e: + print("------------------------ERROR-------------------------------") + print("File not created: " + str(e)) + sys.exit() + +def translate(yarrrml_data): + print("------------------------START TRANSLATING YARRRML TO RML-------------------------------") + list_initial_sources = source_mod.get_initial_sources(yarrrml_data) + final = mapping_mod.add_prefix(yarrrml_data) + try: + for map in yarrrml_data.get("mappings"): + subject_list = subject_mod.add_subject(yarrrml_data, map) + source_list = source_mod.add_source(yarrrml_data, map, list_initial_sources) + pred = predicate_object_mod.add_predicate_object_maps(yarrrml_data, map) it = 0 for source in source_list: for subject in subject_list: - map=mappingmod.addMapping(data,mapp,it) + map_aux = mapping_mod.add_mapping(map, it) if type(source) is list: - final += map + source[0] +subject + pred + source[1] + final += map_aux + source[0] + subject + pred + source[1] else: - final += map + source +subject + pred - final = final[:-3] - final+= ".\n\n\n" - it=it+1 - nuevo_fich.write(final) + final += map_aux + source + subject + pred + final = final[:-2] + final += ".\n\n\n" + it = it + 1 + + print("RML content successfully created!") except Exception as e: - print(str(e)) - print("------------------------END RML-------------------------------") - print("FILE NOT SUCCESSFULY CREATED") - if(".rml" in end): - os.remove(end) - else: - os.remove(end+".rml") - sys.exit() - - - print("------------------------END RML-------------------------------") - print("FILE SUCCESSFULY CREATED!") + print("------------------------ERROR-------------------------------") + print("RML content not generated: "+str(e)) + sys.exit() + + return final + + +if __name__ == "__main__": + yarrrml_data = run_parsing_system_inputs() + rml_content = translate(yarrrml_data) + write_results(rml_content) + + print("------------------------END TRANSLATION-------------------------------") diff --git a/lib/mapping.py b/lib/mapping.py index 415fe4f..6b211fd 100644 --- a/lib/mapping.py +++ b/lib/mapping.py @@ -1,21 +1,38 @@ -import yaml -def addMapping(data, mapping,it): - map_template="<#"+mapping+"_"+str(it)+"> a rr:TriplesMap;\n\n" +import constants + + +def add_mapping(mapping, it): + map_template = "<#" + mapping + "_" + str(it) + "> a " + constants.RML_TRIPLES_MAP + ";\n\n" return map_template -def addPrefix(data): - template="" - if "prefixes" in data: - prefixes=data.get("prefixes") + +def add_prefix(data): + template = "" + common_prefixes = [] + if constants.YARRRML_PREFIXES in data: + prefixes = data.get(constants.YARRRML_PREFIXES) for prefix in prefixes: - template+="@prefix "+ prefix +": <"+data.get("prefixes").get(prefix)+">.\n" - if "rr" not in data.get("prefixes"): - template+="@prefix rr: <http://www.w3.org/ns/r2rml#>.\n" - if "rml" not in data.get("prefixes"): - template+="@prefix rml: <http://semweb.mmlab.be/ns/rml#>.\n" - if "rdf" not in data.get("prefixes"): - template+="@prefix rdf: <http://www.w3.org/1999/02/22-rdf-syntax-ns#>.\n" - template+="\n\n" - return template - else: - return template + prefix_uri = data.get(constants.YARRRML_PREFIXES).get(prefix) + check_common_prefixes(prefix_uri, common_prefixes) + template += constants.RML_PREFIX + " " + prefix + ": <" + prefix_uri + ">.\n" + + if "r2rml" not in common_prefixes: + template += constants.RML_PREFIX + " rr: <" + constants.R2RML_URI + ">.\n" + if "rml" not in common_prefixes: + template += constants.RML_PREFIX + " rml: <" + constants.RML_URI + ">.\n" + if "rdf" not in common_prefixes: + template += constants.RML_PREFIX + " rdf: <" + constants.RDF_URI + ">.\n" + + template += "\n\n" + + return template + + +def check_common_prefixes(prefix_uri, common_prefixes): + + if prefix_uri == constants.R2RML_URI: + common_prefixes.append("r2rml") + elif prefix_uri == constants.RML_URI: + common_prefixes.append("rml") + elif prefix_uri == constants.RDF_URI: + common_prefixes.append("rdf") diff --git a/lib/predicateobject.py b/lib/predicateobject.py index 3fc7a6d..6e91142 100644 --- a/lib/predicateobject.py +++ b/lib/predicateobject.py @@ -1,475 +1,180 @@ -import yaml -import source as sourcemod -import subject as subjectmod +import constants +import source +import subject +import termmap -def addPredicateObject(data,mapping): - po_template = "" - if ("predicateobjects" in data.get("mappings").get(mapping)): - for predob in data.get("mappings").get(mapping).get("predicateobjects"): - if "predicates" in predob: - #full con acceso a predicates - if "objects" in predob: - po_template+=addPredicateObjectFull(data,mapping,predob,"predicates","objects") - elif "o" in predob: - po_template+= addPredicateObjectFull(data,mapping,predob,"predicates","o") - - elif "p" in predob: - #full con acceso a p - if "objects" in predob: - po_template+=addPredicateObjectFull(data,mapping,predob,"p","objects") - elif "o" in predob: - po_template+= addPredicateObjectFull(data,mapping,predob,"p","o") - else: - if type(predob) is list: - po_template+=addPredicateObjectSimplified(data,mapping,predob) - else: - #ERROR - raise Exception("Error: Incorrect predicateObject Map in mapping " + mapping) - elif ("po" in data.get("mappings").get(mapping)): - for predob in data.get("mappings").get(mapping).get("po"): - if "predicates" in predob: - #full - if "objects" in predob: - po_template+=addPredicateObjectFull(data,mapping,predob,"predicates","objects") - elif "o" in predob: - po_template+= addPredicateObjectFull(data,mapping,predob,"predicates","o") - elif "p" in predob: - #full con acceso a p - if "objects" in predob: - po_template+=addPredicateObjectFull(data,mapping,predob,"p","objects") - elif "o" in predob: - po_template+= addPredicateObjectFull(data,mapping,predob,"p","o") - else: - if type(predob) is list: - po_template+=addPredicateObjectSimplified(data,mapping,predob) - else: - #ERROR - raise Exception("Error: Incorrect predicateObject Map in mapping " + mapping) - return po_template +def get_object_access(predicate_object_map): + if constants.YARRRML_OBJECT in predicate_object_map: + object_access = constants.YARRRML_OBJECT + elif constants.YARRRML_SHORTCUT_OBJECT in predicate_object_map: + object_access = constants.YARRRML_SHORTCUT_OBJECT + else: + raise Exception("There isn't an object key correctly specify in " + predicate_object_map) + return object_access -def addPredicateObjectFull(data,mapping,predob,access,accessObject): - template="" - #COMPROBAR SI ES LISTA PRIMERO - if(type(predob.get(access))!= list): - template="\trr:predicateObjectMap [\n" + "\t\ta rr:PredicateObjectMap;\n\t\trr:predicate " + predob.get(access) +";\n" - if accessObject in predob: - if "value" in predob.get(accessObject): - template+="\t\trr:objectMap [ \n\t\t\ta rr:ObjectMap;\n" - object=predob.get(accessObject).get("value") - termMap=getTermMap(object) - if(termMap=='rr:template "'): - object=object.replace("$",'') - object=object.replace("(","{") - object=object.replace(")",'}') - else: - object=object.replace("$(",'') - object=object.replace(")",'') - if(termMap!='rr:constant '): - template+="\t\t\t"+ termMap + object + '"'+ ';\n' - else: - template+="\t\t\t"+ termMap + object + ';\n' - #"\t\t]\n\t];\n\n" - if "datatype" in predob.get(accessObject): - template+= "\t\t\trr:datatype " + predob.get(accessObject).get("datatype") + "\n\t\t]\n\t];" - elif "type" in predob.get(accessObject): - template+= "\t\t\trr:TermType " + predob.get(accessObject).get("type") + "\n\t\t]\n\t];" - elif "language" in predob.get(accessObject): - template+= '\t\t\trr:language "' + predob.get(accessObject).get("language") + '"\n\t\t]\n\t];' - elif "mapping" in predob.get(accessObject): - template+= joinMapping(data,mapping,predob.get(accessObject)) - else: - if(type(predob.get(accessObject))!=list): - template+="\t\trr:objectMap [ \n\t\t\ta rr:ObjectMap;\n" - object=predob.get(accessObject) - termMap=getTermMap(object) - if(termMap=='rr:template "'): - object=object.replace("$",'') - object=object.replace("(","{") - object=object.replace(")",'}') - else: - object=object.replace("$(",'') - object=object.replace(")",'') - if(termMap!='rr:constant '): - template+="\t\t\t"+ termMap + object + '"'+ ';\n\t\t]\n\t];\n\n' - else: - template+="\t\t\t"+ termMap + object + ';\n\t\t]\n\t];\n\n' - else: - for objec in predob.get(accessObject): - if "mapping" in objec: - template+= joinMapping(data,mapping,objec) - continue - else: - template+="\t\trr:objectMap [ \n\t\t\ta rr:ObjectMap;\n" - if(type(objec) is list): - object = objec[0] - termMap=getTermMap(object) - if(termMap=='rr:template "'): - object=object.replace("$",'') - object=object.replace("(","{") - object=object.replace(")",'}') - else: - object=object.replace("$(",'') - object=object.replace(")",'') - else: - termMap=getTermMap(object) - if(termMap=='rr:template "'): - object=object.replace("$",'') - object=object.replace("(","{") - object=object.replace(")",'}') - else: - object=object.replace("$(",'') - object=object.replace(")",'') - if(termMap!='rr:constant '): - template+="\t\t\t"+ termMap + object + '"'+ ';\n' - else: - template+="\t\t\t"+ termMap + object + ';\n' - if type(objec)== list and len(objec)==2: - types = check_type(objec,2) - if(types!="error"): - if(types=="iri"): - template+="\t\t\trr:TermType rr:IRI"+"\n\t\t];\n" - elif(types=="language"): - lenguage=objec[1].replace("~lang","") - template+='\t\t\trr:language "'+ lenguage +'"\n\t\t];\n' - elif(types=="datatype"): - template+="\t\t\trr:datatype "+ objec[1] +"\n\t\t];\n" - else: - raise Exception("Error: incorrect format of predicateObjectMap in mapping " + mapping) - else: - template+="\t\t];\n" - template=template[:-2] - template+= "\n\t];\n\n" - -#CASO LISTA +def get_predicate_access(predicate_object_map): + if constants.YARRRML_PREDICATES in predicate_object_map: + predicate_access = constants.YARRRML_PREDICATES + elif constants.YARRRML_SHORTCUT_PREDICATES in predicate_object_map: + predicate_access = constants.YARRRML_SHORTCUT_PREDICATES else: - if(accessObject in predob): - if (type(predob.get(accessObject))==list): - template="\trr:predicateObjectMap [\n" + "\t\ta rr:PredicateObjectMap;\n" - for predic in predob.get(access): - template+= "\t\trr:predicate " + str(predic) + ";\n" - if(len(predob.get(accessObject))==2 and type(predob.get(accessObject)[1])!=list and type(predob.get(accessObject)[1])!=list): #solo un objeto con tipo - object=predob.get(accessObject)[0] - termMap=getTermMap(object) - if(termMap=='rr:template "'): - object=object.replace("$",'') - object=object.replace("(","{") - object=object.replace(")",'}') - else: - object=object.replace("$(",'') - object=object.replace(")",'') + raise Exception("There isn't a predicate key correctly specify in " + predicate_object_map) + return predicate_access - if(termMap!='rr:constant '): - template+="\t\t\t"+ termMap + object + '"'+ ';\n' - else: - template+="\t\t\t"+ termMap + object + ';\n' - types = check_type_simple(predob.get(accessObject)[1]) - if(types!="error"): - if(types=="iri"): - template+="\t\t\trr:TermType rr:IRI"+"\n\t\t];\n" - elif(types=="language"): - lenguage=predob.get(accessObject)[1].replace("~lang","") - template+='\t\t\trr:language "'+ lenguage +'"\n\t\t];\n' - elif(types=="datatype"): - template+="\t\t\trr:datatype "+ predob.get(accessObject)[1] +"\n\t\t];\n" - else: - raise Exception("Error: incorrect format of predicateObjectMap in mapping " + mapping) - template=template[:-2] - template+= "\n\t];\n\n" - else: - for objec in predob.get(accessObject): - template+="\t\trr:objectMap [ \n\t\t\ta rr:ObjectMap;\n" - if(type(objec) is list): - object = objec[0] - termMap=getTermMap(object) - if(termMap=='rr:template "'): - object=object.replace("$",'') - object=object.replace("(","{") - object=object.replace(")",'}') - else: - object=object.replace("$(",'') - object=object.replace(")",'') - else: - object=objec - termMap=getTermMap(object) - if(termMap=='rr:template "'): - object=object.replace("$",'') - object=object.replace("(","{") - object=object.replace(")",'}') - else: - object=object.replace("$(",'') - object=object.replace(")",'') - if(termMap!='rr:constant '): - template+="\t\t\t"+ termMap + object + '"'+ ';\n' - else: - template+="\t\t\t"+ termMap + object + ';\n' - if type(objec)== list and len(objec)==2: - types = check_type(objec,2) - if(types!="error"): - if(types=="iri"): - template+="\t\t\trr:TermType rr:IRI"+"\n\t\t];\n" - elif(types=="language"): - lenguage=objec[1].replace("~lang","") - template+='\t\t\trr:language "'+ lenguage +'"\n\t\t];\n' - elif(types=="datatype"): - template+="\t\t\trr:datatype "+ objec[1] +"\n\t\t];\n" - else: - raise Exception("Error: incorrect format of predicateObjectMap in mapping " + mapping) - else: - template+="\t\t];\n" - template=template[:-2] - template+= "\n\t];\n\n" - else: - if "value" in predob.get(accessObject): - template+="\t\trr:objectMap [ \n\t\t\ta rr:ObjectMap;\n" - object=predob.get(accessObject).get("value") - termMap=getTermMap(object) - if(termMap=='rr:template "'): - object=object.replace("$",'') - object=object.replace("(","{") - object=object.replace(")",'}') - else: - object=object.replace("$(",'') - object=object.replace(")",'') - if(termMap!='rr:constant '): - template+="\t\t\t"+ termMap + object + '"'+ ';\n' - else: - template+="\t\t\t"+ termMap + object + ';\n' - #"\t\t]\n\t];\n\n" - if "datatype" in predob.get(accessObject): - template+= "\t\t\trr:datatype " + predob.get(accessObject).get("datatype") + "\n\t\t]\n\t];\n\n" - elif "type" in predob.get(accessObject): - template+= "\t\t\trr:TermType " + predob.get(accessObject).get("type") + "\n\t\t]\n\t];\n\n" - elif "language" in predob.get(accessObject): - template+= '\t\t\trr:language "' + predob.get(accessObject).get("language") + '"\n\t\t]\n\t];\n\n' - elif "mapping" in predob.get(accessObject): - template+= joinMapping(data,mapping,predob.get(accessObject)) - else:#creo que sobra - template+="\t\trr:objectMap [ \n\t\t\ta rr:ObjectMap;\n" - object=predob.get(accessObject) - termMap=getTermMap(object) - if(termMap=='rr:template "'): - object=object.replace("$",'') - object=object.replace("(","{") - object=object.replace(")",'}') - else: - object=object.replace("$(",'') - object=object.replace(")",'') - if(termMap!='rr:constant '): - template+="\t\t\t"+ termMap + object + '"'+ ';\n\t\t]\n\t];\n\n' - else: - template+="\t\t\t"+ termMap + object + ';\n\t\t]\n\t];\n\n' - - return template -def addPredicateObjectSimplified(data,mapping,predob): - template= "\trr:predicateObjectMap [\n\t\ta rr:PredicateObjectMap;\n" - if((type(predob[0]) == list or type(predob[1])==list) and len(predob)==2): - if(type(predob[0])==list): #varios predicados - for predicate in predob[0]: - template+="\t\trr:predicate "+str(predicate)+";\n" - else: #un predicado - template+="\t\trr:predicate "+str(predob[0])+";\n" - if(type(predob[1])==list): #varios objetos - for object in predob[1]: - template+="\t\trr:objectMap [ \n\t\t\ta rr:ObjectMap;\n" - object=str(object) - termMap=getTermMap(object) - if(termMap=='rr:template "'): - object=object.replace("$",'') - object=object.replace("(","{") - object=object.replace(")",'}') - else: - object=object.replace("$(",'') - object=object.replace(")",'') - types=check_type_simple(object) - if(types=="error"): - if(termMap!='rr:constant '): - template+="\t\t\t"+ termMap + object + '"'+ ';\n' - else: - template+="\t\t\t"+ termMap + object + ';\n' - else: - if(types=="iri"): - object=object.replace("~iri","") - template+="\t\t\t"+ termMap + object + '";\n' - template+="\t\t\trr:TermType rr:IRI"+"\n\t\t]\n\t];\n\n" - elif(types=="language"): - lenguage=object.replace("~lang","") - template+='\t\t\trr:language "'+ lenguage +'"\n\t\t]\n\t];\n\n' - elif(types=="datatype"): - template+="\t\t\trr:datatype "+ object +"\n\t\t]\n\t];\n\n" - else: #un solo objeto - template+="\t\trr:objectMap [ \n\t\t\ta rr:ObjectMap;\n" - object=str(predob[1]) - termMap=getTermMap(object) - if(termMap=='rr:template "'): - object=object.replace("$",'') - object=object.replace("(","{") - object=object.replace(")",'}') - else: - object=object.replace("$(",'') - object=object.replace(")",'') - types=check_type_simple(object) - if(termMap!='rr:constant '): - template+="\t\t\t"+ termMap + object + '"'+ ';\n' - else: - template+="\t\t\t"+ termMap + object + ';\n' - if(types!="error"): - if(types=="iri"): - template+="\t\t\trr:TermType rr:IRI"+"\n\t\t]\n\t];\n\n" - elif(types=="language"): - lenguage=object.replace("~lang","") - template+='\t\t\trr:language "'+ lenguage +'"\n\t\t]\n\t];\n\n' - elif(types=="datatype"): - template+="\t\t\trr:datatype "+ object +"\n\t\t]\n\t];\n\n" +def get_predicate_object_access(mapping): + if constants.YARRRML_PREDICATEOBJECT in mapping: + predicate_object_access = constants.YARRRML_PREDICATEOBJECT + elif constants.YARRRML_SHORTCUT_PREDICATEOBJECT in mapping: + predicate_object_access = constants.YARRRML_SHORTCUT_PREDICATEOBJECT + else: + raise Exception("There isn't a predicate_object_map key correctly specify in " + mapping) + return predicate_object_access - template = template[:-2] - template+="\n\t];\n" +def add_predicate_object_maps(data, mapping): + po_template = "" + pom_text = "\t" + constants.R2RML_PREDICATE_OBJECT_MAP + " [\n" + mapping_data = data.get(constants.YARRRML_MAPPINGS).get(mapping) + for predicate_object_map in mapping_data.get(get_predicate_object_access(mapping_data)): + if type(predicate_object_map) is list: + po_template += pom_text + add_predicate_object(data, mapping, predicate_object_map) + "\n" + else: + po_template += pom_text + add_predicate_object(data, mapping, predicate_object_map, + get_predicate_access(predicate_object_map), + get_object_access(predicate_object_map)) + "\n" + return po_template - elif(len(predob)==2): - ##solo predicado y objeto singulares - template+="\t\trr:predicate "+str(predob[0])+";\n"+"\t\trr:objectMap [ \n\t\t\ta rr:ObjectMap;\n" - object=str(predob[1]) - if "~" in object: - termMap=getTermMap(object) - if(termMap=='rr:template "'): - object=object.replace("$",'') - object=object.replace("(","{") - object=object.replace(")",'}') - else: - object=object.replace("$(",'') - object=object.replace(")",'') - types=check_type_simple(object) - object1=object.split("~") - if(termMap!='rr:constant '): - template+="\t\t\t"+ termMap + object1[0] + '"'+ ';\n' - else: - template+="\t\t\t"+ termMap + object1[0] + ';\n' - if(types!="error"): - if(types=="iri"): - template+="\t\t\trr:TermType rr:IRI"+"\n\t\t]\n\t];\n\n" - elif(types=="language"): - lenguage=objec[1].replace("~lang","") - template+='\t\t\trr:language "'+ lenguage +'"\n\t\t]\n\t];\n\n' - elif(types=="datatype"): - template+="\t\t\trr:datatype "+ objec[1] +"\n\t\t]\n\t];\n\n" - else: - termMap=getTermMap(object) - if(termMap=='rr:template "'): - object=object.replace("$",'') - object=object.replace("(","{") - object=object.replace(")",'}') - else: - object=object.replace("$(",'') - object=object.replace(")",'') - if(termMap!='rr:constant '): - template+="\t\t\t"+ termMap + object + '"'+ '\n\t\t]\n\t];\n\n' - else: - template+="\t\t\t"+ termMap + object + '\n\t\t]\n\t];\n\n' +def get_object_list(predicate_object, object_access): + object_list = [] + if object_access is not None: + object_maps = predicate_object.get(object_access) + else: + object_maps = predicate_object[1] - elif(len(predob)==3): - #1 pred, 2 obj, 3 datatype, leng - template+="\t\trr:predicate "+str(predob[0])+";\n"+"\t\trr:objectMap [ \n\t\t\ta rr:ObjectMap;\n" - object=str(predob[1]) - termMap=getTermMap(object) - if(termMap=='rr:template "'): - object=object.replace("$",'') - object=object.replace("(","{") - object=object.replace(")",'}') - else: - object=object.replace("$(",'') - object=object.replace(")",'') - if(termMap!='rr:constant '): - template+="\t\t\t"+ termMap + object + '"'+ ';\n' + if type(object_maps) == list: + if len(predicate_object) == 3: + for i in range(len(object_maps)): + object_list.append([object_maps[i], predicate_object[2]]) else: - template+="\t\t\t"+ termMap + object + ';\n' - types = check_type(predob,3) - if(types!="error"): - template+="\t\t\trr:" + types +" "+ predob[2] +"\n\t\t]\n\t];\n\n" - else: - raise Exception("Error: incorrect format of predicateObjectMap in mapping " + mapping) + object_list.extend(object_maps) else: - #ERROR - raise Exception("Error: incorrect format of predicateObjectMap in mapping " + mapping) - - return template - - - + if len(predicate_object) == 3: + object_list.append([object_maps, predicate_object[2]]) + else: + object_list.append(object_maps) + return object_list -def check_type(predob,pos): - type=str(predob[pos-1]) - if "~iri" in type: - return "iri" - elif "~lang" in type: - return "language" - elif ("xsd:" in type) or ("ex:" in type): - return "datatype" +def get_predicate_list(predicate_object, predicate_access): + predicate_list = [] + if predicate_access is not None: + predicate_maps = predicate_object.get(predicate_access) else: - return "error" - -def check_type_simple(object): - - if "~iri" in object: - return "iri" - elif "~lang" in object: - return "language" - elif ("xsd:" in object) or ("ex:" in object): - return "datatype" + predicate_maps = predicate_object[0] + if type(predicate_maps) == list: + predicate_list.extend(predicate_maps) else: - return "error" - - -def joinMapping(data,mapping,objec): - list_mappings=[] - template="" - for mappings in data.get("mappings"): + predicate_list.append(predicate_maps) + return predicate_list + + +def add_predicate_object(data, mapping, predicate_object, predicate_access=None, object_access=None): + template = "" + predicate_list = get_predicate_list(predicate_object, predicate_access) + object_list = get_object_list(predicate_object, object_access) + + for pm in predicate_list: + template += termmap.generate_rml_termmap(constants.R2RML_PREDICATE, constants.R2RML_PREDICATE_CLASS, pm, + "\t\t\t") + for om in object_list: + iri = False + if type(om) == list: + object_value = om[0] + if constants.YARRRML_IRI in om[0]: + object_value = om[0].split(constants.YARRRML_IRI)[0] + iri = True + template += termmap.generate_rml_termmap(constants.R2RML_OBJECT, constants.R2RML_OBJECT_CLASS, + object_value, "\t\t\t") + if len(om) == 2: + types = termmap.check_type(om[1]) + if types != "error": + if types == constants.YARRRML_LANGUAGE: + template = template[0:len(template) - 5] + "\t\t\t" + constants.R2RML_LANGUAGE + " " \ + + om[1].replace(constants.YARRRML_LANG, "") + "\"\n\t\t];\n" + elif types == constants.YARRRML_DATATYPE: + template = template[0:len(template) - 5] + "\t\t\t" + constants.R2RML_DATATYPE + " " \ + + om[1] + "\n\t\t];\n" + if iri: + template = template[0:len(template) - 5] + "\t\t\t" + constants.R2RML_TERMTYPE + " " \ + + constants.R2RML_IRI + "\n\t\t];\n" + elif constants.YARRRML_MAPPING in om: + template += join_mapping(data, mapping, om) + else: + if constants.YARRRML_VALUE in om: + object_value = om.get(constants.YARRRML_VALUE) + else: + object_value = om + if constants.YARRRML_IRI in om: + object_value = om.split(constants.YARRRML_IRI)[0] + iri = True + template += termmap.generate_rml_termmap(constants.R2RML_OBJECT, constants.R2RML_OBJECT_CLASS, + object_value, "\t\t\t") + if type(om) == list and constants.YARRRML_DATATYPE in om: + template = template[0:len(template) - 5] + "\t\t\t" + constants.R2RML_DATATYPE + " " \ + + om.get(constants.YARRRML_DATATYPE) + "\n\t\t];\n" + if (type(om) == list and constants.YARRRML_TYPE in om and om.get(constants.YARRRML_TYPE) == "iri") or iri: + template = template[0:len(template) - 5] + "\t\t\t" + constants.R2RML_TERMTYPE + " " \ + + constants.R2RML_IRI + "\n\t\t];\n" + if type(om) == list and constants.YARRRML_LANGUAGE in om: + template = template[0:len(template) - 5] + "\t\t\t" + constants.R2RML_LANGUAGE + " \"" \ + + om.get(constants.YARRRML_LANGUAGE) + "\"\n\t\t];\n" + + return template + "\t];" + + +def join_mapping(data, mapping, om): + list_mappings = [] + template = "" + for mappings in data.get(constants.YARRRML_MAPPINGS): list_mappings.append(mappings) - if objec.get("mapping") in list_mappings: - subject_list = subjectmod.addSubject(data, objec.get("mapping")) - list_initial_sources= sourcemod.getInitialSources(data) - source_list=sourcemod.addSource(data,objec.get("mapping"),list_initial_sources) - - sumSourSub = len(subject_list) * len(source_list) - for i in range(sumSourSub): - template+="\t\trr:objectMap [ \n\t\t\ta rr:RefObjectMap;\n\t\t\trr:parentTriplesMap <#"+ objec.get("mapping")+"_"+str(i)+">;\n" - if "condition" in objec: - if "parameters" in objec.get("condition"): - list_parameters = objec.get("condition").get("parameters") - if len(list_parameters)==2: - child=list_parameters[0][1] - parent=list_parameters[1][1] - child=child.replace("$(",'"') - child=child.replace(")",'"') - parent=parent.replace("$(",'"') - parent=parent.replace(")",'"') - - template+='\t\t\trr:joinCondition [\n\t\t\t\trr:child ' + child + ';\n\t\t\t\trr:parent ' + parent + ';\n\t\t\t]\n\t\t];\n' + mapping_join = om.get(constants.YARRRML_MAPPING) + + if mapping_join in list_mappings: + subject_list = subject.add_subject(data, mapping_join) + list_initial_sources = source.get_initial_sources(data) + source_list = source.add_source(data, mapping_join, list_initial_sources) + + number_joins_rml = len(subject_list) * len(source_list) + for i in range(number_joins_rml): + template += "\t\t" + constants.R2RML_OBJECT + \ + " [ \n\t\t\ta " + constants.R2RML_REFOBJECT_CLASS + \ + ";\n\t\t\t" + constants.R2RML_PARENT_TRIPLESMAP + " <#" + mapping_join + "_" + str(i) + ">;\n" + if constants.YARRRML_CONDITION in om: + if constants.YARRRML_PARAMETERS in om.get(constants.YARRRML_CONDITION): + list_parameters = om.get(constants.YARRRML_CONDITION).get(constants.YARRRML_PARAMETERS) + if len(list_parameters) == 2: + child = list_parameters[0][1] + parent = list_parameters[1][1] + child = child.replace("$(", '"') + child = child.replace(")", '"') + parent = parent.replace("$(", '"') + parent = parent.replace(")", '"') + + template += "\t\t\t" + constants.R2RML_JOIN_CONITION + \ + " [\n\t\t\t\t" + constants.R2RML_CHILD + " " + child + \ + ";\n\t\t\t\t" + constants.R2RML_PARENT + " " + parent + ";\n\t\t\t]; \n\t\t];\n" else: - raise Exception("Error: more than two parameters in join condition in mapping1 " + mapping) + raise Exception("Error: more than two parameters in join condition in mapping1 " + mapping) else: - template+="\n\t\t]\n\t];" - return template - + template += "\n\t\t]\n" else: raise Exception("Error in reference mapping another mapping in mapping " + mapping) return template - - - -def getTermMap(text): - if("$(" in text and ")" in text): - if text[0]=="$": - return 'rml:reference "' - else: - return 'rr:template "' - else: - return 'rr:constant ' diff --git a/lib/source.py b/lib/source.py index 8f05355..19201ca 100644 --- a/lib/source.py +++ b/lib/source.py @@ -1,181 +1,158 @@ -import yaml +import re +import constants -def getInitialSources(data): - list_initial_sources =[] - if "sources" in data: - for y in data.get("sources"): +def get_initial_sources(data): + list_initial_sources = [] + if constants.YARRRML_SOURCES in data: + for y in data.get(constants.YARRRML_SOURCES): list_initial_sources.append(y) return list_initial_sources -def addSource(data, mapping,list_initial_sources): +def add_source(data, mapping, list_initial_sources): + source_template = "\t" + constants.RML_LOGICAL_SOURCE + " [\n\t\ta " + constants.RML_LOGICAL_SOURCE_PROPERTY + \ + ";\n\t\t" + constants.RML_SOURCE + " " + final_list = [] - source_template="" - list_sources=[] - final_list=[] - #primero se comprueba que este en la lista inicial de sources, sino luego se comprueba si es forma completa o simplificada - - #CASO INITIAL SOURCE: forma de string - if "sources" in data.get("mappings").get(mapping): - if(type(data.get("mappings").get(mapping).get("sources")) is str ): #una unica source - final_list.append(addInitialSource(data,data.get("mappings").get(mapping).get("sources"),mapping)) - else: - for sources in data.get("mappings").get(mapping).get("sources"): #varias sources - list_sources.append(sources) - for sources in list_sources: - if sources in list_initial_sources: - final_list.append(addInitialSource(data,sources,mapping)) - elif "source" in data.get("mappings").get(mapping): - if(type(data.get("mappings").get(mapping).get("source")) is str ): #una unica source - final_list.append(addInitialSource(data,data.get("mappings").get(mapping).get("source"),mapping)) - else: - for sources in data.get("mappings").get(mapping).get("source"): #varias sources - list_sources.append(sources) - for sources in list_sources: - if sources in list_initial_sources: - final_list.append(addInitialSource(data,sources,mapping)) + if constants.YARRRML_SOURCES in data.get(constants.YARRRML_MAPPINGS).get(mapping): + sources = data.get(constants.YARRRML_MAPPINGS).get(mapping).get(constants.YARRRML_SOURCES) + elif constants.YARRRML_SOURCE in data.get(constants.YARRRML_MAPPINGS).get(mapping): # don't find this exist + sources = data.get(constants.YARRRML_MAPPINGS).get(mapping).get(constants.YARRRML_SOURCE) else: raise Exception("ERROR: sources not defined in mapping " + mapping) - #CASO SIMPLIFIED SOURCE Y FULL SOURCE: no tiene forma de string - if "sources" in data.get("mappings").get(mapping): - if (type(data.get("mappings").get(mapping).get("sources")) is not str): - for sources in data.get("mappings").get(mapping).get("sources"): - if "access" in sources: - final_list.append(addSourceFull(data,mapping,sources)) - elif type(sources) is list: - final_list.append(addSourceSimplified(data,mapping,sources)) - elif sources not in list_initial_sources: - raise Exception("ERROR: source " + sources + " in mapping "+ mapping + " not valid") - else: - if (type(data.get("mappings").get(mapping).get("source")) is not str): - for sources in data.get("mappings").get(mapping).get("source"): - if "access" in sources: - final_list.append(addSourceFull(data,mapping,sources)) - elif type(sources) is list: - final_list.append(addSourceSimplified(data,mapping,sources)) - elif sources not in list_initial_sources: - raise Exception("ERROR: source " + sources + " in mapping "+ mapping + " not valid") - return final_list -def addInitialSource(data,sources,mapping): - if type(data.get("sources").get(sources)) is list: - return addSourceSimplified(data,mapping,data.get("sources").get(sources)) + for source in sources: + if source in list_initial_sources: + source = data.get(constants.YARRRML_SOURCES).get(source) - else: - return addSourceFull(data,mapping,data.get("sources").get(sources)) - - -def addSourceSimplified(data,mapping,sources): - source_template = "\trml:logicalSource [\n" +"\t\ta rml:logicalSource;\n" + "\t\trml:source " - source= sources[0] - source_aux = source.split('~') - source_aux1= source_aux[0].split('.') - source_name = source_aux1[0] - source_extension = source_aux1[1] - source_referenceFormulation = source_aux[1] - if(checkExtension(source_extension,source_referenceFormulation)==1): - raise Exception("ERROR: mismatch extension and referenceFormulation in source "+ sources + " in mapping "+mapping) - else: - if len(sources)== 1: - if (source_extension=="csv" or source_extension=="SQL2008"): - source_template+= '"'+source_aux[0]+'"' + ";\n" + "\t\trml:referenceFormulation ql:" + source_referenceFormulation + "\n" + "\t];\n" + if constants.YARRRML_ACCESS in source: + if constants.YARRRML_QUERY in source: + final_list.extend(source_template + database_source(mapping, source)) else: - raise Exception("ERROR: source " + sources + " in mapping " + mapping +" has no iterator") + final_list.append(source_template + add_source_full(mapping, source)) + elif type(source) is list: + final_list.append(source_template + add_source_simplified(mapping, source)) else: - source_delim = sources[1] - source_template+= '"'+source_aux[0]+'"' + ";\n" + "\t\trml:referenceFormulation ql:" + source_referenceFormulation + ";\n" + '\t\trml:iterator "' +source_delim + '"'+ ";\n\t];\n" - return source_template - - -def addSourceFull(data,mapping,sour): - source_final="" - if "query" in sour: - list_source=databaseSource(data,mapping,sour) - return list_source - list_sources=[] - list_sources.append("sources") - list_sources.append(data.get("mappings").get(mapping).get("sources")) - source_template = "\trml:logicalSource [\n" +"\t\ta rml:logicalSource;\n" + "\t\trml:source " - if "access" in sour: - source = str(sour.get("access")) - extension = source.split(".") - else: - raise Exception('ERROR: source' +sour.get("access") + 'in mapping ' + mapping+' has no "access"') - - if "referenceFormulation" in sour: - source_referenceFormulation = str(sour.get("referenceFormulation")) - if checkExtension(extension[1],source_referenceFormulation)==1: - raise Exception("ERROR: mismatch extension and referenceFormulation in source " + sour.get("access") + "in mapping " + mapping) - if getReferenceFOrmulation(source_referenceFormulation)!= "ERROR": - source_referenceFormulation=getReferenceFOrmulation(source_referenceFormulation) + raise Exception("ERROR: source " + source + " in mapping " + mapping + " not valid") + return final_list + + +def add_source_simplified(mapping, source): + source_rdf = "" + file_path = re.sub("~.*", "", source[0]) + reference_formulation = source[0].split('~')[1] + source_extension = file_path.split('.')[1] + ref_formulation_rml = reference_formulation.replace("json", "JSON").replace("csv", "CSV").replace("xpath", "XPath") + if switch_in_reference_formulation(reference_formulation) != source_extension: + raise Exception( + "ERROR: mismatch extension and referenceFormulation in source " + source + " in mapping " + mapping) else: - if(extension[1]=="csv"): - if "iterator" in sour: - source_iterator = str(sour.get("iterator")) - source_template += '"'+source+'"' + ";\n" + '\t\trml:iterator "' +source_iterator + '"\n\t];\n' - return source_template + if len(source) == 1: # si no tiene iterador + if source_extension == "csv" or source_extension == "SQL2008": + source_rdf += '"' + file_path + '"' + ";\n" + "\t\t" + constants.RML_REFERENCE_FORMULATION + " ql:" \ + + ref_formulation_rml + "\n" + "\t];\n" else: - source_template += '"'+source+'"' + ";\n" + '"\n\t];\n' - return source_template + raise Exception("ERROR: source " + source + " in mapping " + mapping + " has no iterator") + else: # source[1] es el iterador en json y xml + source_rdf += "\"" + file_path + "\";\n\t\t" + constants.RML_REFERENCE_FORMULATION + " ql:" \ + + ref_formulation_rml + ";\n\t\t" + constants.RML_ITERATOR + " \"" \ + + source[1] + "\";\n\t];\n" + return source_rdf + + +def add_source_full(mapping, source): + source_rdf = "" + + access = str(source.get(constants.YARRRML_ACCESS)) + extension = source.split(".")[1] + + if constants.YARRRML_REFERENCE_FORMULATION in source: + reference_formulation = str(source.get(constants.YARRRML_REFERENCE_FORMULATION)) + format_from_reference = switch_in_reference_formulation(reference_formulation.lower()) + ref_formulation_rml = reference_formulation.replace("json", "JSON").replace("csv", "CSV").replace("xpath","XPath") + if extension == format_from_reference or format_from_reference == "ERROR": + raise Exception("ERROR: not referenceFormulation found or mismatch between the format and " + "referenceFormulation in source " + access + "in mapping " + mapping) + if constants.YARRRML_ITERATOR in source: + source_iterator = str(source.get(constants.YARRRML_ITERATOR)) + + source_rdf += "\"" + access + "\";\n\t\t" + constants.RML_REFERENCE_FORMULATION + " ql:" \ + + ref_formulation_rml + ";\n\t\t" + constants.RML_ITERATOR + " \"" \ + + source_iterator + "\"\n\t];\n" else: - raise Exception("ERROR: source "+sour.get("access") + "in mapping " + mapping + " has no referenceFormulation") - if "iterator" in sour: - source_iterator = str(sour.get("iterator")) - source_template += '"'+source+'"' + ";\n" + "\t\trml:referenceFormulation ql:"+ source_referenceFormulation +";\n" + '\t\trml:iterator "' +source_iterator + '"\n\t];\n' + if extension == "csv" or extension == "SQL2008": + source_rdf += "\"" + access + "\";\n\t\t" + constants.RML_REFERENCE_FORMULATION + " ql:" \ + + ref_formulation_rml + ";\n\n\t];\n" + else: + raise Exception("ERROR: source " + access + "in mapping " + mapping + " has no referenceFormulation") + else: - if (extension[1]=="csv" or extension[1]=="SQL2008"): - source_template += '"'+source+'"' + ";\n" + "\t\trml:referenceFormulation ql:"+ source_referenceFormulation +";\n" + '"\n\t];\n' + if extension == "csv": + source_rdf += "\"" + access + "\";\n\n\t];\n" else: - raise Exception("ERROR: source "+sour.get("access") + " in mapping " + mapping + " has no iterator") - source_final += source_template - return source_final - -def checkExtension (extension,referenceFormulation): - switcher={ - "json":"jsonpath", - "csv":"csv", - "xml":"xpath" - #para querys - } - reference = switcher.get(extension,"ERROR") - if reference == referenceFormulation: - return 0 - else: - return 1 - -def getReferenceFOrmulation (referenceFormulation): - switcher={ - "jsonpath":"JSONPath", - "csv":"CSV", - "xpath":"XPath" - } - return switcher.get(referenceFormulation,"ERROR") - - -def databaseSource(data,mapping,source): - list=[] - templatelog="\trml:logicalSource [\n\t\ta rml:LogicalSource;\n" - if "access" in source: - if "credentials" in source: - if "type" in source and source.get("type")=="mysql": - templatelog+='\t\trml:source <#DataSource_'+mapping+'>;\n\t\trr:sqlVersion rr:SQL2008;\n\t\trml:query "'+source.get("query") +'";\n\t\t' - if "referenceFormulation" in source: - formu=getReferenceFOrmulation(source.get("referenceFormulation")) - templatelog+="\t\trml:referenceFormulation ql:"+ formu +"\n\t];\n" + raise Exception("ERROR: source " + access + "in mapping " + mapping + " has no referenceFormulation") + + return source_rdf + + +def database_source(mapping, source): + source_database = [] + source_rdf = "" + if constants.YARRRML_ACCESS in source: + if constants.YARRRML_CREDENTIALS in source: + if constants.YARRRML_TYPE in source: + source_rdf += "<#DataSource_" + mapping + ">;\n\t\t" \ + + constants.R2RML_SQL_VERSION + " rr:SQL2008;\n\t\t" \ + + constants.R2RML_SQL_QUERY + " \"" + source.get(constants.YARRRML_QUERY) + "\";\n\t\t" + if constants.YARRRML_REFERENCE_FORMULATION in source: + source_rdf += "\t\t" + constants.RML_REFERENCE_FORMULATION + " ql:" \ + + switch_in_reference_formulation( + source.get(constants.YARRRML_REFERENCE_FORMULATION)) + "\n\t];\n" else: - templatelog+="\n\t];\n" + source_rdf += "\n\t];\n" + source_database.append(source_rdf) + source_database.append(generate_database_connection(mapping, source)) else: - raise Exception("ERROR: no credentials to get access to source in mapping "+ mapping) + raise Exception("ERROR: no credentials to get access to source in mapping " + mapping) else: raise Exception("ERROR: no access to the source in mapping " + mapping) + return source_database - list.append(templatelog) - - templateDat = '<#DataSource_'+mapping+'> a d2rq:Database;\n\tdrr1:jdbcDSN "'+ source.get("access")+'";\n\td2rq:jdbcDriver "com.mysql.jdbc.Driver";\n\td2rq:username "'+source.get("credentials").get("username")+'";\n\td2rq:password "'+source.get("credentials").get("password")+'".\n\n' - list.append(templateDat) - - return list +def switch_in_reference_formulation(value): + value = value.lower() + if value == "csv": + switcher = value + elif "json" in value: + if "path" in value: + switcher = "json" + else: + switcher = "jsonpath" + elif "x" in value: + if "path" in value: + switcher = "xml" + else: + switcher = "xpath" + return switcher + + +def generate_database_connection(mapping, source): + type = source.get(constants.YARRRML_TYPE) + if type == "mysql": + driver = "com.mysql.jdbc.Driver" + elif type == "postgresql": + driver = "org.postgresql.Driver" + elif type == "sqlserver": + driver = "com.microsoft.sqlserver.jdbc.SQLServerDriver" + else: + driver = "" + + database = "<#DataSource_" + mapping + "> a " + constants.D2RQ_DATABASE_CLASS + ";\n\t" \ + + constants.D2RQ_DSN + " \"" + source.get(constants.YARRRML_ACCESS) + "\";\n\t" \ + + constants.D2RQ_DRIVER + " \"" + driver + "\";\n\t" \ + + constants.D2RQ_USER + " \"" + source.get(constants.YARRRML_CREDENTIALS).get(constants.YARRRML_USERNAME) + "\";\n\t" \ + + constants.D2RQ_PASS + " \"" + source.get(constants.YARRRML_CREDENTIALS).get(constants.YARRRML_PASSWORD) + "\".\n\n" + return database diff --git a/lib/subject.py b/lib/subject.py index 9ce2375..3df859c 100644 --- a/lib/subject.py +++ b/lib/subject.py @@ -1,121 +1,40 @@ -import yaml +import sys -def addSubject(data, mapping): - list_subject=[] - final_list=[] +import constants +import termmap - if ("s" in data.get("mappings").get(mapping)): - list_subject=data.get("mappings").get(mapping).get("s") - if (type(list_subject)== list): #varios sujetos - for subject in list_subject: - subject_template = "\trr:subjectMap [\n" + "\t\ta rr:SubjectMap;\n"+ "\t\t" - termMap=getTermMap(subject) - if(termMap=='rr:template "'): - subject=subject.replace("$",'') - subject=subject.replace("(","{") - subject=subject.replace(")",'}') - else: - subject=subject.replace("$(",'') - subject=subject.replace(")",'') - if(termMap!='rr:constant '): - subject_template+= termMap + ' "'+ subject + '"'+ "\n\t];\n\n" - else: - subject_template+= termMap + subject + "\n\t];\n\n" - final_list.append(subject_template) +## Adds a subject or set of subjects to the given TriplesMap +def add_subject(data, mapping): + subjects = [] + rml_subjects = [] + only_one = False - else: #un solo sujeto - subject_template = "\trr:subjectMap [\n" + "\t\ta rr:SubjectMap;\n"+ "\t\t" - subject=list_subject - termMap=getTermMap(subject) - if(termMap=='rr:template "'): - subject=subject.replace("$",'') - subject=subject.replace("(","{") - subject=subject.replace(")",'}') - else: - subject=subject.replace("$(",'') - subject=subject.replace(")",'') - if(termMap!='rr:constant '): - subject_template+= termMap + ' "'+ subject + '"'+ "\n\t];\n\n" - else: - subject_template+= termMap + subject + "\n\t];\n\n" - final_list.append(subject_template) - - - elif ("subjects" in data.get("mappings").get(mapping)): - list_subject= data.get("mappings").get(mapping).get("subjects") - if (type(list_subject)== list): #varios sujetos - for subject in list_subject: - subject_template = "\trr:subjectMap [\n" + "\t\ta rr:SubjectMap;\n"+ "\t\t" - termMap=getTermMap(subject) - if(termMap=='rr:template "'): - subject=subject.replace("$",'') - subject=subject.replace("(","{") - subject=subject.replace(")",'}') - else: - subject=subject.replace("$(",'') - subject=subject.replace(")",'') - if(termMap!='rr:constant '): - subject_template+= termMap + ' "'+ subject + '"'+ "\n\t];\n\n" - else: - subject_template+= termMap + subject + "\n\t];\n\n" - final_list.append(subject_template) - - - else: #un solo sujeto - subject_template = "\trr:subjectMap [\n" + "\t\ta rr:SubjectMap;\n"+ "\t\t" - subject=list_subject - termMap=getTermMap(subject) - if(termMap=='rr:template "'): - subject=subject.replace("$",'') - subject=subject.replace("(","{") - subject=subject.replace(")",'}') - else: - subject=subject.replace("$(",'') - subject=subject.replace(")",'') - if(termMap!='rr:constant '): - subject_template+= termMap + ' "'+ subject + '"'+ "\n\t];\n\n" - else: - subject_template+= termMap + subject + "\n\t];\n\n" - final_list.append(subject_template) - + if constants.YARRRML_SHORTCUT_SUBJECTS in data.get(constants.YARRRML_MAPPINGS).get(mapping): + subject = data.get(constants.YARRRML_MAPPINGS).get(mapping).get(constants.YARRRML_SHORTCUT_SUBJECTS) + elif constants.YARRRML_SUBJECTS in data.get(constants.YARRRML_MAPPINGS).get(mapping): + subject = data.get(constants.YARRRML_MAPPINGS).get(mapping).get(constants.YARRRML_SUBJECTS) + elif constants.YARRRML_SUBJECT in data.get(constants.YARRRML_MAPPINGS).get(mapping): + subject = data.get(constants.YARRRML_MAPPINGS).get(mapping).get(constants.YARRRML_SUBJECT) + only_one = True + else: + raise Exception("ERROR: no subjects in mapping " + mapping) - elif ("subject" in data.get("mappings").get(mapping)): - list_subject=data.get("mappings").get(mapping).get("subject") - if (type(list_subject)== list): #varios sujetos ERROR - raise Exception("ERROR: Only one subject should be in mapping: " + mapping) + if type(subject) != list: + subjects.append(subject) + elif type(subject) == list and only_one: + raise Exception("ERROR: Only one subject can be define using keyword 'subject'") + else: + rml_subjects = subject - else: #un solo sujeto - subject_template = "\trr:subjectMap [\n" + "\t\ta rr:SubjectMap;\n"+ "\t\t" - subject=list_subject - termMap=getTermMap(subject) - if(termMap=='rr:template "'): - subject=subject.replace("$",'') - subject=subject.replace("(","{") - subject=subject.replace(")",'}') - else: - subject=subject.replace("$(",'') - subject=subject.replace(")",'') - if(termMap!='rr:constant '): - subject_template+= termMap + ' "'+ subject + '"'+ "\n\t];\n\n" - else: - subject_template+= termMap + subject + "\n\t];\n\n" - final_list.append(subject_template) + for subject in subjects: + subject_termmap = termmap.generate_rml_termmap(constants.R2RML_SUBJECT_PROPERTY, constants.R2RML_SUBJECT_CLASS, subject, "\t\t") + rml_subjects.append(subject_termmap) + return rml_subjects - else: - raise Exception("ERROR: no subjects in mapping " + mapping) - return final_list -def getTermMap(text): - if("$(" in text and ")" in text): - if text[0]=="$": - return "rml:reference " - else: - return "rr:template " - else: - return "rr:constant " diff --git a/lib/termmap.py b/lib/termmap.py new file mode 100644 index 0000000..666c744 --- /dev/null +++ b/lib/termmap.py @@ -0,0 +1,41 @@ +import constants + + +## return the type of TermMap based on the input text +def get_termmap_type(text): + if "$(" in text and ")" in text: + if text[0] == "$" and text[len(text) - 1] == ")" and text.count("$(") == 1: + return constants.RML_REFERENCE + else: + return constants.R2RML_TEMPLATE + else: + return constants.R2RML_CONSTANT + + +## Generates a TermMap (subject, predicate, object) based on the property, class and the text +def generate_rml_termmap(rml_property, rml_class, text, identation): + template = identation[0:-1] + rml_property + " [\n"+identation+"a " + rml_class + ";\n" + identation + term_map = get_termmap_type(text) + if term_map == constants.R2RML_TEMPLATE: + text = text.replace("$(", "{") + text = text.replace(")", "}") + elif term_map == constants.RML_REFERENCE: + text = text.replace("$(", "") + text = text.replace(")", "") + elif term_map == constants.R2RML_CONSTANT and text == "a": + text = constants.RDF_TYPE + if term_map != "rr:constant": + template += term_map + " \"" + text + "\";\n"+identation[0:-1]+"];\n" + else: + template += term_map + " " + text + ";\n"+identation[0:-1]+"];\n" + + return template + + +def check_type(om): + if "~lang" in om: + return "language" + elif ":" in om: + return "datatype" + else: + return "error" \ No newline at end of file diff --git a/requirements.txt b/requirements.txt new file mode 100644 index 0000000..af95972 --- /dev/null +++ b/requirements.txt @@ -0,0 +1,2 @@ +pyyaml == 5.4.1 +rdflib == 5.0.0 \ No newline at end of file

Does not support json as a source Does not support JSON format as a source ```yaml prefixes: ex: "http://example.com/" mappings: person: sources: - ['data.json~jsonpath', '$.persons[*]'] s: http://example.com/$(firstname) po: - [a, foaf:Person] - [ex:name, $(firstname)] ``` Output: ```yaml @prefix ex: <http://example.com/>. @prefix rr: <http://www.w3.org/ns/r2rml#>. @prefix rml: <http://semweb.mmlab.be/ns/rml#>. @prefix rdf: <http://www.w3.org/1999/02/22-rdf-syntax-ns#>. ```

2021-07-12T07:00:00

0.0

[]

CouncilDataProject/cdp-backend

CouncilDataProject__cdp-backend-236

0982830358df2c62ab00a05bf8f862c4f3ee168c

diff --git a/cdp_backend/utils/file_utils.py b/cdp_backend/utils/file_utils.py index c8c27073..7ea300e4 100644 --- a/cdp_backend/utils/file_utils.py +++ b/cdp_backend/utils/file_utils.py @@ -6,6 +6,7 @@ import math import random import re +import shutil from hashlib import sha256 from pathlib import Path from uuid import uuid4 @@ -225,21 +226,27 @@ def resource_copy( # noqa: C901 ) return str(dst) - # Set custom timeout for http resources + # Common case: http(s) URI if uri.startswith("http"): # The verify=False is passed to any http URIs # It was added because it's very common for SSL certs to be bad # See: https://github.com/CouncilDataProject/cdp-scrapers/pull/85 # And: https://github.com/CouncilDataProject/seattle/runs/5957646032 - with open(dst, "wb") as open_dst: - open_dst.write( - requests.get( - uri, - verify=False, - timeout=1800, - ).content - ) + # Use stream=True to avoid downloading the entire file into memory + # See: https://github.com/CouncilDataProject/cdp-backend/issues/235 + try: + # This response must be closed after the copy is done. But using + # `with requests.get() as response` fails mypy type checking. + # See: https://requests.readthedocs.io/en/latest/user/advanced/#body-content-workflow + response = requests.get(uri, stream=True, verify=False, timeout=1800) + response.raise_for_status() + with open(dst, "wb") as open_dst: + shutil.copyfileobj( + response.raw, open_dst, length=64 * 1024 * 1024 # 64MB chunks + ) + finally: + response.close() else: # TODO: Add explicit use of GCS credentials until public read is fixed diff --git a/pyproject.toml b/pyproject.toml index bf6585a3..0e2bcd29 100644 --- a/pyproject.toml +++ b/pyproject.toml @@ -103,6 +103,7 @@ test = [ # Extras "networkx>=2.5", "pydot>=1.4", + "requests-mock>=1.10.0" ] docs = [ # Sphinx + Doc Gen + Styling

Inefficient usage of requests.get for very large videos causes event gather to fail ### Describe the Bug Usage of [requests](https://github.com/CouncilDataProject/cdp-backend/blob/0982830358df2c62ab00a05bf8f862c4f3ee168c/cdp_backend/utils/file_utils.py#L237) to write the content to the open destination file as a BufferedWriter causes unconstrained memory usage based on the size of the video file being copied from the remote origin. For example, consider the mp4 at https://sg001-media.sliq.net/00309/Media2/2023/04/27/House%20Floor%20Session-House%20Chamber-080505_46177_1113.mp4... ```sh curl -Ss -I https://sg001-media.sliq.net/00309/Media2/2023/04/27/House%20Floor%20Session-House%20Chamber-080505_46177_1113.mp4 | grep ^content-length content-length: 8068641879 ``` The video is ~8GB in size and uses too much memory for n1-standard-4 machine type we use by default in Google Cloud especially when whisper model is already loaded. ### Expected Behavior Downloading video files from remote origin should use a pattern that constrains memory usage in the python process. ### Reproduction ```python import requests with open("movie.mp4", "wb") as f: f.write( requests.get("https://sg001-media.sliq.net/00309/Media2/2023/04/27/House%20Floor%20Session-House%20Chamber-080505_46177_1113.mp4").content ) ``` Running this will download the 8G file and copy to output file. The memory usage will be unconstrained since requests to will load the entire body into memory before copying the file to disk. ### Environment - OS Version: N/A - cdp-backend Version: latest

2023-05-13T18:50:38

0.0

[]

CouncilDataProject/cdp-backend

CouncilDataProject__cdp-backend-214

ab3a65dee89f1c4a0c9ad9ad3abe0881f7c84d99

diff --git a/cdp_backend/pipeline/event_gather_pipeline.py b/cdp_backend/pipeline/event_gather_pipeline.py index 12dca809..e8cdc49f 100644 --- a/cdp_backend/pipeline/event_gather_pipeline.py +++ b/cdp_backend/pipeline/event_gather_pipeline.py @@ -160,7 +160,11 @@ def create_event_gather_flow( if session.caption_uri is not None: # If the caption doesn't exist, remove the property # This will result in Speech-to-Text being used instead - if not resource_exists(session.caption_uri): + if not resource_exists( + session.caption_uri + ) or not file_utils.caption_is_valid( + tmp_video_filepath, session.caption_uri + ): log.warning( f"File not found using provided caption URI: " f"'{session.caption_uri}'. " diff --git a/cdp_backend/utils/file_utils.py b/cdp_backend/utils/file_utils.py index a23bd446..e9fa0ae5 100644 --- a/cdp_backend/utils/file_utils.py +++ b/cdp_backend/utils/file_utils.py @@ -7,13 +7,21 @@ import re from hashlib import sha256 from pathlib import Path +from tempfile import TemporaryDirectory from typing import Optional, Tuple, Union from uuid import uuid4 +import ffmpeg import fireo import fsspec import requests +import webvtt from fsspec.core import url_to_fs +from webvtt.exceptions import ( + InvalidCaptionsError, + MalformedCaptionError, + MalformedFileError, +) from ..database import models as db_models @@ -626,3 +634,50 @@ def clip_and_reformat_video( log.error(ffmpeg_stderr) return output_path + + +def caption_is_valid(video_uri: str, caption_uri: str) -> bool: + """ + Validate the caption file at the URI provided. + + Parameters + ---------- + video_uri: str + The URI for the the target video. + uri: str + The URI to validate caption file for. + + Returns + ------- + status: bool + The validation status. + + Notes + ----- + Duration of the video at video_uri + and the duration of the caption file are compared. + The caption file is accepted if the durations differ by no more than 20%. + """ + try: + ffprobe = ffmpeg.probe(video_uri) + except ffmpeg.Error as e: + log.warning(f"ffprobe({video_uri}): {e.stderr}") + return False + + # Making sure temp copy of the caption file is deleted when finished + with TemporaryDirectory() as dir_path: + local_caption_path = resource_copy(caption_uri, dst=dir_path) + try: + caption_length = webvtt.read(local_caption_path).total_length + except (InvalidCaptionsError, MalformedCaptionError, MalformedFileError) as e: + log.warning(f"webvtt.read({caption_uri}): {str(e)}") + return False + + similar_audio_streams = filter( + lambda s: s.get("codec_type", "") == "audio" + and math.isclose( + float(s.get("duration", "0.0")), caption_length, rel_tol=0.2 + ), + ffprobe.get("streams", []), + ) + return any(similar_audio_streams)

Filter out bad caption files ### Feature Description ```Json {"generator": "CDP WebVTT Conversion -- CDP v3.2.3", "confidence": 0.97, "session_datetime": "2022-09-14T10:00:00-04:00", "created_datetime": "2022-09-17T15:51:32.537666", "sentences": [{"index": 0, "confidence": 0.97, "start_time": 0.0, "end_time": 0.0, "words": [], "text": "", "speaker_index": 0, "speaker_name": null, "annotations": null}], "annotations": null} ``` Note `end_time` in the above output from a PIttsburgh, PA captions processing. Turns out that event does have a caption file in the Legistar data structure for the event _but_ it is in fact empty. We want to filter these out just attempt speech-to-text. ### Use Case Allow for proper transcript generation in `generate_transcript()` by appropriately filtering out empty caption files. ### Solution Compare the lengths of the video and the caption file. If they differ by more than some threshold, e.g. 20%, throw away the caption file. This means the scraper can no longer just hand off the caption file URL as-is. ### Alternatives - Throw away caption files with file size less than some threshold, e.g. 100 bytes. - Throw away caption files less than ~1 minute.

2022-09-23T06:14:34

0.0

[]

CouncilDataProject/cdp-backend

CouncilDataProject__cdp-backend-193

00f98763aab4c9a7e69e69c06015d9ed17d64c78

diff --git a/cdp_backend/sr_models/webvtt_sr_model.py b/cdp_backend/sr_models/webvtt_sr_model.py index e0abb019..5d986a5a 100644 --- a/cdp_backend/sr_models/webvtt_sr_model.py +++ b/cdp_backend/sr_models/webvtt_sr_model.py @@ -164,7 +164,15 @@ def _get_sentences( # List of text, representing a sentence lines: List[str] = [] start_time: Optional[float] = None + for caption in speaker_turn_captions: + + # Clean text of line breaks + caption.text = caption.text.replace("\n", " ") + + # Remove any double spaces as result of line break removal + caption.text = caption.text.replace(" ", " ") + if start_time is None: start_time = caption.start_in_seconds lines.append(caption.text)

WebVTT parsing fails for Boston  ### Describe the Bug _A clear and concise description of the bug._ WebVTT parsing isn't the greatest for Boston: https://councildataproject.org/boston/#/events/3c7f0df0ecd4 ### Expected Behavior _What did you expect to happen instead?_ WebVTT parsing should be the same level or better than Seattle

Boston event gather log for where closed caption is pulled from: https://github.com/CouncilDataProject/boston/runs/6759035563?check_suite_focus=true#step:9:7200 Seems like you should be able to pull that same URL: http://boston.granicus.com//videos/87/captions.vtt and test out changes locally @isaacna or @Shak2000 do either of you have the time or interest to take this one? > @isaacna or @Shak2000 do either of you have the time or interest to take this one? I should have some time later this week or next to take a look

2022-06-13T23:06:45

0.0

[]

CouncilDataProject/cdp-backend

CouncilDataProject__cdp-backend-185

f77ca6d02099dcea71da8378ddf742951a4665a7

diff --git a/.github/workflows/build-docs.yml b/.github/workflows/build-docs.yml index 7ff4d233..bdbfda29 100644 --- a/.github/workflows/build-docs.yml +++ b/.github/workflows/build-docs.yml @@ -15,7 +15,7 @@ jobs: - name: Set up Python uses: actions/setup-python@v2 with: - python-version: 3.9 + python-version: '3.10' cache: 'pip' - name: Install Packages run: | diff --git a/.github/workflows/build-main.yml b/.github/workflows/build-main.yml index 0e922800..f469d390 100644 --- a/.github/workflows/build-main.yml +++ b/.github/workflows/build-main.yml @@ -19,7 +19,7 @@ jobs: strategy: fail-fast: false matrix: - python-version: [3.7, 3.8, 3.9] + python-version: [3.8, 3.9, '3.10'] os: [ubuntu-latest, macOS-latest, windows-latest] steps: @@ -62,7 +62,7 @@ jobs: - name: Set up Python uses: actions/setup-python@v2 with: - python-version: 3.9 + python-version: '3.10' cache: 'pip' - name: Install graphviz run: | @@ -85,7 +85,7 @@ jobs: - name: Set up Python uses: actions/setup-python@v2 with: - python-version: 3.9 + python-version: '3.10' cache: 'pip' - name: Install Dependencies run: | diff --git a/.github/workflows/check-pr.yml b/.github/workflows/check-pr.yml index 8221da74..97842300 100644 --- a/.github/workflows/check-pr.yml +++ b/.github/workflows/check-pr.yml @@ -8,7 +8,7 @@ jobs: strategy: fail-fast: false matrix: - python-version: [3.7, 3.8, 3.9] + python-version: [3.8, 3.9, '3.10'] os: [ubuntu-latest, macOS-latest, windows-latest] steps: @@ -51,7 +51,7 @@ jobs: - name: Set up Python uses: actions/setup-python@v2 with: - python-version: 3.9 + python-version: '3.10' cache: 'pip' - name: Install graphviz run: | @@ -71,7 +71,7 @@ jobs: - name: Set up Python uses: actions/setup-python@v2 with: - python-version: 3.9 + python-version: '3.10' cache: 'pip' - name: Install graphviz run: | diff --git a/.pre-commit-config.yaml b/.pre-commit-config.yaml index bf891b19..c189180f 100644 --- a/.pre-commit-config.yaml +++ b/.pre-commit-config.yaml @@ -5,7 +5,7 @@ repos: hooks: - id: flake8 additional_dependencies: [flake8-typing-imports, flake8-debugger] - args: [--count, --show-source, --statistics, --min-python-version=3.7.0] + args: [--count, --show-source, --statistics, --min-python-version=3.8.0] - repo: https://github.com/myint/autoflake rev: v1.4 hooks: diff --git a/LICENSE b/LICENSE index c65d50b0..c9f5e4f2 100644 --- a/LICENSE +++ b/LICENSE @@ -1,6 +1,6 @@ MIT License -Copyright (c) 2020, Council Data Project Contributors +Copyright (c) 2022, Council Data Project Contributors Permission is hereby granted, free of charge, to any person obtaining a copy of this software and associated documentation files (the "Software"), to deal diff --git a/Makefile b/Makefile index 24b77a20..08a0550e 100644 --- a/Makefile +++ b/Makefile @@ -56,8 +56,8 @@ gen-docs: ## Generate Sphinx HTML documentation, including API docs create_cdp_database_uml \ -o docs/_static/cdp_database_diagram.dot dot \ - -T png \ - -o docs/_static/cdp_database_diagram.png docs/_static/cdp_database_diagram.dot + -T jpg \ + -o docs/_static/cdp_database_diagram.jpg docs/_static/cdp_database_diagram.dot create_cdp_ingestion_models_doc \ -t docs/ingestion_models.template \ -o docs/ingestion_models.md diff --git a/cdp_backend/__init__.py b/cdp_backend/__init__.py index 097173c9..f392cd0d 100644 --- a/cdp_backend/__init__.py +++ b/cdp_backend/__init__.py @@ -2,7 +2,7 @@ """Top-level package for cdp-backend.""" -__author__ = "Jackson Maxfield Brown" +__author__ = "Council Data Project Contributors" __email__ = "[email protected]" from .version import __version__ # noqa: F401 diff --git a/dev-infrastructure/README.md b/dev-infrastructure/README.md index cef61168..0a1de564 100644 --- a/dev-infrastructure/README.md +++ b/dev-infrastructure/README.md @@ -57,9 +57,6 @@ restart your terminal after installation._ After `pulumi` and `gcloud` have both been installed and terminal restarted, run the following commands to setup your local machine with credentials to both services. -**Note:** Pulumi only supports Python 3.7, when creating dev infrastructures you -need to run these scripts in a py37 environment. - ```bash cd cdp-backend/dev-infrastructure make login diff --git a/docs/conf.py b/docs/conf.py index 283f5135..477b6947 100755 --- a/docs/conf.py +++ b/docs/conf.py @@ -21,8 +21,6 @@ import os import sys -import sphinx_rtd_theme # noqa: F401 - import cdp_backend sys.path.insert(0, os.path.abspath("..")) @@ -70,8 +68,8 @@ # General information about the project. project = u"cdp-backend" -copyright = u"2020, Jackson Maxfield Brown" -author = u"Jackson Maxfield Brown" +copyright = u"2022, Council Data Project Contributors" +author = u"Council Data Project Contributors" # The version info for the project you"re documenting, acts as replacement # for |version| and |release|, also used in various other places throughout @@ -106,16 +104,13 @@ # The theme to use for HTML and HTML Help pages. See the documentation for # a list of builtin themes. # -html_theme = "sphinx_rtd_theme" +html_theme = "furo" # Theme options are theme-specific and customize the look and feel of a # theme further. For a list of options available for each theme, see the # documentation. # -html_theme_options = { - "collapse_navigation": False, - "prev_next_buttons_location": "top", -} +html_theme_options = {} # Add any paths that contain custom static files (such as style sheets) here, # relative to this directory. They are copied after the builtin static files, @@ -154,7 +149,7 @@ master_doc, "cdp_backend.tex", u"cdp-backend Documentation", - u"Jackson Maxfield Brown", + u"Council Data Project Contributors", "manual", ), ] diff --git a/docs/database_schema.md b/docs/database_schema.md index 50227c6e..df5b1130 100644 --- a/docs/database_schema.md +++ b/docs/database_schema.md @@ -5,7 +5,7 @@ While CDP utilizes a called [Cloud Firestore](https://firebase.google.com/docs/firestore), it is still useful to view the database schema as a UML diagram. -![cdp-database-schema](./_static/cdp_database_diagram.png) +![cdp-database-schema](./_static/cdp_database_diagram.jpg) ### Notes * `*` denotes that the field is required diff --git a/setup.py b/setup.py index ec288ea4..d291b5bb 100644 --- a/setup.py +++ b/setup.py @@ -13,47 +13,46 @@ ] infrastructure_requirements = [ - "pulumi~=3.3", - "pulumi-google-native~=0.7.0", - "pulumi-gcp~=5.7", + "pulumi~=3.31", + "pulumi-google-native~=0.18", + "pulumi-gcp~=6.0", ] pipeline_requirements = [ "dask[distributed]>=2021.7.0", - "ffmpeg-python~=0.2.0", - "google-cloud-speech~=2.11.0", + "ffmpeg-python==0.2.0", + "google-cloud-speech~=2.13", "graphviz~=0.16", - "imageio~=2.9", - "imageio-ffmpeg~=0.4.4", + "imageio~=2.18", + "imageio-ffmpeg~=0.4", "nltk~=3.6", - "pandas~=1.2", - "prefect~=0.14.0", - "rapidfuzz~=1.4", + "pandas~=1.0", + "prefect~=1.2", + "rapidfuzz~=2.0", "spacy~=3.0", - "truecase~=0.0.12", - "webvtt-py~=0.4.6", - "yt-dlp~=2022.2.4", + "truecase==0.0.14", + "webvtt-py==0.4.6", + "yt-dlp>=2022.2.4", ] test_requirements = [ *infrastructure_requirements, *pipeline_requirements, "black>=22.3.0", - "codecov==2.1.12", - "fastparquet~=0.7", + "codecov>=2.1.12", "flake8>=3.8.3", "flake8-debugger>=3.2.1", "isort>=5.7.0", "mypy>=0.790", "networkx>=2.5", - "pyarrow~=5.0", + "pyarrow>=5.0", "pydot>=1.4", "pytest>=5.4.3", "pytest-cov>=2.9.0", "pytest-raises>=0.11", "tox>=3.15.2", "types-pytz>=2021.1.2", - "types-requests~=0.1.11", + "types-requests", ] dev_requirements = [ @@ -61,25 +60,24 @@ *test_requirements, "bokeh>=2.3.2", "bump2version>=1.0.1", - "coverage==5.4", + "coverage>=5.4", "ipython>=7.15.0", "jinja2>=2.11.2", "m2r2>=0.2.7", - "prefect[viz]~=0.14.0", - "pytest-runner>=5.2", + "prefect[viz]", "Sphinx>=3.4.3", - "sphinx_rtd_theme>=0.5.1", + "furo>=2022.4.7", "twine>=3.1.1", "wheel>=0.34.2", ] requirements = [ - "aiohttp~=3.7.4.post0", - "dataclasses-json~=0.5", - "fireo~=1.4", + "aiohttp>=3.7.4.post0", + "dataclasses-json>=0.5", + "fireo>=1.4", "fsspec", # Version pin set by gcsfs - "gcsfs~=2021.7.0", - "requests~=2.26.0", + "gcsfs>=2021.7.0", + "requests>=2.26.0", ] extra_requirements = { @@ -95,16 +93,18 @@ } setup( - author="Jackson Maxfield Brown, To Huynh, Isaac Na", + author=( + "Jackson Maxfield Brown, To Huynh, Isaac Na, Council Data Project Contributors" + ), author_email="[email protected]", classifiers=[ - "Development Status :: 2 - Pre-Alpha", + "Development Status :: 5 - Production/Stable", "Intended Audience :: Developers", "License :: OSI Approved :: MIT License", "Natural Language :: English", - "Programming Language :: Python :: 3.7", "Programming Language :: Python :: 3.8", "Programming Language :: Python :: 3.9", + "Programming Language :: Python :: 3.10", ], description=( "Data storage utilities and processing pipelines to run on CDP server " @@ -145,7 +145,7 @@ keywords="civic technology, open government", name="cdp-backend", packages=find_packages(exclude=["tests", "*.tests", "*.tests.*"]), - python_requires=">=3.7", + python_requires=">=3.8", setup_requires=setup_requirements, test_suite="cdp_backend/tests", tests_require=test_requirements, diff --git a/tox.ini b/tox.ini index b3c54930..68f11aaa 100644 --- a/tox.ini +++ b/tox.ini @@ -1,5 +1,5 @@ [tox] -envlist = py37, py38, py39, lint +envlist = py38, py39, py310, lint skip_missing_interpreters = true toxworkdir={env:TOX_WORK_DIR:.tox}

Switch docs theme over to Furo Copy docs theme from cdp-data: http://councildataproject.org/cdp-data/index.html

2022-05-01T16:29:25

0.0

[]

CouncilDataProject/cdp-backend

CouncilDataProject__cdp-backend-165

ede007f7be5c42cc7913f1872024dceadcf96f3f

diff --git a/cdp_backend/pipeline/event_index_pipeline.py b/cdp_backend/pipeline/event_index_pipeline.py index c3818d79..831a893b 100644 --- a/cdp_backend/pipeline/event_index_pipeline.py +++ b/cdp_backend/pipeline/event_index_pipeline.py @@ -207,6 +207,7 @@ def read_transcripts_and_generate_grams( cleaned_text=string_utils.clean_text( sentence.text, clean_stop_words=True, + clean_emojis=True, ), n_grams=[], ) diff --git a/cdp_backend/utils/string_utils.py b/cdp_backend/utils/string_utils.py index 457b0d7a..58f81313 100644 --- a/cdp_backend/utils/string_utils.py +++ b/cdp_backend/utils/string_utils.py @@ -12,7 +12,38 @@ ############################################################################### -def clean_text(text: str, clean_stop_words: bool = False) -> str: +def remove_emojis(text: str) -> str: + """ + Minor changes made from this answer on stackoverflow: + https://stackoverflow.com/a/58356570 + """ + emoj_patterns = re.compile( + "[" + "\U0001F600-\U0001F64F" # emoticons + "\U0001F300-\U0001F5FF" # symbols & pictographs + "\U0001F680-\U0001F6FF" # transport & map symbols + "\U0001F1E0-\U0001F1FF" # flags (iOS) + "\U00002702-\U000027B0" + "\U000024C2-\U0001F251" + "\U0001f926-\U0001f937" + "\u2600-\u2B55" + "\u200d" + "\u23cf" + "\u23e9" + "\u231a" + "\ufe0f" # dingbats + "\u3030" + "]+", + re.UNICODE, + ) + return re.sub(emoj_patterns, "", text) + + +def clean_text( + text: str, + clean_stop_words: bool = False, + clean_emojis: bool = False, +) -> str: """ Clean text of common characters and extra formatting. @@ -23,6 +54,9 @@ def clean_text(text: str, clean_stop_words: bool = False) -> str: clean_stop_words: bool Should English stop words be removed from the raw text or not. Default: False (do not remove stop words) + clean_emojis: bool + Should emojis, emoticons, pictograms, and other characters be removed. + Default: False (do not remove pictograms) Returns ------- @@ -57,22 +91,23 @@ def clean_text(text: str, clean_stop_words: bool = False) -> str: STOPWORDS = stopwords.words("english") joined_stopwords = "|".join(STOPWORDS) - cleaned_stopwords = re.sub( + cleaned_text = re.sub( r"\b(" + joined_stopwords + r")\b", "", cleaned_punctuation, ) else: # Update for mypy typing - cleaned_stopwords = cleaned_punctuation + cleaned_text = cleaned_punctuation + + # Remove pictograms + if clean_emojis: + cleaned_text = remove_emojis(cleaned_text) # Remove gaps in string try: - cleaned_doc = re.sub(r" {2,}", " ", cleaned_stopwords) - if cleaned_doc[0] == " ": - cleaned_doc = cleaned_doc[1:] - if cleaned_doc[-1] == " ": - cleaned_doc = cleaned_doc[:-1] + cleaned_doc = re.sub(r" {2,}", " ", cleaned_text) + cleaned_doc = cleaned_doc.strip() # IndexError occurs when the string was cleaned and it contained entirely stop # words or punctuation for some reason

Context span selection fails during index creation  ### Describe the Bug _A clear and concise description of the bug._ https://github.com/CouncilDataProject/seattle-staging/runs/4862736150?check_suite_focus=true Looks like the `closest_term` was `""` and an empty string wasn't found in the list. Regardless of how valuable that is. This was from seattle-staging and we can run and store the index locally so hopefully this will allow us to check out both what term is being processed and what is going on in general.

https://github.com/CouncilDataProject/seattle-staging/runs/4916834417?check_suite_focus=true Another pipeline fail

2022-02-17T07:03:04

0.0

[]

CouncilDataProject/cdp-backend

CouncilDataProject__cdp-backend-132

706b5fedff2ddf9c80dcf8f90a6e33f59beb4ff0

diff --git a/cdp_backend/pipeline/event_gather_pipeline.py b/cdp_backend/pipeline/event_gather_pipeline.py index 7389317f..4359f3cf 100644 --- a/cdp_backend/pipeline/event_gather_pipeline.py +++ b/cdp_backend/pipeline/event_gather_pipeline.py @@ -126,7 +126,15 @@ def create_event_gather_flow( session_processing_results: List[SessionProcessingResult] = [] for session in event.sessions: # Download video to local copy - tmp_video_filepath = resource_copy_task(uri=session.video_uri) + resource_copy_filepath = resource_copy_task(uri=session.video_uri) + + # Convert to mp4 if necessary + tmp_video_filepath = convert_video_to_mp4_and_upload( + video_filepath=resource_copy_filepath, + session=session, + credentials_file=config.google_credentials_file, + bucket=config.validated_gcs_bucket_name, + ) # Get unique session identifier session_content_hash = get_session_content_hash( @@ -274,6 +282,61 @@ def get_session_content_hash( return file_utils.hash_file_contents(uri=tmp_video_filepath) +@task +def convert_video_to_mp4_and_upload( + video_filepath: str, + session: Session, + credentials_file: str, + bucket: str, +) -> str: + """ + Convert a video to MP4 (if necessary), upload it to the file store, and remove + the original non-MP4 file that was resource copied. + + Parameters + ---------- + video_filepath: Union[str, Path] + The local path for video file to convert. + session: Session + The session to append the new MP4 video uri to. + credentials_file: str + Path to Google Service Account Credentials JSON file. + bucket: str + The GCS bucket to store the MP4 file to. + + Returns + ------- + mp4_filepath: str + The local filepath of the converted MP4 file. + """ + + # Get file extension + ext = Path(video_filepath).suffix + + # Convert to mp4 if necessary + if ext != ".mp4": + # Convert video to mp4 + mp4_filepath = file_utils.convert_video_to_mp4(video_filepath) + + # Remove old mkv file + fs_functions.remove_local_file(video_filepath) + + # Upload to gcsfs + mp4_uri = fs_functions.upload_file( + credentials_file=credentials_file, + bucket=bucket, + filepath=mp4_filepath, + ) + + # Set session video_uri to uri in remote file store + session.video_uri = mp4_uri + + # Return converted local mp4 filepath + return mp4_filepath + + return video_filepath + + @task def split_audio( session_content_hash: str, diff --git a/cdp_backend/utils/file_utils.py b/cdp_backend/utils/file_utils.py index a104228a..791a6285 100644 --- a/cdp_backend/utils/file_utils.py +++ b/cdp_backend/utils/file_utils.py @@ -8,6 +8,7 @@ from typing import Optional, Tuple, Union import aiohttp +import ffmpeg import fsspec from fsspec.core import url_to_fs @@ -363,6 +364,27 @@ def hash_file_contents(uri: str, buffer_size: int = 2 ** 16) -> str: return hasher.hexdigest() +def convert_video_to_mp4(video_filepath: str) -> str: + """ + Converts a video to an equivalent MP4 file. + + Parameters + ---------- + video_filepath: str + The filepath of the video to convert. + + Returns + ------- + mp4_filepath: str + The filepath of the converted MP4 video. + """ + mp4_filepath = str(Path(video_filepath).with_suffix(".mp4")) + ffmpeg.input(video_filepath).output(mp4_filepath).overwrite_output().run() + log.info("Finished converting {} to mp4".format(video_filepath)) + + return mp4_filepath + + def generate_file_storage_name(file_uri: str, suffix: str) -> str: """ Generate a filename using the hash of the file contents and some provided suffix.

Ensure or Convert video to MP4 or WebM before processing  ### Feature Description _A clear and concise description of the feature you're requesting._ If the video isn't MP4 / WebM, we won't be able to render the video on the website. Ensure or convert to a "web ready" video format. ### Use Case _Please provide a use case to help us understand your request in context._ Allow more video formats through the pipeline / catch common video recording format errors. (OBS records videos as MKV by default) ### Solution _Please describe your ideal solution._ If the video file isn't MP4 or WEBM, download it, convert, and upload our conversion to file storage. Use our converted video as the basis which would mean hosting it ourselves but :shrug: ### Alternatives _Please describe any alternatives you've considered, even if you've dismissed them._

2021-11-08T05:06:50

0.0

[]

CouncilDataProject/cdp-backend

CouncilDataProject__cdp-backend-128

5b9a478316fe390171c09b59c01ee77714f8db64

diff --git a/cdp_backend/sr_models/google_cloud_sr_model.py b/cdp_backend/sr_models/google_cloud_sr_model.py index c573e8d1..c0b3c5ae 100644 --- a/cdp_backend/sr_models/google_cloud_sr_model.py +++ b/cdp_backend/sr_models/google_cloud_sr_model.py @@ -69,22 +69,22 @@ def transcribe( The transcript model for the supplied media file. """ # Create client - client = speech.SpeechClient.from_service_account_json(self.credentials_file) + client = speech.SpeechClient.from_service_account_file( + filename=str(self.credentials_file) + ) # Create basic metadata - metadata = speech.types.RecognitionMetadata() + metadata = speech.RecognitionMetadata() metadata.interaction_type = ( - speech.enums.RecognitionMetadata.InteractionType.DISCUSSION + speech.RecognitionMetadata.InteractionType.DISCUSSION ) # Add phrases - speech_context = speech.types.SpeechContext( - phrases=self._clean_phrases(phrases) - ) + speech_context = speech.SpeechContext(phrases=self._clean_phrases(phrases)) # Prepare for transcription - config = speech.types.RecognitionConfig( - encoding=speech.enums.RecognitionConfig.AudioEncoding.LINEAR16, + config = speech.RecognitionConfig( + encoding=speech.RecognitionConfig.AudioEncoding.LINEAR16, sample_rate_hertz=16000, language_code="en-US", enable_automatic_punctuation=True, @@ -92,11 +92,13 @@ def transcribe( speech_contexts=[speech_context], metadata=metadata, ) - audio = speech.types.RecognitionAudio(uri=file_uri) + audio = speech.RecognitionAudio(uri=file_uri) # Begin transcription log.debug(f"Beginning transcription for: {file_uri}") - operation = client.long_running_recognize(config, audio) + operation = client.long_running_recognize( + request={"config": config, "audio": audio} + ) # Wait for complete response = operation.result(timeout=10800) @@ -145,10 +147,16 @@ def transcribe( # Extract word from response word = result.alternatives[0].words[w_ind] + # Nanos no longer supported, use microseconds instead + # https://github.com/googleapis/python-speech/issues/71 start_time = ( - word.start_time.seconds + word.start_time.nanos * 1e-9 + word.start_time.seconds + + word.start_time.microseconds * 1e-6 + ) + + end_time = ( + word.end_time.seconds + word.end_time.microseconds * 1e-6 ) - end_time = word.end_time.seconds + word.end_time.nanos * 1e-9 # Add start_time to Sentence if first word if w_ind - w_marker == 0: diff --git a/setup.py b/setup.py index bce2d9c2..079d108e 100644 --- a/setup.py +++ b/setup.py @@ -21,7 +21,7 @@ pipeline_requirements = [ "dask[distributed]>=2021.7.0", "ffmpeg-python~=0.2.0", - "google-cloud-speech~=1.3", + "google-cloud-speech~=2.11.0", "graphviz~=0.16", "imageio~=2.9", "imageio-ffmpeg~=0.4.4",

Upgrade to GCP Speech-to-Text v2 To stay up-to-date we should upgrade to v2 when we get a chance, fortunately they released an upgrading guide here: https://github.com/googleapis/python-speech/blob/release-v2.0.0/UPGRADING.md

2021-11-02T00:19:58

0.0

[]

CouncilDataProject/cdp-backend

CouncilDataProject__cdp-backend-112

8c8dbf4e4d2fdc7e4278206c75d550f345c9cbe6

diff --git a/cdp_backend/pipeline/event_gather_pipeline.py b/cdp_backend/pipeline/event_gather_pipeline.py index 6c609112..ed2c6fe1 100644 --- a/cdp_backend/pipeline/event_gather_pipeline.py +++ b/cdp_backend/pipeline/event_gather_pipeline.py @@ -126,8 +126,10 @@ def create_event_gather_flow( session_processing_results: List[SessionProcessingResult] = [] for session in event.sessions: # Get or create audio + tmp_video_filepath = resource_copy_task(uri=session.video_uri) + session_content_hash, audio_uri = get_video_and_split_audio( - video_uri=session.video_uri, + tmp_video_filepath=tmp_video_filepath, bucket=config.validated_gcs_bucket_name, credentials_file=config.google_credentials_file, ) @@ -167,12 +169,17 @@ def create_event_gather_flow( hover_thumbnail_uri, ) = get_video_and_generate_thumbnails( session_content_hash=session_content_hash, - video_uri=session.video_uri, + tmp_video_path=tmp_video_filepath, event=event, bucket=config.validated_gcs_bucket_name, credentials_file=config.google_credentials_file, ) + # Add audio uri and static thumbnail uri + resource_delete_task( + tmp_video_filepath, upstream_tasks=[audio_uri, static_thumbnail_uri] + ) + # Store all processed and provided data session_processing_results.append( compile_session_processing_result( # type: ignore @@ -197,9 +204,36 @@ def create_event_gather_flow( return flow +@task +def resource_copy_task(uri: str) -> str: + # Copy video file from the internet and store it locally + dirpath = tempfile.mkdtemp() + + file_utils.resource_copy( + uri=uri, + dst=dirpath, + overwrite=True, + ) + + return str(Path(dirpath) / uri.split("/")[-1]) + + +@task +def resource_delete_task(uri: str) -> None: + """ + Remove local file + + Parameters + ---------- + uri: str + The local video file. + """ + fs_functions.remove_local_file(uri) + + @task(nout=2, max_retries=3, retry_delay=timedelta(seconds=60)) def get_video_and_split_audio( - video_uri: str, bucket: str, credentials_file: str + tmp_video_filepath: str, bucket: str, credentials_file: str ) -> Tuple[str, str]: """ Download (or copy) a video file to temp storage, split's the audio, and uploads @@ -207,7 +241,7 @@ def get_video_and_split_audio( Parameters ---------- - video_uri: str + tmp_video_filepath: str The URI to the video file to split audio from. bucket: str The name of the GCS bucket to upload the produced audio to. @@ -226,16 +260,6 @@ def get_video_and_split_audio( We sometimes get file downloading failures when running in parallel so this has two retries attached to it that will run after a failure on a 3 minute delay. """ - # Get just the video filename from the full uri - # TODO: Handle windows file splitting??? - # Generally this is a URI but we do have a goal of a local file processing too... - video_filename = video_uri.split("/")[-1] - tmp_video_filepath = file_utils.resource_copy( - uri=video_uri, - dst=f"audio-splitting-{video_filename}", - overwrite=True, - ) - # Hash the video contents session_content_hash = file_utils.hash_file_contents(uri=tmp_video_filepath) @@ -285,9 +309,6 @@ def get_video_and_split_audio( ]: fs_functions.remove_local_file(local_path) - # Always remove tmp video file - fs_functions.remove_local_file(tmp_video_filepath) - return session_content_hash, audio_uri @@ -715,7 +736,7 @@ def generate_transcript( @task(nout=2) def get_video_and_generate_thumbnails( session_content_hash: str, - video_uri: str, + tmp_video_path: str, event: EventIngestionModel, bucket: str, credentials_file: str, @@ -727,7 +748,7 @@ def get_video_and_generate_thumbnails( ---------- session_content_hash: str The unique key (SHA256 hash of video content) for this session processing. - video_uri: str + tmp_video_path: str The URI to the video file to generate thumbnails from. event: EventIngestionModel The parent event of the session. If no captions are available, @@ -744,12 +765,6 @@ def get_video_and_generate_thumbnails( hover_thumbnail_url: str The URL of the hover thumbnail, stored on GCS. """ - # Create tmp directory to save file in - dirpath = tempfile.mkdtemp() - dst = Path(dirpath) - - tmp_video_path = file_utils.resource_copy(uri=video_uri, dst=dst) - if event.static_thumbnail_uri is None: # Generate new static_thumbnail_file = file_utils.get_static_thumbnail( @@ -784,8 +799,6 @@ def get_video_and_generate_thumbnails( ) fs_functions.remove_local_file(hover_thumbnail_file) - fs_functions.remove_local_file(tmp_video_path) - return ( static_thumbnail_url, hover_thumbnail_url,

Combine all the video processing tasks into a single task  ### Feature Description _A clear and concise description of the feature you're requesting._ Reduce the pipeline computation time by combining all the tasks that require the video on the worker together. ### Use Case _Please provide a use case to help us understand your request in context._ Drastically reducing the pipeline computation time. ### Solution _Please describe your ideal solution._ Move the function calls to generate thumbnails to the start of the pipeline with the strip audio and generate hash. ### Alternatives _Please describe any alternatives you've considered, even if you've dismissed them._

2021-09-27T04:21:23

0.0

[]

CouncilDataProject/cdp-backend

CouncilDataProject__cdp-backend-59

9d0332cdf85747faa2c3c04b8ae991e43d7df4c4

diff --git a/cdp_backend/sr_models/google_cloud_sr_model.py b/cdp_backend/sr_models/google_cloud_sr_model.py index 4a213aeb..c6a1dc0e 100644 --- a/cdp_backend/sr_models/google_cloud_sr_model.py +++ b/cdp_backend/sr_models/google_cloud_sr_model.py @@ -8,6 +8,7 @@ from typing import Any, List, Optional, Union from google.cloud import speech_v1p1beta1 as speech +from spacy.lang.en import English from ..pipeline.transcript_model import Sentence, Transcript, Word from .sr_model import SRModel @@ -107,28 +108,65 @@ def transcribe( # Create timestamped sentences timestamped_sentences: List[Sentence] = [] - current_sentence = None - sentence_index = 0 - word_index = 0 + transcript_sentence_index = 0 + + # Create sentence boundary pipeline + nlp = English() + nlp.add_pipe("sentencizer") for result in response.results: # Some portions of audio may not have text if len(result.alternatives) > 0: - # Check length of transcript result - word_list = result.alternatives[0].words - if len(word_list) > 0: - for word in word_list: - # create Word + # Split transcript into sentences + doc = nlp(result.alternatives[0].transcript) + + # Convert generator to list + sentences = [str(sent) for sent in doc.sents] + + # Index holder for word results of response + w_marker = 0 + + for s_ind in range(0, len(sentences)): + # Sentence text + s_text = sentences[s_ind] + + num_words = len(s_text.split()) + + # Initialize sentence model + timestamped_sentence = Sentence( + index=transcript_sentence_index, + confidence=result.alternatives[0].confidence, + # Start and end time are placeholder values + start_time=0.0, + end_time=0.0, + words=[], + text=s_text, + ) + + for w_ind in range(w_marker, w_marker + num_words): + # Extract word from response + word = result.alternatives[0].words[w_ind] + start_time = ( word.start_time.seconds + word.start_time.nanos * 1e-9 ) end_time = word.end_time.seconds + word.end_time.nanos * 1e-9 + # Add start_time to Sentence if first word + if w_ind - w_marker == 0: + timestamped_sentence.start_time = start_time + + # Add end_time to Sentence if last word + if (w_ind - w_marker) == (num_words - 1): + timestamped_sentence.end_time = end_time + # Clean everything but non-delimiting characters make lowercase regex = re.compile(r"[^a-zA-Z0-9'\-]") cleaned_word = regex.sub("", word.word).lower() + + # Create Word model timestamped_word = Word( - index=word_index, + index=w_ind - w_marker, start_time=start_time, end_time=end_time, text=cleaned_word, @@ -136,43 +174,20 @@ def transcribe( annotations=None, ) - if current_sentence is None: - current_sentence = Sentence( - index=sentence_index, - confidence=result.alternatives[0].confidence, - start_time=word.start_time, - end_time=word.end_time, - # TODO: Add speaker and annotations? - words=[timestamped_word], - text=word.word, - ) - word_index += 1 - - # End current sentence and reset - # TODO: Account for non-sentence ending periods, such as - # prefixes like "Mr." or "Dr." - elif bool(re.match(r"\.|\?", word.word[-1])): - # Finish sentence and append - current_sentence.end_time = word.end_time - current_sentence.text += " {}".format(word.word) - current_sentence.words.append(timestamped_word) - - timestamped_sentences.append(current_sentence) - - # Adjust indices - current_sentence = None - sentence_index += 1 - word_index = 0 - - # Update current sentence - else: - current_sentence.text += " {}".format(word.word) - current_sentence.words.append(timestamped_word) - word_index += 1 - - # Update confidence stats - confidence_sum += result.alternatives[0].confidence - segments += 1 + timestamped_sentence.words.append(timestamped_word) + + # Increment word marker + w_marker += num_words + + # Add Sentence to sentence list + timestamped_sentences.append(timestamped_sentence) + + # Increment transcript sentence index + transcript_sentence_index += 1 + + # Update confidence stats + confidence_sum += result.alternatives[0].confidence + segments += 1 # Compute mean confidence if segments > 0: diff --git a/setup.py b/setup.py index 12666cd3..bbcaed92 100644 --- a/setup.py +++ b/setup.py @@ -58,6 +58,7 @@ "pulumi~=3.0.0", "pulumi-google-native~=0.1.0", "pulumi-gcp~=5.0.0", + "spacy~=3.0.6", "truecase>=0.0.9", "webvtt-py>=0.4.5", ]

Use spacy to separate sentences in SR model transcribe function ## Use Case To account for edge cases in sentence detection (Prefixes like Mr.), we can use the [spacy](https://stackoverflow.com/questions/46290313/how-to-break-up-document-by-sentences-with-with-spacy) library. ## Solution Use the spacy library to separate sentences. Haven't looked into the library too deeply, but would probably require us to change the iteration logic since we create `Word` (and increment index) and potentially end the `Sentence` in the same iteration.

2021-05-31T07:13:54

0.0

[]

CouncilDataProject/cdp-backend

CouncilDataProject__cdp-backend-38

86b9acc5019dd68989e6934805dc7592999b5a0f

diff --git a/.gitignore b/.gitignore index b81f174a..0bb34bc8 100644 --- a/.gitignore +++ b/.gitignore @@ -114,4 +114,5 @@ ENV/ # Project specific cdp_database_diagram.* +cdp_event_gather_flow_* event_data_ingestion_model.md diff --git a/Makefile b/Makefile index 2bf92b8f..5a605fcd 100644 --- a/Makefile +++ b/Makefile @@ -56,6 +56,7 @@ gen-docs: ## generate Sphinx HTML documentation, including API docs create_cdp_database_uml -o docs/_static/cdp_database_diagram.dot dot -T png -o docs/_static/cdp_database_diagram.png docs/_static/cdp_database_diagram.dot create_cdp_ingestion_model_doc -t docs/event_data_ingestion_model.template -o docs/event_data_ingestion_model.md + create_cdp_event_gather_flow_viz -o docs/_static/cdp_event_gather_flow_{ftype}.png sphinx-apidoc -o docs/ cdp_backend **/tests/ $(MAKE) -C docs html diff --git a/cdp_backend/bin/cdp_event_gather.py b/cdp_backend/bin/cdp_event_gather.py index 81f9c46f..51f83b65 100644 --- a/cdp_backend/bin/cdp_event_gather.py +++ b/cdp_backend/bin/cdp_event_gather.py @@ -9,7 +9,7 @@ from pathlib import Path from typing import Callable -from ..pipeline import cdp_event_gather_pipeline as pipeline +from cdp_backend.pipeline import cdp_event_gather_pipeline as pipeline ############################################################################### @@ -34,7 +34,7 @@ def __parse(self) -> None: p.add_argument( "-g", "--google-credentials-file", - default=(Path(__file__).parent.parent.parent / "cdp-dev-infra-creds.json"), + default=(Path(__file__).parent.parent.parent / "cdp-creds.json"), type=Path, dest="google_credentials_file", help="Path to the Google Service Account Credentials JSON file.", diff --git a/cdp_backend/bin/create_cdp_event_gather_flow_viz.py b/cdp_backend/bin/create_cdp_event_gather_flow_viz.py new file mode 100644 index 00000000..7c35f0e4 --- /dev/null +++ b/cdp_backend/bin/create_cdp_event_gather_flow_viz.py @@ -0,0 +1,100 @@ +#!/usr/bin/env python +# -*- coding: utf-8 -*- + +import argparse +import logging +import sys +import traceback +from pathlib import Path +from typing import List + +from cdp_backend.pipeline import cdp_event_gather_pipeline as pipeline +from cdp_backend.pipeline.ingestion_models import ( + EXAMPLE_FILLED_EVENT, + EXAMPLE_MINIMAL_EVENT, + EventIngestionModel, +) + +############################################################################### + +logging.basicConfig( + level=logging.INFO, + format="[%(levelname)4s: %(module)s:%(lineno)4s %(asctime)s] %(message)s", +) +log = logging.getLogger(__name__) + +############################################################################### + + +class Args(argparse.Namespace): + def __init__(self) -> None: + self.__parse() + + def __parse(self) -> None: + p = argparse.ArgumentParser( + prog="cdp_event_gather", + description="Gather, process, and store event data to CDP infrastructure.", + ) + p.add_argument( + "-o", + "--output-file-spec", + type=Path, + default=Path("cdp_event_gather_flow_{ftype}.png"), + dest="output_file", + help=( + "Path spec to where to store the created PNG files. " + "Use `{ftype}` as the formatted spec parameter." + ), + ) + p.parse_args(namespace=self) + + +def fake_get_events_minimal() -> List[EventIngestionModel]: + return [EXAMPLE_MINIMAL_EVENT] + + +def fake_get_events_filled() -> List[EventIngestionModel]: + return [EXAMPLE_FILLED_EVENT] + + +def fake_get_events_many() -> List[EventIngestionModel]: + return [EXAMPLE_MINIMAL_EVENT] * 4 + + +def main() -> None: + try: + args = Args() + minimal_flow = pipeline.create_cdp_event_gather_flow( + fake_get_events_minimal, "" + ) + minimal_flow.visualize( + filename=str(args.output_file.with_suffix("")).format(ftype="minimal"), + format="png", + ) + + filled_flow = pipeline.create_cdp_event_gather_flow(fake_get_events_filled, "") + filled_flow.visualize( + filename=str(args.output_file.with_suffix("")).format(ftype="filled"), + format="png", + ) + + many_flow = pipeline.create_cdp_event_gather_flow(fake_get_events_many, "") + many_flow.visualize( + filename=str(args.output_file.with_suffix("")).format(ftype="many"), + format="png", + ) + + except Exception as e: + log.error("=============================================") + log.error("\n\n" + traceback.format_exc()) + log.error("=============================================") + log.error("\n\n" + str(e) + "\n") + log.error("=============================================") + sys.exit(1) + + +############################################################################### +# Allow caller to directly run this module (usually in development scenarios) + +if __name__ == "__main__": + main() diff --git a/cdp_backend/pipeline/cdp_event_gather_pipeline.py b/cdp_backend/pipeline/cdp_event_gather_pipeline.py index 72b3766d..ed85d218 100644 --- a/cdp_backend/pipeline/cdp_event_gather_pipeline.py +++ b/cdp_backend/pipeline/cdp_event_gather_pipeline.py @@ -29,9 +29,6 @@ def create_cdp_event_gather_flow( get_events_func: Callable, credentials_file: str, ) -> Flow: - # Initialize fireo connection - fireo.connection(from_file=credentials_file) - # Create flow with Flow("CDP Event Gather Pipeline") as flow: events: List[EventIngestionModel] = get_events_func() @@ -42,18 +39,24 @@ def create_cdp_event_gather_flow( # Upload calls for minimal event body_ref = upload_db_model( - create_body_from_ingestion_model(event.body), event.body + create_body_from_ingestion_model(event.body), + event.body, + creds_file=credentials_file, ) # TODO add upload calls for non-minimal event event_ref = upload_db_model( - create_event_from_ingestion_model(event, body_ref), event + create_event_from_ingestion_model(event, body_ref), + event, + creds_file=credentials_file, ) for session in event.sessions: upload_db_model( - create_session_from_ingestion_model(session, event_ref), session + create_session_from_ingestion_model(session, event_ref), + session, + creds_file=credentials_file, ) return flow @@ -72,7 +75,6 @@ def update_db_model(db_model: Model, ingestion_model: Any, db_key: str) -> Model ] needs_update = False - for field in non_primary_db_fields: if hasattr(ingestion_model, field): db_val = getattr(db_model, field) @@ -93,9 +95,11 @@ def update_db_model(db_model: Model, ingestion_model: Any, db_key: str) -> Model @task -def upload_db_model(db_model: Model, ingestion_model: Any) -> Model: - uniqueness_validation = get_model_uniqueness(db_model) +def upload_db_model(db_model: Model, ingestion_model: Any, creds_file: str) -> Model: + # Initialize fireo connection + fireo.connection(from_file=creds_file) + uniqueness_validation = get_model_uniqueness(db_model) if uniqueness_validation.is_unique: db_model.save() log.info( diff --git a/docs/event_gather_pipeline.md b/docs/event_gather_pipeline.md new file mode 100644 index 00000000..23d78432 --- /dev/null +++ b/docs/event_gather_pipeline.md @@ -0,0 +1,64 @@ +# Event Gather Pipeline + +Council Data Project documentation generally shortens the steps that the +event gather pipeline takes into three short and easy to remember key points. + +1. **Gather Events** + This function is created by each instance maintainer and + generally scrapes or requests data from an API and + returns a list of events to process and store. + [Learn more about the event ingestion model](./cdp-backend/event_data_ingestion_model.html) +2. **Transcribe** + If the event was provided with a closed caption file we clean the file up + into the CDP format, if not, CDP tools generate a transcription. +3. **Index** + An entirely different pipeline from the event gather pipeline but in + the context of the short and sweet version, this makes sense to include. + This is the pipeline that runs to generate and index using all the event + transcripts. + +These key points make sense from the thousand-foot-view, but is not +satifactory for the developers and others who are simply interested in +how the event gather pipeline works step-by-step. + +This document gives visualizations of each and every step the event gather +pipeline takes, starting immediately after the maintainer provided function +returns the `List[EventIngestionModel]`. + +### Minimal Event Processing + +Similar to the [event ingestion model](./cdp-backend/event_data_ingestion_model.html) +documentation, if the event is provided back with only a URI to a +video and nothing else, the processing is incredibly simple. + +![minimal cdp event gather flow viz](./_static/cdp_event_gather_flow_minimal.png) + +### Filled Event Processing + +The major difference from the minimal event processing workflow is that the +pipeline will process, prepare, and upload all of the attachments. + +![filled cdp event gather flow viz](./_static/cdp_event_gather_flow_filled.png) + +### Workflow Management and Parallel Processing + +We utilize [Prefect Core](https://github.com/prefecthq/prefect) to generate and +run the workflow. In general, and by default in our cookiecutter-cdp-deployment, +we do not run any part of the workflow in parallel. But, this is simply because we +generally don't need to. As all event gather pipelines are automated to run every day, +or multiple times a day, it is rare for a pipeline to pick up more than one event +to process. + +However, in the case where a maintainer is attempting to backfill large collections of +events (i.e. processing historical events), we recommend attaching a +[Dask Executor](https://docs.prefect.io/api/latest/executors.html#daskexecutor) +to the flow for parallel event processing. +[More details about Dask](https://dask.org/) + +I.E. +If the result of the `get_events` function was a list of four +`EventIngestionModel` objects, the resulting flow would look like the following: +![parallel four wide equal minimal event flows](./_static/cdp_event_gather_flow_many.png) + +And, if a `DaskExecutor` is also provided, all single event processing workflows +would process in parallel (provided enough compute resources). diff --git a/docs/event_gather_pipeline.rst b/docs/event_gather_pipeline.rst new file mode 100644 index 00000000..f1a2ebf2 --- /dev/null +++ b/docs/event_gather_pipeline.rst @@ -0,0 +1,1 @@ +.. mdinclude:: ./event_gather_pipeline.md diff --git a/docs/index.rst b/docs/index.rst index bef157d1..935488c5 100644 --- a/docs/index.rst +++ b/docs/index.rst @@ -10,6 +10,7 @@ Welcome to cdp-backend's documentation! installation Package Modules <modules> Event Data Ingestion Model <event_data_ingestion_model> + Event Gather Pipeline <event_gather_pipeline> Database Schema <database_schema> contributing diff --git a/setup.py b/setup.py index e78e72ce..7b180b66 100644 --- a/setup.py +++ b/setup.py @@ -34,6 +34,7 @@ "ipython>=7.15.0", "jinja2>=2.11.2", "m2r>=0.2.1", + "prefect[viz]~=0.14.0", "pytest-runner>=5.2", "Sphinx>=2.0.0b1,<3", "sphinx_rtd_theme>=0.4.3", @@ -43,15 +44,15 @@ ] requirements = [ - "dask[bag]~=2.30.0", + "dask[bag]~=2020.12.0", "fireo~=1.3.7", "fsspec~=0.8.3", "gcsfs~=0.7.1", "graphviz~=0.14", "pandas~=1.1.3", + "prefect~=0.14.0", "pulumi~=2.16.2", "pulumi-gcp~=4.6.0", - "prefect~=0.14.0" ] extra_requirements = { @@ -87,6 +88,10 @@ "create_cdp_ingestion_model_doc=" "cdp_backend.bin.create_ingestion_model_doc:main" ), + ( + "create_cdp_event_gather_flow_viz=" + "cdp_backend.bin.create_cdp_event_gather_flow_viz:main" + ), "cdp_event_gather=cdp_backend.bin.cdp_event_gather:main", ], },

Create a bin script to generate the flow visualization Create a bin script to generate event gather flow visualization to attach to documentation after #16 is completed. _Originally posted by @JacksonMaxfield in https://github.com/CouncilDataProject/cdp-backend/pull/33#discussion_r547063942_

2021-01-03T03:46:08

0.0

[]